Cellulitis is a common infection causing inflammation of the skin and subcutaneous tissues. Cellulitis has been attributed to gram‐positive organisms through historical evaluations including fine‐needle aspirates and punch biopsies of the infected tissue.[1] Neither of these diagnostic tests is currently used due to their invasiveness, poor diagnostic yield, and availability. Similarly, readily available tests such as blood cultures provide an etiology <5% of the time[1] and are not cost‐effective for most patients for diagnosing cellulitis.[2] In addition, the prevalence of methicillin‐resistant Staphylococcus aureus (MRSA) has steadily increased, complicating decisions about antibiotic selection.[3] The result of this uncertainty is a large variation in practice with respect to antibiotic and imaging selection for patients with a diagnosis of cellulitis.

University of Utah Health Care (UUHC) performed benchmarking for the management of cellulitis using the University HealthSystem Consortium (UHC) database and associated CareFx analytics tool. Benchmarking demonstrated that UUHC had a greater percentage of broad‐spectrum antibiotic use (defined as vancomycin, piperacillin/tazobactam, or carbapenems) than the top 5 performing UHC facilities for International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) diagnoses of cellulitis (vancomycin 83% vs 58% and carbapenem or piperacillin/tazobactam 44% vs 16%). Advanced imaging (computed tomography [CT] or magnetic resonance imaging [MRI]) for the diagnosis of cellulitis was also found to be an opportunity for improvement (CT 27% vs 20% and MRI 8% vs 5%). The hospitalist group (most patients admitted with cellulitis were on this service) believed these data reflected current practice, as there was no standard of treatment for cellulitis despite an active order set. Therefore, cellulitis was considered an opportunity to improve value to our patients. A standardized clinical care pathway was created, as such pathways have demonstrated a reduction in variation in practice and improved efficiency and effectiveness of care for multiple disease states including cellulitis.[4, 5] We hypothesized that implementation of an evidence‐based care pathway would decrease broad‐spectrum antibiotic use, cost, and use of advanced imaging without having any adverse effects on clinical outcomes such as length of stay (LOS) or readmission.

METHODS

Intervention

Initial steps involved the formation of a multidisciplinary team including key stakeholders from the hospitalist group, infectious diseases, the emergency department (ED), and nursing. This multidisciplinary team was charged with developing a clinical care pathway appropriate for local implementation. National guidance for the care pathway was mainly obtained from the Infectious Disease Society of America (IDSA) guidelines on skin and soft tissue infections (SSTIs)[6] and MRSA.[7] Specific attention was paid to recommendations on blood cultures (only when systemically ill), imaging (rarely needed), antibiotic selection (rarely gram‐negative coverage and consideration of MRSA coverage), and patient‐care principles that are often overlooked (elevation of the affected extremity). A distinction of purulent versus nonpurulent cellulitis was adopted based on the guidelines and a prospective evaluation of the care of patients with nonpurulent cellulitis.[8] The 2014 IDSA update on SSTIs incorporates this distinction more clearly in hopes of determining staphylococcal versus streptococcal infections.[9] After multiple iterations, an agreed‐upon care pathway was created that excluded patients with neutropenia, osteomyelitis, diabetic foot ulcerations; hand, perineal, periorbital, or surgical site infections; and human or animal bites (Figure 1). After the care pathway was determined, interventions were performed to implement this change.

Figure 1

Education of all providers involved included discussion of cellulitis as a disease process, presentation of benchmarking data, dissemination of the care pathway to hospitalist and ED physicians, teaching conferences for internal medicine residents and ED residents, and reinforcement of these concepts at the beginning of resident rotations.

Incorporation of the care pathway into the existing electronic order sets for cellulitis care in the inpatient and ED settings, with links to the care pathway, links to excluded disease processes (eg, hand cellulitis), preselection of commonly needed items (eg, elevate leg), and recommendations for antibiotic selection based on categories of purulent or nonpurulent cellulitis. The electronic health record (EHR) did not allow for forced order set usage, so the order set required selection by the admitting physician if indicated. Additionally, an embedded 48‐hour order set could be accessed at any time by the ordering physician and included vancomycin dosing. Specific changes to the preexisting order set included the development of sections for purulent and nonpurulent cellulitis as well as recommended antibiotics. Piperacillin/tazobactam and nafcillin were both removed and vancomycin was limited to the purulent subheading. Additionally, elevation of the extremity was preselected, and orderables for imaging (chest x‐ray and duplex ultrasound), antiulcer prophylaxis, telemetry, and electrocardiograph were all removed.

Audit and feedback of cases of cellulitis and broad‐spectrum antibiotic usage was performed by a senior hospitalist.

RESULTS

Order Set Use Groups Analysis

Descriptive statistics and simple comparison before/after the intervention for the 2 study groups are shown in the last 6 columns of Table 1. Among patients for whom order sets were used, broad‐spectrum antibiotic usage significantly decreased from 78% before the intervention to 49% after the intervention (P < 0.001). In contrast, among patients for whom order sets were not used, broad‐spectrum antibiotic usage remained relatively constant71% before the intervention versus 67% after the intervention (P = 0.50). Figure 2 shows semiannual changes in the prescription of broad‐spectrum antibiotics. There is a noticeable drop after the intervention among patients for whom order sets were used.

Figure 2

Analysis of the interaction between time period and order set usage is provided in Table 3. After the intervention, patients for whom the order sets were used had greater improvement in broad‐spectrum antibiotic selection (75% decrease, P < 0.001) and LOS (25% decrease, P = 0.041) than patients for whom order sets were not used. Pharmacy costs also decreased by 13% more among patients for whom the order sets were used, although the interaction was not statistically significant (P = 0.074). Laboratory costs decreased in both groups, but order set use did not demonstrate an interaction (P = 0.5). Similar results were found for the subgroups of admitted patients and patients seen in the EDOU.

Table 3.Differential Impact of the Intervention on Process Metrics, Clinical, and Cost Outcomes in Two Order Set Use Levels

Logistic Regression
Outcome Variables	Selected Predictor Variables	Odds*	Percent Change	P
Gamma Regression
Outcome Variables	Selected Predictor Variables	Fold Change*	Percent Change	P
* Exponentiation of the parameter for the variable represents odds for categorical variables and fold change in amount for continuous variables. Minus sign represents decrease in percent change in odds or fold change. P values are based on generalized linear models including gender, age at admission, Charlson Comorbidity Index, hospitalization status, order set use, time period, and interaction term between time period and order set use as predictor variables.
Broad spectrum antibiotics	Time period	0.84 (0.50, 1.40)	16% (50% to 40%)	0.50
	Time periodorder set	0.25 (0.12, 0.52)	75% (88% to 48%)	<0.001
MRI done	Time period	1.04 (0.49, 2.20)	4% (51% to 120%)	0.92
	Time periodorder set	0.44 (0.14, 1.38)	56% (86% to 38%)	0.16
CT done	Time period	0.94 (0.55, 1.60)	6% (45% to 60%)	0.81
	Time periodorder set	0.96 (0.44, 2.12)	4% (56% to 112%)	0.93
30‐day readmission	Time period	0.91 (0.33, 2.53)	9% (67% to 153%)	0.86
	Time periodorder set	0.88 (0.20, 3.93)	12% (80% to 293%)	0.87
Clinical length of stay	Time period	1.04 (0.95, 1.14)	4% (5% to 14%)	0.41
	Time periodorder set	0.87 (0.77, 0.99)	13% (23% to 1%)	0.041
Pharmacy cost ($)	Time period	0.88 (0.70, 1.12)	12% (30% to 12%)	0.31
	Time periodorder set	0.75 (0.54, 1.03)	25% (46% to 3%)	0.074
Lab cost ($)	Time period	0.53 (0.42, 0.66)	47% (58% to 34%)	<0.001
	Time periodorder set	1.11 (0.82, 1.50)	11% (18% to 50%)	0.50
Imaging cost ($)	Time period	1.00 (0.71, 1.40)	0% (29% to 40%)	0.98
	Time periodorder set	0.82 (0.51, 1.30)	18% (49% to 30%)	0.39
Facility cost ($)	Time period	0.92 (0.80, 1.05)	8% (20% to 5%)	0.22
	Time periodorder set	0.90 (0.75, 1.09)	10% (25% to 9%)	0.29

Audit and feedback was initially performed for cases of cellulitis using broad‐spectrum antibiotics. However, given the complexity of cellulitis as a disease process and the frequency of broad‐spectrum antibiotic usage, in all cases of review, it was deemed reasonable to use broad‐spectrum antibiotics. Therefore, the audit was not continued.

DISCUSSION

Care pathways have demonstrated improvement across multiple different disease states including cellulitis.[4, 5] They have been noted to reduce variation in practice and improve physician agreement about treatment options.[4] The best method for implementation is not clearly understood,[12] and there remains concern about maintaining flexibility for patient care.[13] Additionally, although implementation of pathways is often well described, evaluations of the processes are noted to frequently be weak.[12] UUHC felt that the literature supported implementing a care pathway for the diagnosis of cellulitis, but that a thorough evaluation was also needed to understand any resulting benefits or harms. Through this study, we found that the implementation of this pathway resulted in a significant decrease in broad‐spectrum antibiotic use, pharmacy costs, and total facility costs. There was also a trend to decrease in imaging cost, and there were no adverse effects on LOS or 30‐day readmissions. Our findings demonstrate that care‐pathway implementation accompanied by education, pathway‐compliant electronic order sets, and audit and feedback can help drive improvements in quality while reducing costs. This finding furthers the evidence supporting standard work through the creation of clinical care pathways for cellulitis as an effective intervention.[4] Additionally, although not measured in this study, reduction of antibiotic use is supported as a measure to help reduce Clostridium difficile infections, a further potential benefit.[14]

This study has several important strengths. First, we included accurate cost analyses using the VDO tool. Given the increasing importance of improving care value, we feel the inclusion of such cost analysis is an increasingly important aspect of health service intervention evaluations. Second, we used a formal benchmarking approach to identify a priority care improvement area and to monitor changes in practice following the rollout of the intervention. We feel this approach provides a useful example on how to systematically improve care quality and value in a broader health system context. Third, we evaluated not order set implementation per se, but rather changing an existing order set. Because studies in this area generally focus on initial order set implementation, our study contributes insights on what can be expected through modifications of existing order sets based on care pathways. Fourth, the analysis accounted for a variety of variables including the CCI. Of interest, our study found that the intervention group (patients for whom order sets were used) had a lower CCI, confirming Allen et al.'s findings that diseases with predictable trajectories are the most likely to benefit from care pathways.[4] As a final strength, the narrative‐based order set intervention was relatively simple, and the inclusion criteria were broad, making the process generalizable.

Limitations of this study include that it was a single center pre‐/postintervention study and not a randomized controlled trial. Related to this limitation, the control group for which order sets were not used reflected a different patient population compared to the intervention group for which order sets were used. Specifically, it was more common for order sets to be used in the EDOU than upon admission, resulting in the order set group consisting of patients with less comorbidities than patients in the nonorder set group. Additionally, patients in the order set intervention group were older than in the baseline group (48.0 vs 44.2 years). However, these differences in population remained relatively stable before and after the intervention, and relevant variables including demographic factors and CCI were accounted for in the regression models. Nevertheless, it remains possible that secular trends existed that we did not capture that affected the 2 populations differently. For example, there was a separate project that overlapped with the intervention period to reduce unnecessary laboratory usage at UUHC. This intervention could have influenced the trend to decreased laboratory utilization in the postintervention period. However, there were no concurrent initiatives to reduce antibiotic use during the study period. As a final limitation, the statistical analyses have not corrected for multiple testing for the secondary outcomes.

CONCLUSION

Using benchmark data from UHC, an academic medical center was able to identify an opportunity for improving the care of patients with cellulitis and subsequently develop an evidence‐based care pathway. The implementation of this pathway correlated with a significant reduction of broad‐spectrum antibiotic use, pharmacy costs, and total facility costs without adverse clinical affects. An important factor in the success of the intervention was the use of electronic order sets for cellulitis, which provided support for the implementation of the care pathway. This study demonstrates that the intervention was not only effective overall, but that it was more effective for those patients for whom the order set was used. This study adds to the growing body of literature suggesting that a well‐defined care pathway can improve outcomes and reduce costs for patients and institutions.

Cellulitis is a common infection causing inflammation of the skin and subcutaneous tissues. Cellulitis has been attributed to gram‐positive organisms through historical evaluations including fine‐needle aspirates and punch biopsies of the infected tissue.[1] Neither of these diagnostic tests is currently used due to their invasiveness, poor diagnostic yield, and availability. Similarly, readily available tests such as blood cultures provide an etiology <5% of the time[1] and are not cost‐effective for most patients for diagnosing cellulitis.[2] In addition, the prevalence of methicillin‐resistant Staphylococcus aureus (MRSA) has steadily increased, complicating decisions about antibiotic selection.[3] The result of this uncertainty is a large variation in practice with respect to antibiotic and imaging selection for patients with a diagnosis of cellulitis.

University of Utah Health Care (UUHC) performed benchmarking for the management of cellulitis using the University HealthSystem Consortium (UHC) database and associated CareFx analytics tool. Benchmarking demonstrated that UUHC had a greater percentage of broad‐spectrum antibiotic use (defined as vancomycin, piperacillin/tazobactam, or carbapenems) than the top 5 performing UHC facilities for International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) diagnoses of cellulitis (vancomycin 83% vs 58% and carbapenem or piperacillin/tazobactam 44% vs 16%). Advanced imaging (computed tomography [CT] or magnetic resonance imaging [MRI]) for the diagnosis of cellulitis was also found to be an opportunity for improvement (CT 27% vs 20% and MRI 8% vs 5%). The hospitalist group (most patients admitted with cellulitis were on this service) believed these data reflected current practice, as there was no standard of treatment for cellulitis despite an active order set. Therefore, cellulitis was considered an opportunity to improve value to our patients. A standardized clinical care pathway was created, as such pathways have demonstrated a reduction in variation in practice and improved efficiency and effectiveness of care for multiple disease states including cellulitis.[4, 5] We hypothesized that implementation of an evidence‐based care pathway would decrease broad‐spectrum antibiotic use, cost, and use of advanced imaging without having any adverse effects on clinical outcomes such as length of stay (LOS) or readmission.

METHODS

Intervention

Initial steps involved the formation of a multidisciplinary team including key stakeholders from the hospitalist group, infectious diseases, the emergency department (ED), and nursing. This multidisciplinary team was charged with developing a clinical care pathway appropriate for local implementation. National guidance for the care pathway was mainly obtained from the Infectious Disease Society of America (IDSA) guidelines on skin and soft tissue infections (SSTIs)[6] and MRSA.[7] Specific attention was paid to recommendations on blood cultures (only when systemically ill), imaging (rarely needed), antibiotic selection (rarely gram‐negative coverage and consideration of MRSA coverage), and patient‐care principles that are often overlooked (elevation of the affected extremity). A distinction of purulent versus nonpurulent cellulitis was adopted based on the guidelines and a prospective evaluation of the care of patients with nonpurulent cellulitis.[8] The 2014 IDSA update on SSTIs incorporates this distinction more clearly in hopes of determining staphylococcal versus streptococcal infections.[9] After multiple iterations, an agreed‐upon care pathway was created that excluded patients with neutropenia, osteomyelitis, diabetic foot ulcerations; hand, perineal, periorbital, or surgical site infections; and human or animal bites (Figure 1). After the care pathway was determined, interventions were performed to implement this change.

Figure 1

Education of all providers involved included discussion of cellulitis as a disease process, presentation of benchmarking data, dissemination of the care pathway to hospitalist and ED physicians, teaching conferences for internal medicine residents and ED residents, and reinforcement of these concepts at the beginning of resident rotations.

Incorporation of the care pathway into the existing electronic order sets for cellulitis care in the inpatient and ED settings, with links to the care pathway, links to excluded disease processes (eg, hand cellulitis), preselection of commonly needed items (eg, elevate leg), and recommendations for antibiotic selection based on categories of purulent or nonpurulent cellulitis. The electronic health record (EHR) did not allow for forced order set usage, so the order set required selection by the admitting physician if indicated. Additionally, an embedded 48‐hour order set could be accessed at any time by the ordering physician and included vancomycin dosing. Specific changes to the preexisting order set included the development of sections for purulent and nonpurulent cellulitis as well as recommended antibiotics. Piperacillin/tazobactam and nafcillin were both removed and vancomycin was limited to the purulent subheading. Additionally, elevation of the extremity was preselected, and orderables for imaging (chest x‐ray and duplex ultrasound), antiulcer prophylaxis, telemetry, and electrocardiograph were all removed.

Audit and feedback of cases of cellulitis and broad‐spectrum antibiotic usage was performed by a senior hospitalist.

RESULTS

Order Set Use Groups Analysis

Descriptive statistics and simple comparison before/after the intervention for the 2 study groups are shown in the last 6 columns of Table 1. Among patients for whom order sets were used, broad‐spectrum antibiotic usage significantly decreased from 78% before the intervention to 49% after the intervention (P < 0.001). In contrast, among patients for whom order sets were not used, broad‐spectrum antibiotic usage remained relatively constant71% before the intervention versus 67% after the intervention (P = 0.50). Figure 2 shows semiannual changes in the prescription of broad‐spectrum antibiotics. There is a noticeable drop after the intervention among patients for whom order sets were used.

Figure 2

Analysis of the interaction between time period and order set usage is provided in Table 3. After the intervention, patients for whom the order sets were used had greater improvement in broad‐spectrum antibiotic selection (75% decrease, P < 0.001) and LOS (25% decrease, P = 0.041) than patients for whom order sets were not used. Pharmacy costs also decreased by 13% more among patients for whom the order sets were used, although the interaction was not statistically significant (P = 0.074). Laboratory costs decreased in both groups, but order set use did not demonstrate an interaction (P = 0.5). Similar results were found for the subgroups of admitted patients and patients seen in the EDOU.

Table 3.Differential Impact of the Intervention on Process Metrics, Clinical, and Cost Outcomes in Two Order Set Use Levels

Logistic Regression
Outcome Variables	Selected Predictor Variables	Odds*	Percent Change	P
Gamma Regression
Outcome Variables	Selected Predictor Variables	Fold Change*	Percent Change	P
* Exponentiation of the parameter for the variable represents odds for categorical variables and fold change in amount for continuous variables. Minus sign represents decrease in percent change in odds or fold change. P values are based on generalized linear models including gender, age at admission, Charlson Comorbidity Index, hospitalization status, order set use, time period, and interaction term between time period and order set use as predictor variables.
Broad spectrum antibiotics	Time period	0.84 (0.50, 1.40)	16% (50% to 40%)	0.50
	Time periodorder set	0.25 (0.12, 0.52)	75% (88% to 48%)	<0.001
MRI done	Time period	1.04 (0.49, 2.20)	4% (51% to 120%)	0.92
	Time periodorder set	0.44 (0.14, 1.38)	56% (86% to 38%)	0.16
CT done	Time period	0.94 (0.55, 1.60)	6% (45% to 60%)	0.81
	Time periodorder set	0.96 (0.44, 2.12)	4% (56% to 112%)	0.93
30‐day readmission	Time period	0.91 (0.33, 2.53)	9% (67% to 153%)	0.86
	Time periodorder set	0.88 (0.20, 3.93)	12% (80% to 293%)	0.87
Clinical length of stay	Time period	1.04 (0.95, 1.14)	4% (5% to 14%)	0.41
	Time periodorder set	0.87 (0.77, 0.99)	13% (23% to 1%)	0.041
Pharmacy cost ($)	Time period	0.88 (0.70, 1.12)	12% (30% to 12%)	0.31
	Time periodorder set	0.75 (0.54, 1.03)	25% (46% to 3%)	0.074
Lab cost ($)	Time period	0.53 (0.42, 0.66)	47% (58% to 34%)	<0.001
	Time periodorder set	1.11 (0.82, 1.50)	11% (18% to 50%)	0.50
Imaging cost ($)	Time period	1.00 (0.71, 1.40)	0% (29% to 40%)	0.98
	Time periodorder set	0.82 (0.51, 1.30)	18% (49% to 30%)	0.39
Facility cost ($)	Time period	0.92 (0.80, 1.05)	8% (20% to 5%)	0.22
	Time periodorder set	0.90 (0.75, 1.09)	10% (25% to 9%)	0.29

Audit and feedback was initially performed for cases of cellulitis using broad‐spectrum antibiotics. However, given the complexity of cellulitis as a disease process and the frequency of broad‐spectrum antibiotic usage, in all cases of review, it was deemed reasonable to use broad‐spectrum antibiotics. Therefore, the audit was not continued.

DISCUSSION

Care pathways have demonstrated improvement across multiple different disease states including cellulitis.[4, 5] They have been noted to reduce variation in practice and improve physician agreement about treatment options.[4] The best method for implementation is not clearly understood,[12] and there remains concern about maintaining flexibility for patient care.[13] Additionally, although implementation of pathways is often well described, evaluations of the processes are noted to frequently be weak.[12] UUHC felt that the literature supported implementing a care pathway for the diagnosis of cellulitis, but that a thorough evaluation was also needed to understand any resulting benefits or harms. Through this study, we found that the implementation of this pathway resulted in a significant decrease in broad‐spectrum antibiotic use, pharmacy costs, and total facility costs. There was also a trend to decrease in imaging cost, and there were no adverse effects on LOS or 30‐day readmissions. Our findings demonstrate that care‐pathway implementation accompanied by education, pathway‐compliant electronic order sets, and audit and feedback can help drive improvements in quality while reducing costs. This finding furthers the evidence supporting standard work through the creation of clinical care pathways for cellulitis as an effective intervention.[4] Additionally, although not measured in this study, reduction of antibiotic use is supported as a measure to help reduce Clostridium difficile infections, a further potential benefit.[14]

This study has several important strengths. First, we included accurate cost analyses using the VDO tool. Given the increasing importance of improving care value, we feel the inclusion of such cost analysis is an increasingly important aspect of health service intervention evaluations. Second, we used a formal benchmarking approach to identify a priority care improvement area and to monitor changes in practice following the rollout of the intervention. We feel this approach provides a useful example on how to systematically improve care quality and value in a broader health system context. Third, we evaluated not order set implementation per se, but rather changing an existing order set. Because studies in this area generally focus on initial order set implementation, our study contributes insights on what can be expected through modifications of existing order sets based on care pathways. Fourth, the analysis accounted for a variety of variables including the CCI. Of interest, our study found that the intervention group (patients for whom order sets were used) had a lower CCI, confirming Allen et al.'s findings that diseases with predictable trajectories are the most likely to benefit from care pathways.[4] As a final strength, the narrative‐based order set intervention was relatively simple, and the inclusion criteria were broad, making the process generalizable.

Limitations of this study include that it was a single center pre‐/postintervention study and not a randomized controlled trial. Related to this limitation, the control group for which order sets were not used reflected a different patient population compared to the intervention group for which order sets were used. Specifically, it was more common for order sets to be used in the EDOU than upon admission, resulting in the order set group consisting of patients with less comorbidities than patients in the nonorder set group. Additionally, patients in the order set intervention group were older than in the baseline group (48.0 vs 44.2 years). However, these differences in population remained relatively stable before and after the intervention, and relevant variables including demographic factors and CCI were accounted for in the regression models. Nevertheless, it remains possible that secular trends existed that we did not capture that affected the 2 populations differently. For example, there was a separate project that overlapped with the intervention period to reduce unnecessary laboratory usage at UUHC. This intervention could have influenced the trend to decreased laboratory utilization in the postintervention period. However, there were no concurrent initiatives to reduce antibiotic use during the study period. As a final limitation, the statistical analyses have not corrected for multiple testing for the secondary outcomes.

CONCLUSION

Using benchmark data from UHC, an academic medical center was able to identify an opportunity for improving the care of patients with cellulitis and subsequently develop an evidence‐based care pathway. The implementation of this pathway correlated with a significant reduction of broad‐spectrum antibiotic use, pharmacy costs, and total facility costs without adverse clinical affects. An important factor in the success of the intervention was the use of electronic order sets for cellulitis, which provided support for the implementation of the care pathway. This study demonstrates that the intervention was not only effective overall, but that it was more effective for those patients for whom the order set was used. This study adds to the growing body of literature suggesting that a well‐defined care pathway can improve outcomes and reduce costs for patients and institutions.

Cellulitis is a common infection causing inflammation of the skin and subcutaneous tissues. Cellulitis has been attributed to gram‐positive organisms through historical evaluations including fine‐needle aspirates and punch biopsies of the infected tissue.[1] Neither of these diagnostic tests is currently used due to their invasiveness, poor diagnostic yield, and availability. Similarly, readily available tests such as blood cultures provide an etiology <5% of the time[1] and are not cost‐effective for most patients for diagnosing cellulitis.[2] In addition, the prevalence of methicillin‐resistant Staphylococcus aureus (MRSA) has steadily increased, complicating decisions about antibiotic selection.[3] The result of this uncertainty is a large variation in practice with respect to antibiotic and imaging selection for patients with a diagnosis of cellulitis.

University of Utah Health Care (UUHC) performed benchmarking for the management of cellulitis using the University HealthSystem Consortium (UHC) database and associated CareFx analytics tool. Benchmarking demonstrated that UUHC had a greater percentage of broad‐spectrum antibiotic use (defined as vancomycin, piperacillin/tazobactam, or carbapenems) than the top 5 performing UHC facilities for International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) diagnoses of cellulitis (vancomycin 83% vs 58% and carbapenem or piperacillin/tazobactam 44% vs 16%). Advanced imaging (computed tomography [CT] or magnetic resonance imaging [MRI]) for the diagnosis of cellulitis was also found to be an opportunity for improvement (CT 27% vs 20% and MRI 8% vs 5%). The hospitalist group (most patients admitted with cellulitis were on this service) believed these data reflected current practice, as there was no standard of treatment for cellulitis despite an active order set. Therefore, cellulitis was considered an opportunity to improve value to our patients. A standardized clinical care pathway was created, as such pathways have demonstrated a reduction in variation in practice and improved efficiency and effectiveness of care for multiple disease states including cellulitis.[4, 5] We hypothesized that implementation of an evidence‐based care pathway would decrease broad‐spectrum antibiotic use, cost, and use of advanced imaging without having any adverse effects on clinical outcomes such as length of stay (LOS) or readmission.

METHODS

Intervention

Initial steps involved the formation of a multidisciplinary team including key stakeholders from the hospitalist group, infectious diseases, the emergency department (ED), and nursing. This multidisciplinary team was charged with developing a clinical care pathway appropriate for local implementation. National guidance for the care pathway was mainly obtained from the Infectious Disease Society of America (IDSA) guidelines on skin and soft tissue infections (SSTIs)[6] and MRSA.[7] Specific attention was paid to recommendations on blood cultures (only when systemically ill), imaging (rarely needed), antibiotic selection (rarely gram‐negative coverage and consideration of MRSA coverage), and patient‐care principles that are often overlooked (elevation of the affected extremity). A distinction of purulent versus nonpurulent cellulitis was adopted based on the guidelines and a prospective evaluation of the care of patients with nonpurulent cellulitis.[8] The 2014 IDSA update on SSTIs incorporates this distinction more clearly in hopes of determining staphylococcal versus streptococcal infections.[9] After multiple iterations, an agreed‐upon care pathway was created that excluded patients with neutropenia, osteomyelitis, diabetic foot ulcerations; hand, perineal, periorbital, or surgical site infections; and human or animal bites (Figure 1). After the care pathway was determined, interventions were performed to implement this change.

Figure 1

Education of all providers involved included discussion of cellulitis as a disease process, presentation of benchmarking data, dissemination of the care pathway to hospitalist and ED physicians, teaching conferences for internal medicine residents and ED residents, and reinforcement of these concepts at the beginning of resident rotations.

Incorporation of the care pathway into the existing electronic order sets for cellulitis care in the inpatient and ED settings, with links to the care pathway, links to excluded disease processes (eg, hand cellulitis), preselection of commonly needed items (eg, elevate leg), and recommendations for antibiotic selection based on categories of purulent or nonpurulent cellulitis. The electronic health record (EHR) did not allow for forced order set usage, so the order set required selection by the admitting physician if indicated. Additionally, an embedded 48‐hour order set could be accessed at any time by the ordering physician and included vancomycin dosing. Specific changes to the preexisting order set included the development of sections for purulent and nonpurulent cellulitis as well as recommended antibiotics. Piperacillin/tazobactam and nafcillin were both removed and vancomycin was limited to the purulent subheading. Additionally, elevation of the extremity was preselected, and orderables for imaging (chest x‐ray and duplex ultrasound), antiulcer prophylaxis, telemetry, and electrocardiograph were all removed.

Audit and feedback of cases of cellulitis and broad‐spectrum antibiotic usage was performed by a senior hospitalist.

RESULTS

Order Set Use Groups Analysis

Descriptive statistics and simple comparison before/after the intervention for the 2 study groups are shown in the last 6 columns of Table 1. Among patients for whom order sets were used, broad‐spectrum antibiotic usage significantly decreased from 78% before the intervention to 49% after the intervention (P < 0.001). In contrast, among patients for whom order sets were not used, broad‐spectrum antibiotic usage remained relatively constant71% before the intervention versus 67% after the intervention (P = 0.50). Figure 2 shows semiannual changes in the prescription of broad‐spectrum antibiotics. There is a noticeable drop after the intervention among patients for whom order sets were used.

Figure 2

Analysis of the interaction between time period and order set usage is provided in Table 3. After the intervention, patients for whom the order sets were used had greater improvement in broad‐spectrum antibiotic selection (75% decrease, P < 0.001) and LOS (25% decrease, P = 0.041) than patients for whom order sets were not used. Pharmacy costs also decreased by 13% more among patients for whom the order sets were used, although the interaction was not statistically significant (P = 0.074). Laboratory costs decreased in both groups, but order set use did not demonstrate an interaction (P = 0.5). Similar results were found for the subgroups of admitted patients and patients seen in the EDOU.

Table 3.Differential Impact of the Intervention on Process Metrics, Clinical, and Cost Outcomes in Two Order Set Use Levels

Logistic Regression
Outcome Variables	Selected Predictor Variables	Odds*	Percent Change	P
Gamma Regression
Outcome Variables	Selected Predictor Variables	Fold Change*	Percent Change	P
* Exponentiation of the parameter for the variable represents odds for categorical variables and fold change in amount for continuous variables. Minus sign represents decrease in percent change in odds or fold change. P values are based on generalized linear models including gender, age at admission, Charlson Comorbidity Index, hospitalization status, order set use, time period, and interaction term between time period and order set use as predictor variables.
Broad spectrum antibiotics	Time period	0.84 (0.50, 1.40)	16% (50% to 40%)	0.50
	Time periodorder set	0.25 (0.12, 0.52)	75% (88% to 48%)	<0.001
MRI done	Time period	1.04 (0.49, 2.20)	4% (51% to 120%)	0.92
	Time periodorder set	0.44 (0.14, 1.38)	56% (86% to 38%)	0.16
CT done	Time period	0.94 (0.55, 1.60)	6% (45% to 60%)	0.81
	Time periodorder set	0.96 (0.44, 2.12)	4% (56% to 112%)	0.93
30‐day readmission	Time period	0.91 (0.33, 2.53)	9% (67% to 153%)	0.86
	Time periodorder set	0.88 (0.20, 3.93)	12% (80% to 293%)	0.87
Clinical length of stay	Time period	1.04 (0.95, 1.14)	4% (5% to 14%)	0.41
	Time periodorder set	0.87 (0.77, 0.99)	13% (23% to 1%)	0.041
Pharmacy cost ($)	Time period	0.88 (0.70, 1.12)	12% (30% to 12%)	0.31
	Time periodorder set	0.75 (0.54, 1.03)	25% (46% to 3%)	0.074
Lab cost ($)	Time period	0.53 (0.42, 0.66)	47% (58% to 34%)	<0.001
	Time periodorder set	1.11 (0.82, 1.50)	11% (18% to 50%)	0.50
Imaging cost ($)	Time period	1.00 (0.71, 1.40)	0% (29% to 40%)	0.98
	Time periodorder set	0.82 (0.51, 1.30)	18% (49% to 30%)	0.39
Facility cost ($)	Time period	0.92 (0.80, 1.05)	8% (20% to 5%)	0.22
	Time periodorder set	0.90 (0.75, 1.09)	10% (25% to 9%)	0.29

Audit and feedback was initially performed for cases of cellulitis using broad‐spectrum antibiotics. However, given the complexity of cellulitis as a disease process and the frequency of broad‐spectrum antibiotic usage, in all cases of review, it was deemed reasonable to use broad‐spectrum antibiotics. Therefore, the audit was not continued.

DISCUSSION

Care pathways have demonstrated improvement across multiple different disease states including cellulitis.[4, 5] They have been noted to reduce variation in practice and improve physician agreement about treatment options.[4] The best method for implementation is not clearly understood,[12] and there remains concern about maintaining flexibility for patient care.[13] Additionally, although implementation of pathways is often well described, evaluations of the processes are noted to frequently be weak.[12] UUHC felt that the literature supported implementing a care pathway for the diagnosis of cellulitis, but that a thorough evaluation was also needed to understand any resulting benefits or harms. Through this study, we found that the implementation of this pathway resulted in a significant decrease in broad‐spectrum antibiotic use, pharmacy costs, and total facility costs. There was also a trend to decrease in imaging cost, and there were no adverse effects on LOS or 30‐day readmissions. Our findings demonstrate that care‐pathway implementation accompanied by education, pathway‐compliant electronic order sets, and audit and feedback can help drive improvements in quality while reducing costs. This finding furthers the evidence supporting standard work through the creation of clinical care pathways for cellulitis as an effective intervention.[4] Additionally, although not measured in this study, reduction of antibiotic use is supported as a measure to help reduce Clostridium difficile infections, a further potential benefit.[14]

This study has several important strengths. First, we included accurate cost analyses using the VDO tool. Given the increasing importance of improving care value, we feel the inclusion of such cost analysis is an increasingly important aspect of health service intervention evaluations. Second, we used a formal benchmarking approach to identify a priority care improvement area and to monitor changes in practice following the rollout of the intervention. We feel this approach provides a useful example on how to systematically improve care quality and value in a broader health system context. Third, we evaluated not order set implementation per se, but rather changing an existing order set. Because studies in this area generally focus on initial order set implementation, our study contributes insights on what can be expected through modifications of existing order sets based on care pathways. Fourth, the analysis accounted for a variety of variables including the CCI. Of interest, our study found that the intervention group (patients for whom order sets were used) had a lower CCI, confirming Allen et al.'s findings that diseases with predictable trajectories are the most likely to benefit from care pathways.[4] As a final strength, the narrative‐based order set intervention was relatively simple, and the inclusion criteria were broad, making the process generalizable.

Limitations of this study include that it was a single center pre‐/postintervention study and not a randomized controlled trial. Related to this limitation, the control group for which order sets were not used reflected a different patient population compared to the intervention group for which order sets were used. Specifically, it was more common for order sets to be used in the EDOU than upon admission, resulting in the order set group consisting of patients with less comorbidities than patients in the nonorder set group. Additionally, patients in the order set intervention group were older than in the baseline group (48.0 vs 44.2 years). However, these differences in population remained relatively stable before and after the intervention, and relevant variables including demographic factors and CCI were accounted for in the regression models. Nevertheless, it remains possible that secular trends existed that we did not capture that affected the 2 populations differently. For example, there was a separate project that overlapped with the intervention period to reduce unnecessary laboratory usage at UUHC. This intervention could have influenced the trend to decreased laboratory utilization in the postintervention period. However, there were no concurrent initiatives to reduce antibiotic use during the study period. As a final limitation, the statistical analyses have not corrected for multiple testing for the secondary outcomes.

CONCLUSION

Using benchmark data from UHC, an academic medical center was able to identify an opportunity for improving the care of patients with cellulitis and subsequently develop an evidence‐based care pathway. The implementation of this pathway correlated with a significant reduction of broad‐spectrum antibiotic use, pharmacy costs, and total facility costs without adverse clinical affects. An important factor in the success of the intervention was the use of electronic order sets for cellulitis, which provided support for the implementation of the care pathway. This study demonstrates that the intervention was not only effective overall, but that it was more effective for those patients for whom the order set was used. This study adds to the growing body of literature suggesting that a well‐defined care pathway can improve outcomes and reduce costs for patients and institutions.

Hospitalists commonly encounter pleural effusions, and their detection and characterization by point‐of‐care ultrasound can guide management. Approximately 44% to 57% of hospitalized patients with bacterial pneumonia,[1, 2] and up to 62% of intensive care unit (ICU) patients[3] have a pleural effusion. For patients with a parapneumonic effusion, hospitalists can use ultrasound to quantify and characterize pleural fluid to determine whether diagnostic or therapeutic drainage is indicated, as well as guide performance of thoracentesis. For patients with lung cancer, detection of a malignant pleural effusion changes staging to stage IV, regardless of tumor size or lymph node involvement, and hospitalists may discuss more appropriate treatment options with patients and consultants.

Routine use of pleural ultrasonography may help hospitalists provide high‐value care by reducing ancillary testing, including computerized tomography (CT) scans that expose patients to ionizing radiation, and reducing complications of thoracentesis. However, many hospitalists may not be familiar with the use of point‐of‐care ultrasound. A national survey in 2012 revealed only 25% of internal medicine residencies have formal curricula to teach point‐of‐care ultrasonography.[4] The purpose of this review is to provide an overview of how point‐of‐care ultrasound can be utilized by hospitalists in the care of patients with pleural effusions. We review the literature on the diagnosis and evaluation of pleural effusions with ultrasound, as well as techniques to examine and drain the pleural space.

DIAGNOSIS OF PLEURAL EFFUSION

PLEURAL ULTRASOUND EXAMINATION

A pleural ultrasound exam may be performed as part of a complete lung ultrasound exam, such as the BLUE (Bedside Lung Ultrasound in Emergency) protocol,[27] or a focused exam to evaluate a suspected or known pleural effusion seen on chest radiograph or CT scan.[27] Free‐flowing pleural effusions accumulate in the most dependent portions of the thorax, most commonly, the posterolateral costophrenic recesses in semirecumbent or seated patients, but anteriorly in mechanically ventilated patients in a prone position.

A low‐frequency (25 MHz) phased‐array transducer is generally preferred for imaging in between the ribs. High‐frequency linear transducers do not provide adequate penetration to visualize deep structures, but do provide superior visualization of the pleural line to assess pleural thickness, measure pleural depth, and evaluate for pneumothorax.

Pleural effusions are best evaluated starting at the level of the diaphragm. Place the transducer in a longitudinal plane on the posterior axillary line at the level of the diaphragm with the transducer orientation marker (notch) pointed cephalad (Figure 1). Five structures must be definitively identified to diagnose a pleural effusion: liver/spleen, diaphragm, pleural fluid, lung, and chest wall (Figure 2A). Large pleural effusions compress the adjacent lung causing atelectasis, which gives the lung a tissue‐like echogenicity similar to the liver (Figure 2B). Static air bronchograms are commonly seen in atelectatic lung bases with pleural effusions.[28]

Figure 1

Figure 2

Color flow Doppler and M‐mode ultrasound may be utilized as adjuncts to routine 2‐dimensional ultrasonography. Free‐flowing pleural effusions will demonstrate flow by color Doppler (Figure 3A). Using M‐mode ultrasound, the lung can been seen moving within a pleural effusion to and from the chest wall (sinusoid sign).[29] Absence of flow or movement is seen with dense pleural loculations, pleural thickening, and peripheral lung or pleural masses (Figure 3B).

Figure 3

CHARACTERIZATION OF PLEURAL EFFUSION

Simple Versus Complex Effusions

Based on its sonographic appearance, pleural effusions are categorized as simple or complex. Simple pleural effusions are anechoic and usually transudative. Complex pleural effusions are subcategorized as homogeneously or heterogeneously echogenic, with or without septations, and are more often exudative.[33]

Effusions with heterogeneous echogenicity with swirling echoes suggest high cellular content that is often associated with malignancy.[34] Fibrinous stranding, septations, and loculations also suggest an exudative effusion (Figure 4A), and are more readily identified and characterized on lung ultrasound than CT scan.[35]

Figure 4

Homogenously echogenic effusions are most often due to hemothorax or empyema.[36] The high cell count of a hemothorax creates a layering effect in costophrenic recesses (hematocrit sign). Empyemas develop from complex effusions that organize into collections of pus and usually have a homogeneously echogenic, speckled appearance (Figure 4B). Sonographic evidence of septations in the presence of empyema predicts the need for intrapleural fibrinolytic therapy, longer duration of drainage, and possible surgical intervention.[37]

Isolated dense loculations may be challenging to differentiate from peripheral lung or pleural lesions, such as abscess or tumor.

PLEURAL FLUID DRAINAGE

Since its first description in 1967, use of ultrasound guidance for thoracentesis has evolved to become the standard of care in many hospitals in the United States.[41] The British Thoracic Society guidelines recommend that all thoracenteses be performed with ultrasound guidance.[42] The American College of Graduate Medical Education now requires proficiency in the use of ultrasound for thoracentesis and pleural catheter insertion by pulmonary and critical care fellows.[43]

The impetus for these recommendations stems from increased procedural success and decreased complications associated with ultrasound‐guided drainage of pleural effusions. A study evaluating thoracentesis site selection based on physical exam and chest radiographs demonstrated inaccurate site selection in 15% of patients, and use of ultrasound for site selection prevented possible accidental organ puncture in 10% of all cases.[44] The success rate of thoracentesis for small pleural effusions has been shown to increase from 66% to 90% with ultrasound guidance.[42] Using ultrasound, the distance from the skin to parietal and visceral pleura can be measured to determine whether thoracentesis can be safely performed, and to guide selection of an adequate length needle (Figure 5). A minimum pleural effusion depth of 1.5 cm between the visceral and parietal pleura has been recommended to perform diagnostic thoracentesis.[28] Diagnostic thoracentesis of complex septated pleural effusions or empyemas may be performed with a straight needle, but therapeutic drainage often requires temporary insertion of a catheter. Traditionally, large‐bore chest tubes (>24 F) had been advocated to drain viscid pus, but recent evidence suggests that small‐bore catheters (1014 F) with instillation of thrombolytics may be as effective and performed with less discomfort.[45] Video‐assisted thoracoscopy to lyse septations and evacuate infected materials is indicated when chest tube drainage has failed.

Figure 5

The most common complication of pleural drainage is pneumothorax. A meta‐analysis demonstrated a reduction in post‐thoracentesis pneumothorax rates from 9% to 4% with use of ultrasound.[46] Transporting patients to radiology for ultrasound marking has not been shown to decrease pneumothorax rates compared to thoracentesis without ultrasound guidance, likely due to changes in patient position and prolonged delays between marking and drainage.[47] Postprocedure pneumothorax can be ruled out if lung sliding is visualized. A meta‐analysis demonstrated superior sensitivity and similar specificity of pleural ultrasonography versus chest radiography to detect pneumothorax (sensitivity 91% vs 50% and specificity 98% vs 99%, respectively).[48] Real‐time ultrasound guidance for thoracentesis, or use of ultrasound to track the needle tip, has not been well studied but may be performed by experienced proceduralists to drain small effusions.

FUTURE RESEARCH

Future research should focus on the clinical effectiveness of point‐of‐care pleural ultrasonography when integrated with other diagnostic tools, and application of new ultrasound technologies to evaluate pleural diseases. Routine use of point‐of‐care ultrasound as the primary imaging modality in a medical ICU demonstrated a highly statistically significant reduction in chest x‐rays (3.75 vs 0.82, P < 0.0001) and chest CT scans (0.10 vs 0.04, P = 0.0007).[49] Similar studies have yet to be performed with the use of ultrasound specifically in the management of pleural diseases. Thus, clinical effectiveness studies are needed to assess the impact of routine use of pleural ultrasound on the initiation of appropriate therapies, length of stay, and costs in the management of pleural disease.

Point‐of‐care pleural ultrasound findings need to be evaluated in the context of other clinical findings and diagnostic tests. Certain ultrasound findings have been associated with exudative pleural effusions, but whether exudative and transudative effusions can be differentiated noninvasively using ultrasound findings alone, or in combination with other clinical data, warrants investigation. Similar to severity of illness scores, models that incorporate clinical, laboratory, and ultrasound findings need to be developed to guide treatment decisions, such as type of drainage procedure, as well as prognostication.

Finally, new technologies may advance the utility of point‐of‐care pleural ultrasonography. Even though 3‐dimensional ultrasonography has been available for over 2 decades, comparative studies of conventional 2‐dimensional versus 3‐dimensional ultrasonography to characterize pleural effusions have yet to be performed. Furthermore, computer‐aided detection has been shown to facilitate interpretation of ultrasound images, but this technology has yet to be applied to pleural ultrasonography.

CONCLUSIONS

Point‐of‐care pleural ultrasound is a powerful bedside tool in the hospitalist's armamentarium that is superior to physical examination and chest radiographs in the detection and characterization of pleural effusions. Furthermore, ultrasound performs similarly when compared to CT scans but offers the advantages of decreased cost, avoidance of ionizing radiation, and availability at the bedside. Ultrasound guidance reduces complications and increases the success rate of pleural drainage procedures, leading to improved patient safety. As clinical effectiveness studies emerge revealing its true value, point‐of‐care pleural ultrasonography is likely to become the standard diagnostic tool to evaluate and manage patients with pleural effusions.

Disclosures: Dr. Restrepo is partially supported by award number K23HL096054 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health. The authors report no conflicts of interest.

Hospitalists commonly encounter pleural effusions, and their detection and characterization by point‐of‐care ultrasound can guide management. Approximately 44% to 57% of hospitalized patients with bacterial pneumonia,[1, 2] and up to 62% of intensive care unit (ICU) patients[3] have a pleural effusion. For patients with a parapneumonic effusion, hospitalists can use ultrasound to quantify and characterize pleural fluid to determine whether diagnostic or therapeutic drainage is indicated, as well as guide performance of thoracentesis. For patients with lung cancer, detection of a malignant pleural effusion changes staging to stage IV, regardless of tumor size or lymph node involvement, and hospitalists may discuss more appropriate treatment options with patients and consultants.

Routine use of pleural ultrasonography may help hospitalists provide high‐value care by reducing ancillary testing, including computerized tomography (CT) scans that expose patients to ionizing radiation, and reducing complications of thoracentesis. However, many hospitalists may not be familiar with the use of point‐of‐care ultrasound. A national survey in 2012 revealed only 25% of internal medicine residencies have formal curricula to teach point‐of‐care ultrasonography.[4] The purpose of this review is to provide an overview of how point‐of‐care ultrasound can be utilized by hospitalists in the care of patients with pleural effusions. We review the literature on the diagnosis and evaluation of pleural effusions with ultrasound, as well as techniques to examine and drain the pleural space.

DIAGNOSIS OF PLEURAL EFFUSION

PLEURAL ULTRASOUND EXAMINATION

A pleural ultrasound exam may be performed as part of a complete lung ultrasound exam, such as the BLUE (Bedside Lung Ultrasound in Emergency) protocol,[27] or a focused exam to evaluate a suspected or known pleural effusion seen on chest radiograph or CT scan.[27] Free‐flowing pleural effusions accumulate in the most dependent portions of the thorax, most commonly, the posterolateral costophrenic recesses in semirecumbent or seated patients, but anteriorly in mechanically ventilated patients in a prone position.

A low‐frequency (25 MHz) phased‐array transducer is generally preferred for imaging in between the ribs. High‐frequency linear transducers do not provide adequate penetration to visualize deep structures, but do provide superior visualization of the pleural line to assess pleural thickness, measure pleural depth, and evaluate for pneumothorax.

Pleural effusions are best evaluated starting at the level of the diaphragm. Place the transducer in a longitudinal plane on the posterior axillary line at the level of the diaphragm with the transducer orientation marker (notch) pointed cephalad (Figure 1). Five structures must be definitively identified to diagnose a pleural effusion: liver/spleen, diaphragm, pleural fluid, lung, and chest wall (Figure 2A). Large pleural effusions compress the adjacent lung causing atelectasis, which gives the lung a tissue‐like echogenicity similar to the liver (Figure 2B). Static air bronchograms are commonly seen in atelectatic lung bases with pleural effusions.[28]

Figure 1

Figure 2

Color flow Doppler and M‐mode ultrasound may be utilized as adjuncts to routine 2‐dimensional ultrasonography. Free‐flowing pleural effusions will demonstrate flow by color Doppler (Figure 3A). Using M‐mode ultrasound, the lung can been seen moving within a pleural effusion to and from the chest wall (sinusoid sign).[29] Absence of flow or movement is seen with dense pleural loculations, pleural thickening, and peripheral lung or pleural masses (Figure 3B).

Figure 3

CHARACTERIZATION OF PLEURAL EFFUSION

Simple Versus Complex Effusions

Based on its sonographic appearance, pleural effusions are categorized as simple or complex. Simple pleural effusions are anechoic and usually transudative. Complex pleural effusions are subcategorized as homogeneously or heterogeneously echogenic, with or without septations, and are more often exudative.[33]

Effusions with heterogeneous echogenicity with swirling echoes suggest high cellular content that is often associated with malignancy.[34] Fibrinous stranding, septations, and loculations also suggest an exudative effusion (Figure 4A), and are more readily identified and characterized on lung ultrasound than CT scan.[35]

Figure 4

Homogenously echogenic effusions are most often due to hemothorax or empyema.[36] The high cell count of a hemothorax creates a layering effect in costophrenic recesses (hematocrit sign). Empyemas develop from complex effusions that organize into collections of pus and usually have a homogeneously echogenic, speckled appearance (Figure 4B). Sonographic evidence of septations in the presence of empyema predicts the need for intrapleural fibrinolytic therapy, longer duration of drainage, and possible surgical intervention.[37]

Isolated dense loculations may be challenging to differentiate from peripheral lung or pleural lesions, such as abscess or tumor.

PLEURAL FLUID DRAINAGE

Since its first description in 1967, use of ultrasound guidance for thoracentesis has evolved to become the standard of care in many hospitals in the United States.[41] The British Thoracic Society guidelines recommend that all thoracenteses be performed with ultrasound guidance.[42] The American College of Graduate Medical Education now requires proficiency in the use of ultrasound for thoracentesis and pleural catheter insertion by pulmonary and critical care fellows.[43]

The impetus for these recommendations stems from increased procedural success and decreased complications associated with ultrasound‐guided drainage of pleural effusions. A study evaluating thoracentesis site selection based on physical exam and chest radiographs demonstrated inaccurate site selection in 15% of patients, and use of ultrasound for site selection prevented possible accidental organ puncture in 10% of all cases.[44] The success rate of thoracentesis for small pleural effusions has been shown to increase from 66% to 90% with ultrasound guidance.[42] Using ultrasound, the distance from the skin to parietal and visceral pleura can be measured to determine whether thoracentesis can be safely performed, and to guide selection of an adequate length needle (Figure 5). A minimum pleural effusion depth of 1.5 cm between the visceral and parietal pleura has been recommended to perform diagnostic thoracentesis.[28] Diagnostic thoracentesis of complex septated pleural effusions or empyemas may be performed with a straight needle, but therapeutic drainage often requires temporary insertion of a catheter. Traditionally, large‐bore chest tubes (>24 F) had been advocated to drain viscid pus, but recent evidence suggests that small‐bore catheters (1014 F) with instillation of thrombolytics may be as effective and performed with less discomfort.[45] Video‐assisted thoracoscopy to lyse septations and evacuate infected materials is indicated when chest tube drainage has failed.

Figure 5

The most common complication of pleural drainage is pneumothorax. A meta‐analysis demonstrated a reduction in post‐thoracentesis pneumothorax rates from 9% to 4% with use of ultrasound.[46] Transporting patients to radiology for ultrasound marking has not been shown to decrease pneumothorax rates compared to thoracentesis without ultrasound guidance, likely due to changes in patient position and prolonged delays between marking and drainage.[47] Postprocedure pneumothorax can be ruled out if lung sliding is visualized. A meta‐analysis demonstrated superior sensitivity and similar specificity of pleural ultrasonography versus chest radiography to detect pneumothorax (sensitivity 91% vs 50% and specificity 98% vs 99%, respectively).[48] Real‐time ultrasound guidance for thoracentesis, or use of ultrasound to track the needle tip, has not been well studied but may be performed by experienced proceduralists to drain small effusions.

FUTURE RESEARCH

Future research should focus on the clinical effectiveness of point‐of‐care pleural ultrasonography when integrated with other diagnostic tools, and application of new ultrasound technologies to evaluate pleural diseases. Routine use of point‐of‐care ultrasound as the primary imaging modality in a medical ICU demonstrated a highly statistically significant reduction in chest x‐rays (3.75 vs 0.82, P < 0.0001) and chest CT scans (0.10 vs 0.04, P = 0.0007).[49] Similar studies have yet to be performed with the use of ultrasound specifically in the management of pleural diseases. Thus, clinical effectiveness studies are needed to assess the impact of routine use of pleural ultrasound on the initiation of appropriate therapies, length of stay, and costs in the management of pleural disease.

Point‐of‐care pleural ultrasound findings need to be evaluated in the context of other clinical findings and diagnostic tests. Certain ultrasound findings have been associated with exudative pleural effusions, but whether exudative and transudative effusions can be differentiated noninvasively using ultrasound findings alone, or in combination with other clinical data, warrants investigation. Similar to severity of illness scores, models that incorporate clinical, laboratory, and ultrasound findings need to be developed to guide treatment decisions, such as type of drainage procedure, as well as prognostication.

Finally, new technologies may advance the utility of point‐of‐care pleural ultrasonography. Even though 3‐dimensional ultrasonography has been available for over 2 decades, comparative studies of conventional 2‐dimensional versus 3‐dimensional ultrasonography to characterize pleural effusions have yet to be performed. Furthermore, computer‐aided detection has been shown to facilitate interpretation of ultrasound images, but this technology has yet to be applied to pleural ultrasonography.

CONCLUSIONS

Point‐of‐care pleural ultrasound is a powerful bedside tool in the hospitalist's armamentarium that is superior to physical examination and chest radiographs in the detection and characterization of pleural effusions. Furthermore, ultrasound performs similarly when compared to CT scans but offers the advantages of decreased cost, avoidance of ionizing radiation, and availability at the bedside. Ultrasound guidance reduces complications and increases the success rate of pleural drainage procedures, leading to improved patient safety. As clinical effectiveness studies emerge revealing its true value, point‐of‐care pleural ultrasonography is likely to become the standard diagnostic tool to evaluate and manage patients with pleural effusions.

Disclosures: Dr. Restrepo is partially supported by award number K23HL096054 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health. The authors report no conflicts of interest.

Hospitalists commonly encounter pleural effusions, and their detection and characterization by point‐of‐care ultrasound can guide management. Approximately 44% to 57% of hospitalized patients with bacterial pneumonia,[1, 2] and up to 62% of intensive care unit (ICU) patients[3] have a pleural effusion. For patients with a parapneumonic effusion, hospitalists can use ultrasound to quantify and characterize pleural fluid to determine whether diagnostic or therapeutic drainage is indicated, as well as guide performance of thoracentesis. For patients with lung cancer, detection of a malignant pleural effusion changes staging to stage IV, regardless of tumor size or lymph node involvement, and hospitalists may discuss more appropriate treatment options with patients and consultants.

Routine use of pleural ultrasonography may help hospitalists provide high‐value care by reducing ancillary testing, including computerized tomography (CT) scans that expose patients to ionizing radiation, and reducing complications of thoracentesis. However, many hospitalists may not be familiar with the use of point‐of‐care ultrasound. A national survey in 2012 revealed only 25% of internal medicine residencies have formal curricula to teach point‐of‐care ultrasonography.[4] The purpose of this review is to provide an overview of how point‐of‐care ultrasound can be utilized by hospitalists in the care of patients with pleural effusions. We review the literature on the diagnosis and evaluation of pleural effusions with ultrasound, as well as techniques to examine and drain the pleural space.

DIAGNOSIS OF PLEURAL EFFUSION

PLEURAL ULTRASOUND EXAMINATION

A pleural ultrasound exam may be performed as part of a complete lung ultrasound exam, such as the BLUE (Bedside Lung Ultrasound in Emergency) protocol,[27] or a focused exam to evaluate a suspected or known pleural effusion seen on chest radiograph or CT scan.[27] Free‐flowing pleural effusions accumulate in the most dependent portions of the thorax, most commonly, the posterolateral costophrenic recesses in semirecumbent or seated patients, but anteriorly in mechanically ventilated patients in a prone position.

A low‐frequency (25 MHz) phased‐array transducer is generally preferred for imaging in between the ribs. High‐frequency linear transducers do not provide adequate penetration to visualize deep structures, but do provide superior visualization of the pleural line to assess pleural thickness, measure pleural depth, and evaluate for pneumothorax.

Pleural effusions are best evaluated starting at the level of the diaphragm. Place the transducer in a longitudinal plane on the posterior axillary line at the level of the diaphragm with the transducer orientation marker (notch) pointed cephalad (Figure 1). Five structures must be definitively identified to diagnose a pleural effusion: liver/spleen, diaphragm, pleural fluid, lung, and chest wall (Figure 2A). Large pleural effusions compress the adjacent lung causing atelectasis, which gives the lung a tissue‐like echogenicity similar to the liver (Figure 2B). Static air bronchograms are commonly seen in atelectatic lung bases with pleural effusions.[28]

Figure 1

Figure 2

Color flow Doppler and M‐mode ultrasound may be utilized as adjuncts to routine 2‐dimensional ultrasonography. Free‐flowing pleural effusions will demonstrate flow by color Doppler (Figure 3A). Using M‐mode ultrasound, the lung can been seen moving within a pleural effusion to and from the chest wall (sinusoid sign).[29] Absence of flow or movement is seen with dense pleural loculations, pleural thickening, and peripheral lung or pleural masses (Figure 3B).

Figure 3

CHARACTERIZATION OF PLEURAL EFFUSION

Simple Versus Complex Effusions

Based on its sonographic appearance, pleural effusions are categorized as simple or complex. Simple pleural effusions are anechoic and usually transudative. Complex pleural effusions are subcategorized as homogeneously or heterogeneously echogenic, with or without septations, and are more often exudative.[33]

Effusions with heterogeneous echogenicity with swirling echoes suggest high cellular content that is often associated with malignancy.[34] Fibrinous stranding, septations, and loculations also suggest an exudative effusion (Figure 4A), and are more readily identified and characterized on lung ultrasound than CT scan.[35]

Figure 4

Homogenously echogenic effusions are most often due to hemothorax or empyema.[36] The high cell count of a hemothorax creates a layering effect in costophrenic recesses (hematocrit sign). Empyemas develop from complex effusions that organize into collections of pus and usually have a homogeneously echogenic, speckled appearance (Figure 4B). Sonographic evidence of septations in the presence of empyema predicts the need for intrapleural fibrinolytic therapy, longer duration of drainage, and possible surgical intervention.[37]

Isolated dense loculations may be challenging to differentiate from peripheral lung or pleural lesions, such as abscess or tumor.

PLEURAL FLUID DRAINAGE

Since its first description in 1967, use of ultrasound guidance for thoracentesis has evolved to become the standard of care in many hospitals in the United States.[41] The British Thoracic Society guidelines recommend that all thoracenteses be performed with ultrasound guidance.[42] The American College of Graduate Medical Education now requires proficiency in the use of ultrasound for thoracentesis and pleural catheter insertion by pulmonary and critical care fellows.[43]

The impetus for these recommendations stems from increased procedural success and decreased complications associated with ultrasound‐guided drainage of pleural effusions. A study evaluating thoracentesis site selection based on physical exam and chest radiographs demonstrated inaccurate site selection in 15% of patients, and use of ultrasound for site selection prevented possible accidental organ puncture in 10% of all cases.[44] The success rate of thoracentesis for small pleural effusions has been shown to increase from 66% to 90% with ultrasound guidance.[42] Using ultrasound, the distance from the skin to parietal and visceral pleura can be measured to determine whether thoracentesis can be safely performed, and to guide selection of an adequate length needle (Figure 5). A minimum pleural effusion depth of 1.5 cm between the visceral and parietal pleura has been recommended to perform diagnostic thoracentesis.[28] Diagnostic thoracentesis of complex septated pleural effusions or empyemas may be performed with a straight needle, but therapeutic drainage often requires temporary insertion of a catheter. Traditionally, large‐bore chest tubes (>24 F) had been advocated to drain viscid pus, but recent evidence suggests that small‐bore catheters (1014 F) with instillation of thrombolytics may be as effective and performed with less discomfort.[45] Video‐assisted thoracoscopy to lyse septations and evacuate infected materials is indicated when chest tube drainage has failed.

Figure 5

The most common complication of pleural drainage is pneumothorax. A meta‐analysis demonstrated a reduction in post‐thoracentesis pneumothorax rates from 9% to 4% with use of ultrasound.[46] Transporting patients to radiology for ultrasound marking has not been shown to decrease pneumothorax rates compared to thoracentesis without ultrasound guidance, likely due to changes in patient position and prolonged delays between marking and drainage.[47] Postprocedure pneumothorax can be ruled out if lung sliding is visualized. A meta‐analysis demonstrated superior sensitivity and similar specificity of pleural ultrasonography versus chest radiography to detect pneumothorax (sensitivity 91% vs 50% and specificity 98% vs 99%, respectively).[48] Real‐time ultrasound guidance for thoracentesis, or use of ultrasound to track the needle tip, has not been well studied but may be performed by experienced proceduralists to drain small effusions.

FUTURE RESEARCH

Future research should focus on the clinical effectiveness of point‐of‐care pleural ultrasonography when integrated with other diagnostic tools, and application of new ultrasound technologies to evaluate pleural diseases. Routine use of point‐of‐care ultrasound as the primary imaging modality in a medical ICU demonstrated a highly statistically significant reduction in chest x‐rays (3.75 vs 0.82, P < 0.0001) and chest CT scans (0.10 vs 0.04, P = 0.0007).[49] Similar studies have yet to be performed with the use of ultrasound specifically in the management of pleural diseases. Thus, clinical effectiveness studies are needed to assess the impact of routine use of pleural ultrasound on the initiation of appropriate therapies, length of stay, and costs in the management of pleural disease.

Point‐of‐care pleural ultrasound findings need to be evaluated in the context of other clinical findings and diagnostic tests. Certain ultrasound findings have been associated with exudative pleural effusions, but whether exudative and transudative effusions can be differentiated noninvasively using ultrasound findings alone, or in combination with other clinical data, warrants investigation. Similar to severity of illness scores, models that incorporate clinical, laboratory, and ultrasound findings need to be developed to guide treatment decisions, such as type of drainage procedure, as well as prognostication.

Finally, new technologies may advance the utility of point‐of‐care pleural ultrasonography. Even though 3‐dimensional ultrasonography has been available for over 2 decades, comparative studies of conventional 2‐dimensional versus 3‐dimensional ultrasonography to characterize pleural effusions have yet to be performed. Furthermore, computer‐aided detection has been shown to facilitate interpretation of ultrasound images, but this technology has yet to be applied to pleural ultrasonography.

CONCLUSIONS

Point‐of‐care pleural ultrasound is a powerful bedside tool in the hospitalist's armamentarium that is superior to physical examination and chest radiographs in the detection and characterization of pleural effusions. Furthermore, ultrasound performs similarly when compared to CT scans but offers the advantages of decreased cost, avoidance of ionizing radiation, and availability at the bedside. Ultrasound guidance reduces complications and increases the success rate of pleural drainage procedures, leading to improved patient safety. As clinical effectiveness studies emerge revealing its true value, point‐of‐care pleural ultrasonography is likely to become the standard diagnostic tool to evaluate and manage patients with pleural effusions.

Disclosures: Dr. Restrepo is partially supported by award number K23HL096054 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health. The authors report no conflicts of interest.

Sepsis is one of the oldest and most elusive syndromes in medicine, and remains a significant contributor to morbidity, mortality, and healthcare expenditure.[1] A 1992 American College of Chest Physicians and Society of Critical Care Medicine consensus conference statement introduced the systemic inflammatory response syndrome (SIRS) into the medical lexicon, along with definitions of sepsis, severe sepsis, and septic shock.[2] A 2003 consensus panel expanded the list of signs and symptoms associated with sepsis, and warned that the findings of SIRS do not differentiate sepsis from other noninfectious conditions.[3] The terminology is important, as these definitions resulted in a shift of the label of the syndrome of infection complicated by end‐organ dysfunction from sepsis to severe sepsis or septic shock. Overlap of these terms has implications for categorizing such infections for the purpose of investigation, estimating epidemiology and outcome, and coding, billing, and reimbursement.[1]

Traditional definitions of the spectrum of sepsis disorders are outlined in Table 1,[2, 3] and it is important to note that an update to these definitions is anticipated in the near future. A recent publication has called into question the sensitivity and categorical requirement of at least 2 SIRS criteria to define severe sepsis.[4] This study of more than 1 million patients from 172 intensive care units (ICUs) in Australia and New Zealand from 2000 to 2013 found that the cutoff of 2 SIRS criteria to define severe sepsis excluded 1 in 8 patients with infection and end‐organ hypoperfusion. SIRS‐negative severe sepsis patients experienced the same mortality as SIRS‐positive patients. In addition, adjusted analysis determined a stepwise increase in mortality risk associated with each additional SIRS criterion without a transition point in risk noted at 2.[4]

From 1979 through 2000, there were over 10 million reported cases of sepsis, which accounted for 1.3% of all hospitalizations in the United States.[5] Normalized to the population distribution of the 2000 US Census, there was an annualized increase in sepsis cases of 8.7%. A 2011 report revealed rates of hospitalization for patients with septicemia or sepsis in the United States more than doubled from 2000 through 2008.[6] Patients with sepsis experienced longer length of stay than other inpatients and were 8 times more likely to die during hospitalization.[6] Estimates of severe sepsis incidence are complicated by how acute organ dysfunction is defined and whether it is related to infection. As of 2001, the number of severe sepsis cases in the United States was believed to exceed 750,000 and comprise approximately 10% of ICU admissions.[1] The incidence of severe sepsis cases in the United States continues to rise.[7, 8, 9] However, a more than doubling of the use of sepsis International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes from 2004 through 2009 has also been noted.[8] Based on ICD‐9‐CM codes indicating the presence of sepsis and organ system failure, the number of severe sepsis hospitalizations per 100,000 persons in the United States increased from 143 in 2000 to 343 in 2007.[7] Total hospital costs for patients with severe sepsis were estimated to increase 57%, from $15.4 billion in 2003 to $24.3 billion in 2007.[9] The Agency for Healthcare Research and Quality considered septicemia the most expensive medical condition in the United States in a 2011 data brief, with annual aggregate hospital costs exceeding $20 billion.[10]

Although many hospitalists care for patients in the ICU and other higher acuity or step‐down units, a significant proportion of patients with severe sepsis receive care on a general medical floor.[11, 12, 13] Sepsis is also clearly not an issue restricted to patients on internal medicine services. Of over 360,000 general surgery patients from 2005 to 2007, the incidences of sepsis (2.3%) and septic shock (1.6%) greatly exceeded those of pulmonary embolism (0.3%) and myocardial infarction (0.2%). In this cohort, the need for emergency surgery and the presence of any comorbidity increased the number of sepsis cases.[14]

Despite difficulties obtaining exact estimates of case numbers, the following appears true: the spectrum of sepsis disorders (including severe sepsis and septic shock) remains a common, costly, and increasing clinical entity that is encountered by hospital medicine physicians in a variety inpatient settings. This review will provide an update for hospitalists based on many important studies that have been published since the last review of this topic in this journal.[15] The expanding evidence base in sepsis includes early goal‐directed therapy (EGDT), clinical endpoints, and bundles of care for sepsis; antibiotics (choice and timing); volume resuscitation; ICU considerations, including the use of insulin and corticosteroids; and mortality, complications, and the advent of the condition of sepsis survivorship.

EARLY GOAL‐DIRECTED THERAPY

A 2001 prospective, randomized trial of EGDT initiated in the emergency department (ED) for patients with severe sepsis and septic shock resulted in an impressive 16% reduction of in‐hospital mortality compared to standard therapy.[16] The intervention protocol included central venous catheter placement and a 500‐mL bolus of crystalloid every 30 minutes to establish a central venous pressure (CVP) of 8 to 12 mm Hg. Vasopressors were used to maintain a mean arterial pressure (MAP) greater than 65 mm Hg, and patients with a MAP greater than 90 mm Hg were given vasodilators. Patients with a central venous oxygen saturation (Scv0₂) of less than 70% received red blood cell transfusion with a goal hematocrit of 30%. If central venous oxygen saturation remained less than 70% despite these interventions, dobutamine was used for inotropic effect until this goal was achieved or was limited by tachycardia or hypotension.[16]

These results prompted inclusion of the specific hemodynamic targets (CVP, MAP, and Scv0₂) into the original 2004 Surviving Sepsis Campaign guidelines and spurred a decade of interest worldwide.[17] The incremental importance of these individual components in managing severe sepsis and septic shock has since come under scrutiny. A recent randomized trial suggested that EGDT guided by venous lactate clearance of >10% was noninferior to the goal Scv0₂ of >70%. However, only 10% of the study population required transfusion or dobutamine.[18, 19] Prospective ICU data on lactate‐guided therapy[20] supported the revised 2012 Surviving Sepsis Campaign (SSC) guidelines to recommend lactate normalization as part of initial resuscitation efforts, particularly when Scv0₂ is not available.[21] Lactate measurement may assist in recognition of cases of severe sepsis or septic shock and provide valuable triage information, as serum lactate has been shown to predict mortality from severe sepsis independent of shock or organ failure.[22] In a retrospective study of patients presenting to the ED with sepsis, a lactate >4 mmol/L was associated with progression to septic shock within 4 to 48 hours.[23]

Our understanding of the specific benefits of EGDT is far from complete, as 3 recent large prospective, multicenter, randomized trials ProCESS (Protocolized Care for Early Septic Shock), ARISE (Australasian Resuscitation In Sepsis Evaluation), and ProMISe (Protocolised Management in Sepsis) did not show EGDT protocols to be superior to usual care.[24, 25, 26] Interpreted collectively, the benefit of EGDT may not be from targeting specific physiologic parameters, but rather from the early recognition of sepsis and the appropriate use of well‐supported interventions like aggressive fluid resuscitation and early/efficacious antibiotics.[27]

Although the precise benefit of EGDT as a package versus its individual components remains in question, we have a decade of experience in delivering this care as an integral component of the bundles put forth in the SSC guidelines.[28] Observational and retrospective studies have shown increased compliance with guidelines and improved mortality after implementing these protocols, although early bundles for severe sepsis included therapies that have subsequently been called into question on an individual basis like drotrecogin alfa (activated) and glucocorticoid therapy.[29, 30, 31] The mortality benefit from sepsis bundles deserves further explanation, although education and early recognition are likely contributory.[32]

Several studies evaluated individual components of EGDT. The TRISS (Transfusion Requirements in Septic Shock) trial randomized ICU patients with septic shock to 2 different red blood cell transfusion strategies, and found no mortality benefit or reduction in ischemic events for patients transfused at a hemoglobin of 9 g/dL compared to the 7 g/dL threshold.[33] The SEPSISPAM (Assessment of Two Levels of Arterial Pressure on Survival in Patients With Septic Shock) trial compared the MAP goal of 65 to 70 mm Hg to 80 to 85 mm Hg for patients with septic shock in a randomized, multicenter trial.[34] Although there was no difference in 28‐day mortality, more atrial fibrillation was diagnosed in the higher target group. For patients with chronic hypertension, targeting the higher MAP led to less renal injury and reduced the need for renal‐replacement therapy.[34, 35] Identifying specific subsets of patients with sepsis who benefit most from particular therapies should help clinicians set patient‐specific goals and targets.

Although we can expect additional studies to provide further guidance, it is reasonable at present to adhere to protocols designed to improve timely sepsis detection and management with aggressive volume resuscitation, early/efficacious antibiotic administration, and effective infection source control.

ANTIBIOTICS AND SOURCE CONTROL

Administration of broad‐spectrum antibiotics has long been the cornerstone of sepsis management. Timely antibiotic infusion is an integral part of the 2004 and 2012 SSC guidelines,[17, 21] with the caveat that blood cultures should be obtained prior to antibiotic therapy provided that no significant delay (>45 minutes) occurs.[21] Recent studies have begun to address fundamental clinical questions, including the timing of antibiotic administration and the efficacy of empiric antibiotic choice. A landmark retrospective cohort study of ICU patients with septic shock demonstrated survival to hospital discharge was highest in patients who received antibiotics within the first hour of hypotension.[36] Survival decreased on average by 7.6% with each hour that antibiotics were delayed. Only 50% of patients with septic shock in this study received effective antibiotic therapy within 6 hours of documented hypotension.[36] A subsequent retrospective, single‐center cohort study of ED patients with severe sepsis or septic shock undergoing EGDT showed a mortality benefit when antibiotics were administered within the first hour. However, it did not demonstrate a statistically significant decline in survival on an hourly basis thereafter.[37]

A prospective, multicenter ED trial that included patients with severe sepsis in addition to septic shock[38] did not show a mortality benefit to administration of antibiotics within the first hour. In‐hospital mortality risk for patients undergoing EGDT was similar across patients in whom time to antibiotics was delayed up to 6 hours after triage.[36, 38] However, patients with severe sepsis in whom antibiotics were delayed until shock was recognized faced a statistically significant increased risk of death (odds ratio = 2.35; 95% confidence interval = 1.12‐4.53).[38, 39] A retrospective study of 28,150 patients from the SSC database demonstrated a statistically significant increase in mortality for each hour that empiric antibiotics were delayed.[40] Importantly, this trend was preserved regardless of location of sepsis diagnosis (ED, ICU, and hospital ward) and across illness severity. Though there remains debate about the critical importance of the golden hour for antibiotic administration, overall current evidence supports early empiric antibiotics in severe sepsis and septic shock.

Choosing an empiric antibiotic regimen, based on infection source and host factors, also plays a key role in sepsis outcomes. A retrospective study of patients with septic shock from 1996 to 2005 showed that inappropriate initial antibiotics (based on eventual in vitro culture sensitivities or evaluation of clinical syndrome) were used 20% of the time and resulted in a 5‐fold reduction in survival.[41] A retrospective cohort study of patients with gram‐negative bacteremia and severe sepsis or septic shock found prior antibiotic exposure within 90 days to be an independent risk factor for drug resistance and in‐hospital mortality.[42] However, careful consideration of side effects should also influence choice of initial antibiotic therapy. A Cochrane review citing 69 trials and containing 7863 subjects with sepsis compared empiric ‐lactam therapy to ‐lactamaminoglycoside combination therapy.[43] All‐cause mortality and clinical failure was similar in both groups, as was the rate of resistance. Importantly, nephrotoxicity was significantly less in the ‐lactam monotherapy group.[43]

Infection source control is an essential component of sepsis management that should occur simultaneously with antibiotic administration. The 2012 SSC guidelines promote infection source control within 12 hours of diagnosis, with consideration of the risks and benefits therein and preference for interventions with the lowest associated physiologic insult.[21] Intravascular access devices should be recognized as a common source of infection, and should be removed after alternative access has been established.[21]

FLUID RESUSCITATION

Volume resuscitation is an essential component of sepsis management, regardless of algorithm or endpoint. Three main types of nonblood product fluid resuscitation have been used: crystalloid (saline and Ringer's solutions), colloid (typically an albumin‐containing solution), and synthetic volume expanders (hydroxyethyl starch [HES] and similar compounds).

Multiple large studies confirmed the lack of a favorable riskbenefit ratio with synthetic volume expanders. Among nearly 800 patients with severe sepsis who were randomized to receive either Ringer's acetate or HES 130/0.4, a significantly higher number of patients receiving HES died (51% vs 43%, relative risk [RR] = 1.17), and required renal‐replacement therapy (22% vs 16%, RR = 1.35). One patient in each group was dialysis dependent at 90 days.[44] An additional multicenter, prospective study of HES versus 0.9% (normal) saline for fluid resuscitation in the ICU found no significant difference in mortality (18% vs 17%, P = 0.26), but did note a higher need for renal‐replacement therapy in the HES group (7.0% vs 5.8%, RR = 1.21).[45] A systematic review incorporating 9 trials that randomized approximately 3400 patients with sepsis receiving either HES, crystalloid, or colloid showed no difference in mortality, although there was an excess risk for renal‐replacement therapy (RR = 1.36), serious adverse events (RR = 1.30), and red blood cell transfusion (RR = 1.29) in patients receiving HES.[46] A second, larger systematic review concluded that HES was associated with an increased mortality compared with crystalloids, albumin, or gelatin (RR = 1.09). Additionally, an increase in renal failure (RR = 1.27) and renal‐replacement therapy (RR = 1.32) was also noted.[47]

The debate between crystalloid and colloid (namely albumin) for fluid resuscitation is ongoing, with recent important additions to the literature. The SAFE (Saline Versus Albumin Fluid Evaluation) study investigators in 2004 randomized nearly 7000 patients to receive either 4% albumin solution or normal saline. At 28 days, no significant differences were found in mortality, new organ failure, ICU and hospital length of stay, days of mechanical ventilation, or days of renal‐replacement therapy.[48] In 2014, another multicenter prospective study of 1800 patients with severe sepsis or septic shock in 100 ICUs in Italy compared 20% albumin and crystalloid solution to crystalloid solution alone. Mortality, end‐organ dysfunction, and ICU length of stay did not differ between groups.[49] Two 2014 systematic reviews and meta‐analyses on fluid resuscitation produced somewhat differing conclusions. Patel et al. evaluated data from 16 randomized trials including more than 4000 patients receiving albumin for volume resuscitation in adults with sepsis. Albumin provided no significant survival advantage in total or in any subgroup, regardless of severity of illness or baseline albumin level, thus arguing against its routine use.[50] Rochwerg et al. evaluated 14 studies with approximately 19,000 patients using Bayesian network meta‐analysis technique. This study concluded that albumin is associated with reduced mortality compared with other fluids, and also that balanced crystalloids (eg, Ringer's lactate and similar) may have lower mortality than normal saline.[51] A chloride‐restrictive resuscitation approach has also been associated with a lower incidence of acute kidney injury in critically ill adults.[52]

The SSC confirmed its recommendations of a minimum of 30 mL/kg of crystalloids as the initial fluid of choice in sepsis in 2012, but added a suggestion for the addition of albumin in patients requiring substantial amounts of crystalloid.[21] The currently available data suggest crystalloid fluids to be the best‐supported initial fluid in the management of sepsis, and that synthetic colloids should be avoided. Prospective data are still required to answer questions regarding the potential advantages of albumin or balanced and/or chloride‐restricted crystalloids.

ICU CONSIDERATIONS

Appropriate management of patients with the syndrome of sepsis, severe sepsis, and septic shock on the hospital ward requires a working knowledge of recent research conducted in the ICU setting. Although conclusions based on data from patients with septic shock might not be generalizable to less severe cases of sepsis, recent trials on glucose control and corticosteroids deserve consideration.

Intensive insulin treatment in the medical ICU is no longer standard practice. In the NICE‐SUGAR (Normoglycaemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation) trial, a large, multicenter randomized controlled trial of ICU patients, close to 20% of patients had severe sepsis at the time of randomization.[53] This subgroup did not benefit from intensive glucose control (a target of 81108 mg/dL) in terms of 90‐day mortality.[53] In contrast to prior studies of glycemic control in the critically ill, the intensive treatment group overall suffered increased mortality.[54] The COIITSS (Combination of Corticotherapy and Intensive Insulin Therapy for Septic Shock) trial looked at intensive insulin therapy in patients with septic shock being treated with corticosteroids, a group particularly at risk for hyperglycemia. In this study, intensive insulin therapy did not improve in‐hospital mortality.[55] Based on these and other ICU data, the current SSC recommendation is to target a blood glucose of 180 mg/dL for patients with severe sepsis.[21, 56]

The use of corticosteroids to treat the host response in septic shock has been re‐evaluated.[57] The 2004 SSC guidelines recommended hydrocortisone therapy for 7 days in patients with septic shock requiring vasopressor support after fluid resuscitation.[17] This recommendation was based on data from a placebo‐controlled multicenter trial in France that showed improved shock reversal and reduced mortality in patients with septic shock who were treated with hydrocortisone and fludricortisone.[58] Of note, these patients were enrolled on the basis of hypotension despite intravenous fluids and the initiation of 1 vasopressor. The benefit of corticosteroids was seen only in patients deemed to have relative adrenal insufficiency based on response corticotropin testing.[58] However, the CORTICUS (Corticosteroid Therapy of Septic Shock) study, a subsequent multicenter, placebo‐controlled, randomized controlled trial failed to show a benefit to corticotropin testing in identifying patients with septic shock who would benefit from corticorsteroids.[59] The corticosteroid treatment arm similarly benefited from faster shock reversal, but at the expense of increased superinfection.[59] Although underpowered, CORTICUS did not show a survival benefit to corticosteroids in septic shock.[59] The most recent SSC guidelines do not recommend corticotropin (adrenocorticotropic hormone) stimulation testing and do not advise corticosteroids in septic shock if fluid resuscitation and vasopressor therapy restore hemodynamic stability during initial resuscitation.[21] Future studies may clarify subpopulations of patients with sepsis who benefit from corticosteroids.

OUTCOMES: MORTALITY AND COMPLICATIONS

An understanding of the currently available information regarding the morbidity and mortality associated with severe sepsis is essential for the practicing hospitalist. Whether transferring care of patients to or receiving patients from the ICU, hospitalists must lead discussions with patients and families regarding prognosis, especially as it informs disposition. Hospitalists are often asked to make projections on outcome as well as the timing and venue of disposition. Clarification of patient wishes and goals of care remains an essential first step in the care of septic patients. Recently published studies provide prognostic information, including mortality (both short and long term) as well as complications associated with severe sepsis.

The attributable mortality for severe sepsis has been predominantly reported to date as short‐term (usually in‐hospital). A meta‐analysis of US patients from 1991 to 2009 demonstrated a 3% annual decline in the short‐term (28 day) mortality from severe sepsis using 2 previously validated algorithms. Data from 36 trials (and approximately 14,000 patients) revealed a decrease in mortality from 47% in the period from 1991 to 1995 to 29% from 2006 to 2009.[60] Although the methods employed (sepsis definitions and ICD‐9‐CM codes) can have a significant impact on estimates of mortality, these results corroborate a progressive decline in short‐term mortality from severe sepsis in the United States between 2004 and 2009 using 4 validated algorithms.[8] Outside the United States, a recent retrospective analysis of more than 1 million patients with severe sepsis treated in the ICU in Australia and New Zealand from 2000 to 2012 also demonstrated a decrease in adjusted in‐hospital mortality. In this study, short‐term mortality declined yearly, with an odds ratio of death of 0.49 in 2012 compared with 2000.[61] Hospital case volume has also been shown to impact rates of inpatient death, with higher‐volume centers demonstrating lower mortality attributed to severe sepsis.[62, 63]

The sufficiency of short‐term mortality as the sole metric for severe sepsis outcome has been more recently questioned.[64, 65] The extent to which full recovery and significant morbidity are affected relative to the change in death rate is unknown, and as such, more data on morbidity and longer‐term mortality are necessary. A Danish study examined data from several registries of patients with severe sepsis. Compared with community‐matched controls, patients with severe sepsis had an increased risk of death at 30 days (hazard ratio [HR] = 90.8), from 30 days to 1 year (HR = 2.7), and 1 to 4 years (HR = 2.3) after discharge.[66] Older survivors of severe sepsis also appear to have higher healthcare utilization in the year following discharge. An analysis of older severe sepsis survivors showed a striking increase in healthcare use relative to their prior resource use, driven primarily by higher number of days in inpatient healthcare facilities. Survivors of severe sepsis also had a significantly higher 90‐day and 1‐year mortality than matched controls.[67]

Increased attention is currently being given to sepsis‐related complications, especially functional and cognitive impairments in older patients. Sepsis survivorship is a swiftly mounting public health issue for older Americans.[68] An 8‐year follow‐up of older sepsis survivors demonstrated a significant increase in the odds of both physical and cognitive dysfunction. During this period, moderate‐to‐severe cognitive dysfunction increased 3‐fold (6.1% before sepsis, 16.7% after).[69] The mechanism by which this dysfunction occurs is unknown, as are the relative contributions of infection site/etiology, ICU length of stay, and extent of organ dysfunction. New functional impairment has been demonstrated in patients with severe sepsis initially admitted to a general floor, even with good baseline function,[12] as well as decreased quality of life in sepsis survivors.[65, 70] Another study showed more admissions complicated by severe sepsis resulted in discharge to a long‐term care facility in 2007 compared to 2000.[7]

Additional organ‐specific consequences of severe sepsis have also been recently suggested. A retrospective analysis showed an increase in the incidence of new‐onset atrial fibrillation in severe sepsis, with an associated increase in risk of in‐hospital stroke and death. New‐onset atrial fibrillation was present in 5.9% of patients with severe sepsis, compared with 0.65% in patients without. Severe sepsis patients with new‐onset atrial fibrillation had an increased risk of in‐hospital stroke (adjusted odds ratio = 2.70) and mortality (adjusted RR = 1.07).[71] These findings suggest association only, and further investigation is warranted. It remains to be seen whether interventions to restore sinus rhythm or anticoagulation are warranted. Preoperative sepsis (within 48 hours) has also been proposed as a risk for postoperative (30 day) arterial (myocardial infarction, stroke) and venous (deep venous thrombosis, pulmonary embolism) thromboembolism. The authors of this study suggest deferral of elective surgery or specific attention to postoperative thromboprophylaxis in patients in whom procedures must occur.[72] This has particular relevance for those septic patients in whom surgical source control is indicated.

Estimates regarding mortality and specific complications attributable to severe sepsis are ongoing, though clearly with a new focus upon metrics other than short‐term mortality. Furthermore, recent data to suggest source of infection as a major driver of mortality in septic shock[73] may contribute to the evolution of the conceptualization of sepsis similar to that of cancer: a heterogeneous collection of disease, among which mortality is determined by specific subtypes. At present, this much appears clear: the previously held notion that survival of a septic insult is unlikely to have future implications is under siege.[74] The extent to which complications and increased longer‐term mortality may reflect generally poorer health at the time of infection versus a sequelae of the survived episode itself is not yet known.

CONCLUSIONS

The past decade of sepsis research has led to significant findings that will change clinical practice for hospital medicine practitioners. Although the incidence of severe sepsis in the United States has continued to rise, in‐hospital mortality has declined; in this context, the management of the spectrum of sepsis disorders is no longer restricted to the ICU, and the entity of sepsis survivorship has blossomed. Prompt recognition of sepsis and improvements in supportive care are likely responsible for improved patient outcomes. EGDT has been called into question as a protocol whose benefit lies not in specific targets or endpoints, but rather in the early recognition of sepsis, appropriate fluid resuscitation, and early/effective antibiotics. Synthetic volume expanders, intensive insulin therapy, and routine use of corticosteroids are no longer recommended.

Hospitalists are a critical link in providing timely, evidence‐based care for patients with sepsis from initial recognition to post‐ICU recovery. Specialized care for the survivors of septic shock is a burgeoning area, and hospitalists are integral in the management of the sequelae of multiorgan failure.

Disclosure: Nothing to report.