Imaging

The vast majority of musculotendinous injuries occur secondary to violent contraction or excessive stretching.¹ Ligamentous injuries, on the other hand, are due to an abnormal motion of joints. The magnitude of inciting forces results in a spectrum of pathology, ranging from a minor tear to a complete disruption of structures.

Ultrasonography provides a detailed assessment of soft tissue anatomy and dynamic functionality, and in some instances can be comparable or even superior to magnetic resonance imaging² because the structural characteristics of certain tendons make them ideal for imaging via ultrasonography. We describe some of these characteristics and highlight their utility in diagnostic imaging.

Anatomical Structure

Tendons consist of tightly packed type I collagen fibers forming subfascicles that are arranged in a parallel distribution as fascicles. These bundles are held together by loose soft tissue, and the entire structure is covered by a thick fibroelastic epitendineum sheath. This linear distribution of structures yields a uniquely linear “fibrillary” pattern when viewed along the longitudinal axis of the structure (Figure 1a). In the short-axis view, the tendon appears as a well-circumscribed structure with speckled pattern of hyperechoic foci (Figure 1b). ³

Imaging Technique

The optimal scanning technique involves the use of a high-frequency linear transducer. Higher frequencies yield more detailed images, but may be limited in patients with deeper structures due to body habitus. A key concept in tendon evaluation is an artifact known as “anisotropy.” This refers to change in appearance of the tendon based on the incident angle of the ultrasound beam. For example, when the probe is held perpendicular to the structure of interest, parallel fibers will reflect the emitted beam toward the probe and thus appear as hyperechoic and speckled, a characteristic of these fibers (Figures 1a and 1b). Contrarily, if the probe is held at a nonperpendicular angle, the reflected beam will not return to the probe, resulting in a hypoechoic appearance (Figure 2).

Pathology

Tendon strains result in varying degrees of fibrous tearing. These tears can range from first-degree tears (a few fibers) to third-degree tears (complete disruption). Partial tears result in focal hematoma formation (Figure 3a) at the region of disruption, appearing on ultrasound as a hypoechoic fluid collection within a hyperechoic fibrillary or speckled tendon structure. If the disruption occurs along the surface of the tendon, a focus of anechoic fluid may be seen surrounding the tendon. Complete tendon ruptures, on the other hand, appear as a hypoechoic void with retracted tendon fragments visualized on either side⁴ (Figures 3b and 3c). Although complete tears can be more apparent clinically in areas in which a group of tendons performs a cohesive movement (ie, rotator cuff), ultrasound can significantly reduce the rate of delayed diagnosis when physical examination is equivocal.

In the appropriate clinical setting, ultrasonography can provide rapid and dynamic assessment of musculotendinous injuries. Lower extremity injuries, including those affecting the Achilles (Figure 1), quadriceps, (Figures 4a and 4b) and patellar tendons (Figure 5), are easier clinical applications.

Assessment of rotator cuff tendons, although more difficult, can provide a specific assessment of shoulder pain.⁵ In such scenarios, ultrasound can serve a very useful role as an adjunct to the physical examination.

Summary

Musculotendinous injuries many times present as nonspecific symptoms of pain and/or swelling. In the case of an equivocal physical examination, musculotendinous injuries can be diagnosed with increased accuracy through the use of ultrasound. Understanding the artifactual component of tendon ultrasound can aid the clinician in diagnosing these injuries, enhancing patient care and satisfaction.

The vast majority of musculotendinous injuries occur secondary to violent contraction or excessive stretching.¹ Ligamentous injuries, on the other hand, are due to an abnormal motion of joints. The magnitude of inciting forces results in a spectrum of pathology, ranging from a minor tear to a complete disruption of structures.

Ultrasonography provides a detailed assessment of soft tissue anatomy and dynamic functionality, and in some instances can be comparable or even superior to magnetic resonance imaging² because the structural characteristics of certain tendons make them ideal for imaging via ultrasonography. We describe some of these characteristics and highlight their utility in diagnostic imaging.

Anatomical Structure

Tendons consist of tightly packed type I collagen fibers forming subfascicles that are arranged in a parallel distribution as fascicles. These bundles are held together by loose soft tissue, and the entire structure is covered by a thick fibroelastic epitendineum sheath. This linear distribution of structures yields a uniquely linear “fibrillary” pattern when viewed along the longitudinal axis of the structure (Figure 1a). In the short-axis view, the tendon appears as a well-circumscribed structure with speckled pattern of hyperechoic foci (Figure 1b). ³

Imaging Technique

The optimal scanning technique involves the use of a high-frequency linear transducer. Higher frequencies yield more detailed images, but may be limited in patients with deeper structures due to body habitus. A key concept in tendon evaluation is an artifact known as “anisotropy.” This refers to change in appearance of the tendon based on the incident angle of the ultrasound beam. For example, when the probe is held perpendicular to the structure of interest, parallel fibers will reflect the emitted beam toward the probe and thus appear as hyperechoic and speckled, a characteristic of these fibers (Figures 1a and 1b). Contrarily, if the probe is held at a nonperpendicular angle, the reflected beam will not return to the probe, resulting in a hypoechoic appearance (Figure 2).

Pathology

Tendon strains result in varying degrees of fibrous tearing. These tears can range from first-degree tears (a few fibers) to third-degree tears (complete disruption). Partial tears result in focal hematoma formation (Figure 3a) at the region of disruption, appearing on ultrasound as a hypoechoic fluid collection within a hyperechoic fibrillary or speckled tendon structure. If the disruption occurs along the surface of the tendon, a focus of anechoic fluid may be seen surrounding the tendon. Complete tendon ruptures, on the other hand, appear as a hypoechoic void with retracted tendon fragments visualized on either side⁴ (Figures 3b and 3c). Although complete tears can be more apparent clinically in areas in which a group of tendons performs a cohesive movement (ie, rotator cuff), ultrasound can significantly reduce the rate of delayed diagnosis when physical examination is equivocal.

In the appropriate clinical setting, ultrasonography can provide rapid and dynamic assessment of musculotendinous injuries. Lower extremity injuries, including those affecting the Achilles (Figure 1), quadriceps, (Figures 4a and 4b) and patellar tendons (Figure 5), are easier clinical applications.

Assessment of rotator cuff tendons, although more difficult, can provide a specific assessment of shoulder pain.⁵ In such scenarios, ultrasound can serve a very useful role as an adjunct to the physical examination.

Summary

Musculotendinous injuries many times present as nonspecific symptoms of pain and/or swelling. In the case of an equivocal physical examination, musculotendinous injuries can be diagnosed with increased accuracy through the use of ultrasound. Understanding the artifactual component of tendon ultrasound can aid the clinician in diagnosing these injuries, enhancing patient care and satisfaction.

The vast majority of musculotendinous injuries occur secondary to violent contraction or excessive stretching.¹ Ligamentous injuries, on the other hand, are due to an abnormal motion of joints. The magnitude of inciting forces results in a spectrum of pathology, ranging from a minor tear to a complete disruption of structures.

Ultrasonography provides a detailed assessment of soft tissue anatomy and dynamic functionality, and in some instances can be comparable or even superior to magnetic resonance imaging² because the structural characteristics of certain tendons make them ideal for imaging via ultrasonography. We describe some of these characteristics and highlight their utility in diagnostic imaging.

Anatomical Structure

Tendons consist of tightly packed type I collagen fibers forming subfascicles that are arranged in a parallel distribution as fascicles. These bundles are held together by loose soft tissue, and the entire structure is covered by a thick fibroelastic epitendineum sheath. This linear distribution of structures yields a uniquely linear “fibrillary” pattern when viewed along the longitudinal axis of the structure (Figure 1a). In the short-axis view, the tendon appears as a well-circumscribed structure with speckled pattern of hyperechoic foci (Figure 1b). ³

Imaging Technique

The optimal scanning technique involves the use of a high-frequency linear transducer. Higher frequencies yield more detailed images, but may be limited in patients with deeper structures due to body habitus. A key concept in tendon evaluation is an artifact known as “anisotropy.” This refers to change in appearance of the tendon based on the incident angle of the ultrasound beam. For example, when the probe is held perpendicular to the structure of interest, parallel fibers will reflect the emitted beam toward the probe and thus appear as hyperechoic and speckled, a characteristic of these fibers (Figures 1a and 1b). Contrarily, if the probe is held at a nonperpendicular angle, the reflected beam will not return to the probe, resulting in a hypoechoic appearance (Figure 2).

Pathology

Tendon strains result in varying degrees of fibrous tearing. These tears can range from first-degree tears (a few fibers) to third-degree tears (complete disruption). Partial tears result in focal hematoma formation (Figure 3a) at the region of disruption, appearing on ultrasound as a hypoechoic fluid collection within a hyperechoic fibrillary or speckled tendon structure. If the disruption occurs along the surface of the tendon, a focus of anechoic fluid may be seen surrounding the tendon. Complete tendon ruptures, on the other hand, appear as a hypoechoic void with retracted tendon fragments visualized on either side⁴ (Figures 3b and 3c). Although complete tears can be more apparent clinically in areas in which a group of tendons performs a cohesive movement (ie, rotator cuff), ultrasound can significantly reduce the rate of delayed diagnosis when physical examination is equivocal.

In the appropriate clinical setting, ultrasonography can provide rapid and dynamic assessment of musculotendinous injuries. Lower extremity injuries, including those affecting the Achilles (Figure 1), quadriceps, (Figures 4a and 4b) and patellar tendons (Figure 5), are easier clinical applications.

Assessment of rotator cuff tendons, although more difficult, can provide a specific assessment of shoulder pain.⁵ In such scenarios, ultrasound can serve a very useful role as an adjunct to the physical examination.

Summary

Musculotendinous injuries many times present as nonspecific symptoms of pain and/or swelling. In the case of an equivocal physical examination, musculotendinous injuries can be diagnosed with increased accuracy through the use of ultrasound. Understanding the artifactual component of tendon ultrasound can aid the clinician in diagnosing these injuries, enhancing patient care and satisfaction.

Hospitalists participated in a double-header of hands-on point-of-care ultrasound training here on Monday, looking to gain an edge in expertise in a role that’s becoming more and more common.

Nearly 100 hospitalists and other health care professionals heard talks on the fundamental principles of ultrasound and cardiac, lung and vascular, and abdominal ultrasound. The highlights of the sessions were two 80-minute hands-on segments using the probes.

“This course has grown and grown – this is the largest we’ve ever done,” said pre-course director Nilam Soni, MD, MS, FHM, associate professor of medicine at the University of Texas Health Science Center San Antonio.

Darnell Scott/Frontline Medical News

Darnell Scott/Frontline Medical News

Hospitalists participated in a double-header of hands-on point-of-care ultrasound training here on Monday, looking to gain an edge in expertise in a role that’s becoming more and more common.

Nearly 100 hospitalists and other health care professionals heard talks on the fundamental principles of ultrasound and cardiac, lung and vascular, and abdominal ultrasound. The highlights of the sessions were two 80-minute hands-on segments using the probes.

“This course has grown and grown – this is the largest we’ve ever done,” said pre-course director Nilam Soni, MD, MS, FHM, associate professor of medicine at the University of Texas Health Science Center San Antonio.

Darnell Scott/Frontline Medical News

Darnell Scott/Frontline Medical News

Hospitalists participated in a double-header of hands-on point-of-care ultrasound training here on Monday, looking to gain an edge in expertise in a role that’s becoming more and more common.

Nearly 100 hospitalists and other health care professionals heard talks on the fundamental principles of ultrasound and cardiac, lung and vascular, and abdominal ultrasound. The highlights of the sessions were two 80-minute hands-on segments using the probes.

“This course has grown and grown – this is the largest we’ve ever done,” said pre-course director Nilam Soni, MD, MS, FHM, associate professor of medicine at the University of Texas Health Science Center San Antonio.

Darnell Scott/Frontline Medical News

Darnell Scott/Frontline Medical News

Take-Home Points

• Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG.
• TT-TG distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.
• TT-TG distance criteria should serve as a guide, rather than a rigid threshold, in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.
• Factors such as knee flexion angle, imaging modality, and landmarks used for the measurements should be considered when using TT-TG distance as an indication for surgery.

• There has been significant variability in reported TT-TG measurements. A surgeon using this measurement should understand how it is obtained because many technical factors are involved.

Assessment of malalignment is an important factor in determining surgical treatment options for patellar instability. Although soft-tissue reconstruction of the medial soft-tissue stabilizers is often performed to address patellar instability, bony malalignment may increase stress on the medial soft tissues; therefore, it must be adequately identified and addressed.

Bony malalignment, which is often thought of as lateralization of the tibial tubercle (TT), can be influenced by tibiofemoral alignment, external tibial torsion, and femoral anteversion.

Clinically, coronal alignment can be assessed with a measurement such as quadriceps (Q) angle, but this has been reported to have low interrater reliability and high variability in the reported optimal conditions and positions in which the measurement should be made.^1-3An anatomically lateralized TT pulls the extensor mechanism laterally with respect to the trochlear groove (TG), and this can accentuate problems related to patellofemoral instability. A recent biomechanical study found that increased TT lateralization significantly increased lateral patellar translation and tilt in the setting of medial patellofemoral ligament (MPFL) deficiency.⁴ Although MPFL reconstruction restored patellar kinematics and contact mechanics, this restoration did not occur when the TT was lateralized more than 10 mm relative to its normal position.

Realigning the extensor mechanism by moving the TT medially decreases the lateralizing forces on the patella and the stress on the soft-tissue restraints. This raises the issues of when to correct a lateralized TT and how to identify and measure malalignment.

Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG. Originally described on radiographs and subsequently on computed tomography (CT) and magnetic resonance imaging (MRI) scans, distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.^5,6However, there has been significant variability in reported TT-TG measurements. Studies have found that TT-TG distance is 3.8 mm larger on CT scans than on MRI scans.⁷ Furthermore, factors such as knee flexion angle at time of imaging have been found to reduce TT-TG distance.¹ More recently, patient size and TT-TG ratios relative to patellar and trochlear width were identified as important factors in assessing TT-TG distance.⁸ Therefore, TT-TG distance measurements should serve as a guide rather than a rigid threshold in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.

What You Need to Know About Measuring Patellofemoral Malalignment

TT-TG distance can guide decisions about performing a medializing TT osteotomy for patellar instability because the measurement can aid in assessing bony malalignment caused by an anatomically lateralized tubercle. TT-TG distance can be used to determine when and how far to move the tubercle in TT osteotomy.

Background

A normal TT-TG value is approximately 10 mm. The measurement originally used bony landmarks, including the deepest part of the bony TG and the anterior-most part of the TT, as described by Goutallier and colleagues.⁹ In their original study, Dejour and colleagues⁵ found that patients with recurrent symptoms of patellar instability had TT-TG distances >20 mm.

Increased TT-TG distance has been shown to correlate with patellar position, including increased lateral shift and lateral tilt of the patella. In a study using dynamic CT scans of patients with recurrent patellar instability, we found that TT-TG distance increased with knee extension, and that this increase correlated with the lateral shift and lateral tilt of the patella.¹⁰An excessively lateralized TT can be corrected with a medializing osteotomy that reduces TT-TG distance to within the normal range. TT surgery can be performed with flat osteotomy, as described by Elmslie and Trillat,¹¹ or with oblique osteotomy, as described by Fulkerson,⁶ to obtain concomitant anteriorization. In a computational study, Elias and colleagues¹² found that medializing TT osteotomy not only reduced TT-TG distance but led to correction of lateral patellar tilt and displacement. Patellofemoral contact forces have also shown to be reduced with anteromedialization.⁶Although reported outcomes of TT osteotomy have been excellent for patients with patellar instability, the procedure has higher risks and longer rehabilitation relative to a soft-tissue procedure alone. Reported risks associated with TT osteotomy include fracture, nonunion, delayed union, painful screws, and deep vein thrombosis.^6,10,13,14Understanding the limitations of and variability in radiographic assessments of TT and TG positions can help when deciding whether to perform TT osteotomy for patellar instability.

Discussion

When considering TT osteotomy for patellar instability, some surgeons use a TT-TG distance of more than 15 mm or 20 mm as a threshold for performing medialization. The variability is based on the multiple patient and imaging factors that can influence TT-TG distance measurement.

Several TG and TT landmarks have been used to measure TT-TG distance. The deepest part of the TG, based on bony anatomy, was used originally, but the cartilaginous landmark at the deepest part of the cartilaginous TG has also been described.¹⁵ Similarly, on the TT, the original description of TT-TG distance, by Goutallier and colleagues,⁹ involved the anterior-most part of the TT on CT scan, though the central part of the TT has also been described.¹⁵ We found a 4.2-mm difference in TT-TG distance with use of different landmarks (central tubercle, anterior tubercle) within the same study population.¹⁶ Therefore, within a practice, the distance used as an indication for TT osteotomy should be measured consistently.

Knee flexion angle at the time of imaging can also affect measurement of TT-TG distance. Several authors have reported smaller TT-TG distance with increased knee flexion angle.^10,16,17 In a study of patients with symptomatic patellar instability, we found that TT-TG distance decreases by an estimated 1 mm for every 4.4° of knee flexion >0°.¹⁰ In measurements of TT-TG distance, the sagittal view can be used to assess knee flexion angle because positioning protocols and patient comfort at the time of imaging may produce variable knee flexion angles.

Given the variability that occurs in TT-TG distance with knee flexion angles, some surgeons use TT–posterior cruciate ligament (PCL) distance as another measurement of TT lateralization.¹⁸ This measurement is made with both tibial landmarks, from the TT to the medial border of the PCL insertion on the tibia, and theoretically eliminates knee flexion angle as a measurement factor. Seitlinger and colleagues¹⁸ found that values >24 mm were associated with symptoms of patellar instability. More study is needed to determine the precise indications for TT osteotomy with use of this measurement.

In addition to patient positioning during knee imaging, patient size should be considered when TT-TG distance is used for malalignment measurement. Camp and colleagues⁸ discussed the importance of “individualizing” TT-TG distance on the basis of patient size and bony structure. They reported that the ratio of TT-TG distance to trochlear width or patellar width more effectively predicted recurrent patellar instability than TT-TG distance alone.

Measurement of TT-TG distance is valuable in planning surgical treatment for patellar instability because it quantifies a component of malalignment and aids in deciding whether to perform TT osteotomy. However, this distance should be understood in the context of many measurement factors to allow for an individualized procedure that addresses the specific contributors to patellar instability in each patient.

Am J Orthop. 2017;46(3):148-151. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

Take-Home Points

• Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG.
• TT-TG distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.
• TT-TG distance criteria should serve as a guide, rather than a rigid threshold, in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.
• Factors such as knee flexion angle, imaging modality, and landmarks used for the measurements should be considered when using TT-TG distance as an indication for surgery.

• There has been significant variability in reported TT-TG measurements. A surgeon using this measurement should understand how it is obtained because many technical factors are involved.

Assessment of malalignment is an important factor in determining surgical treatment options for patellar instability. Although soft-tissue reconstruction of the medial soft-tissue stabilizers is often performed to address patellar instability, bony malalignment may increase stress on the medial soft tissues; therefore, it must be adequately identified and addressed.

Bony malalignment, which is often thought of as lateralization of the tibial tubercle (TT), can be influenced by tibiofemoral alignment, external tibial torsion, and femoral anteversion.

Clinically, coronal alignment can be assessed with a measurement such as quadriceps (Q) angle, but this has been reported to have low interrater reliability and high variability in the reported optimal conditions and positions in which the measurement should be made.^1-3An anatomically lateralized TT pulls the extensor mechanism laterally with respect to the trochlear groove (TG), and this can accentuate problems related to patellofemoral instability. A recent biomechanical study found that increased TT lateralization significantly increased lateral patellar translation and tilt in the setting of medial patellofemoral ligament (MPFL) deficiency.⁴ Although MPFL reconstruction restored patellar kinematics and contact mechanics, this restoration did not occur when the TT was lateralized more than 10 mm relative to its normal position.

Realigning the extensor mechanism by moving the TT medially decreases the lateralizing forces on the patella and the stress on the soft-tissue restraints. This raises the issues of when to correct a lateralized TT and how to identify and measure malalignment.

Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG. Originally described on radiographs and subsequently on computed tomography (CT) and magnetic resonance imaging (MRI) scans, distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.^5,6However, there has been significant variability in reported TT-TG measurements. Studies have found that TT-TG distance is 3.8 mm larger on CT scans than on MRI scans.⁷ Furthermore, factors such as knee flexion angle at time of imaging have been found to reduce TT-TG distance.¹ More recently, patient size and TT-TG ratios relative to patellar and trochlear width were identified as important factors in assessing TT-TG distance.⁸ Therefore, TT-TG distance measurements should serve as a guide rather than a rigid threshold in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.

What You Need to Know About Measuring Patellofemoral Malalignment

TT-TG distance can guide decisions about performing a medializing TT osteotomy for patellar instability because the measurement can aid in assessing bony malalignment caused by an anatomically lateralized tubercle. TT-TG distance can be used to determine when and how far to move the tubercle in TT osteotomy.

Background

A normal TT-TG value is approximately 10 mm. The measurement originally used bony landmarks, including the deepest part of the bony TG and the anterior-most part of the TT, as described by Goutallier and colleagues.⁹ In their original study, Dejour and colleagues⁵ found that patients with recurrent symptoms of patellar instability had TT-TG distances >20 mm.

Increased TT-TG distance has been shown to correlate with patellar position, including increased lateral shift and lateral tilt of the patella. In a study using dynamic CT scans of patients with recurrent patellar instability, we found that TT-TG distance increased with knee extension, and that this increase correlated with the lateral shift and lateral tilt of the patella.¹⁰An excessively lateralized TT can be corrected with a medializing osteotomy that reduces TT-TG distance to within the normal range. TT surgery can be performed with flat osteotomy, as described by Elmslie and Trillat,¹¹ or with oblique osteotomy, as described by Fulkerson,⁶ to obtain concomitant anteriorization. In a computational study, Elias and colleagues¹² found that medializing TT osteotomy not only reduced TT-TG distance but led to correction of lateral patellar tilt and displacement. Patellofemoral contact forces have also shown to be reduced with anteromedialization.⁶Although reported outcomes of TT osteotomy have been excellent for patients with patellar instability, the procedure has higher risks and longer rehabilitation relative to a soft-tissue procedure alone. Reported risks associated with TT osteotomy include fracture, nonunion, delayed union, painful screws, and deep vein thrombosis.^6,10,13,14Understanding the limitations of and variability in radiographic assessments of TT and TG positions can help when deciding whether to perform TT osteotomy for patellar instability.

Discussion

When considering TT osteotomy for patellar instability, some surgeons use a TT-TG distance of more than 15 mm or 20 mm as a threshold for performing medialization. The variability is based on the multiple patient and imaging factors that can influence TT-TG distance measurement.

Several TG and TT landmarks have been used to measure TT-TG distance. The deepest part of the TG, based on bony anatomy, was used originally, but the cartilaginous landmark at the deepest part of the cartilaginous TG has also been described.¹⁵ Similarly, on the TT, the original description of TT-TG distance, by Goutallier and colleagues,⁹ involved the anterior-most part of the TT on CT scan, though the central part of the TT has also been described.¹⁵ We found a 4.2-mm difference in TT-TG distance with use of different landmarks (central tubercle, anterior tubercle) within the same study population.¹⁶ Therefore, within a practice, the distance used as an indication for TT osteotomy should be measured consistently.

Knee flexion angle at the time of imaging can also affect measurement of TT-TG distance. Several authors have reported smaller TT-TG distance with increased knee flexion angle.^10,16,17 In a study of patients with symptomatic patellar instability, we found that TT-TG distance decreases by an estimated 1 mm for every 4.4° of knee flexion >0°.¹⁰ In measurements of TT-TG distance, the sagittal view can be used to assess knee flexion angle because positioning protocols and patient comfort at the time of imaging may produce variable knee flexion angles.

Given the variability that occurs in TT-TG distance with knee flexion angles, some surgeons use TT–posterior cruciate ligament (PCL) distance as another measurement of TT lateralization.¹⁸ This measurement is made with both tibial landmarks, from the TT to the medial border of the PCL insertion on the tibia, and theoretically eliminates knee flexion angle as a measurement factor. Seitlinger and colleagues¹⁸ found that values >24 mm were associated with symptoms of patellar instability. More study is needed to determine the precise indications for TT osteotomy with use of this measurement.

In addition to patient positioning during knee imaging, patient size should be considered when TT-TG distance is used for malalignment measurement. Camp and colleagues⁸ discussed the importance of “individualizing” TT-TG distance on the basis of patient size and bony structure. They reported that the ratio of TT-TG distance to trochlear width or patellar width more effectively predicted recurrent patellar instability than TT-TG distance alone.

Measurement of TT-TG distance is valuable in planning surgical treatment for patellar instability because it quantifies a component of malalignment and aids in deciding whether to perform TT osteotomy. However, this distance should be understood in the context of many measurement factors to allow for an individualized procedure that addresses the specific contributors to patellar instability in each patient.

Am J Orthop. 2017;46(3):148-151. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

Take-Home Points

• Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG.
• TT-TG distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.
• TT-TG distance criteria should serve as a guide, rather than a rigid threshold, in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.
• Factors such as knee flexion angle, imaging modality, and landmarks used for the measurements should be considered when using TT-TG distance as an indication for surgery.

• There has been significant variability in reported TT-TG measurements. A surgeon using this measurement should understand how it is obtained because many technical factors are involved.

Assessment of malalignment is an important factor in determining surgical treatment options for patellar instability. Although soft-tissue reconstruction of the medial soft-tissue stabilizers is often performed to address patellar instability, bony malalignment may increase stress on the medial soft tissues; therefore, it must be adequately identified and addressed.

Bony malalignment, which is often thought of as lateralization of the tibial tubercle (TT), can be influenced by tibiofemoral alignment, external tibial torsion, and femoral anteversion.

Clinically, coronal alignment can be assessed with a measurement such as quadriceps (Q) angle, but this has been reported to have low interrater reliability and high variability in the reported optimal conditions and positions in which the measurement should be made.^1-3An anatomically lateralized TT pulls the extensor mechanism laterally with respect to the trochlear groove (TG), and this can accentuate problems related to patellofemoral instability. A recent biomechanical study found that increased TT lateralization significantly increased lateral patellar translation and tilt in the setting of medial patellofemoral ligament (MPFL) deficiency.⁴ Although MPFL reconstruction restored patellar kinematics and contact mechanics, this restoration did not occur when the TT was lateralized more than 10 mm relative to its normal position.

Realigning the extensor mechanism by moving the TT medially decreases the lateralizing forces on the patella and the stress on the soft-tissue restraints. This raises the issues of when to correct a lateralized TT and how to identify and measure malalignment.

Radiographic assessment of TT position is most commonly performed by measuring TT-TG distance, which is the distance between the extensor mechanism attachment at the TT and the center of the TG. Originally described on radiographs and subsequently on computed tomography (CT) and magnetic resonance imaging (MRI) scans, distances of more than 15 mm or 20 mm have been reported as indications for TT osteotomy.^5,6However, there has been significant variability in reported TT-TG measurements. Studies have found that TT-TG distance is 3.8 mm larger on CT scans than on MRI scans.⁷ Furthermore, factors such as knee flexion angle at time of imaging have been found to reduce TT-TG distance.¹ More recently, patient size and TT-TG ratios relative to patellar and trochlear width were identified as important factors in assessing TT-TG distance.⁸ Therefore, TT-TG distance measurements should serve as a guide rather than a rigid threshold in the context of imaging and patient factors when deciding whether to perform TT osteotomy for patellar instability.

What You Need to Know About Measuring Patellofemoral Malalignment

TT-TG distance can guide decisions about performing a medializing TT osteotomy for patellar instability because the measurement can aid in assessing bony malalignment caused by an anatomically lateralized tubercle. TT-TG distance can be used to determine when and how far to move the tubercle in TT osteotomy.

Background

A normal TT-TG value is approximately 10 mm. The measurement originally used bony landmarks, including the deepest part of the bony TG and the anterior-most part of the TT, as described by Goutallier and colleagues.⁹ In their original study, Dejour and colleagues⁵ found that patients with recurrent symptoms of patellar instability had TT-TG distances >20 mm.

Increased TT-TG distance has been shown to correlate with patellar position, including increased lateral shift and lateral tilt of the patella. In a study using dynamic CT scans of patients with recurrent patellar instability, we found that TT-TG distance increased with knee extension, and that this increase correlated with the lateral shift and lateral tilt of the patella.¹⁰An excessively lateralized TT can be corrected with a medializing osteotomy that reduces TT-TG distance to within the normal range. TT surgery can be performed with flat osteotomy, as described by Elmslie and Trillat,¹¹ or with oblique osteotomy, as described by Fulkerson,⁶ to obtain concomitant anteriorization. In a computational study, Elias and colleagues¹² found that medializing TT osteotomy not only reduced TT-TG distance but led to correction of lateral patellar tilt and displacement. Patellofemoral contact forces have also shown to be reduced with anteromedialization.⁶Although reported outcomes of TT osteotomy have been excellent for patients with patellar instability, the procedure has higher risks and longer rehabilitation relative to a soft-tissue procedure alone. Reported risks associated with TT osteotomy include fracture, nonunion, delayed union, painful screws, and deep vein thrombosis.^6,10,13,14Understanding the limitations of and variability in radiographic assessments of TT and TG positions can help when deciding whether to perform TT osteotomy for patellar instability.

Discussion

When considering TT osteotomy for patellar instability, some surgeons use a TT-TG distance of more than 15 mm or 20 mm as a threshold for performing medialization. The variability is based on the multiple patient and imaging factors that can influence TT-TG distance measurement.

Several TG and TT landmarks have been used to measure TT-TG distance. The deepest part of the TG, based on bony anatomy, was used originally, but the cartilaginous landmark at the deepest part of the cartilaginous TG has also been described.¹⁵ Similarly, on the TT, the original description of TT-TG distance, by Goutallier and colleagues,⁹ involved the anterior-most part of the TT on CT scan, though the central part of the TT has also been described.¹⁵ We found a 4.2-mm difference in TT-TG distance with use of different landmarks (central tubercle, anterior tubercle) within the same study population.¹⁶ Therefore, within a practice, the distance used as an indication for TT osteotomy should be measured consistently.

Knee flexion angle at the time of imaging can also affect measurement of TT-TG distance. Several authors have reported smaller TT-TG distance with increased knee flexion angle.^10,16,17 In a study of patients with symptomatic patellar instability, we found that TT-TG distance decreases by an estimated 1 mm for every 4.4° of knee flexion >0°.¹⁰ In measurements of TT-TG distance, the sagittal view can be used to assess knee flexion angle because positioning protocols and patient comfort at the time of imaging may produce variable knee flexion angles.

Given the variability that occurs in TT-TG distance with knee flexion angles, some surgeons use TT–posterior cruciate ligament (PCL) distance as another measurement of TT lateralization.¹⁸ This measurement is made with both tibial landmarks, from the TT to the medial border of the PCL insertion on the tibia, and theoretically eliminates knee flexion angle as a measurement factor. Seitlinger and colleagues¹⁸ found that values >24 mm were associated with symptoms of patellar instability. More study is needed to determine the precise indications for TT osteotomy with use of this measurement.

In addition to patient positioning during knee imaging, patient size should be considered when TT-TG distance is used for malalignment measurement. Camp and colleagues⁸ discussed the importance of “individualizing” TT-TG distance on the basis of patient size and bony structure. They reported that the ratio of TT-TG distance to trochlear width or patellar width more effectively predicted recurrent patellar instability than TT-TG distance alone.

Measurement of TT-TG distance is valuable in planning surgical treatment for patellar instability because it quantifies a component of malalignment and aids in deciding whether to perform TT osteotomy. However, this distance should be understood in the context of many measurement factors to allow for an individualized procedure that addresses the specific contributors to patellar instability in each patient.

Am J Orthop. 2017;46(3):148-151. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

Take-Home Points

• Use ultrasound to identify integrity and location of MPFL tear.
• Anatomic repair allows native tissue to reintegrate into bone.
• Repairs done early can prevent complications of recurrent instability.
• Repair maintains biological and proprioceptive qualities of tissue.

• 10Ultrasound-guided percutaneous repair is quick and effective.

The medial patellofemoral ligament (MPFL) is the primary passive restraint to lateral patellar excursion^1-5 and helps control patellar tilt and rotation.^6,7 More than 90% of lateral patellar dislocations cause the MPFL to rupture, and roughly 90% of these detachments involve the femoral insertion.⁴ Ensuing patellar instability often results from MPFL insufficiency. It has been suggested that re-creating the anatomy and functionality of this ligament is of utmost importance in restoring normal patellar biomechanics.^1-5,7,8

Anatomical risk factors for recurrent patellar instability include patella alta, increased tibial tuberosity-trochlear groove (TT-TG) distance, trochlear dysplasia, and torsional abnormalities.^1-4,6 A medial reefing technique with a lateral tissue release traditionally was used to restore proper kinematics, but was shown to have associated postoperative issues.⁹

Methods

Patient Demographics

Dr. Hirahara developed this technique in 2013 and performed it 11 times between 2013 and 2016. Of the 11 patients, 1 was excluded from our retrospective analysis because of trochlear dysplasia, now considered a relative contraindication. Of the remaining 10 patients, 5 (50%) had the repair performed on the right knee. Eight patients (80%) were female. Mean (SD) age was 17.21 (3.53) years. One patient had concurrent femur- and patella-side detachments; otherwise, 6 (60%) of 10 repairs were performed exclusively at the patella. We grade patellar instability according to amount of glide based on patellar width and quadrants. Normal lateral displacement was usually 1 to 2 quadrants of lateral glide relative to the contralateral side. Before surgery, 6 (60%) of the 10 patients presented with lateral glide of 3 quadrants, and 3 (30%) presented with lateral glide of 4 quadrants. All had patellar instability apprehension on physical examination.

Surgical Indications

Before surgery, MPFL integrity is determined by ultrasound evaluation. Repair is considered if the MPFL has a femur- or patella-side tear and is of adequate quantity and quality, and if there are minimal or no arthritic changes (Table 2).

Surgical Technique

The patient is brought to the operating room and placed supine. Patellar stability of the affected knee is assessed and compared with that of the contralateral side with patellar glide. The knee is prepared and draped in usual sterile fashion. With the knee flexed at 90º, a tourniquet is inflated. Diagnostic arthroscopy is performed with standard anteromedial and anterolateral portals, and, if necessary, arthroscopic procedures are performed.

Femoral Attachment Repair

With the leg in extension, ultrasound is used to identify the tear at the femoral attachment (watch part 1 of the video). A spinal needle is placed at the femoral insertion, typically just anterior and distal to the adductor tubercle (Figure 4).¹⁰

Patellar Attachment Repair

With the leg in extension, ultrasound is used to identify where the MPFL is detached from the patella (watch part 2 of the video). A spinal needle is placed at the detachment site (Figure 5). A scalpel is used to make a 1-cm incision down to the patella.

In this description, we showcase knotless and knotted techniques for each repair site. Either method is appropriate for the 2 repair sites. Owing to the superficial nature of the attachment sites—they may have very little fat, particularly at the patella—knot stacks are more prominent, can be felt after surgery, and have the potential to irritate surrounding tissues. Therefore, we prefer knotless fixation for both sites.

Rehabilitation

Rehabilitation after MPFL repair is much like rehabilitation after quadriceps tendon repair. The patient is locked in a brace in full extension when up and moving. Early weight-bearing and minimal use of assistive devices (crutches) are allowed because, when the leg is in full extension, there is no tension at the repair sites. Rehabilitation begins within 1 week, and normal daily function is quickly attained. The protocol emphasizes pain-free motion and suitable patellar mobility, and allows the immobilizing brace to be unlocked for exercise and sitting. During the first 4 weeks, quadriceps activation is limited; progression to full ROM occurs by 4 to 6 weeks. During the strengthening phase, loading the knee in early flexion should be avoided. Return to heavy lifting, physical activity, and sports is delayed until after 6 months in order to allow the construct to mature and integrate. Once the patient has satisfied all the strength, ROM, and functional outcome measurements, a brace is no longer required during sports and normal activity.

Results

Mean tourniquet time for each procedure, which includes diagnostic arthroscopy and ultrasound-guided percutaneous repair, was 26.9 minutes.

Discussion

Conservative management typically is recommended for acute patellar dislocations. In the event of failed conservative management or chronic patellar instability, surgical intervention is indicated. Studies have found that conservative management has recurrent-dislocation rates of 35% at 3-year follow-up and 73% at 6-year follow-up, and recurrent dislocations significantly increase patients’ risk of developing chondral and bony damage.¹³ MPFL repair is designed to restore proper patellar tracking and kinematics while maintaining the anatomical tissue. Lateral patellar dislocations often cause the MPFL to rupture; tears are reported in more than 90% of incidents.⁴ The significant rate indicates that, even after a single patellar dislocation, the MPFL should be evaluated. The MPFL contributes 50% to 60% of the medial stabilizing force during patellar tracking^1,7,14 and is the primary restraint to lateral patellar excursion and excessive patellar tilt and rotation.^1-5 Its absence plays a key role in recurrent lateral patellar instability. With this structure being so important, proper identification and intervention are vital. Studies have established that redislocation rates are significantly higher for nonoperatively (vs operatively) treated primary patellar dislocations.¹³ Simple and accurate percutaneous repair of the MPFL should be performed early to avoid the long-term complications of recurrent instability that could damage the cartilage and bone of the patella and trochlea.

The primary advantage of this technique is its novel use of musculoskeletal ultrasound to accurately identify anatomy and pathology and the placement of anatomical repairs. Accurate preoperative and intraoperative assessment of MPFL anatomy is vital to the success of a procedure. Descriptions of MPFL anatomy suggest discrepancies in the exact locations of the femoral and patellar attachments.^{2,5,7,10,12,15,16} Tanaka⁵ noted that, even within paired knees, there was “marked variability” in the MPFL insertions. McCarthy and colleagues¹⁰ contended the femoral attachment of the MPFL is just anterior and distal to the adductor tubercle, the landmark addressed in this technique. Steensen and colleagues¹⁶ described this attachment site as being statistically the “single most important point affecting isometry” of the MPFL. Sallay and colleagues⁴ asserted that an overwhelming majority of MPFL tears (87%) occur at the adductor tubercle. The variable distribution of tear locations and the importance of re-creating patient anatomy further highlight the need for individualized treatment, which is afforded by ultrasound. Fluoroscopy has been inadequate in identifying MPFL anatomy; this modality is difficult, cumbersome, inaccurate, and inconsistent.^11,12 Conversely, ultrasound provides real-time visualization of anatomy and allows for precise identification of MPFL attachments and accurate placement of suture anchors for repair during surgery (Figures 3, 4).

For femur-side and patella-side tears, repairs can and should be performed. For midsubstance tears, however, repair is not feasible, and reconstruction is appropriate. MPFL repair is superior to reconstruction in several ways. Repair is a simple percutaneous procedure that had a mean tourniquet time of 26.9 minutes in this study. For tissue that is quantitatively and qualitatively adequate, repair allows the structure to reintegrate into bone without total reconstruction. In the event of multiple tears, the percutaneous procedure allows for repair of each attachment. As the MPFL sits between the second and third tissue layers of the medial knee, reconstruction can be difficult and invasive and require establishment of a between-layers plane, which can disrupt adjacent tissue.^4,7,17 Repair also maintains native tissue and its neurovascular and proprioceptive properties.

Reconstruction of the MPFL has become the gold-standard treatment for recurrent lateral patellar instability but has limitations and complications.^3,7,12,17 Reconstruction techniques use either surface anatomy palpation (requiring large incisions) or fluoroscopy to identify tunnel placement locations, and accurate placement has often been difficult and inconsistent. Our repair technique has several advantages over reconstruction. It does not burn any bridges; it allows for subsequent reconstruction. It does not require a graft and, using small suture anchors instead of large sockets and anchors, involves less bone loss. It also allows for early repair of tears—patients can return to activities, sports, and work quicker—and avoids the risk of chondral and bony damage with recurrent dislocations. According to our review of the MPFL repairs performed by Dr. Hirahara starting in 2013, the procedure is quick and successful and has outstanding outcomes.

Another treatment option for recurrent lateral patellar instability combines reefing of the medial patellofemoral tissues with a lateral release. This combination has had several postoperative complications and is no longer indicated.⁹ TT transfer and trochleoplasty procedures have been developed to address different aspects of patellar instability, increased TT-TG distance, and dysplastic trochlea (Table 2). Both types of procedures are highly invasive and difficult to perform, requiring technical expertise. They are best used when warranted by the anatomy, but this is uncommon. The technique we have presented allows for easy and reliable repair of dislocations in the absence of associated pathology that would require larger, more complex surgery. The ease of use and accuracy of musculoskeletal ultrasound make this technique superior to others.

Conclusion

The MPFL is a vital static stabilizer of the patella and as such should be evaluated in the setting of patellar injury. The novel preoperative and intraoperative use of musculoskeletal ultrasound described in this article allows for easy real-time identification of the MPFL and simple and accurate percutaneous repair of torn structures. Nonoperative treatments of acute patellar dislocations have higher rates of recurrent dislocations, which put patella and trochlea at risk for bony and chondral damage. Given appropriate tear location and tissue quality, repairs should be considered early and before reconstruction. To our knowledge, a reliable, easily reproducible MPFL repair was not described until now. We have reported on use of such a technique and on its promising patient outcomes, which should be considered when addressing MPFL injuries.

Am J Orthop. 2017;46(3):152-157. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

Take-Home Points

• Use ultrasound to identify integrity and location of MPFL tear.
• Anatomic repair allows native tissue to reintegrate into bone.
• Repairs done early can prevent complications of recurrent instability.
• Repair maintains biological and proprioceptive qualities of tissue.

• 10Ultrasound-guided percutaneous repair is quick and effective.

The medial patellofemoral ligament (MPFL) is the primary passive restraint to lateral patellar excursion^1-5 and helps control patellar tilt and rotation.^6,7 More than 90% of lateral patellar dislocations cause the MPFL to rupture, and roughly 90% of these detachments involve the femoral insertion.⁴ Ensuing patellar instability often results from MPFL insufficiency. It has been suggested that re-creating the anatomy and functionality of this ligament is of utmost importance in restoring normal patellar biomechanics.^1-5,7,8

Anatomical risk factors for recurrent patellar instability include patella alta, increased tibial tuberosity-trochlear groove (TT-TG) distance, trochlear dysplasia, and torsional abnormalities.^1-4,6 A medial reefing technique with a lateral tissue release traditionally was used to restore proper kinematics, but was shown to have associated postoperative issues.⁹

Methods

Patient Demographics

Dr. Hirahara developed this technique in 2013 and performed it 11 times between 2013 and 2016. Of the 11 patients, 1 was excluded from our retrospective analysis because of trochlear dysplasia, now considered a relative contraindication. Of the remaining 10 patients, 5 (50%) had the repair performed on the right knee. Eight patients (80%) were female. Mean (SD) age was 17.21 (3.53) years. One patient had concurrent femur- and patella-side detachments; otherwise, 6 (60%) of 10 repairs were performed exclusively at the patella. We grade patellar instability according to amount of glide based on patellar width and quadrants. Normal lateral displacement was usually 1 to 2 quadrants of lateral glide relative to the contralateral side. Before surgery, 6 (60%) of the 10 patients presented with lateral glide of 3 quadrants, and 3 (30%) presented with lateral glide of 4 quadrants. All had patellar instability apprehension on physical examination.

Surgical Indications

Before surgery, MPFL integrity is determined by ultrasound evaluation. Repair is considered if the MPFL has a femur- or patella-side tear and is of adequate quantity and quality, and if there are minimal or no arthritic changes (Table 2).

Surgical Technique

The patient is brought to the operating room and placed supine. Patellar stability of the affected knee is assessed and compared with that of the contralateral side with patellar glide. The knee is prepared and draped in usual sterile fashion. With the knee flexed at 90º, a tourniquet is inflated. Diagnostic arthroscopy is performed with standard anteromedial and anterolateral portals, and, if necessary, arthroscopic procedures are performed.

Femoral Attachment Repair

With the leg in extension, ultrasound is used to identify the tear at the femoral attachment (watch part 1 of the video). A spinal needle is placed at the femoral insertion, typically just anterior and distal to the adductor tubercle (Figure 4).¹⁰

Patellar Attachment Repair

With the leg in extension, ultrasound is used to identify where the MPFL is detached from the patella (watch part 2 of the video). A spinal needle is placed at the detachment site (Figure 5). A scalpel is used to make a 1-cm incision down to the patella.

In this description, we showcase knotless and knotted techniques for each repair site. Either method is appropriate for the 2 repair sites. Owing to the superficial nature of the attachment sites—they may have very little fat, particularly at the patella—knot stacks are more prominent, can be felt after surgery, and have the potential to irritate surrounding tissues. Therefore, we prefer knotless fixation for both sites.

Rehabilitation

Rehabilitation after MPFL repair is much like rehabilitation after quadriceps tendon repair. The patient is locked in a brace in full extension when up and moving. Early weight-bearing and minimal use of assistive devices (crutches) are allowed because, when the leg is in full extension, there is no tension at the repair sites. Rehabilitation begins within 1 week, and normal daily function is quickly attained. The protocol emphasizes pain-free motion and suitable patellar mobility, and allows the immobilizing brace to be unlocked for exercise and sitting. During the first 4 weeks, quadriceps activation is limited; progression to full ROM occurs by 4 to 6 weeks. During the strengthening phase, loading the knee in early flexion should be avoided. Return to heavy lifting, physical activity, and sports is delayed until after 6 months in order to allow the construct to mature and integrate. Once the patient has satisfied all the strength, ROM, and functional outcome measurements, a brace is no longer required during sports and normal activity.

Results

Mean tourniquet time for each procedure, which includes diagnostic arthroscopy and ultrasound-guided percutaneous repair, was 26.9 minutes.

Discussion

Conservative management typically is recommended for acute patellar dislocations. In the event of failed conservative management or chronic patellar instability, surgical intervention is indicated. Studies have found that conservative management has recurrent-dislocation rates of 35% at 3-year follow-up and 73% at 6-year follow-up, and recurrent dislocations significantly increase patients’ risk of developing chondral and bony damage.¹³ MPFL repair is designed to restore proper patellar tracking and kinematics while maintaining the anatomical tissue. Lateral patellar dislocations often cause the MPFL to rupture; tears are reported in more than 90% of incidents.⁴ The significant rate indicates that, even after a single patellar dislocation, the MPFL should be evaluated. The MPFL contributes 50% to 60% of the medial stabilizing force during patellar tracking^1,7,14 and is the primary restraint to lateral patellar excursion and excessive patellar tilt and rotation.^1-5 Its absence plays a key role in recurrent lateral patellar instability. With this structure being so important, proper identification and intervention are vital. Studies have established that redislocation rates are significantly higher for nonoperatively (vs operatively) treated primary patellar dislocations.¹³ Simple and accurate percutaneous repair of the MPFL should be performed early to avoid the long-term complications of recurrent instability that could damage the cartilage and bone of the patella and trochlea.

The primary advantage of this technique is its novel use of musculoskeletal ultrasound to accurately identify anatomy and pathology and the placement of anatomical repairs. Accurate preoperative and intraoperative assessment of MPFL anatomy is vital to the success of a procedure. Descriptions of MPFL anatomy suggest discrepancies in the exact locations of the femoral and patellar attachments.^{2,5,7,10,12,15,16} Tanaka⁵ noted that, even within paired knees, there was “marked variability” in the MPFL insertions. McCarthy and colleagues¹⁰ contended the femoral attachment of the MPFL is just anterior and distal to the adductor tubercle, the landmark addressed in this technique. Steensen and colleagues¹⁶ described this attachment site as being statistically the “single most important point affecting isometry” of the MPFL. Sallay and colleagues⁴ asserted that an overwhelming majority of MPFL tears (87%) occur at the adductor tubercle. The variable distribution of tear locations and the importance of re-creating patient anatomy further highlight the need for individualized treatment, which is afforded by ultrasound. Fluoroscopy has been inadequate in identifying MPFL anatomy; this modality is difficult, cumbersome, inaccurate, and inconsistent.^11,12 Conversely, ultrasound provides real-time visualization of anatomy and allows for precise identification of MPFL attachments and accurate placement of suture anchors for repair during surgery (Figures 3, 4).

For femur-side and patella-side tears, repairs can and should be performed. For midsubstance tears, however, repair is not feasible, and reconstruction is appropriate. MPFL repair is superior to reconstruction in several ways. Repair is a simple percutaneous procedure that had a mean tourniquet time of 26.9 minutes in this study. For tissue that is quantitatively and qualitatively adequate, repair allows the structure to reintegrate into bone without total reconstruction. In the event of multiple tears, the percutaneous procedure allows for repair of each attachment. As the MPFL sits between the second and third tissue layers of the medial knee, reconstruction can be difficult and invasive and require establishment of a between-layers plane, which can disrupt adjacent tissue.^4,7,17 Repair also maintains native tissue and its neurovascular and proprioceptive properties.

Reconstruction of the MPFL has become the gold-standard treatment for recurrent lateral patellar instability but has limitations and complications.^3,7,12,17 Reconstruction techniques use either surface anatomy palpation (requiring large incisions) or fluoroscopy to identify tunnel placement locations, and accurate placement has often been difficult and inconsistent. Our repair technique has several advantages over reconstruction. It does not burn any bridges; it allows for subsequent reconstruction. It does not require a graft and, using small suture anchors instead of large sockets and anchors, involves less bone loss. It also allows for early repair of tears—patients can return to activities, sports, and work quicker—and avoids the risk of chondral and bony damage with recurrent dislocations. According to our review of the MPFL repairs performed by Dr. Hirahara starting in 2013, the procedure is quick and successful and has outstanding outcomes.

Another treatment option for recurrent lateral patellar instability combines reefing of the medial patellofemoral tissues with a lateral release. This combination has had several postoperative complications and is no longer indicated.⁹ TT transfer and trochleoplasty procedures have been developed to address different aspects of patellar instability, increased TT-TG distance, and dysplastic trochlea (Table 2). Both types of procedures are highly invasive and difficult to perform, requiring technical expertise. They are best used when warranted by the anatomy, but this is uncommon. The technique we have presented allows for easy and reliable repair of dislocations in the absence of associated pathology that would require larger, more complex surgery. The ease of use and accuracy of musculoskeletal ultrasound make this technique superior to others.

Conclusion

The MPFL is a vital static stabilizer of the patella and as such should be evaluated in the setting of patellar injury. The novel preoperative and intraoperative use of musculoskeletal ultrasound described in this article allows for easy real-time identification of the MPFL and simple and accurate percutaneous repair of torn structures. Nonoperative treatments of acute patellar dislocations have higher rates of recurrent dislocations, which put patella and trochlea at risk for bony and chondral damage. Given appropriate tear location and tissue quality, repairs should be considered early and before reconstruction. To our knowledge, a reliable, easily reproducible MPFL repair was not described until now. We have reported on use of such a technique and on its promising patient outcomes, which should be considered when addressing MPFL injuries.

Am J Orthop. 2017;46(3):152-157. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

Take-Home Points

• Use ultrasound to identify integrity and location of MPFL tear.
• Anatomic repair allows native tissue to reintegrate into bone.
• Repairs done early can prevent complications of recurrent instability.
• Repair maintains biological and proprioceptive qualities of tissue.

• 10Ultrasound-guided percutaneous repair is quick and effective.

The medial patellofemoral ligament (MPFL) is the primary passive restraint to lateral patellar excursion^1-5 and helps control patellar tilt and rotation.^6,7 More than 90% of lateral patellar dislocations cause the MPFL to rupture, and roughly 90% of these detachments involve the femoral insertion.⁴ Ensuing patellar instability often results from MPFL insufficiency. It has been suggested that re-creating the anatomy and functionality of this ligament is of utmost importance in restoring normal patellar biomechanics.^1-5,7,8

Anatomical risk factors for recurrent patellar instability include patella alta, increased tibial tuberosity-trochlear groove (TT-TG) distance, trochlear dysplasia, and torsional abnormalities.^1-4,6 A medial reefing technique with a lateral tissue release traditionally was used to restore proper kinematics, but was shown to have associated postoperative issues.⁹

Methods

Patient Demographics

Dr. Hirahara developed this technique in 2013 and performed it 11 times between 2013 and 2016. Of the 11 patients, 1 was excluded from our retrospective analysis because of trochlear dysplasia, now considered a relative contraindication. Of the remaining 10 patients, 5 (50%) had the repair performed on the right knee. Eight patients (80%) were female. Mean (SD) age was 17.21 (3.53) years. One patient had concurrent femur- and patella-side detachments; otherwise, 6 (60%) of 10 repairs were performed exclusively at the patella. We grade patellar instability according to amount of glide based on patellar width and quadrants. Normal lateral displacement was usually 1 to 2 quadrants of lateral glide relative to the contralateral side. Before surgery, 6 (60%) of the 10 patients presented with lateral glide of 3 quadrants, and 3 (30%) presented with lateral glide of 4 quadrants. All had patellar instability apprehension on physical examination.

Surgical Indications

Before surgery, MPFL integrity is determined by ultrasound evaluation. Repair is considered if the MPFL has a femur- or patella-side tear and is of adequate quantity and quality, and if there are minimal or no arthritic changes (Table 2).

Surgical Technique

The patient is brought to the operating room and placed supine. Patellar stability of the affected knee is assessed and compared with that of the contralateral side with patellar glide. The knee is prepared and draped in usual sterile fashion. With the knee flexed at 90º, a tourniquet is inflated. Diagnostic arthroscopy is performed with standard anteromedial and anterolateral portals, and, if necessary, arthroscopic procedures are performed.

Femoral Attachment Repair

With the leg in extension, ultrasound is used to identify the tear at the femoral attachment (watch part 1 of the video). A spinal needle is placed at the femoral insertion, typically just anterior and distal to the adductor tubercle (Figure 4).¹⁰

Patellar Attachment Repair

With the leg in extension, ultrasound is used to identify where the MPFL is detached from the patella (watch part 2 of the video). A spinal needle is placed at the detachment site (Figure 5). A scalpel is used to make a 1-cm incision down to the patella.

In this description, we showcase knotless and knotted techniques for each repair site. Either method is appropriate for the 2 repair sites. Owing to the superficial nature of the attachment sites—they may have very little fat, particularly at the patella—knot stacks are more prominent, can be felt after surgery, and have the potential to irritate surrounding tissues. Therefore, we prefer knotless fixation for both sites.

Rehabilitation

Rehabilitation after MPFL repair is much like rehabilitation after quadriceps tendon repair. The patient is locked in a brace in full extension when up and moving. Early weight-bearing and minimal use of assistive devices (crutches) are allowed because, when the leg is in full extension, there is no tension at the repair sites. Rehabilitation begins within 1 week, and normal daily function is quickly attained. The protocol emphasizes pain-free motion and suitable patellar mobility, and allows the immobilizing brace to be unlocked for exercise and sitting. During the first 4 weeks, quadriceps activation is limited; progression to full ROM occurs by 4 to 6 weeks. During the strengthening phase, loading the knee in early flexion should be avoided. Return to heavy lifting, physical activity, and sports is delayed until after 6 months in order to allow the construct to mature and integrate. Once the patient has satisfied all the strength, ROM, and functional outcome measurements, a brace is no longer required during sports and normal activity.

Results

Mean tourniquet time for each procedure, which includes diagnostic arthroscopy and ultrasound-guided percutaneous repair, was 26.9 minutes.

Discussion

Conservative management typically is recommended for acute patellar dislocations. In the event of failed conservative management or chronic patellar instability, surgical intervention is indicated. Studies have found that conservative management has recurrent-dislocation rates of 35% at 3-year follow-up and 73% at 6-year follow-up, and recurrent dislocations significantly increase patients’ risk of developing chondral and bony damage.¹³ MPFL repair is designed to restore proper patellar tracking and kinematics while maintaining the anatomical tissue. Lateral patellar dislocations often cause the MPFL to rupture; tears are reported in more than 90% of incidents.⁴ The significant rate indicates that, even after a single patellar dislocation, the MPFL should be evaluated. The MPFL contributes 50% to 60% of the medial stabilizing force during patellar tracking^1,7,14 and is the primary restraint to lateral patellar excursion and excessive patellar tilt and rotation.^1-5 Its absence plays a key role in recurrent lateral patellar instability. With this structure being so important, proper identification and intervention are vital. Studies have established that redislocation rates are significantly higher for nonoperatively (vs operatively) treated primary patellar dislocations.¹³ Simple and accurate percutaneous repair of the MPFL should be performed early to avoid the long-term complications of recurrent instability that could damage the cartilage and bone of the patella and trochlea.

The primary advantage of this technique is its novel use of musculoskeletal ultrasound to accurately identify anatomy and pathology and the placement of anatomical repairs. Accurate preoperative and intraoperative assessment of MPFL anatomy is vital to the success of a procedure. Descriptions of MPFL anatomy suggest discrepancies in the exact locations of the femoral and patellar attachments.^{2,5,7,10,12,15,16} Tanaka⁵ noted that, even within paired knees, there was “marked variability” in the MPFL insertions. McCarthy and colleagues¹⁰ contended the femoral attachment of the MPFL is just anterior and distal to the adductor tubercle, the landmark addressed in this technique. Steensen and colleagues¹⁶ described this attachment site as being statistically the “single most important point affecting isometry” of the MPFL. Sallay and colleagues⁴ asserted that an overwhelming majority of MPFL tears (87%) occur at the adductor tubercle. The variable distribution of tear locations and the importance of re-creating patient anatomy further highlight the need for individualized treatment, which is afforded by ultrasound. Fluoroscopy has been inadequate in identifying MPFL anatomy; this modality is difficult, cumbersome, inaccurate, and inconsistent.^11,12 Conversely, ultrasound provides real-time visualization of anatomy and allows for precise identification of MPFL attachments and accurate placement of suture anchors for repair during surgery (Figures 3, 4).

For femur-side and patella-side tears, repairs can and should be performed. For midsubstance tears, however, repair is not feasible, and reconstruction is appropriate. MPFL repair is superior to reconstruction in several ways. Repair is a simple percutaneous procedure that had a mean tourniquet time of 26.9 minutes in this study. For tissue that is quantitatively and qualitatively adequate, repair allows the structure to reintegrate into bone without total reconstruction. In the event of multiple tears, the percutaneous procedure allows for repair of each attachment. As the MPFL sits between the second and third tissue layers of the medial knee, reconstruction can be difficult and invasive and require establishment of a between-layers plane, which can disrupt adjacent tissue.^4,7,17 Repair also maintains native tissue and its neurovascular and proprioceptive properties.

Reconstruction of the MPFL has become the gold-standard treatment for recurrent lateral patellar instability but has limitations and complications.^3,7,12,17 Reconstruction techniques use either surface anatomy palpation (requiring large incisions) or fluoroscopy to identify tunnel placement locations, and accurate placement has often been difficult and inconsistent. Our repair technique has several advantages over reconstruction. It does not burn any bridges; it allows for subsequent reconstruction. It does not require a graft and, using small suture anchors instead of large sockets and anchors, involves less bone loss. It also allows for early repair of tears—patients can return to activities, sports, and work quicker—and avoids the risk of chondral and bony damage with recurrent dislocations. According to our review of the MPFL repairs performed by Dr. Hirahara starting in 2013, the procedure is quick and successful and has outstanding outcomes.

Another treatment option for recurrent lateral patellar instability combines reefing of the medial patellofemoral tissues with a lateral release. This combination has had several postoperative complications and is no longer indicated.⁹ TT transfer and trochleoplasty procedures have been developed to address different aspects of patellar instability, increased TT-TG distance, and dysplastic trochlea (Table 2). Both types of procedures are highly invasive and difficult to perform, requiring technical expertise. They are best used when warranted by the anatomy, but this is uncommon. The technique we have presented allows for easy and reliable repair of dislocations in the absence of associated pathology that would require larger, more complex surgery. The ease of use and accuracy of musculoskeletal ultrasound make this technique superior to others.

Conclusion

The MPFL is a vital static stabilizer of the patella and as such should be evaluated in the setting of patellar injury. The novel preoperative and intraoperative use of musculoskeletal ultrasound described in this article allows for easy real-time identification of the MPFL and simple and accurate percutaneous repair of torn structures. Nonoperative treatments of acute patellar dislocations have higher rates of recurrent dislocations, which put patella and trochlea at risk for bony and chondral damage. Given appropriate tear location and tissue quality, repairs should be considered early and before reconstruction. To our knowledge, a reliable, easily reproducible MPFL repair was not described until now. We have reported on use of such a technique and on its promising patient outcomes, which should be considered when addressing MPFL injuries.

Am J Orthop. 2017;46(3):152-157. Copyright Frontline Medical Communications Inc. 2017. All rights reserved.

A 69-year-old woman presented with fatigue, cough, and lightheadedness. She had a history of atrial fibrillation and complete heart block, for which she had a pacemaker (dual-pacing, dual-sensing, dual-response, and rate-adaptive mode) inserted in 2005. Her heart rate was 30 beats per minute.

A chest radiograph showed a fractured right ventricular pacemaker lead (Figure 1). Electrocardiography showed sinus rhythm with a high-grade atrioventricular block (Figure 2). Pacemaker interrogation confirmed the diagnosis of lead fracture. A new lead was placed, and the old lead was abandoned.

HOW LEADS BREAK

The rate of lead fracture ranges from 0.1% to 4.2% per patient-year, and the annual failure rate increases progressively with time after implantation.^1,2

Extrinsic pressure on the lead can eventually break it. This can happen between the first rib and clavicle, in “subclavian crush” injury, or with any anatomical abnormality that narrows the thoracic outlet. Typically, classic subclavian crush results from entrapment of the pacemaker leads by the subclavius muscle or the costoclavicular ligament as the lead follows the needle course of the antecedent access puncture of the subclavian vein. This results in intermittent flexing of the lead and potential lead fracture³ and was likely the cause of lead fracture in our patient.

The risk of fracture is higher in patients under the age of 50, those who perform intense physical activity, women, and patients with greater left ventricular ejection fraction.^4,5 Certain leads are prone to fracture due to design flaws. One of these was the Medtronic Sprint Fidelis cardioverter defibrillator lead, which was recalled in 2007.⁵

DETECTING LEAD FRACTURE

Symptoms of lead fracture vary, depending on the patient’s pacemaker-dependency and on the degree of loss of capture (ie, the degree to which the heart fails to respond to the pacemaker’s signals), and may include lightheadedness, syncope, and extracardiac stimulation.

The electrical integrity of a lead can be tested by measuring the circuit impedance, which normally ranges from 300 to 1,000 ohms.⁶ An insulation failure results in very low impedance, while a disrupted circuit due to lead fracture commonly causes a sudden rather than gradual increase in impedance.⁶

Simple imaging studies such as chest radiography or fluoroscopy may establish the diagnosis of lead fracture. One should carefully trace every lead along its entire course and look for any conductor discontinuity, kinks, or sharp bends.⁶

REMOVE THE OLD LEAD, OR LEAVE IT IN PLACE?

The treatment for lead fracture is usually to put in a new lead, with or without extracting the old one.

In view of the potential complications of lead removal such as cardiac perforation or vascular tear, lead abandonment with placement of a new lead may be performed.⁷ There are no controlled clinical studies comparing lead abandonment vs lead extraction.⁸ However, extraction is currently recommended only in patients in whom the old lead causes life-threatening arrhythmias, interferes with the operation of implanted cardiac devices, interferes with radiation therapy or needed surgery, or, due to its design or failure, poses an immediate threat to the patient if left in place.⁷ Lead removal is reasonable in patients who require specific imaging studies such as magnetic resonance imaging with no available imaging alternative for the diagnosis.⁷

In our patient, a new lead was placed without removing the fractured lead, with no complications. Afterward, the patient’s heart rhythm was observed to be appropriately paced, and she was discharged home the following day.

A 69-year-old woman presented with fatigue, cough, and lightheadedness. She had a history of atrial fibrillation and complete heart block, for which she had a pacemaker (dual-pacing, dual-sensing, dual-response, and rate-adaptive mode) inserted in 2005. Her heart rate was 30 beats per minute.

A chest radiograph showed a fractured right ventricular pacemaker lead (Figure 1). Electrocardiography showed sinus rhythm with a high-grade atrioventricular block (Figure 2). Pacemaker interrogation confirmed the diagnosis of lead fracture. A new lead was placed, and the old lead was abandoned.

HOW LEADS BREAK

The rate of lead fracture ranges from 0.1% to 4.2% per patient-year, and the annual failure rate increases progressively with time after implantation.^1,2

Extrinsic pressure on the lead can eventually break it. This can happen between the first rib and clavicle, in “subclavian crush” injury, or with any anatomical abnormality that narrows the thoracic outlet. Typically, classic subclavian crush results from entrapment of the pacemaker leads by the subclavius muscle or the costoclavicular ligament as the lead follows the needle course of the antecedent access puncture of the subclavian vein. This results in intermittent flexing of the lead and potential lead fracture³ and was likely the cause of lead fracture in our patient.

The risk of fracture is higher in patients under the age of 50, those who perform intense physical activity, women, and patients with greater left ventricular ejection fraction.^4,5 Certain leads are prone to fracture due to design flaws. One of these was the Medtronic Sprint Fidelis cardioverter defibrillator lead, which was recalled in 2007.⁵

DETECTING LEAD FRACTURE

Symptoms of lead fracture vary, depending on the patient’s pacemaker-dependency and on the degree of loss of capture (ie, the degree to which the heart fails to respond to the pacemaker’s signals), and may include lightheadedness, syncope, and extracardiac stimulation.

The electrical integrity of a lead can be tested by measuring the circuit impedance, which normally ranges from 300 to 1,000 ohms.⁶ An insulation failure results in very low impedance, while a disrupted circuit due to lead fracture commonly causes a sudden rather than gradual increase in impedance.⁶

Simple imaging studies such as chest radiography or fluoroscopy may establish the diagnosis of lead fracture. One should carefully trace every lead along its entire course and look for any conductor discontinuity, kinks, or sharp bends.⁶

REMOVE THE OLD LEAD, OR LEAVE IT IN PLACE?

The treatment for lead fracture is usually to put in a new lead, with or without extracting the old one.

In view of the potential complications of lead removal such as cardiac perforation or vascular tear, lead abandonment with placement of a new lead may be performed.⁷ There are no controlled clinical studies comparing lead abandonment vs lead extraction.⁸ However, extraction is currently recommended only in patients in whom the old lead causes life-threatening arrhythmias, interferes with the operation of implanted cardiac devices, interferes with radiation therapy or needed surgery, or, due to its design or failure, poses an immediate threat to the patient if left in place.⁷ Lead removal is reasonable in patients who require specific imaging studies such as magnetic resonance imaging with no available imaging alternative for the diagnosis.⁷

In our patient, a new lead was placed without removing the fractured lead, with no complications. Afterward, the patient’s heart rhythm was observed to be appropriately paced, and she was discharged home the following day.

A 69-year-old woman presented with fatigue, cough, and lightheadedness. She had a history of atrial fibrillation and complete heart block, for which she had a pacemaker (dual-pacing, dual-sensing, dual-response, and rate-adaptive mode) inserted in 2005. Her heart rate was 30 beats per minute.

A chest radiograph showed a fractured right ventricular pacemaker lead (Figure 1). Electrocardiography showed sinus rhythm with a high-grade atrioventricular block (Figure 2). Pacemaker interrogation confirmed the diagnosis of lead fracture. A new lead was placed, and the old lead was abandoned.

HOW LEADS BREAK

The rate of lead fracture ranges from 0.1% to 4.2% per patient-year, and the annual failure rate increases progressively with time after implantation.^1,2

Extrinsic pressure on the lead can eventually break it. This can happen between the first rib and clavicle, in “subclavian crush” injury, or with any anatomical abnormality that narrows the thoracic outlet. Typically, classic subclavian crush results from entrapment of the pacemaker leads by the subclavius muscle or the costoclavicular ligament as the lead follows the needle course of the antecedent access puncture of the subclavian vein. This results in intermittent flexing of the lead and potential lead fracture³ and was likely the cause of lead fracture in our patient.

The risk of fracture is higher in patients under the age of 50, those who perform intense physical activity, women, and patients with greater left ventricular ejection fraction.^4,5 Certain leads are prone to fracture due to design flaws. One of these was the Medtronic Sprint Fidelis cardioverter defibrillator lead, which was recalled in 2007.⁵

DETECTING LEAD FRACTURE

Symptoms of lead fracture vary, depending on the patient’s pacemaker-dependency and on the degree of loss of capture (ie, the degree to which the heart fails to respond to the pacemaker’s signals), and may include lightheadedness, syncope, and extracardiac stimulation.

The electrical integrity of a lead can be tested by measuring the circuit impedance, which normally ranges from 300 to 1,000 ohms.⁶ An insulation failure results in very low impedance, while a disrupted circuit due to lead fracture commonly causes a sudden rather than gradual increase in impedance.⁶

Simple imaging studies such as chest radiography or fluoroscopy may establish the diagnosis of lead fracture. One should carefully trace every lead along its entire course and look for any conductor discontinuity, kinks, or sharp bends.⁶

REMOVE THE OLD LEAD, OR LEAVE IT IN PLACE?

The treatment for lead fracture is usually to put in a new lead, with or without extracting the old one.

In view of the potential complications of lead removal such as cardiac perforation or vascular tear, lead abandonment with placement of a new lead may be performed.⁷ There are no controlled clinical studies comparing lead abandonment vs lead extraction.⁸ However, extraction is currently recommended only in patients in whom the old lead causes life-threatening arrhythmias, interferes with the operation of implanted cardiac devices, interferes with radiation therapy or needed surgery, or, due to its design or failure, poses an immediate threat to the patient if left in place.⁷ Lead removal is reasonable in patients who require specific imaging studies such as magnetic resonance imaging with no available imaging alternative for the diagnosis.⁷

In our patient, a new lead was placed without removing the fractured lead, with no complications. Afterward, the patient’s heart rhythm was observed to be appropriately paced, and she was discharged home the following day.

We routinely use low-volume (2-L) polyethylene glycol electrolyte preparations such as Moviprep and Miralax in split-dose regimens, in which patients drink half of the preparation the day before the procedure and the other half the day of the procedure. In our experience, these are well tolerated by patients with a history of bariatric surgery and provide adequate colon cleansing before colonoscopy.

RATIONALE

Adequate bowel preparation by the ingestion of a cleansing agent is extremely important before colonoscopy: the quality of colon preparation affects the diagnostic accuracy and safety of the procedure, as inadequate bowel preparation has been associated with failure to detect polyps and with a higher rate of adverse events during the procedure.^1–3

The most commonly used bowel preparations can be divided into high-volume (which require drinking at least 4 L of a cathartic solution) and low-volume (which require drinking about 2 L).⁴ Polyethylene glycol electrolyte solutions are among the most commonly used and are available in both high-volume (eg, Golytely, Nulytely) and low-volume (eg, Moviprep, Miralax) forms.

Other low-volume preparations include sodium picosulfate (Prepopik), magnesium citrate, and sodium phosphate tablets. However, these should be avoided in patients with renal insufficiency.⁴

Prices for bowel preparations vary. For example, the average reported wholesale price of Golytely is $24.56, Moviprep $81.17, and Miralax $10.08.⁴ However, the final cost depends on the patient’s insurance coverage. Generic formulations are available for some preparations.

After bariatric surgery, patients have a smaller stomach

After bariatric surgery, patients have significantly reduced stomach volume, due either to resection of a part of the stomach (such as in partial gastrectomy) or to diversion of the gastrointestinal (GI) tract to bypass most of the stomach (such as in Roux-en-Y gastric bypass). This causes early satiety with smaller amounts of food and leads to weight loss. However, this restriction in stomach volume also makes it more difficult for the patient to tolerate the intake of large volumes of fluids for bowel cleansing before colonoscopy.

Bariatric surgery patients may require colonoscopy for indications such as colorectal cancer screening, chronic diarrhea, or GI bleeding, all of which are commonly encountered during routine clinical practice.

Guidelines are available

Currently, there are no published data available to support the use of one preparation over another in patients with a history of bariatric surgery. However, for patients who have had bariatric surgery, guidelines from the US Multi-Society Task Force on Colorectal Cancer—endorsed by the three major American gastroenterology societies, ie, the American Gastroenterological Association, the American College of Gastroenterology, and the American Society for Gastrointestinal Endoscopy—recommend either a low-volume solution or, if a high-volume solution is used, extending the duration over which the preparation is consumed.⁵ In addition, it is recommended that patients consume sugar-free drinks and liquids to avoid dumping syndrome from high sugar content.⁶

The use of split-dose regimens is also strongly recommended for elective colonoscopy by the US Multi-Society Task Force on Colorectal Cancer.⁵

Our clinical experience has been in line with the above recommendations.

We routinely use low-volume (2-L) polyethylene glycol electrolyte preparations such as Moviprep and Miralax in split-dose regimens, in which patients drink half of the preparation the day before the procedure and the other half the day of the procedure. In our experience, these are well tolerated by patients with a history of bariatric surgery and provide adequate colon cleansing before colonoscopy.

RATIONALE

Adequate bowel preparation by the ingestion of a cleansing agent is extremely important before colonoscopy: the quality of colon preparation affects the diagnostic accuracy and safety of the procedure, as inadequate bowel preparation has been associated with failure to detect polyps and with a higher rate of adverse events during the procedure.^1–3

The most commonly used bowel preparations can be divided into high-volume (which require drinking at least 4 L of a cathartic solution) and low-volume (which require drinking about 2 L).⁴ Polyethylene glycol electrolyte solutions are among the most commonly used and are available in both high-volume (eg, Golytely, Nulytely) and low-volume (eg, Moviprep, Miralax) forms.

Other low-volume preparations include sodium picosulfate (Prepopik), magnesium citrate, and sodium phosphate tablets. However, these should be avoided in patients with renal insufficiency.⁴

Prices for bowel preparations vary. For example, the average reported wholesale price of Golytely is $24.56, Moviprep $81.17, and Miralax $10.08.⁴ However, the final cost depends on the patient’s insurance coverage. Generic formulations are available for some preparations.

After bariatric surgery, patients have a smaller stomach

After bariatric surgery, patients have significantly reduced stomach volume, due either to resection of a part of the stomach (such as in partial gastrectomy) or to diversion of the gastrointestinal (GI) tract to bypass most of the stomach (such as in Roux-en-Y gastric bypass). This causes early satiety with smaller amounts of food and leads to weight loss. However, this restriction in stomach volume also makes it more difficult for the patient to tolerate the intake of large volumes of fluids for bowel cleansing before colonoscopy.

Bariatric surgery patients may require colonoscopy for indications such as colorectal cancer screening, chronic diarrhea, or GI bleeding, all of which are commonly encountered during routine clinical practice.

Guidelines are available

Currently, there are no published data available to support the use of one preparation over another in patients with a history of bariatric surgery. However, for patients who have had bariatric surgery, guidelines from the US Multi-Society Task Force on Colorectal Cancer—endorsed by the three major American gastroenterology societies, ie, the American Gastroenterological Association, the American College of Gastroenterology, and the American Society for Gastrointestinal Endoscopy—recommend either a low-volume solution or, if a high-volume solution is used, extending the duration over which the preparation is consumed.⁵ In addition, it is recommended that patients consume sugar-free drinks and liquids to avoid dumping syndrome from high sugar content.⁶

The use of split-dose regimens is also strongly recommended for elective colonoscopy by the US Multi-Society Task Force on Colorectal Cancer.⁵

Our clinical experience has been in line with the above recommendations.

We routinely use low-volume (2-L) polyethylene glycol electrolyte preparations such as Moviprep and Miralax in split-dose regimens, in which patients drink half of the preparation the day before the procedure and the other half the day of the procedure. In our experience, these are well tolerated by patients with a history of bariatric surgery and provide adequate colon cleansing before colonoscopy.

RATIONALE

Adequate bowel preparation by the ingestion of a cleansing agent is extremely important before colonoscopy: the quality of colon preparation affects the diagnostic accuracy and safety of the procedure, as inadequate bowel preparation has been associated with failure to detect polyps and with a higher rate of adverse events during the procedure.^1–3

The most commonly used bowel preparations can be divided into high-volume (which require drinking at least 4 L of a cathartic solution) and low-volume (which require drinking about 2 L).⁴ Polyethylene glycol electrolyte solutions are among the most commonly used and are available in both high-volume (eg, Golytely, Nulytely) and low-volume (eg, Moviprep, Miralax) forms.

Other low-volume preparations include sodium picosulfate (Prepopik), magnesium citrate, and sodium phosphate tablets. However, these should be avoided in patients with renal insufficiency.⁴

Prices for bowel preparations vary. For example, the average reported wholesale price of Golytely is $24.56, Moviprep $81.17, and Miralax $10.08.⁴ However, the final cost depends on the patient’s insurance coverage. Generic formulations are available for some preparations.

After bariatric surgery, patients have a smaller stomach

After bariatric surgery, patients have significantly reduced stomach volume, due either to resection of a part of the stomach (such as in partial gastrectomy) or to diversion of the gastrointestinal (GI) tract to bypass most of the stomach (such as in Roux-en-Y gastric bypass). This causes early satiety with smaller amounts of food and leads to weight loss. However, this restriction in stomach volume also makes it more difficult for the patient to tolerate the intake of large volumes of fluids for bowel cleansing before colonoscopy.

Bariatric surgery patients may require colonoscopy for indications such as colorectal cancer screening, chronic diarrhea, or GI bleeding, all of which are commonly encountered during routine clinical practice.

Guidelines are available

Currently, there are no published data available to support the use of one preparation over another in patients with a history of bariatric surgery. However, for patients who have had bariatric surgery, guidelines from the US Multi-Society Task Force on Colorectal Cancer—endorsed by the three major American gastroenterology societies, ie, the American Gastroenterological Association, the American College of Gastroenterology, and the American Society for Gastrointestinal Endoscopy—recommend either a low-volume solution or, if a high-volume solution is used, extending the duration over which the preparation is consumed.⁵ In addition, it is recommended that patients consume sugar-free drinks and liquids to avoid dumping syndrome from high sugar content.⁶

The use of split-dose regimens is also strongly recommended for elective colonoscopy by the US Multi-Society Task Force on Colorectal Cancer.⁵

Our clinical experience has been in line with the above recommendations.

A 48-year-old Nicaraguan man underwent mag netic resonance imaging (MRI) 3 days after admission to a South Florida hospital for treatment of cellulitis of the right thigh with vancomycin. MRI had been ordered to evaluate for a possible drainable source of infection, as the clinical picture and duration of illness was worsening and longer than expected for typical uncomplicated cellulitis despite intravenous antibiotic therapy. The MRI showed multiple enlarged inguinal lymph nodes and cellulitis of the superficial soft tissues of the thigh without discrete drainable collections.

After the procedure, the patient was noted to have a bullous lesion on each thigh, each lesion roughly 1 cm in diameter and filled with clear serous fluid (Figure 1). He reported that his legs had been pressed together before entering the MRI machine and that he had felt a burning sensation in both thighs during the test. Examination confirmed that the lesions indeed aligned with each other when he pressed his thighs together.

Study of a biopsy of one of the lesions revealed subepidermal cell blisters with focal epidermal necrosis and coagulative changes in the superficial dermis, consistent with a thermal injury (Figure 2).

HOW BURNS CAN OCCUR DURING MRI

Thermal burns are a potential cause of injury during MRI. Most have been observed in patients connected to external metal-containing monitoring devices, such as electrocardiogram leads and pulse oximeters.^1,2 Thermal burns  in patients unconnected to external devices have occurred when the patient’s body was touching radiofrequency coils³ or, as with our patient, when skin touches skin.^2,4,5 Skin-to-skin contact during MRI can cause the scanner to emit high-power electromagnetic radiofrequency pulses that are conducted through the body, creating heat. Tissue loops are created at points of skin-to-skin contact, thus forming a closed conducting circuit. The current flowing through this circuit can produce second-degree burns.⁴

We believe that during placement in the scanner, our patient inadvertently moved his thighs together, forming a closed loop conduction circuit and resulting in a thermal burn.

This case illustrates the importance of correct positioning during MRI. The MRI technicians had taken standard precautions, placing a sheet over the patient and ensuring no direct contact with the radiofrequency transmitter receiver (MR unit). However, precautions against skin-to-skin contact were not taken, and the patient’s legs were not separated.

LESSONS LEARNED

Appropriate positioning prevents closed skin-to-skin loops, but this may be more challenging in a larger patient.⁶ While the patient is in the scanner, a “squeeze ball” alert system allows the patient to signal the technologist should unexpected distress or heating occur.^6,7 A patient’s inability to utilize the squeeze ball contributes to the risk of a severe injury. Further, in this instance, there may have been a language barrier. Also, proximity of the anatomic skin-to-skin loop to the imaged body part (and therefore the center of the MRI coil) may have conferred risk related to field strength in this patient.

The MRI technologist, under supervision of a radiologist, is primarily responsible for positioning the patient to decrease the risk of this complication and for following institutional MRI safety protocols and professional guidelines.^6–8 MRI technologist certification training highlights all aspects of safety, including skin-to-skin conducting loop prevention.^7,8

A 48-year-old Nicaraguan man underwent mag netic resonance imaging (MRI) 3 days after admission to a South Florida hospital for treatment of cellulitis of the right thigh with vancomycin. MRI had been ordered to evaluate for a possible drainable source of infection, as the clinical picture and duration of illness was worsening and longer than expected for typical uncomplicated cellulitis despite intravenous antibiotic therapy. The MRI showed multiple enlarged inguinal lymph nodes and cellulitis of the superficial soft tissues of the thigh without discrete drainable collections.

After the procedure, the patient was noted to have a bullous lesion on each thigh, each lesion roughly 1 cm in diameter and filled with clear serous fluid (Figure 1). He reported that his legs had been pressed together before entering the MRI machine and that he had felt a burning sensation in both thighs during the test. Examination confirmed that the lesions indeed aligned with each other when he pressed his thighs together.

Study of a biopsy of one of the lesions revealed subepidermal cell blisters with focal epidermal necrosis and coagulative changes in the superficial dermis, consistent with a thermal injury (Figure 2).

HOW BURNS CAN OCCUR DURING MRI

Thermal burns are a potential cause of injury during MRI. Most have been observed in patients connected to external metal-containing monitoring devices, such as electrocardiogram leads and pulse oximeters.^1,2 Thermal burns  in patients unconnected to external devices have occurred when the patient’s body was touching radiofrequency coils³ or, as with our patient, when skin touches skin.^2,4,5 Skin-to-skin contact during MRI can cause the scanner to emit high-power electromagnetic radiofrequency pulses that are conducted through the body, creating heat. Tissue loops are created at points of skin-to-skin contact, thus forming a closed conducting circuit. The current flowing through this circuit can produce second-degree burns.⁴

We believe that during placement in the scanner, our patient inadvertently moved his thighs together, forming a closed loop conduction circuit and resulting in a thermal burn.

This case illustrates the importance of correct positioning during MRI. The MRI technicians had taken standard precautions, placing a sheet over the patient and ensuring no direct contact with the radiofrequency transmitter receiver (MR unit). However, precautions against skin-to-skin contact were not taken, and the patient’s legs were not separated.

LESSONS LEARNED

Appropriate positioning prevents closed skin-to-skin loops, but this may be more challenging in a larger patient.⁶ While the patient is in the scanner, a “squeeze ball” alert system allows the patient to signal the technologist should unexpected distress or heating occur.^6,7 A patient’s inability to utilize the squeeze ball contributes to the risk of a severe injury. Further, in this instance, there may have been a language barrier. Also, proximity of the anatomic skin-to-skin loop to the imaged body part (and therefore the center of the MRI coil) may have conferred risk related to field strength in this patient.

The MRI technologist, under supervision of a radiologist, is primarily responsible for positioning the patient to decrease the risk of this complication and for following institutional MRI safety protocols and professional guidelines.^6–8 MRI technologist certification training highlights all aspects of safety, including skin-to-skin conducting loop prevention.^7,8

A 48-year-old Nicaraguan man underwent mag netic resonance imaging (MRI) 3 days after admission to a South Florida hospital for treatment of cellulitis of the right thigh with vancomycin. MRI had been ordered to evaluate for a possible drainable source of infection, as the clinical picture and duration of illness was worsening and longer than expected for typical uncomplicated cellulitis despite intravenous antibiotic therapy. The MRI showed multiple enlarged inguinal lymph nodes and cellulitis of the superficial soft tissues of the thigh without discrete drainable collections.

After the procedure, the patient was noted to have a bullous lesion on each thigh, each lesion roughly 1 cm in diameter and filled with clear serous fluid (Figure 1). He reported that his legs had been pressed together before entering the MRI machine and that he had felt a burning sensation in both thighs during the test. Examination confirmed that the lesions indeed aligned with each other when he pressed his thighs together.

Study of a biopsy of one of the lesions revealed subepidermal cell blisters with focal epidermal necrosis and coagulative changes in the superficial dermis, consistent with a thermal injury (Figure 2).

HOW BURNS CAN OCCUR DURING MRI

Thermal burns are a potential cause of injury during MRI. Most have been observed in patients connected to external metal-containing monitoring devices, such as electrocardiogram leads and pulse oximeters.^1,2 Thermal burns  in patients unconnected to external devices have occurred when the patient’s body was touching radiofrequency coils³ or, as with our patient, when skin touches skin.^2,4,5 Skin-to-skin contact during MRI can cause the scanner to emit high-power electromagnetic radiofrequency pulses that are conducted through the body, creating heat. Tissue loops are created at points of skin-to-skin contact, thus forming a closed conducting circuit. The current flowing through this circuit can produce second-degree burns.⁴

We believe that during placement in the scanner, our patient inadvertently moved his thighs together, forming a closed loop conduction circuit and resulting in a thermal burn.

This case illustrates the importance of correct positioning during MRI. The MRI technicians had taken standard precautions, placing a sheet over the patient and ensuring no direct contact with the radiofrequency transmitter receiver (MR unit). However, precautions against skin-to-skin contact were not taken, and the patient’s legs were not separated.

LESSONS LEARNED

Appropriate positioning prevents closed skin-to-skin loops, but this may be more challenging in a larger patient.⁶ While the patient is in the scanner, a “squeeze ball” alert system allows the patient to signal the technologist should unexpected distress or heating occur.^6,7 A patient’s inability to utilize the squeeze ball contributes to the risk of a severe injury. Further, in this instance, there may have been a language barrier. Also, proximity of the anatomic skin-to-skin loop to the imaged body part (and therefore the center of the MRI coil) may have conferred risk related to field strength in this patient.

The MRI technologist, under supervision of a radiologist, is primarily responsible for positioning the patient to decrease the risk of this complication and for following institutional MRI safety protocols and professional guidelines.^6–8 MRI technologist certification training highlights all aspects of safety, including skin-to-skin conducting loop prevention.^7,8

SNOWMASS, COLO. – Mounting evidence attests to the value of noninvasive measurement of coronary flow reserve as a means of classifying cardiovascular risk in patients with stable coronary artery disease (CAD) more accurately than is possible via coronary angiography or measurement of fractional flow reserve, Marcelo F. Di Carli, MD, reported at the Annual Cardiovascular Conference at Snowmass.

“We use CFR [coronary flow reserve] as a way to exclude coronary disease. It’s a good practical measure of multivessel ischemic CAD. When the CFR is normal, you can with high confidence exclude the possibility of high-risk CAD,” according to Dr. Di Carli, executive director of the cardiovascular imaging program and chief of the division of nuclear medicine and molecular imaging at Brigham and Women’s Hospital, Boston.

Bruce Jancin/Frontline Medical News

[email protected]

SNOWMASS, COLO. – Mounting evidence attests to the value of noninvasive measurement of coronary flow reserve as a means of classifying cardiovascular risk in patients with stable coronary artery disease (CAD) more accurately than is possible via coronary angiography or measurement of fractional flow reserve, Marcelo F. Di Carli, MD, reported at the Annual Cardiovascular Conference at Snowmass.

“We use CFR [coronary flow reserve] as a way to exclude coronary disease. It’s a good practical measure of multivessel ischemic CAD. When the CFR is normal, you can with high confidence exclude the possibility of high-risk CAD,” according to Dr. Di Carli, executive director of the cardiovascular imaging program and chief of the division of nuclear medicine and molecular imaging at Brigham and Women’s Hospital, Boston.

Bruce Jancin/Frontline Medical News

[email protected]

SNOWMASS, COLO. – Mounting evidence attests to the value of noninvasive measurement of coronary flow reserve as a means of classifying cardiovascular risk in patients with stable coronary artery disease (CAD) more accurately than is possible via coronary angiography or measurement of fractional flow reserve, Marcelo F. Di Carli, MD, reported at the Annual Cardiovascular Conference at Snowmass.

“We use CFR [coronary flow reserve] as a way to exclude coronary disease. It’s a good practical measure of multivessel ischemic CAD. When the CFR is normal, you can with high confidence exclude the possibility of high-risk CAD,” according to Dr. Di Carli, executive director of the cardiovascular imaging program and chief of the division of nuclear medicine and molecular imaging at Brigham and Women’s Hospital, Boston.

Bruce Jancin/Frontline Medical News

[email protected]

“... with the rapid extension of laboratory tests of greater accuracy, there is a tendency for some clinicians and hence for some students in reaching a diagnosis to rely more on laboratory reports and less on the history of the illness, the examination and behavior of the patient and clinical judgment. While in many cases laboratory findings are invaluable for reaching correct conclusions, the student should never be allowed to forget that it takes a man, not a machine, to understand a man.”

—Raymond B. Allen, MD, PhD, 1946¹

From Hippocrates onward, accurate diagnosis has always been the prerequisite for prognosis and treatment. Physicians typically diagnosed through astute interviewing, deductive reasoning, and skillful use of observation and touch. Then, in the past 250 years they added 2 more tools to their diagnostic skill set: percussion and auscultation, the dual foundation of bedside assessment. Intriguingly, both these skills were first envisioned by multifaceted minds: percussion by Leopold Auenbrugger, an Austrian music-lover who even wrote librettos for operas; and stethoscopy by René Laennec, a Breton flutist, poet, and dancer—not exactly the kind of doctors we tend to produce today.

See related article

Still, the point of this preamble is not to say that eclecticism may help creativity (it does), but to remind ourselves that it has only been for a century or so that physicians have been able to rely on laboratory and radiologic studies. In fact, the now ubiquitous and almost obligatory imaging tests (computed tomography, magnetic resonance imaging, positron-emission tomography, and ultrasonography) have been available to practitioners for only threescore years or less. Yet tests have become so dominant in our culture that it is hard to imagine a time when physicians could count only on their wit and senses.

CLINICAL SKILLS ARE STILL RELEVANT

Ironically, many studies tell us that history and bedside examination can still deliver most diagnoses.^2,3 In fact, clinical skills can solve even the most perplexing dilemmas. In an automated analysis of the clinicopathologic conference cases presented in the New England Journal of Medicine,⁴ history and physical examination still yielded a correct diagnosis in 64% of those very challenging patients.

Bedside examination may be especially important in the hospital. In a study of inpatients,⁵ physical examination detected crucial findings in one-fourth of the cases and prompted management changes in many others. As the authors concluded, sick patients need careful examination, the more skilled the better.

Unfortunately, errors in physical examination are common. In a recent review of 208 cases, 63% of oversights were due to failure to perform an examination, while 25% were either missed or misinterpreted findings.⁶ These errors interfered with diagnosis in three-fourths of the cases, and with treatment in half.

Which brings us to the interesting observation by Kondo et al,⁷ who in this issue of the Journal report how the lowly physical examination proved more helpful than expensive magnetic resonance imaging in evaluating a perplexing case of refractory shoulder pain.

This is not an isolated instance. To get back to Laennec, whose stethoscope just turned 200, auscultation too can help the 21st-century physician. For example, posturally induced crackles, a recently discovered phenomenon, are the third-best predictor of outcome following myocardial infarction, immediately after the number of diseased vessels and pulmonary capillary wedge pressure.⁸

The time-honored art of observation can also yield new and important clues. From the earlobe crease of Dr. Frank, to the elfin face of Dr. Williams, there are lots of diseases out there waiting for our name—if only we could see them. As William Osler put it, “The whole art of medicine is in observation.”⁹

But how can residents truly “observe” when they have to spend 40% of their time looking at computer screens and only 12% looking at people?¹⁰ To quote Osler again, “To educate the eye to see, the ear to hear, and the finger to feel takes time.”⁹ Yet time in medicine is at a premium. In a large national survey, the average ambulatory care visit to a general practitioner lasted 16 minutes,¹¹ which makes it difficult to use inexpensive but time-consuming maneuvers. Detection of posturally induced crackles, for example, may require as much as 9 minutes, and a thorough breast examination up to 10.¹² On the other hand, ordering tests costs little time to the physician but a huge sum to patients and society. Paradoxically, “tests” may be quite profitable for the medical-industrial complex. Hence the erosion of clinical skills.

Overreliance on diagnostic technology is particularly concerning when the cost of medicine has skyrocketed. The United States now spends $3.2 trillion a year for healthcare, and much of this money goes into technology.

In fact, high-tech might hurt us even more than in the pocket. It is a sad fact of modern medicine that when unguided by clinical skills, technology can take us down a rabbit hole, wherein tests beget tests, and where at the end there is usually a surgeon, often a lawyer, and sometimes even an undertaker. The literature is full of such cases, to the point that the risk of unnecessary tests has spawned a charming new acronym: VOMIT (victims of modern imaging technology).¹³

I’m not suggesting that we discard appropriate laboratory and radiologic testing. To the contrary. Yet contributions like those of Kondo et al remind us that even in today’s medicine, the bedside remains not only the royal road to diagnosis, but also the best filter for a more judicious and cost-effective use of technology.

That filter starts with history-taking (“Listen to the patient” said Osler, “he is telling you the diagnosis.”),⁹ and continues with the physical examination. In fact, the history typically guides the physical examination. Hence, when the patient’s symptoms point away from a particular organ, the examination of that organ may be reduced to a minimum. For instance, in neurologic patients whose history made certain findings unlikely, a Canadian group was able to cut in half the number of core items of their neurologic examination.¹⁴

Yet when the history flags a system, the clinician needs to go deeper into the examination. It’s very much what we do with laboratory tests, moving from screening tests to more advanced inquiries as we tailor our diagnostic studies to the patient’s presentation. For that we need validated maneuvers. Recent efforts in this direction have turned the art of physical examination into a science.¹⁵

Lastly, patients expect to be examined, and in fact they resent when this doesn’t happen.¹⁶ Lewis Thomas called touching our “real professional secret” and “the oldest and most effective art of doctors.”¹⁷ It may even have therapeutic value.

TEACHING BEDSIDE DIAGNOSIS

So, if bedside diagnosis is important, what can we do to rekindle it? Probably anything but continue in the old ways. Studies have consistently shown that auscultation does not improve with years of training, and that in fact attending physicians may be no more proficient than third-year medical students.¹⁸ Other areas of the examination have shown similarly depressing trends,¹⁹ thus suggesting that the traditional apprenticeship mode of learning from both faculty and senior trainees may not be helpful. In fact, it may be akin to Bruegel the Elder’s painting of the blind leading the blind, and all ending up in a ditch.

Advanced physical diagnosis courses have thus been advocated, and indeed implemented at many institutions, but usually as electives. Faculty development programs have also been recommended. Still, these interventions may not suffice.

Cutting the cord to technology by serving in a developing country

My hunch is that the rekindling of physical diagnosis may require extreme measures, like putting ourselves in a zero-tech, zero-tests environment. Years ago, I had that kind of cold-turkey experience when I spent a month in a remote Nepali clinic with neither electricity nor running water—and, of course, no cell phone and no Internet. In fact, my only tools were a translator, a stethoscope, and my brain and senses. It was both terrifying and instructive, very much like the time my uncle tried to teach me how to swim by suddenly throwing me into the Mediterranean.

Maybe we should offer that kind of “immersion” to our students. A senior rotation in a technology-depleted country might do a lot of good for a young medical mind. For one, it could remind students that physicians are not only the “natural attorneys of the poor,” as Virchow famously put it,²⁰ but also the ultimate citizens of the world. To quote Dr. Osler again, “Distinctions of race, nationality, color, and creed are unknown within the portals of the temple of Æsculapius.”²¹ Such an experience might also foster empathy and tolerance for ambiguity, 2 other traits whose absence we lament in today’s medicine. More importantly, if preceded by an advanced physical diagnosis course, a rotation in a developing country could work miracles for honing bedside skills, especially if the students are accompanied by a faculty member who can be both inspiring and gifted in the art and science of bedside diagnosis.

Ultimately, this experience could remind our young that the art of medicine is much harder to acquire than the science, and that medicine is indeed a calling and not a trade. Osler said it too, and these are indeed provocative thoughts, but short of provocations and out-of-the-box ideas, the tail will continue to wag the dog. And in the end it will cost us more than money. It will cost us the art of medicine.

“... with the rapid extension of laboratory tests of greater accuracy, there is a tendency for some clinicians and hence for some students in reaching a diagnosis to rely more on laboratory reports and less on the history of the illness, the examination and behavior of the patient and clinical judgment. While in many cases laboratory findings are invaluable for reaching correct conclusions, the student should never be allowed to forget that it takes a man, not a machine, to understand a man.”

—Raymond B. Allen, MD, PhD, 1946¹

From Hippocrates onward, accurate diagnosis has always been the prerequisite for prognosis and treatment. Physicians typically diagnosed through astute interviewing, deductive reasoning, and skillful use of observation and touch. Then, in the past 250 years they added 2 more tools to their diagnostic skill set: percussion and auscultation, the dual foundation of bedside assessment. Intriguingly, both these skills were first envisioned by multifaceted minds: percussion by Leopold Auenbrugger, an Austrian music-lover who even wrote librettos for operas; and stethoscopy by René Laennec, a Breton flutist, poet, and dancer—not exactly the kind of doctors we tend to produce today.

See related article

Still, the point of this preamble is not to say that eclecticism may help creativity (it does), but to remind ourselves that it has only been for a century or so that physicians have been able to rely on laboratory and radiologic studies. In fact, the now ubiquitous and almost obligatory imaging tests (computed tomography, magnetic resonance imaging, positron-emission tomography, and ultrasonography) have been available to practitioners for only threescore years or less. Yet tests have become so dominant in our culture that it is hard to imagine a time when physicians could count only on their wit and senses.

CLINICAL SKILLS ARE STILL RELEVANT

Ironically, many studies tell us that history and bedside examination can still deliver most diagnoses.^2,3 In fact, clinical skills can solve even the most perplexing dilemmas. In an automated analysis of the clinicopathologic conference cases presented in the New England Journal of Medicine,⁴ history and physical examination still yielded a correct diagnosis in 64% of those very challenging patients.

Bedside examination may be especially important in the hospital. In a study of inpatients,⁵ physical examination detected crucial findings in one-fourth of the cases and prompted management changes in many others. As the authors concluded, sick patients need careful examination, the more skilled the better.

Unfortunately, errors in physical examination are common. In a recent review of 208 cases, 63% of oversights were due to failure to perform an examination, while 25% were either missed or misinterpreted findings.⁶ These errors interfered with diagnosis in three-fourths of the cases, and with treatment in half.

Which brings us to the interesting observation by Kondo et al,⁷ who in this issue of the Journal report how the lowly physical examination proved more helpful than expensive magnetic resonance imaging in evaluating a perplexing case of refractory shoulder pain.

This is not an isolated instance. To get back to Laennec, whose stethoscope just turned 200, auscultation too can help the 21st-century physician. For example, posturally induced crackles, a recently discovered phenomenon, are the third-best predictor of outcome following myocardial infarction, immediately after the number of diseased vessels and pulmonary capillary wedge pressure.⁸

The time-honored art of observation can also yield new and important clues. From the earlobe crease of Dr. Frank, to the elfin face of Dr. Williams, there are lots of diseases out there waiting for our name—if only we could see them. As William Osler put it, “The whole art of medicine is in observation.”⁹

But how can residents truly “observe” when they have to spend 40% of their time looking at computer screens and only 12% looking at people?¹⁰ To quote Osler again, “To educate the eye to see, the ear to hear, and the finger to feel takes time.”⁹ Yet time in medicine is at a premium. In a large national survey, the average ambulatory care visit to a general practitioner lasted 16 minutes,¹¹ which makes it difficult to use inexpensive but time-consuming maneuvers. Detection of posturally induced crackles, for example, may require as much as 9 minutes, and a thorough breast examination up to 10.¹² On the other hand, ordering tests costs little time to the physician but a huge sum to patients and society. Paradoxically, “tests” may be quite profitable for the medical-industrial complex. Hence the erosion of clinical skills.

Overreliance on diagnostic technology is particularly concerning when the cost of medicine has skyrocketed. The United States now spends $3.2 trillion a year for healthcare, and much of this money goes into technology.

In fact, high-tech might hurt us even more than in the pocket. It is a sad fact of modern medicine that when unguided by clinical skills, technology can take us down a rabbit hole, wherein tests beget tests, and where at the end there is usually a surgeon, often a lawyer, and sometimes even an undertaker. The literature is full of such cases, to the point that the risk of unnecessary tests has spawned a charming new acronym: VOMIT (victims of modern imaging technology).¹³

I’m not suggesting that we discard appropriate laboratory and radiologic testing. To the contrary. Yet contributions like those of Kondo et al remind us that even in today’s medicine, the bedside remains not only the royal road to diagnosis, but also the best filter for a more judicious and cost-effective use of technology.

That filter starts with history-taking (“Listen to the patient” said Osler, “he is telling you the diagnosis.”),⁹ and continues with the physical examination. In fact, the history typically guides the physical examination. Hence, when the patient’s symptoms point away from a particular organ, the examination of that organ may be reduced to a minimum. For instance, in neurologic patients whose history made certain findings unlikely, a Canadian group was able to cut in half the number of core items of their neurologic examination.¹⁴

Yet when the history flags a system, the clinician needs to go deeper into the examination. It’s very much what we do with laboratory tests, moving from screening tests to more advanced inquiries as we tailor our diagnostic studies to the patient’s presentation. For that we need validated maneuvers. Recent efforts in this direction have turned the art of physical examination into a science.¹⁵

Lastly, patients expect to be examined, and in fact they resent when this doesn’t happen.¹⁶ Lewis Thomas called touching our “real professional secret” and “the oldest and most effective art of doctors.”¹⁷ It may even have therapeutic value.

TEACHING BEDSIDE DIAGNOSIS

So, if bedside diagnosis is important, what can we do to rekindle it? Probably anything but continue in the old ways. Studies have consistently shown that auscultation does not improve with years of training, and that in fact attending physicians may be no more proficient than third-year medical students.¹⁸ Other areas of the examination have shown similarly depressing trends,¹⁹ thus suggesting that the traditional apprenticeship mode of learning from both faculty and senior trainees may not be helpful. In fact, it may be akin to Bruegel the Elder’s painting of the blind leading the blind, and all ending up in a ditch.

Advanced physical diagnosis courses have thus been advocated, and indeed implemented at many institutions, but usually as electives. Faculty development programs have also been recommended. Still, these interventions may not suffice.

Cutting the cord to technology by serving in a developing country

My hunch is that the rekindling of physical diagnosis may require extreme measures, like putting ourselves in a zero-tech, zero-tests environment. Years ago, I had that kind of cold-turkey experience when I spent a month in a remote Nepali clinic with neither electricity nor running water—and, of course, no cell phone and no Internet. In fact, my only tools were a translator, a stethoscope, and my brain and senses. It was both terrifying and instructive, very much like the time my uncle tried to teach me how to swim by suddenly throwing me into the Mediterranean.

Maybe we should offer that kind of “immersion” to our students. A senior rotation in a technology-depleted country might do a lot of good for a young medical mind. For one, it could remind students that physicians are not only the “natural attorneys of the poor,” as Virchow famously put it,²⁰ but also the ultimate citizens of the world. To quote Dr. Osler again, “Distinctions of race, nationality, color, and creed are unknown within the portals of the temple of Æsculapius.”²¹ Such an experience might also foster empathy and tolerance for ambiguity, 2 other traits whose absence we lament in today’s medicine. More importantly, if preceded by an advanced physical diagnosis course, a rotation in a developing country could work miracles for honing bedside skills, especially if the students are accompanied by a faculty member who can be both inspiring and gifted in the art and science of bedside diagnosis.

Ultimately, this experience could remind our young that the art of medicine is much harder to acquire than the science, and that medicine is indeed a calling and not a trade. Osler said it too, and these are indeed provocative thoughts, but short of provocations and out-of-the-box ideas, the tail will continue to wag the dog. And in the end it will cost us more than money. It will cost us the art of medicine.

“... with the rapid extension of laboratory tests of greater accuracy, there is a tendency for some clinicians and hence for some students in reaching a diagnosis to rely more on laboratory reports and less on the history of the illness, the examination and behavior of the patient and clinical judgment. While in many cases laboratory findings are invaluable for reaching correct conclusions, the student should never be allowed to forget that it takes a man, not a machine, to understand a man.”

—Raymond B. Allen, MD, PhD, 1946¹

From Hippocrates onward, accurate diagnosis has always been the prerequisite for prognosis and treatment. Physicians typically diagnosed through astute interviewing, deductive reasoning, and skillful use of observation and touch. Then, in the past 250 years they added 2 more tools to their diagnostic skill set: percussion and auscultation, the dual foundation of bedside assessment. Intriguingly, both these skills were first envisioned by multifaceted minds: percussion by Leopold Auenbrugger, an Austrian music-lover who even wrote librettos for operas; and stethoscopy by René Laennec, a Breton flutist, poet, and dancer—not exactly the kind of doctors we tend to produce today.

See related article

Still, the point of this preamble is not to say that eclecticism may help creativity (it does), but to remind ourselves that it has only been for a century or so that physicians have been able to rely on laboratory and radiologic studies. In fact, the now ubiquitous and almost obligatory imaging tests (computed tomography, magnetic resonance imaging, positron-emission tomography, and ultrasonography) have been available to practitioners for only threescore years or less. Yet tests have become so dominant in our culture that it is hard to imagine a time when physicians could count only on their wit and senses.

CLINICAL SKILLS ARE STILL RELEVANT

Ironically, many studies tell us that history and bedside examination can still deliver most diagnoses.^2,3 In fact, clinical skills can solve even the most perplexing dilemmas. In an automated analysis of the clinicopathologic conference cases presented in the New England Journal of Medicine,⁴ history and physical examination still yielded a correct diagnosis in 64% of those very challenging patients.

Bedside examination may be especially important in the hospital. In a study of inpatients,⁵ physical examination detected crucial findings in one-fourth of the cases and prompted management changes in many others. As the authors concluded, sick patients need careful examination, the more skilled the better.

Unfortunately, errors in physical examination are common. In a recent review of 208 cases, 63% of oversights were due to failure to perform an examination, while 25% were either missed or misinterpreted findings.⁶ These errors interfered with diagnosis in three-fourths of the cases, and with treatment in half.

Which brings us to the interesting observation by Kondo et al,⁷ who in this issue of the Journal report how the lowly physical examination proved more helpful than expensive magnetic resonance imaging in evaluating a perplexing case of refractory shoulder pain.

This is not an isolated instance. To get back to Laennec, whose stethoscope just turned 200, auscultation too can help the 21st-century physician. For example, posturally induced crackles, a recently discovered phenomenon, are the third-best predictor of outcome following myocardial infarction, immediately after the number of diseased vessels and pulmonary capillary wedge pressure.⁸

The time-honored art of observation can also yield new and important clues. From the earlobe crease of Dr. Frank, to the elfin face of Dr. Williams, there are lots of diseases out there waiting for our name—if only we could see them. As William Osler put it, “The whole art of medicine is in observation.”⁹

But how can residents truly “observe” when they have to spend 40% of their time looking at computer screens and only 12% looking at people?¹⁰ To quote Osler again, “To educate the eye to see, the ear to hear, and the finger to feel takes time.”⁹ Yet time in medicine is at a premium. In a large national survey, the average ambulatory care visit to a general practitioner lasted 16 minutes,¹¹ which makes it difficult to use inexpensive but time-consuming maneuvers. Detection of posturally induced crackles, for example, may require as much as 9 minutes, and a thorough breast examination up to 10.¹² On the other hand, ordering tests costs little time to the physician but a huge sum to patients and society. Paradoxically, “tests” may be quite profitable for the medical-industrial complex. Hence the erosion of clinical skills.

Overreliance on diagnostic technology is particularly concerning when the cost of medicine has skyrocketed. The United States now spends $3.2 trillion a year for healthcare, and much of this money goes into technology.

In fact, high-tech might hurt us even more than in the pocket. It is a sad fact of modern medicine that when unguided by clinical skills, technology can take us down a rabbit hole, wherein tests beget tests, and where at the end there is usually a surgeon, often a lawyer, and sometimes even an undertaker. The literature is full of such cases, to the point that the risk of unnecessary tests has spawned a charming new acronym: VOMIT (victims of modern imaging technology).¹³

I’m not suggesting that we discard appropriate laboratory and radiologic testing. To the contrary. Yet contributions like those of Kondo et al remind us that even in today’s medicine, the bedside remains not only the royal road to diagnosis, but also the best filter for a more judicious and cost-effective use of technology.

That filter starts with history-taking (“Listen to the patient” said Osler, “he is telling you the diagnosis.”),⁹ and continues with the physical examination. In fact, the history typically guides the physical examination. Hence, when the patient’s symptoms point away from a particular organ, the examination of that organ may be reduced to a minimum. For instance, in neurologic patients whose history made certain findings unlikely, a Canadian group was able to cut in half the number of core items of their neurologic examination.¹⁴

Yet when the history flags a system, the clinician needs to go deeper into the examination. It’s very much what we do with laboratory tests, moving from screening tests to more advanced inquiries as we tailor our diagnostic studies to the patient’s presentation. For that we need validated maneuvers. Recent efforts in this direction have turned the art of physical examination into a science.¹⁵

Lastly, patients expect to be examined, and in fact they resent when this doesn’t happen.¹⁶ Lewis Thomas called touching our “real professional secret” and “the oldest and most effective art of doctors.”¹⁷ It may even have therapeutic value.

TEACHING BEDSIDE DIAGNOSIS

So, if bedside diagnosis is important, what can we do to rekindle it? Probably anything but continue in the old ways. Studies have consistently shown that auscultation does not improve with years of training, and that in fact attending physicians may be no more proficient than third-year medical students.¹⁸ Other areas of the examination have shown similarly depressing trends,¹⁹ thus suggesting that the traditional apprenticeship mode of learning from both faculty and senior trainees may not be helpful. In fact, it may be akin to Bruegel the Elder’s painting of the blind leading the blind, and all ending up in a ditch.

Advanced physical diagnosis courses have thus been advocated, and indeed implemented at many institutions, but usually as electives. Faculty development programs have also been recommended. Still, these interventions may not suffice.

Cutting the cord to technology by serving in a developing country

My hunch is that the rekindling of physical diagnosis may require extreme measures, like putting ourselves in a zero-tech, zero-tests environment. Years ago, I had that kind of cold-turkey experience when I spent a month in a remote Nepali clinic with neither electricity nor running water—and, of course, no cell phone and no Internet. In fact, my only tools were a translator, a stethoscope, and my brain and senses. It was both terrifying and instructive, very much like the time my uncle tried to teach me how to swim by suddenly throwing me into the Mediterranean.

Maybe we should offer that kind of “immersion” to our students. A senior rotation in a technology-depleted country might do a lot of good for a young medical mind. For one, it could remind students that physicians are not only the “natural attorneys of the poor,” as Virchow famously put it,²⁰ but also the ultimate citizens of the world. To quote Dr. Osler again, “Distinctions of race, nationality, color, and creed are unknown within the portals of the temple of Æsculapius.”²¹ Such an experience might also foster empathy and tolerance for ambiguity, 2 other traits whose absence we lament in today’s medicine. More importantly, if preceded by an advanced physical diagnosis course, a rotation in a developing country could work miracles for honing bedside skills, especially if the students are accompanied by a faculty member who can be both inspiring and gifted in the art and science of bedside diagnosis.

Ultimately, this experience could remind our young that the art of medicine is much harder to acquire than the science, and that medicine is indeed a calling and not a trade. Osler said it too, and these are indeed provocative thoughts, but short of provocations and out-of-the-box ideas, the tail will continue to wag the dog. And in the end it will cost us more than money. It will cost us the art of medicine.

Now that we can order MRI studies on a break from rounds walking to Starbucks, utilize portable ultrasounds to direct IV line placement, and use dual-energy CT to detect a gout attack that has not yet occurred, it seems like a romantic anachronism to extol the ongoing virtues of the seemingly lost art of the physical examination. Back “in the day,” the giants of medicine roamed the halls with their natural instruments of palpation and percussion and their skills in observation and auscultation. They were giants because they stood out then, just as skilled diagnosticians stand out today using an upgraded set of tools. Some physicians a few decades ago were able to recognize, describe, and diagnose late-stage endocarditis with a stethoscope, a magnifying glass, and an ophthalmoscope. The giants of today recognize the patient with endocarditis and document its presence using transesophageal echocardiography before the peripheral eponymous stigmata of Janeway and Osler appear or the blood cultures turn positive. The physical examination, history, diagnostic reasoning, and clinical technology are all essential for a blend that provides efficient and effective medical care. The blending is the challenge.

Clinicians are not created equal. We learn and prioritize our skills in different ways. But if we are not taught to value and trust the physical examination, if we don’t have the opportunity to see it influence patient management in positive ways, we may eschew it and instead indiscriminately use easily available laboratory and imaging tests—a more expensive and often misleading strategic approach. Today while in clinic, I saw a 54-year-old woman for evaluation of possible lupus who had arthritis of the hands and a high positive antinuclear antibody titer, but negative or normal results on other, previously ordered tests, including anti-DNA, rheumatoid factor, anti-cyclic citrullinated peptide, hepatitis C studies, complement levels, and another half-dozen immune serologic tests. On examination, she had typical nodular osteoarthritis of the proximal and distal interphalangeal joints of her hand with squaring of her thumbs. The antinuclear antibody was most likely associated with her previously diagnosed autoimmune thyroid disease.

In an editorial in this issue of the Journal, Dr. Salvatore Mangione, the author of a book on physical diagnosis,¹ cites a recent study indicating that the most common recognized diagnostic error related to the physical examination is that the appropriate examination isn’t done.² I would add to that my concerns over the new common custom of cutting and pasting the findings from earlier physical examinations into later progress notes in the electronic record. So much for the value of being able to recognize “changing murmurs” when diagnosing infectious endocarditis.

The apparent efficiency (reflected in length of stay) and availability of technology, as well as a lack of physician skill and time, are often cited as reasons for the demise of the physical examination. Yet this does not need to be the case. If I had trained with portable ultrasonography readily available to confirm or refute my impressions, my skills at detecting low-grade synovitis would surely be better than they are. With a gold standard at hand, which may be technology or at times a skilled mentor, our examinations can be refined if we want them to be.

But the issue of limited physician time must be addressed. Efficiency is a critical concept in preserving how we practice and perform the physical examination. When we know what we are looking for, we are more likely to find it if it is present, or to have confidence that it is not present. I am far more likely to recognize a loud pulmonic second heart sound if I suspect that the dyspneic patient I am examining has pulmonary hypertension associated with her scleroderma than if I am doing a perfunctory cardiac auscultation in a patient admitted with cellulitis. Appropriate focus provides power to the directed physical examination. If I am looking for the cause of unexplained fevers, I will do a purposeful axillary and epitrochlear lymph node examination. I am not mindlessly probing the flesh.

Nishigori and colleagues have written of the “hypothesis-driven” physical examination.³ Busy clinicians, they say, don’t have time to perform a head-to-toe, by-the-book physical examination. Instead, we should, by a dynamic process, formulate a differential diagnosis from the history and other initial information, and then perform the directed physical examination in earnest, looking for evidence to support or refute our diagnostic hypothesis—and thus redirect it. Plus, in a nice break from electronic charting, we can actually explain our thought processes to the patient as we perform the examination.

This approach makes sense to me as both intellectually satisfying and clinically efficient. And then we can consider which lab tests and technologic gadgetry we should order, while walking to get the café latte we ordered with our cell phone app.

New technology can support and not necessarily replace old habits.

Now that we can order MRI studies on a break from rounds walking to Starbucks, utilize portable ultrasounds to direct IV line placement, and use dual-energy CT to detect a gout attack that has not yet occurred, it seems like a romantic anachronism to extol the ongoing virtues of the seemingly lost art of the physical examination. Back “in the day,” the giants of medicine roamed the halls with their natural instruments of palpation and percussion and their skills in observation and auscultation. They were giants because they stood out then, just as skilled diagnosticians stand out today using an upgraded set of tools. Some physicians a few decades ago were able to recognize, describe, and diagnose late-stage endocarditis with a stethoscope, a magnifying glass, and an ophthalmoscope. The giants of today recognize the patient with endocarditis and document its presence using transesophageal echocardiography before the peripheral eponymous stigmata of Janeway and Osler appear or the blood cultures turn positive. The physical examination, history, diagnostic reasoning, and clinical technology are all essential for a blend that provides efficient and effective medical care. The blending is the challenge.

Clinicians are not created equal. We learn and prioritize our skills in different ways. But if we are not taught to value and trust the physical examination, if we don’t have the opportunity to see it influence patient management in positive ways, we may eschew it and instead indiscriminately use easily available laboratory and imaging tests—a more expensive and often misleading strategic approach. Today while in clinic, I saw a 54-year-old woman for evaluation of possible lupus who had arthritis of the hands and a high positive antinuclear antibody titer, but negative or normal results on other, previously ordered tests, including anti-DNA, rheumatoid factor, anti-cyclic citrullinated peptide, hepatitis C studies, complement levels, and another half-dozen immune serologic tests. On examination, she had typical nodular osteoarthritis of the proximal and distal interphalangeal joints of her hand with squaring of her thumbs. The antinuclear antibody was most likely associated with her previously diagnosed autoimmune thyroid disease.

In an editorial in this issue of the Journal, Dr. Salvatore Mangione, the author of a book on physical diagnosis,¹ cites a recent study indicating that the most common recognized diagnostic error related to the physical examination is that the appropriate examination isn’t done.² I would add to that my concerns over the new common custom of cutting and pasting the findings from earlier physical examinations into later progress notes in the electronic record. So much for the value of being able to recognize “changing murmurs” when diagnosing infectious endocarditis.

The apparent efficiency (reflected in length of stay) and availability of technology, as well as a lack of physician skill and time, are often cited as reasons for the demise of the physical examination. Yet this does not need to be the case. If I had trained with portable ultrasonography readily available to confirm or refute my impressions, my skills at detecting low-grade synovitis would surely be better than they are. With a gold standard at hand, which may be technology or at times a skilled mentor, our examinations can be refined if we want them to be.

But the issue of limited physician time must be addressed. Efficiency is a critical concept in preserving how we practice and perform the physical examination. When we know what we are looking for, we are more likely to find it if it is present, or to have confidence that it is not present. I am far more likely to recognize a loud pulmonic second heart sound if I suspect that the dyspneic patient I am examining has pulmonary hypertension associated with her scleroderma than if I am doing a perfunctory cardiac auscultation in a patient admitted with cellulitis. Appropriate focus provides power to the directed physical examination. If I am looking for the cause of unexplained fevers, I will do a purposeful axillary and epitrochlear lymph node examination. I am not mindlessly probing the flesh.

Nishigori and colleagues have written of the “hypothesis-driven” physical examination.³ Busy clinicians, they say, don’t have time to perform a head-to-toe, by-the-book physical examination. Instead, we should, by a dynamic process, formulate a differential diagnosis from the history and other initial information, and then perform the directed physical examination in earnest, looking for evidence to support or refute our diagnostic hypothesis—and thus redirect it. Plus, in a nice break from electronic charting, we can actually explain our thought processes to the patient as we perform the examination.

This approach makes sense to me as both intellectually satisfying and clinically efficient. And then we can consider which lab tests and technologic gadgetry we should order, while walking to get the café latte we ordered with our cell phone app.

New technology can support and not necessarily replace old habits.

Now that we can order MRI studies on a break from rounds walking to Starbucks, utilize portable ultrasounds to direct IV line placement, and use dual-energy CT to detect a gout attack that has not yet occurred, it seems like a romantic anachronism to extol the ongoing virtues of the seemingly lost art of the physical examination. Back “in the day,” the giants of medicine roamed the halls with their natural instruments of palpation and percussion and their skills in observation and auscultation. They were giants because they stood out then, just as skilled diagnosticians stand out today using an upgraded set of tools. Some physicians a few decades ago were able to recognize, describe, and diagnose late-stage endocarditis with a stethoscope, a magnifying glass, and an ophthalmoscope. The giants of today recognize the patient with endocarditis and document its presence using transesophageal echocardiography before the peripheral eponymous stigmata of Janeway and Osler appear or the blood cultures turn positive. The physical examination, history, diagnostic reasoning, and clinical technology are all essential for a blend that provides efficient and effective medical care. The blending is the challenge.

Clinicians are not created equal. We learn and prioritize our skills in different ways. But if we are not taught to value and trust the physical examination, if we don’t have the opportunity to see it influence patient management in positive ways, we may eschew it and instead indiscriminately use easily available laboratory and imaging tests—a more expensive and often misleading strategic approach. Today while in clinic, I saw a 54-year-old woman for evaluation of possible lupus who had arthritis of the hands and a high positive antinuclear antibody titer, but negative or normal results on other, previously ordered tests, including anti-DNA, rheumatoid factor, anti-cyclic citrullinated peptide, hepatitis C studies, complement levels, and another half-dozen immune serologic tests. On examination, she had typical nodular osteoarthritis of the proximal and distal interphalangeal joints of her hand with squaring of her thumbs. The antinuclear antibody was most likely associated with her previously diagnosed autoimmune thyroid disease.

In an editorial in this issue of the Journal, Dr. Salvatore Mangione, the author of a book on physical diagnosis,¹ cites a recent study indicating that the most common recognized diagnostic error related to the physical examination is that the appropriate examination isn’t done.² I would add to that my concerns over the new common custom of cutting and pasting the findings from earlier physical examinations into later progress notes in the electronic record. So much for the value of being able to recognize “changing murmurs” when diagnosing infectious endocarditis.

The apparent efficiency (reflected in length of stay) and availability of technology, as well as a lack of physician skill and time, are often cited as reasons for the demise of the physical examination. Yet this does not need to be the case. If I had trained with portable ultrasonography readily available to confirm or refute my impressions, my skills at detecting low-grade synovitis would surely be better than they are. With a gold standard at hand, which may be technology or at times a skilled mentor, our examinations can be refined if we want them to be.

But the issue of limited physician time must be addressed. Efficiency is a critical concept in preserving how we practice and perform the physical examination. When we know what we are looking for, we are more likely to find it if it is present, or to have confidence that it is not present. I am far more likely to recognize a loud pulmonic second heart sound if I suspect that the dyspneic patient I am examining has pulmonary hypertension associated with her scleroderma than if I am doing a perfunctory cardiac auscultation in a patient admitted with cellulitis. Appropriate focus provides power to the directed physical examination. If I am looking for the cause of unexplained fevers, I will do a purposeful axillary and epitrochlear lymph node examination. I am not mindlessly probing the flesh.

Nishigori and colleagues have written of the “hypothesis-driven” physical examination.³ Busy clinicians, they say, don’t have time to perform a head-to-toe, by-the-book physical examination. Instead, we should, by a dynamic process, formulate a differential diagnosis from the history and other initial information, and then perform the directed physical examination in earnest, looking for evidence to support or refute our diagnostic hypothesis—and thus redirect it. Plus, in a nice break from electronic charting, we can actually explain our thought processes to the patient as we perform the examination.

This approach makes sense to me as both intellectually satisfying and clinically efficient. And then we can consider which lab tests and technologic gadgetry we should order, while walking to get the café latte we ordered with our cell phone app.

New technology can support and not necessarily replace old habits.