Master Class

KEVIN FOLEY, RESEARCH

KEVIN FOLEY, RESEARCH

KEVIN FOLEY, RESEARCH

Stroke is a rare but potentially devastating occurrence during pregnancy and the postpartum period. Maternal mortality is reported to be as high as 26%, and survivors may face long-term neurologic sequelae. Associated fetal mortality and morbidity also remain high. Although stroke is not preventable, early intervention can be key to saving lives and preserving brain function in some patients.

Therefore, even if the likelihood is that an obstetrician will never encounter a patient who suffers a stroke, it is critical that we all make the diagnosis promptly, obtain neurologic and other consultations appropriately, and direct state-of-the-art treatment of cerebrovascular disorders in this special population.

It is important to recognize that stroke occurs in young women of childbearing age at a rate of 10.7 per 100,000. Some have postulated that the risk is elevated during pregnancy for a number of reasons, including hypercoagulability, venous stasis, and blood pressure fluctuations. Indeed, some estimate that the risk of stroke is 13-fold higher in pregnant than in nonpregnant women, although the rarity of the condition makes the true prevalence a matter of debate.

Postpartum Risk Is Higher

The risk of stroke in the postpartum period is almost certainly higher still. Postpartum strokes generally occur from 5 days to 2 weeks after delivery—a vulnerable time when a headache from cerebral vasoconstriction syndrome may be mistaken for postepidural puncture syndrome.

The issue of stroke during pregnancy and the postpartum period is of increasing relevance to obstetricians. The incidence of stroke rises with age, and women are becoming pregnant at older ages than ever before; obstetric patients aged 45–50 years have become increasingly common. Obesity is also a risk factor for stroke, and pregnant women mirror American society at large, in which obesity has become epidemic. Longstanding hypertension and diabetes mellitus, both associated with obesity, further increase the risk of stroke.

Stroke can happen at any time. The clinical presentation is similar to that seen in nonpregnant patients; however, these symptoms can mimic those seen in preeclampsia and eclampsia, and the possibility of stroke may be overlooked.

Another potential delay in diagnosis and treatment may arise in patients who, like many nonpregnant patients, hesitate to seek immediate medical care when experiencing symptoms of stroke. Headache, for example, is common in pregnancy. Some women do not take it seriously unless it is very severe.

It is prudent to counsel all patients—and especially those with relevant risk factors—to seek care for any symptom that may be associated with stroke: headache; visual changes; epigastric pain; seizures; nausea and vomiting; or neurologic defects (focal or global). Severe hypertension and widened pulse pressures are symptomatic of stroke as well.

Cerebrovascular events may be associated with drug ingestion, infection, neoplasms, or trauma, as well as with metabolic factors. Researchers have been unable to determine the etiology of stroke in 23%–32% of cases.

Many medical conditions increase the risk of stroke, including cardioembolic diseases such as rheumatic heart disease or atrial septal defects; atherosclerosis (which may account for 15%–25% of cerebral infarctions in pregnancy); and disorders such as sickle cell disease, thrombophilias, central venous thrombosis, arteriovenous malformations and aneurysms, and cerebral vasoconstrictive syndrome.

Risk factors believed to be associated with pregnancy and the postpartum period include cesarean delivery and pregnancy-related hypertension. The link to hypertensive disorders of pregnancy seems to be an association in numerous studies, yet it is important to recognize that hypertension may be the result of a stroke rather than its cause. Seizures and central neurologic insult can undermine cerebral autoregulation and cause blood pressure to rise. Eclampsia and stroke produce similar clinical laboratory and neuroimaging findings.

Review of Stroke Cases

In a review of our 20-year institutional experience with postpartum stroke, we identified 20 cases among 130,000 deliveries at the E.H. Crump Women's Hospital of the University of Tennessee in Memphis (Am. J. Obstet. Gynecol. 2000;183:83–8).

In all cases, patients were discharged and later readmitted with findings consistent with a diagnosis of stroke. The mean age of these patients was 26 years; the mean gestational age at delivery was 27.3 weeks; and the mean birthweight was 2,617 grams, representing a group of patients with earlier deliveries and lower birthweights who were somewhat older than our obstetric population in general.

Of these 20 patients, 6 had a history of chronic hypertension. Preeclampsia was present in four pregnancies, and eight patients (40%) underwent a cesarean delivery, a rate approximately threefold higher than that in our overall obstetric population.

C-sections were performed for various reasons: three for nonreassuring fetal status, three for failure to progress, one for malpresentation, and one as an elective repeat C-section. Importantly, no intrapartum or postpartum complications presaged stroke in any patient. A review of antepartum, intrapartum, and in-hospital postpartum blood pressure values found normal measurements or well-controlled blood pressure in all women except one, whose blood pressure was elevated on postpartum day 3.

We found no correlation among trial of labor, mode of delivery, and the type of stroke suffered or outcome. We found no association with anesthesia type and stroke.

Presenting symptoms included headache in 16 patients, neurologic signs (motor weakness, blindness, aphasia, or coma) in 12, seizures in 11, and visual changes in 10.

Two deaths occurred, one associated with a cocaine-related hemorrhage and the other with a ruptured anatomical malformation.

A review of the medical records suggested that no case could have been prevented by the recognition of risk factors or alternative medical care.

In the diagnosis of stroke, the history and physical examination are critically important, although imaging is often needed to further clarify the diagnosis and distinguish stroke type and features.

The two basic causes of stroke are ischemia and postpartum hemorrhage.

Ischemia (infarction) may be caused by vascular thrombosis, embolism, vasospasm, or marked reduction in systemic perfusion pressure (hypotensive ischemic injury).

Postpartum Hemorrhage

Postpartum hemorrhage is divided into intracerebral or subarachnoid hemorrhage. Intracerebral hemorrhage refers to bleeding directly into brain tissue from small intracerebral vessels, caused by severe hypertension and/or coagulopathy or vascular malformations. In subarachnoid hemorrhage, blood seeps onto the brain's surface and merges with ambient cerebrospinal fluid, often as a result of a ruptured aneurysm or arteriovenous malformation.

Strokes of arterial origin most often occur during the second or third trimester, or within the first few weeks post partum. Symptoms include acute decompensation of cortical function, aphasia, hemiplegia, and/or hemianopsia.

Strokes of venous origin, on the other hand, are most likely to occur from 3 days to 4 weeks after delivery. Common symptoms include a severe, progressive headache; papilledema; weakness; convulsions; and/or aphasia.

Our review of 20 postpartum cases involved cerebritis (12 women), venous infarction (7 women), arterial infarction (6), intracerebral hemorrhage (5), and atrophy (1).

Among the 18 patients who survived, 12 suffered no residual defects, whereas others suffered hemiparesis, aphasia, or weakness. Although not statistically significant, there appeared to be a trend toward more adverse outcomes in women who suffered intracerebral hemorrhage vs. cerebral infarction.

Neuroimaging studies should be performed in any patient with symptoms that may be consistent with stroke. CT is widely available and may be very useful in confirming the diagnosis; however, a negative CT should not rule out further testing in the face of suspicious symptoms and/or physical examination findings. In our series, the initial CT was negative in 3 of 20 patients, with subsequent MRI or MRI angiography required to accurately diagnose stroke and elucidate its features.

Four-vessel traditional angiography, echocardiography, and lumbar puncture are other diagnostic modalities.

Of course, appropriate consultation is an integral part of stroke management, and may include a maternal-fetal medicine specialist as well as neurologists, neurosurgeons, radiologists, anesthesiologists, and later, rehabilitation specialists, social workers, and physical and occupational therapists.

Treatment hinges on protecting salvageable brain tissue; stabilizing the patient and preventing further complications such as aspiration; controlling blood pressure and other physiologic factors; and initiating physical rehabilitation.

As evidenced in our series, young patients have a great capacity for recovery in many cases.

My overall recommendation for stroke management is to admit the patient to labor and delivery, perform a thorough maternal and fetal evaluation, and use a multidisciplinary approach to care throughout.

Order antihypertensive medication if the systolic blood pressure is 160 mm Hg or greater, the diastolic blood pressure is 110 mm Hg or greater, or if the mean arterial pressure is 125 mm Hg or greater. Antiseizure, antiemetic, and anticoagulation medications should be administered as needed.

Neither medications nor surgery should be withheld because of pregnancy.

Deliver the patient in cases of maternal instability or nonreassuring fetal status; labor or rupture of membranes; or gestational age greater than 34 weeks.

When Intervention Is Necessary

At a gestational age of less than 24 weeks, intervention should be guided by the woman's diagnosis and condition.

Between 24 and 32 weeks' gestation, administer steroids for fetal lung maturity and conduct daily reassessments of the maternal and fetal condition with a planned delivery at 34 weeks, or term delivery if circumstances warrant.

Between 33 and 34 weeks, steroids should be administered and the baby delivered.

The nature of the stroke will determine the best course of medical and surgical management for the cerebrovascular event.

When anticoagulation is needed, keep in mind that warfarin crosses the placenta and has been linked to teratogenicity in the first trimester and bleeding complications in the third trimester. Heparin has been associated with thrombocytopenia, osteoporosis, and bleeding disorders, although it does not cross the placenta. After vaginal delivery, I recommend withholding anticoagulation for 6 hours. I recommend withholding anticoagulation for 12 hours following C-section.

When infection is implicated in stroke, antibiotics should be administered along with anticoagulation.

Low-dose aspirin may be sufficient to treat patients with a single episode of transient ischemic attack.

Hypertensive encephalopathy requires intensive management during labor and for 48 hours post partum. Response to antihypertensive therapy confirms the diagnosis. Volume contraction may be present, evidenced by a sharp drop in diastolic blood pressure and a rise in heart rate on standing from the supine position. Normal saline infusion for 24–48 hours may be considered to achieve volume expansion, decrease the activity of the renin-angiotensin-aldosterone axis, and maintain better blood pressure control. Careful attention should be paid to volume status, blood pressure urinary output, electrocardiographic readings, and mental status. Antepartum patients should have continuous fetal monitoring.

Eclampsia is treated with supportive care, including oxygenation; minimization of aspiration and future injuries; and lowering of blood pressure. Magnesium sulfate is used for the prevention of eclamptic seizures, although seizures may persist in 10% of patients. The medication should be maintained throughout labor and for 24 hours post partum.

Disturbances in the fetal heart rate are commonly seen after an eclamptic seizure, although resolution usually occurs within 5–10 minutes. Proceed to cesarean delivery only for obstetric indications, as vaginal delivery is preferred following a seizure.

Labor can be induced with oxytocin or prostaglandins.

Carefully monitor the patient's overall fluid status. These patients may have profound hemoconcentration, which necessitates close hemodynamic monitoring when epidural anesthesia is used and after severe blood loss. Acute blood loss can be a serious complication in hypovolemic patients. Limit fluids to prevent pulmonary edema secondary to capillary leakage.

Thrombolytic therapy with intraarterial recombinant tissue plasminogen activator has been used to treat ischemic stroke in pregnancy. This treatment must be administered within a window of 6 hours or less to be effective.

Course of Treatment

Patients who receive a diagnosis of arteriovenous malformation or aneurysm before hemorrhage should be referred for surgical embolization or clipping, as it is believed that patients with AVM may be at increased risk of bleeding during pregnancy. Patients with AVMs are also prone to bleed during delivery.

Once an intracerebral hemorrhage has occurred, the extent of the bleeding will determine the course of treatment. If the brain stem is compromised, surgical decompression is necessary. Surgery may also be necessary if the bleed is subarachnoid in origin; however, surgery itself may damage overlying normal brain tissue, and surgical morbidity is high.

If the bleeding and the patient are stable, surgery can be avoided. Blood pressure should be well controlled and seizures prevented. Steroids have not proven beneficial.

In summary, I would encourage obstetricians to become well versed in the symptoms of stroke and to have a low threshold for clinical suspicion of such symptoms, which may mimic common complaints of pregnancy.

A rapid diagnosis and close consultation with an interdisciplinary team of colleagues may maximizing outcome in patients suffering one of the most feared and serious complications of pregnancy.

This CT shows the brain of a pregnant woman who had a hemorrhagic stroke.

This MRI shows the brain of a pregnant woman who had an ischemic stroke. Photos courtesy Dr. Baha Sibai

How to Weigh Stroke History, Pregnancy

With enhanced diagnosis and management of stroke, more patients are recovering well. These patients will increasingly seek advice about the risks of subsequent pregnancy.

We recently conducted a review of 35 pregnancies in 23 women with a history of stroke, including 9 pregnancies in 4 women whose previous stroke had occurred during pregnancy or the postpartum period (Am. J. Obstet. Gynecol. 2004;190:1331–4).

Their risk factors for the prior stroke included thrombophilias, sickle cell disease, cardiac malformations, hypertension, oral contraceptive use, cerebral arteriovenous malformations, head trauma, meningitis, endocarditis, and idiopathic etiologies.

Anticoagulation was prescribed in two pregnancies in patients who had a reported history of pregnancy/postpartum stroke. Our findings were reassuring.

There were no recurrent thrombotic episodes during pregnancy or the postpartum period, although one patient required admission to the ICU for uncontrolled hypertension.

This result aligns with findings from another study (Neurology 2000;55:269–74), which found a 1% recurrence rate. The chance of recurrence in patients with thrombophilia is likely higher, perhaps 20%.

It makes sense to prescribe anticoagulation for women at risk of thromboembolic stroke, either low-dose aspirin plus prophylactic doses of unfractionated heparin, or low-molecular-weight heparin.

All women with a history of stroke deserve close monitoring during pregnancy, delivery, and the postpartum period.

Recognizing Stroke Is a Life or Death Issue

In high-risk obstetrics, we are faced with a variety of critical presentations: threatened early pregnancy loss, pregnant patients with preexisting medical conditions, and the development of acute medical complications during pregnancy.

Fortunately, the most serious complications of pregnancy—seizures, myocardial infarctions, and cerebrovascular events—are rare.

Despite their infrequency, we must become sufficiently familiar with the presenting symptoms of these life-threatening complications of pregnancy to be able to employ our clinical skills at a moment's notice.

Recent advances in stroke management make quick action vital. Medications or surgery not only may save the lives of the woman and fetus, but may preserve brain function in the woman as well.

Very often, but not always, clues to the possibility of a stroke may lie in the woman's medical history of hypertension, preeclampsia, seizures, or a bleeding disorder.

We need to be alert to these risk factors for stroke, and make sure that we include stroke in the differential diagnosis even when the patient's symptoms may be masquerading as less serious complications of pregnancy.

This month, my guest Master Class professor is Baha Sibai, M.D., a world expert on hypertensive disorders and complications in pregnancy who has authored or coauthored more than 350 peer-reviewed publications on the topics of preeclampsia and eclampsia.

Dr. Sibai is professor and director of the department of obstetrics and gynecology at the University of Cincinnati. He formerly directed the division of maternal/fetal medicine at the University of Tennessee, Memphis.

DR. REECE, who specializes in maternal-fetal medicine, is the vice chancellor and dean of the college of medicine at the University of Arkansas in Little Rock.

E. ALBERT REECE, M.D., PH.D., M.B.A.

Stroke is a rare but potentially devastating occurrence during pregnancy and the postpartum period. Maternal mortality is reported to be as high as 26%, and survivors may face long-term neurologic sequelae. Associated fetal mortality and morbidity also remain high. Although stroke is not preventable, early intervention can be key to saving lives and preserving brain function in some patients.

Therefore, even if the likelihood is that an obstetrician will never encounter a patient who suffers a stroke, it is critical that we all make the diagnosis promptly, obtain neurologic and other consultations appropriately, and direct state-of-the-art treatment of cerebrovascular disorders in this special population.

It is important to recognize that stroke occurs in young women of childbearing age at a rate of 10.7 per 100,000. Some have postulated that the risk is elevated during pregnancy for a number of reasons, including hypercoagulability, venous stasis, and blood pressure fluctuations. Indeed, some estimate that the risk of stroke is 13-fold higher in pregnant than in nonpregnant women, although the rarity of the condition makes the true prevalence a matter of debate.

Postpartum Risk Is Higher

The risk of stroke in the postpartum period is almost certainly higher still. Postpartum strokes generally occur from 5 days to 2 weeks after delivery—a vulnerable time when a headache from cerebral vasoconstriction syndrome may be mistaken for postepidural puncture syndrome.

The issue of stroke during pregnancy and the postpartum period is of increasing relevance to obstetricians. The incidence of stroke rises with age, and women are becoming pregnant at older ages than ever before; obstetric patients aged 45–50 years have become increasingly common. Obesity is also a risk factor for stroke, and pregnant women mirror American society at large, in which obesity has become epidemic. Longstanding hypertension and diabetes mellitus, both associated with obesity, further increase the risk of stroke.

Stroke can happen at any time. The clinical presentation is similar to that seen in nonpregnant patients; however, these symptoms can mimic those seen in preeclampsia and eclampsia, and the possibility of stroke may be overlooked.

Another potential delay in diagnosis and treatment may arise in patients who, like many nonpregnant patients, hesitate to seek immediate medical care when experiencing symptoms of stroke. Headache, for example, is common in pregnancy. Some women do not take it seriously unless it is very severe.

It is prudent to counsel all patients—and especially those with relevant risk factors—to seek care for any symptom that may be associated with stroke: headache; visual changes; epigastric pain; seizures; nausea and vomiting; or neurologic defects (focal or global). Severe hypertension and widened pulse pressures are symptomatic of stroke as well.

Cerebrovascular events may be associated with drug ingestion, infection, neoplasms, or trauma, as well as with metabolic factors. Researchers have been unable to determine the etiology of stroke in 23%–32% of cases.

Many medical conditions increase the risk of stroke, including cardioembolic diseases such as rheumatic heart disease or atrial septal defects; atherosclerosis (which may account for 15%–25% of cerebral infarctions in pregnancy); and disorders such as sickle cell disease, thrombophilias, central venous thrombosis, arteriovenous malformations and aneurysms, and cerebral vasoconstrictive syndrome.

Risk factors believed to be associated with pregnancy and the postpartum period include cesarean delivery and pregnancy-related hypertension. The link to hypertensive disorders of pregnancy seems to be an association in numerous studies, yet it is important to recognize that hypertension may be the result of a stroke rather than its cause. Seizures and central neurologic insult can undermine cerebral autoregulation and cause blood pressure to rise. Eclampsia and stroke produce similar clinical laboratory and neuroimaging findings.

Review of Stroke Cases

In a review of our 20-year institutional experience with postpartum stroke, we identified 20 cases among 130,000 deliveries at the E.H. Crump Women's Hospital of the University of Tennessee in Memphis (Am. J. Obstet. Gynecol. 2000;183:83–8).

In all cases, patients were discharged and later readmitted with findings consistent with a diagnosis of stroke. The mean age of these patients was 26 years; the mean gestational age at delivery was 27.3 weeks; and the mean birthweight was 2,617 grams, representing a group of patients with earlier deliveries and lower birthweights who were somewhat older than our obstetric population in general.

Of these 20 patients, 6 had a history of chronic hypertension. Preeclampsia was present in four pregnancies, and eight patients (40%) underwent a cesarean delivery, a rate approximately threefold higher than that in our overall obstetric population.

C-sections were performed for various reasons: three for nonreassuring fetal status, three for failure to progress, one for malpresentation, and one as an elective repeat C-section. Importantly, no intrapartum or postpartum complications presaged stroke in any patient. A review of antepartum, intrapartum, and in-hospital postpartum blood pressure values found normal measurements or well-controlled blood pressure in all women except one, whose blood pressure was elevated on postpartum day 3.

We found no correlation among trial of labor, mode of delivery, and the type of stroke suffered or outcome. We found no association with anesthesia type and stroke.

Presenting symptoms included headache in 16 patients, neurologic signs (motor weakness, blindness, aphasia, or coma) in 12, seizures in 11, and visual changes in 10.

Two deaths occurred, one associated with a cocaine-related hemorrhage and the other with a ruptured anatomical malformation.

A review of the medical records suggested that no case could have been prevented by the recognition of risk factors or alternative medical care.

In the diagnosis of stroke, the history and physical examination are critically important, although imaging is often needed to further clarify the diagnosis and distinguish stroke type and features.

The two basic causes of stroke are ischemia and postpartum hemorrhage.

Ischemia (infarction) may be caused by vascular thrombosis, embolism, vasospasm, or marked reduction in systemic perfusion pressure (hypotensive ischemic injury).

Postpartum Hemorrhage

Postpartum hemorrhage is divided into intracerebral or subarachnoid hemorrhage. Intracerebral hemorrhage refers to bleeding directly into brain tissue from small intracerebral vessels, caused by severe hypertension and/or coagulopathy or vascular malformations. In subarachnoid hemorrhage, blood seeps onto the brain's surface and merges with ambient cerebrospinal fluid, often as a result of a ruptured aneurysm or arteriovenous malformation.

Strokes of arterial origin most often occur during the second or third trimester, or within the first few weeks post partum. Symptoms include acute decompensation of cortical function, aphasia, hemiplegia, and/or hemianopsia.

Strokes of venous origin, on the other hand, are most likely to occur from 3 days to 4 weeks after delivery. Common symptoms include a severe, progressive headache; papilledema; weakness; convulsions; and/or aphasia.

Our review of 20 postpartum cases involved cerebritis (12 women), venous infarction (7 women), arterial infarction (6), intracerebral hemorrhage (5), and atrophy (1).

Among the 18 patients who survived, 12 suffered no residual defects, whereas others suffered hemiparesis, aphasia, or weakness. Although not statistically significant, there appeared to be a trend toward more adverse outcomes in women who suffered intracerebral hemorrhage vs. cerebral infarction.

Neuroimaging studies should be performed in any patient with symptoms that may be consistent with stroke. CT is widely available and may be very useful in confirming the diagnosis; however, a negative CT should not rule out further testing in the face of suspicious symptoms and/or physical examination findings. In our series, the initial CT was negative in 3 of 20 patients, with subsequent MRI or MRI angiography required to accurately diagnose stroke and elucidate its features.

Four-vessel traditional angiography, echocardiography, and lumbar puncture are other diagnostic modalities.

Of course, appropriate consultation is an integral part of stroke management, and may include a maternal-fetal medicine specialist as well as neurologists, neurosurgeons, radiologists, anesthesiologists, and later, rehabilitation specialists, social workers, and physical and occupational therapists.

Treatment hinges on protecting salvageable brain tissue; stabilizing the patient and preventing further complications such as aspiration; controlling blood pressure and other physiologic factors; and initiating physical rehabilitation.

As evidenced in our series, young patients have a great capacity for recovery in many cases.

My overall recommendation for stroke management is to admit the patient to labor and delivery, perform a thorough maternal and fetal evaluation, and use a multidisciplinary approach to care throughout.

Order antihypertensive medication if the systolic blood pressure is 160 mm Hg or greater, the diastolic blood pressure is 110 mm Hg or greater, or if the mean arterial pressure is 125 mm Hg or greater. Antiseizure, antiemetic, and anticoagulation medications should be administered as needed.

Neither medications nor surgery should be withheld because of pregnancy.

Deliver the patient in cases of maternal instability or nonreassuring fetal status; labor or rupture of membranes; or gestational age greater than 34 weeks.

When Intervention Is Necessary

At a gestational age of less than 24 weeks, intervention should be guided by the woman's diagnosis and condition.

Between 24 and 32 weeks' gestation, administer steroids for fetal lung maturity and conduct daily reassessments of the maternal and fetal condition with a planned delivery at 34 weeks, or term delivery if circumstances warrant.

Between 33 and 34 weeks, steroids should be administered and the baby delivered.

The nature of the stroke will determine the best course of medical and surgical management for the cerebrovascular event.

When anticoagulation is needed, keep in mind that warfarin crosses the placenta and has been linked to teratogenicity in the first trimester and bleeding complications in the third trimester. Heparin has been associated with thrombocytopenia, osteoporosis, and bleeding disorders, although it does not cross the placenta. After vaginal delivery, I recommend withholding anticoagulation for 6 hours. I recommend withholding anticoagulation for 12 hours following C-section.

When infection is implicated in stroke, antibiotics should be administered along with anticoagulation.

Low-dose aspirin may be sufficient to treat patients with a single episode of transient ischemic attack.

Hypertensive encephalopathy requires intensive management during labor and for 48 hours post partum. Response to antihypertensive therapy confirms the diagnosis. Volume contraction may be present, evidenced by a sharp drop in diastolic blood pressure and a rise in heart rate on standing from the supine position. Normal saline infusion for 24–48 hours may be considered to achieve volume expansion, decrease the activity of the renin-angiotensin-aldosterone axis, and maintain better blood pressure control. Careful attention should be paid to volume status, blood pressure urinary output, electrocardiographic readings, and mental status. Antepartum patients should have continuous fetal monitoring.

Eclampsia is treated with supportive care, including oxygenation; minimization of aspiration and future injuries; and lowering of blood pressure. Magnesium sulfate is used for the prevention of eclamptic seizures, although seizures may persist in 10% of patients. The medication should be maintained throughout labor and for 24 hours post partum.

Disturbances in the fetal heart rate are commonly seen after an eclamptic seizure, although resolution usually occurs within 5–10 minutes. Proceed to cesarean delivery only for obstetric indications, as vaginal delivery is preferred following a seizure.

Labor can be induced with oxytocin or prostaglandins.

Carefully monitor the patient's overall fluid status. These patients may have profound hemoconcentration, which necessitates close hemodynamic monitoring when epidural anesthesia is used and after severe blood loss. Acute blood loss can be a serious complication in hypovolemic patients. Limit fluids to prevent pulmonary edema secondary to capillary leakage.

Thrombolytic therapy with intraarterial recombinant tissue plasminogen activator has been used to treat ischemic stroke in pregnancy. This treatment must be administered within a window of 6 hours or less to be effective.

Course of Treatment

Patients who receive a diagnosis of arteriovenous malformation or aneurysm before hemorrhage should be referred for surgical embolization or clipping, as it is believed that patients with AVM may be at increased risk of bleeding during pregnancy. Patients with AVMs are also prone to bleed during delivery.

Once an intracerebral hemorrhage has occurred, the extent of the bleeding will determine the course of treatment. If the brain stem is compromised, surgical decompression is necessary. Surgery may also be necessary if the bleed is subarachnoid in origin; however, surgery itself may damage overlying normal brain tissue, and surgical morbidity is high.

If the bleeding and the patient are stable, surgery can be avoided. Blood pressure should be well controlled and seizures prevented. Steroids have not proven beneficial.

In summary, I would encourage obstetricians to become well versed in the symptoms of stroke and to have a low threshold for clinical suspicion of such symptoms, which may mimic common complaints of pregnancy.

A rapid diagnosis and close consultation with an interdisciplinary team of colleagues may maximizing outcome in patients suffering one of the most feared and serious complications of pregnancy.

This CT shows the brain of a pregnant woman who had a hemorrhagic stroke.

This MRI shows the brain of a pregnant woman who had an ischemic stroke. Photos courtesy Dr. Baha Sibai

How to Weigh Stroke History, Pregnancy

With enhanced diagnosis and management of stroke, more patients are recovering well. These patients will increasingly seek advice about the risks of subsequent pregnancy.

We recently conducted a review of 35 pregnancies in 23 women with a history of stroke, including 9 pregnancies in 4 women whose previous stroke had occurred during pregnancy or the postpartum period (Am. J. Obstet. Gynecol. 2004;190:1331–4).

Their risk factors for the prior stroke included thrombophilias, sickle cell disease, cardiac malformations, hypertension, oral contraceptive use, cerebral arteriovenous malformations, head trauma, meningitis, endocarditis, and idiopathic etiologies.

Anticoagulation was prescribed in two pregnancies in patients who had a reported history of pregnancy/postpartum stroke. Our findings were reassuring.

There were no recurrent thrombotic episodes during pregnancy or the postpartum period, although one patient required admission to the ICU for uncontrolled hypertension.

This result aligns with findings from another study (Neurology 2000;55:269–74), which found a 1% recurrence rate. The chance of recurrence in patients with thrombophilia is likely higher, perhaps 20%.

It makes sense to prescribe anticoagulation for women at risk of thromboembolic stroke, either low-dose aspirin plus prophylactic doses of unfractionated heparin, or low-molecular-weight heparin.

All women with a history of stroke deserve close monitoring during pregnancy, delivery, and the postpartum period.

Recognizing Stroke Is a Life or Death Issue

In high-risk obstetrics, we are faced with a variety of critical presentations: threatened early pregnancy loss, pregnant patients with preexisting medical conditions, and the development of acute medical complications during pregnancy.

Fortunately, the most serious complications of pregnancy—seizures, myocardial infarctions, and cerebrovascular events—are rare.

Despite their infrequency, we must become sufficiently familiar with the presenting symptoms of these life-threatening complications of pregnancy to be able to employ our clinical skills at a moment's notice.

Recent advances in stroke management make quick action vital. Medications or surgery not only may save the lives of the woman and fetus, but may preserve brain function in the woman as well.

Very often, but not always, clues to the possibility of a stroke may lie in the woman's medical history of hypertension, preeclampsia, seizures, or a bleeding disorder.

We need to be alert to these risk factors for stroke, and make sure that we include stroke in the differential diagnosis even when the patient's symptoms may be masquerading as less serious complications of pregnancy.

This month, my guest Master Class professor is Baha Sibai, M.D., a world expert on hypertensive disorders and complications in pregnancy who has authored or coauthored more than 350 peer-reviewed publications on the topics of preeclampsia and eclampsia.

Dr. Sibai is professor and director of the department of obstetrics and gynecology at the University of Cincinnati. He formerly directed the division of maternal/fetal medicine at the University of Tennessee, Memphis.

DR. REECE, who specializes in maternal-fetal medicine, is the vice chancellor and dean of the college of medicine at the University of Arkansas in Little Rock.

E. ALBERT REECE, M.D., PH.D., M.B.A.

Stroke is a rare but potentially devastating occurrence during pregnancy and the postpartum period. Maternal mortality is reported to be as high as 26%, and survivors may face long-term neurologic sequelae. Associated fetal mortality and morbidity also remain high. Although stroke is not preventable, early intervention can be key to saving lives and preserving brain function in some patients.

Therefore, even if the likelihood is that an obstetrician will never encounter a patient who suffers a stroke, it is critical that we all make the diagnosis promptly, obtain neurologic and other consultations appropriately, and direct state-of-the-art treatment of cerebrovascular disorders in this special population.

It is important to recognize that stroke occurs in young women of childbearing age at a rate of 10.7 per 100,000. Some have postulated that the risk is elevated during pregnancy for a number of reasons, including hypercoagulability, venous stasis, and blood pressure fluctuations. Indeed, some estimate that the risk of stroke is 13-fold higher in pregnant than in nonpregnant women, although the rarity of the condition makes the true prevalence a matter of debate.

Postpartum Risk Is Higher

The risk of stroke in the postpartum period is almost certainly higher still. Postpartum strokes generally occur from 5 days to 2 weeks after delivery—a vulnerable time when a headache from cerebral vasoconstriction syndrome may be mistaken for postepidural puncture syndrome.

The issue of stroke during pregnancy and the postpartum period is of increasing relevance to obstetricians. The incidence of stroke rises with age, and women are becoming pregnant at older ages than ever before; obstetric patients aged 45–50 years have become increasingly common. Obesity is also a risk factor for stroke, and pregnant women mirror American society at large, in which obesity has become epidemic. Longstanding hypertension and diabetes mellitus, both associated with obesity, further increase the risk of stroke.

Stroke can happen at any time. The clinical presentation is similar to that seen in nonpregnant patients; however, these symptoms can mimic those seen in preeclampsia and eclampsia, and the possibility of stroke may be overlooked.

Another potential delay in diagnosis and treatment may arise in patients who, like many nonpregnant patients, hesitate to seek immediate medical care when experiencing symptoms of stroke. Headache, for example, is common in pregnancy. Some women do not take it seriously unless it is very severe.

It is prudent to counsel all patients—and especially those with relevant risk factors—to seek care for any symptom that may be associated with stroke: headache; visual changes; epigastric pain; seizures; nausea and vomiting; or neurologic defects (focal or global). Severe hypertension and widened pulse pressures are symptomatic of stroke as well.

Cerebrovascular events may be associated with drug ingestion, infection, neoplasms, or trauma, as well as with metabolic factors. Researchers have been unable to determine the etiology of stroke in 23%–32% of cases.

Many medical conditions increase the risk of stroke, including cardioembolic diseases such as rheumatic heart disease or atrial septal defects; atherosclerosis (which may account for 15%–25% of cerebral infarctions in pregnancy); and disorders such as sickle cell disease, thrombophilias, central venous thrombosis, arteriovenous malformations and aneurysms, and cerebral vasoconstrictive syndrome.

Risk factors believed to be associated with pregnancy and the postpartum period include cesarean delivery and pregnancy-related hypertension. The link to hypertensive disorders of pregnancy seems to be an association in numerous studies, yet it is important to recognize that hypertension may be the result of a stroke rather than its cause. Seizures and central neurologic insult can undermine cerebral autoregulation and cause blood pressure to rise. Eclampsia and stroke produce similar clinical laboratory and neuroimaging findings.

Review of Stroke Cases

In a review of our 20-year institutional experience with postpartum stroke, we identified 20 cases among 130,000 deliveries at the E.H. Crump Women's Hospital of the University of Tennessee in Memphis (Am. J. Obstet. Gynecol. 2000;183:83–8).

In all cases, patients were discharged and later readmitted with findings consistent with a diagnosis of stroke. The mean age of these patients was 26 years; the mean gestational age at delivery was 27.3 weeks; and the mean birthweight was 2,617 grams, representing a group of patients with earlier deliveries and lower birthweights who were somewhat older than our obstetric population in general.

Of these 20 patients, 6 had a history of chronic hypertension. Preeclampsia was present in four pregnancies, and eight patients (40%) underwent a cesarean delivery, a rate approximately threefold higher than that in our overall obstetric population.

C-sections were performed for various reasons: three for nonreassuring fetal status, three for failure to progress, one for malpresentation, and one as an elective repeat C-section. Importantly, no intrapartum or postpartum complications presaged stroke in any patient. A review of antepartum, intrapartum, and in-hospital postpartum blood pressure values found normal measurements or well-controlled blood pressure in all women except one, whose blood pressure was elevated on postpartum day 3.

We found no correlation among trial of labor, mode of delivery, and the type of stroke suffered or outcome. We found no association with anesthesia type and stroke.

Presenting symptoms included headache in 16 patients, neurologic signs (motor weakness, blindness, aphasia, or coma) in 12, seizures in 11, and visual changes in 10.

Two deaths occurred, one associated with a cocaine-related hemorrhage and the other with a ruptured anatomical malformation.

A review of the medical records suggested that no case could have been prevented by the recognition of risk factors or alternative medical care.

In the diagnosis of stroke, the history and physical examination are critically important, although imaging is often needed to further clarify the diagnosis and distinguish stroke type and features.

The two basic causes of stroke are ischemia and postpartum hemorrhage.

Ischemia (infarction) may be caused by vascular thrombosis, embolism, vasospasm, or marked reduction in systemic perfusion pressure (hypotensive ischemic injury).

Postpartum Hemorrhage

Postpartum hemorrhage is divided into intracerebral or subarachnoid hemorrhage. Intracerebral hemorrhage refers to bleeding directly into brain tissue from small intracerebral vessels, caused by severe hypertension and/or coagulopathy or vascular malformations. In subarachnoid hemorrhage, blood seeps onto the brain's surface and merges with ambient cerebrospinal fluid, often as a result of a ruptured aneurysm or arteriovenous malformation.

Strokes of arterial origin most often occur during the second or third trimester, or within the first few weeks post partum. Symptoms include acute decompensation of cortical function, aphasia, hemiplegia, and/or hemianopsia.

Strokes of venous origin, on the other hand, are most likely to occur from 3 days to 4 weeks after delivery. Common symptoms include a severe, progressive headache; papilledema; weakness; convulsions; and/or aphasia.

Our review of 20 postpartum cases involved cerebritis (12 women), venous infarction (7 women), arterial infarction (6), intracerebral hemorrhage (5), and atrophy (1).

Among the 18 patients who survived, 12 suffered no residual defects, whereas others suffered hemiparesis, aphasia, or weakness. Although not statistically significant, there appeared to be a trend toward more adverse outcomes in women who suffered intracerebral hemorrhage vs. cerebral infarction.

Neuroimaging studies should be performed in any patient with symptoms that may be consistent with stroke. CT is widely available and may be very useful in confirming the diagnosis; however, a negative CT should not rule out further testing in the face of suspicious symptoms and/or physical examination findings. In our series, the initial CT was negative in 3 of 20 patients, with subsequent MRI or MRI angiography required to accurately diagnose stroke and elucidate its features.

Four-vessel traditional angiography, echocardiography, and lumbar puncture are other diagnostic modalities.

Of course, appropriate consultation is an integral part of stroke management, and may include a maternal-fetal medicine specialist as well as neurologists, neurosurgeons, radiologists, anesthesiologists, and later, rehabilitation specialists, social workers, and physical and occupational therapists.

Treatment hinges on protecting salvageable brain tissue; stabilizing the patient and preventing further complications such as aspiration; controlling blood pressure and other physiologic factors; and initiating physical rehabilitation.

As evidenced in our series, young patients have a great capacity for recovery in many cases.

My overall recommendation for stroke management is to admit the patient to labor and delivery, perform a thorough maternal and fetal evaluation, and use a multidisciplinary approach to care throughout.

Order antihypertensive medication if the systolic blood pressure is 160 mm Hg or greater, the diastolic blood pressure is 110 mm Hg or greater, or if the mean arterial pressure is 125 mm Hg or greater. Antiseizure, antiemetic, and anticoagulation medications should be administered as needed.

Neither medications nor surgery should be withheld because of pregnancy.

Deliver the patient in cases of maternal instability or nonreassuring fetal status; labor or rupture of membranes; or gestational age greater than 34 weeks.

When Intervention Is Necessary

At a gestational age of less than 24 weeks, intervention should be guided by the woman's diagnosis and condition.

Between 24 and 32 weeks' gestation, administer steroids for fetal lung maturity and conduct daily reassessments of the maternal and fetal condition with a planned delivery at 34 weeks, or term delivery if circumstances warrant.

Between 33 and 34 weeks, steroids should be administered and the baby delivered.

The nature of the stroke will determine the best course of medical and surgical management for the cerebrovascular event.

When anticoagulation is needed, keep in mind that warfarin crosses the placenta and has been linked to teratogenicity in the first trimester and bleeding complications in the third trimester. Heparin has been associated with thrombocytopenia, osteoporosis, and bleeding disorders, although it does not cross the placenta. After vaginal delivery, I recommend withholding anticoagulation for 6 hours. I recommend withholding anticoagulation for 12 hours following C-section.

When infection is implicated in stroke, antibiotics should be administered along with anticoagulation.

Low-dose aspirin may be sufficient to treat patients with a single episode of transient ischemic attack.

Hypertensive encephalopathy requires intensive management during labor and for 48 hours post partum. Response to antihypertensive therapy confirms the diagnosis. Volume contraction may be present, evidenced by a sharp drop in diastolic blood pressure and a rise in heart rate on standing from the supine position. Normal saline infusion for 24–48 hours may be considered to achieve volume expansion, decrease the activity of the renin-angiotensin-aldosterone axis, and maintain better blood pressure control. Careful attention should be paid to volume status, blood pressure urinary output, electrocardiographic readings, and mental status. Antepartum patients should have continuous fetal monitoring.

Eclampsia is treated with supportive care, including oxygenation; minimization of aspiration and future injuries; and lowering of blood pressure. Magnesium sulfate is used for the prevention of eclamptic seizures, although seizures may persist in 10% of patients. The medication should be maintained throughout labor and for 24 hours post partum.

Disturbances in the fetal heart rate are commonly seen after an eclamptic seizure, although resolution usually occurs within 5–10 minutes. Proceed to cesarean delivery only for obstetric indications, as vaginal delivery is preferred following a seizure.

Labor can be induced with oxytocin or prostaglandins.

Carefully monitor the patient's overall fluid status. These patients may have profound hemoconcentration, which necessitates close hemodynamic monitoring when epidural anesthesia is used and after severe blood loss. Acute blood loss can be a serious complication in hypovolemic patients. Limit fluids to prevent pulmonary edema secondary to capillary leakage.

Thrombolytic therapy with intraarterial recombinant tissue plasminogen activator has been used to treat ischemic stroke in pregnancy. This treatment must be administered within a window of 6 hours or less to be effective.

Course of Treatment

Patients who receive a diagnosis of arteriovenous malformation or aneurysm before hemorrhage should be referred for surgical embolization or clipping, as it is believed that patients with AVM may be at increased risk of bleeding during pregnancy. Patients with AVMs are also prone to bleed during delivery.

Once an intracerebral hemorrhage has occurred, the extent of the bleeding will determine the course of treatment. If the brain stem is compromised, surgical decompression is necessary. Surgery may also be necessary if the bleed is subarachnoid in origin; however, surgery itself may damage overlying normal brain tissue, and surgical morbidity is high.

If the bleeding and the patient are stable, surgery can be avoided. Blood pressure should be well controlled and seizures prevented. Steroids have not proven beneficial.

In summary, I would encourage obstetricians to become well versed in the symptoms of stroke and to have a low threshold for clinical suspicion of such symptoms, which may mimic common complaints of pregnancy.

A rapid diagnosis and close consultation with an interdisciplinary team of colleagues may maximizing outcome in patients suffering one of the most feared and serious complications of pregnancy.

This CT shows the brain of a pregnant woman who had a hemorrhagic stroke.

This MRI shows the brain of a pregnant woman who had an ischemic stroke. Photos courtesy Dr. Baha Sibai

How to Weigh Stroke History, Pregnancy

With enhanced diagnosis and management of stroke, more patients are recovering well. These patients will increasingly seek advice about the risks of subsequent pregnancy.

We recently conducted a review of 35 pregnancies in 23 women with a history of stroke, including 9 pregnancies in 4 women whose previous stroke had occurred during pregnancy or the postpartum period (Am. J. Obstet. Gynecol. 2004;190:1331–4).

Their risk factors for the prior stroke included thrombophilias, sickle cell disease, cardiac malformations, hypertension, oral contraceptive use, cerebral arteriovenous malformations, head trauma, meningitis, endocarditis, and idiopathic etiologies.

Anticoagulation was prescribed in two pregnancies in patients who had a reported history of pregnancy/postpartum stroke. Our findings were reassuring.

There were no recurrent thrombotic episodes during pregnancy or the postpartum period, although one patient required admission to the ICU for uncontrolled hypertension.

This result aligns with findings from another study (Neurology 2000;55:269–74), which found a 1% recurrence rate. The chance of recurrence in patients with thrombophilia is likely higher, perhaps 20%.

It makes sense to prescribe anticoagulation for women at risk of thromboembolic stroke, either low-dose aspirin plus prophylactic doses of unfractionated heparin, or low-molecular-weight heparin.

All women with a history of stroke deserve close monitoring during pregnancy, delivery, and the postpartum period.

Recognizing Stroke Is a Life or Death Issue

In high-risk obstetrics, we are faced with a variety of critical presentations: threatened early pregnancy loss, pregnant patients with preexisting medical conditions, and the development of acute medical complications during pregnancy.

Fortunately, the most serious complications of pregnancy—seizures, myocardial infarctions, and cerebrovascular events—are rare.

Despite their infrequency, we must become sufficiently familiar with the presenting symptoms of these life-threatening complications of pregnancy to be able to employ our clinical skills at a moment's notice.

Recent advances in stroke management make quick action vital. Medications or surgery not only may save the lives of the woman and fetus, but may preserve brain function in the woman as well.

Very often, but not always, clues to the possibility of a stroke may lie in the woman's medical history of hypertension, preeclampsia, seizures, or a bleeding disorder.

We need to be alert to these risk factors for stroke, and make sure that we include stroke in the differential diagnosis even when the patient's symptoms may be masquerading as less serious complications of pregnancy.

This month, my guest Master Class professor is Baha Sibai, M.D., a world expert on hypertensive disorders and complications in pregnancy who has authored or coauthored more than 350 peer-reviewed publications on the topics of preeclampsia and eclampsia.

Dr. Sibai is professor and director of the department of obstetrics and gynecology at the University of Cincinnati. He formerly directed the division of maternal/fetal medicine at the University of Tennessee, Memphis.

DR. REECE, who specializes in maternal-fetal medicine, is the vice chancellor and dean of the college of medicine at the University of Arkansas in Little Rock.

E. ALBERT REECE, M.D., PH.D., M.B.A.

Generic Allegra-D Approved

A combination of fexofenadine HCl and pseudoephedrine HCl in an extended-release tablet (60 mg/120 mg) is indicated for the relief of symptoms associated with seasonal allergic rhinitis in adults and children 12 years and older. For more information, contact Barr Pharmaceuticals Inc. by calling 800-222-0190 or 201-930-3302.

Rapid Fecal Occult Blood Test

The Clearview ULTRA fecal occult blood test, a rapid two-step immunoassay, is now available. Results are available in less than 5 minutes and only one sample is required from the patient. The test does not require the patient to follow a restricted diet. For more information, contact Wampole Laboratories by calling 800-257-9525 or 609-627-8000.

Dictionary of Drug Names

The 2005 U.S. Pharmacopeia Dictionary of Adopted Names and International Drug Names is available. This 41st edition contains nationally and internationally recognized names for all drugs. The dictionary is available in book form for $299 or in a combination book-online package for $345. For more information, contact U.S. Pharmacopeia by visiting

http://store.usp.org

Fertility Treatment Launched

Menopur (menotropins for injection) is indicated for the development of multiple follicles and pregnancy in patients undergoing assisted reproductive technology procedures, such as in vitro fertilization. This purified human menopausal gonadotropin contains equal amounts of follicle-stimulating hormone and luteinizing hormone. For more information, contact Ferring Pharmaceuticals Inc. by visiting

www.ferring.com

Pigmentation Treatment

Aclaro (4% hydroquinone) emulsion treats ultraviolet-induced dyschromia and discoloration due to oral contraceptive use, pregnancy, hormone therapy, or skin trauma. For more information, contact JSJ Pharmaceuticals by visiting

www.aclaro4.com

Soft, Inflatable Speculum

The FemSpec is a soft, rolled, tamponlike speculum that the physician can inflate for a better fit during gynecologic exams. This disposable product is patient friendly. The speculum comes in small, medium, and large sizes. Contact FemSpec LLC by visiting the Web site

www.femspec.com

Alendronate/Vitamin D in One Tablet

Fosamax Plus D (alendronate sodium/cholecalciferol) is approved for treatment of osteoporosis in postmenopausal women. The once-weekly tablet combines 70-mg alendronate sodium with 2,800-IU vitamin D3. For more information, contact Merck & Co. by calling 800-344-7833.

Generic Allegra-D Approved

A combination of fexofenadine HCl and pseudoephedrine HCl in an extended-release tablet (60 mg/120 mg) is indicated for the relief of symptoms associated with seasonal allergic rhinitis in adults and children 12 years and older. For more information, contact Barr Pharmaceuticals Inc. by calling 800-222-0190 or 201-930-3302.

Rapid Fecal Occult Blood Test

The Clearview ULTRA fecal occult blood test, a rapid two-step immunoassay, is now available. Results are available in less than 5 minutes and only one sample is required from the patient. The test does not require the patient to follow a restricted diet. For more information, contact Wampole Laboratories by calling 800-257-9525 or 609-627-8000.

Dictionary of Drug Names

The 2005 U.S. Pharmacopeia Dictionary of Adopted Names and International Drug Names is available. This 41st edition contains nationally and internationally recognized names for all drugs. The dictionary is available in book form for $299 or in a combination book-online package for $345. For more information, contact U.S. Pharmacopeia by visiting

http://store.usp.org

Fertility Treatment Launched

Menopur (menotropins for injection) is indicated for the development of multiple follicles and pregnancy in patients undergoing assisted reproductive technology procedures, such as in vitro fertilization. This purified human menopausal gonadotropin contains equal amounts of follicle-stimulating hormone and luteinizing hormone. For more information, contact Ferring Pharmaceuticals Inc. by visiting

www.ferring.com

Pigmentation Treatment

Aclaro (4% hydroquinone) emulsion treats ultraviolet-induced dyschromia and discoloration due to oral contraceptive use, pregnancy, hormone therapy, or skin trauma. For more information, contact JSJ Pharmaceuticals by visiting

www.aclaro4.com

Soft, Inflatable Speculum

The FemSpec is a soft, rolled, tamponlike speculum that the physician can inflate for a better fit during gynecologic exams. This disposable product is patient friendly. The speculum comes in small, medium, and large sizes. Contact FemSpec LLC by visiting the Web site

www.femspec.com

Alendronate/Vitamin D in One Tablet

Fosamax Plus D (alendronate sodium/cholecalciferol) is approved for treatment of osteoporosis in postmenopausal women. The once-weekly tablet combines 70-mg alendronate sodium with 2,800-IU vitamin D3. For more information, contact Merck & Co. by calling 800-344-7833.

Generic Allegra-D Approved

A combination of fexofenadine HCl and pseudoephedrine HCl in an extended-release tablet (60 mg/120 mg) is indicated for the relief of symptoms associated with seasonal allergic rhinitis in adults and children 12 years and older. For more information, contact Barr Pharmaceuticals Inc. by calling 800-222-0190 or 201-930-3302.

Rapid Fecal Occult Blood Test

The Clearview ULTRA fecal occult blood test, a rapid two-step immunoassay, is now available. Results are available in less than 5 minutes and only one sample is required from the patient. The test does not require the patient to follow a restricted diet. For more information, contact Wampole Laboratories by calling 800-257-9525 or 609-627-8000.

Dictionary of Drug Names

The 2005 U.S. Pharmacopeia Dictionary of Adopted Names and International Drug Names is available. This 41st edition contains nationally and internationally recognized names for all drugs. The dictionary is available in book form for $299 or in a combination book-online package for $345. For more information, contact U.S. Pharmacopeia by visiting

http://store.usp.org

Fertility Treatment Launched

Menopur (menotropins for injection) is indicated for the development of multiple follicles and pregnancy in patients undergoing assisted reproductive technology procedures, such as in vitro fertilization. This purified human menopausal gonadotropin contains equal amounts of follicle-stimulating hormone and luteinizing hormone. For more information, contact Ferring Pharmaceuticals Inc. by visiting

www.ferring.com

Pigmentation Treatment

Aclaro (4% hydroquinone) emulsion treats ultraviolet-induced dyschromia and discoloration due to oral contraceptive use, pregnancy, hormone therapy, or skin trauma. For more information, contact JSJ Pharmaceuticals by visiting

www.aclaro4.com

Soft, Inflatable Speculum

The FemSpec is a soft, rolled, tamponlike speculum that the physician can inflate for a better fit during gynecologic exams. This disposable product is patient friendly. The speculum comes in small, medium, and large sizes. Contact FemSpec LLC by visiting the Web site

www.femspec.com

Alendronate/Vitamin D in One Tablet

Fosamax Plus D (alendronate sodium/cholecalciferol) is approved for treatment of osteoporosis in postmenopausal women. The once-weekly tablet combines 70-mg alendronate sodium with 2,800-IU vitamin D3. For more information, contact Merck & Co. by calling 800-344-7833.

Safe laparoscopic access begins before an incision is ever made.

It begins when you review the patient's history, which includes any pertinent previous surgeries. It extends to the examination, which should rule out pelvic or abdominal masses, hepatomegaly, or an enlarged spleen. If questions about the patient's anatomy arise, imaging may be helpful. I recently had an in vitro fertilization patient with ovarian torsion for whom ultrasound was needed to locate the limits of the ovaries, which reached the umbilicus.

During surgical preparation in the operating room, the patient's bladder should be drained with a Foley catheter to ensure that it is deflated; otherwise, it may extend into the operative field and be punctured on initial access. The anesthesiologist should drain the stomach contents with oral or nasogastric suction.

Optimal patient positioning is critical.

Keep the table at waist height. You want the movement of the trocar to be as controlled as possible, optimizing proprioception and fine muscle control of your hand so that the entry force arises from small, controlled muscle movements of the forearm, hand, and wrist—not from large, less-controlled movements from your shoulder girdle. By keeping your forearms perpendicular to the patient, the force is directed along the axis of the trocar. It is essential that you maintain control of that force upon entry, so that you neither push the trocar too far nor exert lateral force.

Ensure that the table is flat. Placing the patient into premature Trendelenburg's position will change the physical relationship of the major vessels. The patient's body should lie flat from left to right to allow better anticipation of the side wall and iliac vessels.

There are several schools of thought regarding trocar placement. I enter directly in the center of the umbilicus. Some surgeons prefer an infraumbilical approach; however, I have never seen a wound infection in the center of the umbilicus when it has been meticulously disinfected with swabs during preparation of the patient. Here in the center of the umbilicus, the abdominal wall is at its thinnest, and all layers of the abdominal wall are fused. Above or below the umbilicus, there are thicker, unfused layers of tissue that are more difficult to cut through, increasing the risk of false passage.

Before initial trocar placement, palpate the promontory of the sacrum and the bifurcation of the aorta, which may be above (in most cases), at the level of, or even below the umbilicus, and make a mental map of your entry angle.

When you prepare to enter, place the trocars perfectly centrally from right to left and perpendicular to the skin; do not angle them laterally. Maintain control over the angle as you push. Iliac vessel injuries can occur when the surgeon thinks the trocar is straight, but the angle of his or her hand shifts as more force is applied. It is difficult to push straight down with force. A right-handed surgeon will tend to angle to the left; a left-handed surgeon will tend to angle to the right. Recognize this tendency and avoid it.

Remember that the abdominal-peritoneal cavity is a potential space until you enter and air or gas is allowed to enter. During initial entry, keep the valve to the Veress or trocar open so that air can enter and allow organs to fall away from the abdominal wall. At this point, it is crucial to elevate the anterior abdominal wall as much as possible, either by hand or with towel clips. On my side of the patient, I grasp the anterior wall very firmly with my hand lateral to the umbilicus while, on the other side, my assistant elevates the wall with her hand lateral to the umbilicus. I find that this technique allows sharp retraction more effectively than grasping singly inferior to the umbilicus.

Other surgeons prefer to use two perforating towel clips on the umbilical ring. During your next case, when the laparoscope is already in, try different methods of abdominal wall elevation and see which grasping strategy lifts the abdominal wall most effectively in your hands.

Keep in mind that you may have to adjust your angle of entry depending on the patient's degree of obesity; otherwise, you may find yourself tunneling in subcutaneous fat. The more obese the patient, the more perpendicular the angle of your initial entry should be.

Although you do have to tailor your technique to the obese patient's anatomy, obesity generally is not a contraindication to laparoscopic surgery. Surgical and postsurgical risks are elevated in obese patients; however, open surgery poses significant risks to these patients as well.

With very thin patients, use less force, as the distance from the abdominal wall to retroperitoneal structures is closer, and injuries from hitting the posterior retroperitoneum are more likely.

There are various methods to ensure that the initial entry into the peritoneal cavity is correct and that no injury occurs. None of these methods is foolproof. I most often use a radially dilating trocar inserted over a Veress needle. It is inserted initially with the valve open, so that air can enter and open up the potential space. Then, I inject saline and assess whether it flows easily. Next, I aspirate, checking for blood, feces, or saline from the subcutaneous space. I add a drop of saline as I lift up on the abdominal wall to perform the hanging-drop test. None of these maneuvers has good sensitivity or specificity.

The single most useful test to confirm correct intraperitoneal placement is the observation of low entry pressure from the carbon-dioxide gas. The observation of an initially negative pressure with elevation of the abdominal wall is reassuring as well. Some surgeons prefer to increase distention pressure for subsequent trocar placement.

Another method I sometimes use involves direct entry with a microlaparoscope. Insert a Veress needle with a 2-mm or 3-mm trocar over it and directly look inside with a 2-mm or 3-mm laparoscope before insufflation. In this method, elevation of the abdominal wall by hand is crucial for visualization.

Patients with a history of surgery may have adhesions at or near the umbilicus. Although various alternative sites have been suggested, the one that provides the lowest-risk access in the majority of patients is the left upper quadrant at Palmer's point, the midclavicular line below the left lowermost inferior rib. In almost all patients, even those who've had many previous surgeries, this area will be free of adhesions.

This entry site is close to the stomach, so make sure the anesthesiologist has suctioned stomach contents before surgery. Splenomegaly could complicate entry at this site.

At this site the abdominal wall is much thicker, and you will lose the typical feel of an umbilical entry. Go in carefully and in a completely controlled manner with an adjusted angle of entry. I prefer to use a microlaparoscopic entry here, although some surgeons use a Veress needle and trocar.

The open-entry technique—also called the Hasson technique—was developed in the early 1970s to mimic the steps of entry during an open surgery but with a 10-mm incision. The surgeon makes an incision in the umbilicus and dissects down layer by layer until the peritoneum is entered. The fascia is often tagged with sutures for elevation, and a blunt-tipped trocar is inserted.

The benefit of this approach is that it eliminates the step of blindly inserting something into the abdominal cavity. The drawback is that cutting and dissecting can still cause injury to underlying or adherent vital structures and may result in larger injuries in these cases.

Bowel injuries from adhesions have been reported with all known techniques.

Obviously, it would be helpful to know which technique is safest based on a well-designed, randomized, controlled trial, as most experts believe injuries from all techniques are underreported. However, significant injuries are so rare that one study reported a power analysis indicating that it would require more than 200,000 surgeries to show a 50% reduction in injuries. Most reports suggest that visceral injuries (primarily bowel injuries) occur less than 1% of the time with either an open- or closed-entry technique. A study from the Netherlands found no statistical difference between the two techniques.

Vascular injuries are even rarer than that. Advocates of open techniques suggest fewer vascular injuries occur with these techniques, but not statistically significantly so.

Recently, new trocars have been introduced that are aimed at improving safety. These devices include optical and radially dilating trocars and trocarless systems that screw or dilate their way into the abdomen. As these have come into more widespread use, injuries have been reported with them as well. It remains to be seen whether injury rates will decline significantly with other new alternatives.

Complications of laparoscopic surgery can occur when bowel or vascular injuries go unrecognized at the time of surgery. Vascular injuries can cause hemorrhage and hemodynamic instability. Retroperitoneal hematomas, in particular, can be missed unless the surgeon inspects for the presence of a large, expanding mass.

Bowel injuries may be difficult to see if the injury is small. Urinary tract injuries can also be subtle. Detection of these complications requires diligent postoperative follow-up and instructions to the patient that she should report postoperative fever, increasing pain, abdominal distention, vomiting, or heavy incisional bleeding. Be suspicious if any patient is not recovering as quickly as expected. Although most of us would recognize complications occurring within 24 hours, signs may not appear for up to a week after the procedure. Some patients with bowel perforation had no elevations in white blood count or fever at their initial postoperative checkups. The most serious complications have occurred when injuries went unrecognized.

In laparoscopic surgery, the most damage may occur during entry. Fortunately, such damage is almost totally avoidable if you use the proper techniques, a well-thought-out point of entry, and the safest equipment available.

I know laparoscopic surgeons who use a blind, primary trocar entry technique using no insufflation. They can all recount the number of cases they've done without a major adverse event; one surgeon proudly told me he “put a little nick in the common iliac artery with an accessory port” and solved the problem with rapid action and a quick stitch.

My response is that these surgeons have either been very lucky or simply haven't done enough procedures for the inevitable disasters to catch up with them. I am absolutely opposed to blindly putting large spikes into the abdomen, and I think doing so is an approach that should be condemned.

As laparoscopic surgeons, we generally operate on young, healthy patients, sometimes for elective reasons. The potential for tragedy here is great, and it makes no sense to risk lives and great bodily injury when safe and predictable alternatives are available. I conceptualize safe access by breaking it down into three precise goals:

▸ Avoiding damage to the anterior abdominal wall.

▸ Avoiding intraabdominal structures.

▸ Avoiding retroperitoneal structures. These goals are accomplished by using the most appropriate instruments available, such as a dilating trocar and adjunctive 2-mm laparoscope; choosing the optimal entry site; and properly insufflating the abdomen.

Anterior wall tissue damage can be greatly minimized by using expanding trocars that stretch the tissue rather than cut it. Many companies make disposable dilating trocars.

The reusable radially expanding system from Tyco has a sleeve that is passed into the abdominal cavity over a Veress needle; it is dilated solely with lateral pressure or force. No downward force is applied to the patient's abdomen.

When the tissue is stretched, less damage is done, less pain is caused, recovery is quicker, and the long-term risk of hernia formation is lower.

Avoiding intraabdominal structures is a matter of entry location. I agree with Dr. Palter that the thinnest tissue underlies the umbilicus, and that's where I like to place my primary port. But unfortunately, major vessels—and often adhesions—also directly underlie this spot. It's the place where laparoscopic entry is most likely to kill a patient, and I enter it only under direct visualization.

Several studies have assessed the rate of abdominal adhesions in patients who have had previous surgery. In a 1997 study, none of 45 patients had adhesions after laparoscopy, 17 of 29 had adhesions after a midvertical incision, and 11 of 39 had adhesions after a low transverse incision (J. Am. Assoc. Gynecol. Laparosc. 1997;4:353–6).

French gynecologic surgeon Alain Audebert, M.D., described adhesions in 331 patients with prior surgery. Adhesions were present in half of patients with a prior midline incision, in 21% with a lower transverse incision, and 1.8% of patients with prior laparoscopic surgery had adhesions. Even among 440 patients with no prior surgery, the rate of adhesions was 0.6%.

In a recent study of 100 of my own patients, I found adhesion formation in 16 of 36 who had midline incisions, 20 of 45 who had transverse incisions, and 7 of 19 who had other incisions.

Although it's rare, I recently operated on a patient whose only previous surgery was laparoscopic and even she had an adhesion directly under the umbilicus.

On the other hand, it is extremely rare for surgical adhesions to form over the left upper quadrant of the abdomen, because the gallbladder and appendix lie on the other side. Only trauma surgery or splenectomies are likely to result in adhesions at this anatomical location, which was first described as an entry site by Raoul Palmer, M.D., in 1972. Furthermore, although central obesity can complicate initial access at the umbilicus, the lower margin of the rib cage in the midclavicular line almost always can be palpated, providing a landmark structure for entry at this site.

It is for all these reasons that I favor a left upper quadrant entry—not just in special cases, as Dr. Palter suggested, but as a matter of routine. I have performed approximately 500 laparoscopic procedures using this approach; Dr. Audebert has performed more than 2,000.

Patients receive a nasogastric tube so that their stomachs can be emptied before the procedure to reduce the chance of this underlying organ being damaged, although a needle injury to the stomach or liver is not a major complication.

Splenomegaly is a contraindication.

I palpate the edge of the ribs at the midclavicular line at the costal margin and carefully slide a #11 blade right along the lowermost rib to create an incision only large enough to accommodate a 2-mm scope. This incision will require no stitch—just a Steri-strip—to close.

I insert a Veress needle perpendicularly, feeding it through the abdominal wall with my finger and thumb. It's a very tactile move. You should feel the clicks.

Precisely because entry at this site feels different from entry at the umbilicus, I favor doing the entry this way every time; you learn the feel of doing it right.

I then inject saline through the needle and aspirate. If solution has entered the abdominal cavity, I don't get it back. If I had entered the bowel, the return would be greenish brown or blood-tinged, and further evaluation would be required.

I then insufflate the abdomen with enough carbon dioxide to increase the intraabdominal pressure to 25 mm Hg. The volume required will vary from patient to patient. A postpartum patient may require 12 L of carbon dioxide to reach that degree of pressure; a marathon runner with a tight abdomen may get there with 4 L.

This step accomplishes the third goal: optimizing conditions to avoid retroperitoneal structures, the major vessels.

When intraabdominal pressure is at only 12 mm Hg, the abdominal wall is very close to bowel, leaving no room for error. Increased pressure leaves a bigger space between the anterior abdominal wall and the intraabdominal and retroperitoneal structures, providing a comfortable margin of safety. This is especially critical in very thin patients, whose major vessels may lie very close to the anterior abdominal wall.

Through the same small, upper-left quadrant incision, I insert a 2-mm trocar and cannula for a small laparoscope to inspect the abdomen, identify the inferior epigastric vessel, assess the location of any adhesions, and ensure that no structures have been damaged or are in the path of any intended trocar site.

For some cases, such as tubal ligation, it is possible to use this as the primary port.

More commonly, it is at this point that I determine whether I can place my primary port in the umbilicus, and I do so under direct visualization before proceeding with my surgical procedure.

It is widely believed that laparoscopic surgery injuries are underreported. Nonetheless, patterns can be determined.

A prospective, multicenter study of complications of laparoscopy conducted in the Netherlands identified 29 bowel injuries and 27 major vessel injuries in 25,764 procedures for an identical rate of 0.11%.

More importantly, with regard to entry technique, Richard Soderstrom, M.D., found in a medicolegal review that the primary port was responsible for half of all major vascular injuries during 47 endoscopic cases (J. Am. Assoc. Gynecol. Laparosc. 1997;4:395–8).

A review of malpractice cases after 296 laparoscopic cholecystectomies performed using a primary umbilical entry site found that 86% of major vascular injuries were caused by the primary port and 75% of gastrointestinal injuries were caused by a trocar (J. Gastrointest. Surg. 1997;1:138–45).

Bowel injuries can occur with any laparoscopic technique and with laparotomy as well; some studies suggest a greater risk of injury with the latter approach. Because no method has been devised to completely avoid these injuries, the key is to recognize them, either at the time of surgery, or when the patient fails to improve as expected during the postsurgical course.

We have the potential to entirely avoid major vascular injuries by use of upper left quadrant entry before placement of a central port at the umbilicus and placement of accessory trocars.

By combining appropriate entry tools, a left upper quadrant entry site, and adequate insufflation of the abdomen, risks can be reduced at every step of laparoscopic surgery.

Dr. Duncan Turner, who is performing this procedure, prefers an entry site in the left upper quadrant to avoid major structures and potential adhesions from previous surgery.

Direct visualization and adequate insufflation leave plenty of room for error when the trocar enters the intraabdominal space as shown in this image. Photos courtesy Dr. Duncan Turner

'Do No Harm'

Physician, do no harm is a credo all gynecologic surgeons must live by. Nowhere do these words ring more true than during laparoscopic surgery. With the slightest lapse in technique, the simplest of laparoscopic procedures can quickly become a nightmare both for the patient as well as the surgeon. And when is the greatest risk of injury during a laparoscopic procedure? At entry.

I have invited an expert from each coast to serve as guides to safe access into the pelvic cavity via the laparoscope. Steven F. Palter, M.D., practices reproductive medicine and gynecologic surgery at Gold Coast IVF in Syosset, N.Y.

He is an immediate past member of the board of trustees of the AAGL. Over the years, Dr. Palter has won first prize awards for presentations at three separate world congress venues: at meetings of the American Society for Reproductive Medicine, the American Urological Association, and the AAGL, where he won first prize awards twice.

Dr Palter has been a longtime advocate of safe laparoscopic access, stemming from his background in office-based laparoscopy. From 1995 to 2001, Steve served as the director of the Yale Office Laparoscopy Program.

Duncan Turner, M.D., is a current member of the board of the International Society for Gynecologic Endoscopy. He is in private practice in Santa Barbara, Calif. Duncan first lectured on avoiding trocar-related injuries in 1995 at the World Congress of Gynecologic Laparoscopy in Israel.

Since then, not only has he lectured throughout the world, but he has been involved in trocar development and evaluation as well.

Safe laparoscopic access begins before an incision is ever made.

It begins when you review the patient's history, which includes any pertinent previous surgeries. It extends to the examination, which should rule out pelvic or abdominal masses, hepatomegaly, or an enlarged spleen. If questions about the patient's anatomy arise, imaging may be helpful. I recently had an in vitro fertilization patient with ovarian torsion for whom ultrasound was needed to locate the limits of the ovaries, which reached the umbilicus.

During surgical preparation in the operating room, the patient's bladder should be drained with a Foley catheter to ensure that it is deflated; otherwise, it may extend into the operative field and be punctured on initial access. The anesthesiologist should drain the stomach contents with oral or nasogastric suction.

Optimal patient positioning is critical.

Keep the table at waist height. You want the movement of the trocar to be as controlled as possible, optimizing proprioception and fine muscle control of your hand so that the entry force arises from small, controlled muscle movements of the forearm, hand, and wrist—not from large, less-controlled movements from your shoulder girdle. By keeping your forearms perpendicular to the patient, the force is directed along the axis of the trocar. It is essential that you maintain control of that force upon entry, so that you neither push the trocar too far nor exert lateral force.

Ensure that the table is flat. Placing the patient into premature Trendelenburg's position will change the physical relationship of the major vessels. The patient's body should lie flat from left to right to allow better anticipation of the side wall and iliac vessels.

There are several schools of thought regarding trocar placement. I enter directly in the center of the umbilicus. Some surgeons prefer an infraumbilical approach; however, I have never seen a wound infection in the center of the umbilicus when it has been meticulously disinfected with swabs during preparation of the patient. Here in the center of the umbilicus, the abdominal wall is at its thinnest, and all layers of the abdominal wall are fused. Above or below the umbilicus, there are thicker, unfused layers of tissue that are more difficult to cut through, increasing the risk of false passage.

Before initial trocar placement, palpate the promontory of the sacrum and the bifurcation of the aorta, which may be above (in most cases), at the level of, or even below the umbilicus, and make a mental map of your entry angle.

When you prepare to enter, place the trocars perfectly centrally from right to left and perpendicular to the skin; do not angle them laterally. Maintain control over the angle as you push. Iliac vessel injuries can occur when the surgeon thinks the trocar is straight, but the angle of his or her hand shifts as more force is applied. It is difficult to push straight down with force. A right-handed surgeon will tend to angle to the left; a left-handed surgeon will tend to angle to the right. Recognize this tendency and avoid it.

Remember that the abdominal-peritoneal cavity is a potential space until you enter and air or gas is allowed to enter. During initial entry, keep the valve to the Veress or trocar open so that air can enter and allow organs to fall away from the abdominal wall. At this point, it is crucial to elevate the anterior abdominal wall as much as possible, either by hand or with towel clips. On my side of the patient, I grasp the anterior wall very firmly with my hand lateral to the umbilicus while, on the other side, my assistant elevates the wall with her hand lateral to the umbilicus. I find that this technique allows sharp retraction more effectively than grasping singly inferior to the umbilicus.

Other surgeons prefer to use two perforating towel clips on the umbilical ring. During your next case, when the laparoscope is already in, try different methods of abdominal wall elevation and see which grasping strategy lifts the abdominal wall most effectively in your hands.

Keep in mind that you may have to adjust your angle of entry depending on the patient's degree of obesity; otherwise, you may find yourself tunneling in subcutaneous fat. The more obese the patient, the more perpendicular the angle of your initial entry should be.

Although you do have to tailor your technique to the obese patient's anatomy, obesity generally is not a contraindication to laparoscopic surgery. Surgical and postsurgical risks are elevated in obese patients; however, open surgery poses significant risks to these patients as well.

With very thin patients, use less force, as the distance from the abdominal wall to retroperitoneal structures is closer, and injuries from hitting the posterior retroperitoneum are more likely.

There are various methods to ensure that the initial entry into the peritoneal cavity is correct and that no injury occurs. None of these methods is foolproof. I most often use a radially dilating trocar inserted over a Veress needle. It is inserted initially with the valve open, so that air can enter and open up the potential space. Then, I inject saline and assess whether it flows easily. Next, I aspirate, checking for blood, feces, or saline from the subcutaneous space. I add a drop of saline as I lift up on the abdominal wall to perform the hanging-drop test. None of these maneuvers has good sensitivity or specificity.

The single most useful test to confirm correct intraperitoneal placement is the observation of low entry pressure from the carbon-dioxide gas. The observation of an initially negative pressure with elevation of the abdominal wall is reassuring as well. Some surgeons prefer to increase distention pressure for subsequent trocar placement.

Another method I sometimes use involves direct entry with a microlaparoscope. Insert a Veress needle with a 2-mm or 3-mm trocar over it and directly look inside with a 2-mm or 3-mm laparoscope before insufflation. In this method, elevation of the abdominal wall by hand is crucial for visualization.

Patients with a history of surgery may have adhesions at or near the umbilicus. Although various alternative sites have been suggested, the one that provides the lowest-risk access in the majority of patients is the left upper quadrant at Palmer's point, the midclavicular line below the left lowermost inferior rib. In almost all patients, even those who've had many previous surgeries, this area will be free of adhesions.

This entry site is close to the stomach, so make sure the anesthesiologist has suctioned stomach contents before surgery. Splenomegaly could complicate entry at this site.

At this site the abdominal wall is much thicker, and you will lose the typical feel of an umbilical entry. Go in carefully and in a completely controlled manner with an adjusted angle of entry. I prefer to use a microlaparoscopic entry here, although some surgeons use a Veress needle and trocar.

The open-entry technique—also called the Hasson technique—was developed in the early 1970s to mimic the steps of entry during an open surgery but with a 10-mm incision. The surgeon makes an incision in the umbilicus and dissects down layer by layer until the peritoneum is entered. The fascia is often tagged with sutures for elevation, and a blunt-tipped trocar is inserted.

The benefit of this approach is that it eliminates the step of blindly inserting something into the abdominal cavity. The drawback is that cutting and dissecting can still cause injury to underlying or adherent vital structures and may result in larger injuries in these cases.

Bowel injuries from adhesions have been reported with all known techniques.

Obviously, it would be helpful to know which technique is safest based on a well-designed, randomized, controlled trial, as most experts believe injuries from all techniques are underreported. However, significant injuries are so rare that one study reported a power analysis indicating that it would require more than 200,000 surgeries to show a 50% reduction in injuries. Most reports suggest that visceral injuries (primarily bowel injuries) occur less than 1% of the time with either an open- or closed-entry technique. A study from the Netherlands found no statistical difference between the two techniques.

Vascular injuries are even rarer than that. Advocates of open techniques suggest fewer vascular injuries occur with these techniques, but not statistically significantly so.

Recently, new trocars have been introduced that are aimed at improving safety. These devices include optical and radially dilating trocars and trocarless systems that screw or dilate their way into the abdomen. As these have come into more widespread use, injuries have been reported with them as well. It remains to be seen whether injury rates will decline significantly with other new alternatives.

Complications of laparoscopic surgery can occur when bowel or vascular injuries go unrecognized at the time of surgery. Vascular injuries can cause hemorrhage and hemodynamic instability. Retroperitoneal hematomas, in particular, can be missed unless the surgeon inspects for the presence of a large, expanding mass.

Bowel injuries may be difficult to see if the injury is small. Urinary tract injuries can also be subtle. Detection of these complications requires diligent postoperative follow-up and instructions to the patient that she should report postoperative fever, increasing pain, abdominal distention, vomiting, or heavy incisional bleeding. Be suspicious if any patient is not recovering as quickly as expected. Although most of us would recognize complications occurring within 24 hours, signs may not appear for up to a week after the procedure. Some patients with bowel perforation had no elevations in white blood count or fever at their initial postoperative checkups. The most serious complications have occurred when injuries went unrecognized.

In laparoscopic surgery, the most damage may occur during entry. Fortunately, such damage is almost totally avoidable if you use the proper techniques, a well-thought-out point of entry, and the safest equipment available.

I know laparoscopic surgeons who use a blind, primary trocar entry technique using no insufflation. They can all recount the number of cases they've done without a major adverse event; one surgeon proudly told me he “put a little nick in the common iliac artery with an accessory port” and solved the problem with rapid action and a quick stitch.

My response is that these surgeons have either been very lucky or simply haven't done enough procedures for the inevitable disasters to catch up with them. I am absolutely opposed to blindly putting large spikes into the abdomen, and I think doing so is an approach that should be condemned.

As laparoscopic surgeons, we generally operate on young, healthy patients, sometimes for elective reasons. The potential for tragedy here is great, and it makes no sense to risk lives and great bodily injury when safe and predictable alternatives are available. I conceptualize safe access by breaking it down into three precise goals:

▸ Avoiding damage to the anterior abdominal wall.

▸ Avoiding intraabdominal structures.

▸ Avoiding retroperitoneal structures. These goals are accomplished by using the most appropriate instruments available, such as a dilating trocar and adjunctive 2-mm laparoscope; choosing the optimal entry site; and properly insufflating the abdomen.

Anterior wall tissue damage can be greatly minimized by using expanding trocars that stretch the tissue rather than cut it. Many companies make disposable dilating trocars.

The reusable radially expanding system from Tyco has a sleeve that is passed into the abdominal cavity over a Veress needle; it is dilated solely with lateral pressure or force. No downward force is applied to the patient's abdomen.

When the tissue is stretched, less damage is done, less pain is caused, recovery is quicker, and the long-term risk of hernia formation is lower.

Avoiding intraabdominal structures is a matter of entry location. I agree with Dr. Palter that the thinnest tissue underlies the umbilicus, and that's where I like to place my primary port. But unfortunately, major vessels—and often adhesions—also directly underlie this spot. It's the place where laparoscopic entry is most likely to kill a patient, and I enter it only under direct visualization.

Several studies have assessed the rate of abdominal adhesions in patients who have had previous surgery. In a 1997 study, none of 45 patients had adhesions after laparoscopy, 17 of 29 had adhesions after a midvertical incision, and 11 of 39 had adhesions after a low transverse incision (J. Am. Assoc. Gynecol. Laparosc. 1997;4:353–6).

French gynecologic surgeon Alain Audebert, M.D., described adhesions in 331 patients with prior surgery. Adhesions were present in half of patients with a prior midline incision, in 21% with a lower transverse incision, and 1.8% of patients with prior laparoscopic surgery had adhesions. Even among 440 patients with no prior surgery, the rate of adhesions was 0.6%.

In a recent study of 100 of my own patients, I found adhesion formation in 16 of 36 who had midline incisions, 20 of 45 who had transverse incisions, and 7 of 19 who had other incisions.

Although it's rare, I recently operated on a patient whose only previous surgery was laparoscopic and even she had an adhesion directly under the umbilicus.

On the other hand, it is extremely rare for surgical adhesions to form over the left upper quadrant of the abdomen, because the gallbladder and appendix lie on the other side. Only trauma surgery or splenectomies are likely to result in adhesions at this anatomical location, which was first described as an entry site by Raoul Palmer, M.D., in 1972. Furthermore, although central obesity can complicate initial access at the umbilicus, the lower margin of the rib cage in the midclavicular line almost always can be palpated, providing a landmark structure for entry at this site.

It is for all these reasons that I favor a left upper quadrant entry—not just in special cases, as Dr. Palter suggested, but as a matter of routine. I have performed approximately 500 laparoscopic procedures using this approach; Dr. Audebert has performed more than 2,000.

Patients receive a nasogastric tube so that their stomachs can be emptied before the procedure to reduce the chance of this underlying organ being damaged, although a needle injury to the stomach or liver is not a major complication.

Splenomegaly is a contraindication.

I palpate the edge of the ribs at the midclavicular line at the costal margin and carefully slide a #11 blade right along the lowermost rib to create an incision only large enough to accommodate a 2-mm scope. This incision will require no stitch—just a Steri-strip—to close.

I insert a Veress needle perpendicularly, feeding it through the abdominal wall with my finger and thumb. It's a very tactile move. You should feel the clicks.

Precisely because entry at this site feels different from entry at the umbilicus, I favor doing the entry this way every time; you learn the feel of doing it right.

I then inject saline through the needle and aspirate. If solution has entered the abdominal cavity, I don't get it back. If I had entered the bowel, the return would be greenish brown or blood-tinged, and further evaluation would be required.

I then insufflate the abdomen with enough carbon dioxide to increase the intraabdominal pressure to 25 mm Hg. The volume required will vary from patient to patient. A postpartum patient may require 12 L of carbon dioxide to reach that degree of pressure; a marathon runner with a tight abdomen may get there with 4 L.

This step accomplishes the third goal: optimizing conditions to avoid retroperitoneal structures, the major vessels.

When intraabdominal pressure is at only 12 mm Hg, the abdominal wall is very close to bowel, leaving no room for error. Increased pressure leaves a bigger space between the anterior abdominal wall and the intraabdominal and retroperitoneal structures, providing a comfortable margin of safety. This is especially critical in very thin patients, whose major vessels may lie very close to the anterior abdominal wall.

Through the same small, upper-left quadrant incision, I insert a 2-mm trocar and cannula for a small laparoscope to inspect the abdomen, identify the inferior epigastric vessel, assess the location of any adhesions, and ensure that no structures have been damaged or are in the path of any intended trocar site.

For some cases, such as tubal ligation, it is possible to use this as the primary port.

More commonly, it is at this point that I determine whether I can place my primary port in the umbilicus, and I do so under direct visualization before proceeding with my surgical procedure.

It is widely believed that laparoscopic surgery injuries are underreported. Nonetheless, patterns can be determined.

A prospective, multicenter study of complications of laparoscopy conducted in the Netherlands identified 29 bowel injuries and 27 major vessel injuries in 25,764 procedures for an identical rate of 0.11%.

More importantly, with regard to entry technique, Richard Soderstrom, M.D., found in a medicolegal review that the primary port was responsible for half of all major vascular injuries during 47 endoscopic cases (J. Am. Assoc. Gynecol. Laparosc. 1997;4:395–8).

A review of malpractice cases after 296 laparoscopic cholecystectomies performed using a primary umbilical entry site found that 86% of major vascular injuries were caused by the primary port and 75% of gastrointestinal injuries were caused by a trocar (J. Gastrointest. Surg. 1997;1:138–45).

Bowel injuries can occur with any laparoscopic technique and with laparotomy as well; some studies suggest a greater risk of injury with the latter approach. Because no method has been devised to completely avoid these injuries, the key is to recognize them, either at the time of surgery, or when the patient fails to improve as expected during the postsurgical course.

We have the potential to entirely avoid major vascular injuries by use of upper left quadrant entry before placement of a central port at the umbilicus and placement of accessory trocars.

By combining appropriate entry tools, a left upper quadrant entry site, and adequate insufflation of the abdomen, risks can be reduced at every step of laparoscopic surgery.

Dr. Duncan Turner, who is performing this procedure, prefers an entry site in the left upper quadrant to avoid major structures and potential adhesions from previous surgery.

Direct visualization and adequate insufflation leave plenty of room for error when the trocar enters the intraabdominal space as shown in this image. Photos courtesy Dr. Duncan Turner

'Do No Harm'

Physician, do no harm is a credo all gynecologic surgeons must live by. Nowhere do these words ring more true than during laparoscopic surgery. With the slightest lapse in technique, the simplest of laparoscopic procedures can quickly become a nightmare both for the patient as well as the surgeon. And when is the greatest risk of injury during a laparoscopic procedure? At entry.

I have invited an expert from each coast to serve as guides to safe access into the pelvic cavity via the laparoscope. Steven F. Palter, M.D., practices reproductive medicine and gynecologic surgery at Gold Coast IVF in Syosset, N.Y.

He is an immediate past member of the board of trustees of the AAGL. Over the years, Dr. Palter has won first prize awards for presentations at three separate world congress venues: at meetings of the American Society for Reproductive Medicine, the American Urological Association, and the AAGL, where he won first prize awards twice.

Dr Palter has been a longtime advocate of safe laparoscopic access, stemming from his background in office-based laparoscopy. From 1995 to 2001, Steve served as the director of the Yale Office Laparoscopy Program.

Duncan Turner, M.D., is a current member of the board of the International Society for Gynecologic Endoscopy. He is in private practice in Santa Barbara, Calif. Duncan first lectured on avoiding trocar-related injuries in 1995 at the World Congress of Gynecologic Laparoscopy in Israel.

Since then, not only has he lectured throughout the world, but he has been involved in trocar development and evaluation as well.

Safe laparoscopic access begins before an incision is ever made.

It begins when you review the patient's history, which includes any pertinent previous surgeries. It extends to the examination, which should rule out pelvic or abdominal masses, hepatomegaly, or an enlarged spleen. If questions about the patient's anatomy arise, imaging may be helpful. I recently had an in vitro fertilization patient with ovarian torsion for whom ultrasound was needed to locate the limits of the ovaries, which reached the umbilicus.

During surgical preparation in the operating room, the patient's bladder should be drained with a Foley catheter to ensure that it is deflated; otherwise, it may extend into the operative field and be punctured on initial access. The anesthesiologist should drain the stomach contents with oral or nasogastric suction.

Optimal patient positioning is critical.

Keep the table at waist height. You want the movement of the trocar to be as controlled as possible, optimizing proprioception and fine muscle control of your hand so that the entry force arises from small, controlled muscle movements of the forearm, hand, and wrist—not from large, less-controlled movements from your shoulder girdle. By keeping your forearms perpendicular to the patient, the force is directed along the axis of the trocar. It is essential that you maintain control of that force upon entry, so that you neither push the trocar too far nor exert lateral force.

Ensure that the table is flat. Placing the patient into premature Trendelenburg's position will change the physical relationship of the major vessels. The patient's body should lie flat from left to right to allow better anticipation of the side wall and iliac vessels.

There are several schools of thought regarding trocar placement. I enter directly in the center of the umbilicus. Some surgeons prefer an infraumbilical approach; however, I have never seen a wound infection in the center of the umbilicus when it has been meticulously disinfected with swabs during preparation of the patient. Here in the center of the umbilicus, the abdominal wall is at its thinnest, and all layers of the abdominal wall are fused. Above or below the umbilicus, there are thicker, unfused layers of tissue that are more difficult to cut through, increasing the risk of false passage.

Before initial trocar placement, palpate the promontory of the sacrum and the bifurcation of the aorta, which may be above (in most cases), at the level of, or even below the umbilicus, and make a mental map of your entry angle.

When you prepare to enter, place the trocars perfectly centrally from right to left and perpendicular to the skin; do not angle them laterally. Maintain control over the angle as you push. Iliac vessel injuries can occur when the surgeon thinks the trocar is straight, but the angle of his or her hand shifts as more force is applied. It is difficult to push straight down with force. A right-handed surgeon will tend to angle to the left; a left-handed surgeon will tend to angle to the right. Recognize this tendency and avoid it.

Remember that the abdominal-peritoneal cavity is a potential space until you enter and air or gas is allowed to enter. During initial entry, keep the valve to the Veress or trocar open so that air can enter and allow organs to fall away from the abdominal wall. At this point, it is crucial to elevate the anterior abdominal wall as much as possible, either by hand or with towel clips. On my side of the patient, I grasp the anterior wall very firmly with my hand lateral to the umbilicus while, on the other side, my assistant elevates the wall with her hand lateral to the umbilicus. I find that this technique allows sharp retraction more effectively than grasping singly inferior to the umbilicus.

Other surgeons prefer to use two perforating towel clips on the umbilical ring. During your next case, when the laparoscope is already in, try different methods of abdominal wall elevation and see which grasping strategy lifts the abdominal wall most effectively in your hands.

Keep in mind that you may have to adjust your angle of entry depending on the patient's degree of obesity; otherwise, you may find yourself tunneling in subcutaneous fat. The more obese the patient, the more perpendicular the angle of your initial entry should be.

Although you do have to tailor your technique to the obese patient's anatomy, obesity generally is not a contraindication to laparoscopic surgery. Surgical and postsurgical risks are elevated in obese patients; however, open surgery poses significant risks to these patients as well.

With very thin patients, use less force, as the distance from the abdominal wall to retroperitoneal structures is closer, and injuries from hitting the posterior retroperitoneum are more likely.

There are various methods to ensure that the initial entry into the peritoneal cavity is correct and that no injury occurs. None of these methods is foolproof. I most often use a radially dilating trocar inserted over a Veress needle. It is inserted initially with the valve open, so that air can enter and open up the potential space. Then, I inject saline and assess whether it flows easily. Next, I aspirate, checking for blood, feces, or saline from the subcutaneous space. I add a drop of saline as I lift up on the abdominal wall to perform the hanging-drop test. None of these maneuvers has good sensitivity or specificity.

The single most useful test to confirm correct intraperitoneal placement is the observation of low entry pressure from the carbon-dioxide gas. The observation of an initially negative pressure with elevation of the abdominal wall is reassuring as well. Some surgeons prefer to increase distention pressure for subsequent trocar placement.

Another method I sometimes use involves direct entry with a microlaparoscope. Insert a Veress needle with a 2-mm or 3-mm trocar over it and directly look inside with a 2-mm or 3-mm laparoscope before insufflation. In this method, elevation of the abdominal wall by hand is crucial for visualization.

Patients with a history of surgery may have adhesions at or near the umbilicus. Although various alternative sites have been suggested, the one that provides the lowest-risk access in the majority of patients is the left upper quadrant at Palmer's point, the midclavicular line below the left lowermost inferior rib. In almost all patients, even those who've had many previous surgeries, this area will be free of adhesions.

This entry site is close to the stomach, so make sure the anesthesiologist has suctioned stomach contents before surgery. Splenomegaly could complicate entry at this site.

At this site the abdominal wall is much thicker, and you will lose the typical feel of an umbilical entry. Go in carefully and in a completely controlled manner with an adjusted angle of entry. I prefer to use a microlaparoscopic entry here, although some surgeons use a Veress needle and trocar.

The open-entry technique—also called the Hasson technique—was developed in the early 1970s to mimic the steps of entry during an open surgery but with a 10-mm incision. The surgeon makes an incision in the umbilicus and dissects down layer by layer until the peritoneum is entered. The fascia is often tagged with sutures for elevation, and a blunt-tipped trocar is inserted.

The benefit of this approach is that it eliminates the step of blindly inserting something into the abdominal cavity. The drawback is that cutting and dissecting can still cause injury to underlying or adherent vital structures and may result in larger injuries in these cases.

Bowel injuries from adhesions have been reported with all known techniques.

Obviously, it would be helpful to know which technique is safest based on a well-designed, randomized, controlled trial, as most experts believe injuries from all techniques are underreported. However, significant injuries are so rare that one study reported a power analysis indicating that it would require more than 200,000 surgeries to show a 50% reduction in injuries. Most reports suggest that visceral injuries (primarily bowel injuries) occur less than 1% of the time with either an open- or closed-entry technique. A study from the Netherlands found no statistical difference between the two techniques.

Vascular injuries are even rarer than that. Advocates of open techniques suggest fewer vascular injuries occur with these techniques, but not statistically significantly so.

Recently, new trocars have been introduced that are aimed at improving safety. These devices include optical and radially dilating trocars and trocarless systems that screw or dilate their way into the abdomen. As these have come into more widespread use, injuries have been reported with them as well. It remains to be seen whether injury rates will decline significantly with other new alternatives.

Complications of laparoscopic surgery can occur when bowel or vascular injuries go unrecognized at the time of surgery. Vascular injuries can cause hemorrhage and hemodynamic instability. Retroperitoneal hematomas, in particular, can be missed unless the surgeon inspects for the presence of a large, expanding mass.

Bowel injuries may be difficult to see if the injury is small. Urinary tract injuries can also be subtle. Detection of these complications requires diligent postoperative follow-up and instructions to the patient that she should report postoperative fever, increasing pain, abdominal distention, vomiting, or heavy incisional bleeding. Be suspicious if any patient is not recovering as quickly as expected. Although most of us would recognize complications occurring within 24 hours, signs may not appear for up to a week after the procedure. Some patients with bowel perforation had no elevations in white blood count or fever at their initial postoperative checkups. The most serious complications have occurred when injuries went unrecognized.

In laparoscopic surgery, the most damage may occur during entry. Fortunately, such damage is almost totally avoidable if you use the proper techniques, a well-thought-out point of entry, and the safest equipment available.

I know laparoscopic surgeons who use a blind, primary trocar entry technique using no insufflation. They can all recount the number of cases they've done without a major adverse event; one surgeon proudly told me he “put a little nick in the common iliac artery with an accessory port” and solved the problem with rapid action and a quick stitch.

My response is that these surgeons have either been very lucky or simply haven't done enough procedures for the inevitable disasters to catch up with them. I am absolutely opposed to blindly putting large spikes into the abdomen, and I think doing so is an approach that should be condemned.

As laparoscopic surgeons, we generally operate on young, healthy patients, sometimes for elective reasons. The potential for tragedy here is great, and it makes no sense to risk lives and great bodily injury when safe and predictable alternatives are available. I conceptualize safe access by breaking it down into three precise goals:

▸ Avoiding damage to the anterior abdominal wall.

▸ Avoiding intraabdominal structures.

▸ Avoiding retroperitoneal structures. These goals are accomplished by using the most appropriate instruments available, such as a dilating trocar and adjunctive 2-mm laparoscope; choosing the optimal entry site; and properly insufflating the abdomen.

Anterior wall tissue damage can be greatly minimized by using expanding trocars that stretch the tissue rather than cut it. Many companies make disposable dilating trocars.

The reusable radially expanding system from Tyco has a sleeve that is passed into the abdominal cavity over a Veress needle; it is dilated solely with lateral pressure or force. No downward force is applied to the patient's abdomen.

When the tissue is stretched, less damage is done, less pain is caused, recovery is quicker, and the long-term risk of hernia formation is lower.

Avoiding intraabdominal structures is a matter of entry location. I agree with Dr. Palter that the thinnest tissue underlies the umbilicus, and that's where I like to place my primary port. But unfortunately, major vessels—and often adhesions—also directly underlie this spot. It's the place where laparoscopic entry is most likely to kill a patient, and I enter it only under direct visualization.

Several studies have assessed the rate of abdominal adhesions in patients who have had previous surgery. In a 1997 study, none of 45 patients had adhesions after laparoscopy, 17 of 29 had adhesions after a midvertical incision, and 11 of 39 had adhesions after a low transverse incision (J. Am. Assoc. Gynecol. Laparosc. 1997;4:353–6).

French gynecologic surgeon Alain Audebert, M.D., described adhesions in 331 patients with prior surgery. Adhesions were present in half of patients with a prior midline incision, in 21% with a lower transverse incision, and 1.8% of patients with prior laparoscopic surgery had adhesions. Even among 440 patients with no prior surgery, the rate of adhesions was 0.6%.

In a recent study of 100 of my own patients, I found adhesion formation in 16 of 36 who had midline incisions, 20 of 45 who had transverse incisions, and 7 of 19 who had other incisions.

Although it's rare, I recently operated on a patient whose only previous surgery was laparoscopic and even she had an adhesion directly under the umbilicus.

On the other hand, it is extremely rare for surgical adhesions to form over the left upper quadrant of the abdomen, because the gallbladder and appendix lie on the other side. Only trauma surgery or splenectomies are likely to result in adhesions at this anatomical location, which was first described as an entry site by Raoul Palmer, M.D., in 1972. Furthermore, although central obesity can complicate initial access at the umbilicus, the lower margin of the rib cage in the midclavicular line almost always can be palpated, providing a landmark structure for entry at this site.

It is for all these reasons that I favor a left upper quadrant entry—not just in special cases, as Dr. Palter suggested, but as a matter of routine. I have performed approximately 500 laparoscopic procedures using this approach; Dr. Audebert has performed more than 2,000.

Patients receive a nasogastric tube so that their stomachs can be emptied before the procedure to reduce the chance of this underlying organ being damaged, although a needle injury to the stomach or liver is not a major complication.

Splenomegaly is a contraindication.

I palpate the edge of the ribs at the midclavicular line at the costal margin and carefully slide a #11 blade right along the lowermost rib to create an incision only large enough to accommodate a 2-mm scope. This incision will require no stitch—just a Steri-strip—to close.

I insert a Veress needle perpendicularly, feeding it through the abdominal wall with my finger and thumb. It's a very tactile move. You should feel the clicks.

Precisely because entry at this site feels different from entry at the umbilicus, I favor doing the entry this way every time; you learn the feel of doing it right.

I then inject saline through the needle and aspirate. If solution has entered the abdominal cavity, I don't get it back. If I had entered the bowel, the return would be greenish brown or blood-tinged, and further evaluation would be required.

I then insufflate the abdomen with enough carbon dioxide to increase the intraabdominal pressure to 25 mm Hg. The volume required will vary from patient to patient. A postpartum patient may require 12 L of carbon dioxide to reach that degree of pressure; a marathon runner with a tight abdomen may get there with 4 L.

This step accomplishes the third goal: optimizing conditions to avoid retroperitoneal structures, the major vessels.

When intraabdominal pressure is at only 12 mm Hg, the abdominal wall is very close to bowel, leaving no room for error. Increased pressure leaves a bigger space between the anterior abdominal wall and the intraabdominal and retroperitoneal structures, providing a comfortable margin of safety. This is especially critical in very thin patients, whose major vessels may lie very close to the anterior abdominal wall.

Through the same small, upper-left quadrant incision, I insert a 2-mm trocar and cannula for a small laparoscope to inspect the abdomen, identify the inferior epigastric vessel, assess the location of any adhesions, and ensure that no structures have been damaged or are in the path of any intended trocar site.

For some cases, such as tubal ligation, it is possible to use this as the primary port.

More commonly, it is at this point that I determine whether I can place my primary port in the umbilicus, and I do so under direct visualization before proceeding with my surgical procedure.

It is widely believed that laparoscopic surgery injuries are underreported. Nonetheless, patterns can be determined.

A prospective, multicenter study of complications of laparoscopy conducted in the Netherlands identified 29 bowel injuries and 27 major vessel injuries in 25,764 procedures for an identical rate of 0.11%.

More importantly, with regard to entry technique, Richard Soderstrom, M.D., found in a medicolegal review that the primary port was responsible for half of all major vascular injuries during 47 endoscopic cases (J. Am. Assoc. Gynecol. Laparosc. 1997;4:395–8).

A review of malpractice cases after 296 laparoscopic cholecystectomies performed using a primary umbilical entry site found that 86% of major vascular injuries were caused by the primary port and 75% of gastrointestinal injuries were caused by a trocar (J. Gastrointest. Surg. 1997;1:138–45).

Bowel injuries can occur with any laparoscopic technique and with laparotomy as well; some studies suggest a greater risk of injury with the latter approach. Because no method has been devised to completely avoid these injuries, the key is to recognize them, either at the time of surgery, or when the patient fails to improve as expected during the postsurgical course.

We have the potential to entirely avoid major vascular injuries by use of upper left quadrant entry before placement of a central port at the umbilicus and placement of accessory trocars.

By combining appropriate entry tools, a left upper quadrant entry site, and adequate insufflation of the abdomen, risks can be reduced at every step of laparoscopic surgery.

Dr. Duncan Turner, who is performing this procedure, prefers an entry site in the left upper quadrant to avoid major structures and potential adhesions from previous surgery.

Direct visualization and adequate insufflation leave plenty of room for error when the trocar enters the intraabdominal space as shown in this image. Photos courtesy Dr. Duncan Turner

'Do No Harm'

Physician, do no harm is a credo all gynecologic surgeons must live by. Nowhere do these words ring more true than during laparoscopic surgery. With the slightest lapse in technique, the simplest of laparoscopic procedures can quickly become a nightmare both for the patient as well as the surgeon. And when is the greatest risk of injury during a laparoscopic procedure? At entry.

I have invited an expert from each coast to serve as guides to safe access into the pelvic cavity via the laparoscope. Steven F. Palter, M.D., practices reproductive medicine and gynecologic surgery at Gold Coast IVF in Syosset, N.Y.

He is an immediate past member of the board of trustees of the AAGL. Over the years, Dr. Palter has won first prize awards for presentations at three separate world congress venues: at meetings of the American Society for Reproductive Medicine, the American Urological Association, and the AAGL, where he won first prize awards twice.

Dr Palter has been a longtime advocate of safe laparoscopic access, stemming from his background in office-based laparoscopy. From 1995 to 2001, Steve served as the director of the Yale Office Laparoscopy Program.

Duncan Turner, M.D., is a current member of the board of the International Society for Gynecologic Endoscopy. He is in private practice in Santa Barbara, Calif. Duncan first lectured on avoiding trocar-related injuries in 1995 at the World Congress of Gynecologic Laparoscopy in Israel.

Since then, not only has he lectured throughout the world, but he has been involved in trocar development and evaluation as well.

However, we are far from being “the best” in the world.

Forty nations surpass the United States in infant mortality, including Singapore (2.29 per 1,000), Sweden (2.77 per 1,000), and Japan (3.26 per 1,000).

Because the U.S. and Canada are neighbors and share a border, similar economies, and comparable levels of technologic sophistication, it is of interest that Canada's infant mortality is 30% lower than that of the United States, which was estimated by the CIA to be 6.5 per 1,000 in 2004.

In 2002, infant mortality worsened slightly in both countries, prompting renewed scrutiny of an ever-important issue.

The Centers for Disease Control and Prevention reported that in 2002, U.S. infant mortality edged upward to 7.0 per 1,000 live births from 6.8 per 1,000 in 2001.

This represents the first rise in 44 years. However, even if this anomaly were a one-time occurrence, it's concerning to note this 0.02% increase when we view the statistic in the context of more than 4 million births.

The same trend occurred in Canada, where infant mortality rose from 5.2 per 1,000 in 2001 to 5.4 per 1,000 in 2002, after progressively falling since the 1960s.

There are numerous factors that may help to elucidate this trend. At the very least, this upturn in infant mortality indicates that perhaps we are not progressing at a pace that many believe is one of the most important measures of a nation's health.

Infant mortality can be divided into two categories: neonatal deaths occurring within the first month of life, and postneonatal deaths occurring later in the first year.

Postneonatal deaths have not increased. In fact, tremendous advances in the etiology and prevention of sudden infant death syndrome have substantially reduced postneonatal deaths over the past decade.

The neonatal increases noted in 2002—and the disparity between the United States' and Canada's infant mortality—have occurred in the early weeks of life, when the most common causes of death include congenital anomalies, problems of transition, and complications of preterm birth. Among these factors, preterm birth stands out as a significant contributor to rising infant mortality.

In the United States, preterm births increased to 12.1%, from 11.9% the previous year.

Although the preterm birth rate also rose slightly in Canada, it was 7.6% in 2002—nearly 40% lower than in the United States.

Why is the preterm birth rate trending upward?

The evidence suggests that the trend is being driven by the use of reproductive technology, leading to multiple births; by women giving birth at later ages; by the necessity of earlier obstetrician intervention when the fetus is in jeopardy; and by complications attributed to a lack of early, consistent prenatal care.

The attempt to save an infant via early delivery has allowed many preemies to live who might have been stillborn in years past.

The recent changes in social trends have influenced the ages at which women decide to have their children. In Ottawa, where I practice, over 60% of mothers in 2003 gave birth when they were older than 30 years, and 23.2% when they were older than 35 years. Although women have a right to be informed about their chances of conceiving and delivering healthy singletons at different ages, physicians have no desire to dictate social policy or individual choice. I have a 5-year-old, and I'm not a young man.

We understand that older women have a higher risk of having a preterm baby, in part because they have a higher risk of having multiples, having pregnancy complications, and having babies with congenital anomalies, three factors that contribute to infant mortality.

Older mothers also are more likely to require assisted reproductive technology (ART).

Although ART procedures are similar in the United States and Canada, and are basically patient-funded in both countries, reproductive technology is increasingly subject to oversight in Canada. A bill that recently passed both the House of Commons and the Senate would strictly regulate clinics and procedures, for example.

A great many ART centers in Canada are university-affiliated, not-for-profit programs, rather than independent clinics. As a result, a controversial issue—such as the implantation of multiple embryos—is debated within the wide academic community of endocrinologists, ob.gyns., neonatologists, pediatricians, and ethicists.

When three sets of quadruplets were born in 1 year at the University of Ottawa, the university-affiliated fertility center demonstrated its responsibility by revising its policies to limit the number of embryos transferred during each cycle. Today, we rarely see quadruplets, although triplets are still not a rarity.

Throughout Canada, rates of multiple birth are lower than in the United States, contributing to lower rates of preterm birth. However, in looking at overall preterm birth statistics, it is worth noting that both nations have unequal rates across populations.

The U.S. National Center for Health Statistics reports that African American infants are nearly twice as likely as non-Hispanic white infants to be born prematurely.

In Canada, the disparity is most clearly defined by income, with those in the lowest income quintile having an infant mortality rate two-thirds higher than that of the highest income quintile. As infant mortality secondary to congenital anomalies and other causes has fallen significantly, the differential is largely a result of a higher rate of preterm birth in lower-income families.

Canada's First Nation and Inuit people face serious health problems, including infant mortality in many communities that is twice the national rate, as do America's Native American populations. Although Canada is an increasingly racially diverse country, other racial disparities are less obvious in measures of health care, such as prenatal care or preterm birth. Income is perhaps a more fitting measure of comparison, and deserving of a wider perspective.

One interesting study examined the gross national products and income distribution in 20 poor and 15 rich nations, determining, as one would expect, that overall infant mortality was inversely proportionate to income.

In rich countries, however, the main contributor to higher infant mortality was not income, but income disparity (Lancet 1999;354:2047).

The United Nations, in its annual Innocenti report card for 2000, explored child poverty in rich nations by using consistent indices to identify the percentage of children living in families with incomes below 50% of each nation's median income.

Child poverty levels ranged from 2.6% in Sweden to 26.6% in Mexico. Canada was 15.5%, and the United States was 22.4%.

Some governments, including those in Scandinavia, take a very active role in making sure that people don't live in poverty. Income disparity is low.

Conversely, income inequality is very high in the United States and may contribute to exceedingly elevated preterm birth rates and infant mortality among African Americans.

A related issue, of course, is access to medical care, which varies greatly among the industrialized nations of the world.

In Canada, where we have universal medical care, prenatal care is available at no cost, with no disincentives to seeking care in the system. Even pregnant women who do not have a primary care physician can walk into a clinic in any city and be seen that day.

As a result, in 2000, well over 95% of Canadian women received prenatal care beginning in the first trimester, compared with 83.7% of American women.

Once again, disparity is evident in the U.S. numbers, with only about 75% of African Americans and fewer than 70% of Native Americans receiving early prenatal care. Indeed, 3.6% of women delivered with no prenatal care, or with care initiated only in the last trimester.

Although the U.S. rate of early prenatal care has improved quite dramatically in the last 15 years, rising 10% since 1990, it still falls short of the care rates in most Western nations. Conversely, I should point out that the Swedes think that even Canada's prenatal care numbers are lousy. In Sweden, virtually 100% of women receive prenatal care throughout their pregnancy.

I would also be remiss if I left the impression that the Canadian health care system, the Swedish system, or any system, for that matter, is perfect. In Canada, one faces a long wait for a hip replacement. Far too many Canadians (about 10%) depend on walk-in clinics because they can't find a primary care physician of their own. Canadians who make more money tend to live longer, and certain groups, such as First Nation and Inuit people, have unequal health outcomes despite access to free care.

In Canada, we learn a lot from the United States, from the abundance of medical research and education to the excellent health care available to many. But in the spirit of learning from each other, U.S. physicians may be interested in studying a neighbor that spends less on health care yet produces not only lower preterm birth rates and infant mortality, but also lower mortality overall.

Sources

▸ The National Center for Health Statistics publishes regular reports on infant mortality. The final data for 2002 can be found in Natl. Vital Stat. Rep. 2003;52:1-113. The center's latest annual report on trends in health statistics is “Health, United States, 2004,” which includes a chartbook on trends in the health of Americans as well as interactive links (

www.cdc.gov/nchs/hus.htm

▸ The Central Intelligence Agency publishes the World Factbook each year in printed and Internet versions. Data noted in this Master Class can be found online at

www.cia.gov/cia/publications/factbook/rankorder/2091rank.html

▸ Canadian infant mortality statistics can be found at

www.statcan.ca/start/html

www.phac-aspc.gc.ca/publicat/cphr-rspc03

▸ The United Nations Children's Fund (UNICEF) uses data collected in annual report cards from its Innocenti Research Centre. The first report card was published in June 2000, and—along with more recent report cards—can be accessed at

www.unicef-icdc.org/publications

▸ Simon Hales, M.B., and colleagues published the results of their study of the relationship among infant mortality, gross national product, and income distribution (Lancet 1999;354:2047).

However, we are far from being “the best” in the world.

Forty nations surpass the United States in infant mortality, including Singapore (2.29 per 1,000), Sweden (2.77 per 1,000), and Japan (3.26 per 1,000).

Because the U.S. and Canada are neighbors and share a border, similar economies, and comparable levels of technologic sophistication, it is of interest that Canada's infant mortality is 30% lower than that of the United States, which was estimated by the CIA to be 6.5 per 1,000 in 2004.

In 2002, infant mortality worsened slightly in both countries, prompting renewed scrutiny of an ever-important issue.

The Centers for Disease Control and Prevention reported that in 2002, U.S. infant mortality edged upward to 7.0 per 1,000 live births from 6.8 per 1,000 in 2001.

This represents the first rise in 44 years. However, even if this anomaly were a one-time occurrence, it's concerning to note this 0.02% increase when we view the statistic in the context of more than 4 million births.

The same trend occurred in Canada, where infant mortality rose from 5.2 per 1,000 in 2001 to 5.4 per 1,000 in 2002, after progressively falling since the 1960s.

There are numerous factors that may help to elucidate this trend. At the very least, this upturn in infant mortality indicates that perhaps we are not progressing at a pace that many believe is one of the most important measures of a nation's health.

Infant mortality can be divided into two categories: neonatal deaths occurring within the first month of life, and postneonatal deaths occurring later in the first year.

Postneonatal deaths have not increased. In fact, tremendous advances in the etiology and prevention of sudden infant death syndrome have substantially reduced postneonatal deaths over the past decade.

The neonatal increases noted in 2002—and the disparity between the United States' and Canada's infant mortality—have occurred in the early weeks of life, when the most common causes of death include congenital anomalies, problems of transition, and complications of preterm birth. Among these factors, preterm birth stands out as a significant contributor to rising infant mortality.

In the United States, preterm births increased to 12.1%, from 11.9% the previous year.

Although the preterm birth rate also rose slightly in Canada, it was 7.6% in 2002—nearly 40% lower than in the United States.

Why is the preterm birth rate trending upward?

The evidence suggests that the trend is being driven by the use of reproductive technology, leading to multiple births; by women giving birth at later ages; by the necessity of earlier obstetrician intervention when the fetus is in jeopardy; and by complications attributed to a lack of early, consistent prenatal care.

The attempt to save an infant via early delivery has allowed many preemies to live who might have been stillborn in years past.

The recent changes in social trends have influenced the ages at which women decide to have their children. In Ottawa, where I practice, over 60% of mothers in 2003 gave birth when they were older than 30 years, and 23.2% when they were older than 35 years. Although women have a right to be informed about their chances of conceiving and delivering healthy singletons at different ages, physicians have no desire to dictate social policy or individual choice. I have a 5-year-old, and I'm not a young man.

We understand that older women have a higher risk of having a preterm baby, in part because they have a higher risk of having multiples, having pregnancy complications, and having babies with congenital anomalies, three factors that contribute to infant mortality.

Older mothers also are more likely to require assisted reproductive technology (ART).

Although ART procedures are similar in the United States and Canada, and are basically patient-funded in both countries, reproductive technology is increasingly subject to oversight in Canada. A bill that recently passed both the House of Commons and the Senate would strictly regulate clinics and procedures, for example.

A great many ART centers in Canada are university-affiliated, not-for-profit programs, rather than independent clinics. As a result, a controversial issue—such as the implantation of multiple embryos—is debated within the wide academic community of endocrinologists, ob.gyns., neonatologists, pediatricians, and ethicists.

When three sets of quadruplets were born in 1 year at the University of Ottawa, the university-affiliated fertility center demonstrated its responsibility by revising its policies to limit the number of embryos transferred during each cycle. Today, we rarely see quadruplets, although triplets are still not a rarity.

Throughout Canada, rates of multiple birth are lower than in the United States, contributing to lower rates of preterm birth. However, in looking at overall preterm birth statistics, it is worth noting that both nations have unequal rates across populations.

The U.S. National Center for Health Statistics reports that African American infants are nearly twice as likely as non-Hispanic white infants to be born prematurely.

In Canada, the disparity is most clearly defined by income, with those in the lowest income quintile having an infant mortality rate two-thirds higher than that of the highest income quintile. As infant mortality secondary to congenital anomalies and other causes has fallen significantly, the differential is largely a result of a higher rate of preterm birth in lower-income families.

Canada's First Nation and Inuit people face serious health problems, including infant mortality in many communities that is twice the national rate, as do America's Native American populations. Although Canada is an increasingly racially diverse country, other racial disparities are less obvious in measures of health care, such as prenatal care or preterm birth. Income is perhaps a more fitting measure of comparison, and deserving of a wider perspective.

One interesting study examined the gross national products and income distribution in 20 poor and 15 rich nations, determining, as one would expect, that overall infant mortality was inversely proportionate to income.

In rich countries, however, the main contributor to higher infant mortality was not income, but income disparity (Lancet 1999;354:2047).

The United Nations, in its annual Innocenti report card for 2000, explored child poverty in rich nations by using consistent indices to identify the percentage of children living in families with incomes below 50% of each nation's median income.

Child poverty levels ranged from 2.6% in Sweden to 26.6% in Mexico. Canada was 15.5%, and the United States was 22.4%.

Some governments, including those in Scandinavia, take a very active role in making sure that people don't live in poverty. Income disparity is low.

Conversely, income inequality is very high in the United States and may contribute to exceedingly elevated preterm birth rates and infant mortality among African Americans.

A related issue, of course, is access to medical care, which varies greatly among the industrialized nations of the world.

In Canada, where we have universal medical care, prenatal care is available at no cost, with no disincentives to seeking care in the system. Even pregnant women who do not have a primary care physician can walk into a clinic in any city and be seen that day.

As a result, in 2000, well over 95% of Canadian women received prenatal care beginning in the first trimester, compared with 83.7% of American women.

Once again, disparity is evident in the U.S. numbers, with only about 75% of African Americans and fewer than 70% of Native Americans receiving early prenatal care. Indeed, 3.6% of women delivered with no prenatal care, or with care initiated only in the last trimester.

Although the U.S. rate of early prenatal care has improved quite dramatically in the last 15 years, rising 10% since 1990, it still falls short of the care rates in most Western nations. Conversely, I should point out that the Swedes think that even Canada's prenatal care numbers are lousy. In Sweden, virtually 100% of women receive prenatal care throughout their pregnancy.

I would also be remiss if I left the impression that the Canadian health care system, the Swedish system, or any system, for that matter, is perfect. In Canada, one faces a long wait for a hip replacement. Far too many Canadians (about 10%) depend on walk-in clinics because they can't find a primary care physician of their own. Canadians who make more money tend to live longer, and certain groups, such as First Nation and Inuit people, have unequal health outcomes despite access to free care.

In Canada, we learn a lot from the United States, from the abundance of medical research and education to the excellent health care available to many. But in the spirit of learning from each other, U.S. physicians may be interested in studying a neighbor that spends less on health care yet produces not only lower preterm birth rates and infant mortality, but also lower mortality overall.

Sources

▸ The National Center for Health Statistics publishes regular reports on infant mortality. The final data for 2002 can be found in Natl. Vital Stat. Rep. 2003;52:1-113. The center's latest annual report on trends in health statistics is “Health, United States, 2004,” which includes a chartbook on trends in the health of Americans as well as interactive links (

www.cdc.gov/nchs/hus.htm

▸ The Central Intelligence Agency publishes the World Factbook each year in printed and Internet versions. Data noted in this Master Class can be found online at

www.cia.gov/cia/publications/factbook/rankorder/2091rank.html

▸ Canadian infant mortality statistics can be found at

www.statcan.ca/start/html

www.phac-aspc.gc.ca/publicat/cphr-rspc03

▸ The United Nations Children's Fund (UNICEF) uses data collected in annual report cards from its Innocenti Research Centre. The first report card was published in June 2000, and—along with more recent report cards—can be accessed at

www.unicef-icdc.org/publications

▸ Simon Hales, M.B., and colleagues published the results of their study of the relationship among infant mortality, gross national product, and income distribution (Lancet 1999;354:2047).

However, we are far from being “the best” in the world.

Forty nations surpass the United States in infant mortality, including Singapore (2.29 per 1,000), Sweden (2.77 per 1,000), and Japan (3.26 per 1,000).

Because the U.S. and Canada are neighbors and share a border, similar economies, and comparable levels of technologic sophistication, it is of interest that Canada's infant mortality is 30% lower than that of the United States, which was estimated by the CIA to be 6.5 per 1,000 in 2004.

In 2002, infant mortality worsened slightly in both countries, prompting renewed scrutiny of an ever-important issue.

The Centers for Disease Control and Prevention reported that in 2002, U.S. infant mortality edged upward to 7.0 per 1,000 live births from 6.8 per 1,000 in 2001.

This represents the first rise in 44 years. However, even if this anomaly were a one-time occurrence, it's concerning to note this 0.02% increase when we view the statistic in the context of more than 4 million births.

The same trend occurred in Canada, where infant mortality rose from 5.2 per 1,000 in 2001 to 5.4 per 1,000 in 2002, after progressively falling since the 1960s.

There are numerous factors that may help to elucidate this trend. At the very least, this upturn in infant mortality indicates that perhaps we are not progressing at a pace that many believe is one of the most important measures of a nation's health.

Infant mortality can be divided into two categories: neonatal deaths occurring within the first month of life, and postneonatal deaths occurring later in the first year.

Postneonatal deaths have not increased. In fact, tremendous advances in the etiology and prevention of sudden infant death syndrome have substantially reduced postneonatal deaths over the past decade.

The neonatal increases noted in 2002—and the disparity between the United States' and Canada's infant mortality—have occurred in the early weeks of life, when the most common causes of death include congenital anomalies, problems of transition, and complications of preterm birth. Among these factors, preterm birth stands out as a significant contributor to rising infant mortality.

In the United States, preterm births increased to 12.1%, from 11.9% the previous year.

Although the preterm birth rate also rose slightly in Canada, it was 7.6% in 2002—nearly 40% lower than in the United States.

Why is the preterm birth rate trending upward?

The evidence suggests that the trend is being driven by the use of reproductive technology, leading to multiple births; by women giving birth at later ages; by the necessity of earlier obstetrician intervention when the fetus is in jeopardy; and by complications attributed to a lack of early, consistent prenatal care.

The attempt to save an infant via early delivery has allowed many preemies to live who might have been stillborn in years past.

The recent changes in social trends have influenced the ages at which women decide to have their children. In Ottawa, where I practice, over 60% of mothers in 2003 gave birth when they were older than 30 years, and 23.2% when they were older than 35 years. Although women have a right to be informed about their chances of conceiving and delivering healthy singletons at different ages, physicians have no desire to dictate social policy or individual choice. I have a 5-year-old, and I'm not a young man.

We understand that older women have a higher risk of having a preterm baby, in part because they have a higher risk of having multiples, having pregnancy complications, and having babies with congenital anomalies, three factors that contribute to infant mortality.

Older mothers also are more likely to require assisted reproductive technology (ART).

Although ART procedures are similar in the United States and Canada, and are basically patient-funded in both countries, reproductive technology is increasingly subject to oversight in Canada. A bill that recently passed both the House of Commons and the Senate would strictly regulate clinics and procedures, for example.

A great many ART centers in Canada are university-affiliated, not-for-profit programs, rather than independent clinics. As a result, a controversial issue—such as the implantation of multiple embryos—is debated within the wide academic community of endocrinologists, ob.gyns., neonatologists, pediatricians, and ethicists.

When three sets of quadruplets were born in 1 year at the University of Ottawa, the university-affiliated fertility center demonstrated its responsibility by revising its policies to limit the number of embryos transferred during each cycle. Today, we rarely see quadruplets, although triplets are still not a rarity.

Throughout Canada, rates of multiple birth are lower than in the United States, contributing to lower rates of preterm birth. However, in looking at overall preterm birth statistics, it is worth noting that both nations have unequal rates across populations.

The U.S. National Center for Health Statistics reports that African American infants are nearly twice as likely as non-Hispanic white infants to be born prematurely.

In Canada, the disparity is most clearly defined by income, with those in the lowest income quintile having an infant mortality rate two-thirds higher than that of the highest income quintile. As infant mortality secondary to congenital anomalies and other causes has fallen significantly, the differential is largely a result of a higher rate of preterm birth in lower-income families.

Canada's First Nation and Inuit people face serious health problems, including infant mortality in many communities that is twice the national rate, as do America's Native American populations. Although Canada is an increasingly racially diverse country, other racial disparities are less obvious in measures of health care, such as prenatal care or preterm birth. Income is perhaps a more fitting measure of comparison, and deserving of a wider perspective.

One interesting study examined the gross national products and income distribution in 20 poor and 15 rich nations, determining, as one would expect, that overall infant mortality was inversely proportionate to income.

In rich countries, however, the main contributor to higher infant mortality was not income, but income disparity (Lancet 1999;354:2047).

The United Nations, in its annual Innocenti report card for 2000, explored child poverty in rich nations by using consistent indices to identify the percentage of children living in families with incomes below 50% of each nation's median income.

Child poverty levels ranged from 2.6% in Sweden to 26.6% in Mexico. Canada was 15.5%, and the United States was 22.4%.

Some governments, including those in Scandinavia, take a very active role in making sure that people don't live in poverty. Income disparity is low.

Conversely, income inequality is very high in the United States and may contribute to exceedingly elevated preterm birth rates and infant mortality among African Americans.

A related issue, of course, is access to medical care, which varies greatly among the industrialized nations of the world.

In Canada, where we have universal medical care, prenatal care is available at no cost, with no disincentives to seeking care in the system. Even pregnant women who do not have a primary care physician can walk into a clinic in any city and be seen that day.

As a result, in 2000, well over 95% of Canadian women received prenatal care beginning in the first trimester, compared with 83.7% of American women.

Once again, disparity is evident in the U.S. numbers, with only about 75% of African Americans and fewer than 70% of Native Americans receiving early prenatal care. Indeed, 3.6% of women delivered with no prenatal care, or with care initiated only in the last trimester.

Although the U.S. rate of early prenatal care has improved quite dramatically in the last 15 years, rising 10% since 1990, it still falls short of the care rates in most Western nations. Conversely, I should point out that the Swedes think that even Canada's prenatal care numbers are lousy. In Sweden, virtually 100% of women receive prenatal care throughout their pregnancy.

I would also be remiss if I left the impression that the Canadian health care system, the Swedish system, or any system, for that matter, is perfect. In Canada, one faces a long wait for a hip replacement. Far too many Canadians (about 10%) depend on walk-in clinics because they can't find a primary care physician of their own. Canadians who make more money tend to live longer, and certain groups, such as First Nation and Inuit people, have unequal health outcomes despite access to free care.

In Canada, we learn a lot from the United States, from the abundance of medical research and education to the excellent health care available to many. But in the spirit of learning from each other, U.S. physicians may be interested in studying a neighbor that spends less on health care yet produces not only lower preterm birth rates and infant mortality, but also lower mortality overall.

Sources

▸ The National Center for Health Statistics publishes regular reports on infant mortality. The final data for 2002 can be found in Natl. Vital Stat. Rep. 2003;52:1-113. The center's latest annual report on trends in health statistics is “Health, United States, 2004,” which includes a chartbook on trends in the health of Americans as well as interactive links (

www.cdc.gov/nchs/hus.htm

▸ The Central Intelligence Agency publishes the World Factbook each year in printed and Internet versions. Data noted in this Master Class can be found online at

www.cia.gov/cia/publications/factbook/rankorder/2091rank.html

▸ Canadian infant mortality statistics can be found at

www.statcan.ca/start/html

www.phac-aspc.gc.ca/publicat/cphr-rspc03

▸ The United Nations Children's Fund (UNICEF) uses data collected in annual report cards from its Innocenti Research Centre. The first report card was published in June 2000, and—along with more recent report cards—can be accessed at

www.unicef-icdc.org/publications

▸ Simon Hales, M.B., and colleagues published the results of their study of the relationship among infant mortality, gross national product, and income distribution (Lancet 1999;354:2047).

Infant mortality is one of the dominant measures by which a nation's health is judged. Many factors contribute to the number of babies who survive in a given country, making infant mortality a rather unrefined gauge of overall health. Yet it has been accepted worldwide as a generally fair and realistic reflection of national health.

Leaders in the medical community and government have long recognized that the United States has unacceptably high infant mortality in comparison with other nations. I served, in fact, on the Department of Health and Human Services' Secretary's Committee on Infant Mortality under President George H.W. Bush, as part of a major effort to reduce by half our infant mortality. We still have not succeeded, despite concerted efforts.

Because this is a complex issue that will be solved only by using multiple strategies, we may do well to learn from other countries' successes. The Scandinavian countries, which boast very low infant mortality, have homogeneous populations that are difficult to compare with our own. But right next door is Canada, a country with an increasingly diverse population that may serve as a more analogous example of how programs can work to reduce infant mortality.

For a commentary on this important issue, we turn to C. Robin Walker, M.D., Ch.B., president of the Canadian Paediatric Society and professor of pediatrics at the University of Ottawa. He has studied infant mortality as an international issue, publishing on such topics as population-based approaches to prevention of preterm birth, an important contributor to infant mortality.

We hope his thoughts will provide fresh insight into a very important health measure that we continue to try to improve.

Infant mortality is one of the dominant measures by which a nation's health is judged. Many factors contribute to the number of babies who survive in a given country, making infant mortality a rather unrefined gauge of overall health. Yet it has been accepted worldwide as a generally fair and realistic reflection of national health.

Leaders in the medical community and government have long recognized that the United States has unacceptably high infant mortality in comparison with other nations. I served, in fact, on the Department of Health and Human Services' Secretary's Committee on Infant Mortality under President George H.W. Bush, as part of a major effort to reduce by half our infant mortality. We still have not succeeded, despite concerted efforts.

Because this is a complex issue that will be solved only by using multiple strategies, we may do well to learn from other countries' successes. The Scandinavian countries, which boast very low infant mortality, have homogeneous populations that are difficult to compare with our own. But right next door is Canada, a country with an increasingly diverse population that may serve as a more analogous example of how programs can work to reduce infant mortality.

For a commentary on this important issue, we turn to C. Robin Walker, M.D., Ch.B., president of the Canadian Paediatric Society and professor of pediatrics at the University of Ottawa. He has studied infant mortality as an international issue, publishing on such topics as population-based approaches to prevention of preterm birth, an important contributor to infant mortality.

We hope his thoughts will provide fresh insight into a very important health measure that we continue to try to improve.

Infant mortality is one of the dominant measures by which a nation's health is judged. Many factors contribute to the number of babies who survive in a given country, making infant mortality a rather unrefined gauge of overall health. Yet it has been accepted worldwide as a generally fair and realistic reflection of national health.

Leaders in the medical community and government have long recognized that the United States has unacceptably high infant mortality in comparison with other nations. I served, in fact, on the Department of Health and Human Services' Secretary's Committee on Infant Mortality under President George H.W. Bush, as part of a major effort to reduce by half our infant mortality. We still have not succeeded, despite concerted efforts.

Because this is a complex issue that will be solved only by using multiple strategies, we may do well to learn from other countries' successes. The Scandinavian countries, which boast very low infant mortality, have homogeneous populations that are difficult to compare with our own. But right next door is Canada, a country with an increasingly diverse population that may serve as a more analogous example of how programs can work to reduce infant mortality.

For a commentary on this important issue, we turn to C. Robin Walker, M.D., Ch.B., president of the Canadian Paediatric Society and professor of pediatrics at the University of Ottawa. He has studied infant mortality as an international issue, publishing on such topics as population-based approaches to prevention of preterm birth, an important contributor to infant mortality.

We hope his thoughts will provide fresh insight into a very important health measure that we continue to try to improve.

As editor of the Master Class columns on gynecology, I was very proud to have C.Y. Liu, M.D., present an excellent two-part discourse on pelvic floor prolapse in the October 1, 2004, and November 1, 2004, issues of OB.GYN. NEWS.

I subsequently received a letter to the editor from Marvin H. Terry Grody, M.D. In my mind, Dr. Grody has raised compelling issues, especially in regard to the importance of the perineal body in pelvic floor prolapse. Because of this, I have asked Dr. Liu and a panel of experts to discuss Dr. Grody's concerns.

I trust you will find this discussion both interesting and informative.

Dear Editors:

In the Oct. 1, 2004, issue of Ob.Gyn. News, there appeared Part 1 of a two-part series entitled “Laparoscopic Pelvic Reconstructive Surgery.” The author, C.Y. Liu, M.D., who is a well-reputed and skilled laparoscopic surgeon, acceptably covered the issues of defects of the pelvic supportive and suspensory mechanisms and their effects on associated organs. But from the viewpoint of a vaginal and pelvic reconstructive surgeon, he embodied a major misconception in his statement, “The perineal membrane and perineal body are not very crucial for pelvic organ support.”

He is not only dead wrong, but he is giving misinformation that could be seriously destructive to surgery performed by a myriad of minimally experienced young surgeons whom experts in the field are trying tenaciously to convince otherwise.

Before I go further into this matter, I must first suppress my emotionally charged conviction (shared by many others) that the average gynecologic surgeon will not achieve anywhere near the degree of success working through a telescope that has been thrust through the abdominal wall as she or he could attain much more directly with less time and expense—and probably less risk—by using alternative approaches.

Contrary to Dr. Liu's disregard of any contributive importance of the perineal body (PB), pelvic reconstructive surgeons universally consider a disrupted PB to be a critical obstacle to the achievement of durably effective success in pelvic anatomical and functional restoration. Over a period of 4 decades starting in the 1960s, David H. Nichols, M.D.—whom most of us view as one of the most renowned vaginal surgeons—firmly and repeatedly established the mandatory requirements of restitution of the normal vaginal axis in the correction of the anatomically defective pelvic floor.

For reference, a full description of the normal vaginal axis and its vital role in good pelvic support can be found in my chapter on colpoperineorrhaphy in the ninth edition of TeLinde's Operative Gynecology (Philadelphia: Lippincott Williams & Wilkins, 2003, p. 966-85).

The PB is a key element in the structural composition of the normal vaginal axis. If significant defects in the PB are ignored and not completely repaired to natural configuration in this commonly coexistent lesion in pelvic floor anatomical failure, then no matter how wonderful the surgeon feels about his or her effort in correcting the other defects, the operation is almost certainly doomed to fail in time. Such inevitability relates to the interdependence of all the elements of the connective tissue network running through the pelvis. An ignored, significantly defective PB can become the weak link that will blow the entire chain of support.

Even if we uncover the rare gynecologic surgeon possessed of laparoscopic skill equivalent to that of Dr. Liu, if the patient does not undergo a full perineorrhaphy from the vaginal approach as the last part of the total operation, then that surgeon must be considered stupid.

Finally, I must question the wisdom of publishing this laparoscopy series that focuses on a surgical approach that will unquestionably be within the province of only a highly-specialized, well-trained, innately gifted few when other easier, safer, very effective, and far less costly and time-consuming procedures can be ably pursued by a significantly larger segment of qualified operating practitioners.

Given today's world of astounding technological feats, will such a truly perverse printed exposure stimulate adventurous young gynecologic surgeons who think they are much better than they really are into imprudent undertakings beyond their true capabilities, leading to serious injury to their patients? Goodness knows what difficulties we already find in our cluttered residency programs in getting basic maneuvers (like vaginal hysterectomy) across, let alone highly sophisticated, industry-driven, potentially dangerous operative challenges performed through a spyglass.

If there are critics abroad who think I am wrong, let them please tell me.

Rather than repudiating Dr. Grody's opinion about laparoscopic surgery, I will only respond to his point about the importance of the perineal membrane and PB to pelvic organ support.

All defects should be repaired at the time of pelvic floor reconstructive surgery. Any tear or defect in the area of the perineal membrane or PB should be repaired concurrently with pelvic floor reconstruction. This point was emphasized in the final step outlined in Part 2 of my series: “Repair the rectocele and perform perineorrhaphy vaginally if necessary.”

Based upon my understanding of the functional pelvic support anatomy as well as clinical observation, I maintain my position that “the perineal membrane and perineal body are not very crucial for pelvic organ support.”

The perineal membrane is a single layer of fibromuscular tissue that spans the anterior triangle of the pelvic outlet. Laterally, it attaches to the ischiopubic ramus; medially, it fuses with the sidewalls of the vagina and perineal body. The anterior portion of the perineal membrane is fused with the muscles of the distal urethra. Rather than forming a supportive sheet as it does in the male, the perineal membrane in the female—because of the large opening of the vagina—provides only lateral attachment for the PB and some support for the lower urethra.

The PB is an ill-defined, bordered mass of dense connective tissue lying between the vagina and anus. Fused anteriorly to the posterior vaginal wall and attached laterally to the perineal membrane and bulbocavernosus and superficial transverse perineal muscles, a significant portion of what is clinically called the perineal body is actually the muscle of the external anal sphincter. The strong upward traction of the levator ani muscles is much more important in maintaining vaginal outlet support than are the bulbocavernosus and superficial transverse perineal muscle.

Contrary to Dr. Grody's assertion that the PB makes a substantial contribution to pelvic support, in actuality the support is minimal. Rather, restoration of the PB is important for sexual function and anal/fecal continence. I have examined several patients with no PB as the result of chronic unrepaired fourth-degree obstetric lacerations, yet none of them had prolapse. Similarly, women who have had a radical resection of the anus and rectum for cancer, including the entire removal of the PB, suffered no significant prolapse.

Because considerable descent (up to 1 inch) of the PB is possible during voluntary straining, the perineal membrane and PB cannot be the main supportive layer of the genital outlet. The fact that the PB can move backward 3-4 cm toward the sacrum when a weighted speculum is placed in the posterior vagina likewise indicates that the position of the PB is determined by the levator ani muscle rather than by any inherent importance of its own.

Advances in technology afford greater magnification, visualization, and accuracy—leading to a level of surgical precision heretofore impossible with the relatively “blind” vaginal approach. We must train young surgeons for these state-of-the-art advances.

Although I have the utmost respect for both Dr. Grody and Dr. Liu and I believe that everyone is entitled to his or her own opinion, Dr Liu's article is certainly not worthy of such admonishment.

Dr. Liu not only correctly addresses normal vaginal anatomy, clinical assessment, and one surgeon's approach to the anatomical correction of symptomatic prolapse, he does so in a concise, informative manner.

Dr. Grody's belief that the perineal membrane and PB are crucial for pelvic organ support is indeed just that: his belief. Using the PubMed search term “perineal body surgery,” I found no scientific literature written in the past 40 years that supports the concept that either the perineal membrane or the PB is crucial in the support of any organs of the pelvis. I have yet to read or find an article that suggests that the cure rates of sacrospinous ligament suspension; sacral colpopexy; paravaginal repair; uterosacral ligament suspension; enterocele repair; or Burch, sling, or any other prolapse corrective surgery—including colpocleisis or Lefort procedures—are improved by repairing the PB.

Furthermore, there is no scientific literature that supports the concept that poor perineal support increases the incidence of prolapse. If this were a fact, patients with traumatic or congenital cloaca would also suffer a greater incidence of vaginal prolapse. I have not seen or read of any scientific literature or text that can directly show a cause-and-effect relationship between a damaged PB and vaginal prolapse.

Dr. Grody is a purist in his pursuit of vaginal anatomic correction, but this fine trait does not constitute scientific proof for his allegation. He has the right to theorize that the anatomical correction is essential to improve long-term cure rates of prolapse surgery. But a theory is belief unsupported by substantial fact, and will thus remain just a theory.

After reading Dr. Liu's article on laparoscopic pelvic reconstructive surgery and Dr. Grody's response, I found myself perplexed. How is it that two experienced and respected surgeons can underappreciate each other's perspective on pelvic reconstructive surgery?

For the most part, I agree with most of what each has stated but disagree on the finer points. I must confess that being predominately a laparoscopic or minimally invasive surgeon, I too did not completely comprehend the complexity and functional anatomy of the PB and membrane as an important element in pelvic floor support until more recently.

Thanks to cadaver sections and MRI studies reported by John O.L. Delancey, M.D., at the joint annual meeting of the American Urogynecologic Society and the Society of Gynecologic Surgeons in 2004, we realize that the perineal membrane is a complex 3-D structure composed of a dorsal and ventral portion rather than a trilaminar sheet as previously thought. His description of the anatomical relationship to the compressor urethra, urethra vaginal sphincter, arcus tendineus, pubic bone, and levator muscles underscores the importance of this structure in pelvic support.

We now have level 1 evidence of the laparoscope's benefit in sacral colpopexies compared with an open procedure, as well as its inferiority in treating stress urinary incontinence when comparing a laparoscopic Burch with a transvaginal tape procedure. But the laparoscope is a tool that requires proper training to master. Thanks to the pioneering efforts of Dr. Liu and Dr. Miklos, the development of training centers, and the support of organizations like the American Association of Gynecologic Laparoscopists, it is no longer the gifted few who use this valuable instrument.

The recent articles by Dr. C.Y. Liu in OB.GYN.NEWS on laparoscopic pelvic reconstructive surgery, Parts 1 and 2, are a must-read for any gynecologic surgeon performing reconstructive vaginal surgery. Although the article presents Dr. Liu's laparoscopic approach to problems of vaginal suspension and support, the anatomy presented and the surgical steps discussed are clearly applicable to the repair of any vaginal prolapse via any surgical approach, including vaginal and abdominal.

The anatomy of genital prolapse is up to date, well written, and clearly explained. Part 1 contains many pearls of insight from a master of this anatomy, and it summarizes our current concepts of vaginal suspension and support. The section on clinical assessment of prolapse is practical and very helpful.

The surgical techniques presented are anatomical and readily applicable. Dr. Liu explains how to safely dissect out and investigate the suspensory anatomy to clearly define the anatomical defects that caused the vaginal prolapse. Not only does Dr. Liu address and repair the specific breaks in the continuity of the visceral connective tissue suspensory network, but he presents an excellent dissection technique for safeguarding the ureters.

One point that should have been emphasized is the requirement for cystoscopic confirmation of bilateral ureteral functioning at the end of the case.

The article explains that one of the three supporting layers of the female pelvic organs is “the perineal membrane/external anal sphincter.” What is not said is that the anal sphincter is an important component of the posterior part of the PB.

The lower third of the vagina and the anal canal/anal sphincter are fused with the PB. The PB is shaped roughly like a pyramid, with the base between the vaginal introitus and the anal sphincter. The apex is found at the junction of the lower third and the middle third of the vagina, and at the rectoanal junction. At the apex of the perineal body, the vagina slopes to a more horizontal orientation in the standing patient, whereas the anal canal forms a right angle with the lower rectum.

Portions of the pubococcygeus and puborectalis muscles insert into the apex of the PB. The rectovaginal fascia also inserts into the apex of the PB and helps in its proper anatomical orientation. The intact PB positions itself and the anus just above the level of the ischial tuberosities. The fusion of the anus and anal canal with the PB is important for their anatomical positioning and physiologic functioning in fecal continence. The fusion of the lower third of the vagina with the PB is important for its anatomical positioning and physiologic functioning in pelvic organ support. The PB assists in closing off the genital hiatus at times of increased intrapelvic pressures, supporting the pelvic organs. Another support mechanism is the flap-valve action of the levator plate.

Many women with vaginal prolapse demonstrate abnormal descent of the perineum. Dr. Liu states, “The active support of the pelvic floor comes from the levator ani muscles (the iliococcygeus and pubococcygeus muscles). These muscles close off the pelvic floor so the pelvic organs can rest upon them without tension.” This statement is true.

Dr. Liu does not mention the important action of the levator plate or the action of the PB in this vaginal support mechanism. In fact, with a poorly supportive levator plate, as is frequently seen in vaginal prolapse patients, a well-reconstructed PB will substitute as a backstop against which the resuspended vagina can be compressed for support.

The reconstructed PB will help close off the genital hiatus at times of mechanical pelvic stress. The PB must be reconstructed in shape and bulk to support and orient the anal canal and lower third of the vagina, but also to position itself and the anal canal at or above the level of the ischial tuberosities.

As Dr. Liu implies, we cannot repair or completely rehabilitate damaged and weakened pelvic floor muscles and their innervations. We should surgically reconstruct a disrupted PB. I do feel that Dr. Liu does indeed perform perineoplasty on many of his prolapse patients. He simply emphasized the reconstruction and proper placement of the pericervical ring in his excellent article.

MARVIN H. TERRY GRODY, M.D., is a professor of obstetrics and gynecology and senior attending gynecology consultant, Robert Wood Johnson Medical School at Camden (N.J.).

C.Y. LIU, M.D., is the director of the Manhattan Women's Laser Center, New York.

JOHN R. MIKLOS, M.D., is the director of the Atlanta Center for Laparoscopic Urogynecology.

VINCENT R. LUCENTE, M.D., is chief of urogynecology at Abington (Pa.) Memorial Hospital and associate professor of ob.gyn. at Pennsylvania State Medical Center, Hershey.

ROBERT M. ROGERS JR., M.D., is an attending gynecologist at the Reading Hospital and Medical Center in West Reading, Pa.

As editor of the Master Class columns on gynecology, I was very proud to have C.Y. Liu, M.D., present an excellent two-part discourse on pelvic floor prolapse in the October 1, 2004, and November 1, 2004, issues of OB.GYN. NEWS.

I subsequently received a letter to the editor from Marvin H. Terry Grody, M.D. In my mind, Dr. Grody has raised compelling issues, especially in regard to the importance of the perineal body in pelvic floor prolapse. Because of this, I have asked Dr. Liu and a panel of experts to discuss Dr. Grody's concerns.

I trust you will find this discussion both interesting and informative.

Dear Editors:

In the Oct. 1, 2004, issue of Ob.Gyn. News, there appeared Part 1 of a two-part series entitled “Laparoscopic Pelvic Reconstructive Surgery.” The author, C.Y. Liu, M.D., who is a well-reputed and skilled laparoscopic surgeon, acceptably covered the issues of defects of the pelvic supportive and suspensory mechanisms and their effects on associated organs. But from the viewpoint of a vaginal and pelvic reconstructive surgeon, he embodied a major misconception in his statement, “The perineal membrane and perineal body are not very crucial for pelvic organ support.”

He is not only dead wrong, but he is giving misinformation that could be seriously destructive to surgery performed by a myriad of minimally experienced young surgeons whom experts in the field are trying tenaciously to convince otherwise.

Before I go further into this matter, I must first suppress my emotionally charged conviction (shared by many others) that the average gynecologic surgeon will not achieve anywhere near the degree of success working through a telescope that has been thrust through the abdominal wall as she or he could attain much more directly with less time and expense—and probably less risk—by using alternative approaches.

Contrary to Dr. Liu's disregard of any contributive importance of the perineal body (PB), pelvic reconstructive surgeons universally consider a disrupted PB to be a critical obstacle to the achievement of durably effective success in pelvic anatomical and functional restoration. Over a period of 4 decades starting in the 1960s, David H. Nichols, M.D.—whom most of us view as one of the most renowned vaginal surgeons—firmly and repeatedly established the mandatory requirements of restitution of the normal vaginal axis in the correction of the anatomically defective pelvic floor.

For reference, a full description of the normal vaginal axis and its vital role in good pelvic support can be found in my chapter on colpoperineorrhaphy in the ninth edition of TeLinde's Operative Gynecology (Philadelphia: Lippincott Williams & Wilkins, 2003, p. 966-85).

The PB is a key element in the structural composition of the normal vaginal axis. If significant defects in the PB are ignored and not completely repaired to natural configuration in this commonly coexistent lesion in pelvic floor anatomical failure, then no matter how wonderful the surgeon feels about his or her effort in correcting the other defects, the operation is almost certainly doomed to fail in time. Such inevitability relates to the interdependence of all the elements of the connective tissue network running through the pelvis. An ignored, significantly defective PB can become the weak link that will blow the entire chain of support.

Even if we uncover the rare gynecologic surgeon possessed of laparoscopic skill equivalent to that of Dr. Liu, if the patient does not undergo a full perineorrhaphy from the vaginal approach as the last part of the total operation, then that surgeon must be considered stupid.

Finally, I must question the wisdom of publishing this laparoscopy series that focuses on a surgical approach that will unquestionably be within the province of only a highly-specialized, well-trained, innately gifted few when other easier, safer, very effective, and far less costly and time-consuming procedures can be ably pursued by a significantly larger segment of qualified operating practitioners.

Given today's world of astounding technological feats, will such a truly perverse printed exposure stimulate adventurous young gynecologic surgeons who think they are much better than they really are into imprudent undertakings beyond their true capabilities, leading to serious injury to their patients? Goodness knows what difficulties we already find in our cluttered residency programs in getting basic maneuvers (like vaginal hysterectomy) across, let alone highly sophisticated, industry-driven, potentially dangerous operative challenges performed through a spyglass.

If there are critics abroad who think I am wrong, let them please tell me.

Rather than repudiating Dr. Grody's opinion about laparoscopic surgery, I will only respond to his point about the importance of the perineal membrane and PB to pelvic organ support.

All defects should be repaired at the time of pelvic floor reconstructive surgery. Any tear or defect in the area of the perineal membrane or PB should be repaired concurrently with pelvic floor reconstruction. This point was emphasized in the final step outlined in Part 2 of my series: “Repair the rectocele and perform perineorrhaphy vaginally if necessary.”

Based upon my understanding of the functional pelvic support anatomy as well as clinical observation, I maintain my position that “the perineal membrane and perineal body are not very crucial for pelvic organ support.”

The perineal membrane is a single layer of fibromuscular tissue that spans the anterior triangle of the pelvic outlet. Laterally, it attaches to the ischiopubic ramus; medially, it fuses with the sidewalls of the vagina and perineal body. The anterior portion of the perineal membrane is fused with the muscles of the distal urethra. Rather than forming a supportive sheet as it does in the male, the perineal membrane in the female—because of the large opening of the vagina—provides only lateral attachment for the PB and some support for the lower urethra.

The PB is an ill-defined, bordered mass of dense connective tissue lying between the vagina and anus. Fused anteriorly to the posterior vaginal wall and attached laterally to the perineal membrane and bulbocavernosus and superficial transverse perineal muscles, a significant portion of what is clinically called the perineal body is actually the muscle of the external anal sphincter. The strong upward traction of the levator ani muscles is much more important in maintaining vaginal outlet support than are the bulbocavernosus and superficial transverse perineal muscle.

Contrary to Dr. Grody's assertion that the PB makes a substantial contribution to pelvic support, in actuality the support is minimal. Rather, restoration of the PB is important for sexual function and anal/fecal continence. I have examined several patients with no PB as the result of chronic unrepaired fourth-degree obstetric lacerations, yet none of them had prolapse. Similarly, women who have had a radical resection of the anus and rectum for cancer, including the entire removal of the PB, suffered no significant prolapse.

Because considerable descent (up to 1 inch) of the PB is possible during voluntary straining, the perineal membrane and PB cannot be the main supportive layer of the genital outlet. The fact that the PB can move backward 3-4 cm toward the sacrum when a weighted speculum is placed in the posterior vagina likewise indicates that the position of the PB is determined by the levator ani muscle rather than by any inherent importance of its own.

Advances in technology afford greater magnification, visualization, and accuracy—leading to a level of surgical precision heretofore impossible with the relatively “blind” vaginal approach. We must train young surgeons for these state-of-the-art advances.

Although I have the utmost respect for both Dr. Grody and Dr. Liu and I believe that everyone is entitled to his or her own opinion, Dr Liu's article is certainly not worthy of such admonishment.

Dr. Liu not only correctly addresses normal vaginal anatomy, clinical assessment, and one surgeon's approach to the anatomical correction of symptomatic prolapse, he does so in a concise, informative manner.

Dr. Grody's belief that the perineal membrane and PB are crucial for pelvic organ support is indeed just that: his belief. Using the PubMed search term “perineal body surgery,” I found no scientific literature written in the past 40 years that supports the concept that either the perineal membrane or the PB is crucial in the support of any organs of the pelvis. I have yet to read or find an article that suggests that the cure rates of sacrospinous ligament suspension; sacral colpopexy; paravaginal repair; uterosacral ligament suspension; enterocele repair; or Burch, sling, or any other prolapse corrective surgery—including colpocleisis or Lefort procedures—are improved by repairing the PB.

Furthermore, there is no scientific literature that supports the concept that poor perineal support increases the incidence of prolapse. If this were a fact, patients with traumatic or congenital cloaca would also suffer a greater incidence of vaginal prolapse. I have not seen or read of any scientific literature or text that can directly show a cause-and-effect relationship between a damaged PB and vaginal prolapse.

Dr. Grody is a purist in his pursuit of vaginal anatomic correction, but this fine trait does not constitute scientific proof for his allegation. He has the right to theorize that the anatomical correction is essential to improve long-term cure rates of prolapse surgery. But a theory is belief unsupported by substantial fact, and will thus remain just a theory.

After reading Dr. Liu's article on laparoscopic pelvic reconstructive surgery and Dr. Grody's response, I found myself perplexed. How is it that two experienced and respected surgeons can underappreciate each other's perspective on pelvic reconstructive surgery?

For the most part, I agree with most of what each has stated but disagree on the finer points. I must confess that being predominately a laparoscopic or minimally invasive surgeon, I too did not completely comprehend the complexity and functional anatomy of the PB and membrane as an important element in pelvic floor support until more recently.

Thanks to cadaver sections and MRI studies reported by John O.L. Delancey, M.D., at the joint annual meeting of the American Urogynecologic Society and the Society of Gynecologic Surgeons in 2004, we realize that the perineal membrane is a complex 3-D structure composed of a dorsal and ventral portion rather than a trilaminar sheet as previously thought. His description of the anatomical relationship to the compressor urethra, urethra vaginal sphincter, arcus tendineus, pubic bone, and levator muscles underscores the importance of this structure in pelvic support.

We now have level 1 evidence of the laparoscope's benefit in sacral colpopexies compared with an open procedure, as well as its inferiority in treating stress urinary incontinence when comparing a laparoscopic Burch with a transvaginal tape procedure. But the laparoscope is a tool that requires proper training to master. Thanks to the pioneering efforts of Dr. Liu and Dr. Miklos, the development of training centers, and the support of organizations like the American Association of Gynecologic Laparoscopists, it is no longer the gifted few who use this valuable instrument.

The recent articles by Dr. C.Y. Liu in OB.GYN.NEWS on laparoscopic pelvic reconstructive surgery, Parts 1 and 2, are a must-read for any gynecologic surgeon performing reconstructive vaginal surgery. Although the article presents Dr. Liu's laparoscopic approach to problems of vaginal suspension and support, the anatomy presented and the surgical steps discussed are clearly applicable to the repair of any vaginal prolapse via any surgical approach, including vaginal and abdominal.

The anatomy of genital prolapse is up to date, well written, and clearly explained. Part 1 contains many pearls of insight from a master of this anatomy, and it summarizes our current concepts of vaginal suspension and support. The section on clinical assessment of prolapse is practical and very helpful.

The surgical techniques presented are anatomical and readily applicable. Dr. Liu explains how to safely dissect out and investigate the suspensory anatomy to clearly define the anatomical defects that caused the vaginal prolapse. Not only does Dr. Liu address and repair the specific breaks in the continuity of the visceral connective tissue suspensory network, but he presents an excellent dissection technique for safeguarding the ureters.

One point that should have been emphasized is the requirement for cystoscopic confirmation of bilateral ureteral functioning at the end of the case.

The article explains that one of the three supporting layers of the female pelvic organs is “the perineal membrane/external anal sphincter.” What is not said is that the anal sphincter is an important component of the posterior part of the PB.

The lower third of the vagina and the anal canal/anal sphincter are fused with the PB. The PB is shaped roughly like a pyramid, with the base between the vaginal introitus and the anal sphincter. The apex is found at the junction of the lower third and the middle third of the vagina, and at the rectoanal junction. At the apex of the perineal body, the vagina slopes to a more horizontal orientation in the standing patient, whereas the anal canal forms a right angle with the lower rectum.

Portions of the pubococcygeus and puborectalis muscles insert into the apex of the PB. The rectovaginal fascia also inserts into the apex of the PB and helps in its proper anatomical orientation. The intact PB positions itself and the anus just above the level of the ischial tuberosities. The fusion of the anus and anal canal with the PB is important for their anatomical positioning and physiologic functioning in fecal continence. The fusion of the lower third of the vagina with the PB is important for its anatomical positioning and physiologic functioning in pelvic organ support. The PB assists in closing off the genital hiatus at times of increased intrapelvic pressures, supporting the pelvic organs. Another support mechanism is the flap-valve action of the levator plate.

Many women with vaginal prolapse demonstrate abnormal descent of the perineum. Dr. Liu states, “The active support of the pelvic floor comes from the levator ani muscles (the iliococcygeus and pubococcygeus muscles). These muscles close off the pelvic floor so the pelvic organs can rest upon them without tension.” This statement is true.

Dr. Liu does not mention the important action of the levator plate or the action of the PB in this vaginal support mechanism. In fact, with a poorly supportive levator plate, as is frequently seen in vaginal prolapse patients, a well-reconstructed PB will substitute as a backstop against which the resuspended vagina can be compressed for support.

The reconstructed PB will help close off the genital hiatus at times of mechanical pelvic stress. The PB must be reconstructed in shape and bulk to support and orient the anal canal and lower third of the vagina, but also to position itself and the anal canal at or above the level of the ischial tuberosities.

As Dr. Liu implies, we cannot repair or completely rehabilitate damaged and weakened pelvic floor muscles and their innervations. We should surgically reconstruct a disrupted PB. I do feel that Dr. Liu does indeed perform perineoplasty on many of his prolapse patients. He simply emphasized the reconstruction and proper placement of the pericervical ring in his excellent article.

MARVIN H. TERRY GRODY, M.D., is a professor of obstetrics and gynecology and senior attending gynecology consultant, Robert Wood Johnson Medical School at Camden (N.J.).

C.Y. LIU, M.D., is the director of the Manhattan Women's Laser Center, New York.

JOHN R. MIKLOS, M.D., is the director of the Atlanta Center for Laparoscopic Urogynecology.

VINCENT R. LUCENTE, M.D., is chief of urogynecology at Abington (Pa.) Memorial Hospital and associate professor of ob.gyn. at Pennsylvania State Medical Center, Hershey.

ROBERT M. ROGERS JR., M.D., is an attending gynecologist at the Reading Hospital and Medical Center in West Reading, Pa.

As editor of the Master Class columns on gynecology, I was very proud to have C.Y. Liu, M.D., present an excellent two-part discourse on pelvic floor prolapse in the October 1, 2004, and November 1, 2004, issues of OB.GYN. NEWS.

I subsequently received a letter to the editor from Marvin H. Terry Grody, M.D. In my mind, Dr. Grody has raised compelling issues, especially in regard to the importance of the perineal body in pelvic floor prolapse. Because of this, I have asked Dr. Liu and a panel of experts to discuss Dr. Grody's concerns.

I trust you will find this discussion both interesting and informative.

Dear Editors:

In the Oct. 1, 2004, issue of Ob.Gyn. News, there appeared Part 1 of a two-part series entitled “Laparoscopic Pelvic Reconstructive Surgery.” The author, C.Y. Liu, M.D., who is a well-reputed and skilled laparoscopic surgeon, acceptably covered the issues of defects of the pelvic supportive and suspensory mechanisms and their effects on associated organs. But from the viewpoint of a vaginal and pelvic reconstructive surgeon, he embodied a major misconception in his statement, “The perineal membrane and perineal body are not very crucial for pelvic organ support.”

He is not only dead wrong, but he is giving misinformation that could be seriously destructive to surgery performed by a myriad of minimally experienced young surgeons whom experts in the field are trying tenaciously to convince otherwise.

Before I go further into this matter, I must first suppress my emotionally charged conviction (shared by many others) that the average gynecologic surgeon will not achieve anywhere near the degree of success working through a telescope that has been thrust through the abdominal wall as she or he could attain much more directly with less time and expense—and probably less risk—by using alternative approaches.

Contrary to Dr. Liu's disregard of any contributive importance of the perineal body (PB), pelvic reconstructive surgeons universally consider a disrupted PB to be a critical obstacle to the achievement of durably effective success in pelvic anatomical and functional restoration. Over a period of 4 decades starting in the 1960s, David H. Nichols, M.D.—whom most of us view as one of the most renowned vaginal surgeons—firmly and repeatedly established the mandatory requirements of restitution of the normal vaginal axis in the correction of the anatomically defective pelvic floor.

For reference, a full description of the normal vaginal axis and its vital role in good pelvic support can be found in my chapter on colpoperineorrhaphy in the ninth edition of TeLinde's Operative Gynecology (Philadelphia: Lippincott Williams & Wilkins, 2003, p. 966-85).

The PB is a key element in the structural composition of the normal vaginal axis. If significant defects in the PB are ignored and not completely repaired to natural configuration in this commonly coexistent lesion in pelvic floor anatomical failure, then no matter how wonderful the surgeon feels about his or her effort in correcting the other defects, the operation is almost certainly doomed to fail in time. Such inevitability relates to the interdependence of all the elements of the connective tissue network running through the pelvis. An ignored, significantly defective PB can become the weak link that will blow the entire chain of support.

Even if we uncover the rare gynecologic surgeon possessed of laparoscopic skill equivalent to that of Dr. Liu, if the patient does not undergo a full perineorrhaphy from the vaginal approach as the last part of the total operation, then that surgeon must be considered stupid.

Finally, I must question the wisdom of publishing this laparoscopy series that focuses on a surgical approach that will unquestionably be within the province of only a highly-specialized, well-trained, innately gifted few when other easier, safer, very effective, and far less costly and time-consuming procedures can be ably pursued by a significantly larger segment of qualified operating practitioners.

Given today's world of astounding technological feats, will such a truly perverse printed exposure stimulate adventurous young gynecologic surgeons who think they are much better than they really are into imprudent undertakings beyond their true capabilities, leading to serious injury to their patients? Goodness knows what difficulties we already find in our cluttered residency programs in getting basic maneuvers (like vaginal hysterectomy) across, let alone highly sophisticated, industry-driven, potentially dangerous operative challenges performed through a spyglass.

If there are critics abroad who think I am wrong, let them please tell me.

Rather than repudiating Dr. Grody's opinion about laparoscopic surgery, I will only respond to his point about the importance of the perineal membrane and PB to pelvic organ support.

All defects should be repaired at the time of pelvic floor reconstructive surgery. Any tear or defect in the area of the perineal membrane or PB should be repaired concurrently with pelvic floor reconstruction. This point was emphasized in the final step outlined in Part 2 of my series: “Repair the rectocele and perform perineorrhaphy vaginally if necessary.”

Based upon my understanding of the functional pelvic support anatomy as well as clinical observation, I maintain my position that “the perineal membrane and perineal body are not very crucial for pelvic organ support.”

The perineal membrane is a single layer of fibromuscular tissue that spans the anterior triangle of the pelvic outlet. Laterally, it attaches to the ischiopubic ramus; medially, it fuses with the sidewalls of the vagina and perineal body. The anterior portion of the perineal membrane is fused with the muscles of the distal urethra. Rather than forming a supportive sheet as it does in the male, the perineal membrane in the female—because of the large opening of the vagina—provides only lateral attachment for the PB and some support for the lower urethra.

The PB is an ill-defined, bordered mass of dense connective tissue lying between the vagina and anus. Fused anteriorly to the posterior vaginal wall and attached laterally to the perineal membrane and bulbocavernosus and superficial transverse perineal muscles, a significant portion of what is clinically called the perineal body is actually the muscle of the external anal sphincter. The strong upward traction of the levator ani muscles is much more important in maintaining vaginal outlet support than are the bulbocavernosus and superficial transverse perineal muscle.

Contrary to Dr. Grody's assertion that the PB makes a substantial contribution to pelvic support, in actuality the support is minimal. Rather, restoration of the PB is important for sexual function and anal/fecal continence. I have examined several patients with no PB as the result of chronic unrepaired fourth-degree obstetric lacerations, yet none of them had prolapse. Similarly, women who have had a radical resection of the anus and rectum for cancer, including the entire removal of the PB, suffered no significant prolapse.

Because considerable descent (up to 1 inch) of the PB is possible during voluntary straining, the perineal membrane and PB cannot be the main supportive layer of the genital outlet. The fact that the PB can move backward 3-4 cm toward the sacrum when a weighted speculum is placed in the posterior vagina likewise indicates that the position of the PB is determined by the levator ani muscle rather than by any inherent importance of its own.

Advances in technology afford greater magnification, visualization, and accuracy—leading to a level of surgical precision heretofore impossible with the relatively “blind” vaginal approach. We must train young surgeons for these state-of-the-art advances.

Although I have the utmost respect for both Dr. Grody and Dr. Liu and I believe that everyone is entitled to his or her own opinion, Dr Liu's article is certainly not worthy of such admonishment.

Dr. Liu not only correctly addresses normal vaginal anatomy, clinical assessment, and one surgeon's approach to the anatomical correction of symptomatic prolapse, he does so in a concise, informative manner.

Dr. Grody's belief that the perineal membrane and PB are crucial for pelvic organ support is indeed just that: his belief. Using the PubMed search term “perineal body surgery,” I found no scientific literature written in the past 40 years that supports the concept that either the perineal membrane or the PB is crucial in the support of any organs of the pelvis. I have yet to read or find an article that suggests that the cure rates of sacrospinous ligament suspension; sacral colpopexy; paravaginal repair; uterosacral ligament suspension; enterocele repair; or Burch, sling, or any other prolapse corrective surgery—including colpocleisis or Lefort procedures—are improved by repairing the PB.

Furthermore, there is no scientific literature that supports the concept that poor perineal support increases the incidence of prolapse. If this were a fact, patients with traumatic or congenital cloaca would also suffer a greater incidence of vaginal prolapse. I have not seen or read of any scientific literature or text that can directly show a cause-and-effect relationship between a damaged PB and vaginal prolapse.

Dr. Grody is a purist in his pursuit of vaginal anatomic correction, but this fine trait does not constitute scientific proof for his allegation. He has the right to theorize that the anatomical correction is essential to improve long-term cure rates of prolapse surgery. But a theory is belief unsupported by substantial fact, and will thus remain just a theory.

After reading Dr. Liu's article on laparoscopic pelvic reconstructive surgery and Dr. Grody's response, I found myself perplexed. How is it that two experienced and respected surgeons can underappreciate each other's perspective on pelvic reconstructive surgery?

For the most part, I agree with most of what each has stated but disagree on the finer points. I must confess that being predominately a laparoscopic or minimally invasive surgeon, I too did not completely comprehend the complexity and functional anatomy of the PB and membrane as an important element in pelvic floor support until more recently.

Thanks to cadaver sections and MRI studies reported by John O.L. Delancey, M.D., at the joint annual meeting of the American Urogynecologic Society and the Society of Gynecologic Surgeons in 2004, we realize that the perineal membrane is a complex 3-D structure composed of a dorsal and ventral portion rather than a trilaminar sheet as previously thought. His description of the anatomical relationship to the compressor urethra, urethra vaginal sphincter, arcus tendineus, pubic bone, and levator muscles underscores the importance of this structure in pelvic support.

We now have level 1 evidence of the laparoscope's benefit in sacral colpopexies compared with an open procedure, as well as its inferiority in treating stress urinary incontinence when comparing a laparoscopic Burch with a transvaginal tape procedure. But the laparoscope is a tool that requires proper training to master. Thanks to the pioneering efforts of Dr. Liu and Dr. Miklos, the development of training centers, and the support of organizations like the American Association of Gynecologic Laparoscopists, it is no longer the gifted few who use this valuable instrument.

The recent articles by Dr. C.Y. Liu in OB.GYN.NEWS on laparoscopic pelvic reconstructive surgery, Parts 1 and 2, are a must-read for any gynecologic surgeon performing reconstructive vaginal surgery. Although the article presents Dr. Liu's laparoscopic approach to problems of vaginal suspension and support, the anatomy presented and the surgical steps discussed are clearly applicable to the repair of any vaginal prolapse via any surgical approach, including vaginal and abdominal.

The anatomy of genital prolapse is up to date, well written, and clearly explained. Part 1 contains many pearls of insight from a master of this anatomy, and it summarizes our current concepts of vaginal suspension and support. The section on clinical assessment of prolapse is practical and very helpful.

The surgical techniques presented are anatomical and readily applicable. Dr. Liu explains how to safely dissect out and investigate the suspensory anatomy to clearly define the anatomical defects that caused the vaginal prolapse. Not only does Dr. Liu address and repair the specific breaks in the continuity of the visceral connective tissue suspensory network, but he presents an excellent dissection technique for safeguarding the ureters.

One point that should have been emphasized is the requirement for cystoscopic confirmation of bilateral ureteral functioning at the end of the case.

The article explains that one of the three supporting layers of the female pelvic organs is “the perineal membrane/external anal sphincter.” What is not said is that the anal sphincter is an important component of the posterior part of the PB.

The lower third of the vagina and the anal canal/anal sphincter are fused with the PB. The PB is shaped roughly like a pyramid, with the base between the vaginal introitus and the anal sphincter. The apex is found at the junction of the lower third and the middle third of the vagina, and at the rectoanal junction. At the apex of the perineal body, the vagina slopes to a more horizontal orientation in the standing patient, whereas the anal canal forms a right angle with the lower rectum.

Portions of the pubococcygeus and puborectalis muscles insert into the apex of the PB. The rectovaginal fascia also inserts into the apex of the PB and helps in its proper anatomical orientation. The intact PB positions itself and the anus just above the level of the ischial tuberosities. The fusion of the anus and anal canal with the PB is important for their anatomical positioning and physiologic functioning in fecal continence. The fusion of the lower third of the vagina with the PB is important for its anatomical positioning and physiologic functioning in pelvic organ support. The PB assists in closing off the genital hiatus at times of increased intrapelvic pressures, supporting the pelvic organs. Another support mechanism is the flap-valve action of the levator plate.

Many women with vaginal prolapse demonstrate abnormal descent of the perineum. Dr. Liu states, “The active support of the pelvic floor comes from the levator ani muscles (the iliococcygeus and pubococcygeus muscles). These muscles close off the pelvic floor so the pelvic organs can rest upon them without tension.” This statement is true.

Dr. Liu does not mention the important action of the levator plate or the action of the PB in this vaginal support mechanism. In fact, with a poorly supportive levator plate, as is frequently seen in vaginal prolapse patients, a well-reconstructed PB will substitute as a backstop against which the resuspended vagina can be compressed for support.

The reconstructed PB will help close off the genital hiatus at times of mechanical pelvic stress. The PB must be reconstructed in shape and bulk to support and orient the anal canal and lower third of the vagina, but also to position itself and the anal canal at or above the level of the ischial tuberosities.

As Dr. Liu implies, we cannot repair or completely rehabilitate damaged and weakened pelvic floor muscles and their innervations. We should surgically reconstruct a disrupted PB. I do feel that Dr. Liu does indeed perform perineoplasty on many of his prolapse patients. He simply emphasized the reconstruction and proper placement of the pericervical ring in his excellent article.

MARVIN H. TERRY GRODY, M.D., is a professor of obstetrics and gynecology and senior attending gynecology consultant, Robert Wood Johnson Medical School at Camden (N.J.).

C.Y. LIU, M.D., is the director of the Manhattan Women's Laser Center, New York.

JOHN R. MIKLOS, M.D., is the director of the Atlanta Center for Laparoscopic Urogynecology.

VINCENT R. LUCENTE, M.D., is chief of urogynecology at Abington (Pa.) Memorial Hospital and associate professor of ob.gyn. at Pennsylvania State Medical Center, Hershey.

ROBERT M. ROGERS JR., M.D., is an attending gynecologist at the Reading Hospital and Medical Center in West Reading, Pa.

Up to 70% of children who have been diagnosed with attention-deficit hyperactivity disorder continue to experience symptoms into adulthood, and as a result, as many as 8 million adults are affected. That figure may come as a surprise to clinicians who still think of ADHD as a childhood disorder.

Studies suggest that only 20% of adults with ADHD are diagnosed and treated for the disorder. And there are a number of possible reasons for this treatment gap. First, the adult form of the illness was not recognized as a diagnostic entity until the late 1980s. Second, patients who were never diagnosed may have accepted their symptoms as personality flaws or developed ways to compensate. Finally, others are regulars in the psychiatrist's office and are being treated unsuccessfully for comorbid conditions such as depression or anxiety.

Diagnosis

Most adults diagnosed with ADHD are self-referred. One of the most frequent stimuli for seeking help is that their own child has been diagnosed with the disorder. A combination of hyperactivity and inattention forms the most common subtype of the disorder; 70% of adult patients manifest that combination. A largely inattentive subtype occurs in 25%, and about 5% show the hyperactive/impulsive subtype.

Adults with ADHD are likely to display poor tolerance for frustration, temper outbursts, lack of social judgment, inability to organize daily tasks, lack of motivation, procrastination, risk-taking, and low stimulation. They are poor listeners, tending to interrupt. They frequently misplace items or forget appointments.

Adults with ADHD are highly likely to have comorbid depression, anxiety, or bipolar disorder. If untreated, they also have an increased risk of substance abuse, which many researchers believe is an attempt to self-medicate.

Several screening tools are available to assess the likelihood of ADHD. One of these includes a six-question patient self-report that can be supplemented with a longer physician screen. The screening tool identifies individuals at risk for ADHD so that they can be evaluated by their doctors. It's important to use a screen designed for adults, with questions that pinpoint very specific areas of impairment an adult is likely to encounter. One such screen and a self-assessment tool are available at

www.med.nyu.edu/psych/training/adhd.html

Management

Adults with ADHD respond readily to the same medications used to treat childhood ADHD. Stimulants (methylphenidate and amphetamine products) are very highly studied in children, producing a robust, prompt response and a significant decrease in symptoms. But until recently, amphetamines had not been well studied in adults.

As a result, most adults were significantly underdosed. Additionally, the short-term action of the drugs made them a less-than-ideal choice.

In 2004, the Food and Drug Administration approved a mixed salts preparation of a single-entity amphetamine (Adderall XR) as a once-daily treatment for adults with ADHD. With the approval of this extended-release formulation, stimulant therapy became more appropriate for adults.

Atomoxetine (Strattera) is the first nonstimulant therapy for adult ADHD. A selective norepinephrine reuptake inhibitor, atomoxetine relieves symptoms for as long as 12 hours and does not exacerbate comorbid tics or mood disorders. The drug is not a controlled substance, as are amphetamines. This is an advantage when treating a patient with a history of serious drug abuse. Some improvement may be noted in 5-7 days, but the full effect usually takes about 2 weeks to appear. The FDA recently stated that patients should discontinue the drug if they develop jaundice or have laboratory evidence of liver injury. Atomoxetine has been linked to two cases of severe liver injury.

Both classes of drug should be started at a low dosage and titrated upward until the patient achieves good symptom relief with minimum side effects. Most patients benefit from some form of adjunctive psychosocial therapy.

Medication compliance can be challenging in patients who have trouble remembering. Patients piggyback their medication onto another daily task, preferably one that is habitual. Segmented medication organizers can be helpful, especially if a family member or roommate can ensure the doses are taken at the correct times.

Sources: LENARD A. ADLER, M.D., head of the adult ADHD program at New York University, New York; RICHARD H. WEISLER, M.D., department of psychiatry, University of North Carolina, Chapel Hill, and Duke University, Durham, N.C.

Up to 70% of children who have been diagnosed with attention-deficit hyperactivity disorder continue to experience symptoms into adulthood, and as a result, as many as 8 million adults are affected. That figure may come as a surprise to clinicians who still think of ADHD as a childhood disorder.

Studies suggest that only 20% of adults with ADHD are diagnosed and treated for the disorder. And there are a number of possible reasons for this treatment gap. First, the adult form of the illness was not recognized as a diagnostic entity until the late 1980s. Second, patients who were never diagnosed may have accepted their symptoms as personality flaws or developed ways to compensate. Finally, others are regulars in the psychiatrist's office and are being treated unsuccessfully for comorbid conditions such as depression or anxiety.

Diagnosis

Most adults diagnosed with ADHD are self-referred. One of the most frequent stimuli for seeking help is that their own child has been diagnosed with the disorder. A combination of hyperactivity and inattention forms the most common subtype of the disorder; 70% of adult patients manifest that combination. A largely inattentive subtype occurs in 25%, and about 5% show the hyperactive/impulsive subtype.

Adults with ADHD are likely to display poor tolerance for frustration, temper outbursts, lack of social judgment, inability to organize daily tasks, lack of motivation, procrastination, risk-taking, and low stimulation. They are poor listeners, tending to interrupt. They frequently misplace items or forget appointments.

Adults with ADHD are highly likely to have comorbid depression, anxiety, or bipolar disorder. If untreated, they also have an increased risk of substance abuse, which many researchers believe is an attempt to self-medicate.

Several screening tools are available to assess the likelihood of ADHD. One of these includes a six-question patient self-report that can be supplemented with a longer physician screen. The screening tool identifies individuals at risk for ADHD so that they can be evaluated by their doctors. It's important to use a screen designed for adults, with questions that pinpoint very specific areas of impairment an adult is likely to encounter. One such screen and a self-assessment tool are available at

www.med.nyu.edu/psych/training/adhd.html

Management

Adults with ADHD respond readily to the same medications used to treat childhood ADHD. Stimulants (methylphenidate and amphetamine products) are very highly studied in children, producing a robust, prompt response and a significant decrease in symptoms. But until recently, amphetamines had not been well studied in adults.

As a result, most adults were significantly underdosed. Additionally, the short-term action of the drugs made them a less-than-ideal choice.

In 2004, the Food and Drug Administration approved a mixed salts preparation of a single-entity amphetamine (Adderall XR) as a once-daily treatment for adults with ADHD. With the approval of this extended-release formulation, stimulant therapy became more appropriate for adults.

Atomoxetine (Strattera) is the first nonstimulant therapy for adult ADHD. A selective norepinephrine reuptake inhibitor, atomoxetine relieves symptoms for as long as 12 hours and does not exacerbate comorbid tics or mood disorders. The drug is not a controlled substance, as are amphetamines. This is an advantage when treating a patient with a history of serious drug abuse. Some improvement may be noted in 5-7 days, but the full effect usually takes about 2 weeks to appear. The FDA recently stated that patients should discontinue the drug if they develop jaundice or have laboratory evidence of liver injury. Atomoxetine has been linked to two cases of severe liver injury.

Both classes of drug should be started at a low dosage and titrated upward until the patient achieves good symptom relief with minimum side effects. Most patients benefit from some form of adjunctive psychosocial therapy.

Medication compliance can be challenging in patients who have trouble remembering. Patients piggyback their medication onto another daily task, preferably one that is habitual. Segmented medication organizers can be helpful, especially if a family member or roommate can ensure the doses are taken at the correct times.

Sources: LENARD A. ADLER, M.D., head of the adult ADHD program at New York University, New York; RICHARD H. WEISLER, M.D., department of psychiatry, University of North Carolina, Chapel Hill, and Duke University, Durham, N.C.

Up to 70% of children who have been diagnosed with attention-deficit hyperactivity disorder continue to experience symptoms into adulthood, and as a result, as many as 8 million adults are affected. That figure may come as a surprise to clinicians who still think of ADHD as a childhood disorder.

Studies suggest that only 20% of adults with ADHD are diagnosed and treated for the disorder. And there are a number of possible reasons for this treatment gap. First, the adult form of the illness was not recognized as a diagnostic entity until the late 1980s. Second, patients who were never diagnosed may have accepted their symptoms as personality flaws or developed ways to compensate. Finally, others are regulars in the psychiatrist's office and are being treated unsuccessfully for comorbid conditions such as depression or anxiety.

Diagnosis

Most adults diagnosed with ADHD are self-referred. One of the most frequent stimuli for seeking help is that their own child has been diagnosed with the disorder. A combination of hyperactivity and inattention forms the most common subtype of the disorder; 70% of adult patients manifest that combination. A largely inattentive subtype occurs in 25%, and about 5% show the hyperactive/impulsive subtype.

Adults with ADHD are likely to display poor tolerance for frustration, temper outbursts, lack of social judgment, inability to organize daily tasks, lack of motivation, procrastination, risk-taking, and low stimulation. They are poor listeners, tending to interrupt. They frequently misplace items or forget appointments.

Adults with ADHD are highly likely to have comorbid depression, anxiety, or bipolar disorder. If untreated, they also have an increased risk of substance abuse, which many researchers believe is an attempt to self-medicate.

Several screening tools are available to assess the likelihood of ADHD. One of these includes a six-question patient self-report that can be supplemented with a longer physician screen. The screening tool identifies individuals at risk for ADHD so that they can be evaluated by their doctors. It's important to use a screen designed for adults, with questions that pinpoint very specific areas of impairment an adult is likely to encounter. One such screen and a self-assessment tool are available at

www.med.nyu.edu/psych/training/adhd.html

Management

Adults with ADHD respond readily to the same medications used to treat childhood ADHD. Stimulants (methylphenidate and amphetamine products) are very highly studied in children, producing a robust, prompt response and a significant decrease in symptoms. But until recently, amphetamines had not been well studied in adults.

As a result, most adults were significantly underdosed. Additionally, the short-term action of the drugs made them a less-than-ideal choice.

In 2004, the Food and Drug Administration approved a mixed salts preparation of a single-entity amphetamine (Adderall XR) as a once-daily treatment for adults with ADHD. With the approval of this extended-release formulation, stimulant therapy became more appropriate for adults.

Atomoxetine (Strattera) is the first nonstimulant therapy for adult ADHD. A selective norepinephrine reuptake inhibitor, atomoxetine relieves symptoms for as long as 12 hours and does not exacerbate comorbid tics or mood disorders. The drug is not a controlled substance, as are amphetamines. This is an advantage when treating a patient with a history of serious drug abuse. Some improvement may be noted in 5-7 days, but the full effect usually takes about 2 weeks to appear. The FDA recently stated that patients should discontinue the drug if they develop jaundice or have laboratory evidence of liver injury. Atomoxetine has been linked to two cases of severe liver injury.

Both classes of drug should be started at a low dosage and titrated upward until the patient achieves good symptom relief with minimum side effects. Most patients benefit from some form of adjunctive psychosocial therapy.

Medication compliance can be challenging in patients who have trouble remembering. Patients piggyback their medication onto another daily task, preferably one that is habitual. Segmented medication organizers can be helpful, especially if a family member or roommate can ensure the doses are taken at the correct times.

Sources: LENARD A. ADLER, M.D., head of the adult ADHD program at New York University, New York; RICHARD H. WEISLER, M.D., department of psychiatry, University of North Carolina, Chapel Hill, and Duke University, Durham, N.C.

Lumbar spinal stenosis occurs in the lower back, where most back movement takes place. The condition develops when osteoarthritis causes a narrowing of the opening in the vertebrae through which the spinal nerves pass. In some advanced cases, bones press on the spinal nerves when the body is in a certain position or positions, which causes a sharp pain.

Symptoms of this condition are similar to those of disk disease; patients may complain of numbness in the legs and pain in the lower spine. Older patients may report shooting pains in the legs when they first wake up in the morning.

Patients with lumbar stenosis may report that walking seems more difficult although they have no apparent pain in their knees. They also may find that their calves hurt after walking but that the calf pain goes away when they sit down. These symptoms are due to a combination of body position and tight muscles that further close off the already narrowed vertebral openings. When the patient sits down, the openings become slightly enlarged, which may reduce the pain temporarily.

Lumbar spinal stenosis is often mistakenly diagnosed as sciatica, peripheral vascular disease, or simply the aches and pains of old age. It tends to develop with age and is more common among people who have worked in a physically demanding job, such as construction, for most of their adult lives. The condition will become more common among the population in general as people live longer.

One way to diagnose lumbar stenosis is a treadmill test, but not the kind you would use for a heart examination. Have the patient walk on a level surface, and then change it to an incline. While walking uphill, patients with spinal stenosis can experience pain relief, but patients with peripheral vascular disease tend to have more severe pain.

In this month's column, we'll look at some exercises to help relieve the discomfort of lumbar spinal stenosis. (See illustrations and instructions for patients below.)

Surgery, although feasible, is difficult and not always successful. Loose hips are important protectors against pain. In addition, strong oblique stomach muscles and strong legs can make patients more comfortable.

Discourage patients with lumbar spinal stenosis from running or walking downhill; this angles the spine in a way that aggravates their pain. Walking on a treadmill on a slight upward incline is a safe activity, however.

Next month: Exercises for patients with type 2 diabetes.

Lumbar Stenosis

Piriformis muscle stretch. The piriformis is a muscle in the hip that often tightens in patients with spinal stenosis. To stretch it, lie on your back with both arms extended at shoulder height. Bring your right foot over your left leg, turning the right knee to the left. Try to keep your trunk and shoulders flat on the floor. Hold for 6 seconds, then return to starting position. Repeat 6-8 times on each side.

Supine paraspinal stretch. Lie on your back on a carpeted floor with legs extended in front of you. Exhale slowly as you lift your head and shoulders and grasp your knees while gently bringing your knees to your chest. Relax. Repeat 6-8 times.

Abdominal strengthening. Lie on your back on a carpeted floor with knees bent and feet flat on the floor. Tuck your chin to your chest, extend your arms, and reach each arm and shoulder across the opposite knee, creating a slight twist. Hold for 6 seconds, then relax. Repeat 6-8 times each side.

Low back extensor. Sit in a chair, with feet flat on the floor and knees shoulder-width apart. Bend forward, and roll the left shoulder toward the right knee, reaching your arms toward your right foot and bringing your head toward the outside of the right knee. Hold for 6 seconds. Contract your abdominal and gluteal muscles as you rise to a seated position. Repeat 4-6 times. Perform on the other side, reaching toward the left knee with the right shoulder. Breathe out as you bend forward. EMILY BRENNAN, ILLUSTRATIONS

Lumbar spinal stenosis occurs in the lower back, where most back movement takes place. The condition develops when osteoarthritis causes a narrowing of the opening in the vertebrae through which the spinal nerves pass. In some advanced cases, bones press on the spinal nerves when the body is in a certain position or positions, which causes a sharp pain.

Symptoms of this condition are similar to those of disk disease; patients may complain of numbness in the legs and pain in the lower spine. Older patients may report shooting pains in the legs when they first wake up in the morning.

Patients with lumbar stenosis may report that walking seems more difficult although they have no apparent pain in their knees. They also may find that their calves hurt after walking but that the calf pain goes away when they sit down. These symptoms are due to a combination of body position and tight muscles that further close off the already narrowed vertebral openings. When the patient sits down, the openings become slightly enlarged, which may reduce the pain temporarily.

Lumbar spinal stenosis is often mistakenly diagnosed as sciatica, peripheral vascular disease, or simply the aches and pains of old age. It tends to develop with age and is more common among people who have worked in a physically demanding job, such as construction, for most of their adult lives. The condition will become more common among the population in general as people live longer.

One way to diagnose lumbar stenosis is a treadmill test, but not the kind you would use for a heart examination. Have the patient walk on a level surface, and then change it to an incline. While walking uphill, patients with spinal stenosis can experience pain relief, but patients with peripheral vascular disease tend to have more severe pain.

In this month's column, we'll look at some exercises to help relieve the discomfort of lumbar spinal stenosis. (See illustrations and instructions for patients below.)

Surgery, although feasible, is difficult and not always successful. Loose hips are important protectors against pain. In addition, strong oblique stomach muscles and strong legs can make patients more comfortable.

Discourage patients with lumbar spinal stenosis from running or walking downhill; this angles the spine in a way that aggravates their pain. Walking on a treadmill on a slight upward incline is a safe activity, however.

Next month: Exercises for patients with type 2 diabetes.

Lumbar Stenosis

Piriformis muscle stretch. The piriformis is a muscle in the hip that often tightens in patients with spinal stenosis. To stretch it, lie on your back with both arms extended at shoulder height. Bring your right foot over your left leg, turning the right knee to the left. Try to keep your trunk and shoulders flat on the floor. Hold for 6 seconds, then return to starting position. Repeat 6-8 times on each side.

Supine paraspinal stretch. Lie on your back on a carpeted floor with legs extended in front of you. Exhale slowly as you lift your head and shoulders and grasp your knees while gently bringing your knees to your chest. Relax. Repeat 6-8 times.

Abdominal strengthening. Lie on your back on a carpeted floor with knees bent and feet flat on the floor. Tuck your chin to your chest, extend your arms, and reach each arm and shoulder across the opposite knee, creating a slight twist. Hold for 6 seconds, then relax. Repeat 6-8 times each side.

Low back extensor. Sit in a chair, with feet flat on the floor and knees shoulder-width apart. Bend forward, and roll the left shoulder toward the right knee, reaching your arms toward your right foot and bringing your head toward the outside of the right knee. Hold for 6 seconds. Contract your abdominal and gluteal muscles as you rise to a seated position. Repeat 4-6 times. Perform on the other side, reaching toward the left knee with the right shoulder. Breathe out as you bend forward. EMILY BRENNAN, ILLUSTRATIONS

Lumbar spinal stenosis occurs in the lower back, where most back movement takes place. The condition develops when osteoarthritis causes a narrowing of the opening in the vertebrae through which the spinal nerves pass. In some advanced cases, bones press on the spinal nerves when the body is in a certain position or positions, which causes a sharp pain.

Symptoms of this condition are similar to those of disk disease; patients may complain of numbness in the legs and pain in the lower spine. Older patients may report shooting pains in the legs when they first wake up in the morning.

Patients with lumbar stenosis may report that walking seems more difficult although they have no apparent pain in their knees. They also may find that their calves hurt after walking but that the calf pain goes away when they sit down. These symptoms are due to a combination of body position and tight muscles that further close off the already narrowed vertebral openings. When the patient sits down, the openings become slightly enlarged, which may reduce the pain temporarily.

Lumbar spinal stenosis is often mistakenly diagnosed as sciatica, peripheral vascular disease, or simply the aches and pains of old age. It tends to develop with age and is more common among people who have worked in a physically demanding job, such as construction, for most of their adult lives. The condition will become more common among the population in general as people live longer.

One way to diagnose lumbar stenosis is a treadmill test, but not the kind you would use for a heart examination. Have the patient walk on a level surface, and then change it to an incline. While walking uphill, patients with spinal stenosis can experience pain relief, but patients with peripheral vascular disease tend to have more severe pain.

In this month's column, we'll look at some exercises to help relieve the discomfort of lumbar spinal stenosis. (See illustrations and instructions for patients below.)

Surgery, although feasible, is difficult and not always successful. Loose hips are important protectors against pain. In addition, strong oblique stomach muscles and strong legs can make patients more comfortable.

Discourage patients with lumbar spinal stenosis from running or walking downhill; this angles the spine in a way that aggravates their pain. Walking on a treadmill on a slight upward incline is a safe activity, however.

Next month: Exercises for patients with type 2 diabetes.

Lumbar Stenosis

Piriformis muscle stretch. The piriformis is a muscle in the hip that often tightens in patients with spinal stenosis. To stretch it, lie on your back with both arms extended at shoulder height. Bring your right foot over your left leg, turning the right knee to the left. Try to keep your trunk and shoulders flat on the floor. Hold for 6 seconds, then return to starting position. Repeat 6-8 times on each side.

Supine paraspinal stretch. Lie on your back on a carpeted floor with legs extended in front of you. Exhale slowly as you lift your head and shoulders and grasp your knees while gently bringing your knees to your chest. Relax. Repeat 6-8 times.

Abdominal strengthening. Lie on your back on a carpeted floor with knees bent and feet flat on the floor. Tuck your chin to your chest, extend your arms, and reach each arm and shoulder across the opposite knee, creating a slight twist. Hold for 6 seconds, then relax. Repeat 6-8 times each side.

Low back extensor. Sit in a chair, with feet flat on the floor and knees shoulder-width apart. Bend forward, and roll the left shoulder toward the right knee, reaching your arms toward your right foot and bringing your head toward the outside of the right knee. Hold for 6 seconds. Contract your abdominal and gluteal muscles as you rise to a seated position. Repeat 4-6 times. Perform on the other side, reaching toward the left knee with the right shoulder. Breathe out as you bend forward. EMILY BRENNAN, ILLUSTRATIONS

Preterm labor is one of the most important and vexing challenges complicating pregnancy. Premature babies account for an estimated 6%–10% of births, yet they account for 70%–85% of neonatal morbidity and mortality.

In multiple gestations, which are increasingly common as a result of delayed childbearing and the use of assisted reproductive technologies, preterm labor is an even greater risk. The literature points to an incidence of preterm labor of 20%–75% in multiple gestations. Although these figures may be somewhat high, I think it is safe to say that at least 1 in 10 multiple gestations seen at my institution are complicated by preterm labor. Not all of these patients will be admitted or will deliver early, but the very common nature of this problem and the potentially lethal consequences of premature multiple deliveries make this an issue that every physician and institution should approach carefully.

First, it is important to consider the delivery goal of a multiple gestation pregnancy. Overall, most twins are delivered at 37–38 weeks. For triplets, the gestational age is closer to 34 weeks, and quadruplets are born at around 30 weeks. These are reasonable numbers applicable to community practice.

If a patient arrives in preterm labor, you have to decide what to do, considering her situation and the capabilities of your local hospital and medical staff. It is clear that premature babies fare best when they are cared for in the institution where they are born. If someone needs to be transferred, it should be the mother, not the baby.

Obviously, if a patient carrying twins presents in labor at 35 or 36 weeks, most obstetricians would be inclined to do very little to cut short the labor, because—in the absence of other complications—these babies are likely to do well. However, if she presents at 29 weeks, it would make sense to be more aggressive.

Can the patient be safely and aggressively managed for preterm labor in her local community? The answer hinges on the plan for delivery if the treatment fails. Each hospital and service has to pick a gestational age at which neonatal survival is acceptably high. Then, options for the mother should be discussed with her and with the neonatal intensive care unit at your hospital or the institution to which she will be transferred.

Depending on the circumstances, the treatment of preterm labor may be undertaken for several reasons:

▸ To delay delivery until the patient can be transferred to a tertiary medical center with a high-level neonatal intensive care unit.

▸ To delay delivery 24–48 hours for the administration of corticosteroid therapy.

▸ To reduce the strength and frequency of uterine contractions, enabling the fetus to further develop in the uterus.

▸ To minimize hospital stays for the mother and the neonate.

▸ To reduce the risk of neonatal morbidity and mortality by preventing preterm delivery, the most dangerous complication of multiple gestation pregnancies.

When I consult with a woman in preterm labor, I go through a list of available options. Unfortunately, a careful review of the literature reveals few really good, effective treatments.

Although new ideas emerge every few years, not many interventional strategies have withstood attempts to corroborate results from single institutions. It may be tempting to “just do something,” but we owe it to our patients to stick to scientifically valid and efficacious treatments.

Bed Rest

Bed rest or activity restriction will not prevent preterm labor. Rest neither lengthens gestation nor reduces neonatal morbidity in multiple gestation pregnancies. In some studies these patients did worse.

If a patient carrying multiples has a short cervix and threatened preterm labor, there is some evidence to support getting her off her feet rather than having her continue working at a very active job.

Once a multiple pregnancy is complicated by preterm labor, hospitalization may be necessary for observation and implementation of a treatment course.

Hydration

There is no evidence that hydration is an effective treatment for preterm labor. In fact, the initial administration of bolus intravenous fluids may pose some risk to patients with multiple gestations. These patients already have an increased blood volume and could develop pulmonary edema from fluid overload if tocolytic therapy is initiated after unnecessary fluids are administered.

Progesterone

A study by Paul J. Meis, M.D., and colleagues (N. Engl. J. Med. 2003;348:2379–85) suggests recurrent preterm birth in singleton pregnancies can be prevented by 17 δ-hydroxyprogesterone caproate. The jury is still out on whether progesterone can be useful in managing active or threatened preterm labor in a multiple gestation pregnancy.

Studies are underway that may provide us with more guidance in the use of this agent. However, no evidence exists that it is safe and efficacious in multiple gestation pregnancies, so I suggest its use be reserved for patients in clinical trials.

Antibiotics

It is tantalizing to believe antibiotics would be helpful in preventing or treating preterm labor. Many researchers theorize that intrauterine infection or fetal infection may be responsible for preterm labor, particularly in pregnancies that are not complicated by multiple fetuses. However, the data do not show that antibiotic treatment is any more efficacious than placebo in prolonging pregnancy or preventing preterm delivery.

Tocolytics

The use of tocolytics to decrease or halt preterm labor is controversial in multiple gestations as well as in singleton pregnancies because the drugs pose risks to the mother and, in some cases, to the fetus. However, the available data support the position that tocolytic agents work for a short period—about 48 hours, although Roger B. Newman, M.D., and colleagues have shown that some multiple gestations can be prolonged for more than 7 days (Am. J. Obstet. Gynecol. 1989;161:547–55; “Multifetal Pregnancy: A Handbook for Care of the Pregnant Patient” [Philadelphia: Lippincott Williams & Wilkins, 2000]).

Each tocolytic agent carries its own benefits, risks, contraindications, and adverse effects profile. Numerous sources are available for this information; for quick reference; I recently published a summary in chart form (Clin. Obstet. Gynecol. 2004;47:216–26). Keep in mind that women with multiple gestations have an elevated risk of cardiovascular complications, such as pulmonary edema resulting from anemia, lower colloid oncotic pressure, and higher blood volume.

I would take a middle-of-the-road approach in choosing an agent or agents for tocolysis. For example, oral terbutaline, oral calcium channel blockers, and oral Indocin have been well-studied and widely used, with varying levels of success.

John P. Elliott, M.D., and Tari Radin, Ph.D., studied a small number of high-order multiple gestations and found similar levels of serum magnesium in triplets and quadruplets and in singleton pregnancies after the administration of magnesium sulfate for tocolysis (J. Reprod. Med. 1995;40:450–2). However, they concluded that higher levels of magnesium sulfate are needed in multiple gestations to inhibit labor. They suggested administration at infusion rates of 4–5 g/h in triplet and quadruplet pregnancies.

Research on combination tocolytic therapy has produced conflicting results. Most of the studies on this topic are dated, limited in scope, and not specifically focused on multiple gestations. Some concerns, however, have been raised. For example, magnesium sulfate administered in conjunction with nifedipine can result in significant neuromuscular blockage and a subsequent marked hypotensive effect.

My recommendation is to administer combination tocolytic therapy only with great caution, using agents such as intravenous magnesium sulfate with oral terbutaline or indomethacin. The mother and fetus also should be monitored closely by professionals well-versed in side effects linked to this form of therapy.

Should tocolytic therapy be maintained after successful cessation of labor? A careful reading of the available evidence suggests the answer is no. In well-designed studies, maintenance tocolytic therapy has reduced neither preterm deliveries nor perinatal morbidity or mortality.

Corticosteroids

Corticosteroids are among the few noncontroversial agents for use during preterm labor. These agents—which clearly reduce the incidence and severity of neonatal respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, and neonatal mortality—should be administered to any patient in preterm labor at 24–34 weeks, using the same regimen as in singleton gestations.

The accepted corticosteroid treatment consists of two intramuscular doses of betamethasone or four intramuscular doses of dexamethasone.

In higher-order gestations, Dr. Elliott and Dr. Radin have observed that beta-methasone can increase uterine contractions and advance labor in patients with frequent contractions. They therefore recommend that corticosteroids be reserved in these pregnancies for patients having fewer than three contractions per hour (Obstet. Gynecol. 1995;85:250–54). Uterine activity in higher-order gestations should be closely monitored following the administration of corticosteroids.

It once was suggested that multiple courses of prenatal corticosteroids might be of benefit. This recommendation is no longer made, because there is no good evidence for enhanced efficacy and because repeated doses are associated with clear risks, including decreased fetal growth and reduced birth weight.

Adjunctive Treatments

Isolated studies have suggested a role in preterm labor for such interventions as vaginal lever pessaries, blood transfusions, and cerclage. None of these have been put to rigorous scientific scrutiny. Cerclage obviously has no role in patients who do not have an incompetent cervix, or in such specific cases as the previable delivery of one dichorionic twin when a delayed-interval delivery of the second twin is desired.

Treating the Whole Patient

Not infrequently, preterm labor in a multiple gestation pregnancy requires the prolonged hospitalization of the mother, who may be a busy professional woman or the parent of other small children. Spending 5–10 weeks in a hospital is a long, frustrating time, replete with boredom and anxiety.

At my institution, a clinical nurse specialist who was concerned with the psychosocial effect of hospitalization on the women in our unit proposed what has become a highly successful multidimensional program: Mom Matters. The program offers support, diversion, and empathy in the form of wheelchair outings, Internet access, flexible visitation for family members (including children), manicures, pedicures, movies, and get-togethers.

The success of our program and the gratitude expressed by the mothers on our service have convinced me that this is a kind and therapeutic approach worth considering and implementing elsewhere.

The Future

The road to safe and effective therapy for preterm labor has been a long and frustrating one. I am hopeful that researchers are now on the right track, focusing on subtle indications of infection and other potential causes of preterm labor.

In multiple gestations, of course, preterm labor is often caused simply because there are too many fetuses in the womb. Uterine overdistension will not be an easy problem to overcome, except by reducing multiple gestations. Many of our colleagues in reproductive endocrinology have gotten the message that implanting too many embryos is unwise and unethical.

I believe that in the coming years, we will gain a better understanding of organic causes of preterm labor, permitting us to customize therapy according to individual circumstances of each pregnancy.

A twin pregnancy at 25 weeks' gestation shows discordancy for fetal size. The patient presented in preterm labor and will be followed closely. Courtesy Dr. Washington Hill

A Practical Perspective on a Complicated Problem

The health of any country is judged by the survival of its infants. The United States spends 15% of its gross national product on health care, yet it ranks 21st in the world in its infant mortality rate of 8 deaths per 1,000 live births, according to the World Health Organization. The two main contributors to this death rate are prematurity and birth defects.

Aggressive research programs are aimed at trying to understand the pathophysiology of preterm birth, and clinical interventions have been introduced in an attempt to reduce this unacceptably high rate of preterm birth.

Washington Clark Hill, M.D., the guest expert for the Master Class this month, has long studied preterm labor in the context of both singleton and multiple gestations. He has published comprehensive overviews of research on the complications of tocolysis and the prevention and treatment of preterm labor.

A graduate of Temple University School of Medicine in Philadelphia, Dr. Hill did his residency training at William Beaumont Army Medical Center in El Paso, Tex., before completing a fellowship in maternal-fetal medicine at the University of California, San Francisco.

He is director of the perinatal center and the division of maternal-fetal medicine at the Sarasota (Fla.) Memorial Hospital. Dr. Hill brings a clinical and practical perspective to this complicated problem. His insights will allow us to disentangle fact from fiction and what works from what doesn't.

Preterm labor is one of the most important and vexing challenges complicating pregnancy. Premature babies account for an estimated 6%–10% of births, yet they account for 70%–85% of neonatal morbidity and mortality.

In multiple gestations, which are increasingly common as a result of delayed childbearing and the use of assisted reproductive technologies, preterm labor is an even greater risk. The literature points to an incidence of preterm labor of 20%–75% in multiple gestations. Although these figures may be somewhat high, I think it is safe to say that at least 1 in 10 multiple gestations seen at my institution are complicated by preterm labor. Not all of these patients will be admitted or will deliver early, but the very common nature of this problem and the potentially lethal consequences of premature multiple deliveries make this an issue that every physician and institution should approach carefully.

First, it is important to consider the delivery goal of a multiple gestation pregnancy. Overall, most twins are delivered at 37–38 weeks. For triplets, the gestational age is closer to 34 weeks, and quadruplets are born at around 30 weeks. These are reasonable numbers applicable to community practice.

If a patient arrives in preterm labor, you have to decide what to do, considering her situation and the capabilities of your local hospital and medical staff. It is clear that premature babies fare best when they are cared for in the institution where they are born. If someone needs to be transferred, it should be the mother, not the baby.

Obviously, if a patient carrying twins presents in labor at 35 or 36 weeks, most obstetricians would be inclined to do very little to cut short the labor, because—in the absence of other complications—these babies are likely to do well. However, if she presents at 29 weeks, it would make sense to be more aggressive.

Can the patient be safely and aggressively managed for preterm labor in her local community? The answer hinges on the plan for delivery if the treatment fails. Each hospital and service has to pick a gestational age at which neonatal survival is acceptably high. Then, options for the mother should be discussed with her and with the neonatal intensive care unit at your hospital or the institution to which she will be transferred.

Depending on the circumstances, the treatment of preterm labor may be undertaken for several reasons:

▸ To delay delivery until the patient can be transferred to a tertiary medical center with a high-level neonatal intensive care unit.

▸ To delay delivery 24–48 hours for the administration of corticosteroid therapy.

▸ To reduce the strength and frequency of uterine contractions, enabling the fetus to further develop in the uterus.

▸ To minimize hospital stays for the mother and the neonate.

▸ To reduce the risk of neonatal morbidity and mortality by preventing preterm delivery, the most dangerous complication of multiple gestation pregnancies.

When I consult with a woman in preterm labor, I go through a list of available options. Unfortunately, a careful review of the literature reveals few really good, effective treatments.

Although new ideas emerge every few years, not many interventional strategies have withstood attempts to corroborate results from single institutions. It may be tempting to “just do something,” but we owe it to our patients to stick to scientifically valid and efficacious treatments.

Bed Rest

Bed rest or activity restriction will not prevent preterm labor. Rest neither lengthens gestation nor reduces neonatal morbidity in multiple gestation pregnancies. In some studies these patients did worse.

If a patient carrying multiples has a short cervix and threatened preterm labor, there is some evidence to support getting her off her feet rather than having her continue working at a very active job.

Once a multiple pregnancy is complicated by preterm labor, hospitalization may be necessary for observation and implementation of a treatment course.

Hydration

There is no evidence that hydration is an effective treatment for preterm labor. In fact, the initial administration of bolus intravenous fluids may pose some risk to patients with multiple gestations. These patients already have an increased blood volume and could develop pulmonary edema from fluid overload if tocolytic therapy is initiated after unnecessary fluids are administered.

Progesterone

A study by Paul J. Meis, M.D., and colleagues (N. Engl. J. Med. 2003;348:2379–85) suggests recurrent preterm birth in singleton pregnancies can be prevented by 17 δ-hydroxyprogesterone caproate. The jury is still out on whether progesterone can be useful in managing active or threatened preterm labor in a multiple gestation pregnancy.

Studies are underway that may provide us with more guidance in the use of this agent. However, no evidence exists that it is safe and efficacious in multiple gestation pregnancies, so I suggest its use be reserved for patients in clinical trials.

Antibiotics

It is tantalizing to believe antibiotics would be helpful in preventing or treating preterm labor. Many researchers theorize that intrauterine infection or fetal infection may be responsible for preterm labor, particularly in pregnancies that are not complicated by multiple fetuses. However, the data do not show that antibiotic treatment is any more efficacious than placebo in prolonging pregnancy or preventing preterm delivery.

Tocolytics

The use of tocolytics to decrease or halt preterm labor is controversial in multiple gestations as well as in singleton pregnancies because the drugs pose risks to the mother and, in some cases, to the fetus. However, the available data support the position that tocolytic agents work for a short period—about 48 hours, although Roger B. Newman, M.D., and colleagues have shown that some multiple gestations can be prolonged for more than 7 days (Am. J. Obstet. Gynecol. 1989;161:547–55; “Multifetal Pregnancy: A Handbook for Care of the Pregnant Patient” [Philadelphia: Lippincott Williams & Wilkins, 2000]).

Each tocolytic agent carries its own benefits, risks, contraindications, and adverse effects profile. Numerous sources are available for this information; for quick reference; I recently published a summary in chart form (Clin. Obstet. Gynecol. 2004;47:216–26). Keep in mind that women with multiple gestations have an elevated risk of cardiovascular complications, such as pulmonary edema resulting from anemia, lower colloid oncotic pressure, and higher blood volume.

I would take a middle-of-the-road approach in choosing an agent or agents for tocolysis. For example, oral terbutaline, oral calcium channel blockers, and oral Indocin have been well-studied and widely used, with varying levels of success.

John P. Elliott, M.D., and Tari Radin, Ph.D., studied a small number of high-order multiple gestations and found similar levels of serum magnesium in triplets and quadruplets and in singleton pregnancies after the administration of magnesium sulfate for tocolysis (J. Reprod. Med. 1995;40:450–2). However, they concluded that higher levels of magnesium sulfate are needed in multiple gestations to inhibit labor. They suggested administration at infusion rates of 4–5 g/h in triplet and quadruplet pregnancies.

Research on combination tocolytic therapy has produced conflicting results. Most of the studies on this topic are dated, limited in scope, and not specifically focused on multiple gestations. Some concerns, however, have been raised. For example, magnesium sulfate administered in conjunction with nifedipine can result in significant neuromuscular blockage and a subsequent marked hypotensive effect.

My recommendation is to administer combination tocolytic therapy only with great caution, using agents such as intravenous magnesium sulfate with oral terbutaline or indomethacin. The mother and fetus also should be monitored closely by professionals well-versed in side effects linked to this form of therapy.

Should tocolytic therapy be maintained after successful cessation of labor? A careful reading of the available evidence suggests the answer is no. In well-designed studies, maintenance tocolytic therapy has reduced neither preterm deliveries nor perinatal morbidity or mortality.

Corticosteroids

Corticosteroids are among the few noncontroversial agents for use during preterm labor. These agents—which clearly reduce the incidence and severity of neonatal respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, and neonatal mortality—should be administered to any patient in preterm labor at 24–34 weeks, using the same regimen as in singleton gestations.

The accepted corticosteroid treatment consists of two intramuscular doses of betamethasone or four intramuscular doses of dexamethasone.

In higher-order gestations, Dr. Elliott and Dr. Radin have observed that beta-methasone can increase uterine contractions and advance labor in patients with frequent contractions. They therefore recommend that corticosteroids be reserved in these pregnancies for patients having fewer than three contractions per hour (Obstet. Gynecol. 1995;85:250–54). Uterine activity in higher-order gestations should be closely monitored following the administration of corticosteroids.

It once was suggested that multiple courses of prenatal corticosteroids might be of benefit. This recommendation is no longer made, because there is no good evidence for enhanced efficacy and because repeated doses are associated with clear risks, including decreased fetal growth and reduced birth weight.

Adjunctive Treatments

Isolated studies have suggested a role in preterm labor for such interventions as vaginal lever pessaries, blood transfusions, and cerclage. None of these have been put to rigorous scientific scrutiny. Cerclage obviously has no role in patients who do not have an incompetent cervix, or in such specific cases as the previable delivery of one dichorionic twin when a delayed-interval delivery of the second twin is desired.

Treating the Whole Patient

Not infrequently, preterm labor in a multiple gestation pregnancy requires the prolonged hospitalization of the mother, who may be a busy professional woman or the parent of other small children. Spending 5–10 weeks in a hospital is a long, frustrating time, replete with boredom and anxiety.

At my institution, a clinical nurse specialist who was concerned with the psychosocial effect of hospitalization on the women in our unit proposed what has become a highly successful multidimensional program: Mom Matters. The program offers support, diversion, and empathy in the form of wheelchair outings, Internet access, flexible visitation for family members (including children), manicures, pedicures, movies, and get-togethers.

The success of our program and the gratitude expressed by the mothers on our service have convinced me that this is a kind and therapeutic approach worth considering and implementing elsewhere.

The Future

The road to safe and effective therapy for preterm labor has been a long and frustrating one. I am hopeful that researchers are now on the right track, focusing on subtle indications of infection and other potential causes of preterm labor.

In multiple gestations, of course, preterm labor is often caused simply because there are too many fetuses in the womb. Uterine overdistension will not be an easy problem to overcome, except by reducing multiple gestations. Many of our colleagues in reproductive endocrinology have gotten the message that implanting too many embryos is unwise and unethical.

I believe that in the coming years, we will gain a better understanding of organic causes of preterm labor, permitting us to customize therapy according to individual circumstances of each pregnancy.

A twin pregnancy at 25 weeks' gestation shows discordancy for fetal size. The patient presented in preterm labor and will be followed closely. Courtesy Dr. Washington Hill

A Practical Perspective on a Complicated Problem

The health of any country is judged by the survival of its infants. The United States spends 15% of its gross national product on health care, yet it ranks 21st in the world in its infant mortality rate of 8 deaths per 1,000 live births, according to the World Health Organization. The two main contributors to this death rate are prematurity and birth defects.

Aggressive research programs are aimed at trying to understand the pathophysiology of preterm birth, and clinical interventions have been introduced in an attempt to reduce this unacceptably high rate of preterm birth.

Washington Clark Hill, M.D., the guest expert for the Master Class this month, has long studied preterm labor in the context of both singleton and multiple gestations. He has published comprehensive overviews of research on the complications of tocolysis and the prevention and treatment of preterm labor.

A graduate of Temple University School of Medicine in Philadelphia, Dr. Hill did his residency training at William Beaumont Army Medical Center in El Paso, Tex., before completing a fellowship in maternal-fetal medicine at the University of California, San Francisco.

He is director of the perinatal center and the division of maternal-fetal medicine at the Sarasota (Fla.) Memorial Hospital. Dr. Hill brings a clinical and practical perspective to this complicated problem. His insights will allow us to disentangle fact from fiction and what works from what doesn't.

Preterm labor is one of the most important and vexing challenges complicating pregnancy. Premature babies account for an estimated 6%–10% of births, yet they account for 70%–85% of neonatal morbidity and mortality.

In multiple gestations, which are increasingly common as a result of delayed childbearing and the use of assisted reproductive technologies, preterm labor is an even greater risk. The literature points to an incidence of preterm labor of 20%–75% in multiple gestations. Although these figures may be somewhat high, I think it is safe to say that at least 1 in 10 multiple gestations seen at my institution are complicated by preterm labor. Not all of these patients will be admitted or will deliver early, but the very common nature of this problem and the potentially lethal consequences of premature multiple deliveries make this an issue that every physician and institution should approach carefully.

First, it is important to consider the delivery goal of a multiple gestation pregnancy. Overall, most twins are delivered at 37–38 weeks. For triplets, the gestational age is closer to 34 weeks, and quadruplets are born at around 30 weeks. These are reasonable numbers applicable to community practice.

If a patient arrives in preterm labor, you have to decide what to do, considering her situation and the capabilities of your local hospital and medical staff. It is clear that premature babies fare best when they are cared for in the institution where they are born. If someone needs to be transferred, it should be the mother, not the baby.

Obviously, if a patient carrying twins presents in labor at 35 or 36 weeks, most obstetricians would be inclined to do very little to cut short the labor, because—in the absence of other complications—these babies are likely to do well. However, if she presents at 29 weeks, it would make sense to be more aggressive.

Can the patient be safely and aggressively managed for preterm labor in her local community? The answer hinges on the plan for delivery if the treatment fails. Each hospital and service has to pick a gestational age at which neonatal survival is acceptably high. Then, options for the mother should be discussed with her and with the neonatal intensive care unit at your hospital or the institution to which she will be transferred.

Depending on the circumstances, the treatment of preterm labor may be undertaken for several reasons:

▸ To delay delivery until the patient can be transferred to a tertiary medical center with a high-level neonatal intensive care unit.

▸ To delay delivery 24–48 hours for the administration of corticosteroid therapy.

▸ To reduce the strength and frequency of uterine contractions, enabling the fetus to further develop in the uterus.

▸ To minimize hospital stays for the mother and the neonate.

▸ To reduce the risk of neonatal morbidity and mortality by preventing preterm delivery, the most dangerous complication of multiple gestation pregnancies.

When I consult with a woman in preterm labor, I go through a list of available options. Unfortunately, a careful review of the literature reveals few really good, effective treatments.

Although new ideas emerge every few years, not many interventional strategies have withstood attempts to corroborate results from single institutions. It may be tempting to “just do something,” but we owe it to our patients to stick to scientifically valid and efficacious treatments.

Bed Rest

Bed rest or activity restriction will not prevent preterm labor. Rest neither lengthens gestation nor reduces neonatal morbidity in multiple gestation pregnancies. In some studies these patients did worse.

If a patient carrying multiples has a short cervix and threatened preterm labor, there is some evidence to support getting her off her feet rather than having her continue working at a very active job.

Once a multiple pregnancy is complicated by preterm labor, hospitalization may be necessary for observation and implementation of a treatment course.

Hydration

There is no evidence that hydration is an effective treatment for preterm labor. In fact, the initial administration of bolus intravenous fluids may pose some risk to patients with multiple gestations. These patients already have an increased blood volume and could develop pulmonary edema from fluid overload if tocolytic therapy is initiated after unnecessary fluids are administered.

Progesterone

A study by Paul J. Meis, M.D., and colleagues (N. Engl. J. Med. 2003;348:2379–85) suggests recurrent preterm birth in singleton pregnancies can be prevented by 17 δ-hydroxyprogesterone caproate. The jury is still out on whether progesterone can be useful in managing active or threatened preterm labor in a multiple gestation pregnancy.

Studies are underway that may provide us with more guidance in the use of this agent. However, no evidence exists that it is safe and efficacious in multiple gestation pregnancies, so I suggest its use be reserved for patients in clinical trials.

Antibiotics

It is tantalizing to believe antibiotics would be helpful in preventing or treating preterm labor. Many researchers theorize that intrauterine infection or fetal infection may be responsible for preterm labor, particularly in pregnancies that are not complicated by multiple fetuses. However, the data do not show that antibiotic treatment is any more efficacious than placebo in prolonging pregnancy or preventing preterm delivery.

Tocolytics

The use of tocolytics to decrease or halt preterm labor is controversial in multiple gestations as well as in singleton pregnancies because the drugs pose risks to the mother and, in some cases, to the fetus. However, the available data support the position that tocolytic agents work for a short period—about 48 hours, although Roger B. Newman, M.D., and colleagues have shown that some multiple gestations can be prolonged for more than 7 days (Am. J. Obstet. Gynecol. 1989;161:547–55; “Multifetal Pregnancy: A Handbook for Care of the Pregnant Patient” [Philadelphia: Lippincott Williams & Wilkins, 2000]).

Each tocolytic agent carries its own benefits, risks, contraindications, and adverse effects profile. Numerous sources are available for this information; for quick reference; I recently published a summary in chart form (Clin. Obstet. Gynecol. 2004;47:216–26). Keep in mind that women with multiple gestations have an elevated risk of cardiovascular complications, such as pulmonary edema resulting from anemia, lower colloid oncotic pressure, and higher blood volume.

I would take a middle-of-the-road approach in choosing an agent or agents for tocolysis. For example, oral terbutaline, oral calcium channel blockers, and oral Indocin have been well-studied and widely used, with varying levels of success.

John P. Elliott, M.D., and Tari Radin, Ph.D., studied a small number of high-order multiple gestations and found similar levels of serum magnesium in triplets and quadruplets and in singleton pregnancies after the administration of magnesium sulfate for tocolysis (J. Reprod. Med. 1995;40:450–2). However, they concluded that higher levels of magnesium sulfate are needed in multiple gestations to inhibit labor. They suggested administration at infusion rates of 4–5 g/h in triplet and quadruplet pregnancies.

Research on combination tocolytic therapy has produced conflicting results. Most of the studies on this topic are dated, limited in scope, and not specifically focused on multiple gestations. Some concerns, however, have been raised. For example, magnesium sulfate administered in conjunction with nifedipine can result in significant neuromuscular blockage and a subsequent marked hypotensive effect.

My recommendation is to administer combination tocolytic therapy only with great caution, using agents such as intravenous magnesium sulfate with oral terbutaline or indomethacin. The mother and fetus also should be monitored closely by professionals well-versed in side effects linked to this form of therapy.

Should tocolytic therapy be maintained after successful cessation of labor? A careful reading of the available evidence suggests the answer is no. In well-designed studies, maintenance tocolytic therapy has reduced neither preterm deliveries nor perinatal morbidity or mortality.

Corticosteroids

Corticosteroids are among the few noncontroversial agents for use during preterm labor. These agents—which clearly reduce the incidence and severity of neonatal respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, and neonatal mortality—should be administered to any patient in preterm labor at 24–34 weeks, using the same regimen as in singleton gestations.

The accepted corticosteroid treatment consists of two intramuscular doses of betamethasone or four intramuscular doses of dexamethasone.

In higher-order gestations, Dr. Elliott and Dr. Radin have observed that beta-methasone can increase uterine contractions and advance labor in patients with frequent contractions. They therefore recommend that corticosteroids be reserved in these pregnancies for patients having fewer than three contractions per hour (Obstet. Gynecol. 1995;85:250–54). Uterine activity in higher-order gestations should be closely monitored following the administration of corticosteroids.

It once was suggested that multiple courses of prenatal corticosteroids might be of benefit. This recommendation is no longer made, because there is no good evidence for enhanced efficacy and because repeated doses are associated with clear risks, including decreased fetal growth and reduced birth weight.

Adjunctive Treatments

Isolated studies have suggested a role in preterm labor for such interventions as vaginal lever pessaries, blood transfusions, and cerclage. None of these have been put to rigorous scientific scrutiny. Cerclage obviously has no role in patients who do not have an incompetent cervix, or in such specific cases as the previable delivery of one dichorionic twin when a delayed-interval delivery of the second twin is desired.

Treating the Whole Patient

Not infrequently, preterm labor in a multiple gestation pregnancy requires the prolonged hospitalization of the mother, who may be a busy professional woman or the parent of other small children. Spending 5–10 weeks in a hospital is a long, frustrating time, replete with boredom and anxiety.

At my institution, a clinical nurse specialist who was concerned with the psychosocial effect of hospitalization on the women in our unit proposed what has become a highly successful multidimensional program: Mom Matters. The program offers support, diversion, and empathy in the form of wheelchair outings, Internet access, flexible visitation for family members (including children), manicures, pedicures, movies, and get-togethers.

The success of our program and the gratitude expressed by the mothers on our service have convinced me that this is a kind and therapeutic approach worth considering and implementing elsewhere.

The Future

The road to safe and effective therapy for preterm labor has been a long and frustrating one. I am hopeful that researchers are now on the right track, focusing on subtle indications of infection and other potential causes of preterm labor.

In multiple gestations, of course, preterm labor is often caused simply because there are too many fetuses in the womb. Uterine overdistension will not be an easy problem to overcome, except by reducing multiple gestations. Many of our colleagues in reproductive endocrinology have gotten the message that implanting too many embryos is unwise and unethical.

I believe that in the coming years, we will gain a better understanding of organic causes of preterm labor, permitting us to customize therapy according to individual circumstances of each pregnancy.

A twin pregnancy at 25 weeks' gestation shows discordancy for fetal size. The patient presented in preterm labor and will be followed closely. Courtesy Dr. Washington Hill

A Practical Perspective on a Complicated Problem

The health of any country is judged by the survival of its infants. The United States spends 15% of its gross national product on health care, yet it ranks 21st in the world in its infant mortality rate of 8 deaths per 1,000 live births, according to the World Health Organization. The two main contributors to this death rate are prematurity and birth defects.

Aggressive research programs are aimed at trying to understand the pathophysiology of preterm birth, and clinical interventions have been introduced in an attempt to reduce this unacceptably high rate of preterm birth.

Washington Clark Hill, M.D., the guest expert for the Master Class this month, has long studied preterm labor in the context of both singleton and multiple gestations. He has published comprehensive overviews of research on the complications of tocolysis and the prevention and treatment of preterm labor.

A graduate of Temple University School of Medicine in Philadelphia, Dr. Hill did his residency training at William Beaumont Army Medical Center in El Paso, Tex., before completing a fellowship in maternal-fetal medicine at the University of California, San Francisco.

He is director of the perinatal center and the division of maternal-fetal medicine at the Sarasota (Fla.) Memorial Hospital. Dr. Hill brings a clinical and practical perspective to this complicated problem. His insights will allow us to disentangle fact from fiction and what works from what doesn't.