Acute upper gastrointestinal hemorrhage (UGIH) is one of the most common hospital admissions for acute care. Estimates indicate that 300,000 patients (100‐150 cases per 100,000 adults) are admitted annually with an associated economic impact of $2.5 billion.15 The current standard management of UGIH requires hospital admission and esophagogastroduodenoscopy (EGD) by a gastroenterologist for diagnosis and/or treatment. This management strategy results in a high consumption of hospital resources and costs.

Simultaneously, hospitalists have dramatically changed the delivery of inpatient care in the United States and are recognized as a location‐driven subspecialty for the care of acute hospitalized patients, similar to emergency medicine. Currently there are 20,000 hospitalists, and more than one‐third of general medicine inpatients are cared for by hospitalists.6, 7

Previous studies have shown that hospitalist care offers better or comparable outcomes, with lower overall length of stay (LOS) and costs compared to traditional providers.810 However, most of these studies were performed in single institutions, had weak designs or little‐to‐no adjustment for severity of illness, or were limited to 7 specific diseases (pneumonia, congestive heart failure [CHF], chest pain, ischemic stroke, urinary tract infection, chronic obstructive lung disease [COPD], and acute myocardial infarction [AMI]).8

Furthermore, less is known about the effect of hospitalists on conditions that may be dependent upon specialist consultation for procedures and/or treatment plans. In this study, gastroenterologists performed diagnostic and/or therapeutic endoscopy work as consultants to the attending physicians in the management of acute inpatient UGIH.

To explore the effects of hospitalists on care of patients with acute UGIH, we examined data from the Multicenter Hospitalist (MCH) trial. The objectives of our study were to compare clinical outcomesin‐hospital mortality and complications (ie, recurrent bleeding, intensive care unit [ICU] transfer, decompensation, transfusion, reendoscopy, 30‐day readmission)and efficiency (LOS and costs) in hospitalized acute UGIH patients cared for by hospitalists and nonhospitalists in 6 academic centers in the United States during a 2‐year period.

Patients and Methods

Results

Discussion

This is the first study that has looked at the effect of hospitalists on clinical outcomes and efficiency in patients admitted for acute UGIH, a condition highly dependent upon another specialty for procedures and management. This is also one of only a few studies on UGIH that adjusted for severity of illness (Rockall score), comorbidity, performance of early endoscopypatient‐level confounders usually unaccounted for in prior research.

We show that hospitalists and nonhospitalists caring for acute UGIH patients had overall similar unadjusted outcomes; except for blood transfusion and 30‐day readmission rates. Unfortunately, due to the small number of events for readmissions, we were unable to perform adjusted analysis for readmission. Differences between hospitalists and nonhospitalists on blood transfusion rates were not substantiated on multivariable adjustments.

As for efficiency, univariable and multivariable analyses revealed that LOS was similar between provider types while costs were greater in UGIH patients attended by hospitalists.

Reductions in resource use, particularly costs, may be achieved by increasing throughput (eg, reducing LOS) or by decreasing service intensity (eg, using fewer ancillary services and specialty consultations).26 Specifically in acute UGIH, LOS is significantly affected by performance of early EGD.27, 28 In these studies, gastroenterologist‐led teams, compared to internists and surgeons, have easier access to endoscopy, thus reducing LOS and overall costs.27, 28

Similarly, prior studies have shown that the mechanism by which hospitalists lower costs is by decreasing LOS.810, 29 There are several hypotheses on how hospitalists affect LOS. Hospitalists, by being available all day, are thought to respond quickly to acute symptoms or new test results, are more efficient in navigating the complex hospital environment, or develop greater expertise as a result of added inpatient experience.8 On the downside, although the hospitalist model reduces overall LOS and costs, they also provide higher intensity of care as reflected by greater costs when broken down per hospital day.29 Thus, the cost differential we found may represent higher intensity of care by hospitalists in their management of acute UGIH, as higher intensity care without decreasing LOS can translate to higher costs.

In addition, patients with acute UGIH are unique in several respects. In contrast to diseases like heart failure, COPD, and pneumonia, in which the admitting provider has the option to request a subspecialist consultation, all patients with acute UGIH need a gastroenterologist to perform endoscopy as part of the management. These patients are usually admitted to general medicine wards, aggressively resuscitated with intravenous fluids, with a nonurgent gastroenterology consult or EGD performed on the next available schedule.

Aside from LOS being greatly affected by performance of early EGD and/or delay in consulting gastroenterology, sicker patients require longer hospitalization and drive LOS and healthcare costs up. It was therefore crucial that we accounted for severity of illness, comorbidity, and performance of early EGD in our regression models for LOS and costs. This approach allows us to acquire a more accurate estimate on the effects of hospitalist on LOS and costs in patients admitted with acute UGIH.

Our findings suggest that the academic hospitalist model of care may not have as great of an impact on hospital efficiency in certain patient groups that require nonurgent subspecialty consultations. Future studies should focus on elucidating these relationships.

Limitations

This study has several limitations. First, clinical data were abstracted at 6 sites by different abstractors so it is possible there were variations in how data were collected. To reduce variation, a standardized abstraction form with instructions was developed and the primary investigator (PI) was available for specific questions during the abstraction process. Second, only 5 out of the 6 sites used TSI accounting systems. Although similar, interhospital costs captured by TSI may vary among sites in terms of classifying direct and indirect costs, potentially resulting in misclassification bias in our cost estimates.17 We addressed these issues by including the hospital site variable in our regression models, regardless of its significance. Third, consent rates across sites vary from 70% to 85%. It is possible that patients who refused enrollment in the MCH trial are systematically different and may introduce bias in our analysis.

Furthermore, the study was designed as a natural experiment based on a rotational call cycle between hospitalist‐led and nonhospitalist‐led teams. It is possible that the order of patient assignment might not be completely naturally random as we intended. However, the study period was for 2 years and we expect the effect of order would have averaged out in time.

There are many hospitalist models of care. In terms of generalizability, the study pertains only to academic hospitalists and may not be applicable to hospitalists practicing in community hospitals. For example, the nonhospitalist comparison group is likely different in the community and academic settings. Community nonhospitalists (traditional practitioners) are usually internists covering both inpatient and outpatient responsibilities at the same time. In contrast, academic nonhospitalists are internists or subspecialists serving as ward attendings for a limited period (usually 1 month) with considerable variation in their nonattending responsibilities (eg, research, clinic, administration). Furthermore, academic nonhospitalist providers might be a self‐selected group by their willingness to serve as a ward attending, making them more hospitalist‐like. Changes and variability of inpatient attendings may also affect our findings when compared to prior work. Finally, it is also possible that having residents at academic medical centers may attenuate the effect of hospitalists more than in community‐based models.

Conclusions/Implications

Compared to nonhospitalists, academic hospitalist care of acute UGIH patients had similar overall clinical outcomes. However, our finding of similar LOS yet higher costs for patients cared for by hospitalists support 1 proposed mechanism in which hospitalists decrease healthcare costs: providing higher intensity of care per day of hospitalization. However, in academic hospitalist models, this higher intensity hypothesis should be revisited, especially in certain patient groups in which timing and involvement of subspecialists may influence discharge decisions, affecting LOS and overall costs.

Due to inherent limitations in this observational study, future studies should focus on verifying and elucidating these relationships further. Lastly, understanding which patient groups receive the greatest potential benefit from this model will help guide both organizational efforts and quality improvement strategies.

Acute upper gastrointestinal hemorrhage (UGIH) is one of the most common hospital admissions for acute care. Estimates indicate that 300,000 patients (100‐150 cases per 100,000 adults) are admitted annually with an associated economic impact of $2.5 billion.15 The current standard management of UGIH requires hospital admission and esophagogastroduodenoscopy (EGD) by a gastroenterologist for diagnosis and/or treatment. This management strategy results in a high consumption of hospital resources and costs.

Simultaneously, hospitalists have dramatically changed the delivery of inpatient care in the United States and are recognized as a location‐driven subspecialty for the care of acute hospitalized patients, similar to emergency medicine. Currently there are 20,000 hospitalists, and more than one‐third of general medicine inpatients are cared for by hospitalists.6, 7

Previous studies have shown that hospitalist care offers better or comparable outcomes, with lower overall length of stay (LOS) and costs compared to traditional providers.810 However, most of these studies were performed in single institutions, had weak designs or little‐to‐no adjustment for severity of illness, or were limited to 7 specific diseases (pneumonia, congestive heart failure [CHF], chest pain, ischemic stroke, urinary tract infection, chronic obstructive lung disease [COPD], and acute myocardial infarction [AMI]).8

Furthermore, less is known about the effect of hospitalists on conditions that may be dependent upon specialist consultation for procedures and/or treatment plans. In this study, gastroenterologists performed diagnostic and/or therapeutic endoscopy work as consultants to the attending physicians in the management of acute inpatient UGIH.

To explore the effects of hospitalists on care of patients with acute UGIH, we examined data from the Multicenter Hospitalist (MCH) trial. The objectives of our study were to compare clinical outcomesin‐hospital mortality and complications (ie, recurrent bleeding, intensive care unit [ICU] transfer, decompensation, transfusion, reendoscopy, 30‐day readmission)and efficiency (LOS and costs) in hospitalized acute UGIH patients cared for by hospitalists and nonhospitalists in 6 academic centers in the United States during a 2‐year period.

Patients and Methods

Results

Discussion

This is the first study that has looked at the effect of hospitalists on clinical outcomes and efficiency in patients admitted for acute UGIH, a condition highly dependent upon another specialty for procedures and management. This is also one of only a few studies on UGIH that adjusted for severity of illness (Rockall score), comorbidity, performance of early endoscopypatient‐level confounders usually unaccounted for in prior research.

We show that hospitalists and nonhospitalists caring for acute UGIH patients had overall similar unadjusted outcomes; except for blood transfusion and 30‐day readmission rates. Unfortunately, due to the small number of events for readmissions, we were unable to perform adjusted analysis for readmission. Differences between hospitalists and nonhospitalists on blood transfusion rates were not substantiated on multivariable adjustments.

As for efficiency, univariable and multivariable analyses revealed that LOS was similar between provider types while costs were greater in UGIH patients attended by hospitalists.

Reductions in resource use, particularly costs, may be achieved by increasing throughput (eg, reducing LOS) or by decreasing service intensity (eg, using fewer ancillary services and specialty consultations).26 Specifically in acute UGIH, LOS is significantly affected by performance of early EGD.27, 28 In these studies, gastroenterologist‐led teams, compared to internists and surgeons, have easier access to endoscopy, thus reducing LOS and overall costs.27, 28

Similarly, prior studies have shown that the mechanism by which hospitalists lower costs is by decreasing LOS.810, 29 There are several hypotheses on how hospitalists affect LOS. Hospitalists, by being available all day, are thought to respond quickly to acute symptoms or new test results, are more efficient in navigating the complex hospital environment, or develop greater expertise as a result of added inpatient experience.8 On the downside, although the hospitalist model reduces overall LOS and costs, they also provide higher intensity of care as reflected by greater costs when broken down per hospital day.29 Thus, the cost differential we found may represent higher intensity of care by hospitalists in their management of acute UGIH, as higher intensity care without decreasing LOS can translate to higher costs.

In addition, patients with acute UGIH are unique in several respects. In contrast to diseases like heart failure, COPD, and pneumonia, in which the admitting provider has the option to request a subspecialist consultation, all patients with acute UGIH need a gastroenterologist to perform endoscopy as part of the management. These patients are usually admitted to general medicine wards, aggressively resuscitated with intravenous fluids, with a nonurgent gastroenterology consult or EGD performed on the next available schedule.

Aside from LOS being greatly affected by performance of early EGD and/or delay in consulting gastroenterology, sicker patients require longer hospitalization and drive LOS and healthcare costs up. It was therefore crucial that we accounted for severity of illness, comorbidity, and performance of early EGD in our regression models for LOS and costs. This approach allows us to acquire a more accurate estimate on the effects of hospitalist on LOS and costs in patients admitted with acute UGIH.

Our findings suggest that the academic hospitalist model of care may not have as great of an impact on hospital efficiency in certain patient groups that require nonurgent subspecialty consultations. Future studies should focus on elucidating these relationships.

Limitations

This study has several limitations. First, clinical data were abstracted at 6 sites by different abstractors so it is possible there were variations in how data were collected. To reduce variation, a standardized abstraction form with instructions was developed and the primary investigator (PI) was available for specific questions during the abstraction process. Second, only 5 out of the 6 sites used TSI accounting systems. Although similar, interhospital costs captured by TSI may vary among sites in terms of classifying direct and indirect costs, potentially resulting in misclassification bias in our cost estimates.17 We addressed these issues by including the hospital site variable in our regression models, regardless of its significance. Third, consent rates across sites vary from 70% to 85%. It is possible that patients who refused enrollment in the MCH trial are systematically different and may introduce bias in our analysis.

Furthermore, the study was designed as a natural experiment based on a rotational call cycle between hospitalist‐led and nonhospitalist‐led teams. It is possible that the order of patient assignment might not be completely naturally random as we intended. However, the study period was for 2 years and we expect the effect of order would have averaged out in time.

There are many hospitalist models of care. In terms of generalizability, the study pertains only to academic hospitalists and may not be applicable to hospitalists practicing in community hospitals. For example, the nonhospitalist comparison group is likely different in the community and academic settings. Community nonhospitalists (traditional practitioners) are usually internists covering both inpatient and outpatient responsibilities at the same time. In contrast, academic nonhospitalists are internists or subspecialists serving as ward attendings for a limited period (usually 1 month) with considerable variation in their nonattending responsibilities (eg, research, clinic, administration). Furthermore, academic nonhospitalist providers might be a self‐selected group by their willingness to serve as a ward attending, making them more hospitalist‐like. Changes and variability of inpatient attendings may also affect our findings when compared to prior work. Finally, it is also possible that having residents at academic medical centers may attenuate the effect of hospitalists more than in community‐based models.

Conclusions/Implications

Compared to nonhospitalists, academic hospitalist care of acute UGIH patients had similar overall clinical outcomes. However, our finding of similar LOS yet higher costs for patients cared for by hospitalists support 1 proposed mechanism in which hospitalists decrease healthcare costs: providing higher intensity of care per day of hospitalization. However, in academic hospitalist models, this higher intensity hypothesis should be revisited, especially in certain patient groups in which timing and involvement of subspecialists may influence discharge decisions, affecting LOS and overall costs.

Due to inherent limitations in this observational study, future studies should focus on verifying and elucidating these relationships further. Lastly, understanding which patient groups receive the greatest potential benefit from this model will help guide both organizational efforts and quality improvement strategies.

Acute upper gastrointestinal hemorrhage (UGIH) is one of the most common hospital admissions for acute care. Estimates indicate that 300,000 patients (100‐150 cases per 100,000 adults) are admitted annually with an associated economic impact of $2.5 billion.15 The current standard management of UGIH requires hospital admission and esophagogastroduodenoscopy (EGD) by a gastroenterologist for diagnosis and/or treatment. This management strategy results in a high consumption of hospital resources and costs.

Simultaneously, hospitalists have dramatically changed the delivery of inpatient care in the United States and are recognized as a location‐driven subspecialty for the care of acute hospitalized patients, similar to emergency medicine. Currently there are 20,000 hospitalists, and more than one‐third of general medicine inpatients are cared for by hospitalists.6, 7

Previous studies have shown that hospitalist care offers better or comparable outcomes, with lower overall length of stay (LOS) and costs compared to traditional providers.810 However, most of these studies were performed in single institutions, had weak designs or little‐to‐no adjustment for severity of illness, or were limited to 7 specific diseases (pneumonia, congestive heart failure [CHF], chest pain, ischemic stroke, urinary tract infection, chronic obstructive lung disease [COPD], and acute myocardial infarction [AMI]).8

Furthermore, less is known about the effect of hospitalists on conditions that may be dependent upon specialist consultation for procedures and/or treatment plans. In this study, gastroenterologists performed diagnostic and/or therapeutic endoscopy work as consultants to the attending physicians in the management of acute inpatient UGIH.

To explore the effects of hospitalists on care of patients with acute UGIH, we examined data from the Multicenter Hospitalist (MCH) trial. The objectives of our study were to compare clinical outcomesin‐hospital mortality and complications (ie, recurrent bleeding, intensive care unit [ICU] transfer, decompensation, transfusion, reendoscopy, 30‐day readmission)and efficiency (LOS and costs) in hospitalized acute UGIH patients cared for by hospitalists and nonhospitalists in 6 academic centers in the United States during a 2‐year period.

Patients and Methods

Results

Discussion

This is the first study that has looked at the effect of hospitalists on clinical outcomes and efficiency in patients admitted for acute UGIH, a condition highly dependent upon another specialty for procedures and management. This is also one of only a few studies on UGIH that adjusted for severity of illness (Rockall score), comorbidity, performance of early endoscopypatient‐level confounders usually unaccounted for in prior research.

We show that hospitalists and nonhospitalists caring for acute UGIH patients had overall similar unadjusted outcomes; except for blood transfusion and 30‐day readmission rates. Unfortunately, due to the small number of events for readmissions, we were unable to perform adjusted analysis for readmission. Differences between hospitalists and nonhospitalists on blood transfusion rates were not substantiated on multivariable adjustments.

As for efficiency, univariable and multivariable analyses revealed that LOS was similar between provider types while costs were greater in UGIH patients attended by hospitalists.

Reductions in resource use, particularly costs, may be achieved by increasing throughput (eg, reducing LOS) or by decreasing service intensity (eg, using fewer ancillary services and specialty consultations).26 Specifically in acute UGIH, LOS is significantly affected by performance of early EGD.27, 28 In these studies, gastroenterologist‐led teams, compared to internists and surgeons, have easier access to endoscopy, thus reducing LOS and overall costs.27, 28

Similarly, prior studies have shown that the mechanism by which hospitalists lower costs is by decreasing LOS.810, 29 There are several hypotheses on how hospitalists affect LOS. Hospitalists, by being available all day, are thought to respond quickly to acute symptoms or new test results, are more efficient in navigating the complex hospital environment, or develop greater expertise as a result of added inpatient experience.8 On the downside, although the hospitalist model reduces overall LOS and costs, they also provide higher intensity of care as reflected by greater costs when broken down per hospital day.29 Thus, the cost differential we found may represent higher intensity of care by hospitalists in their management of acute UGIH, as higher intensity care without decreasing LOS can translate to higher costs.

In addition, patients with acute UGIH are unique in several respects. In contrast to diseases like heart failure, COPD, and pneumonia, in which the admitting provider has the option to request a subspecialist consultation, all patients with acute UGIH need a gastroenterologist to perform endoscopy as part of the management. These patients are usually admitted to general medicine wards, aggressively resuscitated with intravenous fluids, with a nonurgent gastroenterology consult or EGD performed on the next available schedule.

Aside from LOS being greatly affected by performance of early EGD and/or delay in consulting gastroenterology, sicker patients require longer hospitalization and drive LOS and healthcare costs up. It was therefore crucial that we accounted for severity of illness, comorbidity, and performance of early EGD in our regression models for LOS and costs. This approach allows us to acquire a more accurate estimate on the effects of hospitalist on LOS and costs in patients admitted with acute UGIH.

Our findings suggest that the academic hospitalist model of care may not have as great of an impact on hospital efficiency in certain patient groups that require nonurgent subspecialty consultations. Future studies should focus on elucidating these relationships.

Limitations

This study has several limitations. First, clinical data were abstracted at 6 sites by different abstractors so it is possible there were variations in how data were collected. To reduce variation, a standardized abstraction form with instructions was developed and the primary investigator (PI) was available for specific questions during the abstraction process. Second, only 5 out of the 6 sites used TSI accounting systems. Although similar, interhospital costs captured by TSI may vary among sites in terms of classifying direct and indirect costs, potentially resulting in misclassification bias in our cost estimates.17 We addressed these issues by including the hospital site variable in our regression models, regardless of its significance. Third, consent rates across sites vary from 70% to 85%. It is possible that patients who refused enrollment in the MCH trial are systematically different and may introduce bias in our analysis.

Furthermore, the study was designed as a natural experiment based on a rotational call cycle between hospitalist‐led and nonhospitalist‐led teams. It is possible that the order of patient assignment might not be completely naturally random as we intended. However, the study period was for 2 years and we expect the effect of order would have averaged out in time.

There are many hospitalist models of care. In terms of generalizability, the study pertains only to academic hospitalists and may not be applicable to hospitalists practicing in community hospitals. For example, the nonhospitalist comparison group is likely different in the community and academic settings. Community nonhospitalists (traditional practitioners) are usually internists covering both inpatient and outpatient responsibilities at the same time. In contrast, academic nonhospitalists are internists or subspecialists serving as ward attendings for a limited period (usually 1 month) with considerable variation in their nonattending responsibilities (eg, research, clinic, administration). Furthermore, academic nonhospitalist providers might be a self‐selected group by their willingness to serve as a ward attending, making them more hospitalist‐like. Changes and variability of inpatient attendings may also affect our findings when compared to prior work. Finally, it is also possible that having residents at academic medical centers may attenuate the effect of hospitalists more than in community‐based models.

Conclusions/Implications

Compared to nonhospitalists, academic hospitalist care of acute UGIH patients had similar overall clinical outcomes. However, our finding of similar LOS yet higher costs for patients cared for by hospitalists support 1 proposed mechanism in which hospitalists decrease healthcare costs: providing higher intensity of care per day of hospitalization. However, in academic hospitalist models, this higher intensity hypothesis should be revisited, especially in certain patient groups in which timing and involvement of subspecialists may influence discharge decisions, affecting LOS and overall costs.

Due to inherent limitations in this observational study, future studies should focus on verifying and elucidating these relationships further. Lastly, understanding which patient groups receive the greatest potential benefit from this model will help guide both organizational efforts and quality improvement strategies.

A 40‐year‐old man presented to the emergency department with a 5‐day history of fever, productive cough, and a painful rash. The rash was composed of grouped, raised, fluid‐filled vesicles with an erythematous base and honey‐colored crusting (Figure 1). The patient had a history of prior tuberculosis infection, illicit drug use, and human immunodeficiency virus (HIV). On initial physical examination, oral temperature was 98.5F, pulse was 110 beats/minute, respiratory rate was 22 breaths/minute, blood pressure was 151/79 mm Hg, and oxygen saturation (SpO₂) was 94%. Laboratory test results at admission were as follows: hemoglobin 10.6 g/dL; platelets 364,000 cells/L; white blood cell count 8,300 cells/L; CD4 count 132 cells/L; total serum protein 9.1 g/dL; albumin 2.0 g/dL; and lactate dehydrogenase (LDH) 158 IU/L. Chest radiograph showed diffuse pulmonary infiltrates bilaterally (Figure 2). Sputum cultures showed regular respiratory flora and no acid fast bacilli. Viral cultures of the vesicular lesions were positive for varicella zoster virus (VZV). The patient was started on acyclovir for VZV. After 2 days the patient's symptoms improved and he was subsequently discharged.

Figure 1

Figure 2

Herpes zoster is caused by the reactivation of a latent VZV infection in the dorsal root ganglion or cranial nerve ganglion. Zoster is characterized by an erythematous, vesicular, pustular rash in a unilateral dermatomal distribution. Immunosuppression is a risk factor for zoster; however, 92% of cases are in immunocompetent patients.1 Further, immunocompromised patients are more likely to develop disseminated zoster infection, including pneumonia, hepatitis, and encephalitis. These patients are also more likely to have secondary bacterial superinfections of cutaneous lesions and respiratory tracts.2 While zoster is often self‐limiting, it can lead to significant morbidity and mortality in immunocompromised patients. Intravenous acyclovir should be considered in patients with disseminated disease or visceral involvement, with advanced HIV, and in transplant patients being treated for rejection.2 Unilateral erythematous, vesicular rashes in a dermatomal distribution should yield a high clinical suspicion for herpes zoster, especially in immunocompromised patients.

A 40‐year‐old man presented to the emergency department with a 5‐day history of fever, productive cough, and a painful rash. The rash was composed of grouped, raised, fluid‐filled vesicles with an erythematous base and honey‐colored crusting (Figure 1). The patient had a history of prior tuberculosis infection, illicit drug use, and human immunodeficiency virus (HIV). On initial physical examination, oral temperature was 98.5F, pulse was 110 beats/minute, respiratory rate was 22 breaths/minute, blood pressure was 151/79 mm Hg, and oxygen saturation (SpO₂) was 94%. Laboratory test results at admission were as follows: hemoglobin 10.6 g/dL; platelets 364,000 cells/L; white blood cell count 8,300 cells/L; CD4 count 132 cells/L; total serum protein 9.1 g/dL; albumin 2.0 g/dL; and lactate dehydrogenase (LDH) 158 IU/L. Chest radiograph showed diffuse pulmonary infiltrates bilaterally (Figure 2). Sputum cultures showed regular respiratory flora and no acid fast bacilli. Viral cultures of the vesicular lesions were positive for varicella zoster virus (VZV). The patient was started on acyclovir for VZV. After 2 days the patient's symptoms improved and he was subsequently discharged.

Figure 1

Figure 2

Herpes zoster is caused by the reactivation of a latent VZV infection in the dorsal root ganglion or cranial nerve ganglion. Zoster is characterized by an erythematous, vesicular, pustular rash in a unilateral dermatomal distribution. Immunosuppression is a risk factor for zoster; however, 92% of cases are in immunocompetent patients.1 Further, immunocompromised patients are more likely to develop disseminated zoster infection, including pneumonia, hepatitis, and encephalitis. These patients are also more likely to have secondary bacterial superinfections of cutaneous lesions and respiratory tracts.2 While zoster is often self‐limiting, it can lead to significant morbidity and mortality in immunocompromised patients. Intravenous acyclovir should be considered in patients with disseminated disease or visceral involvement, with advanced HIV, and in transplant patients being treated for rejection.2 Unilateral erythematous, vesicular rashes in a dermatomal distribution should yield a high clinical suspicion for herpes zoster, especially in immunocompromised patients.

A 40‐year‐old man presented to the emergency department with a 5‐day history of fever, productive cough, and a painful rash. The rash was composed of grouped, raised, fluid‐filled vesicles with an erythematous base and honey‐colored crusting (Figure 1). The patient had a history of prior tuberculosis infection, illicit drug use, and human immunodeficiency virus (HIV). On initial physical examination, oral temperature was 98.5F, pulse was 110 beats/minute, respiratory rate was 22 breaths/minute, blood pressure was 151/79 mm Hg, and oxygen saturation (SpO₂) was 94%. Laboratory test results at admission were as follows: hemoglobin 10.6 g/dL; platelets 364,000 cells/L; white blood cell count 8,300 cells/L; CD4 count 132 cells/L; total serum protein 9.1 g/dL; albumin 2.0 g/dL; and lactate dehydrogenase (LDH) 158 IU/L. Chest radiograph showed diffuse pulmonary infiltrates bilaterally (Figure 2). Sputum cultures showed regular respiratory flora and no acid fast bacilli. Viral cultures of the vesicular lesions were positive for varicella zoster virus (VZV). The patient was started on acyclovir for VZV. After 2 days the patient's symptoms improved and he was subsequently discharged.

Figure 1

Figure 2

Herpes zoster is caused by the reactivation of a latent VZV infection in the dorsal root ganglion or cranial nerve ganglion. Zoster is characterized by an erythematous, vesicular, pustular rash in a unilateral dermatomal distribution. Immunosuppression is a risk factor for zoster; however, 92% of cases are in immunocompetent patients.1 Further, immunocompromised patients are more likely to develop disseminated zoster infection, including pneumonia, hepatitis, and encephalitis. These patients are also more likely to have secondary bacterial superinfections of cutaneous lesions and respiratory tracts.2 While zoster is often self‐limiting, it can lead to significant morbidity and mortality in immunocompromised patients. Intravenous acyclovir should be considered in patients with disseminated disease or visceral involvement, with advanced HIV, and in transplant patients being treated for rejection.2 Unilateral erythematous, vesicular rashes in a dermatomal distribution should yield a high clinical suspicion for herpes zoster, especially in immunocompromised patients.

Heparin induced thrombocytopenia (HIT) is a significant, potentially life‐threatening immune‐mediated adverse event that occurs several days after commencement of therapy with unfractionated or low‐molecular weight heparin. There are several potential sequelae of HIT, the most frequent of these is thrombosis, including but not limited to deep venous thrombosis (DVT), pulmonary embolism (PE), myocardial infarction, limb arterial occlusion, and disseminated intravascular coagulation. The prothrombotic state induced by HIT can be very significant, generating a thrombosis risk 30 times that of the general population and a mortality risk of 17% to 30% in those patients who develop thrombosis.1, 2

Case Report

A 51‐year‐old female was transferred to our institution for further management of a prothrombotic state. Six days prior to transfer, she presented to an outside hospital with significant edema and discomfort of her left lower extremity. She was found to have bilateral pulmonary emboli and a left lower extremity DVT. She was anticoagulated with unfractionated heparin and transitioned to coumadin. Upon preparation for discharge she developed drastically increased edema of her left lower extremity. Coumadin was discontinued and she was transferred to our institution for alternate anticoagulation and potential interventional vascular treatments.

On examination, the patient reported pain in her legs bilaterally but was in no distress. She had marked edema of the left lower extremity with tender erythematous skin over the anterior thigh with mild cyanosis and pallor of the left toes. Pulses were not palpable but could be identified by handheld Doppler scan. Urgent bilateral lower extremity venous and arterial duplex studies were completed, revealing extensive thrombosis involving the entire deep and superficial venous system on the left and the superficial femoral, popliteal, and peroneal veins on the right.

She was treated with an argatroban drip and a complete thrombophilia evaluation commenced. The following day she was mildly obtunded and slow to mentate. A noninfused computed tomography (CT) scan of the head revealed multiple acute left middle cerebral artery ischemic infarctions (Figure 1). CT scans of the chest, abdomen, and pelvis were done to assess for further thrombosis; bilateral renal infarcts were discovered.

Figure 1

The hypercoagulable workup revealed prothrombin and Factor V Leiden gene mutations and anticardiolipin immunoglobulin (Ig)G, IgA and IgM that were all negative; however, heparin‐dependent antibody platelet factor 4 (PF4) enzyme‐linked immunosorbent assay (ELISA) was positive. Her preheparin platelet count was 149,000/L, 185,000/L at the time of her transfer and thrombosis extension, and 117,000/L at its nadir, 11 days after initial heparin exposure.

Despite lower extremity thrombectomy, right common femoral endarterectomy, and therapeutic anticoagulation, the patient continued to develop massive thrombosis and she expired. The patient underwent autopsy, which confirmed her extensive thrombosis and cited multisystem organ failure as the cause of death. Additionally, this examination revealed an occult high‐grade cervical cancer with lymphatic invasion.

Discussion

This case is an example of multiorgan failure as a result of the prothrombotic state induced by HIT. The thrombocytopenia of HIT is defined as either a platelet count of less than 150,000/L or a decrease of greater than 50% from baseline.3, 4 Despite the eventual confirmation of the diagnosis by PF4 ELISA (sensitivity 80%‐90%), the patient was not thrombocytopenic by definition at the time of extension of her thrombosis.5

Greinacher et al.3 retrospectively evaluated 408 patients with thrombosis associated with HIT and found that at the time of their thrombosis 40.2% became thrombocytopenic (>50% decrease in their platelet count) 1 or more days prior to their initial thrombosis, 26% became thrombocytopenic on the day of their initial thrombosis, and 33.5% had thrombosis that preceded their thrombocytopenia with a 3‐day median delay between thrombosis and thrombocytopenia. Our patient fell in the latter category, developing her thrombocytopenia 5 days after the extension of her thrombosis. The time course of this presentation places emphasis on the need for clinicians to be aware of this pattern and to have a suspicion for HIT in patients on heparin who develop thrombosis regardless of their platelet count at the time of the thrombotic event.

In addition, our patient had the occult diagnosis of cervical cancer. In a retrospective review, Opatrny and Warner6 found that thrombotic complications associated with HIT, venous thrombosis, and PE specifically, occurred more frequently in patients with malignancy than those without malignant disease. They evaluated 64 patients with the diagnosis of HIT, made by heparin‐PF4 ELISA, and discovered the incidence of thrombosis to be 73% in the patients with malignancy compared to 30% in the patients without malignancy. However, since our patient's cancer diagnosis was unknown at the time of the case events, it could not be considered.

There have been rare case reports published describing patients who develop thrombosis secondary to heparin‐dependent antibodies (HDA) without meeting the above definition of thrombocytopenia in HIT. Bream‐Rouwenhorst and Hobbs7 recently reported a similar case in which a 35‐year‐old woman with bilateral lower extremity arterial thrombosis had additional thrombotic events after reexposure to heparin; the patient had a positive heparin‐PF4 ELISA with a platelet count that remained consistently above 200,000/L and never fell below 75% of her baseline. They cite only 22 additional cases of patients with HDA without thrombocytopenia reported in the literature since 1965 and suggest that the term heparin‐associated thrombosis without HIT may be a more appropriate terminology to describe similar cases.

Conclusions

Early recognition and initiation of alternate anticoagulation are essential to the effective management of HIT and prevention of its sequelae. The possible diagnosis of HIT is important for clinicians to keep in mind for all patients that are receiving any form of heparin, not only those patients who present with thrombocytopenia but also those with otherwise unexplainable thrombosis regardless of the platelet count.

Heparin induced thrombocytopenia (HIT) is a significant, potentially life‐threatening immune‐mediated adverse event that occurs several days after commencement of therapy with unfractionated or low‐molecular weight heparin. There are several potential sequelae of HIT, the most frequent of these is thrombosis, including but not limited to deep venous thrombosis (DVT), pulmonary embolism (PE), myocardial infarction, limb arterial occlusion, and disseminated intravascular coagulation. The prothrombotic state induced by HIT can be very significant, generating a thrombosis risk 30 times that of the general population and a mortality risk of 17% to 30% in those patients who develop thrombosis.1, 2

Case Report

A 51‐year‐old female was transferred to our institution for further management of a prothrombotic state. Six days prior to transfer, she presented to an outside hospital with significant edema and discomfort of her left lower extremity. She was found to have bilateral pulmonary emboli and a left lower extremity DVT. She was anticoagulated with unfractionated heparin and transitioned to coumadin. Upon preparation for discharge she developed drastically increased edema of her left lower extremity. Coumadin was discontinued and she was transferred to our institution for alternate anticoagulation and potential interventional vascular treatments.

On examination, the patient reported pain in her legs bilaterally but was in no distress. She had marked edema of the left lower extremity with tender erythematous skin over the anterior thigh with mild cyanosis and pallor of the left toes. Pulses were not palpable but could be identified by handheld Doppler scan. Urgent bilateral lower extremity venous and arterial duplex studies were completed, revealing extensive thrombosis involving the entire deep and superficial venous system on the left and the superficial femoral, popliteal, and peroneal veins on the right.

She was treated with an argatroban drip and a complete thrombophilia evaluation commenced. The following day she was mildly obtunded and slow to mentate. A noninfused computed tomography (CT) scan of the head revealed multiple acute left middle cerebral artery ischemic infarctions (Figure 1). CT scans of the chest, abdomen, and pelvis were done to assess for further thrombosis; bilateral renal infarcts were discovered.

Figure 1

The hypercoagulable workup revealed prothrombin and Factor V Leiden gene mutations and anticardiolipin immunoglobulin (Ig)G, IgA and IgM that were all negative; however, heparin‐dependent antibody platelet factor 4 (PF4) enzyme‐linked immunosorbent assay (ELISA) was positive. Her preheparin platelet count was 149,000/L, 185,000/L at the time of her transfer and thrombosis extension, and 117,000/L at its nadir, 11 days after initial heparin exposure.

Despite lower extremity thrombectomy, right common femoral endarterectomy, and therapeutic anticoagulation, the patient continued to develop massive thrombosis and she expired. The patient underwent autopsy, which confirmed her extensive thrombosis and cited multisystem organ failure as the cause of death. Additionally, this examination revealed an occult high‐grade cervical cancer with lymphatic invasion.

Discussion

This case is an example of multiorgan failure as a result of the prothrombotic state induced by HIT. The thrombocytopenia of HIT is defined as either a platelet count of less than 150,000/L or a decrease of greater than 50% from baseline.3, 4 Despite the eventual confirmation of the diagnosis by PF4 ELISA (sensitivity 80%‐90%), the patient was not thrombocytopenic by definition at the time of extension of her thrombosis.5

Greinacher et al.3 retrospectively evaluated 408 patients with thrombosis associated with HIT and found that at the time of their thrombosis 40.2% became thrombocytopenic (>50% decrease in their platelet count) 1 or more days prior to their initial thrombosis, 26% became thrombocytopenic on the day of their initial thrombosis, and 33.5% had thrombosis that preceded their thrombocytopenia with a 3‐day median delay between thrombosis and thrombocytopenia. Our patient fell in the latter category, developing her thrombocytopenia 5 days after the extension of her thrombosis. The time course of this presentation places emphasis on the need for clinicians to be aware of this pattern and to have a suspicion for HIT in patients on heparin who develop thrombosis regardless of their platelet count at the time of the thrombotic event.

In addition, our patient had the occult diagnosis of cervical cancer. In a retrospective review, Opatrny and Warner6 found that thrombotic complications associated with HIT, venous thrombosis, and PE specifically, occurred more frequently in patients with malignancy than those without malignant disease. They evaluated 64 patients with the diagnosis of HIT, made by heparin‐PF4 ELISA, and discovered the incidence of thrombosis to be 73% in the patients with malignancy compared to 30% in the patients without malignancy. However, since our patient's cancer diagnosis was unknown at the time of the case events, it could not be considered.

There have been rare case reports published describing patients who develop thrombosis secondary to heparin‐dependent antibodies (HDA) without meeting the above definition of thrombocytopenia in HIT. Bream‐Rouwenhorst and Hobbs7 recently reported a similar case in which a 35‐year‐old woman with bilateral lower extremity arterial thrombosis had additional thrombotic events after reexposure to heparin; the patient had a positive heparin‐PF4 ELISA with a platelet count that remained consistently above 200,000/L and never fell below 75% of her baseline. They cite only 22 additional cases of patients with HDA without thrombocytopenia reported in the literature since 1965 and suggest that the term heparin‐associated thrombosis without HIT may be a more appropriate terminology to describe similar cases.

Conclusions

Early recognition and initiation of alternate anticoagulation are essential to the effective management of HIT and prevention of its sequelae. The possible diagnosis of HIT is important for clinicians to keep in mind for all patients that are receiving any form of heparin, not only those patients who present with thrombocytopenia but also those with otherwise unexplainable thrombosis regardless of the platelet count.

Heparin induced thrombocytopenia (HIT) is a significant, potentially life‐threatening immune‐mediated adverse event that occurs several days after commencement of therapy with unfractionated or low‐molecular weight heparin. There are several potential sequelae of HIT, the most frequent of these is thrombosis, including but not limited to deep venous thrombosis (DVT), pulmonary embolism (PE), myocardial infarction, limb arterial occlusion, and disseminated intravascular coagulation. The prothrombotic state induced by HIT can be very significant, generating a thrombosis risk 30 times that of the general population and a mortality risk of 17% to 30% in those patients who develop thrombosis.1, 2

Case Report

A 51‐year‐old female was transferred to our institution for further management of a prothrombotic state. Six days prior to transfer, she presented to an outside hospital with significant edema and discomfort of her left lower extremity. She was found to have bilateral pulmonary emboli and a left lower extremity DVT. She was anticoagulated with unfractionated heparin and transitioned to coumadin. Upon preparation for discharge she developed drastically increased edema of her left lower extremity. Coumadin was discontinued and she was transferred to our institution for alternate anticoagulation and potential interventional vascular treatments.

On examination, the patient reported pain in her legs bilaterally but was in no distress. She had marked edema of the left lower extremity with tender erythematous skin over the anterior thigh with mild cyanosis and pallor of the left toes. Pulses were not palpable but could be identified by handheld Doppler scan. Urgent bilateral lower extremity venous and arterial duplex studies were completed, revealing extensive thrombosis involving the entire deep and superficial venous system on the left and the superficial femoral, popliteal, and peroneal veins on the right.

She was treated with an argatroban drip and a complete thrombophilia evaluation commenced. The following day she was mildly obtunded and slow to mentate. A noninfused computed tomography (CT) scan of the head revealed multiple acute left middle cerebral artery ischemic infarctions (Figure 1). CT scans of the chest, abdomen, and pelvis were done to assess for further thrombosis; bilateral renal infarcts were discovered.

Figure 1

The hypercoagulable workup revealed prothrombin and Factor V Leiden gene mutations and anticardiolipin immunoglobulin (Ig)G, IgA and IgM that were all negative; however, heparin‐dependent antibody platelet factor 4 (PF4) enzyme‐linked immunosorbent assay (ELISA) was positive. Her preheparin platelet count was 149,000/L, 185,000/L at the time of her transfer and thrombosis extension, and 117,000/L at its nadir, 11 days after initial heparin exposure.

Despite lower extremity thrombectomy, right common femoral endarterectomy, and therapeutic anticoagulation, the patient continued to develop massive thrombosis and she expired. The patient underwent autopsy, which confirmed her extensive thrombosis and cited multisystem organ failure as the cause of death. Additionally, this examination revealed an occult high‐grade cervical cancer with lymphatic invasion.

Discussion

This case is an example of multiorgan failure as a result of the prothrombotic state induced by HIT. The thrombocytopenia of HIT is defined as either a platelet count of less than 150,000/L or a decrease of greater than 50% from baseline.3, 4 Despite the eventual confirmation of the diagnosis by PF4 ELISA (sensitivity 80%‐90%), the patient was not thrombocytopenic by definition at the time of extension of her thrombosis.5

Greinacher et al.3 retrospectively evaluated 408 patients with thrombosis associated with HIT and found that at the time of their thrombosis 40.2% became thrombocytopenic (>50% decrease in their platelet count) 1 or more days prior to their initial thrombosis, 26% became thrombocytopenic on the day of their initial thrombosis, and 33.5% had thrombosis that preceded their thrombocytopenia with a 3‐day median delay between thrombosis and thrombocytopenia. Our patient fell in the latter category, developing her thrombocytopenia 5 days after the extension of her thrombosis. The time course of this presentation places emphasis on the need for clinicians to be aware of this pattern and to have a suspicion for HIT in patients on heparin who develop thrombosis regardless of their platelet count at the time of the thrombotic event.

In addition, our patient had the occult diagnosis of cervical cancer. In a retrospective review, Opatrny and Warner6 found that thrombotic complications associated with HIT, venous thrombosis, and PE specifically, occurred more frequently in patients with malignancy than those without malignant disease. They evaluated 64 patients with the diagnosis of HIT, made by heparin‐PF4 ELISA, and discovered the incidence of thrombosis to be 73% in the patients with malignancy compared to 30% in the patients without malignancy. However, since our patient's cancer diagnosis was unknown at the time of the case events, it could not be considered.

There have been rare case reports published describing patients who develop thrombosis secondary to heparin‐dependent antibodies (HDA) without meeting the above definition of thrombocytopenia in HIT. Bream‐Rouwenhorst and Hobbs7 recently reported a similar case in which a 35‐year‐old woman with bilateral lower extremity arterial thrombosis had additional thrombotic events after reexposure to heparin; the patient had a positive heparin‐PF4 ELISA with a platelet count that remained consistently above 200,000/L and never fell below 75% of her baseline. They cite only 22 additional cases of patients with HDA without thrombocytopenia reported in the literature since 1965 and suggest that the term heparin‐associated thrombosis without HIT may be a more appropriate terminology to describe similar cases.

Conclusions

Early recognition and initiation of alternate anticoagulation are essential to the effective management of HIT and prevention of its sequelae. The possible diagnosis of HIT is important for clinicians to keep in mind for all patients that are receiving any form of heparin, not only those patients who present with thrombocytopenia but also those with otherwise unexplainable thrombosis regardless of the platelet count.

A 34‐year‐old man was admitted for evaluation of elevated blood glucose despite extremely high subcutaneous (SQ) insulin requirements. He had a 12‐year history of Type 2 diabetes mellitus (T2DM) without episodes of ketoacidosis, managed initially with oral medications (metformin with various sulfonylureas and thiazolidinediones). Three months prior to admission, he was transitioned to SQ insulin and thereafter his requirements escalated rapidly. By the time of his admission, his blood glucose measurements were consistently above 300 mg/dL despite injecting more than 4100 units of insulin daily. His regimen included 300 units of insulin glargine (Lantus) 2 times per day (BID) and 1.75 mL of Humilin U‐500 Insulin (875 units) 4 times per day (QID). Past medical history included metabolic syndrome, nonalcoholic steatohepatitis, and diabetic neuropathy. Physical exam was remarkable for centripetal obesity (body mass index [BMI] = 38.9 kg/m²), acanthosis nigricans, and necrobiosis lipoidica diabeticorum (NLD) (Figure 1).

Figure 1

We undertook an investigation to characterize this extreme insulin resistance. After 24 hours without insulin supplementation, and 12 hours of nothing by mouth (NPO), his blood glucose level was 280 mg/dL and his serum insulin was 133.5 IU/mL. We injected 12 units of insulin Aspart and subsequently measured his serum glucose and insulin once more. His blood glucose level had risen to 289 mg/dL and his serum insulin fell to 110.7 IU/mL. We then transitioned the patient to intravenous (IV) insulin. After a series of boluses totaling 400 units, his blood glucose normalized (90 mg/dL) and was maintained in normal range on a rate of 48 units per hour. Over 24 hours, we had infused over 1400 units.

During this time, we also drew several labs. Serum antiinsulin antibodies were undetectable (ARUP Laboratories, Salt Lake City, UT). A full rheumatologic workup was negative for systemic lupus erythematosus (SLE), rheumatoid factor, Sjgren's syndrome (SS)‐A and SS‐B. Androgen levels were normal, as were 24‐hour urine collections for cortisol and metanephrines. The patient was discharged on a regimen of U‐500 without glargine.

By 5 months after discharge, his blood glucose remained uncontrolled despite increasing doses of U‐500 (with or without metformin and thiazolidinediones). The patient was offered a gastric bypass operation. Now, 4 months postoperative, his blood glucose is controlled, no greater than 90 mg/dL in the morning and 125 mg/dL in the evening. He is off insulin, taking 30 mg pioglitazone (Actos) daily and 500 mg metformin 3 times per day (TID).

Discussion

Extreme insulin resistance (EIR), defined by daily insulin requirements in excess of 200 U, is a rare and frustrating condition.1 Rarer still is extreme subcutaneous insulin resistance (ESIR). A systematic Medline review revealed only 29 reported cases of ESIR, all of which involved patients that maintained IV sensitivity to insulin. Classic diagnostic criteria for ESIR include preserved sensitivity to IV insulin, failure to increase serum insulin with subcutaneous injection, and insulin degrading activity of subcutaneous tissue.2, 3 However, there are, at present, no laboratory tests that can test the final criterion. Indeed, very few of the published reports of ESIR satisfy it, with most studies considering as diagnostic of ESIR the constellation of EIR with failure to raise serum insulin after injection and preserved intravenous insulin sensitivity.

As was evident in the high doses of IV insulin required for blood glucose normalization, our patient also had a proven receptor‐level peripheral resistance. Beyond the common, multifactorial insulin resistance of T2DM, the published reports of patients with extreme peripheral resistance are of 2 types: (A) genetic (eg, Leprechaunism) and (B) acquired autoimmune (Table 1).4 This patient fits neither category. Patients with Type A are very sick, with a syndromic disease that sharply curtails their life expectancy. Patients with Type B acquire antibodies directed against their insulin receptors and are almost invariably elderly African‐American women with severe rheumatological disease, namely SLE. We could not test our patient for an insulin‐receptor antibody secondary to prohibitive cost. This is probably moot, given that his autoimmune workup was negative and, as above, patients with such antibodies are vastly different compared to our patients.

Based on SQ insulin requirements, our patient had EIR. As his insulin levels failed to rise following an insulin injection, his EIR is thus subcutaneous in nature. However, among patients with this condition his failure to respond to IV insulin is unique. He does not fit criteria for types A or B insulin resistance; his condition is likely also due to an extreme version of the more common, multifactorial peripheral insulin resistance. This is supported by his successful response to the gastric bypass operation.5

The standard treatments for ESIR include: (1) concentrated regular insulin (U‐500) and (2) implantable intraperitoneal delivery; our patient received the former.6 U‐500 use in EIR has been shown to be more cost‐effective.1 Several reports have suggested success with protease inhibitors (aprotinin, nafamostat ointment), plasmapheresis, and intravenous immunoglobulin for extreme SQ resistance. Our case also represents the first treated successfully with a gastric bypass operation.

CONCLUSIONS

EIR can present a significant challenge for both the patient and hospitalist. The approach to this condition should begin with the determination of 24‐hour IV insulin requirement utilizing an insulin drip; serum insulin antibody evaluation; and endocrinology consultation. Our case also highlights a few important points about the broader management of diabetes mellitus. First, there are dermatological manifestations of diabetes that serve as potential markers for disease (namely acanthosis nigricans and NLD). Second, for patients with extreme insulin requirements, an extensive workup should be initiated and the patient should be transitioned to a concentrated regular insulin or intraperitoneal delivery. Third, our experience suggests a role for other measures such as gastric bypass that ought to be studied further.

A 34‐year‐old man was admitted for evaluation of elevated blood glucose despite extremely high subcutaneous (SQ) insulin requirements. He had a 12‐year history of Type 2 diabetes mellitus (T2DM) without episodes of ketoacidosis, managed initially with oral medications (metformin with various sulfonylureas and thiazolidinediones). Three months prior to admission, he was transitioned to SQ insulin and thereafter his requirements escalated rapidly. By the time of his admission, his blood glucose measurements were consistently above 300 mg/dL despite injecting more than 4100 units of insulin daily. His regimen included 300 units of insulin glargine (Lantus) 2 times per day (BID) and 1.75 mL of Humilin U‐500 Insulin (875 units) 4 times per day (QID). Past medical history included metabolic syndrome, nonalcoholic steatohepatitis, and diabetic neuropathy. Physical exam was remarkable for centripetal obesity (body mass index [BMI] = 38.9 kg/m²), acanthosis nigricans, and necrobiosis lipoidica diabeticorum (NLD) (Figure 1).

Figure 1

We undertook an investigation to characterize this extreme insulin resistance. After 24 hours without insulin supplementation, and 12 hours of nothing by mouth (NPO), his blood glucose level was 280 mg/dL and his serum insulin was 133.5 IU/mL. We injected 12 units of insulin Aspart and subsequently measured his serum glucose and insulin once more. His blood glucose level had risen to 289 mg/dL and his serum insulin fell to 110.7 IU/mL. We then transitioned the patient to intravenous (IV) insulin. After a series of boluses totaling 400 units, his blood glucose normalized (90 mg/dL) and was maintained in normal range on a rate of 48 units per hour. Over 24 hours, we had infused over 1400 units.

During this time, we also drew several labs. Serum antiinsulin antibodies were undetectable (ARUP Laboratories, Salt Lake City, UT). A full rheumatologic workup was negative for systemic lupus erythematosus (SLE), rheumatoid factor, Sjgren's syndrome (SS)‐A and SS‐B. Androgen levels were normal, as were 24‐hour urine collections for cortisol and metanephrines. The patient was discharged on a regimen of U‐500 without glargine.

By 5 months after discharge, his blood glucose remained uncontrolled despite increasing doses of U‐500 (with or without metformin and thiazolidinediones). The patient was offered a gastric bypass operation. Now, 4 months postoperative, his blood glucose is controlled, no greater than 90 mg/dL in the morning and 125 mg/dL in the evening. He is off insulin, taking 30 mg pioglitazone (Actos) daily and 500 mg metformin 3 times per day (TID).

Discussion

Extreme insulin resistance (EIR), defined by daily insulin requirements in excess of 200 U, is a rare and frustrating condition.1 Rarer still is extreme subcutaneous insulin resistance (ESIR). A systematic Medline review revealed only 29 reported cases of ESIR, all of which involved patients that maintained IV sensitivity to insulin. Classic diagnostic criteria for ESIR include preserved sensitivity to IV insulin, failure to increase serum insulin with subcutaneous injection, and insulin degrading activity of subcutaneous tissue.2, 3 However, there are, at present, no laboratory tests that can test the final criterion. Indeed, very few of the published reports of ESIR satisfy it, with most studies considering as diagnostic of ESIR the constellation of EIR with failure to raise serum insulin after injection and preserved intravenous insulin sensitivity.

As was evident in the high doses of IV insulin required for blood glucose normalization, our patient also had a proven receptor‐level peripheral resistance. Beyond the common, multifactorial insulin resistance of T2DM, the published reports of patients with extreme peripheral resistance are of 2 types: (A) genetic (eg, Leprechaunism) and (B) acquired autoimmune (Table 1).4 This patient fits neither category. Patients with Type A are very sick, with a syndromic disease that sharply curtails their life expectancy. Patients with Type B acquire antibodies directed against their insulin receptors and are almost invariably elderly African‐American women with severe rheumatological disease, namely SLE. We could not test our patient for an insulin‐receptor antibody secondary to prohibitive cost. This is probably moot, given that his autoimmune workup was negative and, as above, patients with such antibodies are vastly different compared to our patients.

Based on SQ insulin requirements, our patient had EIR. As his insulin levels failed to rise following an insulin injection, his EIR is thus subcutaneous in nature. However, among patients with this condition his failure to respond to IV insulin is unique. He does not fit criteria for types A or B insulin resistance; his condition is likely also due to an extreme version of the more common, multifactorial peripheral insulin resistance. This is supported by his successful response to the gastric bypass operation.5

The standard treatments for ESIR include: (1) concentrated regular insulin (U‐500) and (2) implantable intraperitoneal delivery; our patient received the former.6 U‐500 use in EIR has been shown to be more cost‐effective.1 Several reports have suggested success with protease inhibitors (aprotinin, nafamostat ointment), plasmapheresis, and intravenous immunoglobulin for extreme SQ resistance. Our case also represents the first treated successfully with a gastric bypass operation.

CONCLUSIONS

EIR can present a significant challenge for both the patient and hospitalist. The approach to this condition should begin with the determination of 24‐hour IV insulin requirement utilizing an insulin drip; serum insulin antibody evaluation; and endocrinology consultation. Our case also highlights a few important points about the broader management of diabetes mellitus. First, there are dermatological manifestations of diabetes that serve as potential markers for disease (namely acanthosis nigricans and NLD). Second, for patients with extreme insulin requirements, an extensive workup should be initiated and the patient should be transitioned to a concentrated regular insulin or intraperitoneal delivery. Third, our experience suggests a role for other measures such as gastric bypass that ought to be studied further.

A 34‐year‐old man was admitted for evaluation of elevated blood glucose despite extremely high subcutaneous (SQ) insulin requirements. He had a 12‐year history of Type 2 diabetes mellitus (T2DM) without episodes of ketoacidosis, managed initially with oral medications (metformin with various sulfonylureas and thiazolidinediones). Three months prior to admission, he was transitioned to SQ insulin and thereafter his requirements escalated rapidly. By the time of his admission, his blood glucose measurements were consistently above 300 mg/dL despite injecting more than 4100 units of insulin daily. His regimen included 300 units of insulin glargine (Lantus) 2 times per day (BID) and 1.75 mL of Humilin U‐500 Insulin (875 units) 4 times per day (QID). Past medical history included metabolic syndrome, nonalcoholic steatohepatitis, and diabetic neuropathy. Physical exam was remarkable for centripetal obesity (body mass index [BMI] = 38.9 kg/m²), acanthosis nigricans, and necrobiosis lipoidica diabeticorum (NLD) (Figure 1).

Figure 1

We undertook an investigation to characterize this extreme insulin resistance. After 24 hours without insulin supplementation, and 12 hours of nothing by mouth (NPO), his blood glucose level was 280 mg/dL and his serum insulin was 133.5 IU/mL. We injected 12 units of insulin Aspart and subsequently measured his serum glucose and insulin once more. His blood glucose level had risen to 289 mg/dL and his serum insulin fell to 110.7 IU/mL. We then transitioned the patient to intravenous (IV) insulin. After a series of boluses totaling 400 units, his blood glucose normalized (90 mg/dL) and was maintained in normal range on a rate of 48 units per hour. Over 24 hours, we had infused over 1400 units.

During this time, we also drew several labs. Serum antiinsulin antibodies were undetectable (ARUP Laboratories, Salt Lake City, UT). A full rheumatologic workup was negative for systemic lupus erythematosus (SLE), rheumatoid factor, Sjgren's syndrome (SS)‐A and SS‐B. Androgen levels were normal, as were 24‐hour urine collections for cortisol and metanephrines. The patient was discharged on a regimen of U‐500 without glargine.

By 5 months after discharge, his blood glucose remained uncontrolled despite increasing doses of U‐500 (with or without metformin and thiazolidinediones). The patient was offered a gastric bypass operation. Now, 4 months postoperative, his blood glucose is controlled, no greater than 90 mg/dL in the morning and 125 mg/dL in the evening. He is off insulin, taking 30 mg pioglitazone (Actos) daily and 500 mg metformin 3 times per day (TID).

Discussion

Extreme insulin resistance (EIR), defined by daily insulin requirements in excess of 200 U, is a rare and frustrating condition.1 Rarer still is extreme subcutaneous insulin resistance (ESIR). A systematic Medline review revealed only 29 reported cases of ESIR, all of which involved patients that maintained IV sensitivity to insulin. Classic diagnostic criteria for ESIR include preserved sensitivity to IV insulin, failure to increase serum insulin with subcutaneous injection, and insulin degrading activity of subcutaneous tissue.2, 3 However, there are, at present, no laboratory tests that can test the final criterion. Indeed, very few of the published reports of ESIR satisfy it, with most studies considering as diagnostic of ESIR the constellation of EIR with failure to raise serum insulin after injection and preserved intravenous insulin sensitivity.

As was evident in the high doses of IV insulin required for blood glucose normalization, our patient also had a proven receptor‐level peripheral resistance. Beyond the common, multifactorial insulin resistance of T2DM, the published reports of patients with extreme peripheral resistance are of 2 types: (A) genetic (eg, Leprechaunism) and (B) acquired autoimmune (Table 1).4 This patient fits neither category. Patients with Type A are very sick, with a syndromic disease that sharply curtails their life expectancy. Patients with Type B acquire antibodies directed against their insulin receptors and are almost invariably elderly African‐American women with severe rheumatological disease, namely SLE. We could not test our patient for an insulin‐receptor antibody secondary to prohibitive cost. This is probably moot, given that his autoimmune workup was negative and, as above, patients with such antibodies are vastly different compared to our patients.

Based on SQ insulin requirements, our patient had EIR. As his insulin levels failed to rise following an insulin injection, his EIR is thus subcutaneous in nature. However, among patients with this condition his failure to respond to IV insulin is unique. He does not fit criteria for types A or B insulin resistance; his condition is likely also due to an extreme version of the more common, multifactorial peripheral insulin resistance. This is supported by his successful response to the gastric bypass operation.5

The standard treatments for ESIR include: (1) concentrated regular insulin (U‐500) and (2) implantable intraperitoneal delivery; our patient received the former.6 U‐500 use in EIR has been shown to be more cost‐effective.1 Several reports have suggested success with protease inhibitors (aprotinin, nafamostat ointment), plasmapheresis, and intravenous immunoglobulin for extreme SQ resistance. Our case also represents the first treated successfully with a gastric bypass operation.

CONCLUSIONS

EIR can present a significant challenge for both the patient and hospitalist. The approach to this condition should begin with the determination of 24‐hour IV insulin requirement utilizing an insulin drip; serum insulin antibody evaluation; and endocrinology consultation. Our case also highlights a few important points about the broader management of diabetes mellitus. First, there are dermatological manifestations of diabetes that serve as potential markers for disease (namely acanthosis nigricans and NLD). Second, for patients with extreme insulin requirements, an extensive workup should be initiated and the patient should be transitioned to a concentrated regular insulin or intraperitoneal delivery. Third, our experience suggests a role for other measures such as gastric bypass that ought to be studied further.

Ultrasound, one of the most reliable diagnostic technologies in medicine, has a unique long‐term safety profile across a wide spectrum of applications. In line with the trend toward the miniaturization of many other technologies, increasingly sophisticated hand‐held or hand‐carried ultrasound (HCU) devices have become widely available. To date, the U.S. Food and Drug Administration (FDA) has approved more than 10 new‐generation portable (1.0‐4.5 kg) ultrasound devices, and a recent industry report projected that the HCU market will see revenues in excess of $1 billion by 2011.1

Although cardiovascular assessment remains its primary use, hospitalist physicians are increasingly turning to this technology for the localization of fluid and other abnormalities prior to paracentesis and thoracentesis. While there are other potential uses (eg, managing acute scrotal pain, diagnosing meniscal tears, measuring carotid intimal thickness), the higher‐quality studies of hospitalist‐physicians' use of HCU have focused on cardiovascular assessment. HCU confers a number of potential workflow‐related advantages, including coordinated point‐of‐care evaluation at short notice when formal ultrasound may be unavailable, as well as circumvention of the need to call on radiology or cardiology specialists.2 Even for experienced cardiologists, heart failure can be difficult to identify using any modality, and the clinical diagnosis of cardiovascular disease by hospital physicians has been documented as poor.3, 4 Thus, the addition of HCU to the palette of diagnostic and teaching tools available to frontline physicians potentially offers improvements over stethoscope‐assisted physical examination alone (including visual inspection, palpation, and auscultation), which has remained essentially unaltered for 150 years.57

Evidence Base for HCU Use by Hospitalists

The few primary studies on HCU use by hospitalists have focused on the potential utility of this technology as a valuable adjunct to the physical exam for the detection of cardiovascular disease (eg, asymptomatic left ventricular [LV] dysfunction, cardiomegaly, pericardial effusion) in the ambulatory or acute care setting.8, 9 Operation of HCU by hospitalists is not clearly indicated for the evaluation of valvular disease (eg, aortic and mitral regurgitation), in part due to the limited Doppler capabilities of the smaller devices.911 The risk of a gradual erosion of physical exam skills accompanying expansion of HCU use by hospitalists could itself become a potential disadvantage of a premature replacement of the stethoscope, since the results obtained by hospitalists performing a standard physical exam have been shown to be better than those obtained with HCU.8, 9

The lack of large, multicenter studies of HCU use by hospitalists leaves many questions unanswered, including whether or not the relatively low initial cost of an HCU device ($9,000‐$50,000) vs. that of a full‐sized hospital ultrasound system ($250,000) will eventually translate into overall cost‐effectiveness or actual patient‐centered benefit.10 While cautious advocates have insisted that HCU provides additive information in conjunction with the physical exam, this approach is not meant to serve as a substitute for standard echocardiography in patients requiring full evaluation in inpatient settings relevant for hospitalists.1114 Referral for additional testing or specialist opinionsand the associated costs incurredcannot necessarily be circumvented by hospitalist‐operated HCU.

A major problem with the HCU literature in general is its lack of standardization betweenand withinstudies, which renders it nearly impossible to generalize findings about important clinical outcomes, patient satisfaction, quality‐of‐life, symptoms, physical functioning, and morbidity and mortality. There are a preponderance of underpowered, methodologically inconsistent, single‐center case series that do not evaluate diagnostic accuracy in terms of patient outcomes. For example, although one study did find a modest (22‐29%) reduction in department workload with HCU, the authors omitted important information regarding blinding, and no power calculations were reported; thus, it was not possible to ascertain whether or not the reported results were due to the intervention or to chance.15 There clearly remains a need to convincingly demonstrate that patient care, shortening of length of stay, long‐term prognosis, or potential financial savings could occur with use of these devices by hospitalists.5 The process of device acquisition and resource allocation is, at least in part, based on accumulated evidence from studies that have ill‐defined relevant outcomes (eg, left ventricular function). However, even if such outcomes were to be more closely examined, medical decision‐making would still suffer from discrepant findings due to numerous differences in study design, including parameters involving patient population and selection, setting (eg, echocardiography laboratory vs. critical care unit), provider background, and specific device(s) used.

Training Issues

Hospitalist proficiency across HCU imaging skills (ie, acquisition, measurement, interpretation) has been found to be inconsistent.9 Endorsement and expansion of hospitalist use of HCU may to some extent reflect an overgeneralization from disparate comparative studies showing moderate success obtained with HCU (vs. physical exam) by other practitioner groups such as medical students and fellows with limited experience.16, 17 Whereas in 2005, Hellmann et al.18 concluded that medical residents with minimal training can learn to perform some of the basic functions of HCU with reasonable accuracy, Martin et al.8, 9 (in 2007 and 2009) reported conflicting results from a study of hospitalists trained at the same institution.

Concern about switching from standard to nonstandard HCU operators is raised by studies in which specialized operators (eg, echocardiography technicians) obtained better results than hospitalists using these devices.8, 9 In 2004, Borges et al.19 reported the results of 315 patients referred to specialists at a cardiology clinic for preoperative assessment prior to noncardiac surgery; the results (94.8% and 96.7% agreement with standard echocardiography on the main echocardiographic finding and detection of valve disease, respectively) were attributed to the fact that experienced cardiologists were working under ideal conditions using only the most advanced HCU devices with Doppler as well as harmonic imaging capabilities. Likewise, in 2004, Tsutsui et al.20 studied 44 consecutive hospitalized patients who underwent comprehensive echocardiography and bedside HCU. They reported that hemodynamic assessment by HCU was poor, even when performed by practitioners with relatively high levels of training.20 In 2003, DeCara et al.12 performed standard echocardiography on 300 adult inpatients referred for imaging, and concluded that standardized training, competency testing, and quality assurance guidelines need to be established before these devices can be utilized for clinical decision‐making by physicians without formal training in echocardiography. Although there have been numerous calls for training guidelines, it has not yet been determined how much training would be optimalor even necessaryfor professionals of each subspecialty to achieve levels of accuracy that are acceptable. Furthermore, it is well known that skill level declines unless a technique is regularly reinforced with practice, and therefore, recertification or procedure volume standards should be established.

The issue of potential harm needs to be raised, if hospitalists with access to HCU are indeed less accurate in their diagnoses than trained cardiologists interpreting images acquired by an established alternative such as echocardiography. False negatives can lead to delayed treatment, and false positives to unwarranted treatment. Given that the treatment effects of HCU use by hospitalists have not been closely scrutinized, the expansion of such use appears unwarranted, at least until further randomized studies with well‐defined outcomes have been conducted. Although the HCU devices themselves have a good safety profile, their potential benefits and harms (eg, possibility of increased nosocomial infection) will ultimately reflect operator skill and their impact on patient management relative to the gold‐standard diagnostic modalities for which there is abundant evidence of safety and efficacy.21

Premarketing and Postmarketing Concerns

The controversy regarding hospitalist use of HCU exposes gaps in the FDA approval process for medical devices, which are subjected to much less rigorous scrutiny during the premarketing approval process than pharmaceuticals.22 Moreover, the aggressive marketing of newly approved devices (and drugs) can drive medically unwarranted overuse, or indication creep, which justifies calls for the establishment of rigorous standards of clinical relevance and practice.23, 24 While the available literature on HCU operation by hospitalists is focused on cardiovascular indications for the technology, hospital medicine physicians are increasingly using HCU to guide paracentesis and thoracentesis. Given how commonplace the expansion of such practices has become, it is noteworthy that HCU operation by hospitalists has not yet been evaluated and endorsed in larger, controlled trials demonstrating appropriate outcomes.25

Across all fields of medicine, the transition from traditional to newer modalities remains a slippery slope in terms of demonstration of persuasive evidence of patient‐centered benefit.26 Fascination with emerging technologies (so‐called gizmo idolatry) and increased reimbursement potential threaten to distract patients and their providers from legitimate concerns about how medical device manufacturers and for‐profit corporations increasingly influence device acquisition and clinical practice.2731 While we lack strong evidence demonstrating that diagnostic tests such as HCU are beneficial when performed by hospitalists, the expanded use of these handy new devices by hospitalists is simultaneously generating increased incidental and equivocal findings, which in turn render it necessary to go back and perform secondary verification studies by specialists using older, gold‐standard modalities. This vicious cycle, coupled with the current lack of evidence, will continue to degrade confidence in the initiation of either acute or chronic treatment on the basis of HCU results obtained by hospitalist physicians.

Eventually, the increased use of HCU by hospitalists might lead to demonstrations of improved hospital workflow management, but it may just as easily represent another new coupling of technology and practitioner that prematurely becomes the standard of care in the absence of any demonstration of added value. The initially enthusiastic application of pulmonary artery catheters (PACs) serves as a cautionary tale in which the acquisition of additional clinical data did not necessarily lead to improved clinical outcomes: whereas PACs did enhance the clinical understanding of hemodynamics, they were not associated with an overall advantage in terms of mortality, length of hospital stay, or cost.3235 Ultimately, more information is not necessarily better information. Although new medical technologies can produce extremely useful diagnostic results that aid in the management of critically ill patients, poor data interpretation resulting from lack of targeted training and experience can nullify point‐of‐care advantages, and perhaps lead to excess morbidity and mortality.14 In clinical practice, it is generally best to avoid reliance on assumptions of added value in lieu of demonstrations of the same.

Conclusions

Hospital practitioners should not yet put away their stethoscopes. New technologies such as HCU need to be embraced in parallel with accumulating evidence of benefit. In the hands of hospitalists, the smaller HCU devices may very well prove handy, but at present, the literature simply does not support the use of HCU by hospitalist physicians.

Ultrasound, one of the most reliable diagnostic technologies in medicine, has a unique long‐term safety profile across a wide spectrum of applications. In line with the trend toward the miniaturization of many other technologies, increasingly sophisticated hand‐held or hand‐carried ultrasound (HCU) devices have become widely available. To date, the U.S. Food and Drug Administration (FDA) has approved more than 10 new‐generation portable (1.0‐4.5 kg) ultrasound devices, and a recent industry report projected that the HCU market will see revenues in excess of $1 billion by 2011.1

Although cardiovascular assessment remains its primary use, hospitalist physicians are increasingly turning to this technology for the localization of fluid and other abnormalities prior to paracentesis and thoracentesis. While there are other potential uses (eg, managing acute scrotal pain, diagnosing meniscal tears, measuring carotid intimal thickness), the higher‐quality studies of hospitalist‐physicians' use of HCU have focused on cardiovascular assessment. HCU confers a number of potential workflow‐related advantages, including coordinated point‐of‐care evaluation at short notice when formal ultrasound may be unavailable, as well as circumvention of the need to call on radiology or cardiology specialists.2 Even for experienced cardiologists, heart failure can be difficult to identify using any modality, and the clinical diagnosis of cardiovascular disease by hospital physicians has been documented as poor.3, 4 Thus, the addition of HCU to the palette of diagnostic and teaching tools available to frontline physicians potentially offers improvements over stethoscope‐assisted physical examination alone (including visual inspection, palpation, and auscultation), which has remained essentially unaltered for 150 years.57

Evidence Base for HCU Use by Hospitalists

The few primary studies on HCU use by hospitalists have focused on the potential utility of this technology as a valuable adjunct to the physical exam for the detection of cardiovascular disease (eg, asymptomatic left ventricular [LV] dysfunction, cardiomegaly, pericardial effusion) in the ambulatory or acute care setting.8, 9 Operation of HCU by hospitalists is not clearly indicated for the evaluation of valvular disease (eg, aortic and mitral regurgitation), in part due to the limited Doppler capabilities of the smaller devices.911 The risk of a gradual erosion of physical exam skills accompanying expansion of HCU use by hospitalists could itself become a potential disadvantage of a premature replacement of the stethoscope, since the results obtained by hospitalists performing a standard physical exam have been shown to be better than those obtained with HCU.8, 9

The lack of large, multicenter studies of HCU use by hospitalists leaves many questions unanswered, including whether or not the relatively low initial cost of an HCU device ($9,000‐$50,000) vs. that of a full‐sized hospital ultrasound system ($250,000) will eventually translate into overall cost‐effectiveness or actual patient‐centered benefit.10 While cautious advocates have insisted that HCU provides additive information in conjunction with the physical exam, this approach is not meant to serve as a substitute for standard echocardiography in patients requiring full evaluation in inpatient settings relevant for hospitalists.1114 Referral for additional testing or specialist opinionsand the associated costs incurredcannot necessarily be circumvented by hospitalist‐operated HCU.

A major problem with the HCU literature in general is its lack of standardization betweenand withinstudies, which renders it nearly impossible to generalize findings about important clinical outcomes, patient satisfaction, quality‐of‐life, symptoms, physical functioning, and morbidity and mortality. There are a preponderance of underpowered, methodologically inconsistent, single‐center case series that do not evaluate diagnostic accuracy in terms of patient outcomes. For example, although one study did find a modest (22‐29%) reduction in department workload with HCU, the authors omitted important information regarding blinding, and no power calculations were reported; thus, it was not possible to ascertain whether or not the reported results were due to the intervention or to chance.15 There clearly remains a need to convincingly demonstrate that patient care, shortening of length of stay, long‐term prognosis, or potential financial savings could occur with use of these devices by hospitalists.5 The process of device acquisition and resource allocation is, at least in part, based on accumulated evidence from studies that have ill‐defined relevant outcomes (eg, left ventricular function). However, even if such outcomes were to be more closely examined, medical decision‐making would still suffer from discrepant findings due to numerous differences in study design, including parameters involving patient population and selection, setting (eg, echocardiography laboratory vs. critical care unit), provider background, and specific device(s) used.

Training Issues

Hospitalist proficiency across HCU imaging skills (ie, acquisition, measurement, interpretation) has been found to be inconsistent.9 Endorsement and expansion of hospitalist use of HCU may to some extent reflect an overgeneralization from disparate comparative studies showing moderate success obtained with HCU (vs. physical exam) by other practitioner groups such as medical students and fellows with limited experience.16, 17 Whereas in 2005, Hellmann et al.18 concluded that medical residents with minimal training can learn to perform some of the basic functions of HCU with reasonable accuracy, Martin et al.8, 9 (in 2007 and 2009) reported conflicting results from a study of hospitalists trained at the same institution.

Concern about switching from standard to nonstandard HCU operators is raised by studies in which specialized operators (eg, echocardiography technicians) obtained better results than hospitalists using these devices.8, 9 In 2004, Borges et al.19 reported the results of 315 patients referred to specialists at a cardiology clinic for preoperative assessment prior to noncardiac surgery; the results (94.8% and 96.7% agreement with standard echocardiography on the main echocardiographic finding and detection of valve disease, respectively) were attributed to the fact that experienced cardiologists were working under ideal conditions using only the most advanced HCU devices with Doppler as well as harmonic imaging capabilities. Likewise, in 2004, Tsutsui et al.20 studied 44 consecutive hospitalized patients who underwent comprehensive echocardiography and bedside HCU. They reported that hemodynamic assessment by HCU was poor, even when performed by practitioners with relatively high levels of training.20 In 2003, DeCara et al.12 performed standard echocardiography on 300 adult inpatients referred for imaging, and concluded that standardized training, competency testing, and quality assurance guidelines need to be established before these devices can be utilized for clinical decision‐making by physicians without formal training in echocardiography. Although there have been numerous calls for training guidelines, it has not yet been determined how much training would be optimalor even necessaryfor professionals of each subspecialty to achieve levels of accuracy that are acceptable. Furthermore, it is well known that skill level declines unless a technique is regularly reinforced with practice, and therefore, recertification or procedure volume standards should be established.

The issue of potential harm needs to be raised, if hospitalists with access to HCU are indeed less accurate in their diagnoses than trained cardiologists interpreting images acquired by an established alternative such as echocardiography. False negatives can lead to delayed treatment, and false positives to unwarranted treatment. Given that the treatment effects of HCU use by hospitalists have not been closely scrutinized, the expansion of such use appears unwarranted, at least until further randomized studies with well‐defined outcomes have been conducted. Although the HCU devices themselves have a good safety profile, their potential benefits and harms (eg, possibility of increased nosocomial infection) will ultimately reflect operator skill and their impact on patient management relative to the gold‐standard diagnostic modalities for which there is abundant evidence of safety and efficacy.21

Premarketing and Postmarketing Concerns

The controversy regarding hospitalist use of HCU exposes gaps in the FDA approval process for medical devices, which are subjected to much less rigorous scrutiny during the premarketing approval process than pharmaceuticals.22 Moreover, the aggressive marketing of newly approved devices (and drugs) can drive medically unwarranted overuse, or indication creep, which justifies calls for the establishment of rigorous standards of clinical relevance and practice.23, 24 While the available literature on HCU operation by hospitalists is focused on cardiovascular indications for the technology, hospital medicine physicians are increasingly using HCU to guide paracentesis and thoracentesis. Given how commonplace the expansion of such practices has become, it is noteworthy that HCU operation by hospitalists has not yet been evaluated and endorsed in larger, controlled trials demonstrating appropriate outcomes.25

Across all fields of medicine, the transition from traditional to newer modalities remains a slippery slope in terms of demonstration of persuasive evidence of patient‐centered benefit.26 Fascination with emerging technologies (so‐called gizmo idolatry) and increased reimbursement potential threaten to distract patients and their providers from legitimate concerns about how medical device manufacturers and for‐profit corporations increasingly influence device acquisition and clinical practice.2731 While we lack strong evidence demonstrating that diagnostic tests such as HCU are beneficial when performed by hospitalists, the expanded use of these handy new devices by hospitalists is simultaneously generating increased incidental and equivocal findings, which in turn render it necessary to go back and perform secondary verification studies by specialists using older, gold‐standard modalities. This vicious cycle, coupled with the current lack of evidence, will continue to degrade confidence in the initiation of either acute or chronic treatment on the basis of HCU results obtained by hospitalist physicians.

Eventually, the increased use of HCU by hospitalists might lead to demonstrations of improved hospital workflow management, but it may just as easily represent another new coupling of technology and practitioner that prematurely becomes the standard of care in the absence of any demonstration of added value. The initially enthusiastic application of pulmonary artery catheters (PACs) serves as a cautionary tale in which the acquisition of additional clinical data did not necessarily lead to improved clinical outcomes: whereas PACs did enhance the clinical understanding of hemodynamics, they were not associated with an overall advantage in terms of mortality, length of hospital stay, or cost.3235 Ultimately, more information is not necessarily better information. Although new medical technologies can produce extremely useful diagnostic results that aid in the management of critically ill patients, poor data interpretation resulting from lack of targeted training and experience can nullify point‐of‐care advantages, and perhaps lead to excess morbidity and mortality.14 In clinical practice, it is generally best to avoid reliance on assumptions of added value in lieu of demonstrations of the same.

Conclusions

Hospital practitioners should not yet put away their stethoscopes. New technologies such as HCU need to be embraced in parallel with accumulating evidence of benefit. In the hands of hospitalists, the smaller HCU devices may very well prove handy, but at present, the literature simply does not support the use of HCU by hospitalist physicians.

Ultrasound, one of the most reliable diagnostic technologies in medicine, has a unique long‐term safety profile across a wide spectrum of applications. In line with the trend toward the miniaturization of many other technologies, increasingly sophisticated hand‐held or hand‐carried ultrasound (HCU) devices have become widely available. To date, the U.S. Food and Drug Administration (FDA) has approved more than 10 new‐generation portable (1.0‐4.5 kg) ultrasound devices, and a recent industry report projected that the HCU market will see revenues in excess of $1 billion by 2011.1

Although cardiovascular assessment remains its primary use, hospitalist physicians are increasingly turning to this technology for the localization of fluid and other abnormalities prior to paracentesis and thoracentesis. While there are other potential uses (eg, managing acute scrotal pain, diagnosing meniscal tears, measuring carotid intimal thickness), the higher‐quality studies of hospitalist‐physicians' use of HCU have focused on cardiovascular assessment. HCU confers a number of potential workflow‐related advantages, including coordinated point‐of‐care evaluation at short notice when formal ultrasound may be unavailable, as well as circumvention of the need to call on radiology or cardiology specialists.2 Even for experienced cardiologists, heart failure can be difficult to identify using any modality, and the clinical diagnosis of cardiovascular disease by hospital physicians has been documented as poor.3, 4 Thus, the addition of HCU to the palette of diagnostic and teaching tools available to frontline physicians potentially offers improvements over stethoscope‐assisted physical examination alone (including visual inspection, palpation, and auscultation), which has remained essentially unaltered for 150 years.57

Evidence Base for HCU Use by Hospitalists

The few primary studies on HCU use by hospitalists have focused on the potential utility of this technology as a valuable adjunct to the physical exam for the detection of cardiovascular disease (eg, asymptomatic left ventricular [LV] dysfunction, cardiomegaly, pericardial effusion) in the ambulatory or acute care setting.8, 9 Operation of HCU by hospitalists is not clearly indicated for the evaluation of valvular disease (eg, aortic and mitral regurgitation), in part due to the limited Doppler capabilities of the smaller devices.911 The risk of a gradual erosion of physical exam skills accompanying expansion of HCU use by hospitalists could itself become a potential disadvantage of a premature replacement of the stethoscope, since the results obtained by hospitalists performing a standard physical exam have been shown to be better than those obtained with HCU.8, 9

The lack of large, multicenter studies of HCU use by hospitalists leaves many questions unanswered, including whether or not the relatively low initial cost of an HCU device ($9,000‐$50,000) vs. that of a full‐sized hospital ultrasound system ($250,000) will eventually translate into overall cost‐effectiveness or actual patient‐centered benefit.10 While cautious advocates have insisted that HCU provides additive information in conjunction with the physical exam, this approach is not meant to serve as a substitute for standard echocardiography in patients requiring full evaluation in inpatient settings relevant for hospitalists.1114 Referral for additional testing or specialist opinionsand the associated costs incurredcannot necessarily be circumvented by hospitalist‐operated HCU.

A major problem with the HCU literature in general is its lack of standardization betweenand withinstudies, which renders it nearly impossible to generalize findings about important clinical outcomes, patient satisfaction, quality‐of‐life, symptoms, physical functioning, and morbidity and mortality. There are a preponderance of underpowered, methodologically inconsistent, single‐center case series that do not evaluate diagnostic accuracy in terms of patient outcomes. For example, although one study did find a modest (22‐29%) reduction in department workload with HCU, the authors omitted important information regarding blinding, and no power calculations were reported; thus, it was not possible to ascertain whether or not the reported results were due to the intervention or to chance.15 There clearly remains a need to convincingly demonstrate that patient care, shortening of length of stay, long‐term prognosis, or potential financial savings could occur with use of these devices by hospitalists.5 The process of device acquisition and resource allocation is, at least in part, based on accumulated evidence from studies that have ill‐defined relevant outcomes (eg, left ventricular function). However, even if such outcomes were to be more closely examined, medical decision‐making would still suffer from discrepant findings due to numerous differences in study design, including parameters involving patient population and selection, setting (eg, echocardiography laboratory vs. critical care unit), provider background, and specific device(s) used.

Training Issues

Hospitalist proficiency across HCU imaging skills (ie, acquisition, measurement, interpretation) has been found to be inconsistent.9 Endorsement and expansion of hospitalist use of HCU may to some extent reflect an overgeneralization from disparate comparative studies showing moderate success obtained with HCU (vs. physical exam) by other practitioner groups such as medical students and fellows with limited experience.16, 17 Whereas in 2005, Hellmann et al.18 concluded that medical residents with minimal training can learn to perform some of the basic functions of HCU with reasonable accuracy, Martin et al.8, 9 (in 2007 and 2009) reported conflicting results from a study of hospitalists trained at the same institution.

Concern about switching from standard to nonstandard HCU operators is raised by studies in which specialized operators (eg, echocardiography technicians) obtained better results than hospitalists using these devices.8, 9 In 2004, Borges et al.19 reported the results of 315 patients referred to specialists at a cardiology clinic for preoperative assessment prior to noncardiac surgery; the results (94.8% and 96.7% agreement with standard echocardiography on the main echocardiographic finding and detection of valve disease, respectively) were attributed to the fact that experienced cardiologists were working under ideal conditions using only the most advanced HCU devices with Doppler as well as harmonic imaging capabilities. Likewise, in 2004, Tsutsui et al.20 studied 44 consecutive hospitalized patients who underwent comprehensive echocardiography and bedside HCU. They reported that hemodynamic assessment by HCU was poor, even when performed by practitioners with relatively high levels of training.20 In 2003, DeCara et al.12 performed standard echocardiography on 300 adult inpatients referred for imaging, and concluded that standardized training, competency testing, and quality assurance guidelines need to be established before these devices can be utilized for clinical decision‐making by physicians without formal training in echocardiography. Although there have been numerous calls for training guidelines, it has not yet been determined how much training would be optimalor even necessaryfor professionals of each subspecialty to achieve levels of accuracy that are acceptable. Furthermore, it is well known that skill level declines unless a technique is regularly reinforced with practice, and therefore, recertification or procedure volume standards should be established.

The issue of potential harm needs to be raised, if hospitalists with access to HCU are indeed less accurate in their diagnoses than trained cardiologists interpreting images acquired by an established alternative such as echocardiography. False negatives can lead to delayed treatment, and false positives to unwarranted treatment. Given that the treatment effects of HCU use by hospitalists have not been closely scrutinized, the expansion of such use appears unwarranted, at least until further randomized studies with well‐defined outcomes have been conducted. Although the HCU devices themselves have a good safety profile, their potential benefits and harms (eg, possibility of increased nosocomial infection) will ultimately reflect operator skill and their impact on patient management relative to the gold‐standard diagnostic modalities for which there is abundant evidence of safety and efficacy.21

Premarketing and Postmarketing Concerns

The controversy regarding hospitalist use of HCU exposes gaps in the FDA approval process for medical devices, which are subjected to much less rigorous scrutiny during the premarketing approval process than pharmaceuticals.22 Moreover, the aggressive marketing of newly approved devices (and drugs) can drive medically unwarranted overuse, or indication creep, which justifies calls for the establishment of rigorous standards of clinical relevance and practice.23, 24 While the available literature on HCU operation by hospitalists is focused on cardiovascular indications for the technology, hospital medicine physicians are increasingly using HCU to guide paracentesis and thoracentesis. Given how commonplace the expansion of such practices has become, it is noteworthy that HCU operation by hospitalists has not yet been evaluated and endorsed in larger, controlled trials demonstrating appropriate outcomes.25

Across all fields of medicine, the transition from traditional to newer modalities remains a slippery slope in terms of demonstration of persuasive evidence of patient‐centered benefit.26 Fascination with emerging technologies (so‐called gizmo idolatry) and increased reimbursement potential threaten to distract patients and their providers from legitimate concerns about how medical device manufacturers and for‐profit corporations increasingly influence device acquisition and clinical practice.2731 While we lack strong evidence demonstrating that diagnostic tests such as HCU are beneficial when performed by hospitalists, the expanded use of these handy new devices by hospitalists is simultaneously generating increased incidental and equivocal findings, which in turn render it necessary to go back and perform secondary verification studies by specialists using older, gold‐standard modalities. This vicious cycle, coupled with the current lack of evidence, will continue to degrade confidence in the initiation of either acute or chronic treatment on the basis of HCU results obtained by hospitalist physicians.

Eventually, the increased use of HCU by hospitalists might lead to demonstrations of improved hospital workflow management, but it may just as easily represent another new coupling of technology and practitioner that prematurely becomes the standard of care in the absence of any demonstration of added value. The initially enthusiastic application of pulmonary artery catheters (PACs) serves as a cautionary tale in which the acquisition of additional clinical data did not necessarily lead to improved clinical outcomes: whereas PACs did enhance the clinical understanding of hemodynamics, they were not associated with an overall advantage in terms of mortality, length of hospital stay, or cost.3235 Ultimately, more information is not necessarily better information. Although new medical technologies can produce extremely useful diagnostic results that aid in the management of critically ill patients, poor data interpretation resulting from lack of targeted training and experience can nullify point‐of‐care advantages, and perhaps lead to excess morbidity and mortality.14 In clinical practice, it is generally best to avoid reliance on assumptions of added value in lieu of demonstrations of the same.

Conclusions

Hospital practitioners should not yet put away their stethoscopes. New technologies such as HCU need to be embraced in parallel with accumulating evidence of benefit. In the hands of hospitalists, the smaller HCU devices may very well prove handy, but at present, the literature simply does not support the use of HCU by hospitalist physicians.

Physicians assume myriad leadership roles within medical institutions. Clinically‐oriented leadership roles can range from managing a small group of providers, to leading entire health systems, to heading up national quality improvement initiatives. While often competent in the practice of medicine, many physicians have not pursued structured management or administrative training. In a survey of Medicine Department Chairs at academic medical centers, none had advanced management degrees despite spending an average of 55% of their time on administrative duties. It is not uncommon for physicians to attend leadership development programs or management seminars, as evidenced by the increasing demand for education.1 Various methods for skill enhancement have been described24; however, the most effective approaches have yet to be determined.

Miller and Dollard5 and Bandura6, 7 have explained that behavioral contracts have evolved from social cognitive theory principles. These contracts are formal written agreements, often negotiated between 2 individuals, to facilitate behavior change. Typically, they involve a clear definition of expected behaviors with specific consequences (usually positive reinforcement).810 Their use in modifying physician behavior, particularly those related to leadership, has not been studied.

Hospitalist physicians represent the fastest growing specialty in the United States.11, 12 Among other responsibilities, they have taken on roles as leaders in hospital administration, education, quality improvement, and public health.1315 The Society of Hospital Medicine (SHM), the largest US organization committed to the practice of hospital medicine,16 has established Leadership Academies to prepare hospitalists for these duties. The goal of this study was to assess how hospitalist physicians' commitment to grow as leaders was expressed using behavioral contacts as a vehicle to clarify their intentions and whether behavioral change occurred over time.

Methods

Results

Follow‐up Data About Adherence to Plans Delineated in Behavioral Contracts

Out of 65 completed behavioral contracts from the 2007 Level I participants, 32 returned a follow‐up survey (response rate 49.3%). Figure 1 shows the extent to which respondents believed that they were compliant with their proposed plans for change or improvement. Degree of adherence was displayed as a proportion of total goals. Out of those who returned a follow‐up survey, all but 1 respondent either strongly agreed or agreed that they adhered to at least one of their goals (96.9%).

Figure 1

Select representative comments that illustrate the physicians' appreciation of using behavioral contracts include:

• my approach to problems is a bit more analytical.

• simple changes in how I approach people and interact with them has greatly improved my skills as a leader and allowed me to accomplish my goals with much less effort.

 

Discussion

Through the qualitative analysis of the behavioral contracts completed by participants of a Leadership Academy for hospitalists, we characterized the ways that hospitalist practitioners hoped to evolve as leaders. The major themes that emerged relate not only to their own growth and development but also their pledge to advance the success of the group or division. The level of commitment and impact of the behavioral contracts appear to be reinforced by an overwhelmingly positive response to adherence to personal goals one year after course participation. Communication and interpersonal development were most frequently cited in the behavioral contracts as areas for which the hospitalist leaders acknowledged a desire to grow. In a study of academic department of medicine chairs, communication skills were identified as being vital for effective leadership.3 The Chairs also recognized other proficiencies required for leading that were consistent with those outlined in the behavioral contracts: strategic planning, change management, team building, personnel management, and systems thinking. McDade et al.17 examined the effects of participation in an executive leadership program developed for female academic faculty in medical and dental schools in the United States and Canada. They noted increased self‐assessed leadership capabilities at 18 months after attending the program, across 10 leadership constructs taught in the classes. These leadership constructs resonate with the themes found in the plans for change described by our informants.

Hospitalists are assuming leadership roles in an increasing number and with greater scope; however, until now their perspectives on what skill sets are required to be successful have not been well documented. Significant time, effort, and money are invested into the development of hospitalists as leaders.4 The behavioral contract appears to be a tool acceptable to hospitalist physicians; perhaps it can be used as part annual reviews with hospitalists aspiring to be leaders.

Several limitations of the study shall be considered. First, not all participants attending the Leadership Academy opted to fill out the behavioral contracts. Second, this qualitative study is limited to those practitioners who are genuinely interested in growing as leaders as evidenced by their willingness to invest in going to the course. Third, follow‐up surveys relied on self‐assessment and it is not known whether actual realization of these goals occurred or the extent to which behavioral contracts were responsible. Further, follow‐up data were only completed by 49% percent of those targeted. However, hospitalists may be fairly resistant to being surveyed as evidenced by the fact that SHM's 2005‐2006 membership survey yielded a response rate of only 26%.18 Finally, many of the thematic goals were described by fewer than 50% of informants. However, it is important to note that the elements included on each person's behavioral contract emerged spontaneously. If subjects were specifically asked about each theme, the number of comments related to each would certainly be much higher. Qualitative analysis does not really allow us to know whether one theme is more important than another merely because it was mentioned more frequently.

Hospitalist leaders appear to be committed to professional growth and they have reported realization of goals delineated in their behavioral contracts. While varied methods are being used as part of physician leadership training programs, behavioral contracts may enhance promise for change.

Physicians assume myriad leadership roles within medical institutions. Clinically‐oriented leadership roles can range from managing a small group of providers, to leading entire health systems, to heading up national quality improvement initiatives. While often competent in the practice of medicine, many physicians have not pursued structured management or administrative training. In a survey of Medicine Department Chairs at academic medical centers, none had advanced management degrees despite spending an average of 55% of their time on administrative duties. It is not uncommon for physicians to attend leadership development programs or management seminars, as evidenced by the increasing demand for education.1 Various methods for skill enhancement have been described24; however, the most effective approaches have yet to be determined.

Miller and Dollard5 and Bandura6, 7 have explained that behavioral contracts have evolved from social cognitive theory principles. These contracts are formal written agreements, often negotiated between 2 individuals, to facilitate behavior change. Typically, they involve a clear definition of expected behaviors with specific consequences (usually positive reinforcement).810 Their use in modifying physician behavior, particularly those related to leadership, has not been studied.

Hospitalist physicians represent the fastest growing specialty in the United States.11, 12 Among other responsibilities, they have taken on roles as leaders in hospital administration, education, quality improvement, and public health.1315 The Society of Hospital Medicine (SHM), the largest US organization committed to the practice of hospital medicine,16 has established Leadership Academies to prepare hospitalists for these duties. The goal of this study was to assess how hospitalist physicians' commitment to grow as leaders was expressed using behavioral contacts as a vehicle to clarify their intentions and whether behavioral change occurred over time.

Methods

Results

Follow‐up Data About Adherence to Plans Delineated in Behavioral Contracts

Out of 65 completed behavioral contracts from the 2007 Level I participants, 32 returned a follow‐up survey (response rate 49.3%). Figure 1 shows the extent to which respondents believed that they were compliant with their proposed plans for change or improvement. Degree of adherence was displayed as a proportion of total goals. Out of those who returned a follow‐up survey, all but 1 respondent either strongly agreed or agreed that they adhered to at least one of their goals (96.9%).

Figure 1

Select representative comments that illustrate the physicians' appreciation of using behavioral contracts include:

• my approach to problems is a bit more analytical.

• simple changes in how I approach people and interact with them has greatly improved my skills as a leader and allowed me to accomplish my goals with much less effort.

 

Discussion

Through the qualitative analysis of the behavioral contracts completed by participants of a Leadership Academy for hospitalists, we characterized the ways that hospitalist practitioners hoped to evolve as leaders. The major themes that emerged relate not only to their own growth and development but also their pledge to advance the success of the group or division. The level of commitment and impact of the behavioral contracts appear to be reinforced by an overwhelmingly positive response to adherence to personal goals one year after course participation. Communication and interpersonal development were most frequently cited in the behavioral contracts as areas for which the hospitalist leaders acknowledged a desire to grow. In a study of academic department of medicine chairs, communication skills were identified as being vital for effective leadership.3 The Chairs also recognized other proficiencies required for leading that were consistent with those outlined in the behavioral contracts: strategic planning, change management, team building, personnel management, and systems thinking. McDade et al.17 examined the effects of participation in an executive leadership program developed for female academic faculty in medical and dental schools in the United States and Canada. They noted increased self‐assessed leadership capabilities at 18 months after attending the program, across 10 leadership constructs taught in the classes. These leadership constructs resonate with the themes found in the plans for change described by our informants.

Hospitalists are assuming leadership roles in an increasing number and with greater scope; however, until now their perspectives on what skill sets are required to be successful have not been well documented. Significant time, effort, and money are invested into the development of hospitalists as leaders.4 The behavioral contract appears to be a tool acceptable to hospitalist physicians; perhaps it can be used as part annual reviews with hospitalists aspiring to be leaders.

Several limitations of the study shall be considered. First, not all participants attending the Leadership Academy opted to fill out the behavioral contracts. Second, this qualitative study is limited to those practitioners who are genuinely interested in growing as leaders as evidenced by their willingness to invest in going to the course. Third, follow‐up surveys relied on self‐assessment and it is not known whether actual realization of these goals occurred or the extent to which behavioral contracts were responsible. Further, follow‐up data were only completed by 49% percent of those targeted. However, hospitalists may be fairly resistant to being surveyed as evidenced by the fact that SHM's 2005‐2006 membership survey yielded a response rate of only 26%.18 Finally, many of the thematic goals were described by fewer than 50% of informants. However, it is important to note that the elements included on each person's behavioral contract emerged spontaneously. If subjects were specifically asked about each theme, the number of comments related to each would certainly be much higher. Qualitative analysis does not really allow us to know whether one theme is more important than another merely because it was mentioned more frequently.

Hospitalist leaders appear to be committed to professional growth and they have reported realization of goals delineated in their behavioral contracts. While varied methods are being used as part of physician leadership training programs, behavioral contracts may enhance promise for change.

Physicians assume myriad leadership roles within medical institutions. Clinically‐oriented leadership roles can range from managing a small group of providers, to leading entire health systems, to heading up national quality improvement initiatives. While often competent in the practice of medicine, many physicians have not pursued structured management or administrative training. In a survey of Medicine Department Chairs at academic medical centers, none had advanced management degrees despite spending an average of 55% of their time on administrative duties. It is not uncommon for physicians to attend leadership development programs or management seminars, as evidenced by the increasing demand for education.1 Various methods for skill enhancement have been described24; however, the most effective approaches have yet to be determined.

Miller and Dollard5 and Bandura6, 7 have explained that behavioral contracts have evolved from social cognitive theory principles. These contracts are formal written agreements, often negotiated between 2 individuals, to facilitate behavior change. Typically, they involve a clear definition of expected behaviors with specific consequences (usually positive reinforcement).810 Their use in modifying physician behavior, particularly those related to leadership, has not been studied.

Hospitalist physicians represent the fastest growing specialty in the United States.11, 12 Among other responsibilities, they have taken on roles as leaders in hospital administration, education, quality improvement, and public health.1315 The Society of Hospital Medicine (SHM), the largest US organization committed to the practice of hospital medicine,16 has established Leadership Academies to prepare hospitalists for these duties. The goal of this study was to assess how hospitalist physicians' commitment to grow as leaders was expressed using behavioral contacts as a vehicle to clarify their intentions and whether behavioral change occurred over time.

Methods

Results

Follow‐up Data About Adherence to Plans Delineated in Behavioral Contracts

Out of 65 completed behavioral contracts from the 2007 Level I participants, 32 returned a follow‐up survey (response rate 49.3%). Figure 1 shows the extent to which respondents believed that they were compliant with their proposed plans for change or improvement. Degree of adherence was displayed as a proportion of total goals. Out of those who returned a follow‐up survey, all but 1 respondent either strongly agreed or agreed that they adhered to at least one of their goals (96.9%).

Figure 1

Select representative comments that illustrate the physicians' appreciation of using behavioral contracts include:

• my approach to problems is a bit more analytical.

• simple changes in how I approach people and interact with them has greatly improved my skills as a leader and allowed me to accomplish my goals with much less effort.

 

Discussion

Through the qualitative analysis of the behavioral contracts completed by participants of a Leadership Academy for hospitalists, we characterized the ways that hospitalist practitioners hoped to evolve as leaders. The major themes that emerged relate not only to their own growth and development but also their pledge to advance the success of the group or division. The level of commitment and impact of the behavioral contracts appear to be reinforced by an overwhelmingly positive response to adherence to personal goals one year after course participation. Communication and interpersonal development were most frequently cited in the behavioral contracts as areas for which the hospitalist leaders acknowledged a desire to grow. In a study of academic department of medicine chairs, communication skills were identified as being vital for effective leadership.3 The Chairs also recognized other proficiencies required for leading that were consistent with those outlined in the behavioral contracts: strategic planning, change management, team building, personnel management, and systems thinking. McDade et al.17 examined the effects of participation in an executive leadership program developed for female academic faculty in medical and dental schools in the United States and Canada. They noted increased self‐assessed leadership capabilities at 18 months after attending the program, across 10 leadership constructs taught in the classes. These leadership constructs resonate with the themes found in the plans for change described by our informants.

Hospitalists are assuming leadership roles in an increasing number and with greater scope; however, until now their perspectives on what skill sets are required to be successful have not been well documented. Significant time, effort, and money are invested into the development of hospitalists as leaders.4 The behavioral contract appears to be a tool acceptable to hospitalist physicians; perhaps it can be used as part annual reviews with hospitalists aspiring to be leaders.

Several limitations of the study shall be considered. First, not all participants attending the Leadership Academy opted to fill out the behavioral contracts. Second, this qualitative study is limited to those practitioners who are genuinely interested in growing as leaders as evidenced by their willingness to invest in going to the course. Third, follow‐up surveys relied on self‐assessment and it is not known whether actual realization of these goals occurred or the extent to which behavioral contracts were responsible. Further, follow‐up data were only completed by 49% percent of those targeted. However, hospitalists may be fairly resistant to being surveyed as evidenced by the fact that SHM's 2005‐2006 membership survey yielded a response rate of only 26%.18 Finally, many of the thematic goals were described by fewer than 50% of informants. However, it is important to note that the elements included on each person's behavioral contract emerged spontaneously. If subjects were specifically asked about each theme, the number of comments related to each would certainly be much higher. Qualitative analysis does not really allow us to know whether one theme is more important than another merely because it was mentioned more frequently.

Hospitalist leaders appear to be committed to professional growth and they have reported realization of goals delineated in their behavioral contracts. While varied methods are being used as part of physician leadership training programs, behavioral contracts may enhance promise for change.

Hand‐carried ultrasound (HCU) is a field technique. Originally intended for military triage, the advent of small, portable, ultrasound devices has brought ultrasound imaging to the patient's bedside to guide procedures and evaluate life‐threatening conditions. Although many recently‐trained physicians in emergency or critical care medicine now routinely use HCU to place central lines1 and tap effusions,2, 3 the capability of this technique to augment physical examination by all physicians has far greater potential value in medicine. When applied in acute critical scenarios, HCU techniques can quickly demonstrate findings regarding abdominal aortic aneurysm,4 deep vein thrombosis,5 pericardial fluid, or hemoperitoneum6 in patients with unexplained hypotension, and examine inferior vena cava collapsibility7 or brachial artery velocity variation8 to help determine the need for volume resuscitation in sepsis. In patients with unexplained dyspnea, HCU can search for ultrasound lung comet‐tail artifacts as a sign of pulmonary edema,9 or use the presence of pleural sliding to exclude pneumothorax.10 In addition, numerous less urgent applications for HCU imaging are emerging such as cardiac, lung, vascular, musculoskeletal, nerve, thyroid, gallbladder, liver, spleen, renal, testicular, and bladder imaging.

Medical or surgical subspecialties familiar with ultrasound have developed limited HCU examinations that serve specific purposes within the relatively narrow clinical indications encountered by these specialties. As a consequence, overall expertise in bedside HCU currently requires the mastery of multiple unrelated ultrasound views and diagnostic criteria. Without central leadership within this burgeoning field, HCU has found no consensus on its use or development within general medical practice. No one has yet validated a single ultrasound imaging protocol for augmenting the physical examination on all patients akin to the use of the stethoscope. This review discusses the importance of the internisthospitalist at this critical point in the early development of bedside HCU examination, focusing on the cardiopulmonary component as a prototype that has universal application across medical practice. Involvement by hospitalists in pioneering the overall technique will direct research in clinical outcome, restructure internal medicine education, change perception of the physical examination, and spur industry in device development specific for general medicine.

The role of the hospitalist as the leading in‐house diagnostician is unique in medicine, requiring breadth in medical knowledge and unprecedented communication skills in the seamless care of the most medically ill patients in the community.11 Ideally, the hospitalist quickly recognizes disease, discriminately uses consultation or expensive diagnostic testing, chooses cost‐effective therapies, and shortens length of hospital stay. Early accurate diagnosis afforded by HCU imaging has the potential to improve efficiency of medical care across a wide spectrum of clinical presentations. Although to date there are no outcome studies using a mortality endpoint, small individual studies have demonstrated that specific HCU findings improve diagnostic accuracy and relate to hospital stay length12 and readmission.13 The hospitalist position is in theory well‐suited for learning and applying bedside ultrasound, having both expert resources in the hospital to guide training and a clinical objective to reduce unnecessary hospital costs.

Saving the Bedside Examination: The Laying‐on of Ultrasound

Bedside examination is a vital component of the initial hospitalist‐patient interaction, adding objective data to the patient's history. In this era of physician surrogates and telemedicine, physical examination remains a nonnegotiable reason why physicians must appear in person at the patient's bedside to lay on hands. However, bedside cardiovascular examination skills have greatly diminished over the past decade for a variety of reasons.14 In particular, physical examination is impaired in the environment in which the hospitalist must practice. The admitting physician must oftentimes hurriedly examine the patient on the gurney in the noisy emergency department or in bed in an alarm‐filled intensive care unit (ICU) or hospital room. Ambient noise levels often preclude auscultation of acute aortic and mitral valve regurgitation, splitting of valve sounds, low diastolic rumbles, soft gallops, and fine rales. Patient positioning is limited in ventilated patients or those in respiratory or circulatory distress. Although medical education still honors the value of teaching the traditional cardiac examination, no outcome data exist to justify the application of the various maneuvers and techniques learned in medical school to contemporary, commonly encountered inpatient care scenarios. For example, few physical examination data exist on how to evaluate central venous pressures of an obese patient on the ventilator or assess the severity of aortic stenosis in the elderly hypertensive patient. Furthermore, many important cardiopulmonary abnormalities that are easily detected by ultrasound, such as pericardial fluid, well‐compensated left ventricular systolic dysfunction, small pleural effusion, and left atrial enlargement, make no characteristic sound for auscultation. The effect of undiagnosed cardiac abnormalities on the patient's immediate hospital course is unknown, but is likely related to the clinical presentation and long‐term outcome. Today, the hospitalist's suspicion of cardiovascular abnormalities is more often generated from elements in the patient's initial history, serum biomarkers, chest radiography, or electrocardiogram, and less from auscultation. Accordingly, cardiac physical examination is only adjunctively used in determining the general direction of the ensuing evaluation and when abnormal, often generates additional diagnostic testing for confirmation.

The optimal role of HCU for the internist‐hospitalist is in augmentation of bedside physical diagnosis.15, 16 Unlike x‐ray or even rapid serum biomarkers, ultrasound is a safe, immediate, noninvasive modality and has been particularly effective in delineating cardiac structure and physiology. Accurate HCU estimation of a patient's central venous pressure,17 left atrial size,18 or left ventricular ejection fraction19, 20 is of particular value in those with unexplained respiratory distress or circulatory collapse, or in those in whom referral for echocardiography or cardiac consultation is not obvious. Asymptomatic left ventricular systolic dysfunction has an estimated prevalence of 5% in adult populations,21 and its detection would have immediate implications in regard to etiology, volume management, and drug therapy. Multiple studies have shown the prognostic importance of left atrial enlargement in ischemic cardiac disease, congestive heart failure, atrial arrhythmias, and stroke.22 The inferior vena cava diameter has been related to central venous pressure and prognosis in congestive heart failure. A recent study13 using medical residents employing HCU demonstrated that persistent dilatation of the inferior vena cava at discharge related to a higher readmission rate in patients with congestive heart failure. The potential exists to follow and guide a patient's response to therapy with HCU during daily rounds. Comparative studies2325 confirm that HCU examinations are better than expert auscultation and improve overall exam accuracy when added to traditional physical exam techniques. Entering into the modern‐day emergency room with a pocket‐sized ultrasound device that provides the immediate capability of detecting left ventricular dysfunction, left atrial enlargement, pericardial effusion, or abnormalities in volume status, provides an additional sense of being prepared for battle.

Deriving Limited Ultrasound Applications: Time Well Spent

However, in order for a hospitalist to use HCU, easily applied limited imaging protocols must be derived from standard ultrasound examination techniques for each organ. For the heart, studies from our laboratory have shown that it is feasible to distill the comprehensive echocardiogram down to simple cardiac screening examinations for rapid bedside HCU use.2628 We found that a limited cardiac ultrasound study consisting of a single parasternal long‐axis (PLAX) view (Figure 1) requires only seconds to perform and can identify those patients who have significant cardiac abnormalities. In an outpatient population (n = 196) followed in an internal medicine clinic, the PLAX component of an HCU cardiac screening protocol uncovered left atrial enlargement in 4 patients and left ventricular systolic dysfunction in 4 patients that had not been suspected by the patients' primary physicians.29 In a study of 124 patients in the emergency department with suspected cardiac disease,12 abnormal cardiac findings were noted 3 times more frequently by PLAX than by clinical evaluation, and an abnormal PLAX was significantly associated with a longer hospital length of stay. In other preliminary studies using cardiologists, limited imaging has been shown to reduce costs of unnecessary echo referral.28, 3032 Cost analysis has yet to be performed in nonexpert HCU users, but benefit is likely related to the difference between the user's own accuracy with the stethoscope and the HCU device.

Figure 1

Although experts in ultrasound exist in radiology and cardiology, it is unlikely these subspecialists will spontaneously create and optimize a full‐body HCU imaging protocol for hospitalists. Similar to the use of ultrasound in emergency medicine, anesthesiology, and critical care medicine, the derivation of a bedside ultrasound exam appropriate for the in‐hospital physical examination should be developed within the specialty itself, by those acquainted with the clinical scenarios in which HCU would be deployed. For example, the question of whether the gallbladder should be routinely imaged by a quick HCU exam in the evaluation of chest pain is similar to the question of whether the Valsalva maneuver should be performed in the evaluation of every murmurboth require Bayesian knowledge of disease prevalence, exam difficulty, and test accuracy. With the collaboration of experts in ultrasound, internists can derive brief, easily learned, limited ultrasound exams for left ventricular dysfunction, left atrial enlargement, carotid atherosclerosis, interstitial lung disease, hepatosplenomegaly, cholelithiasis, hydronephrosis, renal atrophy, pleural or pericardial effusion, ascites, deep vein thrombosis, and abdominal aortic aneurysm. The discovery of these disease states has clinical value for long‐term care, even if incidental to the patient's acute presentation. The lasting implications of a more comprehensive general examination will likely differentiate the use of HCU in internal medicine practice from that of emergency medicine.

Basic Training in HCU

A significant challenge to medical education will be in physician training in HCU. Over 15 studies12, 13, 15, 1720, 22, 23, 3343 have now shown the ability of briefly trained medical students, residents, and physicians in internal medicine to perform a limited cardiovascular ultrasound examination. Not surprisingly, these studies show variable degrees of training proficiency, apparently dependent upon the complexity of the imaging protocol. In a recent pair of studies from 1 institution,42, 43 10 hospitalists were trained to perform an extensive HCU echocardiogram including 4 views, color and spectral Doppler, and interpret severity of valvular disease, ventricular function, pericardial effusion. In 345 patients already referred for formal echocardiography, which later served as the gold standard, HCU improved the hospitalists' physical examination for left ventricular dysfunction, cardiomegaly, and pericardial effusion, but not for valvular disease. Notably, despite a focused training program including didactic teaching, self‐study cases, 5 training studies, and the imaging of 35 patients with assistance as needed, image acquisition was inferior to standard examination and image interpretation was inferior to that of cardiology fellows. Such data reemphasize the fact that the scope of each body‐system imaging protocol must be narrow in order to make the learning of a full‐body HCU exam feasible and to incorporate training into time already allocated to the bedside physical examination curriculum or continuing medical education activities.

At our institution, internal medical residents are trained in bedside cardiovascular ultrasound to blend results with their auscultative findings during bedside examination. We have developed 2 cardiovascular limited ultrasound examinations (CLUEs) that can be performed in 5 minutes and have evidence‐basis for their clinical use through pilot training studies.18, 19, 29, 35 Our basic CLUE, designed for general cardiovascular examination, includes screening the carotid bulb for subclinical atherosclerosis, PLAX imaging for left atrial enlargement and systolic dysfunction of the left ventricle, and abdominal scanning for abdominal aortic aneurysm. In this imaging protocol consisting of only 4 targets, atherosclerotic risk increases from top to bottom (cephalad to caudal), making the exam easy to remember. The CLUEparasternal, lung, and subcostal (CLUE‐PLUS), designed for the urgent evaluation of unexplained dyspnea or hypotension, uses a work backward imaging format (from left ventricle to right atrium) and a single cardiac transducer for simplicity. The PLAX view screens for left ventricular systolic dysfunction and then left atrial enlargement. Next, a brief 4‐point lung exam screens for ultrasonic lung comets and pleural effusion. A subcostal view of the heart is used to evaluate right ventricular size and pericardial effusion, and finally the inferior vena cava is evaluated for central venous pressures. CLUEs are taught in bedside and didactic formats over the 3 years of residency with formal competency testing after lecture attendance, practice imaging in our echo‐vascular laboratories, participation in rounds, and completion of at least 30 supervised examinations.

Reaffirming the Role of the Internist

Although emergency44 and critical care45 medical subspecialties have begun to train their constituencies in HCU, general diagnostic techniques that have wide‐ranging application in medical illness should be the evidence‐based tools of the internist. The rejuvenation of bedside examination using HCU on multiple organ systems should be orchestrated within internal medicine and not simply evolve as an unedited collection of all subspecialty organ ultrasound examinations. Device development can then be customized and made affordable for use in general internal medicine, perhaps limiting the unnecessary production costs and training requirements for advanced Doppler or multiple transducers.

Concern has been raised about the medical and economic impact of training internists in HCU. Although training costs can be incorporated in residency or hospital‐based continuing medical education, discussions regarding reimbursement for cardiac imaging require a distinction between the brief application of ultrasound using a small device by a nontraditional user and a limited echocardiogram as defined by payers and professional societies.46 To date, no procedural code or reimbursement has yet been approved for ultrasound‐assisted physical examination using HCU devices and likely awaits outcome data. There is also concern about the possibility of errors being made by HCU use by briefly trained physicians. Patient care and cost‐savings depend on HCU accuracy, being liable both for unnecessary referrals due to false‐positive screening HCU exams and delays in diagnosis due to false‐negative examinations. However, such errors are commonplace and accepted with standard physical examination techniques and the current use of the stethoscope, both of which lack sensitivity when compared to HCU.

HCU is a disruptive technology.47 However, unlike the successful disruption that small desktop computers had on their mainframe counterparts, HCU devices appeared before the operating system of their clinical application had been formulated, making dissemination to new users nearly impossible. Furthermore, placing ultrasound transducers into the hands of nontraditional users often alienates or displaces established users of ultrasound as well as established untrained members within the profession. Competency requirements will have to be derived, preferably from studies performed within the profession for specific uses in internal medicine. Perhaps championed by hospitalists and driven by hospital‐based outcome studies, the use of HCU by internists as a physical exam technique will require advocacy by internists themselves. The alternative, having the hospitalist ask the emergency department physician for help in examining the patient, is difficult to imagine. The answer to whether the hospitalist should use HCU should be a resounding yesbased upon the benefit of earlier, more accurate examination and the value of preserving the diagnostic role of the internist at the bedside. In regard to the latter, it is a concept worth fighting for.

Hand‐carried ultrasound (HCU) is a field technique. Originally intended for military triage, the advent of small, portable, ultrasound devices has brought ultrasound imaging to the patient's bedside to guide procedures and evaluate life‐threatening conditions. Although many recently‐trained physicians in emergency or critical care medicine now routinely use HCU to place central lines1 and tap effusions,2, 3 the capability of this technique to augment physical examination by all physicians has far greater potential value in medicine. When applied in acute critical scenarios, HCU techniques can quickly demonstrate findings regarding abdominal aortic aneurysm,4 deep vein thrombosis,5 pericardial fluid, or hemoperitoneum6 in patients with unexplained hypotension, and examine inferior vena cava collapsibility7 or brachial artery velocity variation8 to help determine the need for volume resuscitation in sepsis. In patients with unexplained dyspnea, HCU can search for ultrasound lung comet‐tail artifacts as a sign of pulmonary edema,9 or use the presence of pleural sliding to exclude pneumothorax.10 In addition, numerous less urgent applications for HCU imaging are emerging such as cardiac, lung, vascular, musculoskeletal, nerve, thyroid, gallbladder, liver, spleen, renal, testicular, and bladder imaging.

Medical or surgical subspecialties familiar with ultrasound have developed limited HCU examinations that serve specific purposes within the relatively narrow clinical indications encountered by these specialties. As a consequence, overall expertise in bedside HCU currently requires the mastery of multiple unrelated ultrasound views and diagnostic criteria. Without central leadership within this burgeoning field, HCU has found no consensus on its use or development within general medical practice. No one has yet validated a single ultrasound imaging protocol for augmenting the physical examination on all patients akin to the use of the stethoscope. This review discusses the importance of the internisthospitalist at this critical point in the early development of bedside HCU examination, focusing on the cardiopulmonary component as a prototype that has universal application across medical practice. Involvement by hospitalists in pioneering the overall technique will direct research in clinical outcome, restructure internal medicine education, change perception of the physical examination, and spur industry in device development specific for general medicine.

The role of the hospitalist as the leading in‐house diagnostician is unique in medicine, requiring breadth in medical knowledge and unprecedented communication skills in the seamless care of the most medically ill patients in the community.11 Ideally, the hospitalist quickly recognizes disease, discriminately uses consultation or expensive diagnostic testing, chooses cost‐effective therapies, and shortens length of hospital stay. Early accurate diagnosis afforded by HCU imaging has the potential to improve efficiency of medical care across a wide spectrum of clinical presentations. Although to date there are no outcome studies using a mortality endpoint, small individual studies have demonstrated that specific HCU findings improve diagnostic accuracy and relate to hospital stay length12 and readmission.13 The hospitalist position is in theory well‐suited for learning and applying bedside ultrasound, having both expert resources in the hospital to guide training and a clinical objective to reduce unnecessary hospital costs.

Saving the Bedside Examination: The Laying‐on of Ultrasound

Bedside examination is a vital component of the initial hospitalist‐patient interaction, adding objective data to the patient's history. In this era of physician surrogates and telemedicine, physical examination remains a nonnegotiable reason why physicians must appear in person at the patient's bedside to lay on hands. However, bedside cardiovascular examination skills have greatly diminished over the past decade for a variety of reasons.14 In particular, physical examination is impaired in the environment in which the hospitalist must practice. The admitting physician must oftentimes hurriedly examine the patient on the gurney in the noisy emergency department or in bed in an alarm‐filled intensive care unit (ICU) or hospital room. Ambient noise levels often preclude auscultation of acute aortic and mitral valve regurgitation, splitting of valve sounds, low diastolic rumbles, soft gallops, and fine rales. Patient positioning is limited in ventilated patients or those in respiratory or circulatory distress. Although medical education still honors the value of teaching the traditional cardiac examination, no outcome data exist to justify the application of the various maneuvers and techniques learned in medical school to contemporary, commonly encountered inpatient care scenarios. For example, few physical examination data exist on how to evaluate central venous pressures of an obese patient on the ventilator or assess the severity of aortic stenosis in the elderly hypertensive patient. Furthermore, many important cardiopulmonary abnormalities that are easily detected by ultrasound, such as pericardial fluid, well‐compensated left ventricular systolic dysfunction, small pleural effusion, and left atrial enlargement, make no characteristic sound for auscultation. The effect of undiagnosed cardiac abnormalities on the patient's immediate hospital course is unknown, but is likely related to the clinical presentation and long‐term outcome. Today, the hospitalist's suspicion of cardiovascular abnormalities is more often generated from elements in the patient's initial history, serum biomarkers, chest radiography, or electrocardiogram, and less from auscultation. Accordingly, cardiac physical examination is only adjunctively used in determining the general direction of the ensuing evaluation and when abnormal, often generates additional diagnostic testing for confirmation.

The optimal role of HCU for the internist‐hospitalist is in augmentation of bedside physical diagnosis.15, 16 Unlike x‐ray or even rapid serum biomarkers, ultrasound is a safe, immediate, noninvasive modality and has been particularly effective in delineating cardiac structure and physiology. Accurate HCU estimation of a patient's central venous pressure,17 left atrial size,18 or left ventricular ejection fraction19, 20 is of particular value in those with unexplained respiratory distress or circulatory collapse, or in those in whom referral for echocardiography or cardiac consultation is not obvious. Asymptomatic left ventricular systolic dysfunction has an estimated prevalence of 5% in adult populations,21 and its detection would have immediate implications in regard to etiology, volume management, and drug therapy. Multiple studies have shown the prognostic importance of left atrial enlargement in ischemic cardiac disease, congestive heart failure, atrial arrhythmias, and stroke.22 The inferior vena cava diameter has been related to central venous pressure and prognosis in congestive heart failure. A recent study13 using medical residents employing HCU demonstrated that persistent dilatation of the inferior vena cava at discharge related to a higher readmission rate in patients with congestive heart failure. The potential exists to follow and guide a patient's response to therapy with HCU during daily rounds. Comparative studies2325 confirm that HCU examinations are better than expert auscultation and improve overall exam accuracy when added to traditional physical exam techniques. Entering into the modern‐day emergency room with a pocket‐sized ultrasound device that provides the immediate capability of detecting left ventricular dysfunction, left atrial enlargement, pericardial effusion, or abnormalities in volume status, provides an additional sense of being prepared for battle.

Deriving Limited Ultrasound Applications: Time Well Spent

However, in order for a hospitalist to use HCU, easily applied limited imaging protocols must be derived from standard ultrasound examination techniques for each organ. For the heart, studies from our laboratory have shown that it is feasible to distill the comprehensive echocardiogram down to simple cardiac screening examinations for rapid bedside HCU use.2628 We found that a limited cardiac ultrasound study consisting of a single parasternal long‐axis (PLAX) view (Figure 1) requires only seconds to perform and can identify those patients who have significant cardiac abnormalities. In an outpatient population (n = 196) followed in an internal medicine clinic, the PLAX component of an HCU cardiac screening protocol uncovered left atrial enlargement in 4 patients and left ventricular systolic dysfunction in 4 patients that had not been suspected by the patients' primary physicians.29 In a study of 124 patients in the emergency department with suspected cardiac disease,12 abnormal cardiac findings were noted 3 times more frequently by PLAX than by clinical evaluation, and an abnormal PLAX was significantly associated with a longer hospital length of stay. In other preliminary studies using cardiologists, limited imaging has been shown to reduce costs of unnecessary echo referral.28, 3032 Cost analysis has yet to be performed in nonexpert HCU users, but benefit is likely related to the difference between the user's own accuracy with the stethoscope and the HCU device.

Figure 1

Although experts in ultrasound exist in radiology and cardiology, it is unlikely these subspecialists will spontaneously create and optimize a full‐body HCU imaging protocol for hospitalists. Similar to the use of ultrasound in emergency medicine, anesthesiology, and critical care medicine, the derivation of a bedside ultrasound exam appropriate for the in‐hospital physical examination should be developed within the specialty itself, by those acquainted with the clinical scenarios in which HCU would be deployed. For example, the question of whether the gallbladder should be routinely imaged by a quick HCU exam in the evaluation of chest pain is similar to the question of whether the Valsalva maneuver should be performed in the evaluation of every murmurboth require Bayesian knowledge of disease prevalence, exam difficulty, and test accuracy. With the collaboration of experts in ultrasound, internists can derive brief, easily learned, limited ultrasound exams for left ventricular dysfunction, left atrial enlargement, carotid atherosclerosis, interstitial lung disease, hepatosplenomegaly, cholelithiasis, hydronephrosis, renal atrophy, pleural or pericardial effusion, ascites, deep vein thrombosis, and abdominal aortic aneurysm. The discovery of these disease states has clinical value for long‐term care, even if incidental to the patient's acute presentation. The lasting implications of a more comprehensive general examination will likely differentiate the use of HCU in internal medicine practice from that of emergency medicine.

Basic Training in HCU

A significant challenge to medical education will be in physician training in HCU. Over 15 studies12, 13, 15, 1720, 22, 23, 3343 have now shown the ability of briefly trained medical students, residents, and physicians in internal medicine to perform a limited cardiovascular ultrasound examination. Not surprisingly, these studies show variable degrees of training proficiency, apparently dependent upon the complexity of the imaging protocol. In a recent pair of studies from 1 institution,42, 43 10 hospitalists were trained to perform an extensive HCU echocardiogram including 4 views, color and spectral Doppler, and interpret severity of valvular disease, ventricular function, pericardial effusion. In 345 patients already referred for formal echocardiography, which later served as the gold standard, HCU improved the hospitalists' physical examination for left ventricular dysfunction, cardiomegaly, and pericardial effusion, but not for valvular disease. Notably, despite a focused training program including didactic teaching, self‐study cases, 5 training studies, and the imaging of 35 patients with assistance as needed, image acquisition was inferior to standard examination and image interpretation was inferior to that of cardiology fellows. Such data reemphasize the fact that the scope of each body‐system imaging protocol must be narrow in order to make the learning of a full‐body HCU exam feasible and to incorporate training into time already allocated to the bedside physical examination curriculum or continuing medical education activities.

At our institution, internal medical residents are trained in bedside cardiovascular ultrasound to blend results with their auscultative findings during bedside examination. We have developed 2 cardiovascular limited ultrasound examinations (CLUEs) that can be performed in 5 minutes and have evidence‐basis for their clinical use through pilot training studies.18, 19, 29, 35 Our basic CLUE, designed for general cardiovascular examination, includes screening the carotid bulb for subclinical atherosclerosis, PLAX imaging for left atrial enlargement and systolic dysfunction of the left ventricle, and abdominal scanning for abdominal aortic aneurysm. In this imaging protocol consisting of only 4 targets, atherosclerotic risk increases from top to bottom (cephalad to caudal), making the exam easy to remember. The CLUEparasternal, lung, and subcostal (CLUE‐PLUS), designed for the urgent evaluation of unexplained dyspnea or hypotension, uses a work backward imaging format (from left ventricle to right atrium) and a single cardiac transducer for simplicity. The PLAX view screens for left ventricular systolic dysfunction and then left atrial enlargement. Next, a brief 4‐point lung exam screens for ultrasonic lung comets and pleural effusion. A subcostal view of the heart is used to evaluate right ventricular size and pericardial effusion, and finally the inferior vena cava is evaluated for central venous pressures. CLUEs are taught in bedside and didactic formats over the 3 years of residency with formal competency testing after lecture attendance, practice imaging in our echo‐vascular laboratories, participation in rounds, and completion of at least 30 supervised examinations.

Reaffirming the Role of the Internist

Although emergency44 and critical care45 medical subspecialties have begun to train their constituencies in HCU, general diagnostic techniques that have wide‐ranging application in medical illness should be the evidence‐based tools of the internist. The rejuvenation of bedside examination using HCU on multiple organ systems should be orchestrated within internal medicine and not simply evolve as an unedited collection of all subspecialty organ ultrasound examinations. Device development can then be customized and made affordable for use in general internal medicine, perhaps limiting the unnecessary production costs and training requirements for advanced Doppler or multiple transducers.

Concern has been raised about the medical and economic impact of training internists in HCU. Although training costs can be incorporated in residency or hospital‐based continuing medical education, discussions regarding reimbursement for cardiac imaging require a distinction between the brief application of ultrasound using a small device by a nontraditional user and a limited echocardiogram as defined by payers and professional societies.46 To date, no procedural code or reimbursement has yet been approved for ultrasound‐assisted physical examination using HCU devices and likely awaits outcome data. There is also concern about the possibility of errors being made by HCU use by briefly trained physicians. Patient care and cost‐savings depend on HCU accuracy, being liable both for unnecessary referrals due to false‐positive screening HCU exams and delays in diagnosis due to false‐negative examinations. However, such errors are commonplace and accepted with standard physical examination techniques and the current use of the stethoscope, both of which lack sensitivity when compared to HCU.

HCU is a disruptive technology.47 However, unlike the successful disruption that small desktop computers had on their mainframe counterparts, HCU devices appeared before the operating system of their clinical application had been formulated, making dissemination to new users nearly impossible. Furthermore, placing ultrasound transducers into the hands of nontraditional users often alienates or displaces established users of ultrasound as well as established untrained members within the profession. Competency requirements will have to be derived, preferably from studies performed within the profession for specific uses in internal medicine. Perhaps championed by hospitalists and driven by hospital‐based outcome studies, the use of HCU by internists as a physical exam technique will require advocacy by internists themselves. The alternative, having the hospitalist ask the emergency department physician for help in examining the patient, is difficult to imagine. The answer to whether the hospitalist should use HCU should be a resounding yesbased upon the benefit of earlier, more accurate examination and the value of preserving the diagnostic role of the internist at the bedside. In regard to the latter, it is a concept worth fighting for.

Hand‐carried ultrasound (HCU) is a field technique. Originally intended for military triage, the advent of small, portable, ultrasound devices has brought ultrasound imaging to the patient's bedside to guide procedures and evaluate life‐threatening conditions. Although many recently‐trained physicians in emergency or critical care medicine now routinely use HCU to place central lines1 and tap effusions,2, 3 the capability of this technique to augment physical examination by all physicians has far greater potential value in medicine. When applied in acute critical scenarios, HCU techniques can quickly demonstrate findings regarding abdominal aortic aneurysm,4 deep vein thrombosis,5 pericardial fluid, or hemoperitoneum6 in patients with unexplained hypotension, and examine inferior vena cava collapsibility7 or brachial artery velocity variation8 to help determine the need for volume resuscitation in sepsis. In patients with unexplained dyspnea, HCU can search for ultrasound lung comet‐tail artifacts as a sign of pulmonary edema,9 or use the presence of pleural sliding to exclude pneumothorax.10 In addition, numerous less urgent applications for HCU imaging are emerging such as cardiac, lung, vascular, musculoskeletal, nerve, thyroid, gallbladder, liver, spleen, renal, testicular, and bladder imaging.

Medical or surgical subspecialties familiar with ultrasound have developed limited HCU examinations that serve specific purposes within the relatively narrow clinical indications encountered by these specialties. As a consequence, overall expertise in bedside HCU currently requires the mastery of multiple unrelated ultrasound views and diagnostic criteria. Without central leadership within this burgeoning field, HCU has found no consensus on its use or development within general medical practice. No one has yet validated a single ultrasound imaging protocol for augmenting the physical examination on all patients akin to the use of the stethoscope. This review discusses the importance of the internisthospitalist at this critical point in the early development of bedside HCU examination, focusing on the cardiopulmonary component as a prototype that has universal application across medical practice. Involvement by hospitalists in pioneering the overall technique will direct research in clinical outcome, restructure internal medicine education, change perception of the physical examination, and spur industry in device development specific for general medicine.

The role of the hospitalist as the leading in‐house diagnostician is unique in medicine, requiring breadth in medical knowledge and unprecedented communication skills in the seamless care of the most medically ill patients in the community.11 Ideally, the hospitalist quickly recognizes disease, discriminately uses consultation or expensive diagnostic testing, chooses cost‐effective therapies, and shortens length of hospital stay. Early accurate diagnosis afforded by HCU imaging has the potential to improve efficiency of medical care across a wide spectrum of clinical presentations. Although to date there are no outcome studies using a mortality endpoint, small individual studies have demonstrated that specific HCU findings improve diagnostic accuracy and relate to hospital stay length12 and readmission.13 The hospitalist position is in theory well‐suited for learning and applying bedside ultrasound, having both expert resources in the hospital to guide training and a clinical objective to reduce unnecessary hospital costs.

Saving the Bedside Examination: The Laying‐on of Ultrasound

Bedside examination is a vital component of the initial hospitalist‐patient interaction, adding objective data to the patient's history. In this era of physician surrogates and telemedicine, physical examination remains a nonnegotiable reason why physicians must appear in person at the patient's bedside to lay on hands. However, bedside cardiovascular examination skills have greatly diminished over the past decade for a variety of reasons.14 In particular, physical examination is impaired in the environment in which the hospitalist must practice. The admitting physician must oftentimes hurriedly examine the patient on the gurney in the noisy emergency department or in bed in an alarm‐filled intensive care unit (ICU) or hospital room. Ambient noise levels often preclude auscultation of acute aortic and mitral valve regurgitation, splitting of valve sounds, low diastolic rumbles, soft gallops, and fine rales. Patient positioning is limited in ventilated patients or those in respiratory or circulatory distress. Although medical education still honors the value of teaching the traditional cardiac examination, no outcome data exist to justify the application of the various maneuvers and techniques learned in medical school to contemporary, commonly encountered inpatient care scenarios. For example, few physical examination data exist on how to evaluate central venous pressures of an obese patient on the ventilator or assess the severity of aortic stenosis in the elderly hypertensive patient. Furthermore, many important cardiopulmonary abnormalities that are easily detected by ultrasound, such as pericardial fluid, well‐compensated left ventricular systolic dysfunction, small pleural effusion, and left atrial enlargement, make no characteristic sound for auscultation. The effect of undiagnosed cardiac abnormalities on the patient's immediate hospital course is unknown, but is likely related to the clinical presentation and long‐term outcome. Today, the hospitalist's suspicion of cardiovascular abnormalities is more often generated from elements in the patient's initial history, serum biomarkers, chest radiography, or electrocardiogram, and less from auscultation. Accordingly, cardiac physical examination is only adjunctively used in determining the general direction of the ensuing evaluation and when abnormal, often generates additional diagnostic testing for confirmation.

The optimal role of HCU for the internist‐hospitalist is in augmentation of bedside physical diagnosis.15, 16 Unlike x‐ray or even rapid serum biomarkers, ultrasound is a safe, immediate, noninvasive modality and has been particularly effective in delineating cardiac structure and physiology. Accurate HCU estimation of a patient's central venous pressure,17 left atrial size,18 or left ventricular ejection fraction19, 20 is of particular value in those with unexplained respiratory distress or circulatory collapse, or in those in whom referral for echocardiography or cardiac consultation is not obvious. Asymptomatic left ventricular systolic dysfunction has an estimated prevalence of 5% in adult populations,21 and its detection would have immediate implications in regard to etiology, volume management, and drug therapy. Multiple studies have shown the prognostic importance of left atrial enlargement in ischemic cardiac disease, congestive heart failure, atrial arrhythmias, and stroke.22 The inferior vena cava diameter has been related to central venous pressure and prognosis in congestive heart failure. A recent study13 using medical residents employing HCU demonstrated that persistent dilatation of the inferior vena cava at discharge related to a higher readmission rate in patients with congestive heart failure. The potential exists to follow and guide a patient's response to therapy with HCU during daily rounds. Comparative studies2325 confirm that HCU examinations are better than expert auscultation and improve overall exam accuracy when added to traditional physical exam techniques. Entering into the modern‐day emergency room with a pocket‐sized ultrasound device that provides the immediate capability of detecting left ventricular dysfunction, left atrial enlargement, pericardial effusion, or abnormalities in volume status, provides an additional sense of being prepared for battle.

Deriving Limited Ultrasound Applications: Time Well Spent

However, in order for a hospitalist to use HCU, easily applied limited imaging protocols must be derived from standard ultrasound examination techniques for each organ. For the heart, studies from our laboratory have shown that it is feasible to distill the comprehensive echocardiogram down to simple cardiac screening examinations for rapid bedside HCU use.2628 We found that a limited cardiac ultrasound study consisting of a single parasternal long‐axis (PLAX) view (Figure 1) requires only seconds to perform and can identify those patients who have significant cardiac abnormalities. In an outpatient population (n = 196) followed in an internal medicine clinic, the PLAX component of an HCU cardiac screening protocol uncovered left atrial enlargement in 4 patients and left ventricular systolic dysfunction in 4 patients that had not been suspected by the patients' primary physicians.29 In a study of 124 patients in the emergency department with suspected cardiac disease,12 abnormal cardiac findings were noted 3 times more frequently by PLAX than by clinical evaluation, and an abnormal PLAX was significantly associated with a longer hospital length of stay. In other preliminary studies using cardiologists, limited imaging has been shown to reduce costs of unnecessary echo referral.28, 3032 Cost analysis has yet to be performed in nonexpert HCU users, but benefit is likely related to the difference between the user's own accuracy with the stethoscope and the HCU device.

Figure 1

Although experts in ultrasound exist in radiology and cardiology, it is unlikely these subspecialists will spontaneously create and optimize a full‐body HCU imaging protocol for hospitalists. Similar to the use of ultrasound in emergency medicine, anesthesiology, and critical care medicine, the derivation of a bedside ultrasound exam appropriate for the in‐hospital physical examination should be developed within the specialty itself, by those acquainted with the clinical scenarios in which HCU would be deployed. For example, the question of whether the gallbladder should be routinely imaged by a quick HCU exam in the evaluation of chest pain is similar to the question of whether the Valsalva maneuver should be performed in the evaluation of every murmurboth require Bayesian knowledge of disease prevalence, exam difficulty, and test accuracy. With the collaboration of experts in ultrasound, internists can derive brief, easily learned, limited ultrasound exams for left ventricular dysfunction, left atrial enlargement, carotid atherosclerosis, interstitial lung disease, hepatosplenomegaly, cholelithiasis, hydronephrosis, renal atrophy, pleural or pericardial effusion, ascites, deep vein thrombosis, and abdominal aortic aneurysm. The discovery of these disease states has clinical value for long‐term care, even if incidental to the patient's acute presentation. The lasting implications of a more comprehensive general examination will likely differentiate the use of HCU in internal medicine practice from that of emergency medicine.

Basic Training in HCU

A significant challenge to medical education will be in physician training in HCU. Over 15 studies12, 13, 15, 1720, 22, 23, 3343 have now shown the ability of briefly trained medical students, residents, and physicians in internal medicine to perform a limited cardiovascular ultrasound examination. Not surprisingly, these studies show variable degrees of training proficiency, apparently dependent upon the complexity of the imaging protocol. In a recent pair of studies from 1 institution,42, 43 10 hospitalists were trained to perform an extensive HCU echocardiogram including 4 views, color and spectral Doppler, and interpret severity of valvular disease, ventricular function, pericardial effusion. In 345 patients already referred for formal echocardiography, which later served as the gold standard, HCU improved the hospitalists' physical examination for left ventricular dysfunction, cardiomegaly, and pericardial effusion, but not for valvular disease. Notably, despite a focused training program including didactic teaching, self‐study cases, 5 training studies, and the imaging of 35 patients with assistance as needed, image acquisition was inferior to standard examination and image interpretation was inferior to that of cardiology fellows. Such data reemphasize the fact that the scope of each body‐system imaging protocol must be narrow in order to make the learning of a full‐body HCU exam feasible and to incorporate training into time already allocated to the bedside physical examination curriculum or continuing medical education activities.

At our institution, internal medical residents are trained in bedside cardiovascular ultrasound to blend results with their auscultative findings during bedside examination. We have developed 2 cardiovascular limited ultrasound examinations (CLUEs) that can be performed in 5 minutes and have evidence‐basis for their clinical use through pilot training studies.18, 19, 29, 35 Our basic CLUE, designed for general cardiovascular examination, includes screening the carotid bulb for subclinical atherosclerosis, PLAX imaging for left atrial enlargement and systolic dysfunction of the left ventricle, and abdominal scanning for abdominal aortic aneurysm. In this imaging protocol consisting of only 4 targets, atherosclerotic risk increases from top to bottom (cephalad to caudal), making the exam easy to remember. The CLUEparasternal, lung, and subcostal (CLUE‐PLUS), designed for the urgent evaluation of unexplained dyspnea or hypotension, uses a work backward imaging format (from left ventricle to right atrium) and a single cardiac transducer for simplicity. The PLAX view screens for left ventricular systolic dysfunction and then left atrial enlargement. Next, a brief 4‐point lung exam screens for ultrasonic lung comets and pleural effusion. A subcostal view of the heart is used to evaluate right ventricular size and pericardial effusion, and finally the inferior vena cava is evaluated for central venous pressures. CLUEs are taught in bedside and didactic formats over the 3 years of residency with formal competency testing after lecture attendance, practice imaging in our echo‐vascular laboratories, participation in rounds, and completion of at least 30 supervised examinations.

Reaffirming the Role of the Internist

Although emergency44 and critical care45 medical subspecialties have begun to train their constituencies in HCU, general diagnostic techniques that have wide‐ranging application in medical illness should be the evidence‐based tools of the internist. The rejuvenation of bedside examination using HCU on multiple organ systems should be orchestrated within internal medicine and not simply evolve as an unedited collection of all subspecialty organ ultrasound examinations. Device development can then be customized and made affordable for use in general internal medicine, perhaps limiting the unnecessary production costs and training requirements for advanced Doppler or multiple transducers.

Concern has been raised about the medical and economic impact of training internists in HCU. Although training costs can be incorporated in residency or hospital‐based continuing medical education, discussions regarding reimbursement for cardiac imaging require a distinction between the brief application of ultrasound using a small device by a nontraditional user and a limited echocardiogram as defined by payers and professional societies.46 To date, no procedural code or reimbursement has yet been approved for ultrasound‐assisted physical examination using HCU devices and likely awaits outcome data. There is also concern about the possibility of errors being made by HCU use by briefly trained physicians. Patient care and cost‐savings depend on HCU accuracy, being liable both for unnecessary referrals due to false‐positive screening HCU exams and delays in diagnosis due to false‐negative examinations. However, such errors are commonplace and accepted with standard physical examination techniques and the current use of the stethoscope, both of which lack sensitivity when compared to HCU.

HCU is a disruptive technology.47 However, unlike the successful disruption that small desktop computers had on their mainframe counterparts, HCU devices appeared before the operating system of their clinical application had been formulated, making dissemination to new users nearly impossible. Furthermore, placing ultrasound transducers into the hands of nontraditional users often alienates or displaces established users of ultrasound as well as established untrained members within the profession. Competency requirements will have to be derived, preferably from studies performed within the profession for specific uses in internal medicine. Perhaps championed by hospitalists and driven by hospital‐based outcome studies, the use of HCU by internists as a physical exam technique will require advocacy by internists themselves. The alternative, having the hospitalist ask the emergency department physician for help in examining the patient, is difficult to imagine. The answer to whether the hospitalist should use HCU should be a resounding yesbased upon the benefit of earlier, more accurate examination and the value of preserving the diagnostic role of the internist at the bedside. In regard to the latter, it is a concept worth fighting for.