News for Your Practice

Although laparotomy is still considered the standard for ovarian cyst removal, over the past 15 years minimally invasive surgery has gained wider acceptance in cases where preoperative assessment suggests an adnexal mass is benign.

Unfortunately, minimally invasive management of a large ovarian cyst (greater than 10 cm) is particularly challenging for several reasons:

• The cyst can rupture and spill its contents into the peritoneum,
• the cyst’s size limits the surgical field, and
• an unexpected malignancy may be revealed.

An innovative minilaparotomy technique for the removal of benign ovarian cysts offers the advantages of laparoscopic and laparoscopic-assisted procedures while bypassing the major disadvantages: the necessity for specialized and expensive equipment, lengthy operative time, and long learning curves.¹ (The minimally invasive procedures currently available for the treatment of ovarian cysts include laparoscopic cystectomy, laparoscopic-assisted minilaparotomy cystectomy, laparoscopic-assisted vaginal cystectomy, combined percutaneous ultrasound cyst aspiration and laparoscopic cystectomy, transvaginal cystectomy, and the traditional minilaparotomy cystectomy.^2-10)

The procedure is faster, less expensive, carries fewer potential risks than traditional alternatives, and offers these advantages:

• can be performed under regional anesthesia
• relies on standard open techniques
• uses inexpensive instrumentation
• is easy to learn
• can be used for very large cysts
• eliminates the risk of intraperitoneal spillage of cyst contents
• offers similar postoperative convalescence and mean time to return to work as laparoscopic or laparoscopic-assisted management of large ovarian cysts

General Ob/Gyns—not gynecologic oncologists—perform most surgeries on patients with adnexal masses, since ovarian cancer is relatively uncommon in the absence of preoperative risk factors for malignancy. Our approach offers an appealing option to Ob/Gyns reluctant to abandon routine traditional laparotomy for such ovarian cysts.

Selecting the right patient

Adequate preoperative assessment diminishes the risk of unexpected malignancy in a patient undergoing surgery for an ovarian mass to less than 1%.⁴ At this time, the combination of menopausal status, cancer antigen (CA) 125 level, physical examination, and ultrasound is the best strategy for evaluating the patient with an ovarian cyst.¹¹

Signs of malignancy. Ultrasound features that suggest malignancy include irregular borders, thick septa, solid areas, internal and external excrescences, matted bowel, and ascites. Benign cysts, on the other hand, are usually unilateral and have regular borders, thin septa, no solid areas, and no internal excrescences.⁴ The measurement of blood flow within the mass by color Doppler may improve the accuracy of ultrasound in differentiating benign from malignant cysts.^4,12

On physical examination, an adnexal mass that is fixed, irregular, or solid also suggests a neoplasm. An elevated CA 125 combined with a complex adnexal mass is likely to be associated with malignancy. The test is even more specific in postmenopausal women with adnexal masses.^4,12

However, plasma levels of CA 125 also can be elevated in several benign gynecologic conditions such as endometriosis, simple ovarian cysts, pelvic inflammatory disease, ovarian torsion, fibroids, and in physiologic conditions such as menstruation and pregnancy.¹³

Anticipate the need to convert to laparotomy. Every patient’s surgical consent should include a possible conversion to laparotomy. To avoid incomplete surgical treatment and significant delays in proper therapy, a gynecologic surgeon experienced in the management of ovarian cancer should be readily available, in the event an unexpected malignancy is encountered. Ideally, the staging surgery and definite treatment should be performed at the time of initial minilaparotomy. Comprehensive surgical staging and treatment include thorough exploration of the pelvis and abdomen, omentectomy, pelvic and paraaortic lymph node sampling, multiple peritoneal biopsies and washings, bilateral salpingo-oophorectomy, hysterectomy, and debulking, when indicated.

Prepare for surgery with position, incision, and retraction

Before beginning, it is crucial to correctly position the patient, make the appropriate incision, and insert the right retractor.¹

Position. After administering regional or general anesthesia, place the patient in a modified lithotomy, as for laparoscopic surgery. Tuck the arms alongside the torso and place the legs in boot stirrups. Avoid hip flexion and allow adequate thigh abduction to expose the vagina. Perform a careful pelvic examination to determine the size and mobility of the adnexal mass.

When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure

Place an indwelling transurethral catheter, and pass a sturdy, hinged uterine manipulator such as the Pelosi Uterine Manipulator (Apple Medical Corp, Marlboro, Mass) transcervically into the uterine cavity (FIGURE 1).

 

The cruciate incision. With a conventional scalpel make a small suprapubic transverse incision through the skin and subcutaneous tissue (FIGURE 2). After clearing subcutaneous fat from the midline, incise the rectus fascia and the peritoneum in a vertical direction.

A vertical skin incision can be selected if the preoperative workup suggests a later extension of the original minilaparotomy incision may be required, or if there is a prior vertical incision.

Retraction. Use a soft sleeve-type self-retaining plastic retractor, such as Mobius (Apple Medical Corp) (FIGURE 3 ). When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure of the pelvis (FIGURE 4).

During surgery it may be necessary to adjust the outer ring if the sleeve loosens. Narrow Deaver or Richardson retractors, if required, provide additional retraction. The bowel may be gently packed, if necessary, but typically it is adequately displaced by the large ovarian cyst.

The atraumatic retraction provided by the soft, self-retaining abdominal retractor minimizes the possibility of tissue trauma, nerve damage, bruising, and postoperative pain. At the same time, the continuous 360° retraction force on the incision maximizes surgical exposure, providing a significantly larger working area than conventional retractors. For example, when applied to a 6-cm incision, the self-retaining retractor creates a 28-cm² exposed working area, compared with only 18 cm² provided by a conventional 4-point metal retractor.

The adjustable height of the self-retaining retractor adapts to wounds of varying depth and works on virtually any tissue thickness—a feature that makes the device effective for obese patients. Further, by lining the abdominal incision, the retractor’s plastic sleeve protects the wound’s edges from contamination and potential implantation of malignant cells, making the device ideal for managing ovarian cysts.

FIGURE 1 Hinged uterine manipulator

The manipulator facilitates exposure of the cyst and contralateral adnexa, as well as uterine elevation/rotation.

FIGURE 2 Cruciate incision

Make a 2.5- to 5-cm suprapubic transverse skin incision. Using the Bovie device, incise the subcutaneous fat transversely along the full length of the skin incision down to the level of the anterior rectus fascia. Clear the subcutaneous fat from the midline superiorly and inferiorly to expose 5 to 6 cm of the rectus fascia in the vertical axis. Use blunt digital dissection to assist in mobilizing the subcutaneous fat. Incise the anterior rectus fascia vertically through the full length of the cleared area. Retract the rectus muscles from the midline to expose the underlying transversalis fascia and the peritoneum. Control small bleeding points with the Bovie device. Enter the peritoneum either digitally or with scissors above the bladder dome and extend the peritoneal incision to the full length of the fascial incision.

FIGURE 3 Self-retaining retractor

The retractor consists of a flexible plastic inner ring and a firmer outer ring connected by a soft plastic sleeve.

FIGURE 4360° retraction force

Squeeze the inner ring of the soft, sleeve-type, self-retaining retractor into the peritoneal cavity through the cruciate abdominal incision and allow it to spring open against the parietal peritoneum. Make a digital assessment to ensure that viscera is not trapped between the inner ring and the abdominal wall. Place the entire sleeve on traction by lifting the outer ring. Then roll the outer ring onto the sleeve, collecting excess length, until it sits firmly against the skin. The atraumatic retraction provided by the sleeve-type retractor maximizes exposure of the pelvis. Note how a portion of the large cyst is clearly seen through the atraumatic, circular area of retraction.

Cyst assessment

Visually and digitally inspect the cyst and carefully evaluate the uterus, pelvis, and contralateral adnexa. Determine the extent of adhesions and any unexpected pelvic pathology. When needed, use traditional small retractors or gentle packing to gain additional exposure. If the cystic mass appears suspicious (internal and external excrescences on the cyst, ovaries, or peritoneal surfaces, or ascites), obtain pelvic washings with a suction-irrigation cannula, send the fluid for cytologic examination, and convert the minilaparotomy to a standard exploratory laparotomy. Extensive adhesions to the bowel, broad ligament, or pelvic sidewall and unexpected extensive endometriosis may also require a conversion to standard laparotomy.

Reduce cyst size by decompression

Simple aspiration is inadvisable. To remove a large ovarian cyst using the Pelosi minilaparotomy, reduce the size of the cyst to permit safe and effective mobilization and resection through the small abdominal incision. Simple aspiration of the cyst, with its potential for spilling the contents, is not a wise strategy for several reasons. First, many ovarian cysts contain functional epithelium with a high recurrence rate (8% to 67%). Second, studies show that 10% to 66% of ovarian cyst fluid aspirates initially diagnosed as benign actually are malignant. Further, relying on negative cytology from the aspirate of an ovarian cyst without tissue biopsy may delay appropriate surgery, and the puncture of an unexpected malignant cyst may seed the peritoneal cavity and possibly worsen the patient’s prognosis.^11,12

 

This technique, suitable for all ovarian cysts, makes it possible to aspirate a large cyst without leakage.

Whether spillage of cancer cells actually worsens the prognosis of a patient with a neoplastic cyst remains controversial because of conflicting study results. Nonetheless, the possibility of intraperitoneal dissemination of neoplastic cells from a ruptured cyst cannot be considered innocuous, and the potential negative effect on a patient’s prognosis should not be ignored.¹⁴ Make every attempt, therefore, to avoid rupturing the cyst and spilling the fluid into the peritoneal cavity.

A shared flaw plagues aspiration devices. Different devices are available for intraoperative cyst aspiration during laparoscopic, transvaginal, or laparotomy approaches.^3,4,6 In addition to long needles, drainage trocars, suction cannulas, and suprapubic bladder catheters, special aspiration instruments have been developed.¹⁵ They include a metal vacuum system with an aspirator trocar that seals the surface of the cyst, and a catheter system that pinches the punctured cyst wall between double balloons to prevent spillage.¹⁶ In addition, several commercial bags are available to prevent intraperitoneal spillage during removal of ovarian cysts.

All these devices have a universal flaw, however: After a thin-walled cyst initially is punctured, none of these products can prevent the spontaneous dehiscence of the cyst and the resulting spillage of its contents into the abdominal cavity. Vacuum systems work well for large cysts with smooth, round surfaces, but in those with irregular surfaces, both application and maintaining the seal are difficult.¹⁷ Fortunately, our technique makes it possible to aspirate a large ovarian cyst without leakage, and the method is suitable for all ovarian cysts regardless of surface type or wall thickness.

Glue the dressing to the cyst to capture leakage. Using a gauze pad, carefully dry the area of the ovarian cyst that is visible through the self-retaining retractor. Then generously spread sterile surgical glue such as Dermabond (Ethicon, Somerville, NJ) on the cyst wall surface (FIGURE 5A). Dermabond is the commercial name for 2-octyl cyanoacrylate, a sterile skin adhesive used as an alternative to stitches to close the edges of small wounds. It is similar to commercial adhesives such as Super Glue and Krazy Glue.

Remove the paper cover of a transparent plastic surgical dressing and place the adhesive side directly onto the glued cyst surface until you are sure the adhesive is completely fixed (FIGURES 5B AND 5C). The 35 cm x 35 cm Steri-Drape or the transparent Tegaderm (both from 3M Health Care, St. Paul, Minn) dressing is effective. A standard nonadhesive plastic dressing or a sterile plastic bag also can be used, as long as the free edges extend beyond the outer rim of the self-retaining retractor.

With a needle aspirator, pierce the cyst through the glued plastic dressing and carefully aspirate the fluid (FIGURES 6A AND 6B). Any leakage is trapped inside the plastic dressing rather than the abdominal cavity. Continue the aspiration until the collapsed cyst and ovary can be gradually delivered through the abdominal incision (FIGURE 6C). Note that the selfretaining retractor also protects the abdominal incision from potential contamination and implantation of neoplastic cells.

Once the cyst and ovary are extracted, you can readily perform an extracorporeal cystectomy, after which the repaired ovary is returned to the abdominal cavity. Be careful to avoid letting any fluid flow back into the peritoneal cavity during cyst removal. When indicated, an extracorporeal adnexectomy can readily be performed.

The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst.

Dermoid cysts require extra care. Preventing intraperitoneal spillage is especially important when removing a large dermoid cyst, to avoid the possibility of chemical peritonitis, dense adhesions, and fistulas. The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst. Quite frequently, a large-diameter suction cannula is required to remove the waxy contents. If the dermoid cyst is very large, aspiration alone will not empty it entirely. Complete emptying is not necessary, however. The primary goal of aspiration is to reduce the size and tension of the cyst to permit delivery through the abdominal incision.

Closing the cruciate incision is quicker and requires less exposure than a scaled-down Pfannenstiel’s incision.

Before concluding the procedure, irrigate the peritoneal cavity to remove any remnants of cyst contents that may have spilled—especially important with a dermoid cyst. Perform any additional indicated procedures through the minilaparotomy incision, such as a contralateral ovarian cystectomy, salpingo-oophorectomy, or hysterectomy. After the surgery is completed, remove the self-retaining retractor by hooking a finger through the bottom ring and pulling it gently out of the incision. Closing the cruciate incision is quicker and requires less exposure to complete than a scaled-down Pfannenstiel’s incision. Apply a vertical pressure dressing over the incision to prevent postoperative wound hematoma or seroma formation. Remove the dressing 24 hours later.

 

FIGURE 5A ‘Leak-proofing’ the aspiration

After drying the surface of the cyst, apply sterile surgical glue to the cyst wall (2 to 3 ampules are usually required to cover the cyst surface).

FIGURE 5B ‘Leak-proofing’ the aspiration

Apply the adhesive side of a clear plastic wound dressing directly onto the cyst’s surface, making sure that the dressing is large enough to fully cover the self-retaining retractor. Using either a piece of folded gauze or your fingers, press the plastic dressing against the adhesive-coated cyst for about 3 to 5 minutes until the adhesive is completely fixed.

FIGURE 5C ‘Leak-proofing’ the aspiration

Remove the paper cover of the wound dressing.

FIGURE 6A Aspirating the cyst

Pierce the dressing and carefully aspirate the fluid until the cyst is partially collapsed.

FIGURE 6B Aspirating the cyst

Place atraumatic clamps on the cyst wall to further control drainage. Any leakage is trapped inside the plastic dressing rather than draining into the abdominal cavity.

FIGURE 6C Aspirating the cyst

Continue the aspiration until the collapsed cyst and ovary can be delivered gradually through the abdominal incision, after detaching the adhesive plastic dressing from the edges of the self-retaining retractor. The portion of the plastic sleeve that is glued to the cyst surface remains attached to the cyst until the cyst is extracted. Following extraction, perform an extracorporeal cystectomy and return the repaired ovary to the abdominal cavity.

Good results

Using our approach, we have treated 38 patients with ovarian cysts of diameters greater than 20 cm thought to be benign by preoperative workup (FIGURE 7). We encountered no malignancies. All surgeries were successfully completed without laparoscopy or conversion to traditional laparotomy and with good cosmesis (FIGURE 8).

Other procedures performed in some of these patients using the same technique included contralateral cystectomy, salpingo-oophorectomy, subtotal and total hysterectomy, adhesiolysis, and appendectomy. We encountered no intraoperative or postoperative complications. Operating times ranged from 18 to 65 minutes. All patients were discharged home within 36 hours and returned to work in a mean of 12 days. Pathology findings of the ovarian cysts included endometrioma, dermoid cyst, serous cystadenoma, and mucinous cystadenoma.

FIGURE 7 Effective for large cysts

Following removal, the collapsed cyst was refilled with 1,300 mL of normal saline to demonstrate its actual size.

FIGURE 8 Good cosmetic results

Cosmetic appearance of the small cruciate abdominal incision 10 days after surgery.Dr. Pelosi II reports that he is a consultant for Apple Medical Corporation. Dr. Pelosi III reports no affiliations or financial arrangements with any companies whose products are mentioned in this article.

Although laparotomy is still considered the standard for ovarian cyst removal, over the past 15 years minimally invasive surgery has gained wider acceptance in cases where preoperative assessment suggests an adnexal mass is benign.

Unfortunately, minimally invasive management of a large ovarian cyst (greater than 10 cm) is particularly challenging for several reasons:

• The cyst can rupture and spill its contents into the peritoneum,
• the cyst’s size limits the surgical field, and
• an unexpected malignancy may be revealed.

An innovative minilaparotomy technique for the removal of benign ovarian cysts offers the advantages of laparoscopic and laparoscopic-assisted procedures while bypassing the major disadvantages: the necessity for specialized and expensive equipment, lengthy operative time, and long learning curves.¹ (The minimally invasive procedures currently available for the treatment of ovarian cysts include laparoscopic cystectomy, laparoscopic-assisted minilaparotomy cystectomy, laparoscopic-assisted vaginal cystectomy, combined percutaneous ultrasound cyst aspiration and laparoscopic cystectomy, transvaginal cystectomy, and the traditional minilaparotomy cystectomy.^2-10)

The procedure is faster, less expensive, carries fewer potential risks than traditional alternatives, and offers these advantages:

• can be performed under regional anesthesia
• relies on standard open techniques
• uses inexpensive instrumentation
• is easy to learn
• can be used for very large cysts
• eliminates the risk of intraperitoneal spillage of cyst contents
• offers similar postoperative convalescence and mean time to return to work as laparoscopic or laparoscopic-assisted management of large ovarian cysts

General Ob/Gyns—not gynecologic oncologists—perform most surgeries on patients with adnexal masses, since ovarian cancer is relatively uncommon in the absence of preoperative risk factors for malignancy. Our approach offers an appealing option to Ob/Gyns reluctant to abandon routine traditional laparotomy for such ovarian cysts.

Selecting the right patient

Adequate preoperative assessment diminishes the risk of unexpected malignancy in a patient undergoing surgery for an ovarian mass to less than 1%.⁴ At this time, the combination of menopausal status, cancer antigen (CA) 125 level, physical examination, and ultrasound is the best strategy for evaluating the patient with an ovarian cyst.¹¹

Signs of malignancy. Ultrasound features that suggest malignancy include irregular borders, thick septa, solid areas, internal and external excrescences, matted bowel, and ascites. Benign cysts, on the other hand, are usually unilateral and have regular borders, thin septa, no solid areas, and no internal excrescences.⁴ The measurement of blood flow within the mass by color Doppler may improve the accuracy of ultrasound in differentiating benign from malignant cysts.^4,12

On physical examination, an adnexal mass that is fixed, irregular, or solid also suggests a neoplasm. An elevated CA 125 combined with a complex adnexal mass is likely to be associated with malignancy. The test is even more specific in postmenopausal women with adnexal masses.^4,12

However, plasma levels of CA 125 also can be elevated in several benign gynecologic conditions such as endometriosis, simple ovarian cysts, pelvic inflammatory disease, ovarian torsion, fibroids, and in physiologic conditions such as menstruation and pregnancy.¹³

Anticipate the need to convert to laparotomy. Every patient’s surgical consent should include a possible conversion to laparotomy. To avoid incomplete surgical treatment and significant delays in proper therapy, a gynecologic surgeon experienced in the management of ovarian cancer should be readily available, in the event an unexpected malignancy is encountered. Ideally, the staging surgery and definite treatment should be performed at the time of initial minilaparotomy. Comprehensive surgical staging and treatment include thorough exploration of the pelvis and abdomen, omentectomy, pelvic and paraaortic lymph node sampling, multiple peritoneal biopsies and washings, bilateral salpingo-oophorectomy, hysterectomy, and debulking, when indicated.

Prepare for surgery with position, incision, and retraction

Before beginning, it is crucial to correctly position the patient, make the appropriate incision, and insert the right retractor.¹

Position. After administering regional or general anesthesia, place the patient in a modified lithotomy, as for laparoscopic surgery. Tuck the arms alongside the torso and place the legs in boot stirrups. Avoid hip flexion and allow adequate thigh abduction to expose the vagina. Perform a careful pelvic examination to determine the size and mobility of the adnexal mass.

When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure

Place an indwelling transurethral catheter, and pass a sturdy, hinged uterine manipulator such as the Pelosi Uterine Manipulator (Apple Medical Corp, Marlboro, Mass) transcervically into the uterine cavity (FIGURE 1).

 

The cruciate incision. With a conventional scalpel make a small suprapubic transverse incision through the skin and subcutaneous tissue (FIGURE 2). After clearing subcutaneous fat from the midline, incise the rectus fascia and the peritoneum in a vertical direction.

A vertical skin incision can be selected if the preoperative workup suggests a later extension of the original minilaparotomy incision may be required, or if there is a prior vertical incision.

Retraction. Use a soft sleeve-type self-retaining plastic retractor, such as Mobius (Apple Medical Corp) (FIGURE 3 ). When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure of the pelvis (FIGURE 4).

During surgery it may be necessary to adjust the outer ring if the sleeve loosens. Narrow Deaver or Richardson retractors, if required, provide additional retraction. The bowel may be gently packed, if necessary, but typically it is adequately displaced by the large ovarian cyst.

The atraumatic retraction provided by the soft, self-retaining abdominal retractor minimizes the possibility of tissue trauma, nerve damage, bruising, and postoperative pain. At the same time, the continuous 360° retraction force on the incision maximizes surgical exposure, providing a significantly larger working area than conventional retractors. For example, when applied to a 6-cm incision, the self-retaining retractor creates a 28-cm² exposed working area, compared with only 18 cm² provided by a conventional 4-point metal retractor.

The adjustable height of the self-retaining retractor adapts to wounds of varying depth and works on virtually any tissue thickness—a feature that makes the device effective for obese patients. Further, by lining the abdominal incision, the retractor’s plastic sleeve protects the wound’s edges from contamination and potential implantation of malignant cells, making the device ideal for managing ovarian cysts.

FIGURE 1 Hinged uterine manipulator

The manipulator facilitates exposure of the cyst and contralateral adnexa, as well as uterine elevation/rotation.

FIGURE 2 Cruciate incision

Make a 2.5- to 5-cm suprapubic transverse skin incision. Using the Bovie device, incise the subcutaneous fat transversely along the full length of the skin incision down to the level of the anterior rectus fascia. Clear the subcutaneous fat from the midline superiorly and inferiorly to expose 5 to 6 cm of the rectus fascia in the vertical axis. Use blunt digital dissection to assist in mobilizing the subcutaneous fat. Incise the anterior rectus fascia vertically through the full length of the cleared area. Retract the rectus muscles from the midline to expose the underlying transversalis fascia and the peritoneum. Control small bleeding points with the Bovie device. Enter the peritoneum either digitally or with scissors above the bladder dome and extend the peritoneal incision to the full length of the fascial incision.

FIGURE 3 Self-retaining retractor

The retractor consists of a flexible plastic inner ring and a firmer outer ring connected by a soft plastic sleeve.

FIGURE 4360° retraction force

Squeeze the inner ring of the soft, sleeve-type, self-retaining retractor into the peritoneal cavity through the cruciate abdominal incision and allow it to spring open against the parietal peritoneum. Make a digital assessment to ensure that viscera is not trapped between the inner ring and the abdominal wall. Place the entire sleeve on traction by lifting the outer ring. Then roll the outer ring onto the sleeve, collecting excess length, until it sits firmly against the skin. The atraumatic retraction provided by the sleeve-type retractor maximizes exposure of the pelvis. Note how a portion of the large cyst is clearly seen through the atraumatic, circular area of retraction.

Cyst assessment

Visually and digitally inspect the cyst and carefully evaluate the uterus, pelvis, and contralateral adnexa. Determine the extent of adhesions and any unexpected pelvic pathology. When needed, use traditional small retractors or gentle packing to gain additional exposure. If the cystic mass appears suspicious (internal and external excrescences on the cyst, ovaries, or peritoneal surfaces, or ascites), obtain pelvic washings with a suction-irrigation cannula, send the fluid for cytologic examination, and convert the minilaparotomy to a standard exploratory laparotomy. Extensive adhesions to the bowel, broad ligament, or pelvic sidewall and unexpected extensive endometriosis may also require a conversion to standard laparotomy.

Reduce cyst size by decompression

Simple aspiration is inadvisable. To remove a large ovarian cyst using the Pelosi minilaparotomy, reduce the size of the cyst to permit safe and effective mobilization and resection through the small abdominal incision. Simple aspiration of the cyst, with its potential for spilling the contents, is not a wise strategy for several reasons. First, many ovarian cysts contain functional epithelium with a high recurrence rate (8% to 67%). Second, studies show that 10% to 66% of ovarian cyst fluid aspirates initially diagnosed as benign actually are malignant. Further, relying on negative cytology from the aspirate of an ovarian cyst without tissue biopsy may delay appropriate surgery, and the puncture of an unexpected malignant cyst may seed the peritoneal cavity and possibly worsen the patient’s prognosis.^11,12

 

This technique, suitable for all ovarian cysts, makes it possible to aspirate a large cyst without leakage.

Whether spillage of cancer cells actually worsens the prognosis of a patient with a neoplastic cyst remains controversial because of conflicting study results. Nonetheless, the possibility of intraperitoneal dissemination of neoplastic cells from a ruptured cyst cannot be considered innocuous, and the potential negative effect on a patient’s prognosis should not be ignored.¹⁴ Make every attempt, therefore, to avoid rupturing the cyst and spilling the fluid into the peritoneal cavity.

A shared flaw plagues aspiration devices. Different devices are available for intraoperative cyst aspiration during laparoscopic, transvaginal, or laparotomy approaches.^3,4,6 In addition to long needles, drainage trocars, suction cannulas, and suprapubic bladder catheters, special aspiration instruments have been developed.¹⁵ They include a metal vacuum system with an aspirator trocar that seals the surface of the cyst, and a catheter system that pinches the punctured cyst wall between double balloons to prevent spillage.¹⁶ In addition, several commercial bags are available to prevent intraperitoneal spillage during removal of ovarian cysts.

All these devices have a universal flaw, however: After a thin-walled cyst initially is punctured, none of these products can prevent the spontaneous dehiscence of the cyst and the resulting spillage of its contents into the abdominal cavity. Vacuum systems work well for large cysts with smooth, round surfaces, but in those with irregular surfaces, both application and maintaining the seal are difficult.¹⁷ Fortunately, our technique makes it possible to aspirate a large ovarian cyst without leakage, and the method is suitable for all ovarian cysts regardless of surface type or wall thickness.

Glue the dressing to the cyst to capture leakage. Using a gauze pad, carefully dry the area of the ovarian cyst that is visible through the self-retaining retractor. Then generously spread sterile surgical glue such as Dermabond (Ethicon, Somerville, NJ) on the cyst wall surface (FIGURE 5A). Dermabond is the commercial name for 2-octyl cyanoacrylate, a sterile skin adhesive used as an alternative to stitches to close the edges of small wounds. It is similar to commercial adhesives such as Super Glue and Krazy Glue.

Remove the paper cover of a transparent plastic surgical dressing and place the adhesive side directly onto the glued cyst surface until you are sure the adhesive is completely fixed (FIGURES 5B AND 5C). The 35 cm x 35 cm Steri-Drape or the transparent Tegaderm (both from 3M Health Care, St. Paul, Minn) dressing is effective. A standard nonadhesive plastic dressing or a sterile plastic bag also can be used, as long as the free edges extend beyond the outer rim of the self-retaining retractor.

With a needle aspirator, pierce the cyst through the glued plastic dressing and carefully aspirate the fluid (FIGURES 6A AND 6B). Any leakage is trapped inside the plastic dressing rather than the abdominal cavity. Continue the aspiration until the collapsed cyst and ovary can be gradually delivered through the abdominal incision (FIGURE 6C). Note that the selfretaining retractor also protects the abdominal incision from potential contamination and implantation of neoplastic cells.

Once the cyst and ovary are extracted, you can readily perform an extracorporeal cystectomy, after which the repaired ovary is returned to the abdominal cavity. Be careful to avoid letting any fluid flow back into the peritoneal cavity during cyst removal. When indicated, an extracorporeal adnexectomy can readily be performed.

The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst.

Dermoid cysts require extra care. Preventing intraperitoneal spillage is especially important when removing a large dermoid cyst, to avoid the possibility of chemical peritonitis, dense adhesions, and fistulas. The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst. Quite frequently, a large-diameter suction cannula is required to remove the waxy contents. If the dermoid cyst is very large, aspiration alone will not empty it entirely. Complete emptying is not necessary, however. The primary goal of aspiration is to reduce the size and tension of the cyst to permit delivery through the abdominal incision.

Closing the cruciate incision is quicker and requires less exposure than a scaled-down Pfannenstiel’s incision.

Before concluding the procedure, irrigate the peritoneal cavity to remove any remnants of cyst contents that may have spilled—especially important with a dermoid cyst. Perform any additional indicated procedures through the minilaparotomy incision, such as a contralateral ovarian cystectomy, salpingo-oophorectomy, or hysterectomy. After the surgery is completed, remove the self-retaining retractor by hooking a finger through the bottom ring and pulling it gently out of the incision. Closing the cruciate incision is quicker and requires less exposure to complete than a scaled-down Pfannenstiel’s incision. Apply a vertical pressure dressing over the incision to prevent postoperative wound hematoma or seroma formation. Remove the dressing 24 hours later.

 

FIGURE 5A ‘Leak-proofing’ the aspiration

After drying the surface of the cyst, apply sterile surgical glue to the cyst wall (2 to 3 ampules are usually required to cover the cyst surface).

FIGURE 5B ‘Leak-proofing’ the aspiration

Apply the adhesive side of a clear plastic wound dressing directly onto the cyst’s surface, making sure that the dressing is large enough to fully cover the self-retaining retractor. Using either a piece of folded gauze or your fingers, press the plastic dressing against the adhesive-coated cyst for about 3 to 5 minutes until the adhesive is completely fixed.

FIGURE 5C ‘Leak-proofing’ the aspiration

Remove the paper cover of the wound dressing.

FIGURE 6A Aspirating the cyst

Pierce the dressing and carefully aspirate the fluid until the cyst is partially collapsed.

FIGURE 6B Aspirating the cyst

Place atraumatic clamps on the cyst wall to further control drainage. Any leakage is trapped inside the plastic dressing rather than draining into the abdominal cavity.

FIGURE 6C Aspirating the cyst

Continue the aspiration until the collapsed cyst and ovary can be delivered gradually through the abdominal incision, after detaching the adhesive plastic dressing from the edges of the self-retaining retractor. The portion of the plastic sleeve that is glued to the cyst surface remains attached to the cyst until the cyst is extracted. Following extraction, perform an extracorporeal cystectomy and return the repaired ovary to the abdominal cavity.

Good results

Using our approach, we have treated 38 patients with ovarian cysts of diameters greater than 20 cm thought to be benign by preoperative workup (FIGURE 7). We encountered no malignancies. All surgeries were successfully completed without laparoscopy or conversion to traditional laparotomy and with good cosmesis (FIGURE 8).

Other procedures performed in some of these patients using the same technique included contralateral cystectomy, salpingo-oophorectomy, subtotal and total hysterectomy, adhesiolysis, and appendectomy. We encountered no intraoperative or postoperative complications. Operating times ranged from 18 to 65 minutes. All patients were discharged home within 36 hours and returned to work in a mean of 12 days. Pathology findings of the ovarian cysts included endometrioma, dermoid cyst, serous cystadenoma, and mucinous cystadenoma.

FIGURE 7 Effective for large cysts

Following removal, the collapsed cyst was refilled with 1,300 mL of normal saline to demonstrate its actual size.

FIGURE 8 Good cosmetic results

Cosmetic appearance of the small cruciate abdominal incision 10 days after surgery.Dr. Pelosi II reports that he is a consultant for Apple Medical Corporation. Dr. Pelosi III reports no affiliations or financial arrangements with any companies whose products are mentioned in this article.

Although laparotomy is still considered the standard for ovarian cyst removal, over the past 15 years minimally invasive surgery has gained wider acceptance in cases where preoperative assessment suggests an adnexal mass is benign.

Unfortunately, minimally invasive management of a large ovarian cyst (greater than 10 cm) is particularly challenging for several reasons:

• The cyst can rupture and spill its contents into the peritoneum,
• the cyst’s size limits the surgical field, and
• an unexpected malignancy may be revealed.

An innovative minilaparotomy technique for the removal of benign ovarian cysts offers the advantages of laparoscopic and laparoscopic-assisted procedures while bypassing the major disadvantages: the necessity for specialized and expensive equipment, lengthy operative time, and long learning curves.¹ (The minimally invasive procedures currently available for the treatment of ovarian cysts include laparoscopic cystectomy, laparoscopic-assisted minilaparotomy cystectomy, laparoscopic-assisted vaginal cystectomy, combined percutaneous ultrasound cyst aspiration and laparoscopic cystectomy, transvaginal cystectomy, and the traditional minilaparotomy cystectomy.^2-10)

The procedure is faster, less expensive, carries fewer potential risks than traditional alternatives, and offers these advantages:

• can be performed under regional anesthesia
• relies on standard open techniques
• uses inexpensive instrumentation
• is easy to learn
• can be used for very large cysts
• eliminates the risk of intraperitoneal spillage of cyst contents
• offers similar postoperative convalescence and mean time to return to work as laparoscopic or laparoscopic-assisted management of large ovarian cysts

General Ob/Gyns—not gynecologic oncologists—perform most surgeries on patients with adnexal masses, since ovarian cancer is relatively uncommon in the absence of preoperative risk factors for malignancy. Our approach offers an appealing option to Ob/Gyns reluctant to abandon routine traditional laparotomy for such ovarian cysts.

Selecting the right patient

Adequate preoperative assessment diminishes the risk of unexpected malignancy in a patient undergoing surgery for an ovarian mass to less than 1%.⁴ At this time, the combination of menopausal status, cancer antigen (CA) 125 level, physical examination, and ultrasound is the best strategy for evaluating the patient with an ovarian cyst.¹¹

Signs of malignancy. Ultrasound features that suggest malignancy include irregular borders, thick septa, solid areas, internal and external excrescences, matted bowel, and ascites. Benign cysts, on the other hand, are usually unilateral and have regular borders, thin septa, no solid areas, and no internal excrescences.⁴ The measurement of blood flow within the mass by color Doppler may improve the accuracy of ultrasound in differentiating benign from malignant cysts.^4,12

On physical examination, an adnexal mass that is fixed, irregular, or solid also suggests a neoplasm. An elevated CA 125 combined with a complex adnexal mass is likely to be associated with malignancy. The test is even more specific in postmenopausal women with adnexal masses.^4,12

However, plasma levels of CA 125 also can be elevated in several benign gynecologic conditions such as endometriosis, simple ovarian cysts, pelvic inflammatory disease, ovarian torsion, fibroids, and in physiologic conditions such as menstruation and pregnancy.¹³

Anticipate the need to convert to laparotomy. Every patient’s surgical consent should include a possible conversion to laparotomy. To avoid incomplete surgical treatment and significant delays in proper therapy, a gynecologic surgeon experienced in the management of ovarian cancer should be readily available, in the event an unexpected malignancy is encountered. Ideally, the staging surgery and definite treatment should be performed at the time of initial minilaparotomy. Comprehensive surgical staging and treatment include thorough exploration of the pelvis and abdomen, omentectomy, pelvic and paraaortic lymph node sampling, multiple peritoneal biopsies and washings, bilateral salpingo-oophorectomy, hysterectomy, and debulking, when indicated.

Prepare for surgery with position, incision, and retraction

Before beginning, it is crucial to correctly position the patient, make the appropriate incision, and insert the right retractor.¹

Position. After administering regional or general anesthesia, place the patient in a modified lithotomy, as for laparoscopic surgery. Tuck the arms alongside the torso and place the legs in boot stirrups. Avoid hip flexion and allow adequate thigh abduction to expose the vagina. Perform a careful pelvic examination to determine the size and mobility of the adnexal mass.

When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure

Place an indwelling transurethral catheter, and pass a sturdy, hinged uterine manipulator such as the Pelosi Uterine Manipulator (Apple Medical Corp, Marlboro, Mass) transcervically into the uterine cavity (FIGURE 1).

 

The cruciate incision. With a conventional scalpel make a small suprapubic transverse incision through the skin and subcutaneous tissue (FIGURE 2). After clearing subcutaneous fat from the midline, incise the rectus fascia and the peritoneum in a vertical direction.

A vertical skin incision can be selected if the preoperative workup suggests a later extension of the original minilaparotomy incision may be required, or if there is a prior vertical incision.

Retraction. Use a soft sleeve-type self-retaining plastic retractor, such as Mobius (Apple Medical Corp) (FIGURE 3 ). When properly placed, this retractor creates an atraumatic, circular area of self-retraction, enabling superior exposure of the pelvis (FIGURE 4).

During surgery it may be necessary to adjust the outer ring if the sleeve loosens. Narrow Deaver or Richardson retractors, if required, provide additional retraction. The bowel may be gently packed, if necessary, but typically it is adequately displaced by the large ovarian cyst.

The atraumatic retraction provided by the soft, self-retaining abdominal retractor minimizes the possibility of tissue trauma, nerve damage, bruising, and postoperative pain. At the same time, the continuous 360° retraction force on the incision maximizes surgical exposure, providing a significantly larger working area than conventional retractors. For example, when applied to a 6-cm incision, the self-retaining retractor creates a 28-cm² exposed working area, compared with only 18 cm² provided by a conventional 4-point metal retractor.

The adjustable height of the self-retaining retractor adapts to wounds of varying depth and works on virtually any tissue thickness—a feature that makes the device effective for obese patients. Further, by lining the abdominal incision, the retractor’s plastic sleeve protects the wound’s edges from contamination and potential implantation of malignant cells, making the device ideal for managing ovarian cysts.

FIGURE 1 Hinged uterine manipulator

The manipulator facilitates exposure of the cyst and contralateral adnexa, as well as uterine elevation/rotation.

FIGURE 2 Cruciate incision

Make a 2.5- to 5-cm suprapubic transverse skin incision. Using the Bovie device, incise the subcutaneous fat transversely along the full length of the skin incision down to the level of the anterior rectus fascia. Clear the subcutaneous fat from the midline superiorly and inferiorly to expose 5 to 6 cm of the rectus fascia in the vertical axis. Use blunt digital dissection to assist in mobilizing the subcutaneous fat. Incise the anterior rectus fascia vertically through the full length of the cleared area. Retract the rectus muscles from the midline to expose the underlying transversalis fascia and the peritoneum. Control small bleeding points with the Bovie device. Enter the peritoneum either digitally or with scissors above the bladder dome and extend the peritoneal incision to the full length of the fascial incision.

FIGURE 3 Self-retaining retractor

The retractor consists of a flexible plastic inner ring and a firmer outer ring connected by a soft plastic sleeve.

FIGURE 4360° retraction force

Squeeze the inner ring of the soft, sleeve-type, self-retaining retractor into the peritoneal cavity through the cruciate abdominal incision and allow it to spring open against the parietal peritoneum. Make a digital assessment to ensure that viscera is not trapped between the inner ring and the abdominal wall. Place the entire sleeve on traction by lifting the outer ring. Then roll the outer ring onto the sleeve, collecting excess length, until it sits firmly against the skin. The atraumatic retraction provided by the sleeve-type retractor maximizes exposure of the pelvis. Note how a portion of the large cyst is clearly seen through the atraumatic, circular area of retraction.

Cyst assessment

Visually and digitally inspect the cyst and carefully evaluate the uterus, pelvis, and contralateral adnexa. Determine the extent of adhesions and any unexpected pelvic pathology. When needed, use traditional small retractors or gentle packing to gain additional exposure. If the cystic mass appears suspicious (internal and external excrescences on the cyst, ovaries, or peritoneal surfaces, or ascites), obtain pelvic washings with a suction-irrigation cannula, send the fluid for cytologic examination, and convert the minilaparotomy to a standard exploratory laparotomy. Extensive adhesions to the bowel, broad ligament, or pelvic sidewall and unexpected extensive endometriosis may also require a conversion to standard laparotomy.

Reduce cyst size by decompression

Simple aspiration is inadvisable. To remove a large ovarian cyst using the Pelosi minilaparotomy, reduce the size of the cyst to permit safe and effective mobilization and resection through the small abdominal incision. Simple aspiration of the cyst, with its potential for spilling the contents, is not a wise strategy for several reasons. First, many ovarian cysts contain functional epithelium with a high recurrence rate (8% to 67%). Second, studies show that 10% to 66% of ovarian cyst fluid aspirates initially diagnosed as benign actually are malignant. Further, relying on negative cytology from the aspirate of an ovarian cyst without tissue biopsy may delay appropriate surgery, and the puncture of an unexpected malignant cyst may seed the peritoneal cavity and possibly worsen the patient’s prognosis.^11,12

 

This technique, suitable for all ovarian cysts, makes it possible to aspirate a large cyst without leakage.

Whether spillage of cancer cells actually worsens the prognosis of a patient with a neoplastic cyst remains controversial because of conflicting study results. Nonetheless, the possibility of intraperitoneal dissemination of neoplastic cells from a ruptured cyst cannot be considered innocuous, and the potential negative effect on a patient’s prognosis should not be ignored.¹⁴ Make every attempt, therefore, to avoid rupturing the cyst and spilling the fluid into the peritoneal cavity.

A shared flaw plagues aspiration devices. Different devices are available for intraoperative cyst aspiration during laparoscopic, transvaginal, or laparotomy approaches.^3,4,6 In addition to long needles, drainage trocars, suction cannulas, and suprapubic bladder catheters, special aspiration instruments have been developed.¹⁵ They include a metal vacuum system with an aspirator trocar that seals the surface of the cyst, and a catheter system that pinches the punctured cyst wall between double balloons to prevent spillage.¹⁶ In addition, several commercial bags are available to prevent intraperitoneal spillage during removal of ovarian cysts.

All these devices have a universal flaw, however: After a thin-walled cyst initially is punctured, none of these products can prevent the spontaneous dehiscence of the cyst and the resulting spillage of its contents into the abdominal cavity. Vacuum systems work well for large cysts with smooth, round surfaces, but in those with irregular surfaces, both application and maintaining the seal are difficult.¹⁷ Fortunately, our technique makes it possible to aspirate a large ovarian cyst without leakage, and the method is suitable for all ovarian cysts regardless of surface type or wall thickness.

Glue the dressing to the cyst to capture leakage. Using a gauze pad, carefully dry the area of the ovarian cyst that is visible through the self-retaining retractor. Then generously spread sterile surgical glue such as Dermabond (Ethicon, Somerville, NJ) on the cyst wall surface (FIGURE 5A). Dermabond is the commercial name for 2-octyl cyanoacrylate, a sterile skin adhesive used as an alternative to stitches to close the edges of small wounds. It is similar to commercial adhesives such as Super Glue and Krazy Glue.

Remove the paper cover of a transparent plastic surgical dressing and place the adhesive side directly onto the glued cyst surface until you are sure the adhesive is completely fixed (FIGURES 5B AND 5C). The 35 cm x 35 cm Steri-Drape or the transparent Tegaderm (both from 3M Health Care, St. Paul, Minn) dressing is effective. A standard nonadhesive plastic dressing or a sterile plastic bag also can be used, as long as the free edges extend beyond the outer rim of the self-retaining retractor.

With a needle aspirator, pierce the cyst through the glued plastic dressing and carefully aspirate the fluid (FIGURES 6A AND 6B). Any leakage is trapped inside the plastic dressing rather than the abdominal cavity. Continue the aspiration until the collapsed cyst and ovary can be gradually delivered through the abdominal incision (FIGURE 6C). Note that the selfretaining retractor also protects the abdominal incision from potential contamination and implantation of neoplastic cells.

Once the cyst and ovary are extracted, you can readily perform an extracorporeal cystectomy, after which the repaired ovary is returned to the abdominal cavity. Be careful to avoid letting any fluid flow back into the peritoneal cavity during cyst removal. When indicated, an extracorporeal adnexectomy can readily be performed.

The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst.

Dermoid cysts require extra care. Preventing intraperitoneal spillage is especially important when removing a large dermoid cyst, to avoid the possibility of chemical peritonitis, dense adhesions, and fistulas. The presence of oily material or hairs in the aspirated fluid and on the needle tip readily identifies a dermoid cyst. Quite frequently, a large-diameter suction cannula is required to remove the waxy contents. If the dermoid cyst is very large, aspiration alone will not empty it entirely. Complete emptying is not necessary, however. The primary goal of aspiration is to reduce the size and tension of the cyst to permit delivery through the abdominal incision.

Closing the cruciate incision is quicker and requires less exposure than a scaled-down Pfannenstiel’s incision.

Before concluding the procedure, irrigate the peritoneal cavity to remove any remnants of cyst contents that may have spilled—especially important with a dermoid cyst. Perform any additional indicated procedures through the minilaparotomy incision, such as a contralateral ovarian cystectomy, salpingo-oophorectomy, or hysterectomy. After the surgery is completed, remove the self-retaining retractor by hooking a finger through the bottom ring and pulling it gently out of the incision. Closing the cruciate incision is quicker and requires less exposure to complete than a scaled-down Pfannenstiel’s incision. Apply a vertical pressure dressing over the incision to prevent postoperative wound hematoma or seroma formation. Remove the dressing 24 hours later.

 

FIGURE 5A ‘Leak-proofing’ the aspiration

After drying the surface of the cyst, apply sterile surgical glue to the cyst wall (2 to 3 ampules are usually required to cover the cyst surface).

FIGURE 5B ‘Leak-proofing’ the aspiration

Apply the adhesive side of a clear plastic wound dressing directly onto the cyst’s surface, making sure that the dressing is large enough to fully cover the self-retaining retractor. Using either a piece of folded gauze or your fingers, press the plastic dressing against the adhesive-coated cyst for about 3 to 5 minutes until the adhesive is completely fixed.

FIGURE 5C ‘Leak-proofing’ the aspiration

Remove the paper cover of the wound dressing.

FIGURE 6A Aspirating the cyst

Pierce the dressing and carefully aspirate the fluid until the cyst is partially collapsed.

FIGURE 6B Aspirating the cyst

Place atraumatic clamps on the cyst wall to further control drainage. Any leakage is trapped inside the plastic dressing rather than draining into the abdominal cavity.

FIGURE 6C Aspirating the cyst

Continue the aspiration until the collapsed cyst and ovary can be delivered gradually through the abdominal incision, after detaching the adhesive plastic dressing from the edges of the self-retaining retractor. The portion of the plastic sleeve that is glued to the cyst surface remains attached to the cyst until the cyst is extracted. Following extraction, perform an extracorporeal cystectomy and return the repaired ovary to the abdominal cavity.

Good results

Using our approach, we have treated 38 patients with ovarian cysts of diameters greater than 20 cm thought to be benign by preoperative workup (FIGURE 7). We encountered no malignancies. All surgeries were successfully completed without laparoscopy or conversion to traditional laparotomy and with good cosmesis (FIGURE 8).

Other procedures performed in some of these patients using the same technique included contralateral cystectomy, salpingo-oophorectomy, subtotal and total hysterectomy, adhesiolysis, and appendectomy. We encountered no intraoperative or postoperative complications. Operating times ranged from 18 to 65 minutes. All patients were discharged home within 36 hours and returned to work in a mean of 12 days. Pathology findings of the ovarian cysts included endometrioma, dermoid cyst, serous cystadenoma, and mucinous cystadenoma.

FIGURE 7 Effective for large cysts

Following removal, the collapsed cyst was refilled with 1,300 mL of normal saline to demonstrate its actual size.

FIGURE 8 Good cosmetic results

Cosmetic appearance of the small cruciate abdominal incision 10 days after surgery.Dr. Pelosi II reports that he is a consultant for Apple Medical Corporation. Dr. Pelosi III reports no affiliations or financial arrangements with any companies whose products are mentioned in this article.

Why did it take us 2 centuries to learn how to do it right? Posterior pelvic repair for the correction of rectocele combined with restoration of the perineal body has been on gynecologic OR schedules since time immemorial, so one would think we knew what we were doing. Yet only in the past 10 years has there been progressive general recognition of the true nature of the anatomic lesions responsible for rectocele formation, thereby finally pointing us in the right direction.

Although it was always assumed, correctly, that the major factor initiating rectoceles is tissue trauma sustained in vaginal birth, traditionally we were taught, and we believed, that the end result was attenuation and deterioration of the rectovaginal septum (RVS) connective tissue. Similar tissue changes also initiated at vaginal delivery were ascribed to the proximal and distal connective tissue attachments of the RVS, respectively; the cardinal-uterosacral ligament complex; and the perineal body—but tears were never entertained in our perceptions. We know, as emphasized by Nichols in the 1960s and 1970s,^1-3 that all of these elements, together with the bilateral RVS connection to levator fascia plus the levator plate of the pelvic floor, constitute the normal vaginal axis when all elements are intact.

To be concise, the disrupted vaginal axis, including its 2 major anatomic abnormalities, rectocele and perineal body defects, was thought to be the direct result of disintegration over time after the initiating shock of vaginal delivery. Specific tissue tears were, until now, an unrealized concept. As a result, we routinely resorted to makeshift methods in attempts to restore posterior pelvic configuration. Today, thanks to A. Cullen Richardson, we are much better informed.^4,5

Sutured reunion of RVS tears key to rectocele eradication

This article centers on today’s concepts of the defects isolated within the RVS, a critical part of the vaginal axis. No longer is there any question about the key role of sutured reunion of RVS tears in eradicating rectocele. We must acknowledge, however, that such anatomic restoration alone is insufficient without simultaneous fixed suspension of the posterior vaginal vault above and perineal body reconstitution below, as Nichols emphasized.³

The overall goal in corrective posterior pelvic surgery must always be restoration of the normal vaginal axis in order to preserve the pelvic valve mechanism, as described by Porges.⁶

The primary reason for connective tissue defects is tears, not attenuation. Richardson dramatically and conclusively demonstrated this fact in the early 1990s, after prolonged and intensive investigation of the anatomy.^4,5,7 This astonishing disclosure was nowhere more obvious than in the defective RVS, where discrete separations in otherwise intact tissue could be uncovered and reattached by simple standard stitching technique. We had simply never looked for these torn segments (which were never obvious), so we never found them. Once we were convinced of Richardson’s concepts, the torn edges became easily identifiable in practically every case of rectocele. Suddenly, we were able to restore the natural anatomic integrity of the RVS by reuniting these torn edges. With absolute confidence, we eliminated the rectocele.^8-16

The critical digital technique

The key to this logical operative technique was the rectal examining finger. Using this digital strategy, not only could we definitively outline the rectocele on outpatient examination, but we could delineate the exact extent of the rectocele intraoperatively. This is accomplished by performing the first rectal exam just before vaginal mucosal dissection, and then, most accurately, after full dissection and exposure of the entire operative field, thus clearly exposing the torn edges in the RVS. A final anorectal exam is performed to check the restored anatomic configuration of the RVS before vaginal mucosal reunification.⁸

Transverse tears

Transverse tears are the most common: either detachment from below, adjacent to the perineal body, or separation at the top in juxtaposition to the fibrous uterosacral extensions in the area of the cul-de-sac. FIGURES 1 AND 2 convey the “before and after” aspects, showing Allis clamps drawing the retracted edge to the site of the tear followed by the revelation of the actual repair accomplished by interrupted O-caliber sutures, either delayed absorbable or nonabsorbable. A rectally inserted index finger initially demonstrates that, except for scraps of areolar tissue, the anorectal wall is the only layer that prevents the examining finger from falling into the operative field as it produces rectocele simulation with forward pressure. When the clamps draw the long-torn RVS segments into normal anatomic positions, the anorectal finger cannot advance at all, despite strong effort, because of the natural firm barrier effect of the anatomically restored RVS.⁸

 

FIGURE 1 Lower transverse tear

FIGURE 2 Upper transverse tear

U-shaped tears

Another not uncommon type of defect can be U-shaped, either at the bottom, as depicted in FIGURE 3, or at the top of the posterior pelvic compartment. Similarly linear (longitudinal) tears near one side or the other directly adjacent to the pelvic sidewalls, rarely in the midline, are not unusual and can be repaired in the same manner.

FIGURE 3 U-shaped tear

Hockey-stick tears

Occasionally, a hockey-stick lesion, combining a longitudinal tear and a transverse tear in continuity, as shown in FIGURE 4, is discovered.

FIGURE 4 Hockey-stick tear

Double defect

A more uncommon type of combined or double defect, in which the Denonvilliers fascia (RVS) has been torn both adjacent to the vault and in the perineal area but retains strong attachments bilaterally, is shown in FIGURE 5.

FIGURE 5 Double defect

A contented vaginal environment

The major breakthrough is the concept that tears in the Denonvilliers fascia, not attenuation, are the cause of rectoceles. These tears—transverse, longitudinal, U-shaped, multidirectional, even stellate—can be identified almost always without difficulty with the aid of a rectal examining finger and edge-grasping clamps at surgery. The repair itself is not only anatomically logical but also much easier and more confidence-inspiring than the traditional, now archaic, method of fishing around for scraps of levator fascia and muscle to approximate, under tension, in the midline.

This defect-directed repair, born of common sense (ie, comprehension and application of normal anatomy), is structurally nonconstrictive and functionally nonrestrictive—truly a contented vaginal environment.

All sketches from TeLinde’s Operative Gynecology, 9th edition, with permission from Lippincott, Williams, and Wilkins, Publishers.

Dr. Grody reports no financial relationship with any companies whose products are mentioned in this article.

Why did it take us 2 centuries to learn how to do it right? Posterior pelvic repair for the correction of rectocele combined with restoration of the perineal body has been on gynecologic OR schedules since time immemorial, so one would think we knew what we were doing. Yet only in the past 10 years has there been progressive general recognition of the true nature of the anatomic lesions responsible for rectocele formation, thereby finally pointing us in the right direction.

Although it was always assumed, correctly, that the major factor initiating rectoceles is tissue trauma sustained in vaginal birth, traditionally we were taught, and we believed, that the end result was attenuation and deterioration of the rectovaginal septum (RVS) connective tissue. Similar tissue changes also initiated at vaginal delivery were ascribed to the proximal and distal connective tissue attachments of the RVS, respectively; the cardinal-uterosacral ligament complex; and the perineal body—but tears were never entertained in our perceptions. We know, as emphasized by Nichols in the 1960s and 1970s,^1-3 that all of these elements, together with the bilateral RVS connection to levator fascia plus the levator plate of the pelvic floor, constitute the normal vaginal axis when all elements are intact.

To be concise, the disrupted vaginal axis, including its 2 major anatomic abnormalities, rectocele and perineal body defects, was thought to be the direct result of disintegration over time after the initiating shock of vaginal delivery. Specific tissue tears were, until now, an unrealized concept. As a result, we routinely resorted to makeshift methods in attempts to restore posterior pelvic configuration. Today, thanks to A. Cullen Richardson, we are much better informed.^4,5

Sutured reunion of RVS tears key to rectocele eradication

This article centers on today’s concepts of the defects isolated within the RVS, a critical part of the vaginal axis. No longer is there any question about the key role of sutured reunion of RVS tears in eradicating rectocele. We must acknowledge, however, that such anatomic restoration alone is insufficient without simultaneous fixed suspension of the posterior vaginal vault above and perineal body reconstitution below, as Nichols emphasized.³

The overall goal in corrective posterior pelvic surgery must always be restoration of the normal vaginal axis in order to preserve the pelvic valve mechanism, as described by Porges.⁶

The primary reason for connective tissue defects is tears, not attenuation. Richardson dramatically and conclusively demonstrated this fact in the early 1990s, after prolonged and intensive investigation of the anatomy.^4,5,7 This astonishing disclosure was nowhere more obvious than in the defective RVS, where discrete separations in otherwise intact tissue could be uncovered and reattached by simple standard stitching technique. We had simply never looked for these torn segments (which were never obvious), so we never found them. Once we were convinced of Richardson’s concepts, the torn edges became easily identifiable in practically every case of rectocele. Suddenly, we were able to restore the natural anatomic integrity of the RVS by reuniting these torn edges. With absolute confidence, we eliminated the rectocele.^8-16

The critical digital technique

The key to this logical operative technique was the rectal examining finger. Using this digital strategy, not only could we definitively outline the rectocele on outpatient examination, but we could delineate the exact extent of the rectocele intraoperatively. This is accomplished by performing the first rectal exam just before vaginal mucosal dissection, and then, most accurately, after full dissection and exposure of the entire operative field, thus clearly exposing the torn edges in the RVS. A final anorectal exam is performed to check the restored anatomic configuration of the RVS before vaginal mucosal reunification.⁸

Transverse tears

Transverse tears are the most common: either detachment from below, adjacent to the perineal body, or separation at the top in juxtaposition to the fibrous uterosacral extensions in the area of the cul-de-sac. FIGURES 1 AND 2 convey the “before and after” aspects, showing Allis clamps drawing the retracted edge to the site of the tear followed by the revelation of the actual repair accomplished by interrupted O-caliber sutures, either delayed absorbable or nonabsorbable. A rectally inserted index finger initially demonstrates that, except for scraps of areolar tissue, the anorectal wall is the only layer that prevents the examining finger from falling into the operative field as it produces rectocele simulation with forward pressure. When the clamps draw the long-torn RVS segments into normal anatomic positions, the anorectal finger cannot advance at all, despite strong effort, because of the natural firm barrier effect of the anatomically restored RVS.⁸

 

FIGURE 1 Lower transverse tear

FIGURE 2 Upper transverse tear

U-shaped tears

Another not uncommon type of defect can be U-shaped, either at the bottom, as depicted in FIGURE 3, or at the top of the posterior pelvic compartment. Similarly linear (longitudinal) tears near one side or the other directly adjacent to the pelvic sidewalls, rarely in the midline, are not unusual and can be repaired in the same manner.

FIGURE 3 U-shaped tear

Hockey-stick tears

Occasionally, a hockey-stick lesion, combining a longitudinal tear and a transverse tear in continuity, as shown in FIGURE 4, is discovered.

FIGURE 4 Hockey-stick tear

Double defect

A more uncommon type of combined or double defect, in which the Denonvilliers fascia (RVS) has been torn both adjacent to the vault and in the perineal area but retains strong attachments bilaterally, is shown in FIGURE 5.

FIGURE 5 Double defect

A contented vaginal environment

The major breakthrough is the concept that tears in the Denonvilliers fascia, not attenuation, are the cause of rectoceles. These tears—transverse, longitudinal, U-shaped, multidirectional, even stellate—can be identified almost always without difficulty with the aid of a rectal examining finger and edge-grasping clamps at surgery. The repair itself is not only anatomically logical but also much easier and more confidence-inspiring than the traditional, now archaic, method of fishing around for scraps of levator fascia and muscle to approximate, under tension, in the midline.

This defect-directed repair, born of common sense (ie, comprehension and application of normal anatomy), is structurally nonconstrictive and functionally nonrestrictive—truly a contented vaginal environment.

All sketches from TeLinde’s Operative Gynecology, 9th edition, with permission from Lippincott, Williams, and Wilkins, Publishers.

Dr. Grody reports no financial relationship with any companies whose products are mentioned in this article.

Why did it take us 2 centuries to learn how to do it right? Posterior pelvic repair for the correction of rectocele combined with restoration of the perineal body has been on gynecologic OR schedules since time immemorial, so one would think we knew what we were doing. Yet only in the past 10 years has there been progressive general recognition of the true nature of the anatomic lesions responsible for rectocele formation, thereby finally pointing us in the right direction.

Although it was always assumed, correctly, that the major factor initiating rectoceles is tissue trauma sustained in vaginal birth, traditionally we were taught, and we believed, that the end result was attenuation and deterioration of the rectovaginal septum (RVS) connective tissue. Similar tissue changes also initiated at vaginal delivery were ascribed to the proximal and distal connective tissue attachments of the RVS, respectively; the cardinal-uterosacral ligament complex; and the perineal body—but tears were never entertained in our perceptions. We know, as emphasized by Nichols in the 1960s and 1970s,^1-3 that all of these elements, together with the bilateral RVS connection to levator fascia plus the levator plate of the pelvic floor, constitute the normal vaginal axis when all elements are intact.

To be concise, the disrupted vaginal axis, including its 2 major anatomic abnormalities, rectocele and perineal body defects, was thought to be the direct result of disintegration over time after the initiating shock of vaginal delivery. Specific tissue tears were, until now, an unrealized concept. As a result, we routinely resorted to makeshift methods in attempts to restore posterior pelvic configuration. Today, thanks to A. Cullen Richardson, we are much better informed.^4,5

Sutured reunion of RVS tears key to rectocele eradication

This article centers on today’s concepts of the defects isolated within the RVS, a critical part of the vaginal axis. No longer is there any question about the key role of sutured reunion of RVS tears in eradicating rectocele. We must acknowledge, however, that such anatomic restoration alone is insufficient without simultaneous fixed suspension of the posterior vaginal vault above and perineal body reconstitution below, as Nichols emphasized.³

The overall goal in corrective posterior pelvic surgery must always be restoration of the normal vaginal axis in order to preserve the pelvic valve mechanism, as described by Porges.⁶

The primary reason for connective tissue defects is tears, not attenuation. Richardson dramatically and conclusively demonstrated this fact in the early 1990s, after prolonged and intensive investigation of the anatomy.^4,5,7 This astonishing disclosure was nowhere more obvious than in the defective RVS, where discrete separations in otherwise intact tissue could be uncovered and reattached by simple standard stitching technique. We had simply never looked for these torn segments (which were never obvious), so we never found them. Once we were convinced of Richardson’s concepts, the torn edges became easily identifiable in practically every case of rectocele. Suddenly, we were able to restore the natural anatomic integrity of the RVS by reuniting these torn edges. With absolute confidence, we eliminated the rectocele.^8-16

The critical digital technique

The key to this logical operative technique was the rectal examining finger. Using this digital strategy, not only could we definitively outline the rectocele on outpatient examination, but we could delineate the exact extent of the rectocele intraoperatively. This is accomplished by performing the first rectal exam just before vaginal mucosal dissection, and then, most accurately, after full dissection and exposure of the entire operative field, thus clearly exposing the torn edges in the RVS. A final anorectal exam is performed to check the restored anatomic configuration of the RVS before vaginal mucosal reunification.⁸

Transverse tears

Transverse tears are the most common: either detachment from below, adjacent to the perineal body, or separation at the top in juxtaposition to the fibrous uterosacral extensions in the area of the cul-de-sac. FIGURES 1 AND 2 convey the “before and after” aspects, showing Allis clamps drawing the retracted edge to the site of the tear followed by the revelation of the actual repair accomplished by interrupted O-caliber sutures, either delayed absorbable or nonabsorbable. A rectally inserted index finger initially demonstrates that, except for scraps of areolar tissue, the anorectal wall is the only layer that prevents the examining finger from falling into the operative field as it produces rectocele simulation with forward pressure. When the clamps draw the long-torn RVS segments into normal anatomic positions, the anorectal finger cannot advance at all, despite strong effort, because of the natural firm barrier effect of the anatomically restored RVS.⁸

 

FIGURE 1 Lower transverse tear

FIGURE 2 Upper transverse tear

U-shaped tears

Another not uncommon type of defect can be U-shaped, either at the bottom, as depicted in FIGURE 3, or at the top of the posterior pelvic compartment. Similarly linear (longitudinal) tears near one side or the other directly adjacent to the pelvic sidewalls, rarely in the midline, are not unusual and can be repaired in the same manner.

FIGURE 3 U-shaped tear

Hockey-stick tears

Occasionally, a hockey-stick lesion, combining a longitudinal tear and a transverse tear in continuity, as shown in FIGURE 4, is discovered.

FIGURE 4 Hockey-stick tear

Double defect

A more uncommon type of combined or double defect, in which the Denonvilliers fascia (RVS) has been torn both adjacent to the vault and in the perineal area but retains strong attachments bilaterally, is shown in FIGURE 5.

FIGURE 5 Double defect

A contented vaginal environment

The major breakthrough is the concept that tears in the Denonvilliers fascia, not attenuation, are the cause of rectoceles. These tears—transverse, longitudinal, U-shaped, multidirectional, even stellate—can be identified almost always without difficulty with the aid of a rectal examining finger and edge-grasping clamps at surgery. The repair itself is not only anatomically logical but also much easier and more confidence-inspiring than the traditional, now archaic, method of fishing around for scraps of levator fascia and muscle to approximate, under tension, in the midline.

This defect-directed repair, born of common sense (ie, comprehension and application of normal anatomy), is structurally nonconstrictive and functionally nonrestrictive—truly a contented vaginal environment.

All sketches from TeLinde’s Operative Gynecology, 9th edition, with permission from Lippincott, Williams, and Wilkins, Publishers.

Dr. Grody reports no financial relationship with any companies whose products are mentioned in this article.

The state of the art can be simply put: Hysteroscopy is underutilized. Most studies consistently demonstrate the safety and efficacy of operative hysteroscopy, as well as high patient satisfaction.

It is a valuable tool in the evaluation and treatment of infertility, recurrent pregnancy loss, and abnormal and postmenopausal uterine bleeding, and is useful when saline infusion sonography findings are equivocal.

Further, if a global ablation device fails, the surgeon can convert to hysteroscopic ablation rather than abandon the procedure altogether. This is not as unusual as it might appear: In US Food and Drug Administration trials, there was a staggering 10% to 22% malfunction of global ablation technology.¹

Safe, easily learned

Although gynecologists are beginning to embrace this modality, many physicians avoid it because of inadequate training or exaggerated fears of complications. In reality, hysteroscopy is one of the safest and most easily acquired surgical skills in gynecology. For example, in a prospective evaluation of 13,600 diagnostic and operative hysteroscopic procedures performed at 63 hospitals in the Netherlands—which involved both established surgeons and residents—Jansen et al² found an astonishingly low complication rate of 0.28%, with no deaths.

Proper selection and treatment of patients and prompt intervention minimize complications as well as legal risks. Surgical misadventures and lawsuits occur with delayed intervention, failure to recognize pathology or risky conditions, and inadequate preventive maneuvers.

Overall, emphasis on safety is vital to success, and thorough awareness of potential complications is just as important.

Three types of complications

Complications fall into 3 categories (TABLE):

• Procedure-related
  
• Media-related
  
• Postoperative
  

PROCEDURE-RELATED COMPLICATIONS

Complication rates

In a retrospective investigation, Propst et al³ determined the rate of complications associated with specific hysteroscopic procedures. Demographic data and medical histories were collected for 925 women who had operative hysteroscopy in 1995 and 1996. The overall complication rate was 2.7%. Myomectomy and resection of uterine septa carried the greatest odds of complications; polypectomy and endometrial ablation had the lowest. Preoperative treatment with a gonadotropinreleasing hormone (GnRH) agonist increased the odds of complications by a factor of 4 to 7. Women under age 50 were more likely to experience complications than those over 50.

In the study by Jansen et al,² 38 complica-tions occurred in the 13,600 procedures. The greatest risk of complications occurred with adhesiolysis (4.48%), followed by endometrial resection (0.81%), myomectomy (0.75%), and polypectomy (0.38%).

Cervical entry requires special attention

Almost half of the complications in the Jansen study were related to cervical entry, so caution and, perhaps, preoperative cervical ripening are advised. Many premenopausal subjects were given GnRH analogues, which may render the cervix more resistant to dilation. Complications associated with a stenotic cervix include a cervical tear, creation of a false cervical passage, and uterine perforation.

Cervical ripening may help prevent uterine perforation. The most common complication, occurring in 14.2 cases per thousand, is uterine perforation.² The risk of this is highest in postpartum procedures, followed by procedures in postmenopausal, then perimenopausal, women. Patients with endometrial cancer also have a higher rate of perforation.

Risk factors for uterine perforation include:

• nulliparity
  
• menopause
  
• use of GnRH agonists
  
• prior cone biopsy
  
• markedly retroverted uterus
  
• undue force
  

Modern operative hysteroscopes often require dilation of the cervix to a number 8-10 Hegar dilator. Navigation of the internal os is critical before operative instruments can be inserted and the surgical procedure performed. In the past, use of preoperative laminaria was recommended to soften the cervix, except in women with marked cervical stenosis and iodine allergy. Preoperative cervical softening still should be considered in high-risk patients.

Vaginal or oral misoprostol for cervical ripening prior to operative hysteroscopy was evaluated in a randomized trial.⁴ Researchers found a reduced need for cervical dilation, a minimum of cervical complications, and reduced operative time in study patients compared with controls.

When 400 μg oral misoprostol is given 12 and 24 hours before surgery, it also softens the cervix and eases dilation.⁵ Although misoprostol has several bothersome side effects (such as lower abdominal pain and slight vaginal bleeding), few if any prevent its use.

Signs of perforation. Patients who sustain uterine perforation with subsequent intraperitoneal bleeding often complain of pain in the abdomen and shoulder, and experience hemodynamic instability. A quick sonographic survey of the abdomen will demonstrate free intraperitoneal fluid. (It is rare for much intraperitoneal fluid to accumulate by transtubal regurgitation during operative hysteroscopy, despite the quantity of fluid used.)

 

If perforation is suspected, laparoscopy or laparotomy is necessary to clarify the cause of pain, unstable vital signs, or free fluid visualized by ultrasound.⁶

Exercise extra care and precautions in women who have had a prior cesarean section, myomectomy, or uterine perforation. Complete visualization of uterine landmarks is necessary during operative hysteroscopy to exclude uterine dehiscence, sacculation, and perforation. Prior uterine surgery may cause myometrial weakness and lead to possible perforation. Do not proceed if abnormal uterine morphology is detected. If uterine perforation occurs, injury to bladder and bowel is possible when electrical energy is applied to a uterine wall compromised by prior surgery. Strict visualization of uterine anatomy is critical in this population so that bowel or bladder burns can be avoided.

MEDIA-RELATED COMPLICATIONS

Notorious complications and several recent lawsuits have stemmed from fluid overload. A common element has been the physician’s lack of awareness of how rapidly complications can arise, and what signs and symptoms are specific to the fluid used.

Monitor fluids vigilantly

Operative hysteroscopy must be performed in a fluid medium. The type of fluid depends on the surgeon’s preference and the instrument utilized, but any fluid can be associated with complications. Fluid choices with monopolar instruments include glycine 1.5%, a mixture of sorbitol 3% and mannitol 0.54%, and mannitol 5%. These are frequently used with the continuous-flow resectoscope. Bipolar operative hysteroscopy can be performed using saline.

The solution to media-related complications is basic: vigilant monitoring of fluids. A cavalier attitude, poor fluid documentation, and failure to respond to complications can lead to trouble. If fluid overload occurs, comanagement and consultation with an intensive care specialist is advised.

Distention media

Among the options for distention media in operative and diagnostic hysteroscopy are high-viscosity dextran 70 and low-viscosity fluids such as hypotonic, electrolyte-free and isotonic, electrolyte-containing solutions. The popularity of dextran 70 is waning, however. While it is immiscible with blood, significant complications have been reported.

Signs of anaphylactic reactions to dextran 70 include acute hypotension, hypoxia, pulmonary edema, fluid overload, fulminant coagulopathies, and anemia. The surgeon must operate quickly, minimize endometrial trauma, use continuous pulse oximetry, and obtain a preoperative coagulation panel.

Dextran 70 also can ruin operative hysteroscopes if they are not cleaned promptly and thoroughly after use.

Hypotonic, electrolyte-free solutions. With hypotonic, electrolyte-free solutions such as glycine 1.5%, early recognition of possible complications, including hyponatremic hypervolemia, is vital. For example, when glycine and sorbitol are metabolized, free water accumulates and the body attempts to achieve homeostasis through compensatory mechanisms such as osmosis, which moves free water into extracellular and intracellular spaces. This can lead to increased free water in the brain, resulting in cerebral edema, rising intracranial pressure, and cellular necrosis.

The cerebral cation pump normally pumps osmotically active cations into the extracellular space, thereby minimizing cerebral edema. However, this pump is inhibited by estrogen, so the compensatory mechanism is diminished.

Classic clinical features of hyponatremic hypervolemia include apprehension, confusion, fatigue, headache, mental agitation, nausea, visual disturbances (including blindness), vomiting, and weakness. These complications are more readily apparent when regional anesthesia is used rather than general anesthesia.

If hyponatremic hypervolemia goes unrecognized, bradycardia and hypertension can ensue, followed rapidly by cerebral and pulmonary edema and cardiovascular collapse. In addition, glycine 1.5% is metabolized to glycolic acid and ammonia. Free ammonia is associated with central nervous system disorders. Recognition and prompt treatment by an intensivist may prevent permanent neurologic sequelae, death, and lawsuits.⁷

Isotonic, electrolyte-containing solutions. Mannitol 5% is electrolyte poor but isotonic, creating less risk for hypo-osmolality. However, dilutional hyponatremia (ie, low sodium levels) can still occur.

Advantages of bipolar instruments. To minimize complications from hypotonic, electrolyte-free solutions, manufacturers developed operative hysteroscopes that can function in a bipolar environment. Bipolar instruments can operate in isotonic, physiologic, electrolyte-containing media. Hyponatremia and hypo-osmolality cannot occur with normal saline or Ringer’s lactate, but fluid overload can. (Fluid overload with saline can cause pulmonary edema and congestive heart failure.)

How much fluid will be absorbed? The answer depends on factors including surface area of the surgical field, duration of surgery, opened venous channels, type of irrigation fluid used, and pressure of the delivery system. Modern gynecologic suites employ fluid irrigation systems that continuously measure input and output, with alarms that signal a predetermined fluid deficit. The alarm indicates the need to halt the procedure and quickly evaluate the patient. Careful attention to the recommendations of Loffer et al⁸ would lead to fewer complications from fluid mismanagement.

Appropriate use of CO₂

High risk of embolism with CO₂ in operative procedures. Although diagnostic hysteroscopic procedures often are performed with carbon dioxide (CO₂), operative procedures never should be. The reason: the high risk of CO₂ embolism that occurs with open venous channels and vascular endometrium. The choice between CO₂ and fluid medium for diagnostic hysteroscopy often is determined by physician preference and the presence of uterine bleeding. Many gynecologists prefer CO₂ for its optical clarity and patient comfort during insufflation.⁹

 

Purge tubing of room air before each procedure. Embolic complications with CO₂ have been recorded with use of the neodymium: yttrium aluminum garnet (Nd:YAG) laser and during operative procedures. Less well known are the adverse sequelae that can occur with room air prior to beginning the procedure. It is critical to purge the entire tubal system with CO₂ prior to instrumentation, since up to 40 cm³ of room air may be insufflated into a patient when 200 cm of connective tubing with a 0.5-cm lumen is used.¹⁰ Wait for several minutes before starting the procedure so that the whole system is purged.

Advantages of CO₂in diagnostic hysteroscopy. Although any number of gaseous solutions could be used during diagnostic hysteroscopy, CO2is the most ideal due to its solubility in blood, the rapidity with which it is dissolved, and the large quantities (1 L/min or more) that would be necessary to cause a fatality. (A hysteroinsufflator utilizing 100 mL/min or less is used for diagnostic hysteroscopes.) When Bradner et al¹¹ reviewed the incidence of severe nonfatal embolism in 3,932 diagnostic hysteroscopic procedures using CO2, the risk of subclinical embolic events was 0.51%; 0.03% of patients experienced severe events. When special precautions were taken to deaerate the supply tubing and instruments, no events occurred in the next 1,000 cases. Thus, it is possible that ambient air trapped in tubing—rather than CO2—could be the culprit in gas emboli.

6 most common symptoms of venous or air emboli. Anesthesiologists and gynecologists must be vigilant to prevent venous or air emboli. Munro et al¹² succinctly outline the 6 most common symptoms:

• pulmonary hypertension
  
• hypercarbia
  
• hypoxia
  
• arrhythmias
  
• tachypnea
  
• systemic hypotension
  

Morbidity and mortality can be prevented when these symptoms are quickly recognized and promptly treated.¹³

Beware of a drop in end-tidal CO₂. The most common sign of impending cardiovascular collapse is a sudden decrease in end-tidal CO₂, when the right cardiac outflow tract is obstructed by CO₂, which leads to arterial oxygen (O₂) desaturation. If such a decrease is suspected, stop the procedure immediately and administer 100% O₂. (Also stop nitrous oxide, if used.) Turn the patient to the left lateral decubitis position and use a central venous catheter to aspirate gas, if necessary. Cardiac massage and a precordial thump may dislodge CO₂; unfortunately, high false-positive rates of pre-cordial Doppler make its use impractical.

 

How to minimize risks

• Avoid coaxial gas cooling tips associated with Nd:YAG crystal lasers
  
• Avoid a steep Trendelenburg position
  
• Keep cervix covered with sponge or dilator when operative hysteroscope is removed to minimize air embolism
  
• Deaerate the equipment prior to surgery Use a low-pressure hysteroscopic CO₂ insufflator
  
• Carefully monitor the patient
  
• Be highly suspicious when vital signs are unstable
  

POSTOPERATIVE COMPLICATIONS

Some complications of hysteroscopy may not become clinically evident for months or even years. The most common complications of hysteroscopic endometrial ablation include pregnancy, postablation tubal sterilization syndrome, new or worsening dysmenorrhea, hematometra, endometrial cancer, and failure to completely treat symptoms.

Patients scheduled for hysteroscopy must be informed of potential delayed risks of the procedure. In addition, all reproductive-aged women should be advised that pregnancy is possible after endometrial ablation or operative removal of an intracavitary mass; thus, contraception is crucial. The endometrial tissue is resilient and may regenerate after ablation.

Hematometra: Avoid cervical canal

Hematometra is an infrequent late complication of operative hysteroscopy. If menstruating women or those taking hormone replacement therapy experience cyclic or chronic lower pelvic pain after surgery, scarring or narrowing of the endometrial cavity may be the cause. Approximately 1% to 2% of women who undergo operative hysteroscopy experience this phenomenon. Most cases can be treated with cervical dilation alone.

Since the cervical canal contains no endometrial glands, there is no need to treat this area in women undergoing endometrial ablation. In fact, avoiding this area during treatment is a critical component of successful surgery.

Tubal sterilization syndrome possible after endometrial ablation

Consider this syndrome when a patient undergoing endometrial ablation complains of crampy, cyclic, unilateral or bilateral pelvic pain, possibly accompanied by vaginal spotting. Sometimes a unilateral mass can be palpated, but more commonly tenderness is elicited on pelvic examination.

Ultrasound may demonstrate fluid near the cornual region. Laparoscopy confirms the diagnosis by visualizing a swollen, edematous proximal fallopian tube. Salpingectomy may confirm hematosalpinx, chronic or acute inflammation, or hemosiderin deposits.

Treatment includes bilateral cornual resection and reablation of proximal endometrium, or hysterectomy.¹⁴

Pregnancy complications

Endometrial ablation is not to be regarded as a method of contraception. Patients needing birth control should consider concurrent tubal ligation or other reliable methods after this procedure.

The frequency of pregnancy after endometrial ablation ranges from 0.2% to 1.6%, though this data may represent underreporting. Pregnancy outcomes have been dismal in women conceiving after endometrial ablation. Complications include preterm labor, premature delivery, intrauterine growth retardation, prenatal death, postpartum hemorrhage, and placentation problems such as placenta accreta, increta, or percreta, as well as placental abruption.¹⁵

Uterine dehiscence and sacculation and extremely thin myometrium have been reported after uterine adhesiolysis, uterine perforation during operative hysteroscopy, and with myoma resections. A high index of suspicion is vital when a gravida presents with pelvic pain, decreased fetal movement, vaginal bleeding, or abnormal uterine masses detected ultrasonographically.

Signs of uterine rupture. Pregnancyrelated complications of operative hysteroscopy can be dramatic and fatal if not recognized quickly, as in the case of uterine rupture. Kerimis et al¹⁶ describe uterine rupture in a term pregnancy after hysteroscopic resection of a uterine septum. Severe fetal distress, maternal shoulder pain, and abdominal pain led to an emergency cesarean section. Intraoperative findings included a 7-cm tear from left cornua to right cornua. The original metroplasty, performed with cutting diathermy and laparoscopy, was not accompanied by complications or perforation.

Patients who experience intraoperative complications during metroplasty or deep resection of intramural fibroids should be informed of the risk of uterine rupture so they may consider elective cesarean. Regardless of the mode of delivery, prompt attention is vital if fetal distress is suspected.

Postablation warning signs

Patients undergoing endometrial ablation generally have a quick postoperative return to activity, minimal need for postoperative pain medication, and limited complaints. Beware of patients who make frequent postoperative phone calls and have escalating requirements for pain medication. While bowel and bladder injuries are infrequent—as is postoperative endometritis—these must be vigilantly considered and evaluated when patients complain of persistent pain, fever, and general malaise. Office evaluation is necessary, including thorough abdominal and pelvic examinations. Laboratory testing should include electrolytes, complete blood count, sedimentation rate, ultrasound, and a flat plate of the abdomen (kidneys, ureter, and bladder; upright) may be required. Sometimes a computed tomography scan of the pelvis/abdomen may be needed if perforation with bowel or bladder injury is suspected.

Hysteroscopic fibroid removal may be necessary after UAE

Uterine artery embolization (UAE) is gaining popularity for the treatment of symptomatic uterine fibroids. Transcatheter embolization of the uterine artery leads to occlusion of the fibroid, ischemic shrinkage of the fibroid, and shrinkage of residual myometrial tissue. Fibroids may migrate weeks to months after the procedure as the myometrium contracts and the treated fibroid degenerates, leading to delayed discharge, passage of necrotic fibroids, cramps, and heavy bleeding if the fibroid migrates to a submucosal location. Hysteroscopic removal is an obvious option.

 

Recently, De Iaco et al¹⁷ reported the development of a uterine fistula and discontinuity of the myometrium after hysteroscopic resection of an embolized migrated fibroid. They speculated this was due to the development of an avascular myometrium after UAE. The patient was asymptomatic, but routine diagnostic hysteroscopy revealed a 2-cm discontinuity of the uterine wall at the site of the previous resectoscopic myomectomy. The myometrium was white and less than the full thickness.

Ultrasound guidance improves outcomes

Coccia et al¹⁸ described the benefits of intraoperative ultrasound guidance during operative hysteroscopy in fibroid treatment and uterine septum removal. Prospective evaluation of 81 patients involved an experienced ultrasonographer who mapped the limits of treatment. Patients were compared to 45 historical controls who had been similarly treated with laparoscopic monitoring. Satisfactory outcomes included relief of menorrhagia, complete resection of fibroids (including full resection of intramural fibroids), and thorough metroplasty of uterine septum.

Ultrasound guidance made it possible to extend the resection beyond the limit conventionally defined by hysteroscopy; none of the patients in the ultrasound group required reintervention. Among controls, a second operation was necessary in 4 cases. Investigators concluded that a wider resection (10 to 15 mm distance from the external surface of the uterus) of fibroids was achieved using ultrasound guidance.

Dr. Bradley reports that she serves as a consultant to Karl Storz, ACMI, Olympus, and Gynecare, and as a lecturer for Novacept.

The state of the art can be simply put: Hysteroscopy is underutilized. Most studies consistently demonstrate the safety and efficacy of operative hysteroscopy, as well as high patient satisfaction.

It is a valuable tool in the evaluation and treatment of infertility, recurrent pregnancy loss, and abnormal and postmenopausal uterine bleeding, and is useful when saline infusion sonography findings are equivocal.

Further, if a global ablation device fails, the surgeon can convert to hysteroscopic ablation rather than abandon the procedure altogether. This is not as unusual as it might appear: In US Food and Drug Administration trials, there was a staggering 10% to 22% malfunction of global ablation technology.¹

Safe, easily learned

Although gynecologists are beginning to embrace this modality, many physicians avoid it because of inadequate training or exaggerated fears of complications. In reality, hysteroscopy is one of the safest and most easily acquired surgical skills in gynecology. For example, in a prospective evaluation of 13,600 diagnostic and operative hysteroscopic procedures performed at 63 hospitals in the Netherlands—which involved both established surgeons and residents—Jansen et al² found an astonishingly low complication rate of 0.28%, with no deaths.

Proper selection and treatment of patients and prompt intervention minimize complications as well as legal risks. Surgical misadventures and lawsuits occur with delayed intervention, failure to recognize pathology or risky conditions, and inadequate preventive maneuvers.

Overall, emphasis on safety is vital to success, and thorough awareness of potential complications is just as important.

Three types of complications

Complications fall into 3 categories (TABLE):

• Procedure-related
  
• Media-related
  
• Postoperative
  

PROCEDURE-RELATED COMPLICATIONS

Complication rates

In a retrospective investigation, Propst et al³ determined the rate of complications associated with specific hysteroscopic procedures. Demographic data and medical histories were collected for 925 women who had operative hysteroscopy in 1995 and 1996. The overall complication rate was 2.7%. Myomectomy and resection of uterine septa carried the greatest odds of complications; polypectomy and endometrial ablation had the lowest. Preoperative treatment with a gonadotropinreleasing hormone (GnRH) agonist increased the odds of complications by a factor of 4 to 7. Women under age 50 were more likely to experience complications than those over 50.

In the study by Jansen et al,² 38 complica-tions occurred in the 13,600 procedures. The greatest risk of complications occurred with adhesiolysis (4.48%), followed by endometrial resection (0.81%), myomectomy (0.75%), and polypectomy (0.38%).

Cervical entry requires special attention

Almost half of the complications in the Jansen study were related to cervical entry, so caution and, perhaps, preoperative cervical ripening are advised. Many premenopausal subjects were given GnRH analogues, which may render the cervix more resistant to dilation. Complications associated with a stenotic cervix include a cervical tear, creation of a false cervical passage, and uterine perforation.

Cervical ripening may help prevent uterine perforation. The most common complication, occurring in 14.2 cases per thousand, is uterine perforation.² The risk of this is highest in postpartum procedures, followed by procedures in postmenopausal, then perimenopausal, women. Patients with endometrial cancer also have a higher rate of perforation.

Risk factors for uterine perforation include:

• nulliparity
  
• menopause
  
• use of GnRH agonists
  
• prior cone biopsy
  
• markedly retroverted uterus
  
• undue force
  

Modern operative hysteroscopes often require dilation of the cervix to a number 8-10 Hegar dilator. Navigation of the internal os is critical before operative instruments can be inserted and the surgical procedure performed. In the past, use of preoperative laminaria was recommended to soften the cervix, except in women with marked cervical stenosis and iodine allergy. Preoperative cervical softening still should be considered in high-risk patients.

Vaginal or oral misoprostol for cervical ripening prior to operative hysteroscopy was evaluated in a randomized trial.⁴ Researchers found a reduced need for cervical dilation, a minimum of cervical complications, and reduced operative time in study patients compared with controls.

When 400 μg oral misoprostol is given 12 and 24 hours before surgery, it also softens the cervix and eases dilation.⁵ Although misoprostol has several bothersome side effects (such as lower abdominal pain and slight vaginal bleeding), few if any prevent its use.

Signs of perforation. Patients who sustain uterine perforation with subsequent intraperitoneal bleeding often complain of pain in the abdomen and shoulder, and experience hemodynamic instability. A quick sonographic survey of the abdomen will demonstrate free intraperitoneal fluid. (It is rare for much intraperitoneal fluid to accumulate by transtubal regurgitation during operative hysteroscopy, despite the quantity of fluid used.)

 

If perforation is suspected, laparoscopy or laparotomy is necessary to clarify the cause of pain, unstable vital signs, or free fluid visualized by ultrasound.⁶

Exercise extra care and precautions in women who have had a prior cesarean section, myomectomy, or uterine perforation. Complete visualization of uterine landmarks is necessary during operative hysteroscopy to exclude uterine dehiscence, sacculation, and perforation. Prior uterine surgery may cause myometrial weakness and lead to possible perforation. Do not proceed if abnormal uterine morphology is detected. If uterine perforation occurs, injury to bladder and bowel is possible when electrical energy is applied to a uterine wall compromised by prior surgery. Strict visualization of uterine anatomy is critical in this population so that bowel or bladder burns can be avoided.

MEDIA-RELATED COMPLICATIONS

Notorious complications and several recent lawsuits have stemmed from fluid overload. A common element has been the physician’s lack of awareness of how rapidly complications can arise, and what signs and symptoms are specific to the fluid used.

Monitor fluids vigilantly

Operative hysteroscopy must be performed in a fluid medium. The type of fluid depends on the surgeon’s preference and the instrument utilized, but any fluid can be associated with complications. Fluid choices with monopolar instruments include glycine 1.5%, a mixture of sorbitol 3% and mannitol 0.54%, and mannitol 5%. These are frequently used with the continuous-flow resectoscope. Bipolar operative hysteroscopy can be performed using saline.

The solution to media-related complications is basic: vigilant monitoring of fluids. A cavalier attitude, poor fluid documentation, and failure to respond to complications can lead to trouble. If fluid overload occurs, comanagement and consultation with an intensive care specialist is advised.

Distention media

Among the options for distention media in operative and diagnostic hysteroscopy are high-viscosity dextran 70 and low-viscosity fluids such as hypotonic, electrolyte-free and isotonic, electrolyte-containing solutions. The popularity of dextran 70 is waning, however. While it is immiscible with blood, significant complications have been reported.

Signs of anaphylactic reactions to dextran 70 include acute hypotension, hypoxia, pulmonary edema, fluid overload, fulminant coagulopathies, and anemia. The surgeon must operate quickly, minimize endometrial trauma, use continuous pulse oximetry, and obtain a preoperative coagulation panel.

Dextran 70 also can ruin operative hysteroscopes if they are not cleaned promptly and thoroughly after use.

Hypotonic, electrolyte-free solutions. With hypotonic, electrolyte-free solutions such as glycine 1.5%, early recognition of possible complications, including hyponatremic hypervolemia, is vital. For example, when glycine and sorbitol are metabolized, free water accumulates and the body attempts to achieve homeostasis through compensatory mechanisms such as osmosis, which moves free water into extracellular and intracellular spaces. This can lead to increased free water in the brain, resulting in cerebral edema, rising intracranial pressure, and cellular necrosis.

The cerebral cation pump normally pumps osmotically active cations into the extracellular space, thereby minimizing cerebral edema. However, this pump is inhibited by estrogen, so the compensatory mechanism is diminished.

Classic clinical features of hyponatremic hypervolemia include apprehension, confusion, fatigue, headache, mental agitation, nausea, visual disturbances (including blindness), vomiting, and weakness. These complications are more readily apparent when regional anesthesia is used rather than general anesthesia.

If hyponatremic hypervolemia goes unrecognized, bradycardia and hypertension can ensue, followed rapidly by cerebral and pulmonary edema and cardiovascular collapse. In addition, glycine 1.5% is metabolized to glycolic acid and ammonia. Free ammonia is associated with central nervous system disorders. Recognition and prompt treatment by an intensivist may prevent permanent neurologic sequelae, death, and lawsuits.⁷

Isotonic, electrolyte-containing solutions. Mannitol 5% is electrolyte poor but isotonic, creating less risk for hypo-osmolality. However, dilutional hyponatremia (ie, low sodium levels) can still occur.

Advantages of bipolar instruments. To minimize complications from hypotonic, electrolyte-free solutions, manufacturers developed operative hysteroscopes that can function in a bipolar environment. Bipolar instruments can operate in isotonic, physiologic, electrolyte-containing media. Hyponatremia and hypo-osmolality cannot occur with normal saline or Ringer’s lactate, but fluid overload can. (Fluid overload with saline can cause pulmonary edema and congestive heart failure.)

How much fluid will be absorbed? The answer depends on factors including surface area of the surgical field, duration of surgery, opened venous channels, type of irrigation fluid used, and pressure of the delivery system. Modern gynecologic suites employ fluid irrigation systems that continuously measure input and output, with alarms that signal a predetermined fluid deficit. The alarm indicates the need to halt the procedure and quickly evaluate the patient. Careful attention to the recommendations of Loffer et al⁸ would lead to fewer complications from fluid mismanagement.

Appropriate use of CO₂

High risk of embolism with CO₂ in operative procedures. Although diagnostic hysteroscopic procedures often are performed with carbon dioxide (CO₂), operative procedures never should be. The reason: the high risk of CO₂ embolism that occurs with open venous channels and vascular endometrium. The choice between CO₂ and fluid medium for diagnostic hysteroscopy often is determined by physician preference and the presence of uterine bleeding. Many gynecologists prefer CO₂ for its optical clarity and patient comfort during insufflation.⁹

 

Purge tubing of room air before each procedure. Embolic complications with CO₂ have been recorded with use of the neodymium: yttrium aluminum garnet (Nd:YAG) laser and during operative procedures. Less well known are the adverse sequelae that can occur with room air prior to beginning the procedure. It is critical to purge the entire tubal system with CO₂ prior to instrumentation, since up to 40 cm³ of room air may be insufflated into a patient when 200 cm of connective tubing with a 0.5-cm lumen is used.¹⁰ Wait for several minutes before starting the procedure so that the whole system is purged.

Advantages of CO₂in diagnostic hysteroscopy. Although any number of gaseous solutions could be used during diagnostic hysteroscopy, CO2is the most ideal due to its solubility in blood, the rapidity with which it is dissolved, and the large quantities (1 L/min or more) that would be necessary to cause a fatality. (A hysteroinsufflator utilizing 100 mL/min or less is used for diagnostic hysteroscopes.) When Bradner et al¹¹ reviewed the incidence of severe nonfatal embolism in 3,932 diagnostic hysteroscopic procedures using CO2, the risk of subclinical embolic events was 0.51%; 0.03% of patients experienced severe events. When special precautions were taken to deaerate the supply tubing and instruments, no events occurred in the next 1,000 cases. Thus, it is possible that ambient air trapped in tubing—rather than CO2—could be the culprit in gas emboli.

6 most common symptoms of venous or air emboli. Anesthesiologists and gynecologists must be vigilant to prevent venous or air emboli. Munro et al¹² succinctly outline the 6 most common symptoms:

• pulmonary hypertension
  
• hypercarbia
  
• hypoxia
  
• arrhythmias
  
• tachypnea
  
• systemic hypotension
  

Morbidity and mortality can be prevented when these symptoms are quickly recognized and promptly treated.¹³

Beware of a drop in end-tidal CO₂. The most common sign of impending cardiovascular collapse is a sudden decrease in end-tidal CO₂, when the right cardiac outflow tract is obstructed by CO₂, which leads to arterial oxygen (O₂) desaturation. If such a decrease is suspected, stop the procedure immediately and administer 100% O₂. (Also stop nitrous oxide, if used.) Turn the patient to the left lateral decubitis position and use a central venous catheter to aspirate gas, if necessary. Cardiac massage and a precordial thump may dislodge CO₂; unfortunately, high false-positive rates of pre-cordial Doppler make its use impractical.

 

How to minimize risks

• Avoid coaxial gas cooling tips associated with Nd:YAG crystal lasers
  
• Avoid a steep Trendelenburg position
  
• Keep cervix covered with sponge or dilator when operative hysteroscope is removed to minimize air embolism
  
• Deaerate the equipment prior to surgery Use a low-pressure hysteroscopic CO₂ insufflator
  
• Carefully monitor the patient
  
• Be highly suspicious when vital signs are unstable
  

POSTOPERATIVE COMPLICATIONS

Some complications of hysteroscopy may not become clinically evident for months or even years. The most common complications of hysteroscopic endometrial ablation include pregnancy, postablation tubal sterilization syndrome, new or worsening dysmenorrhea, hematometra, endometrial cancer, and failure to completely treat symptoms.

Patients scheduled for hysteroscopy must be informed of potential delayed risks of the procedure. In addition, all reproductive-aged women should be advised that pregnancy is possible after endometrial ablation or operative removal of an intracavitary mass; thus, contraception is crucial. The endometrial tissue is resilient and may regenerate after ablation.

Hematometra: Avoid cervical canal

Hematometra is an infrequent late complication of operative hysteroscopy. If menstruating women or those taking hormone replacement therapy experience cyclic or chronic lower pelvic pain after surgery, scarring or narrowing of the endometrial cavity may be the cause. Approximately 1% to 2% of women who undergo operative hysteroscopy experience this phenomenon. Most cases can be treated with cervical dilation alone.

Since the cervical canal contains no endometrial glands, there is no need to treat this area in women undergoing endometrial ablation. In fact, avoiding this area during treatment is a critical component of successful surgery.

Tubal sterilization syndrome possible after endometrial ablation

Consider this syndrome when a patient undergoing endometrial ablation complains of crampy, cyclic, unilateral or bilateral pelvic pain, possibly accompanied by vaginal spotting. Sometimes a unilateral mass can be palpated, but more commonly tenderness is elicited on pelvic examination.

Ultrasound may demonstrate fluid near the cornual region. Laparoscopy confirms the diagnosis by visualizing a swollen, edematous proximal fallopian tube. Salpingectomy may confirm hematosalpinx, chronic or acute inflammation, or hemosiderin deposits.

Treatment includes bilateral cornual resection and reablation of proximal endometrium, or hysterectomy.¹⁴

Pregnancy complications

Endometrial ablation is not to be regarded as a method of contraception. Patients needing birth control should consider concurrent tubal ligation or other reliable methods after this procedure.

The frequency of pregnancy after endometrial ablation ranges from 0.2% to 1.6%, though this data may represent underreporting. Pregnancy outcomes have been dismal in women conceiving after endometrial ablation. Complications include preterm labor, premature delivery, intrauterine growth retardation, prenatal death, postpartum hemorrhage, and placentation problems such as placenta accreta, increta, or percreta, as well as placental abruption.¹⁵

Uterine dehiscence and sacculation and extremely thin myometrium have been reported after uterine adhesiolysis, uterine perforation during operative hysteroscopy, and with myoma resections. A high index of suspicion is vital when a gravida presents with pelvic pain, decreased fetal movement, vaginal bleeding, or abnormal uterine masses detected ultrasonographically.

Signs of uterine rupture. Pregnancyrelated complications of operative hysteroscopy can be dramatic and fatal if not recognized quickly, as in the case of uterine rupture. Kerimis et al¹⁶ describe uterine rupture in a term pregnancy after hysteroscopic resection of a uterine septum. Severe fetal distress, maternal shoulder pain, and abdominal pain led to an emergency cesarean section. Intraoperative findings included a 7-cm tear from left cornua to right cornua. The original metroplasty, performed with cutting diathermy and laparoscopy, was not accompanied by complications or perforation.

Patients who experience intraoperative complications during metroplasty or deep resection of intramural fibroids should be informed of the risk of uterine rupture so they may consider elective cesarean. Regardless of the mode of delivery, prompt attention is vital if fetal distress is suspected.

Postablation warning signs

Patients undergoing endometrial ablation generally have a quick postoperative return to activity, minimal need for postoperative pain medication, and limited complaints. Beware of patients who make frequent postoperative phone calls and have escalating requirements for pain medication. While bowel and bladder injuries are infrequent—as is postoperative endometritis—these must be vigilantly considered and evaluated when patients complain of persistent pain, fever, and general malaise. Office evaluation is necessary, including thorough abdominal and pelvic examinations. Laboratory testing should include electrolytes, complete blood count, sedimentation rate, ultrasound, and a flat plate of the abdomen (kidneys, ureter, and bladder; upright) may be required. Sometimes a computed tomography scan of the pelvis/abdomen may be needed if perforation with bowel or bladder injury is suspected.

Hysteroscopic fibroid removal may be necessary after UAE

Uterine artery embolization (UAE) is gaining popularity for the treatment of symptomatic uterine fibroids. Transcatheter embolization of the uterine artery leads to occlusion of the fibroid, ischemic shrinkage of the fibroid, and shrinkage of residual myometrial tissue. Fibroids may migrate weeks to months after the procedure as the myometrium contracts and the treated fibroid degenerates, leading to delayed discharge, passage of necrotic fibroids, cramps, and heavy bleeding if the fibroid migrates to a submucosal location. Hysteroscopic removal is an obvious option.

 

Recently, De Iaco et al¹⁷ reported the development of a uterine fistula and discontinuity of the myometrium after hysteroscopic resection of an embolized migrated fibroid. They speculated this was due to the development of an avascular myometrium after UAE. The patient was asymptomatic, but routine diagnostic hysteroscopy revealed a 2-cm discontinuity of the uterine wall at the site of the previous resectoscopic myomectomy. The myometrium was white and less than the full thickness.

Ultrasound guidance improves outcomes

Coccia et al¹⁸ described the benefits of intraoperative ultrasound guidance during operative hysteroscopy in fibroid treatment and uterine septum removal. Prospective evaluation of 81 patients involved an experienced ultrasonographer who mapped the limits of treatment. Patients were compared to 45 historical controls who had been similarly treated with laparoscopic monitoring. Satisfactory outcomes included relief of menorrhagia, complete resection of fibroids (including full resection of intramural fibroids), and thorough metroplasty of uterine septum.

Ultrasound guidance made it possible to extend the resection beyond the limit conventionally defined by hysteroscopy; none of the patients in the ultrasound group required reintervention. Among controls, a second operation was necessary in 4 cases. Investigators concluded that a wider resection (10 to 15 mm distance from the external surface of the uterus) of fibroids was achieved using ultrasound guidance.

Dr. Bradley reports that she serves as a consultant to Karl Storz, ACMI, Olympus, and Gynecare, and as a lecturer for Novacept.

The state of the art can be simply put: Hysteroscopy is underutilized. Most studies consistently demonstrate the safety and efficacy of operative hysteroscopy, as well as high patient satisfaction.

It is a valuable tool in the evaluation and treatment of infertility, recurrent pregnancy loss, and abnormal and postmenopausal uterine bleeding, and is useful when saline infusion sonography findings are equivocal.

Further, if a global ablation device fails, the surgeon can convert to hysteroscopic ablation rather than abandon the procedure altogether. This is not as unusual as it might appear: In US Food and Drug Administration trials, there was a staggering 10% to 22% malfunction of global ablation technology.¹

Safe, easily learned

Although gynecologists are beginning to embrace this modality, many physicians avoid it because of inadequate training or exaggerated fears of complications. In reality, hysteroscopy is one of the safest and most easily acquired surgical skills in gynecology. For example, in a prospective evaluation of 13,600 diagnostic and operative hysteroscopic procedures performed at 63 hospitals in the Netherlands—which involved both established surgeons and residents—Jansen et al² found an astonishingly low complication rate of 0.28%, with no deaths.

Proper selection and treatment of patients and prompt intervention minimize complications as well as legal risks. Surgical misadventures and lawsuits occur with delayed intervention, failure to recognize pathology or risky conditions, and inadequate preventive maneuvers.

Overall, emphasis on safety is vital to success, and thorough awareness of potential complications is just as important.

Three types of complications

Complications fall into 3 categories (TABLE):

• Procedure-related
  
• Media-related
  
• Postoperative
  

PROCEDURE-RELATED COMPLICATIONS

Complication rates

In a retrospective investigation, Propst et al³ determined the rate of complications associated with specific hysteroscopic procedures. Demographic data and medical histories were collected for 925 women who had operative hysteroscopy in 1995 and 1996. The overall complication rate was 2.7%. Myomectomy and resection of uterine septa carried the greatest odds of complications; polypectomy and endometrial ablation had the lowest. Preoperative treatment with a gonadotropinreleasing hormone (GnRH) agonist increased the odds of complications by a factor of 4 to 7. Women under age 50 were more likely to experience complications than those over 50.

In the study by Jansen et al,² 38 complica-tions occurred in the 13,600 procedures. The greatest risk of complications occurred with adhesiolysis (4.48%), followed by endometrial resection (0.81%), myomectomy (0.75%), and polypectomy (0.38%).

Cervical entry requires special attention

Almost half of the complications in the Jansen study were related to cervical entry, so caution and, perhaps, preoperative cervical ripening are advised. Many premenopausal subjects were given GnRH analogues, which may render the cervix more resistant to dilation. Complications associated with a stenotic cervix include a cervical tear, creation of a false cervical passage, and uterine perforation.

Cervical ripening may help prevent uterine perforation. The most common complication, occurring in 14.2 cases per thousand, is uterine perforation.² The risk of this is highest in postpartum procedures, followed by procedures in postmenopausal, then perimenopausal, women. Patients with endometrial cancer also have a higher rate of perforation.

Risk factors for uterine perforation include:

• nulliparity
  
• menopause
  
• use of GnRH agonists
  
• prior cone biopsy
  
• markedly retroverted uterus
  
• undue force
  

Modern operative hysteroscopes often require dilation of the cervix to a number 8-10 Hegar dilator. Navigation of the internal os is critical before operative instruments can be inserted and the surgical procedure performed. In the past, use of preoperative laminaria was recommended to soften the cervix, except in women with marked cervical stenosis and iodine allergy. Preoperative cervical softening still should be considered in high-risk patients.

Vaginal or oral misoprostol for cervical ripening prior to operative hysteroscopy was evaluated in a randomized trial.⁴ Researchers found a reduced need for cervical dilation, a minimum of cervical complications, and reduced operative time in study patients compared with controls.

When 400 μg oral misoprostol is given 12 and 24 hours before surgery, it also softens the cervix and eases dilation.⁵ Although misoprostol has several bothersome side effects (such as lower abdominal pain and slight vaginal bleeding), few if any prevent its use.

Signs of perforation. Patients who sustain uterine perforation with subsequent intraperitoneal bleeding often complain of pain in the abdomen and shoulder, and experience hemodynamic instability. A quick sonographic survey of the abdomen will demonstrate free intraperitoneal fluid. (It is rare for much intraperitoneal fluid to accumulate by transtubal regurgitation during operative hysteroscopy, despite the quantity of fluid used.)

 

If perforation is suspected, laparoscopy or laparotomy is necessary to clarify the cause of pain, unstable vital signs, or free fluid visualized by ultrasound.⁶

Exercise extra care and precautions in women who have had a prior cesarean section, myomectomy, or uterine perforation. Complete visualization of uterine landmarks is necessary during operative hysteroscopy to exclude uterine dehiscence, sacculation, and perforation. Prior uterine surgery may cause myometrial weakness and lead to possible perforation. Do not proceed if abnormal uterine morphology is detected. If uterine perforation occurs, injury to bladder and bowel is possible when electrical energy is applied to a uterine wall compromised by prior surgery. Strict visualization of uterine anatomy is critical in this population so that bowel or bladder burns can be avoided.

MEDIA-RELATED COMPLICATIONS

Notorious complications and several recent lawsuits have stemmed from fluid overload. A common element has been the physician’s lack of awareness of how rapidly complications can arise, and what signs and symptoms are specific to the fluid used.

Monitor fluids vigilantly

Operative hysteroscopy must be performed in a fluid medium. The type of fluid depends on the surgeon’s preference and the instrument utilized, but any fluid can be associated with complications. Fluid choices with monopolar instruments include glycine 1.5%, a mixture of sorbitol 3% and mannitol 0.54%, and mannitol 5%. These are frequently used with the continuous-flow resectoscope. Bipolar operative hysteroscopy can be performed using saline.

The solution to media-related complications is basic: vigilant monitoring of fluids. A cavalier attitude, poor fluid documentation, and failure to respond to complications can lead to trouble. If fluid overload occurs, comanagement and consultation with an intensive care specialist is advised.

Distention media

Among the options for distention media in operative and diagnostic hysteroscopy are high-viscosity dextran 70 and low-viscosity fluids such as hypotonic, electrolyte-free and isotonic, electrolyte-containing solutions. The popularity of dextran 70 is waning, however. While it is immiscible with blood, significant complications have been reported.

Signs of anaphylactic reactions to dextran 70 include acute hypotension, hypoxia, pulmonary edema, fluid overload, fulminant coagulopathies, and anemia. The surgeon must operate quickly, minimize endometrial trauma, use continuous pulse oximetry, and obtain a preoperative coagulation panel.

Dextran 70 also can ruin operative hysteroscopes if they are not cleaned promptly and thoroughly after use.

Hypotonic, electrolyte-free solutions. With hypotonic, electrolyte-free solutions such as glycine 1.5%, early recognition of possible complications, including hyponatremic hypervolemia, is vital. For example, when glycine and sorbitol are metabolized, free water accumulates and the body attempts to achieve homeostasis through compensatory mechanisms such as osmosis, which moves free water into extracellular and intracellular spaces. This can lead to increased free water in the brain, resulting in cerebral edema, rising intracranial pressure, and cellular necrosis.

The cerebral cation pump normally pumps osmotically active cations into the extracellular space, thereby minimizing cerebral edema. However, this pump is inhibited by estrogen, so the compensatory mechanism is diminished.

Classic clinical features of hyponatremic hypervolemia include apprehension, confusion, fatigue, headache, mental agitation, nausea, visual disturbances (including blindness), vomiting, and weakness. These complications are more readily apparent when regional anesthesia is used rather than general anesthesia.

If hyponatremic hypervolemia goes unrecognized, bradycardia and hypertension can ensue, followed rapidly by cerebral and pulmonary edema and cardiovascular collapse. In addition, glycine 1.5% is metabolized to glycolic acid and ammonia. Free ammonia is associated with central nervous system disorders. Recognition and prompt treatment by an intensivist may prevent permanent neurologic sequelae, death, and lawsuits.⁷

Isotonic, electrolyte-containing solutions. Mannitol 5% is electrolyte poor but isotonic, creating less risk for hypo-osmolality. However, dilutional hyponatremia (ie, low sodium levels) can still occur.

Advantages of bipolar instruments. To minimize complications from hypotonic, electrolyte-free solutions, manufacturers developed operative hysteroscopes that can function in a bipolar environment. Bipolar instruments can operate in isotonic, physiologic, electrolyte-containing media. Hyponatremia and hypo-osmolality cannot occur with normal saline or Ringer’s lactate, but fluid overload can. (Fluid overload with saline can cause pulmonary edema and congestive heart failure.)

How much fluid will be absorbed? The answer depends on factors including surface area of the surgical field, duration of surgery, opened venous channels, type of irrigation fluid used, and pressure of the delivery system. Modern gynecologic suites employ fluid irrigation systems that continuously measure input and output, with alarms that signal a predetermined fluid deficit. The alarm indicates the need to halt the procedure and quickly evaluate the patient. Careful attention to the recommendations of Loffer et al⁸ would lead to fewer complications from fluid mismanagement.

Appropriate use of CO₂

High risk of embolism with CO₂ in operative procedures. Although diagnostic hysteroscopic procedures often are performed with carbon dioxide (CO₂), operative procedures never should be. The reason: the high risk of CO₂ embolism that occurs with open venous channels and vascular endometrium. The choice between CO₂ and fluid medium for diagnostic hysteroscopy often is determined by physician preference and the presence of uterine bleeding. Many gynecologists prefer CO₂ for its optical clarity and patient comfort during insufflation.⁹

 

Purge tubing of room air before each procedure. Embolic complications with CO₂ have been recorded with use of the neodymium: yttrium aluminum garnet (Nd:YAG) laser and during operative procedures. Less well known are the adverse sequelae that can occur with room air prior to beginning the procedure. It is critical to purge the entire tubal system with CO₂ prior to instrumentation, since up to 40 cm³ of room air may be insufflated into a patient when 200 cm of connective tubing with a 0.5-cm lumen is used.¹⁰ Wait for several minutes before starting the procedure so that the whole system is purged.

Advantages of CO₂in diagnostic hysteroscopy. Although any number of gaseous solutions could be used during diagnostic hysteroscopy, CO2is the most ideal due to its solubility in blood, the rapidity with which it is dissolved, and the large quantities (1 L/min or more) that would be necessary to cause a fatality. (A hysteroinsufflator utilizing 100 mL/min or less is used for diagnostic hysteroscopes.) When Bradner et al¹¹ reviewed the incidence of severe nonfatal embolism in 3,932 diagnostic hysteroscopic procedures using CO2, the risk of subclinical embolic events was 0.51%; 0.03% of patients experienced severe events. When special precautions were taken to deaerate the supply tubing and instruments, no events occurred in the next 1,000 cases. Thus, it is possible that ambient air trapped in tubing—rather than CO2—could be the culprit in gas emboli.

6 most common symptoms of venous or air emboli. Anesthesiologists and gynecologists must be vigilant to prevent venous or air emboli. Munro et al¹² succinctly outline the 6 most common symptoms:

• pulmonary hypertension
  
• hypercarbia
  
• hypoxia
  
• arrhythmias
  
• tachypnea
  
• systemic hypotension
  

Morbidity and mortality can be prevented when these symptoms are quickly recognized and promptly treated.¹³

Beware of a drop in end-tidal CO₂. The most common sign of impending cardiovascular collapse is a sudden decrease in end-tidal CO₂, when the right cardiac outflow tract is obstructed by CO₂, which leads to arterial oxygen (O₂) desaturation. If such a decrease is suspected, stop the procedure immediately and administer 100% O₂. (Also stop nitrous oxide, if used.) Turn the patient to the left lateral decubitis position and use a central venous catheter to aspirate gas, if necessary. Cardiac massage and a precordial thump may dislodge CO₂; unfortunately, high false-positive rates of pre-cordial Doppler make its use impractical.

 

How to minimize risks

• Avoid coaxial gas cooling tips associated with Nd:YAG crystal lasers
  
• Avoid a steep Trendelenburg position
  
• Keep cervix covered with sponge or dilator when operative hysteroscope is removed to minimize air embolism
  
• Deaerate the equipment prior to surgery Use a low-pressure hysteroscopic CO₂ insufflator
  
• Carefully monitor the patient
  
• Be highly suspicious when vital signs are unstable
  

POSTOPERATIVE COMPLICATIONS

Some complications of hysteroscopy may not become clinically evident for months or even years. The most common complications of hysteroscopic endometrial ablation include pregnancy, postablation tubal sterilization syndrome, new or worsening dysmenorrhea, hematometra, endometrial cancer, and failure to completely treat symptoms.

Patients scheduled for hysteroscopy must be informed of potential delayed risks of the procedure. In addition, all reproductive-aged women should be advised that pregnancy is possible after endometrial ablation or operative removal of an intracavitary mass; thus, contraception is crucial. The endometrial tissue is resilient and may regenerate after ablation.

Hematometra: Avoid cervical canal

Hematometra is an infrequent late complication of operative hysteroscopy. If menstruating women or those taking hormone replacement therapy experience cyclic or chronic lower pelvic pain after surgery, scarring or narrowing of the endometrial cavity may be the cause. Approximately 1% to 2% of women who undergo operative hysteroscopy experience this phenomenon. Most cases can be treated with cervical dilation alone.

Since the cervical canal contains no endometrial glands, there is no need to treat this area in women undergoing endometrial ablation. In fact, avoiding this area during treatment is a critical component of successful surgery.

Tubal sterilization syndrome possible after endometrial ablation

Consider this syndrome when a patient undergoing endometrial ablation complains of crampy, cyclic, unilateral or bilateral pelvic pain, possibly accompanied by vaginal spotting. Sometimes a unilateral mass can be palpated, but more commonly tenderness is elicited on pelvic examination.

Ultrasound may demonstrate fluid near the cornual region. Laparoscopy confirms the diagnosis by visualizing a swollen, edematous proximal fallopian tube. Salpingectomy may confirm hematosalpinx, chronic or acute inflammation, or hemosiderin deposits.

Treatment includes bilateral cornual resection and reablation of proximal endometrium, or hysterectomy.¹⁴

Pregnancy complications

Endometrial ablation is not to be regarded as a method of contraception. Patients needing birth control should consider concurrent tubal ligation or other reliable methods after this procedure.

The frequency of pregnancy after endometrial ablation ranges from 0.2% to 1.6%, though this data may represent underreporting. Pregnancy outcomes have been dismal in women conceiving after endometrial ablation. Complications include preterm labor, premature delivery, intrauterine growth retardation, prenatal death, postpartum hemorrhage, and placentation problems such as placenta accreta, increta, or percreta, as well as placental abruption.¹⁵

Uterine dehiscence and sacculation and extremely thin myometrium have been reported after uterine adhesiolysis, uterine perforation during operative hysteroscopy, and with myoma resections. A high index of suspicion is vital when a gravida presents with pelvic pain, decreased fetal movement, vaginal bleeding, or abnormal uterine masses detected ultrasonographically.

Signs of uterine rupture. Pregnancyrelated complications of operative hysteroscopy can be dramatic and fatal if not recognized quickly, as in the case of uterine rupture. Kerimis et al¹⁶ describe uterine rupture in a term pregnancy after hysteroscopic resection of a uterine septum. Severe fetal distress, maternal shoulder pain, and abdominal pain led to an emergency cesarean section. Intraoperative findings included a 7-cm tear from left cornua to right cornua. The original metroplasty, performed with cutting diathermy and laparoscopy, was not accompanied by complications or perforation.

Patients who experience intraoperative complications during metroplasty or deep resection of intramural fibroids should be informed of the risk of uterine rupture so they may consider elective cesarean. Regardless of the mode of delivery, prompt attention is vital if fetal distress is suspected.

Postablation warning signs

Patients undergoing endometrial ablation generally have a quick postoperative return to activity, minimal need for postoperative pain medication, and limited complaints. Beware of patients who make frequent postoperative phone calls and have escalating requirements for pain medication. While bowel and bladder injuries are infrequent—as is postoperative endometritis—these must be vigilantly considered and evaluated when patients complain of persistent pain, fever, and general malaise. Office evaluation is necessary, including thorough abdominal and pelvic examinations. Laboratory testing should include electrolytes, complete blood count, sedimentation rate, ultrasound, and a flat plate of the abdomen (kidneys, ureter, and bladder; upright) may be required. Sometimes a computed tomography scan of the pelvis/abdomen may be needed if perforation with bowel or bladder injury is suspected.

Hysteroscopic fibroid removal may be necessary after UAE

Uterine artery embolization (UAE) is gaining popularity for the treatment of symptomatic uterine fibroids. Transcatheter embolization of the uterine artery leads to occlusion of the fibroid, ischemic shrinkage of the fibroid, and shrinkage of residual myometrial tissue. Fibroids may migrate weeks to months after the procedure as the myometrium contracts and the treated fibroid degenerates, leading to delayed discharge, passage of necrotic fibroids, cramps, and heavy bleeding if the fibroid migrates to a submucosal location. Hysteroscopic removal is an obvious option.

 

Recently, De Iaco et al¹⁷ reported the development of a uterine fistula and discontinuity of the myometrium after hysteroscopic resection of an embolized migrated fibroid. They speculated this was due to the development of an avascular myometrium after UAE. The patient was asymptomatic, but routine diagnostic hysteroscopy revealed a 2-cm discontinuity of the uterine wall at the site of the previous resectoscopic myomectomy. The myometrium was white and less than the full thickness.

Ultrasound guidance improves outcomes

Coccia et al¹⁸ described the benefits of intraoperative ultrasound guidance during operative hysteroscopy in fibroid treatment and uterine septum removal. Prospective evaluation of 81 patients involved an experienced ultrasonographer who mapped the limits of treatment. Patients were compared to 45 historical controls who had been similarly treated with laparoscopic monitoring. Satisfactory outcomes included relief of menorrhagia, complete resection of fibroids (including full resection of intramural fibroids), and thorough metroplasty of uterine septum.

Ultrasound guidance made it possible to extend the resection beyond the limit conventionally defined by hysteroscopy; none of the patients in the ultrasound group required reintervention. Among controls, a second operation was necessary in 4 cases. Investigators concluded that a wider resection (10 to 15 mm distance from the external surface of the uterus) of fibroids was achieved using ultrasound guidance.

Dr. Bradley reports that she serves as a consultant to Karl Storz, ACMI, Olympus, and Gynecare, and as a lecturer for Novacept.

Are adhesions a pathologic response to injury or a normal aspect of healing? Can they be avoided, or are preventive efforts part of the problem? How useful are the different barriers in gynecologic surgery? What is the ideal adjuvant?

OBG Management convened a panel of experts to explore these and other questions.

Common problem, high recurrence rate

DECHERNEY: Adhesion formation is serious because it is associated with clinical entities such as infertility, pelvic pain, and bowel obstruction. We all agree that approximately 40% of people who undergo primary surgery develop adhesions and that 80% to 90% of patients who undergo lysis develop recurrent adhesions.

SANFILIPPO: One study several years ago explored adhesion formation.¹ Unfortunately, no matter how meticulous the surgeon is, adhesions will form, even with microsurgical techniques and carefully ensured hemostasis.

HURD: The number of patients with significant adhesion formation after some gynecologic procedures has been reported to be greater than 90%.²

DECHERNEY: That higher incidence usually occurs after general surgery—and there’s a reason it is so high: General surgeons don’t use adjunctive therapy. They are critical of it. It is to our credit as gynecologic surgeons that we adopted adjunctive therapies about 15 years ago with the introduction of dextran 70 (Hyskon; Medisan Pharmaceuticals, Parsippany, NJ).

PAGIDAS: If anything, the pelvis seems to have even more of a predilection for adhesion formation than the abdomen, probably because of the close proximity of structures.

How and why adhesions form

DECHERNEY: What is the pathophysiology of adhesion formation? Let’s say you have 2 raw surface areas. What happens?

The process

PAGIDAS: The increase in leukotrienes and prostaglandins and the decrease in plasminogen activity (which actually initiates the inflammation) appear to be significant.

HURD: Vessel permeability also increases, and inflammatory cells leak through the vessels and set up a matrix for adhesion formation.

DECHERNEY: So we have 2 raw surface areas with fibrin leaking out and forming bridges between them.

PAGIDAS: The key is that it takes 2 surfaces to form these bridges. As I mentioned, the greater proximity of pelvic structures—particularly around the tube and ovary—probably contributes to adhesion formation.

DECHERNEY: Macrophage activity also is important. The macrophage “migrates” along these fibrin bridges and lays down collagen over a period of time. Then the collagen becomes organized and, eventually, vascular.

Window of opportunity

SANFILIPPO: Adhesion formation probably occurs and is pretty well established within 5 to 7 days of the precipitating event—usually surgery. Once that process is under way, attempts to halt it yield diminishing returns. Unfortunately, we don’t know how to interfere with it in a positive way.

HURD: Under normal conditions, there seems to be a balance between fibrin deposition and fibrinolysis. In some tissues, however, these functions become imbalanced. This disparity may contribute more to adhesions than the actual laying down of fibrin—especially in tissue that is hypoxic.

DECHERNEY: Would you say that adhesion formation represents normal or abnormal healing?

HURD: It is one of the body’s normal protective mechanisms and an important part of healing. Without it, any abdominal injury would likely result in death.

SANFILIPPO: I don’t think it differs that much from processes that occur externally. For example, if you get cut deeply enough, you develop a scar. Is that scar part of the normal healing process? It is.

PAGIDAS: Right. It is a normal process of tissue remodeling. The question is: What allows it to go astray?

What surgical techniques help prevent adhesions?

DECHERNEY: Let’s review the aspects of surgical technique that are important for adhesion prevention.

 

PAGIDAS: I emphasize the value of microsurgical techniques, which help to minimize severe tissue handling. It also is important to keep surfaces moistened so they don’t desiccate.

SANFILIPPO: I agree with Dr. Pagidas about microsurgical techniques such as gentle tissue handling, careful hemostasis, and keeping tissues moist. If we follow these principles, we create an environment that minimizes the potential for adhesion formation.

HURD: The findings of many well-controlled animal studies have been surprising. For example, it is difficult to demonstrate that drying of tissue increases adhesions.³ Probably the greatest contributor to adhesions in these models was abrasion.⁴ One way that laparoscopic surgery decreases adhesions is by avoiding abrasion of the bowel mucosa, which occurs specifically with packing.

In open cases, one thing we can do to minimize the risk of adhesions is to pack gently when needed. Also, we should avoid using packing to reposition the bowel.

Another factor frequently overlooked is the application of heat, which appears to be a very effective way to create adhesions. This probably isn’t an issue for laparoscopic cases, but when you use irrigation fluid in an open case, watch the temperature. If it feels hot to you, you need to worry about potential injury to the bowel surfaces.

PAGIDAS: That is critical. In abdominal cases you want to make sure irrigation fluid is warm, but not too warm, because heat increases the vascular permeability of vessels and leads to more macrophages, more prostaglandins, and more leukotrienes.

HURD: Another important element is the type of suture material used.

DECHERNEY: Overall, we need to minimize the use of sutures. For example, when I am operating laparoscopically on an ovarian cyst, I try to apply bipolar energy to the edges so that they will coapt without a stitch.

HURD: When it first became clear that suturing ovaries increased adhesion formation, we conducted a controlled trial of different kinds of sutures in animals. Not surprisingly, we found that the less reactive the suture, the fewer adhesions.⁵ Sutures that are absorbed more slowly, such as polydioxanone, seem to be less reactive.

Obviously, inert sutures like nylon are the least reactive, but they are permanent. It is assumed that animal-protein sutures such as chromic and plain gut are the most reactive, although I am not sure there are sufficient data to support that conclusion.

 

Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why. It may be related to decreased suturing.

DECHERNEY: Bulk is important, too—that is, the number of throws in the suture. When Vicryl (polyglactin 910) became available, we conducted a study in mice using proportionately small Vicryl plaques to determine whether this would be good a barrier (A. DeCherney, MD, unpublished data). It caused a tremendous amount of adhesion because so much foreign matter was applied.

We also did a study using human-size titanium clips in rats (A. DeCherney, MD, unpublished data). Not surprisingly, there was a lot of adhesion formation.

FIGURE 1 Pelvic adhesions: How they develop, problems they cause

Adhesions occur when 2 or more raw surfaces are exposed to leaking fibrin, which forms a bridge between the surfaces. Macrophages “migrate” along these bridges, depositing collagen.

The pelvis has a greater predilection for adhesions than the abdomen because of the close proximity of structures.

Although adhesions represent one of the body’s protective mechanisms, they may also cause pain or interfere with fertility, bowel function, or other processes.

Laparoscopy versus laparotomy: More adhesions in open cases?

DECHERNEY: Based on all the techniques we have learned from microsurgery—with the exception of magnification—it appears that laparoscopic procedures are less likely to cause adhesions than laparotomy. Do you agree?

PAGIDAS: I think so. As Dr. Hurd noted, a main reason is the diminished tissue handling, because there is no packing.

HURD: Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why.^6,7 It may be related to decreased suturing.

DECHERNEY: Less bleeding occurs because surgeons are less aggressive laparoscopically than in laparotomy.

I have stopped doing difficult cases laparoscopically. For example, it is rare for me to operate laparoscopically on a patient with stage IV endometriosis, at least when it comes to infertility—I might consider laparoscopy for pain.

I think case selection plays a role as well, although there are few data to back that up. It is purely clinical opinion.

 

SANFILIPPO: We need a well-designed prospective study to explore the effects of laparotomy versus laparoscopy. Existing data are not clear. You would assume laparoscopy would be associated with less adhesion formation. But genetic or other factors may explain why patient A is more prone to adhesions than patient B.

Does anybody think carbon dioxide plays a role in adhesion formation?

HURD: In the laboratory, carbon dioxide increases cell growth.⁸ Without an increased carbon dioxide concentration in the atmosphere, cell cultures don’t grow well. This might suggest that the carbon dioxide used for laparoscopy could actually enhance adhesion growth. Fortunately, this does not appear to be the case clinically.

With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision.

DECHERNEY: What about second-look laparoscopies? Do any of you perform them after a patient has undergone lysis of adhesions?

SANFILIPPO: Only as part of a research protocol. It amazes me how rapidly adhesions can form and how dense they are 2 or more weeks after the initial laparoscopic surgery.

PAGIDAS: We tend to limit second-look laparoscopy to a research protocol, although it is sometimes valuable after laparoscopic or abdominal myomectomy, which has the highest incidence of adhesions. If the surgeon can perform a second look and lyse adhesions, he or she may potentially alter the reproductive outcome. However, with assisted reproductive techniques becoming integral to every infertility case, that approach has begun to go out of style.

SANFILIPPO: That’s a good point. With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision. If adhesions do occur, it is best if they occur toward the bladder rather than in the area of the tubes and ovaries.

HURD: For second-look laparoscopy, we must keep in mind the cost and the small but real risks of surgery. Until good controlled studies show a reasonable clinical advantage, this approach probably should remain a research protocol.

SANFILIPPO: The literature suggests it is helpful, but does not help fertility, so second-look laparoscopy is used mainly to evaluate adjunctive therapies. I don’t think anybody uses it as a primary therapy anymore.

Bowel obstruction still a risk, though rarely seen by Ob/Gyns

DECHERNEY: Although bowel obstruction is fairly common, Ob/Gyns do not often encounter it because it occurs relatively distant from the index surgery. Even though a patient may not experience bowel obstruction in the first year, an obstruction related to the index surgery is just as likely to develop 20 years later as 2 years later. These patients usually are treated by general surgeons. Still, we should beware of the potential for bowel obstruction and take steps to prevent it, if at all possible. Do you agree?

PAGIDAS: Yes. We tend to forget about bowel obstruction because we rarely follow patients past pregnancy or the first trimester if they are seeking infertility treatment.

HURD: The primary problem seems to be the abdominal wall incision. Fortunately, cesarean section seems to carry a decreased risk of abdominal wall adhesions, probably because the uterus serves as a splint over the incision.

The Pfannenstiel incision also appears to have some advantage. Both human and animal models suggest little advantage or disadvantage when peritoneal closure is compared to nonclosure.^9,10

SANFILIPPO: I’m curious about how the panelists manage loose clips. If you are using an EndoGIA (US Surgical, Norwalk, Conn) or other stapling device and you have free-floating clips, do you make a concerted effort to find them? In some cases, they have been implicated in bowel adhesion and obstruction. I try to retrieve loose clips, whether open or closed.

HURD: The advantage of those devices is minimal tissue damage, and the clips are inert. In general, inert, nonreactive clips have not been implicated as much in adhesion formation. I retrieve them if I see them, but I don’t search them out.

PAGIDAS: I do the same. If the clips are visible, I remove them. But I would not repack the bowel or do anything more heroic than look in locations where they might be.

What drugs may inhibit inflammatory response?

DECHERNEY: What about use of pharmacologic agents to prevent adhesions? Is there reason to think research should focus on inhibiting the inflammatory response? How important is polymorphonuclear cell infiltration?

Cyclooxygenase (COX) 2 agents could be helpful for inhibition of platelet function, since they are low in side effects. Thus, high doses of these drugs might be effective. At one time, aspirin was proposed, but you’d have to give a human so much aspirin that her ears would ring.

 

PAGIDAS: Pharmacologic agents have a role, especially for dampening the inflammatory immune response. But you don’t want to dampen it completely because, as we observed, it is an important part of healing. The difficulty is finding a balance between allowing the tissue to heal and preventing adhesions.

HURD: We studied the ability of a water-soluble prostaglandin inhibitor to prevent postoperative adhesion formation. Like many other agents, we found only a partial response.¹¹

DECHERNEY: With current options, the best you can aim for is a 50% reduction.

Adjunctive therapy likely to limit adhesion rate Hydroflotation

DECHERNEY: The original adjunctive therapy was 20 mg dexamethasone and 25 mg femergin in 200 cc of Ringer’s lactate, with an equivalent amount of dexamethasone and femergin every 4 hours for a total of 6 doses. I prescribed that regimen because I was trained to do so. I stopped after it became clear that hydroflotation from the fluid—not the medication itself—was responsible for the improvement.

I must admit I gave it up reluctantly; patients felt fabulous with those higher load doses of glucocorticoids after surgery.

Do any of you use crystalloids as adjunctive therapy?

SANFILIPPO: In the animal model, they are so rapidly absorbed that they aren’t effective. I was a strong advocate, but now I don’t use them at all.

HURD: A lot depends on the kind of case. For instance, at the end of an open myomectomy, the patient often is oozing, so you want to use a barrier that blood won’t affect.

For ovarian surgery, you might want to specifically target the ovaries with some kind of coverage. But when you are doing a broad lysis of adhesions, you have few choices to cover the pelvis. In those cases I use hydroflotation with Ringer’s lactate. Both human and animal studies have shown some benefit in preventing adhesions, and it appears to have little risk.^12,13

It’s better than nothing, in my opinion.

DECHERNEY: Do you use dextran 70 or crystalloids?

HURD: I use Ringer’s lactate solution. I was trained in the dextran 70 era, and there were certain problems with that approach. Since studies have failed to show a consistent effect of dextran 70, I no longer use this solution.^14,15

DECHERNEY: Another problem with crystalloids is that they leak, which is disconcerting to the patient.

HURD: They also can mask an injury to the bladder in difficult cases.

DECHERNEY: I agree that dextran 70 is only appropriate in certain cases, but it is a good hydroflotation agent. Every cubic centimeter of dextran 70 brings in 1.2 cc of transudate, so it hangs around for at least 4 days.

It is appropriate only for certain surface areas—mainly the cul-de-sac. It is harmful on raw surface areas on the lateral pelvic sidewall because it tends to push the ovary and tube to those areas. Unless you are doing a lot of work in the deep pelvis, I would avoid dextran 70.

In addition, there have been allergic reactions, most of which seem to occur in patients with fluid overload; a lot of the dextran 70 is absorbed.

 

With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal.

SANFILIPPO: Dextran 70 is not recommended for patients with sugar beet allergy, either.

We completed a study in a rabbit model, in which the peritoneal cavity was lavaged with either chlorhexidine or iodine.¹⁶ At the time of second-look surgery, the rate of adhesion formation was decreased, especially with the iodine preparation. I would hope that this has potential in humans.

DECHERNEY: In your lavage procedures to prevent adhesions, do any of you use heparin?

HURD: No.

PAGIDAS: I don’t think any evidence suggests that local administration changes the outcome.

DECHERNEY: I agree. When heparin has been used, the doses have been so low that it was not terribly helpful. And when you consider that hemorrhage can be a problem, heparin is probably deleterious rather than helpful.

Barriers

DECHERNEY: What about barriers? The first to become available, Interceed (Gynecare, a division of Ethicon, Somerville, NJ), is oxidized cellulose, similar to Surgicell (Johnson & Johnson, New Brunswick, NJ). Since it gelates quickly, there is no fenestration, so the fibrin is unable to penetrate. However, if the patient has bleeding by capillary action, the raw surface just moves from one side of the Interceed barrier to the other.

What has been your experience? Do you use it?

PAGIDAS: I do not use Interceed, although prospective randomized trials and a meta-analysis confirmed its benefits in de novo formation and reformation.^17-19 I don’t use it because it requires complete hemostasis. Also, with the surfaces we work on—notably, the ovary and tube—it is difficult to apply to just 1 surface area. From a clinical perspective, I appreciate the data, but it is hard to ensure a good application to optimize its effectiveness.

SANFILIPPO: I use Interceed, but I agree with you about its limitations. Meticulous hemostasis is a prerequisite.

HURD: With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal. If those patients have dense adhesions of the ovary or the sidewall, I generally leave them alone, and I try to avoid putting Interceed around the ovaries. No study has shown that using Interceed improves pregnancy.

In contrast, when a chronic pain patient’s ovaries are densely adherent to the cul-de-sac, which appears to be highly associated with dyspareunia, I lyse the adhesions, achieve meticulous hemostasis, and then use Interceed. It is hard to demonstrate in a study that this approach decreases the chance of pain. Even so, it certainly does decrease the chance of the ovaries being adherent.

DECHERNEY: One issue with Interceed is that we don’t know what happens to it once the abdomen is closed. It may migrate significantly.

Psychological issues may also be involved. For example, patients with multiple somatic complaints may be less likely to benefit from lysis of adhesions.

PAGIDAS: Right. Interestingly, a meta-analysis of all the randomized trials involving mechanical barriers found no correlation to pregnancy outcome or pelvic pain.¹⁸ If we were to consider new trials, the psychological aspect would be worth looking into.

Seprafilm

DECHERNEY: Let’s move on to Seprafilm (Genzyme, Cambridge, Mass). What is it and how useful is it?

HURD: Seprafilm is modified hyaluronic acid, which forms a brittle, thin plastic layer. It is somewhat difficult to work with but, once it is in place, seems to adhere well. The presence of blood does not appear to be a problem, since the Seprafilm forms an impermeable barrier—unless it breaks. I have found it especially useful in myomectomies, which produce postoperative oozing through the incisions no matter how hard you try to prevent it.

In addition, in open cases, surfaces can easily be covered with this material. Unfortunately, it can’t be used laparoscopically because it is so brittle.

 

DECHERNEY: In my opinion, that is its major drawback.

PAGIDAS: In cardiac surgery, Seprafilm appears to work quite effectively.

SANFILIPPO: The manufacturer initially focused on surgeons in the context of sigmoid colon surgery, and it seems to work well in that setting.

Intergel

DECHERNEY: That brings us to the current state of the art: gels. I’m sure you all are familiar with Intergel (Lifecore Biomedical, Chaska, Minn), which is a ferrous derivative of hyaluronic acid that works by coating the raw surface areas. It also has the theoretical advantage of ease of use. Have any of you used Intergel?

HURD: As you know it was only recently approved by the US Food and Drug Administration, but not for laparoscopic use. It may work best on abraded bowel, which is avoided by laparoscopic surgery.

As you are probably also aware, the manufacturer recently took it off the market because of unusual side effects, namely a chemical peritonitis. Although peritonitis was cited as being rare, we encountered it in probably half the patients we operated on.

PAGIDAS: When we used it on hospitalized patients, the peritonitis wasn’t that obvious, since there was an expectation of significant pain. However, when we used Intergel on short-stay patients, we had to readmit them and do a full workup because we were concerned about bowel perforations. I’m surprised the manufacturer didn’t take it off the market sooner.

DECHERNEY: It seems strange, since Intergel has been used in Europe for a while now. I’ve used it in only 1 case and didn’t have adverse effects. It seemed to work well.

SANFILIPPO: It had all the right ingredients for success. It is unfortunate that these side effects have prohibited its use.

There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery.

Gels and the cost factor

DECHERNEY: Other gels are in the pipeline. I’m reminded of plasminogen activator, which is a powerful antiadhesive agent that lyses fibrin effectively. Unfortunately, it is prohibitively expensive.

The next phase likely will involve the so-called polymers. If you spray them on your hand, they are activated by light or another chemical and become a cellophane-like substance. The problem is viscosity. If sprayed on the sidewall, for instance, they run halfway down before they are activated, so the entire surface does not get covered.

PAGIDAS: One concern with polymers is that they could actually bring surfaces together when they polymerize. We still have a lot to learn.

DECHERNEY: Let’s say a new gel comes on the market that takes reformation adhesions from 90% to 10%, as opposed to 40% recurrence. Would you use it in all 4 of the cases we discuss here?

HURD: If it was that effective and had no adverse effects, it would be wonderful.

The cesarean-delivery case is different, as healing in a pregnant patient is 1 concern; the size of the uterus also has an effect. If the patient is breastfeeding, you would want to make sure the gel didn’t interfere.

PAGIDAS: We desperately need a product that can minimize adhesions regardless of the route of access or type of procedure. Even though we lack data on outcomes, I predict wide use of such a product, assuming it is nontoxic and effective.

DECHERNEY: What if it costs $1,000 a case?

SANFILIPPO: If it prevents 1 bowel obstruction, it still would be cost-effective.

DECHERNEY: The incidence of bowel obstruction for total abdominal hysterectomy is 2%, and 5% for radical hysterectomy.

HURD: We must be careful of the cost-benefit ratio. Bowel obstruction after gynecologic surgery is uncommon.

Is the gel worth $100? $1,000? $3,000? It’s difficult to say, but the more expensive it is, the less likely it will find widespread use.

PAGIDAS: I agree. We should remember that we still need to maintain microsurgical techniques and appropriate tissue handling, as well as avoid ischemia and infection.

Looking for the magic bullet

DECHERNEY: What is the future of adjunctive therapy?

SANFILIPPO: I would focus on noxythiolin; it has potential. Calcium channel blockers for adhesion prevention have also been studied.²⁰ In 1 investigation involving a rat model, the calcium channel blocker verapamil as well as several other agents—including vitamin E, carboxymethylcellulose, cyclosporin, aprotinin, and tenoxicam—were compared with respect to tissue effects. A beneficial effect was noted with all agents except cyclosporin and carboxymethylcellulose.

Whoever succeeds in manufacturing an effective preventive will be a winner.

HURD: There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery. The most effective agent would be applied intraabdominally, since any systemic agent that stops adhesion formation would probably decrease wound healing as well.

 

I hope the most effective agents can be used in both laparoscopy and laparotomy, and that they will decrease the adhesion-formation rate by more than 50%. We need to find the magic bullet that can cover the entire pelvis—if not the entire abdomen.

Dr. DeCherney reports small holdings with Lifecore Biomedical and Johnson & Johnson. Drs. Hurd and Pagidas report no financial relationship with any companies whose products are mentioned in this article. Dr. Sanfilippo serves on the speaker’s bureau for Berlex, Ortho Pharmaceuticals, and Wyeth, and receives grant support from Eli Lilly and Wyeth.

Are adhesions a pathologic response to injury or a normal aspect of healing? Can they be avoided, or are preventive efforts part of the problem? How useful are the different barriers in gynecologic surgery? What is the ideal adjuvant?

OBG Management convened a panel of experts to explore these and other questions.

Common problem, high recurrence rate

DECHERNEY: Adhesion formation is serious because it is associated with clinical entities such as infertility, pelvic pain, and bowel obstruction. We all agree that approximately 40% of people who undergo primary surgery develop adhesions and that 80% to 90% of patients who undergo lysis develop recurrent adhesions.

SANFILIPPO: One study several years ago explored adhesion formation.¹ Unfortunately, no matter how meticulous the surgeon is, adhesions will form, even with microsurgical techniques and carefully ensured hemostasis.

HURD: The number of patients with significant adhesion formation after some gynecologic procedures has been reported to be greater than 90%.²

DECHERNEY: That higher incidence usually occurs after general surgery—and there’s a reason it is so high: General surgeons don’t use adjunctive therapy. They are critical of it. It is to our credit as gynecologic surgeons that we adopted adjunctive therapies about 15 years ago with the introduction of dextran 70 (Hyskon; Medisan Pharmaceuticals, Parsippany, NJ).

PAGIDAS: If anything, the pelvis seems to have even more of a predilection for adhesion formation than the abdomen, probably because of the close proximity of structures.

How and why adhesions form

DECHERNEY: What is the pathophysiology of adhesion formation? Let’s say you have 2 raw surface areas. What happens?

The process

PAGIDAS: The increase in leukotrienes and prostaglandins and the decrease in plasminogen activity (which actually initiates the inflammation) appear to be significant.

HURD: Vessel permeability also increases, and inflammatory cells leak through the vessels and set up a matrix for adhesion formation.

DECHERNEY: So we have 2 raw surface areas with fibrin leaking out and forming bridges between them.

PAGIDAS: The key is that it takes 2 surfaces to form these bridges. As I mentioned, the greater proximity of pelvic structures—particularly around the tube and ovary—probably contributes to adhesion formation.

DECHERNEY: Macrophage activity also is important. The macrophage “migrates” along these fibrin bridges and lays down collagen over a period of time. Then the collagen becomes organized and, eventually, vascular.

Window of opportunity

SANFILIPPO: Adhesion formation probably occurs and is pretty well established within 5 to 7 days of the precipitating event—usually surgery. Once that process is under way, attempts to halt it yield diminishing returns. Unfortunately, we don’t know how to interfere with it in a positive way.

HURD: Under normal conditions, there seems to be a balance between fibrin deposition and fibrinolysis. In some tissues, however, these functions become imbalanced. This disparity may contribute more to adhesions than the actual laying down of fibrin—especially in tissue that is hypoxic.

DECHERNEY: Would you say that adhesion formation represents normal or abnormal healing?

HURD: It is one of the body’s normal protective mechanisms and an important part of healing. Without it, any abdominal injury would likely result in death.

SANFILIPPO: I don’t think it differs that much from processes that occur externally. For example, if you get cut deeply enough, you develop a scar. Is that scar part of the normal healing process? It is.

PAGIDAS: Right. It is a normal process of tissue remodeling. The question is: What allows it to go astray?

What surgical techniques help prevent adhesions?

DECHERNEY: Let’s review the aspects of surgical technique that are important for adhesion prevention.

 

PAGIDAS: I emphasize the value of microsurgical techniques, which help to minimize severe tissue handling. It also is important to keep surfaces moistened so they don’t desiccate.

SANFILIPPO: I agree with Dr. Pagidas about microsurgical techniques such as gentle tissue handling, careful hemostasis, and keeping tissues moist. If we follow these principles, we create an environment that minimizes the potential for adhesion formation.

HURD: The findings of many well-controlled animal studies have been surprising. For example, it is difficult to demonstrate that drying of tissue increases adhesions.³ Probably the greatest contributor to adhesions in these models was abrasion.⁴ One way that laparoscopic surgery decreases adhesions is by avoiding abrasion of the bowel mucosa, which occurs specifically with packing.

In open cases, one thing we can do to minimize the risk of adhesions is to pack gently when needed. Also, we should avoid using packing to reposition the bowel.

Another factor frequently overlooked is the application of heat, which appears to be a very effective way to create adhesions. This probably isn’t an issue for laparoscopic cases, but when you use irrigation fluid in an open case, watch the temperature. If it feels hot to you, you need to worry about potential injury to the bowel surfaces.

PAGIDAS: That is critical. In abdominal cases you want to make sure irrigation fluid is warm, but not too warm, because heat increases the vascular permeability of vessels and leads to more macrophages, more prostaglandins, and more leukotrienes.

HURD: Another important element is the type of suture material used.

DECHERNEY: Overall, we need to minimize the use of sutures. For example, when I am operating laparoscopically on an ovarian cyst, I try to apply bipolar energy to the edges so that they will coapt without a stitch.

HURD: When it first became clear that suturing ovaries increased adhesion formation, we conducted a controlled trial of different kinds of sutures in animals. Not surprisingly, we found that the less reactive the suture, the fewer adhesions.⁵ Sutures that are absorbed more slowly, such as polydioxanone, seem to be less reactive.

Obviously, inert sutures like nylon are the least reactive, but they are permanent. It is assumed that animal-protein sutures such as chromic and plain gut are the most reactive, although I am not sure there are sufficient data to support that conclusion.

 

Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why. It may be related to decreased suturing.

DECHERNEY: Bulk is important, too—that is, the number of throws in the suture. When Vicryl (polyglactin 910) became available, we conducted a study in mice using proportionately small Vicryl plaques to determine whether this would be good a barrier (A. DeCherney, MD, unpublished data). It caused a tremendous amount of adhesion because so much foreign matter was applied.

We also did a study using human-size titanium clips in rats (A. DeCherney, MD, unpublished data). Not surprisingly, there was a lot of adhesion formation.

FIGURE 1 Pelvic adhesions: How they develop, problems they cause

Adhesions occur when 2 or more raw surfaces are exposed to leaking fibrin, which forms a bridge between the surfaces. Macrophages “migrate” along these bridges, depositing collagen.

The pelvis has a greater predilection for adhesions than the abdomen because of the close proximity of structures.

Although adhesions represent one of the body’s protective mechanisms, they may also cause pain or interfere with fertility, bowel function, or other processes.

Laparoscopy versus laparotomy: More adhesions in open cases?

DECHERNEY: Based on all the techniques we have learned from microsurgery—with the exception of magnification—it appears that laparoscopic procedures are less likely to cause adhesions than laparotomy. Do you agree?

PAGIDAS: I think so. As Dr. Hurd noted, a main reason is the diminished tissue handling, because there is no packing.

HURD: Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why.^6,7 It may be related to decreased suturing.

DECHERNEY: Less bleeding occurs because surgeons are less aggressive laparoscopically than in laparotomy.

I have stopped doing difficult cases laparoscopically. For example, it is rare for me to operate laparoscopically on a patient with stage IV endometriosis, at least when it comes to infertility—I might consider laparoscopy for pain.

I think case selection plays a role as well, although there are few data to back that up. It is purely clinical opinion.

 

SANFILIPPO: We need a well-designed prospective study to explore the effects of laparotomy versus laparoscopy. Existing data are not clear. You would assume laparoscopy would be associated with less adhesion formation. But genetic or other factors may explain why patient A is more prone to adhesions than patient B.

Does anybody think carbon dioxide plays a role in adhesion formation?

HURD: In the laboratory, carbon dioxide increases cell growth.⁸ Without an increased carbon dioxide concentration in the atmosphere, cell cultures don’t grow well. This might suggest that the carbon dioxide used for laparoscopy could actually enhance adhesion growth. Fortunately, this does not appear to be the case clinically.

With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision.

DECHERNEY: What about second-look laparoscopies? Do any of you perform them after a patient has undergone lysis of adhesions?

SANFILIPPO: Only as part of a research protocol. It amazes me how rapidly adhesions can form and how dense they are 2 or more weeks after the initial laparoscopic surgery.

PAGIDAS: We tend to limit second-look laparoscopy to a research protocol, although it is sometimes valuable after laparoscopic or abdominal myomectomy, which has the highest incidence of adhesions. If the surgeon can perform a second look and lyse adhesions, he or she may potentially alter the reproductive outcome. However, with assisted reproductive techniques becoming integral to every infertility case, that approach has begun to go out of style.

SANFILIPPO: That’s a good point. With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision. If adhesions do occur, it is best if they occur toward the bladder rather than in the area of the tubes and ovaries.

HURD: For second-look laparoscopy, we must keep in mind the cost and the small but real risks of surgery. Until good controlled studies show a reasonable clinical advantage, this approach probably should remain a research protocol.

SANFILIPPO: The literature suggests it is helpful, but does not help fertility, so second-look laparoscopy is used mainly to evaluate adjunctive therapies. I don’t think anybody uses it as a primary therapy anymore.

Bowel obstruction still a risk, though rarely seen by Ob/Gyns

DECHERNEY: Although bowel obstruction is fairly common, Ob/Gyns do not often encounter it because it occurs relatively distant from the index surgery. Even though a patient may not experience bowel obstruction in the first year, an obstruction related to the index surgery is just as likely to develop 20 years later as 2 years later. These patients usually are treated by general surgeons. Still, we should beware of the potential for bowel obstruction and take steps to prevent it, if at all possible. Do you agree?

PAGIDAS: Yes. We tend to forget about bowel obstruction because we rarely follow patients past pregnancy or the first trimester if they are seeking infertility treatment.

HURD: The primary problem seems to be the abdominal wall incision. Fortunately, cesarean section seems to carry a decreased risk of abdominal wall adhesions, probably because the uterus serves as a splint over the incision.

The Pfannenstiel incision also appears to have some advantage. Both human and animal models suggest little advantage or disadvantage when peritoneal closure is compared to nonclosure.^9,10

SANFILIPPO: I’m curious about how the panelists manage loose clips. If you are using an EndoGIA (US Surgical, Norwalk, Conn) or other stapling device and you have free-floating clips, do you make a concerted effort to find them? In some cases, they have been implicated in bowel adhesion and obstruction. I try to retrieve loose clips, whether open or closed.

HURD: The advantage of those devices is minimal tissue damage, and the clips are inert. In general, inert, nonreactive clips have not been implicated as much in adhesion formation. I retrieve them if I see them, but I don’t search them out.

PAGIDAS: I do the same. If the clips are visible, I remove them. But I would not repack the bowel or do anything more heroic than look in locations where they might be.

What drugs may inhibit inflammatory response?

DECHERNEY: What about use of pharmacologic agents to prevent adhesions? Is there reason to think research should focus on inhibiting the inflammatory response? How important is polymorphonuclear cell infiltration?

Cyclooxygenase (COX) 2 agents could be helpful for inhibition of platelet function, since they are low in side effects. Thus, high doses of these drugs might be effective. At one time, aspirin was proposed, but you’d have to give a human so much aspirin that her ears would ring.

 

PAGIDAS: Pharmacologic agents have a role, especially for dampening the inflammatory immune response. But you don’t want to dampen it completely because, as we observed, it is an important part of healing. The difficulty is finding a balance between allowing the tissue to heal and preventing adhesions.

HURD: We studied the ability of a water-soluble prostaglandin inhibitor to prevent postoperative adhesion formation. Like many other agents, we found only a partial response.¹¹

DECHERNEY: With current options, the best you can aim for is a 50% reduction.

Adjunctive therapy likely to limit adhesion rate Hydroflotation

DECHERNEY: The original adjunctive therapy was 20 mg dexamethasone and 25 mg femergin in 200 cc of Ringer’s lactate, with an equivalent amount of dexamethasone and femergin every 4 hours for a total of 6 doses. I prescribed that regimen because I was trained to do so. I stopped after it became clear that hydroflotation from the fluid—not the medication itself—was responsible for the improvement.

I must admit I gave it up reluctantly; patients felt fabulous with those higher load doses of glucocorticoids after surgery.

Do any of you use crystalloids as adjunctive therapy?

SANFILIPPO: In the animal model, they are so rapidly absorbed that they aren’t effective. I was a strong advocate, but now I don’t use them at all.

HURD: A lot depends on the kind of case. For instance, at the end of an open myomectomy, the patient often is oozing, so you want to use a barrier that blood won’t affect.

For ovarian surgery, you might want to specifically target the ovaries with some kind of coverage. But when you are doing a broad lysis of adhesions, you have few choices to cover the pelvis. In those cases I use hydroflotation with Ringer’s lactate. Both human and animal studies have shown some benefit in preventing adhesions, and it appears to have little risk.^12,13

It’s better than nothing, in my opinion.

DECHERNEY: Do you use dextran 70 or crystalloids?

HURD: I use Ringer’s lactate solution. I was trained in the dextran 70 era, and there were certain problems with that approach. Since studies have failed to show a consistent effect of dextran 70, I no longer use this solution.^14,15

DECHERNEY: Another problem with crystalloids is that they leak, which is disconcerting to the patient.

HURD: They also can mask an injury to the bladder in difficult cases.

DECHERNEY: I agree that dextran 70 is only appropriate in certain cases, but it is a good hydroflotation agent. Every cubic centimeter of dextran 70 brings in 1.2 cc of transudate, so it hangs around for at least 4 days.

It is appropriate only for certain surface areas—mainly the cul-de-sac. It is harmful on raw surface areas on the lateral pelvic sidewall because it tends to push the ovary and tube to those areas. Unless you are doing a lot of work in the deep pelvis, I would avoid dextran 70.

In addition, there have been allergic reactions, most of which seem to occur in patients with fluid overload; a lot of the dextran 70 is absorbed.

 

With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal.

SANFILIPPO: Dextran 70 is not recommended for patients with sugar beet allergy, either.

We completed a study in a rabbit model, in which the peritoneal cavity was lavaged with either chlorhexidine or iodine.¹⁶ At the time of second-look surgery, the rate of adhesion formation was decreased, especially with the iodine preparation. I would hope that this has potential in humans.

DECHERNEY: In your lavage procedures to prevent adhesions, do any of you use heparin?

HURD: No.

PAGIDAS: I don’t think any evidence suggests that local administration changes the outcome.

DECHERNEY: I agree. When heparin has been used, the doses have been so low that it was not terribly helpful. And when you consider that hemorrhage can be a problem, heparin is probably deleterious rather than helpful.

Barriers

DECHERNEY: What about barriers? The first to become available, Interceed (Gynecare, a division of Ethicon, Somerville, NJ), is oxidized cellulose, similar to Surgicell (Johnson & Johnson, New Brunswick, NJ). Since it gelates quickly, there is no fenestration, so the fibrin is unable to penetrate. However, if the patient has bleeding by capillary action, the raw surface just moves from one side of the Interceed barrier to the other.

What has been your experience? Do you use it?

PAGIDAS: I do not use Interceed, although prospective randomized trials and a meta-analysis confirmed its benefits in de novo formation and reformation.^17-19 I don’t use it because it requires complete hemostasis. Also, with the surfaces we work on—notably, the ovary and tube—it is difficult to apply to just 1 surface area. From a clinical perspective, I appreciate the data, but it is hard to ensure a good application to optimize its effectiveness.

SANFILIPPO: I use Interceed, but I agree with you about its limitations. Meticulous hemostasis is a prerequisite.

HURD: With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal. If those patients have dense adhesions of the ovary or the sidewall, I generally leave them alone, and I try to avoid putting Interceed around the ovaries. No study has shown that using Interceed improves pregnancy.

In contrast, when a chronic pain patient’s ovaries are densely adherent to the cul-de-sac, which appears to be highly associated with dyspareunia, I lyse the adhesions, achieve meticulous hemostasis, and then use Interceed. It is hard to demonstrate in a study that this approach decreases the chance of pain. Even so, it certainly does decrease the chance of the ovaries being adherent.

DECHERNEY: One issue with Interceed is that we don’t know what happens to it once the abdomen is closed. It may migrate significantly.

Psychological issues may also be involved. For example, patients with multiple somatic complaints may be less likely to benefit from lysis of adhesions.

PAGIDAS: Right. Interestingly, a meta-analysis of all the randomized trials involving mechanical barriers found no correlation to pregnancy outcome or pelvic pain.¹⁸ If we were to consider new trials, the psychological aspect would be worth looking into.

Seprafilm

DECHERNEY: Let’s move on to Seprafilm (Genzyme, Cambridge, Mass). What is it and how useful is it?

HURD: Seprafilm is modified hyaluronic acid, which forms a brittle, thin plastic layer. It is somewhat difficult to work with but, once it is in place, seems to adhere well. The presence of blood does not appear to be a problem, since the Seprafilm forms an impermeable barrier—unless it breaks. I have found it especially useful in myomectomies, which produce postoperative oozing through the incisions no matter how hard you try to prevent it.

In addition, in open cases, surfaces can easily be covered with this material. Unfortunately, it can’t be used laparoscopically because it is so brittle.

 

DECHERNEY: In my opinion, that is its major drawback.

PAGIDAS: In cardiac surgery, Seprafilm appears to work quite effectively.

SANFILIPPO: The manufacturer initially focused on surgeons in the context of sigmoid colon surgery, and it seems to work well in that setting.

Intergel

DECHERNEY: That brings us to the current state of the art: gels. I’m sure you all are familiar with Intergel (Lifecore Biomedical, Chaska, Minn), which is a ferrous derivative of hyaluronic acid that works by coating the raw surface areas. It also has the theoretical advantage of ease of use. Have any of you used Intergel?

HURD: As you know it was only recently approved by the US Food and Drug Administration, but not for laparoscopic use. It may work best on abraded bowel, which is avoided by laparoscopic surgery.

As you are probably also aware, the manufacturer recently took it off the market because of unusual side effects, namely a chemical peritonitis. Although peritonitis was cited as being rare, we encountered it in probably half the patients we operated on.

PAGIDAS: When we used it on hospitalized patients, the peritonitis wasn’t that obvious, since there was an expectation of significant pain. However, when we used Intergel on short-stay patients, we had to readmit them and do a full workup because we were concerned about bowel perforations. I’m surprised the manufacturer didn’t take it off the market sooner.

DECHERNEY: It seems strange, since Intergel has been used in Europe for a while now. I’ve used it in only 1 case and didn’t have adverse effects. It seemed to work well.

SANFILIPPO: It had all the right ingredients for success. It is unfortunate that these side effects have prohibited its use.

There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery.

Gels and the cost factor

DECHERNEY: Other gels are in the pipeline. I’m reminded of plasminogen activator, which is a powerful antiadhesive agent that lyses fibrin effectively. Unfortunately, it is prohibitively expensive.

The next phase likely will involve the so-called polymers. If you spray them on your hand, they are activated by light or another chemical and become a cellophane-like substance. The problem is viscosity. If sprayed on the sidewall, for instance, they run halfway down before they are activated, so the entire surface does not get covered.

PAGIDAS: One concern with polymers is that they could actually bring surfaces together when they polymerize. We still have a lot to learn.

DECHERNEY: Let’s say a new gel comes on the market that takes reformation adhesions from 90% to 10%, as opposed to 40% recurrence. Would you use it in all 4 of the cases we discuss here?

HURD: If it was that effective and had no adverse effects, it would be wonderful.

The cesarean-delivery case is different, as healing in a pregnant patient is 1 concern; the size of the uterus also has an effect. If the patient is breastfeeding, you would want to make sure the gel didn’t interfere.

PAGIDAS: We desperately need a product that can minimize adhesions regardless of the route of access or type of procedure. Even though we lack data on outcomes, I predict wide use of such a product, assuming it is nontoxic and effective.

DECHERNEY: What if it costs $1,000 a case?

SANFILIPPO: If it prevents 1 bowel obstruction, it still would be cost-effective.

DECHERNEY: The incidence of bowel obstruction for total abdominal hysterectomy is 2%, and 5% for radical hysterectomy.

HURD: We must be careful of the cost-benefit ratio. Bowel obstruction after gynecologic surgery is uncommon.

Is the gel worth $100? $1,000? $3,000? It’s difficult to say, but the more expensive it is, the less likely it will find widespread use.

PAGIDAS: I agree. We should remember that we still need to maintain microsurgical techniques and appropriate tissue handling, as well as avoid ischemia and infection.

Looking for the magic bullet

DECHERNEY: What is the future of adjunctive therapy?

SANFILIPPO: I would focus on noxythiolin; it has potential. Calcium channel blockers for adhesion prevention have also been studied.²⁰ In 1 investigation involving a rat model, the calcium channel blocker verapamil as well as several other agents—including vitamin E, carboxymethylcellulose, cyclosporin, aprotinin, and tenoxicam—were compared with respect to tissue effects. A beneficial effect was noted with all agents except cyclosporin and carboxymethylcellulose.

Whoever succeeds in manufacturing an effective preventive will be a winner.

HURD: There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery. The most effective agent would be applied intraabdominally, since any systemic agent that stops adhesion formation would probably decrease wound healing as well.

 

I hope the most effective agents can be used in both laparoscopy and laparotomy, and that they will decrease the adhesion-formation rate by more than 50%. We need to find the magic bullet that can cover the entire pelvis—if not the entire abdomen.

Dr. DeCherney reports small holdings with Lifecore Biomedical and Johnson & Johnson. Drs. Hurd and Pagidas report no financial relationship with any companies whose products are mentioned in this article. Dr. Sanfilippo serves on the speaker’s bureau for Berlex, Ortho Pharmaceuticals, and Wyeth, and receives grant support from Eli Lilly and Wyeth.

Are adhesions a pathologic response to injury or a normal aspect of healing? Can they be avoided, or are preventive efforts part of the problem? How useful are the different barriers in gynecologic surgery? What is the ideal adjuvant?

OBG Management convened a panel of experts to explore these and other questions.

Common problem, high recurrence rate

DECHERNEY: Adhesion formation is serious because it is associated with clinical entities such as infertility, pelvic pain, and bowel obstruction. We all agree that approximately 40% of people who undergo primary surgery develop adhesions and that 80% to 90% of patients who undergo lysis develop recurrent adhesions.

SANFILIPPO: One study several years ago explored adhesion formation.¹ Unfortunately, no matter how meticulous the surgeon is, adhesions will form, even with microsurgical techniques and carefully ensured hemostasis.

HURD: The number of patients with significant adhesion formation after some gynecologic procedures has been reported to be greater than 90%.²

DECHERNEY: That higher incidence usually occurs after general surgery—and there’s a reason it is so high: General surgeons don’t use adjunctive therapy. They are critical of it. It is to our credit as gynecologic surgeons that we adopted adjunctive therapies about 15 years ago with the introduction of dextran 70 (Hyskon; Medisan Pharmaceuticals, Parsippany, NJ).

PAGIDAS: If anything, the pelvis seems to have even more of a predilection for adhesion formation than the abdomen, probably because of the close proximity of structures.

How and why adhesions form

DECHERNEY: What is the pathophysiology of adhesion formation? Let’s say you have 2 raw surface areas. What happens?

The process

PAGIDAS: The increase in leukotrienes and prostaglandins and the decrease in plasminogen activity (which actually initiates the inflammation) appear to be significant.

HURD: Vessel permeability also increases, and inflammatory cells leak through the vessels and set up a matrix for adhesion formation.

DECHERNEY: So we have 2 raw surface areas with fibrin leaking out and forming bridges between them.

PAGIDAS: The key is that it takes 2 surfaces to form these bridges. As I mentioned, the greater proximity of pelvic structures—particularly around the tube and ovary—probably contributes to adhesion formation.

DECHERNEY: Macrophage activity also is important. The macrophage “migrates” along these fibrin bridges and lays down collagen over a period of time. Then the collagen becomes organized and, eventually, vascular.

Window of opportunity

SANFILIPPO: Adhesion formation probably occurs and is pretty well established within 5 to 7 days of the precipitating event—usually surgery. Once that process is under way, attempts to halt it yield diminishing returns. Unfortunately, we don’t know how to interfere with it in a positive way.

HURD: Under normal conditions, there seems to be a balance between fibrin deposition and fibrinolysis. In some tissues, however, these functions become imbalanced. This disparity may contribute more to adhesions than the actual laying down of fibrin—especially in tissue that is hypoxic.

DECHERNEY: Would you say that adhesion formation represents normal or abnormal healing?

HURD: It is one of the body’s normal protective mechanisms and an important part of healing. Without it, any abdominal injury would likely result in death.

SANFILIPPO: I don’t think it differs that much from processes that occur externally. For example, if you get cut deeply enough, you develop a scar. Is that scar part of the normal healing process? It is.

PAGIDAS: Right. It is a normal process of tissue remodeling. The question is: What allows it to go astray?

What surgical techniques help prevent adhesions?

DECHERNEY: Let’s review the aspects of surgical technique that are important for adhesion prevention.

 

PAGIDAS: I emphasize the value of microsurgical techniques, which help to minimize severe tissue handling. It also is important to keep surfaces moistened so they don’t desiccate.

SANFILIPPO: I agree with Dr. Pagidas about microsurgical techniques such as gentle tissue handling, careful hemostasis, and keeping tissues moist. If we follow these principles, we create an environment that minimizes the potential for adhesion formation.

HURD: The findings of many well-controlled animal studies have been surprising. For example, it is difficult to demonstrate that drying of tissue increases adhesions.³ Probably the greatest contributor to adhesions in these models was abrasion.⁴ One way that laparoscopic surgery decreases adhesions is by avoiding abrasion of the bowel mucosa, which occurs specifically with packing.

In open cases, one thing we can do to minimize the risk of adhesions is to pack gently when needed. Also, we should avoid using packing to reposition the bowel.

Another factor frequently overlooked is the application of heat, which appears to be a very effective way to create adhesions. This probably isn’t an issue for laparoscopic cases, but when you use irrigation fluid in an open case, watch the temperature. If it feels hot to you, you need to worry about potential injury to the bowel surfaces.

PAGIDAS: That is critical. In abdominal cases you want to make sure irrigation fluid is warm, but not too warm, because heat increases the vascular permeability of vessels and leads to more macrophages, more prostaglandins, and more leukotrienes.

HURD: Another important element is the type of suture material used.

DECHERNEY: Overall, we need to minimize the use of sutures. For example, when I am operating laparoscopically on an ovarian cyst, I try to apply bipolar energy to the edges so that they will coapt without a stitch.

HURD: When it first became clear that suturing ovaries increased adhesion formation, we conducted a controlled trial of different kinds of sutures in animals. Not surprisingly, we found that the less reactive the suture, the fewer adhesions.⁵ Sutures that are absorbed more slowly, such as polydioxanone, seem to be less reactive.

Obviously, inert sutures like nylon are the least reactive, but they are permanent. It is assumed that animal-protein sutures such as chromic and plain gut are the most reactive, although I am not sure there are sufficient data to support that conclusion.

 

Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why. It may be related to decreased suturing.

DECHERNEY: Bulk is important, too—that is, the number of throws in the suture. When Vicryl (polyglactin 910) became available, we conducted a study in mice using proportionately small Vicryl plaques to determine whether this would be good a barrier (A. DeCherney, MD, unpublished data). It caused a tremendous amount of adhesion because so much foreign matter was applied.

We also did a study using human-size titanium clips in rats (A. DeCherney, MD, unpublished data). Not surprisingly, there was a lot of adhesion formation.

FIGURE 1 Pelvic adhesions: How they develop, problems they cause

Adhesions occur when 2 or more raw surfaces are exposed to leaking fibrin, which forms a bridge between the surfaces. Macrophages “migrate” along these bridges, depositing collagen.

The pelvis has a greater predilection for adhesions than the abdomen because of the close proximity of structures.

Although adhesions represent one of the body’s protective mechanisms, they may also cause pain or interfere with fertility, bowel function, or other processes.

Laparoscopy versus laparotomy: More adhesions in open cases?

DECHERNEY: Based on all the techniques we have learned from microsurgery—with the exception of magnification—it appears that laparoscopic procedures are less likely to cause adhesions than laparotomy. Do you agree?

PAGIDAS: I think so. As Dr. Hurd noted, a main reason is the diminished tissue handling, because there is no packing.

HURD: Multiple clinical studies have shown laparoscopy to be associated with a lower adhesion rate, although it isn’t clear why.^6,7 It may be related to decreased suturing.

DECHERNEY: Less bleeding occurs because surgeons are less aggressive laparoscopically than in laparotomy.

I have stopped doing difficult cases laparoscopically. For example, it is rare for me to operate laparoscopically on a patient with stage IV endometriosis, at least when it comes to infertility—I might consider laparoscopy for pain.

I think case selection plays a role as well, although there are few data to back that up. It is purely clinical opinion.

 

SANFILIPPO: We need a well-designed prospective study to explore the effects of laparotomy versus laparoscopy. Existing data are not clear. You would assume laparoscopy would be associated with less adhesion formation. But genetic or other factors may explain why patient A is more prone to adhesions than patient B.

Does anybody think carbon dioxide plays a role in adhesion formation?

HURD: In the laboratory, carbon dioxide increases cell growth.⁸ Without an increased carbon dioxide concentration in the atmosphere, cell cultures don’t grow well. This might suggest that the carbon dioxide used for laparoscopy could actually enhance adhesion growth. Fortunately, this does not appear to be the case clinically.

With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision.

DECHERNEY: What about second-look laparoscopies? Do any of you perform them after a patient has undergone lysis of adhesions?

SANFILIPPO: Only as part of a research protocol. It amazes me how rapidly adhesions can form and how dense they are 2 or more weeks after the initial laparoscopic surgery.

PAGIDAS: We tend to limit second-look laparoscopy to a research protocol, although it is sometimes valuable after laparoscopic or abdominal myomectomy, which has the highest incidence of adhesions. If the surgeon can perform a second look and lyse adhesions, he or she may potentially alter the reproductive outcome. However, with assisted reproductive techniques becoming integral to every infertility case, that approach has begun to go out of style.

SANFILIPPO: That’s a good point. With myomectomy, the surgeon needs to plan ahead to maximize the number of myomas removed from a single incision. If adhesions do occur, it is best if they occur toward the bladder rather than in the area of the tubes and ovaries.

HURD: For second-look laparoscopy, we must keep in mind the cost and the small but real risks of surgery. Until good controlled studies show a reasonable clinical advantage, this approach probably should remain a research protocol.

SANFILIPPO: The literature suggests it is helpful, but does not help fertility, so second-look laparoscopy is used mainly to evaluate adjunctive therapies. I don’t think anybody uses it as a primary therapy anymore.

Bowel obstruction still a risk, though rarely seen by Ob/Gyns

DECHERNEY: Although bowel obstruction is fairly common, Ob/Gyns do not often encounter it because it occurs relatively distant from the index surgery. Even though a patient may not experience bowel obstruction in the first year, an obstruction related to the index surgery is just as likely to develop 20 years later as 2 years later. These patients usually are treated by general surgeons. Still, we should beware of the potential for bowel obstruction and take steps to prevent it, if at all possible. Do you agree?

PAGIDAS: Yes. We tend to forget about bowel obstruction because we rarely follow patients past pregnancy or the first trimester if they are seeking infertility treatment.

HURD: The primary problem seems to be the abdominal wall incision. Fortunately, cesarean section seems to carry a decreased risk of abdominal wall adhesions, probably because the uterus serves as a splint over the incision.

The Pfannenstiel incision also appears to have some advantage. Both human and animal models suggest little advantage or disadvantage when peritoneal closure is compared to nonclosure.^9,10

SANFILIPPO: I’m curious about how the panelists manage loose clips. If you are using an EndoGIA (US Surgical, Norwalk, Conn) or other stapling device and you have free-floating clips, do you make a concerted effort to find them? In some cases, they have been implicated in bowel adhesion and obstruction. I try to retrieve loose clips, whether open or closed.

HURD: The advantage of those devices is minimal tissue damage, and the clips are inert. In general, inert, nonreactive clips have not been implicated as much in adhesion formation. I retrieve them if I see them, but I don’t search them out.

PAGIDAS: I do the same. If the clips are visible, I remove them. But I would not repack the bowel or do anything more heroic than look in locations where they might be.

What drugs may inhibit inflammatory response?

DECHERNEY: What about use of pharmacologic agents to prevent adhesions? Is there reason to think research should focus on inhibiting the inflammatory response? How important is polymorphonuclear cell infiltration?

Cyclooxygenase (COX) 2 agents could be helpful for inhibition of platelet function, since they are low in side effects. Thus, high doses of these drugs might be effective. At one time, aspirin was proposed, but you’d have to give a human so much aspirin that her ears would ring.

 

PAGIDAS: Pharmacologic agents have a role, especially for dampening the inflammatory immune response. But you don’t want to dampen it completely because, as we observed, it is an important part of healing. The difficulty is finding a balance between allowing the tissue to heal and preventing adhesions.

HURD: We studied the ability of a water-soluble prostaglandin inhibitor to prevent postoperative adhesion formation. Like many other agents, we found only a partial response.¹¹

DECHERNEY: With current options, the best you can aim for is a 50% reduction.

Adjunctive therapy likely to limit adhesion rate Hydroflotation

DECHERNEY: The original adjunctive therapy was 20 mg dexamethasone and 25 mg femergin in 200 cc of Ringer’s lactate, with an equivalent amount of dexamethasone and femergin every 4 hours for a total of 6 doses. I prescribed that regimen because I was trained to do so. I stopped after it became clear that hydroflotation from the fluid—not the medication itself—was responsible for the improvement.

I must admit I gave it up reluctantly; patients felt fabulous with those higher load doses of glucocorticoids after surgery.

Do any of you use crystalloids as adjunctive therapy?

SANFILIPPO: In the animal model, they are so rapidly absorbed that they aren’t effective. I was a strong advocate, but now I don’t use them at all.

HURD: A lot depends on the kind of case. For instance, at the end of an open myomectomy, the patient often is oozing, so you want to use a barrier that blood won’t affect.

For ovarian surgery, you might want to specifically target the ovaries with some kind of coverage. But when you are doing a broad lysis of adhesions, you have few choices to cover the pelvis. In those cases I use hydroflotation with Ringer’s lactate. Both human and animal studies have shown some benefit in preventing adhesions, and it appears to have little risk.^12,13

It’s better than nothing, in my opinion.

DECHERNEY: Do you use dextran 70 or crystalloids?

HURD: I use Ringer’s lactate solution. I was trained in the dextran 70 era, and there were certain problems with that approach. Since studies have failed to show a consistent effect of dextran 70, I no longer use this solution.^14,15

DECHERNEY: Another problem with crystalloids is that they leak, which is disconcerting to the patient.

HURD: They also can mask an injury to the bladder in difficult cases.

DECHERNEY: I agree that dextran 70 is only appropriate in certain cases, but it is a good hydroflotation agent. Every cubic centimeter of dextran 70 brings in 1.2 cc of transudate, so it hangs around for at least 4 days.

It is appropriate only for certain surface areas—mainly the cul-de-sac. It is harmful on raw surface areas on the lateral pelvic sidewall because it tends to push the ovary and tube to those areas. Unless you are doing a lot of work in the deep pelvis, I would avoid dextran 70.

In addition, there have been allergic reactions, most of which seem to occur in patients with fluid overload; a lot of the dextran 70 is absorbed.

 

With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal.

SANFILIPPO: Dextran 70 is not recommended for patients with sugar beet allergy, either.

We completed a study in a rabbit model, in which the peritoneal cavity was lavaged with either chlorhexidine or iodine.¹⁶ At the time of second-look surgery, the rate of adhesion formation was decreased, especially with the iodine preparation. I would hope that this has potential in humans.

DECHERNEY: In your lavage procedures to prevent adhesions, do any of you use heparin?

HURD: No.

PAGIDAS: I don’t think any evidence suggests that local administration changes the outcome.

DECHERNEY: I agree. When heparin has been used, the doses have been so low that it was not terribly helpful. And when you consider that hemorrhage can be a problem, heparin is probably deleterious rather than helpful.

Barriers

DECHERNEY: What about barriers? The first to become available, Interceed (Gynecare, a division of Ethicon, Somerville, NJ), is oxidized cellulose, similar to Surgicell (Johnson & Johnson, New Brunswick, NJ). Since it gelates quickly, there is no fenestration, so the fibrin is unable to penetrate. However, if the patient has bleeding by capillary action, the raw surface just moves from one side of the Interceed barrier to the other.

What has been your experience? Do you use it?

PAGIDAS: I do not use Interceed, although prospective randomized trials and a meta-analysis confirmed its benefits in de novo formation and reformation.^17-19 I don’t use it because it requires complete hemostasis. Also, with the surfaces we work on—notably, the ovary and tube—it is difficult to apply to just 1 surface area. From a clinical perspective, I appreciate the data, but it is hard to ensure a good application to optimize its effectiveness.

SANFILIPPO: I use Interceed, but I agree with you about its limitations. Meticulous hemostasis is a prerequisite.

HURD: With infertility patients, even if you lyse dense adhesions, you do not render the ovarian surface normal. If those patients have dense adhesions of the ovary or the sidewall, I generally leave them alone, and I try to avoid putting Interceed around the ovaries. No study has shown that using Interceed improves pregnancy.

In contrast, when a chronic pain patient’s ovaries are densely adherent to the cul-de-sac, which appears to be highly associated with dyspareunia, I lyse the adhesions, achieve meticulous hemostasis, and then use Interceed. It is hard to demonstrate in a study that this approach decreases the chance of pain. Even so, it certainly does decrease the chance of the ovaries being adherent.

DECHERNEY: One issue with Interceed is that we don’t know what happens to it once the abdomen is closed. It may migrate significantly.

Psychological issues may also be involved. For example, patients with multiple somatic complaints may be less likely to benefit from lysis of adhesions.

PAGIDAS: Right. Interestingly, a meta-analysis of all the randomized trials involving mechanical barriers found no correlation to pregnancy outcome or pelvic pain.¹⁸ If we were to consider new trials, the psychological aspect would be worth looking into.

Seprafilm

DECHERNEY: Let’s move on to Seprafilm (Genzyme, Cambridge, Mass). What is it and how useful is it?

HURD: Seprafilm is modified hyaluronic acid, which forms a brittle, thin plastic layer. It is somewhat difficult to work with but, once it is in place, seems to adhere well. The presence of blood does not appear to be a problem, since the Seprafilm forms an impermeable barrier—unless it breaks. I have found it especially useful in myomectomies, which produce postoperative oozing through the incisions no matter how hard you try to prevent it.

In addition, in open cases, surfaces can easily be covered with this material. Unfortunately, it can’t be used laparoscopically because it is so brittle.

 

DECHERNEY: In my opinion, that is its major drawback.

PAGIDAS: In cardiac surgery, Seprafilm appears to work quite effectively.

SANFILIPPO: The manufacturer initially focused on surgeons in the context of sigmoid colon surgery, and it seems to work well in that setting.

Intergel

DECHERNEY: That brings us to the current state of the art: gels. I’m sure you all are familiar with Intergel (Lifecore Biomedical, Chaska, Minn), which is a ferrous derivative of hyaluronic acid that works by coating the raw surface areas. It also has the theoretical advantage of ease of use. Have any of you used Intergel?

HURD: As you know it was only recently approved by the US Food and Drug Administration, but not for laparoscopic use. It may work best on abraded bowel, which is avoided by laparoscopic surgery.

As you are probably also aware, the manufacturer recently took it off the market because of unusual side effects, namely a chemical peritonitis. Although peritonitis was cited as being rare, we encountered it in probably half the patients we operated on.

PAGIDAS: When we used it on hospitalized patients, the peritonitis wasn’t that obvious, since there was an expectation of significant pain. However, when we used Intergel on short-stay patients, we had to readmit them and do a full workup because we were concerned about bowel perforations. I’m surprised the manufacturer didn’t take it off the market sooner.

DECHERNEY: It seems strange, since Intergel has been used in Europe for a while now. I’ve used it in only 1 case and didn’t have adverse effects. It seemed to work well.

SANFILIPPO: It had all the right ingredients for success. It is unfortunate that these side effects have prohibited its use.

There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery.

Gels and the cost factor

DECHERNEY: Other gels are in the pipeline. I’m reminded of plasminogen activator, which is a powerful antiadhesive agent that lyses fibrin effectively. Unfortunately, it is prohibitively expensive.

The next phase likely will involve the so-called polymers. If you spray them on your hand, they are activated by light or another chemical and become a cellophane-like substance. The problem is viscosity. If sprayed on the sidewall, for instance, they run halfway down before they are activated, so the entire surface does not get covered.

PAGIDAS: One concern with polymers is that they could actually bring surfaces together when they polymerize. We still have a lot to learn.

DECHERNEY: Let’s say a new gel comes on the market that takes reformation adhesions from 90% to 10%, as opposed to 40% recurrence. Would you use it in all 4 of the cases we discuss here?

HURD: If it was that effective and had no adverse effects, it would be wonderful.

The cesarean-delivery case is different, as healing in a pregnant patient is 1 concern; the size of the uterus also has an effect. If the patient is breastfeeding, you would want to make sure the gel didn’t interfere.

PAGIDAS: We desperately need a product that can minimize adhesions regardless of the route of access or type of procedure. Even though we lack data on outcomes, I predict wide use of such a product, assuming it is nontoxic and effective.

DECHERNEY: What if it costs $1,000 a case?

SANFILIPPO: If it prevents 1 bowel obstruction, it still would be cost-effective.

DECHERNEY: The incidence of bowel obstruction for total abdominal hysterectomy is 2%, and 5% for radical hysterectomy.

HURD: We must be careful of the cost-benefit ratio. Bowel obstruction after gynecologic surgery is uncommon.

Is the gel worth $100? $1,000? $3,000? It’s difficult to say, but the more expensive it is, the less likely it will find widespread use.

PAGIDAS: I agree. We should remember that we still need to maintain microsurgical techniques and appropriate tissue handling, as well as avoid ischemia and infection.

Looking for the magic bullet

DECHERNEY: What is the future of adjunctive therapy?

SANFILIPPO: I would focus on noxythiolin; it has potential. Calcium channel blockers for adhesion prevention have also been studied.²⁰ In 1 investigation involving a rat model, the calcium channel blocker verapamil as well as several other agents—including vitamin E, carboxymethylcellulose, cyclosporin, aprotinin, and tenoxicam—were compared with respect to tissue effects. A beneficial effect was noted with all agents except cyclosporin and carboxymethylcellulose.

Whoever succeeds in manufacturing an effective preventive will be a winner.

HURD: There is no question that adhesion prevention is one of the unmet challenges in all surgeries, especially reproductive surgery. The most effective agent would be applied intraabdominally, since any systemic agent that stops adhesion formation would probably decrease wound healing as well.

 

I hope the most effective agents can be used in both laparoscopy and laparotomy, and that they will decrease the adhesion-formation rate by more than 50%. We need to find the magic bullet that can cover the entire pelvis—if not the entire abdomen.

Dr. DeCherney reports small holdings with Lifecore Biomedical and Johnson & Johnson. Drs. Hurd and Pagidas report no financial relationship with any companies whose products are mentioned in this article. Dr. Sanfilippo serves on the speaker’s bureau for Berlex, Ortho Pharmaceuticals, and Wyeth, and receives grant support from Eli Lilly and Wyeth.

Systematic study has established that so-called routine episiotomy should be abandoned, and restrictive-use protocols should be developed that aim, initially, for a rate of less than 30%.^1,2 Yet episiotomy (or more correctly, perineotomy) remains perhaps the most commonly performed surgical procedure in obstetrics.³ Its rate—more than 60% of vaginal deliveries in the United States—has not declined since Thacker and Banta’s landmark 1983 review.⁴

Although a large body of evidence indicates reassessment is in order, prophylactic episiotomy is a contentious issue. Indeed, it has been controversial ever since the procedure first became “routine” in the United States, in 1920. Still, advocates and dissenters share the same goal: to prevent severe perineal tears and their potential for urinary and fecal incontinence and sexual dysfunction.

This article reviews research findings that indicate:

• Data are inadequate to recommend one method of episiotomy over another.
  
• Timing of episiotomy to shorten the second stage of labor may be less relevant in an era of decreasing forceps utilization and without evidence of improved neonatal outcomes.⁵
  
• Episiotomy, particularly midline episiotomy, remains the single greatest risk that a patient will sustain a third- or fourth-degree laceration. When such lacerations occur spontaneously, recovery is equivalent to episiotomy extension or deliberate proctoepisiotomy.
  

TABLE

Incidence of third- or fourth-degree laceration with and without episiotomy

	NO. STUDIES COMPILED	NO. PATIENTS	% WITH 3RD- OR 4TH-DEGREE LACERATION
Midline episiotomy	12	49,395	6.5
No episiotomy	13	38,961	1.4
Adapted from Thorp JM.3

Research does not support presumed indications

Episiotomy was first described in 1742 as a procedure that could assist the obstetrician in difficult vaginal deliveries.³ It was not until the work of DeLee⁶ and Pomeroy⁷ was published in 1920—coincident with deliveries beginning to move from home to hospital—that the procedure became “routine.” Still, some leaders in the field—specifically, J. Whitridge Williams of Johns Hopkins—vigorously dissented.⁸

Historically, episiotomy has been used to facilitate delivery in cases of protracted second stage, instrumented vaginal delivery, and suspected fetal compromise. However, data supporting episiotomy as a facilitating procedure are sparse, and evidence endorsing prophylactic episiotomy is largely anecdotal or descriptive.

Agreement is widespread that episiotomy is warranted under certain circumstances: Shoulder dystocia, operative vaginal delivery, and a “short” perineal body have been presumed indications. Data are inadequate to support these claims, however.

Shoulder dystocia. While it might seem to make sense to perform an episiotomy (or more likely, a proctoepisiotomy) in cases of shoulder dystocia, no data from controlled trials support this theory. Given the relative rarity of severe shoulder dystocia and the inability to conduct a truly randomized trial, physicians are left with only their clinical judgment as a guide in this circumstance.

Operative delivery. Many clinicians have advocated routine episiotomy before operative vaginal delivery, particularly with forceps. The intent is to increase the space available for delivery that has been diminished by the introduction of forceps. This rationale does not hold up as well for vacuum extraction; 1 study noted that when episiotomy is performed in cases of vacuum extraction, the likelihood of severe perineal trauma is increased.⁹

It has been reported¹⁰ that the greatest risk factor for both perineal trauma and third- or fourth-degree perineal laceration is episiotomy itself (TABLE), independent of mode of delivery (spontaneous or operative).

Short perineum. Many physicians, myself included, have performed episiotomies because they perceived that the perineum was short and that even a controlled delivery with optimal use of the Ritgen maneuver probably would not prevent a perineal laceration. That said, data on anal and flatus incontinence and postpartum sexual functioning suggest that spontaneous recovery from second-degree lacerations is no worse than recovery from midline episiotomy^11,12 and, as stated, episiotomy itself is the leading risk factor for incurring a third- or fourth-degree extension—which imposes significantly greater recovery problems.

Two recent studies^11,12 identified episiotomy as a specific, independent risk factor for fecal incontinence and delayed return of sexual activity postpartum. When matched for degree of perineal trauma, episiotomy without extension still resulted in poorer outcomes at 3 and 6 months postpartum than did spontaneous second-degree lacerations, suggesting that routine episiotomy not only fails to prevent, but may actually increase risk of perineal injury and impaired function.

 

‘Prophylactic’ episiotomy is not preventive

Much debate has centered on optimal utilization of so-called prophylactic episiotomy. The intent of routine prophylactic episiotomy is to protect the pelvic floor, thus minimizing the risk of urinary incontinence and pelvic floor dysfunction. Data have suggested that absence of labor and cesarean delivery may protect against pelvic floor dysfunction; however, the role of episiotomy in preventing such dysfunction remains to be determined.

Cochrane Database review. This review¹ found no differences in vaginal or perineal trauma, dyspareunia, or urinary incontinence between patients with and without episiotomy. Patients who had an episiotomy had less risk of anterior perineal trauma but an overall greater risk of posterior perineal trauma and other complications. The reviewers concluded that restrictive episiotomy utilization is preferable to routine utilization.

The reviewers selected a total of 6 randomized trials; these examined:

• restrictive versus routine use of episiotomy;
  
• restrictive versus mediolateral episiotomy;
  
• restrictive versus routine midline episiotomy; and
  
• midline versus mediolateral episiotomy.
  

In the routine episiotomy group, 72.7% (1,752 of 2,409) of women underwent the procedure, versus 27.6% (673 of 2,441) in the restrictive episiotomy group.

Compared with routine use, restrictive episiotomy involved less posterior perineal trauma (relative risk [RR], 0.88; 95% confidence interval [CI], 0.84 to 0.920), less suturing (RR, 0.74; 95% CI, 0.71 to 0.77), and fewer healing complications (RR, 0.69; 95% CI, 0.56 to 0.85). Restrictive episiotomy was associated with more anterior perineal trauma (RR, 1.79; 95% CI, 1.55 to 2.07).

There was no difference in severe vaginal or perineal trauma (RR, 1.11; 95% CI, 0.83 to 1.50), dyspareunia (RR, 1.02; 95% CI, 0.90 to 1.16), urinary incontinence (RR 0.98; 95% CI, 0.79 to 1.20), or several pain measures.

Results for restrictive versus routine mediolateral and midline episiotomies were similar to the overall comparison.

Reviewers concluded that a policy of restrictive episiotomy appears to have several benefits over routine episiotomy: less posterior perineal trauma, less suturing, fewer complications, and no difference for most pain measures and severe vaginal or perineal trauma.

Risk of anterior perineal trauma with restrictive episiotomy was increased, however. Restrictive-use protocols, likely to be institution-specific, essentially curb episiotomy use by stating that the procedure should not be “routinely performed.” Instead, episiotomy is restricted to cases in which the clinician believes it is warranted. Examples of such situations include use of forceps, shoulder dystocia, and an estimated fetal weight above 4,000 g. As discussed, the data cannot support the value of episiotomy use even in these circumstances; however, simply discouraging routine episiotomy would effectively lower the rate to the desired 30% range.

Midline versus mediolateral incision. The most vocal debates focus on which type of episiotomy to perform and whether it should be performed earlier or later in the second stage of labor.

It has been proposed that by abandoning midline episiotomies in favor of the mediolateral technique, physicians can avoid injury to the sphincter and improve immediate birth outcome without compromising long-term function—though pros and cons of this approach are a subject of debate (see “Comparison of mid-line versus mediolateral episiotomy”).

Still, the data suggest that, when properly performed, median and mediolateral episiotomy have equivalent rates of satisfactory recovery,¹³ though the latter technique may require more technical skill for both its performance and repair.

Early versus late incision. Proponents argue that an episiotomy at the time the presenting part is crowning is “too little, too late.” They maintain that for the procedure to be truly protective, it should be utilized earlier in the second stage of labor.

Data are insufficient to confirm or refute the efficacy of early episiotomy. One would do well to remember, however, that early episiotomy was endorsed as a method to help shorten the second stage of labor when used in conjunction with prophylactic forceps delivery—a method that is now less prevalent in obstetric practice.

 

Does vaginal birth trauma cause pelvic floor dysfunction?

The relationship between vaginal birth trauma, irrespective of episiotomy, and pelvic floor dysfunction remains a topic of investigation. A recent report generated much interest in the potentially protective role of prophylactic cesarean section, particularly if performed prior to the onset of labor.¹⁴

Dr. Repke reports no financial relationship with any companies whose products are mentioned in this article.

Systematic study has established that so-called routine episiotomy should be abandoned, and restrictive-use protocols should be developed that aim, initially, for a rate of less than 30%.^1,2 Yet episiotomy (or more correctly, perineotomy) remains perhaps the most commonly performed surgical procedure in obstetrics.³ Its rate—more than 60% of vaginal deliveries in the United States—has not declined since Thacker and Banta’s landmark 1983 review.⁴

Although a large body of evidence indicates reassessment is in order, prophylactic episiotomy is a contentious issue. Indeed, it has been controversial ever since the procedure first became “routine” in the United States, in 1920. Still, advocates and dissenters share the same goal: to prevent severe perineal tears and their potential for urinary and fecal incontinence and sexual dysfunction.

This article reviews research findings that indicate:

• Data are inadequate to recommend one method of episiotomy over another.
  
• Timing of episiotomy to shorten the second stage of labor may be less relevant in an era of decreasing forceps utilization and without evidence of improved neonatal outcomes.⁵
  
• Episiotomy, particularly midline episiotomy, remains the single greatest risk that a patient will sustain a third- or fourth-degree laceration. When such lacerations occur spontaneously, recovery is equivalent to episiotomy extension or deliberate proctoepisiotomy.
  

TABLE

Incidence of third- or fourth-degree laceration with and without episiotomy

	NO. STUDIES COMPILED	NO. PATIENTS	% WITH 3RD- OR 4TH-DEGREE LACERATION
Midline episiotomy	12	49,395	6.5
No episiotomy	13	38,961	1.4
Adapted from Thorp JM.3

Research does not support presumed indications

Episiotomy was first described in 1742 as a procedure that could assist the obstetrician in difficult vaginal deliveries.³ It was not until the work of DeLee⁶ and Pomeroy⁷ was published in 1920—coincident with deliveries beginning to move from home to hospital—that the procedure became “routine.” Still, some leaders in the field—specifically, J. Whitridge Williams of Johns Hopkins—vigorously dissented.⁸

Historically, episiotomy has been used to facilitate delivery in cases of protracted second stage, instrumented vaginal delivery, and suspected fetal compromise. However, data supporting episiotomy as a facilitating procedure are sparse, and evidence endorsing prophylactic episiotomy is largely anecdotal or descriptive.

Agreement is widespread that episiotomy is warranted under certain circumstances: Shoulder dystocia, operative vaginal delivery, and a “short” perineal body have been presumed indications. Data are inadequate to support these claims, however.

Shoulder dystocia. While it might seem to make sense to perform an episiotomy (or more likely, a proctoepisiotomy) in cases of shoulder dystocia, no data from controlled trials support this theory. Given the relative rarity of severe shoulder dystocia and the inability to conduct a truly randomized trial, physicians are left with only their clinical judgment as a guide in this circumstance.

Operative delivery. Many clinicians have advocated routine episiotomy before operative vaginal delivery, particularly with forceps. The intent is to increase the space available for delivery that has been diminished by the introduction of forceps. This rationale does not hold up as well for vacuum extraction; 1 study noted that when episiotomy is performed in cases of vacuum extraction, the likelihood of severe perineal trauma is increased.⁹

It has been reported¹⁰ that the greatest risk factor for both perineal trauma and third- or fourth-degree perineal laceration is episiotomy itself (TABLE), independent of mode of delivery (spontaneous or operative).

Short perineum. Many physicians, myself included, have performed episiotomies because they perceived that the perineum was short and that even a controlled delivery with optimal use of the Ritgen maneuver probably would not prevent a perineal laceration. That said, data on anal and flatus incontinence and postpartum sexual functioning suggest that spontaneous recovery from second-degree lacerations is no worse than recovery from midline episiotomy^11,12 and, as stated, episiotomy itself is the leading risk factor for incurring a third- or fourth-degree extension—which imposes significantly greater recovery problems.

Two recent studies^11,12 identified episiotomy as a specific, independent risk factor for fecal incontinence and delayed return of sexual activity postpartum. When matched for degree of perineal trauma, episiotomy without extension still resulted in poorer outcomes at 3 and 6 months postpartum than did spontaneous second-degree lacerations, suggesting that routine episiotomy not only fails to prevent, but may actually increase risk of perineal injury and impaired function.

 

‘Prophylactic’ episiotomy is not preventive

Much debate has centered on optimal utilization of so-called prophylactic episiotomy. The intent of routine prophylactic episiotomy is to protect the pelvic floor, thus minimizing the risk of urinary incontinence and pelvic floor dysfunction. Data have suggested that absence of labor and cesarean delivery may protect against pelvic floor dysfunction; however, the role of episiotomy in preventing such dysfunction remains to be determined.

Cochrane Database review. This review¹ found no differences in vaginal or perineal trauma, dyspareunia, or urinary incontinence between patients with and without episiotomy. Patients who had an episiotomy had less risk of anterior perineal trauma but an overall greater risk of posterior perineal trauma and other complications. The reviewers concluded that restrictive episiotomy utilization is preferable to routine utilization.

The reviewers selected a total of 6 randomized trials; these examined:

• restrictive versus routine use of episiotomy;
  
• restrictive versus mediolateral episiotomy;
  
• restrictive versus routine midline episiotomy; and
  
• midline versus mediolateral episiotomy.
  

In the routine episiotomy group, 72.7% (1,752 of 2,409) of women underwent the procedure, versus 27.6% (673 of 2,441) in the restrictive episiotomy group.

Compared with routine use, restrictive episiotomy involved less posterior perineal trauma (relative risk [RR], 0.88; 95% confidence interval [CI], 0.84 to 0.920), less suturing (RR, 0.74; 95% CI, 0.71 to 0.77), and fewer healing complications (RR, 0.69; 95% CI, 0.56 to 0.85). Restrictive episiotomy was associated with more anterior perineal trauma (RR, 1.79; 95% CI, 1.55 to 2.07).

There was no difference in severe vaginal or perineal trauma (RR, 1.11; 95% CI, 0.83 to 1.50), dyspareunia (RR, 1.02; 95% CI, 0.90 to 1.16), urinary incontinence (RR 0.98; 95% CI, 0.79 to 1.20), or several pain measures.

Results for restrictive versus routine mediolateral and midline episiotomies were similar to the overall comparison.

Reviewers concluded that a policy of restrictive episiotomy appears to have several benefits over routine episiotomy: less posterior perineal trauma, less suturing, fewer complications, and no difference for most pain measures and severe vaginal or perineal trauma.

Risk of anterior perineal trauma with restrictive episiotomy was increased, however. Restrictive-use protocols, likely to be institution-specific, essentially curb episiotomy use by stating that the procedure should not be “routinely performed.” Instead, episiotomy is restricted to cases in which the clinician believes it is warranted. Examples of such situations include use of forceps, shoulder dystocia, and an estimated fetal weight above 4,000 g. As discussed, the data cannot support the value of episiotomy use even in these circumstances; however, simply discouraging routine episiotomy would effectively lower the rate to the desired 30% range.

Midline versus mediolateral incision. The most vocal debates focus on which type of episiotomy to perform and whether it should be performed earlier or later in the second stage of labor.

It has been proposed that by abandoning midline episiotomies in favor of the mediolateral technique, physicians can avoid injury to the sphincter and improve immediate birth outcome without compromising long-term function—though pros and cons of this approach are a subject of debate (see “Comparison of mid-line versus mediolateral episiotomy”).

Still, the data suggest that, when properly performed, median and mediolateral episiotomy have equivalent rates of satisfactory recovery,¹³ though the latter technique may require more technical skill for both its performance and repair.

Early versus late incision. Proponents argue that an episiotomy at the time the presenting part is crowning is “too little, too late.” They maintain that for the procedure to be truly protective, it should be utilized earlier in the second stage of labor.

Data are insufficient to confirm or refute the efficacy of early episiotomy. One would do well to remember, however, that early episiotomy was endorsed as a method to help shorten the second stage of labor when used in conjunction with prophylactic forceps delivery—a method that is now less prevalent in obstetric practice.

 

Does vaginal birth trauma cause pelvic floor dysfunction?

The relationship between vaginal birth trauma, irrespective of episiotomy, and pelvic floor dysfunction remains a topic of investigation. A recent report generated much interest in the potentially protective role of prophylactic cesarean section, particularly if performed prior to the onset of labor.¹⁴

Dr. Repke reports no financial relationship with any companies whose products are mentioned in this article.

Systematic study has established that so-called routine episiotomy should be abandoned, and restrictive-use protocols should be developed that aim, initially, for a rate of less than 30%.^1,2 Yet episiotomy (or more correctly, perineotomy) remains perhaps the most commonly performed surgical procedure in obstetrics.³ Its rate—more than 60% of vaginal deliveries in the United States—has not declined since Thacker and Banta’s landmark 1983 review.⁴

Although a large body of evidence indicates reassessment is in order, prophylactic episiotomy is a contentious issue. Indeed, it has been controversial ever since the procedure first became “routine” in the United States, in 1920. Still, advocates and dissenters share the same goal: to prevent severe perineal tears and their potential for urinary and fecal incontinence and sexual dysfunction.

This article reviews research findings that indicate:

• Data are inadequate to recommend one method of episiotomy over another.
  
• Timing of episiotomy to shorten the second stage of labor may be less relevant in an era of decreasing forceps utilization and without evidence of improved neonatal outcomes.⁵
  
• Episiotomy, particularly midline episiotomy, remains the single greatest risk that a patient will sustain a third- or fourth-degree laceration. When such lacerations occur spontaneously, recovery is equivalent to episiotomy extension or deliberate proctoepisiotomy.
  

TABLE

Incidence of third- or fourth-degree laceration with and without episiotomy

	NO. STUDIES COMPILED	NO. PATIENTS	% WITH 3RD- OR 4TH-DEGREE LACERATION
Midline episiotomy	12	49,395	6.5
No episiotomy	13	38,961	1.4
Adapted from Thorp JM.3

Research does not support presumed indications

Episiotomy was first described in 1742 as a procedure that could assist the obstetrician in difficult vaginal deliveries.³ It was not until the work of DeLee⁶ and Pomeroy⁷ was published in 1920—coincident with deliveries beginning to move from home to hospital—that the procedure became “routine.” Still, some leaders in the field—specifically, J. Whitridge Williams of Johns Hopkins—vigorously dissented.⁸

Historically, episiotomy has been used to facilitate delivery in cases of protracted second stage, instrumented vaginal delivery, and suspected fetal compromise. However, data supporting episiotomy as a facilitating procedure are sparse, and evidence endorsing prophylactic episiotomy is largely anecdotal or descriptive.

Agreement is widespread that episiotomy is warranted under certain circumstances: Shoulder dystocia, operative vaginal delivery, and a “short” perineal body have been presumed indications. Data are inadequate to support these claims, however.

Shoulder dystocia. While it might seem to make sense to perform an episiotomy (or more likely, a proctoepisiotomy) in cases of shoulder dystocia, no data from controlled trials support this theory. Given the relative rarity of severe shoulder dystocia and the inability to conduct a truly randomized trial, physicians are left with only their clinical judgment as a guide in this circumstance.

Operative delivery. Many clinicians have advocated routine episiotomy before operative vaginal delivery, particularly with forceps. The intent is to increase the space available for delivery that has been diminished by the introduction of forceps. This rationale does not hold up as well for vacuum extraction; 1 study noted that when episiotomy is performed in cases of vacuum extraction, the likelihood of severe perineal trauma is increased.⁹

It has been reported¹⁰ that the greatest risk factor for both perineal trauma and third- or fourth-degree perineal laceration is episiotomy itself (TABLE), independent of mode of delivery (spontaneous or operative).

Short perineum. Many physicians, myself included, have performed episiotomies because they perceived that the perineum was short and that even a controlled delivery with optimal use of the Ritgen maneuver probably would not prevent a perineal laceration. That said, data on anal and flatus incontinence and postpartum sexual functioning suggest that spontaneous recovery from second-degree lacerations is no worse than recovery from midline episiotomy^11,12 and, as stated, episiotomy itself is the leading risk factor for incurring a third- or fourth-degree extension—which imposes significantly greater recovery problems.

Two recent studies^11,12 identified episiotomy as a specific, independent risk factor for fecal incontinence and delayed return of sexual activity postpartum. When matched for degree of perineal trauma, episiotomy without extension still resulted in poorer outcomes at 3 and 6 months postpartum than did spontaneous second-degree lacerations, suggesting that routine episiotomy not only fails to prevent, but may actually increase risk of perineal injury and impaired function.

 

‘Prophylactic’ episiotomy is not preventive

Much debate has centered on optimal utilization of so-called prophylactic episiotomy. The intent of routine prophylactic episiotomy is to protect the pelvic floor, thus minimizing the risk of urinary incontinence and pelvic floor dysfunction. Data have suggested that absence of labor and cesarean delivery may protect against pelvic floor dysfunction; however, the role of episiotomy in preventing such dysfunction remains to be determined.

Cochrane Database review. This review¹ found no differences in vaginal or perineal trauma, dyspareunia, or urinary incontinence between patients with and without episiotomy. Patients who had an episiotomy had less risk of anterior perineal trauma but an overall greater risk of posterior perineal trauma and other complications. The reviewers concluded that restrictive episiotomy utilization is preferable to routine utilization.

The reviewers selected a total of 6 randomized trials; these examined:

• restrictive versus routine use of episiotomy;
  
• restrictive versus mediolateral episiotomy;
  
• restrictive versus routine midline episiotomy; and
  
• midline versus mediolateral episiotomy.
  

In the routine episiotomy group, 72.7% (1,752 of 2,409) of women underwent the procedure, versus 27.6% (673 of 2,441) in the restrictive episiotomy group.

Compared with routine use, restrictive episiotomy involved less posterior perineal trauma (relative risk [RR], 0.88; 95% confidence interval [CI], 0.84 to 0.920), less suturing (RR, 0.74; 95% CI, 0.71 to 0.77), and fewer healing complications (RR, 0.69; 95% CI, 0.56 to 0.85). Restrictive episiotomy was associated with more anterior perineal trauma (RR, 1.79; 95% CI, 1.55 to 2.07).

There was no difference in severe vaginal or perineal trauma (RR, 1.11; 95% CI, 0.83 to 1.50), dyspareunia (RR, 1.02; 95% CI, 0.90 to 1.16), urinary incontinence (RR 0.98; 95% CI, 0.79 to 1.20), or several pain measures.

Results for restrictive versus routine mediolateral and midline episiotomies were similar to the overall comparison.

Reviewers concluded that a policy of restrictive episiotomy appears to have several benefits over routine episiotomy: less posterior perineal trauma, less suturing, fewer complications, and no difference for most pain measures and severe vaginal or perineal trauma.

Risk of anterior perineal trauma with restrictive episiotomy was increased, however. Restrictive-use protocols, likely to be institution-specific, essentially curb episiotomy use by stating that the procedure should not be “routinely performed.” Instead, episiotomy is restricted to cases in which the clinician believes it is warranted. Examples of such situations include use of forceps, shoulder dystocia, and an estimated fetal weight above 4,000 g. As discussed, the data cannot support the value of episiotomy use even in these circumstances; however, simply discouraging routine episiotomy would effectively lower the rate to the desired 30% range.

Midline versus mediolateral incision. The most vocal debates focus on which type of episiotomy to perform and whether it should be performed earlier or later in the second stage of labor.

It has been proposed that by abandoning midline episiotomies in favor of the mediolateral technique, physicians can avoid injury to the sphincter and improve immediate birth outcome without compromising long-term function—though pros and cons of this approach are a subject of debate (see “Comparison of mid-line versus mediolateral episiotomy”).

Still, the data suggest that, when properly performed, median and mediolateral episiotomy have equivalent rates of satisfactory recovery,¹³ though the latter technique may require more technical skill for both its performance and repair.

Early versus late incision. Proponents argue that an episiotomy at the time the presenting part is crowning is “too little, too late.” They maintain that for the procedure to be truly protective, it should be utilized earlier in the second stage of labor.

Data are insufficient to confirm or refute the efficacy of early episiotomy. One would do well to remember, however, that early episiotomy was endorsed as a method to help shorten the second stage of labor when used in conjunction with prophylactic forceps delivery—a method that is now less prevalent in obstetric practice.

 

Does vaginal birth trauma cause pelvic floor dysfunction?

The relationship between vaginal birth trauma, irrespective of episiotomy, and pelvic floor dysfunction remains a topic of investigation. A recent report generated much interest in the potentially protective role of prophylactic cesarean section, particularly if performed prior to the onset of labor.¹⁴

Dr. Repke reports no financial relationship with any companies whose products are mentioned in this article.

For many years, a single standard of care governed the prevention of postoperative ileus following cesarean delivery and other abdominal surgeries; now it appears the thinking behind that strategy is outdated.

Traditionally, the routine approach to avoiding this complication consisted of placing a nasogastric (NG) tube to decompress the bowel and delaying feeding until bowel function resumed.

More recent studies indicate that a different tactic may be preferable. These suggest that postoperative ileus—which has an estimated annual cost of $750 million¹—can be significantly reduced with a simple 3-step process:

• withholding the nasogastric (NG) tube,
  
• feeding the patient early in the recovery process, and
  
• continuing epidural local anesthesia postoperatively.
  

Pathogenesis of ileus

We now know that the return of bowel function following surgery is an orderly event. The return of the small intestine’s action begins first, usually 4 to 8 hours postoperatively, and generally becomes complete around 24 hours. The colon resumes its function between 48 and 72 hours postoperatively.²

Very little has been written about the pathogenesis of postoperative ileus, but multiple causes have been suggested: sympathetic reflexes; inhibitory humoral agents; release of norepinephrine from the bowel wall; and the effects of anesthesia agents, opiates, and inflammation.³ The 2 most frequently mentioned etiologies are:

• the inhibitory neural reflex and
  
• inflammatory mediators released from the site of injury. (Inflammation is thought to trigger the release of macrophages, cytokines, and other inflammatory mediators, causing neutrophil infiltration.⁴)
  

We also know that many types of anesthesia can affect bowel motility. Delayed gastric emptying—which can cause aspiration, postoperative nausea and vomiting, and delayed absorption of medicine^4,5—is observed after exposure to systemic anesthesia. Atropine, halothane, and enflurane all decrease gastric emptying.

Controversy remains as to what actually initiates the ileus. Is it manipulation of the bowel or the rigors of surgery and anesthesia? Kalff et al⁶ conducted bowel muscle studies in rats and concluded that manipulation of the bowel—and not the laparotomy per se—causes a failure of gut circular muscle function 24 hours later. They also noted an increase of phagocytes and mast cells. Their data support the hypothesis that abdominal surgery initiates a cascade of inflammatory events that leads to postoperative ileus.²

The case for early feeding

In the past 10 years, several studies have demonstrated that—rather than reduce the incidence of ileus—inserting an NG tube and withholding regular feeding following abdominal surgery can cause an ileus or prolong a preexisting one. Other trials have shown that feeding a patient early in the postoperative period can actually prevent ileus.^7,8

Physiologic studies have shown that neither electrical activity of the bowel nor motor activities in the stomach are affected by surgery.⁸ Schilder et al⁹ reported bowel activity before the passage of flatus, indicating that the bowel is on its way to recovery much earlier than had been assumed. Thus, early postoperative feeding is well tolerated in most patients and associated with reduced discomfort and a more rapid recovery.⁷

For example, MacMillan et al studied 139 women undergoing gynecologic surgery for benign conditions; 67 were randomized to “early” feeding and 72 to traditional management. Early feeding involved a low-residue diet given 6 hours postoperatively, while traditional feeding consisted of clear liquids, which were withheld until the return of normal bowel sounds. Patients progressed to a regular diet with the passage of flatus. No increase in gastrointestinal complaints occurred in the early feeding group.⁷

Pearl et al⁸ compared similar groups of patients (TABLE 1). Patients in the first group were fed a clear liquid diet on the first postoperative day and progressed to a regular diet as soon as it could be tolerated. The traditional group was not fed until the return of bowel function, which was defined as the passage of flatus and no abdominal distension or vomiting; they were then started on clear liquids and, later, solid foods. While the incidence of complications was the same in both groups, hospitalization was shorter in the early feeding group.

A Cochrane review compared early versus delayed oral fluids and food after cesarean delivery. Of 12 studies considered, 6 were included in the review. No evidence was found to justify a policy of withholding oral fluids after uncomplicated cesarean sections.⁵

Simple versus complex procedures. Early feeding is not only safe in standardized surgeries such as cesarean section, but extends to complicated surgeries as well, as demonstrated in the trials by MacMillan et al⁷ and Pearl et al,⁸ which involved major gynecologic surgery. Trials with gynecologic oncology patients have shown the same result.^2,8 Even patients undergoing colorectal surgery can tolerate oral feeding very early in their postoperative course without bowel complications.^2,8

 

Not all studies have reported similar findings. Several concluded that other, nonmedical reasons, such as insurance requirements, accounted for the shorter hospital stays in many patients receiving early feeding.

TABLE 1

Complication rates associated with early feeding versus traditional management

COMPLICATION	EARLY FEEDING* (N=92)	TRADITIONAL MANAGEMENT· (N= 103)
Nausea	43.5%	24.3%
Nasogastric tube use	3.3%	6.7%
Febrile morbidity	54.3%	55.3%
Pneumonia	0%	1.9%
Wound complications	21.7%	21.4%
Atelectasis	8.7%	10.7%
Length of hospital stay (mean±standard deviation)	4.6±2.1	5.8±2.7
*Clear liquid diet on postoperative day 1
†Feeding delayed until return of bowel function
Reprinted from Obstet Gyn; vol 92; Pearl ML, Valea FA, Fischer M, Mahler L, Chalas E. A randomized controlled trial of early postoperative feeding in gynecologic oncology patients undergoing intraabdominal surgery; pages 94-97; copyright 1998; with permission from American College of Obstetricians and Gynecologists.

Nasogastric tubes only when indicated

The use of NG tubes after laparotomy has been studied extensively. A review of the literature suggests that routine placement of the tubes in asymptomatic patients is not justified and may possibly be harmful.¹⁰

In their meta-analysis of the issue, Cheatham et al¹¹ showed that although abdominal distension and vomiting are more frequent in patients who forgo NG tubes postoperatively, fever, atelectasis, and pneumonia are less common, and the interval between surgery and oral feeding is reduced (TABLE 2).

The authors concluded that for every NG tube inserted after abdominal surgery, at least 20 patients can be managed without it.

Forgoing an NG tube also lowers the risk of pulmonary complications, which increases 10-fold when a tube is inserted.¹²

TABLE 2

Complications associated with selective versus routine NG tube placement

	SELECTIVE PLACEMENT (N)	ROUTINE PLACEMENT (N)	P VALUE	RELATIVE RISK
Patients	1,986	1,978
Tubes placed/replaced	103	36	<.001>	2.9
Complications	833	1,084	.03	0.76
Deaths	13	25	.22	0.36
Pneumonia	53	119	<.0001>	0.49
Atelectasis	44	94	.001	0.46
Fever	108	212	.02	0.51
Vomiting	201	168	.11	1.19
Oral feedings (days)	3.53	4.59	.04
Length of stay (days)	9.32	10.1	.22
NG=nasogastric
Reprinted with permission from Cheatham ML, Chapman WC, Key SP, Sawyer JL. A meta-analysis of selective versus routine nasogastric decompression after elective laparotomy. Ann Surg. 1995:221:469-478.

Continue the epidural anesthetic

A number of experts believe that postoperative ileus is caused by stimulation of neural reflexes, which appear to be of 2 kinds: afferent stimuli to the spinal cord and efferent stimuli to the intestines through the sympathetic nervous system. The latter inhibits motility of the intestinal tract. Numerous studies demonstrate that this sympathetic reflex can be blocked by the use of epidural anesthesia.¹⁰

For example, Holte et al⁴ found that postoperative administration of thoracic epidural blockade with local anesthesia significantly reduced both ileus and pulmonary complications. They concluded that continuous epidural anesthesia with local anesthesia and minimally invasive surgery are the 2 most critical events in reducing postoperative ileus.

In a Cochrane review, Jorgensen and colleagues¹² compared the effects of epidural local anesthesia and opioid-based analgesic regimens on postoperative gastrointestinal paralysis, nausea and vomiting, and pain after abdominal surgery (TABLE 3). Epidural local anesthetics reduced gastrointestinal paralysis, as compared with systemic or epidural opioids, but provided the same postoperative pain relief. They also found that the addition of opioids to local epidural anesthesia provided superior postoperative analgesia—compared with epidural local anesthetics alone—without increasing the likelihood of ileus.

A study¹⁰ of patients undergoing colectomy found postoperative ileus was prevented or decreased with a 2-day regimen that included:

• continuous thoracic epidural anesthesia for 48 hours;
  
• withholding NG tubes;
  
• having the patient drink a liter of fluid on the day of surgery;
  
• initiating feeding after 24 hours;
  
• administering milk of magnesia; and
  
• mobilization after 8 hours, if possible.
  

Other studies have suggested that movement of the patient does not help eliminate postoperative ileus. It does, however, help prevent other postoperative complications, especially deep vein thrombosis.

Physicians in this study also used transverse surgical incisions to reduce pain and pulmonary problems.

TABLE 3

Anesthetic effect on GI function, postoperative pain, and nausea and vomiting: A comparison

ANESTHETIC	GI FUNCTION RETURNS	PAIN RELIEF	NAUSEA AND VOMITING
Epidural plus local	24 hr	Comparable	No significant difference
Epidural plus opioids	37 hr	Comparable	No significant difference
Systemic plus opioids	37 hr	Comparable	No significant difference
GI=gastrointestinal
Data from Jorgensen H, et al12

Clinical recommendations

The deregulation of the autonomic nervous system during surgery alters the gastrointestinal tract postoperatively, with neurotransmitters, local factors, and hormones playing a large role. Some forms of anesthesia also contribute to postoperative ileus, as does the use of narcotic analgesia after surgery.

The most efficient ways to activate the bowel postoperatively are:

• Continuing the thoracic epidural from 24 to 48 hours, which increases the splanchnic blood flow and blocks afferent and efferent sympathetic inhibitory nerve impulses. Note, however, that comparative studies of thoracic epidural anesthesia with local anesthesia are needed to quantify its impact.
  
• Hydrating the patient with a large amount of fluid in the first 24 hours after surgery.
  

Early feeding also seems to stimulate propulsive bowel motility.

Following these steps routinely can significantly decrease the risk of postoperative ileus and thus its resulting complications.

 

Dr. Rosenman reports no financial relationship with any companies whose products are mentioned in this article.

For many years, a single standard of care governed the prevention of postoperative ileus following cesarean delivery and other abdominal surgeries; now it appears the thinking behind that strategy is outdated.

Traditionally, the routine approach to avoiding this complication consisted of placing a nasogastric (NG) tube to decompress the bowel and delaying feeding until bowel function resumed.

More recent studies indicate that a different tactic may be preferable. These suggest that postoperative ileus—which has an estimated annual cost of $750 million¹—can be significantly reduced with a simple 3-step process:

• withholding the nasogastric (NG) tube,
  
• feeding the patient early in the recovery process, and
  
• continuing epidural local anesthesia postoperatively.
  

Pathogenesis of ileus

We now know that the return of bowel function following surgery is an orderly event. The return of the small intestine’s action begins first, usually 4 to 8 hours postoperatively, and generally becomes complete around 24 hours. The colon resumes its function between 48 and 72 hours postoperatively.²

Very little has been written about the pathogenesis of postoperative ileus, but multiple causes have been suggested: sympathetic reflexes; inhibitory humoral agents; release of norepinephrine from the bowel wall; and the effects of anesthesia agents, opiates, and inflammation.³ The 2 most frequently mentioned etiologies are:

• the inhibitory neural reflex and
  
• inflammatory mediators released from the site of injury. (Inflammation is thought to trigger the release of macrophages, cytokines, and other inflammatory mediators, causing neutrophil infiltration.⁴)
  

We also know that many types of anesthesia can affect bowel motility. Delayed gastric emptying—which can cause aspiration, postoperative nausea and vomiting, and delayed absorption of medicine^4,5—is observed after exposure to systemic anesthesia. Atropine, halothane, and enflurane all decrease gastric emptying.

Controversy remains as to what actually initiates the ileus. Is it manipulation of the bowel or the rigors of surgery and anesthesia? Kalff et al⁶ conducted bowel muscle studies in rats and concluded that manipulation of the bowel—and not the laparotomy per se—causes a failure of gut circular muscle function 24 hours later. They also noted an increase of phagocytes and mast cells. Their data support the hypothesis that abdominal surgery initiates a cascade of inflammatory events that leads to postoperative ileus.²

The case for early feeding

In the past 10 years, several studies have demonstrated that—rather than reduce the incidence of ileus—inserting an NG tube and withholding regular feeding following abdominal surgery can cause an ileus or prolong a preexisting one. Other trials have shown that feeding a patient early in the postoperative period can actually prevent ileus.^7,8

Physiologic studies have shown that neither electrical activity of the bowel nor motor activities in the stomach are affected by surgery.⁸ Schilder et al⁹ reported bowel activity before the passage of flatus, indicating that the bowel is on its way to recovery much earlier than had been assumed. Thus, early postoperative feeding is well tolerated in most patients and associated with reduced discomfort and a more rapid recovery.⁷

For example, MacMillan et al studied 139 women undergoing gynecologic surgery for benign conditions; 67 were randomized to “early” feeding and 72 to traditional management. Early feeding involved a low-residue diet given 6 hours postoperatively, while traditional feeding consisted of clear liquids, which were withheld until the return of normal bowel sounds. Patients progressed to a regular diet with the passage of flatus. No increase in gastrointestinal complaints occurred in the early feeding group.⁷

Pearl et al⁸ compared similar groups of patients (TABLE 1). Patients in the first group were fed a clear liquid diet on the first postoperative day and progressed to a regular diet as soon as it could be tolerated. The traditional group was not fed until the return of bowel function, which was defined as the passage of flatus and no abdominal distension or vomiting; they were then started on clear liquids and, later, solid foods. While the incidence of complications was the same in both groups, hospitalization was shorter in the early feeding group.

A Cochrane review compared early versus delayed oral fluids and food after cesarean delivery. Of 12 studies considered, 6 were included in the review. No evidence was found to justify a policy of withholding oral fluids after uncomplicated cesarean sections.⁵

Simple versus complex procedures. Early feeding is not only safe in standardized surgeries such as cesarean section, but extends to complicated surgeries as well, as demonstrated in the trials by MacMillan et al⁷ and Pearl et al,⁸ which involved major gynecologic surgery. Trials with gynecologic oncology patients have shown the same result.^2,8 Even patients undergoing colorectal surgery can tolerate oral feeding very early in their postoperative course without bowel complications.^2,8

 

Not all studies have reported similar findings. Several concluded that other, nonmedical reasons, such as insurance requirements, accounted for the shorter hospital stays in many patients receiving early feeding.

TABLE 1

Complication rates associated with early feeding versus traditional management

COMPLICATION	EARLY FEEDING* (N=92)	TRADITIONAL MANAGEMENT· (N= 103)
Nausea	43.5%	24.3%
Nasogastric tube use	3.3%	6.7%
Febrile morbidity	54.3%	55.3%
Pneumonia	0%	1.9%
Wound complications	21.7%	21.4%
Atelectasis	8.7%	10.7%
Length of hospital stay (mean±standard deviation)	4.6±2.1	5.8±2.7
*Clear liquid diet on postoperative day 1
†Feeding delayed until return of bowel function
Reprinted from Obstet Gyn; vol 92; Pearl ML, Valea FA, Fischer M, Mahler L, Chalas E. A randomized controlled trial of early postoperative feeding in gynecologic oncology patients undergoing intraabdominal surgery; pages 94-97; copyright 1998; with permission from American College of Obstetricians and Gynecologists.

Nasogastric tubes only when indicated

The use of NG tubes after laparotomy has been studied extensively. A review of the literature suggests that routine placement of the tubes in asymptomatic patients is not justified and may possibly be harmful.¹⁰

In their meta-analysis of the issue, Cheatham et al¹¹ showed that although abdominal distension and vomiting are more frequent in patients who forgo NG tubes postoperatively, fever, atelectasis, and pneumonia are less common, and the interval between surgery and oral feeding is reduced (TABLE 2).

The authors concluded that for every NG tube inserted after abdominal surgery, at least 20 patients can be managed without it.

Forgoing an NG tube also lowers the risk of pulmonary complications, which increases 10-fold when a tube is inserted.¹²

TABLE 2

Complications associated with selective versus routine NG tube placement

	SELECTIVE PLACEMENT (N)	ROUTINE PLACEMENT (N)	P VALUE	RELATIVE RISK
Patients	1,986	1,978
Tubes placed/replaced	103	36	<.001>	2.9
Complications	833	1,084	.03	0.76
Deaths	13	25	.22	0.36
Pneumonia	53	119	<.0001>	0.49
Atelectasis	44	94	.001	0.46
Fever	108	212	.02	0.51
Vomiting	201	168	.11	1.19
Oral feedings (days)	3.53	4.59	.04
Length of stay (days)	9.32	10.1	.22
NG=nasogastric
Reprinted with permission from Cheatham ML, Chapman WC, Key SP, Sawyer JL. A meta-analysis of selective versus routine nasogastric decompression after elective laparotomy. Ann Surg. 1995:221:469-478.

Continue the epidural anesthetic

A number of experts believe that postoperative ileus is caused by stimulation of neural reflexes, which appear to be of 2 kinds: afferent stimuli to the spinal cord and efferent stimuli to the intestines through the sympathetic nervous system. The latter inhibits motility of the intestinal tract. Numerous studies demonstrate that this sympathetic reflex can be blocked by the use of epidural anesthesia.¹⁰

For example, Holte et al⁴ found that postoperative administration of thoracic epidural blockade with local anesthesia significantly reduced both ileus and pulmonary complications. They concluded that continuous epidural anesthesia with local anesthesia and minimally invasive surgery are the 2 most critical events in reducing postoperative ileus.

In a Cochrane review, Jorgensen and colleagues¹² compared the effects of epidural local anesthesia and opioid-based analgesic regimens on postoperative gastrointestinal paralysis, nausea and vomiting, and pain after abdominal surgery (TABLE 3). Epidural local anesthetics reduced gastrointestinal paralysis, as compared with systemic or epidural opioids, but provided the same postoperative pain relief. They also found that the addition of opioids to local epidural anesthesia provided superior postoperative analgesia—compared with epidural local anesthetics alone—without increasing the likelihood of ileus.

A study¹⁰ of patients undergoing colectomy found postoperative ileus was prevented or decreased with a 2-day regimen that included:

• continuous thoracic epidural anesthesia for 48 hours;
  
• withholding NG tubes;
  
• having the patient drink a liter of fluid on the day of surgery;
  
• initiating feeding after 24 hours;
  
• administering milk of magnesia; and
  
• mobilization after 8 hours, if possible.
  

Other studies have suggested that movement of the patient does not help eliminate postoperative ileus. It does, however, help prevent other postoperative complications, especially deep vein thrombosis.

Physicians in this study also used transverse surgical incisions to reduce pain and pulmonary problems.

TABLE 3

Anesthetic effect on GI function, postoperative pain, and nausea and vomiting: A comparison

ANESTHETIC	GI FUNCTION RETURNS	PAIN RELIEF	NAUSEA AND VOMITING
Epidural plus local	24 hr	Comparable	No significant difference
Epidural plus opioids	37 hr	Comparable	No significant difference
Systemic plus opioids	37 hr	Comparable	No significant difference
GI=gastrointestinal
Data from Jorgensen H, et al12

Clinical recommendations

The deregulation of the autonomic nervous system during surgery alters the gastrointestinal tract postoperatively, with neurotransmitters, local factors, and hormones playing a large role. Some forms of anesthesia also contribute to postoperative ileus, as does the use of narcotic analgesia after surgery.

The most efficient ways to activate the bowel postoperatively are:

• Continuing the thoracic epidural from 24 to 48 hours, which increases the splanchnic blood flow and blocks afferent and efferent sympathetic inhibitory nerve impulses. Note, however, that comparative studies of thoracic epidural anesthesia with local anesthesia are needed to quantify its impact.
  
• Hydrating the patient with a large amount of fluid in the first 24 hours after surgery.
  

Early feeding also seems to stimulate propulsive bowel motility.

Following these steps routinely can significantly decrease the risk of postoperative ileus and thus its resulting complications.

 

Dr. Rosenman reports no financial relationship with any companies whose products are mentioned in this article.

For many years, a single standard of care governed the prevention of postoperative ileus following cesarean delivery and other abdominal surgeries; now it appears the thinking behind that strategy is outdated.

Traditionally, the routine approach to avoiding this complication consisted of placing a nasogastric (NG) tube to decompress the bowel and delaying feeding until bowel function resumed.

More recent studies indicate that a different tactic may be preferable. These suggest that postoperative ileus—which has an estimated annual cost of $750 million¹—can be significantly reduced with a simple 3-step process:

• withholding the nasogastric (NG) tube,
  
• feeding the patient early in the recovery process, and
  
• continuing epidural local anesthesia postoperatively.
  

Pathogenesis of ileus

We now know that the return of bowel function following surgery is an orderly event. The return of the small intestine’s action begins first, usually 4 to 8 hours postoperatively, and generally becomes complete around 24 hours. The colon resumes its function between 48 and 72 hours postoperatively.²

Very little has been written about the pathogenesis of postoperative ileus, but multiple causes have been suggested: sympathetic reflexes; inhibitory humoral agents; release of norepinephrine from the bowel wall; and the effects of anesthesia agents, opiates, and inflammation.³ The 2 most frequently mentioned etiologies are:

• the inhibitory neural reflex and
  
• inflammatory mediators released from the site of injury. (Inflammation is thought to trigger the release of macrophages, cytokines, and other inflammatory mediators, causing neutrophil infiltration.⁴)
  

We also know that many types of anesthesia can affect bowel motility. Delayed gastric emptying—which can cause aspiration, postoperative nausea and vomiting, and delayed absorption of medicine^4,5—is observed after exposure to systemic anesthesia. Atropine, halothane, and enflurane all decrease gastric emptying.

Controversy remains as to what actually initiates the ileus. Is it manipulation of the bowel or the rigors of surgery and anesthesia? Kalff et al⁶ conducted bowel muscle studies in rats and concluded that manipulation of the bowel—and not the laparotomy per se—causes a failure of gut circular muscle function 24 hours later. They also noted an increase of phagocytes and mast cells. Their data support the hypothesis that abdominal surgery initiates a cascade of inflammatory events that leads to postoperative ileus.²

The case for early feeding

In the past 10 years, several studies have demonstrated that—rather than reduce the incidence of ileus—inserting an NG tube and withholding regular feeding following abdominal surgery can cause an ileus or prolong a preexisting one. Other trials have shown that feeding a patient early in the postoperative period can actually prevent ileus.^7,8

Physiologic studies have shown that neither electrical activity of the bowel nor motor activities in the stomach are affected by surgery.⁸ Schilder et al⁹ reported bowel activity before the passage of flatus, indicating that the bowel is on its way to recovery much earlier than had been assumed. Thus, early postoperative feeding is well tolerated in most patients and associated with reduced discomfort and a more rapid recovery.⁷

For example, MacMillan et al studied 139 women undergoing gynecologic surgery for benign conditions; 67 were randomized to “early” feeding and 72 to traditional management. Early feeding involved a low-residue diet given 6 hours postoperatively, while traditional feeding consisted of clear liquids, which were withheld until the return of normal bowel sounds. Patients progressed to a regular diet with the passage of flatus. No increase in gastrointestinal complaints occurred in the early feeding group.⁷

Pearl et al⁸ compared similar groups of patients (TABLE 1). Patients in the first group were fed a clear liquid diet on the first postoperative day and progressed to a regular diet as soon as it could be tolerated. The traditional group was not fed until the return of bowel function, which was defined as the passage of flatus and no abdominal distension or vomiting; they were then started on clear liquids and, later, solid foods. While the incidence of complications was the same in both groups, hospitalization was shorter in the early feeding group.

A Cochrane review compared early versus delayed oral fluids and food after cesarean delivery. Of 12 studies considered, 6 were included in the review. No evidence was found to justify a policy of withholding oral fluids after uncomplicated cesarean sections.⁵

Simple versus complex procedures. Early feeding is not only safe in standardized surgeries such as cesarean section, but extends to complicated surgeries as well, as demonstrated in the trials by MacMillan et al⁷ and Pearl et al,⁸ which involved major gynecologic surgery. Trials with gynecologic oncology patients have shown the same result.^2,8 Even patients undergoing colorectal surgery can tolerate oral feeding very early in their postoperative course without bowel complications.^2,8

 

Not all studies have reported similar findings. Several concluded that other, nonmedical reasons, such as insurance requirements, accounted for the shorter hospital stays in many patients receiving early feeding.

TABLE 1

Complication rates associated with early feeding versus traditional management

COMPLICATION	EARLY FEEDING* (N=92)	TRADITIONAL MANAGEMENT· (N= 103)
Nausea	43.5%	24.3%
Nasogastric tube use	3.3%	6.7%
Febrile morbidity	54.3%	55.3%
Pneumonia	0%	1.9%
Wound complications	21.7%	21.4%
Atelectasis	8.7%	10.7%
Length of hospital stay (mean±standard deviation)	4.6±2.1	5.8±2.7
*Clear liquid diet on postoperative day 1
†Feeding delayed until return of bowel function
Reprinted from Obstet Gyn; vol 92; Pearl ML, Valea FA, Fischer M, Mahler L, Chalas E. A randomized controlled trial of early postoperative feeding in gynecologic oncology patients undergoing intraabdominal surgery; pages 94-97; copyright 1998; with permission from American College of Obstetricians and Gynecologists.

Nasogastric tubes only when indicated

The use of NG tubes after laparotomy has been studied extensively. A review of the literature suggests that routine placement of the tubes in asymptomatic patients is not justified and may possibly be harmful.¹⁰

In their meta-analysis of the issue, Cheatham et al¹¹ showed that although abdominal distension and vomiting are more frequent in patients who forgo NG tubes postoperatively, fever, atelectasis, and pneumonia are less common, and the interval between surgery and oral feeding is reduced (TABLE 2).

The authors concluded that for every NG tube inserted after abdominal surgery, at least 20 patients can be managed without it.

Forgoing an NG tube also lowers the risk of pulmonary complications, which increases 10-fold when a tube is inserted.¹²

TABLE 2

Complications associated with selective versus routine NG tube placement

	SELECTIVE PLACEMENT (N)	ROUTINE PLACEMENT (N)	P VALUE	RELATIVE RISK
Patients	1,986	1,978
Tubes placed/replaced	103	36	<.001>	2.9
Complications	833	1,084	.03	0.76
Deaths	13	25	.22	0.36
Pneumonia	53	119	<.0001>	0.49
Atelectasis	44	94	.001	0.46
Fever	108	212	.02	0.51
Vomiting	201	168	.11	1.19
Oral feedings (days)	3.53	4.59	.04
Length of stay (days)	9.32	10.1	.22
NG=nasogastric
Reprinted with permission from Cheatham ML, Chapman WC, Key SP, Sawyer JL. A meta-analysis of selective versus routine nasogastric decompression after elective laparotomy. Ann Surg. 1995:221:469-478.

Continue the epidural anesthetic

A number of experts believe that postoperative ileus is caused by stimulation of neural reflexes, which appear to be of 2 kinds: afferent stimuli to the spinal cord and efferent stimuli to the intestines through the sympathetic nervous system. The latter inhibits motility of the intestinal tract. Numerous studies demonstrate that this sympathetic reflex can be blocked by the use of epidural anesthesia.¹⁰

For example, Holte et al⁴ found that postoperative administration of thoracic epidural blockade with local anesthesia significantly reduced both ileus and pulmonary complications. They concluded that continuous epidural anesthesia with local anesthesia and minimally invasive surgery are the 2 most critical events in reducing postoperative ileus.

In a Cochrane review, Jorgensen and colleagues¹² compared the effects of epidural local anesthesia and opioid-based analgesic regimens on postoperative gastrointestinal paralysis, nausea and vomiting, and pain after abdominal surgery (TABLE 3). Epidural local anesthetics reduced gastrointestinal paralysis, as compared with systemic or epidural opioids, but provided the same postoperative pain relief. They also found that the addition of opioids to local epidural anesthesia provided superior postoperative analgesia—compared with epidural local anesthetics alone—without increasing the likelihood of ileus.

A study¹⁰ of patients undergoing colectomy found postoperative ileus was prevented or decreased with a 2-day regimen that included:

• continuous thoracic epidural anesthesia for 48 hours;
  
• withholding NG tubes;
  
• having the patient drink a liter of fluid on the day of surgery;
  
• initiating feeding after 24 hours;
  
• administering milk of magnesia; and
  
• mobilization after 8 hours, if possible.
  

Other studies have suggested that movement of the patient does not help eliminate postoperative ileus. It does, however, help prevent other postoperative complications, especially deep vein thrombosis.

Physicians in this study also used transverse surgical incisions to reduce pain and pulmonary problems.

TABLE 3

Anesthetic effect on GI function, postoperative pain, and nausea and vomiting: A comparison

ANESTHETIC	GI FUNCTION RETURNS	PAIN RELIEF	NAUSEA AND VOMITING
Epidural plus local	24 hr	Comparable	No significant difference
Epidural plus opioids	37 hr	Comparable	No significant difference
Systemic plus opioids	37 hr	Comparable	No significant difference
GI=gastrointestinal
Data from Jorgensen H, et al12

Clinical recommendations

The deregulation of the autonomic nervous system during surgery alters the gastrointestinal tract postoperatively, with neurotransmitters, local factors, and hormones playing a large role. Some forms of anesthesia also contribute to postoperative ileus, as does the use of narcotic analgesia after surgery.

The most efficient ways to activate the bowel postoperatively are:

• Continuing the thoracic epidural from 24 to 48 hours, which increases the splanchnic blood flow and blocks afferent and efferent sympathetic inhibitory nerve impulses. Note, however, that comparative studies of thoracic epidural anesthesia with local anesthesia are needed to quantify its impact.
  
• Hydrating the patient with a large amount of fluid in the first 24 hours after surgery.
  

Early feeding also seems to stimulate propulsive bowel motility.

Following these steps routinely can significantly decrease the risk of postoperative ileus and thus its resulting complications.

 

Dr. Rosenman reports no financial relationship with any companies whose products are mentioned in this article.