Lack of sleep is a common problem in hospitalized patients and is associated with poorer health outcomes, especially in older patients.[1, 2, 3] Prior studies highlight a multitude of factors that can result in sleep loss in the hospital[3, 4, 5, 6] with 1 of the most common causes of sleep disruption in the hospital being noise.[7, 8, 9]

In addition to external factors, such as hospital noise, there may be inherent characteristics that predispose certain patients to greater sleep loss when hospitalized. One such measure is the construct of perceived control or the psychological measure of how much individuals expect themselves to be capable of bringing about desired outcomes.[10] Among older patients, low perceived control is associated with increased rates of physician visits, hospitalizations, and death.[11, 12] In contrast, patients who feel more in control of their environment may experience positive health benefits.[13]

Yet, when patients are placed in a hospital setting, they experience a significant reduction in control over their environment along with an increase in dependency on medical staff and therapies.[14, 15] For example, hospitalized patients are restricted in their personal decisions, such as what clothes they can wear and what they can eat and are not in charge of their own schedules, including their sleep time.

Although prior studies suggest that perceived control over sleep is related to actual sleep among community‐dwelling adults,[16, 17] no study has examined this relationship in hospitalized adults. Therefore, the aim of our study was to examine the possible association between perceived control, noise levels, and sleep in hospitalized middle‐aged and older patients.

METHODS

RESULTS

From April 2010 to May 2012, 1134 patients were screened by study personnel for this study via an ongoing study of hospitalized patients on the inpatient general medicine ward. Of the 361 (31.8%) eligible patients, 206 (57.1%) consented to participate. Of the subjects enrolled in the study, 118 were able to complete at least 1 night of actigraphy, sound monitoring, and subjective assessment for a total of 185 patient nights (Figure 1).

Figure 1

The majority of patients were female (57%), African American (67%), and non‐Hispanic (97%). The mean age was 65 years (standard deviation [SD], 11.6 years), and the median length of stay was 4 days (interquartile range [IQR], 36). The majority of patients also had hypertension (67%), with chronic obstructive pulmonary disease [COPD] (31%) and congestive heart failure (31%) being the next most common comorbidities. About two‐thirds of subjects (64%) were characterized as average or above average sleepers with Epworth Sleepiness Scale scores 9[20] (Table 1). Only 5% of patients received pharmacological sleep aids.

The mean baseline SLOC score was 30.4 (SD, 6.7), with a median of 31 (IQR, 2735). The mean baseline SSE score was 32.1 (SD, 9.4), with a median of 34 (IQR, 2441). Fifty‐four patients were categorized as having high sleep self‐efficacy (high SSE), which we defined as scoring above the median of 34.

Average in‐hospital sleep was 5.5 hours (333 minutes; SD, 128 minutes) which was significantly shorter than the self‐reported sleep duration of 6.5 hours prior to admission (387 minutes, SD, 125 minutes; P=0.0001). The mean sleep efficiency was 73% (SD, 19%) with 55% of actigraphy nights below the normal range of 80% efficiency for adults.[19] Median KSQI was 3.5 (IQR, 2.254.75), with 41% of the patients with a KSQI 3, putting them in the insomniac range.[32] The median score on the noise disruptiveness question was 1 (IQR, 14) with 42% of reports coded as disruptive defined as a score >1 on the 5‐point scale. The median score on the roommate disruptiveness question was 1 (IQR, 11) with 77% of responses coded as not disruptive defined as a score of 1 on the 5‐point scale.

A 2‐sample t test with equal variances showed that those patients reporting high SSE were more likely to sleep longer in the hospital than those reporting low SSE (364 minutes 95% confidence interval [CI]: 340, 388 vs 309 minutes 95% CI: 283, 336; P=0.003) (Figure 2). Patients with high SSE were also more likely to have a normal sleep efficiency (above 80%) compared to those with low SSE (54% 95% CI: 43, 65 vs 38% 95% CI: 28,47; P=0.028). Last, there was a trend toward patients reporting higher SSE to also report less noise disruption compared to those patients with low SSE ([42%] 95% CI: 31, 53 vs [56%] 95% CI: 46, 65; P=0.063) (Figure 3).

Figure 2

Figure 3

Linear regression clustered by subject showed that high SSE was associated with longer sleep duration (55 minutes 95% CI: 14, 97; P=0.010). Furthermore, high SSE was significantly associated with longer sleep duration after controlling for both objective noise level and patient demographics in the model using stepwise forward regression (50 minutes 95% CI: 11, 90; P=0.014) (Table 2).

Logistic regression clustered by subject demonstrated that patients with high SSE had 2 times higher odds of having a KSQI score above 3 (95% CI: 1.12, 3.71; P=0.020). This association was still significant after controlling for noise and patient demographics (OR: 2.01; 95% CI: 1.06, 3.79; P=0.032). After controlling for noise levels and patient demographics, there was a statistically significant association between high SSE and lower odds of noise complaints (OR: 0.49; 95% CI: 0.25, 0.96; P=0.039) (Table 2). Although demographic characteristics were not associated with high SSE, those patients with high SSE had lower odds of being in the loudest tertile rooms (OR: 0.34; 95% CI: 0.15, 0.74; P=0.007).

In multivariate linear regression analyses, there were no significant relationships between SLOC scores and KSQI, reported noise disruptiveness, and markers of sleep (sleep duration or sleep efficiency).

DISCUSSION

This study is the first to examine the relationship between perceived control, noise levels, and objective measurements of sleep in a hospital setting. One measure of perceived control, namely SSE, was associated with objective sleep duration, subjective and objective sleep quality, noise levels in patient rooms, and perhaps also patient complaints of noise. These associations remained significant after controlling for objective noise levels and patient demographics, suggesting that SSE is independently related to sleep.

In contrast to SSE, SLOC was not found to be significantly associated with either subjective or objective measures of sleep quality. The lack of association may be due to the fact that the SLOC questionnaire does not translate as well to the inpatient setting as the SSE questionnaire. The SLOC questionnaire focuses on general beliefs about sleep whereas the SSE questionnaire focuses on personal beliefs about one's own ability sleep in the immediate future, which may make it more relevant in the inpatient setting (see Supporting Information, Appendix 1 and 2, in the online version of this article).

Given our findings, it is important to identify why patients with high SSE have better sleep and fewer noise complaints. One possibility is that sleep self‐efficacy is an inherited trait unique to each person that is also predictive of a patient's sleep patterns. However, is it also possible that those patients with high SSE feel more empowered to take control of their environment, allowing them to advocate for better sleep? This hypothesis is further strengthened by the finding that those patients with high SSE on study entry were less likely to be in the noisiest rooms. This raises the possibility that at least 1 of the mechanisms by which high SSE may be protective against sleep loss is through patients taking an active role in noise reduction, such as closing the door or advocating for their sleep with staff. However, we did not directly observe or ask patients whether doors of patient rooms were open or closed or whether the patients took other measures to advocate for their own sleep. Thus, further work is necessary to understand the mechanisms by which sleep self‐efficacy may influence sleep.

One potential avenue for future research is to explore possible interventions for boosting sleep self‐efficacy in the hospital. Although most interventions have focused on environmental noise and staff‐based education, empowering patients through boosting SSE may be a helpful adjunct to improving hospital sleep.[33, 34] Currently, the SSE scale is not commonly used in the inpatient setting. Motivational interviewing and patient coaching could be explored as potential tools for boosting SSE. Furthermore, even if SSE is not easily changed, measuring SSE in patients newly admitted to the hospital may be useful in identifying patients most susceptible to sleep disruptions. Efforts to identify patients with low SSE should go hand‐in‐hand with measures to reduce noise. Addressing both patient‐level and environmental factors simultaneously may be the best strategy for improving sleep in an inpatient hospital setting.

In contrast to our prior study, it is worth noting that we did not find any significant relationships between overall noise levels and sleep.[9] In this dataset, nighttime noise is still a predictor of sleep loss in the hospital. However, when we restrict our sample to those who answered the SSE questionnaire and had nighttime noise recorded, we lose a significant number of observations. Because of our interest in testing the relationship between SSE and sleep, we chose to control for overall noise (which enabled us to retain more observations). We also did not find any interactions between SSE and noise in our regression models. Further work is warranted with larger sample sizes to better understand the role of SSE in the context of sleep and noise levels. In addition, females also received more sleep than males in our study.

There are several limitations to this study. This study was carried out at a single service at a single institution, limiting the ability to generalize the findings to other hospital settings. This study had a relatively high rate of patients who were unable to complete at least 1 night of data collection (42%), often due to watch removal for imaging or procedures, which may also affect the representativeness of our sample. Moreover, we can only examine associations and not causal relationships. The SSE scale has never been used in hospitalized patients, making comparisons between scores from hospitalized patients and population controls difficult. In addition, the SSE scale also has not been dichotomized in previous studies into high and low SSE. However, a sensitivity analysis with raw SSE scores did not change the results of our study. It can be difficult to perform actigraphy measurements in the hospital because many patients spend most of their time in bed. Because we chose a relatively healthy cohort of patients without significant limitations in mobility, actigraphy could still be used to differentiate time spent awake from time spent sleeping. Because we did not perform polysomnography, we cannot explore the role of sleep architecture which is an important component of sleep quality. Although the use of pharmacologic sleep aids is a potential confounding factor, the rate of use was very low in our cohort and unlikely to significantly affect our results. Continued study of this patient population is warranted to further develop the findings.

In conclusion, patients with high SSE sleep better in the hospital, tend to be in quieter rooms, and may report fewer noise complaints. Our findings suggest that a greater confidence in the ability to sleep may be beneficial in hospitalized adults. In addition to noise control, hospitals should also consider targeting patients with low SSE when designing novel interventions to improve in‐hospital sleep.

Disclosures

This work was supported by funding from the National Institute on Aging through a Short‐Term Aging‐Related Research Program (1 T35 AG029795), National Institute on Aging career development award (K23AG033763), a midcareer career development award (1K24AG031326), a program project (P01AG‐11412), an Agency for Healthcare Research and Quality Centers for Education and Research on Therapeutics grant (1U18HS016967), and a National Institute on Aging Clinical Translational Sciences award (UL1 RR024999). Dr. Arora had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the statistical analysis. The funding agencies had no role in the design of the study; the collection, analysis, and interpretation of the data; or the decision to approve publication of the finished manuscript. The authors report no conflicts of interest.

Lack of sleep is a common problem in hospitalized patients and is associated with poorer health outcomes, especially in older patients.[1, 2, 3] Prior studies highlight a multitude of factors that can result in sleep loss in the hospital[3, 4, 5, 6] with 1 of the most common causes of sleep disruption in the hospital being noise.[7, 8, 9]

In addition to external factors, such as hospital noise, there may be inherent characteristics that predispose certain patients to greater sleep loss when hospitalized. One such measure is the construct of perceived control or the psychological measure of how much individuals expect themselves to be capable of bringing about desired outcomes.[10] Among older patients, low perceived control is associated with increased rates of physician visits, hospitalizations, and death.[11, 12] In contrast, patients who feel more in control of their environment may experience positive health benefits.[13]

Yet, when patients are placed in a hospital setting, they experience a significant reduction in control over their environment along with an increase in dependency on medical staff and therapies.[14, 15] For example, hospitalized patients are restricted in their personal decisions, such as what clothes they can wear and what they can eat and are not in charge of their own schedules, including their sleep time.

Although prior studies suggest that perceived control over sleep is related to actual sleep among community‐dwelling adults,[16, 17] no study has examined this relationship in hospitalized adults. Therefore, the aim of our study was to examine the possible association between perceived control, noise levels, and sleep in hospitalized middle‐aged and older patients.

METHODS

RESULTS

From April 2010 to May 2012, 1134 patients were screened by study personnel for this study via an ongoing study of hospitalized patients on the inpatient general medicine ward. Of the 361 (31.8%) eligible patients, 206 (57.1%) consented to participate. Of the subjects enrolled in the study, 118 were able to complete at least 1 night of actigraphy, sound monitoring, and subjective assessment for a total of 185 patient nights (Figure 1).

Figure 1

The majority of patients were female (57%), African American (67%), and non‐Hispanic (97%). The mean age was 65 years (standard deviation [SD], 11.6 years), and the median length of stay was 4 days (interquartile range [IQR], 36). The majority of patients also had hypertension (67%), with chronic obstructive pulmonary disease [COPD] (31%) and congestive heart failure (31%) being the next most common comorbidities. About two‐thirds of subjects (64%) were characterized as average or above average sleepers with Epworth Sleepiness Scale scores 9[20] (Table 1). Only 5% of patients received pharmacological sleep aids.

The mean baseline SLOC score was 30.4 (SD, 6.7), with a median of 31 (IQR, 2735). The mean baseline SSE score was 32.1 (SD, 9.4), with a median of 34 (IQR, 2441). Fifty‐four patients were categorized as having high sleep self‐efficacy (high SSE), which we defined as scoring above the median of 34.

Average in‐hospital sleep was 5.5 hours (333 minutes; SD, 128 minutes) which was significantly shorter than the self‐reported sleep duration of 6.5 hours prior to admission (387 minutes, SD, 125 minutes; P=0.0001). The mean sleep efficiency was 73% (SD, 19%) with 55% of actigraphy nights below the normal range of 80% efficiency for adults.[19] Median KSQI was 3.5 (IQR, 2.254.75), with 41% of the patients with a KSQI 3, putting them in the insomniac range.[32] The median score on the noise disruptiveness question was 1 (IQR, 14) with 42% of reports coded as disruptive defined as a score >1 on the 5‐point scale. The median score on the roommate disruptiveness question was 1 (IQR, 11) with 77% of responses coded as not disruptive defined as a score of 1 on the 5‐point scale.

A 2‐sample t test with equal variances showed that those patients reporting high SSE were more likely to sleep longer in the hospital than those reporting low SSE (364 minutes 95% confidence interval [CI]: 340, 388 vs 309 minutes 95% CI: 283, 336; P=0.003) (Figure 2). Patients with high SSE were also more likely to have a normal sleep efficiency (above 80%) compared to those with low SSE (54% 95% CI: 43, 65 vs 38% 95% CI: 28,47; P=0.028). Last, there was a trend toward patients reporting higher SSE to also report less noise disruption compared to those patients with low SSE ([42%] 95% CI: 31, 53 vs [56%] 95% CI: 46, 65; P=0.063) (Figure 3).

Figure 2

Figure 3

Linear regression clustered by subject showed that high SSE was associated with longer sleep duration (55 minutes 95% CI: 14, 97; P=0.010). Furthermore, high SSE was significantly associated with longer sleep duration after controlling for both objective noise level and patient demographics in the model using stepwise forward regression (50 minutes 95% CI: 11, 90; P=0.014) (Table 2).

Logistic regression clustered by subject demonstrated that patients with high SSE had 2 times higher odds of having a KSQI score above 3 (95% CI: 1.12, 3.71; P=0.020). This association was still significant after controlling for noise and patient demographics (OR: 2.01; 95% CI: 1.06, 3.79; P=0.032). After controlling for noise levels and patient demographics, there was a statistically significant association between high SSE and lower odds of noise complaints (OR: 0.49; 95% CI: 0.25, 0.96; P=0.039) (Table 2). Although demographic characteristics were not associated with high SSE, those patients with high SSE had lower odds of being in the loudest tertile rooms (OR: 0.34; 95% CI: 0.15, 0.74; P=0.007).

In multivariate linear regression analyses, there were no significant relationships between SLOC scores and KSQI, reported noise disruptiveness, and markers of sleep (sleep duration or sleep efficiency).

DISCUSSION

This study is the first to examine the relationship between perceived control, noise levels, and objective measurements of sleep in a hospital setting. One measure of perceived control, namely SSE, was associated with objective sleep duration, subjective and objective sleep quality, noise levels in patient rooms, and perhaps also patient complaints of noise. These associations remained significant after controlling for objective noise levels and patient demographics, suggesting that SSE is independently related to sleep.

In contrast to SSE, SLOC was not found to be significantly associated with either subjective or objective measures of sleep quality. The lack of association may be due to the fact that the SLOC questionnaire does not translate as well to the inpatient setting as the SSE questionnaire. The SLOC questionnaire focuses on general beliefs about sleep whereas the SSE questionnaire focuses on personal beliefs about one's own ability sleep in the immediate future, which may make it more relevant in the inpatient setting (see Supporting Information, Appendix 1 and 2, in the online version of this article).

Given our findings, it is important to identify why patients with high SSE have better sleep and fewer noise complaints. One possibility is that sleep self‐efficacy is an inherited trait unique to each person that is also predictive of a patient's sleep patterns. However, is it also possible that those patients with high SSE feel more empowered to take control of their environment, allowing them to advocate for better sleep? This hypothesis is further strengthened by the finding that those patients with high SSE on study entry were less likely to be in the noisiest rooms. This raises the possibility that at least 1 of the mechanisms by which high SSE may be protective against sleep loss is through patients taking an active role in noise reduction, such as closing the door or advocating for their sleep with staff. However, we did not directly observe or ask patients whether doors of patient rooms were open or closed or whether the patients took other measures to advocate for their own sleep. Thus, further work is necessary to understand the mechanisms by which sleep self‐efficacy may influence sleep.

One potential avenue for future research is to explore possible interventions for boosting sleep self‐efficacy in the hospital. Although most interventions have focused on environmental noise and staff‐based education, empowering patients through boosting SSE may be a helpful adjunct to improving hospital sleep.[33, 34] Currently, the SSE scale is not commonly used in the inpatient setting. Motivational interviewing and patient coaching could be explored as potential tools for boosting SSE. Furthermore, even if SSE is not easily changed, measuring SSE in patients newly admitted to the hospital may be useful in identifying patients most susceptible to sleep disruptions. Efforts to identify patients with low SSE should go hand‐in‐hand with measures to reduce noise. Addressing both patient‐level and environmental factors simultaneously may be the best strategy for improving sleep in an inpatient hospital setting.

In contrast to our prior study, it is worth noting that we did not find any significant relationships between overall noise levels and sleep.[9] In this dataset, nighttime noise is still a predictor of sleep loss in the hospital. However, when we restrict our sample to those who answered the SSE questionnaire and had nighttime noise recorded, we lose a significant number of observations. Because of our interest in testing the relationship between SSE and sleep, we chose to control for overall noise (which enabled us to retain more observations). We also did not find any interactions between SSE and noise in our regression models. Further work is warranted with larger sample sizes to better understand the role of SSE in the context of sleep and noise levels. In addition, females also received more sleep than males in our study.

There are several limitations to this study. This study was carried out at a single service at a single institution, limiting the ability to generalize the findings to other hospital settings. This study had a relatively high rate of patients who were unable to complete at least 1 night of data collection (42%), often due to watch removal for imaging or procedures, which may also affect the representativeness of our sample. Moreover, we can only examine associations and not causal relationships. The SSE scale has never been used in hospitalized patients, making comparisons between scores from hospitalized patients and population controls difficult. In addition, the SSE scale also has not been dichotomized in previous studies into high and low SSE. However, a sensitivity analysis with raw SSE scores did not change the results of our study. It can be difficult to perform actigraphy measurements in the hospital because many patients spend most of their time in bed. Because we chose a relatively healthy cohort of patients without significant limitations in mobility, actigraphy could still be used to differentiate time spent awake from time spent sleeping. Because we did not perform polysomnography, we cannot explore the role of sleep architecture which is an important component of sleep quality. Although the use of pharmacologic sleep aids is a potential confounding factor, the rate of use was very low in our cohort and unlikely to significantly affect our results. Continued study of this patient population is warranted to further develop the findings.

In conclusion, patients with high SSE sleep better in the hospital, tend to be in quieter rooms, and may report fewer noise complaints. Our findings suggest that a greater confidence in the ability to sleep may be beneficial in hospitalized adults. In addition to noise control, hospitals should also consider targeting patients with low SSE when designing novel interventions to improve in‐hospital sleep.

Disclosures

This work was supported by funding from the National Institute on Aging through a Short‐Term Aging‐Related Research Program (1 T35 AG029795), National Institute on Aging career development award (K23AG033763), a midcareer career development award (1K24AG031326), a program project (P01AG‐11412), an Agency for Healthcare Research and Quality Centers for Education and Research on Therapeutics grant (1U18HS016967), and a National Institute on Aging Clinical Translational Sciences award (UL1 RR024999). Dr. Arora had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the statistical analysis. The funding agencies had no role in the design of the study; the collection, analysis, and interpretation of the data; or the decision to approve publication of the finished manuscript. The authors report no conflicts of interest.

Lack of sleep is a common problem in hospitalized patients and is associated with poorer health outcomes, especially in older patients.[1, 2, 3] Prior studies highlight a multitude of factors that can result in sleep loss in the hospital[3, 4, 5, 6] with 1 of the most common causes of sleep disruption in the hospital being noise.[7, 8, 9]

In addition to external factors, such as hospital noise, there may be inherent characteristics that predispose certain patients to greater sleep loss when hospitalized. One such measure is the construct of perceived control or the psychological measure of how much individuals expect themselves to be capable of bringing about desired outcomes.[10] Among older patients, low perceived control is associated with increased rates of physician visits, hospitalizations, and death.[11, 12] In contrast, patients who feel more in control of their environment may experience positive health benefits.[13]

Yet, when patients are placed in a hospital setting, they experience a significant reduction in control over their environment along with an increase in dependency on medical staff and therapies.[14, 15] For example, hospitalized patients are restricted in their personal decisions, such as what clothes they can wear and what they can eat and are not in charge of their own schedules, including their sleep time.

Although prior studies suggest that perceived control over sleep is related to actual sleep among community‐dwelling adults,[16, 17] no study has examined this relationship in hospitalized adults. Therefore, the aim of our study was to examine the possible association between perceived control, noise levels, and sleep in hospitalized middle‐aged and older patients.

METHODS

RESULTS

From April 2010 to May 2012, 1134 patients were screened by study personnel for this study via an ongoing study of hospitalized patients on the inpatient general medicine ward. Of the 361 (31.8%) eligible patients, 206 (57.1%) consented to participate. Of the subjects enrolled in the study, 118 were able to complete at least 1 night of actigraphy, sound monitoring, and subjective assessment for a total of 185 patient nights (Figure 1).

Figure 1

The majority of patients were female (57%), African American (67%), and non‐Hispanic (97%). The mean age was 65 years (standard deviation [SD], 11.6 years), and the median length of stay was 4 days (interquartile range [IQR], 36). The majority of patients also had hypertension (67%), with chronic obstructive pulmonary disease [COPD] (31%) and congestive heart failure (31%) being the next most common comorbidities. About two‐thirds of subjects (64%) were characterized as average or above average sleepers with Epworth Sleepiness Scale scores 9[20] (Table 1). Only 5% of patients received pharmacological sleep aids.

The mean baseline SLOC score was 30.4 (SD, 6.7), with a median of 31 (IQR, 2735). The mean baseline SSE score was 32.1 (SD, 9.4), with a median of 34 (IQR, 2441). Fifty‐four patients were categorized as having high sleep self‐efficacy (high SSE), which we defined as scoring above the median of 34.

Average in‐hospital sleep was 5.5 hours (333 minutes; SD, 128 minutes) which was significantly shorter than the self‐reported sleep duration of 6.5 hours prior to admission (387 minutes, SD, 125 minutes; P=0.0001). The mean sleep efficiency was 73% (SD, 19%) with 55% of actigraphy nights below the normal range of 80% efficiency for adults.[19] Median KSQI was 3.5 (IQR, 2.254.75), with 41% of the patients with a KSQI 3, putting them in the insomniac range.[32] The median score on the noise disruptiveness question was 1 (IQR, 14) with 42% of reports coded as disruptive defined as a score >1 on the 5‐point scale. The median score on the roommate disruptiveness question was 1 (IQR, 11) with 77% of responses coded as not disruptive defined as a score of 1 on the 5‐point scale.

A 2‐sample t test with equal variances showed that those patients reporting high SSE were more likely to sleep longer in the hospital than those reporting low SSE (364 minutes 95% confidence interval [CI]: 340, 388 vs 309 minutes 95% CI: 283, 336; P=0.003) (Figure 2). Patients with high SSE were also more likely to have a normal sleep efficiency (above 80%) compared to those with low SSE (54% 95% CI: 43, 65 vs 38% 95% CI: 28,47; P=0.028). Last, there was a trend toward patients reporting higher SSE to also report less noise disruption compared to those patients with low SSE ([42%] 95% CI: 31, 53 vs [56%] 95% CI: 46, 65; P=0.063) (Figure 3).

Figure 2

Figure 3

Linear regression clustered by subject showed that high SSE was associated with longer sleep duration (55 minutes 95% CI: 14, 97; P=0.010). Furthermore, high SSE was significantly associated with longer sleep duration after controlling for both objective noise level and patient demographics in the model using stepwise forward regression (50 minutes 95% CI: 11, 90; P=0.014) (Table 2).

Logistic regression clustered by subject demonstrated that patients with high SSE had 2 times higher odds of having a KSQI score above 3 (95% CI: 1.12, 3.71; P=0.020). This association was still significant after controlling for noise and patient demographics (OR: 2.01; 95% CI: 1.06, 3.79; P=0.032). After controlling for noise levels and patient demographics, there was a statistically significant association between high SSE and lower odds of noise complaints (OR: 0.49; 95% CI: 0.25, 0.96; P=0.039) (Table 2). Although demographic characteristics were not associated with high SSE, those patients with high SSE had lower odds of being in the loudest tertile rooms (OR: 0.34; 95% CI: 0.15, 0.74; P=0.007).

In multivariate linear regression analyses, there were no significant relationships between SLOC scores and KSQI, reported noise disruptiveness, and markers of sleep (sleep duration or sleep efficiency).

DISCUSSION

This study is the first to examine the relationship between perceived control, noise levels, and objective measurements of sleep in a hospital setting. One measure of perceived control, namely SSE, was associated with objective sleep duration, subjective and objective sleep quality, noise levels in patient rooms, and perhaps also patient complaints of noise. These associations remained significant after controlling for objective noise levels and patient demographics, suggesting that SSE is independently related to sleep.

In contrast to SSE, SLOC was not found to be significantly associated with either subjective or objective measures of sleep quality. The lack of association may be due to the fact that the SLOC questionnaire does not translate as well to the inpatient setting as the SSE questionnaire. The SLOC questionnaire focuses on general beliefs about sleep whereas the SSE questionnaire focuses on personal beliefs about one's own ability sleep in the immediate future, which may make it more relevant in the inpatient setting (see Supporting Information, Appendix 1 and 2, in the online version of this article).

Given our findings, it is important to identify why patients with high SSE have better sleep and fewer noise complaints. One possibility is that sleep self‐efficacy is an inherited trait unique to each person that is also predictive of a patient's sleep patterns. However, is it also possible that those patients with high SSE feel more empowered to take control of their environment, allowing them to advocate for better sleep? This hypothesis is further strengthened by the finding that those patients with high SSE on study entry were less likely to be in the noisiest rooms. This raises the possibility that at least 1 of the mechanisms by which high SSE may be protective against sleep loss is through patients taking an active role in noise reduction, such as closing the door or advocating for their sleep with staff. However, we did not directly observe or ask patients whether doors of patient rooms were open or closed or whether the patients took other measures to advocate for their own sleep. Thus, further work is necessary to understand the mechanisms by which sleep self‐efficacy may influence sleep.

One potential avenue for future research is to explore possible interventions for boosting sleep self‐efficacy in the hospital. Although most interventions have focused on environmental noise and staff‐based education, empowering patients through boosting SSE may be a helpful adjunct to improving hospital sleep.[33, 34] Currently, the SSE scale is not commonly used in the inpatient setting. Motivational interviewing and patient coaching could be explored as potential tools for boosting SSE. Furthermore, even if SSE is not easily changed, measuring SSE in patients newly admitted to the hospital may be useful in identifying patients most susceptible to sleep disruptions. Efforts to identify patients with low SSE should go hand‐in‐hand with measures to reduce noise. Addressing both patient‐level and environmental factors simultaneously may be the best strategy for improving sleep in an inpatient hospital setting.

In contrast to our prior study, it is worth noting that we did not find any significant relationships between overall noise levels and sleep.[9] In this dataset, nighttime noise is still a predictor of sleep loss in the hospital. However, when we restrict our sample to those who answered the SSE questionnaire and had nighttime noise recorded, we lose a significant number of observations. Because of our interest in testing the relationship between SSE and sleep, we chose to control for overall noise (which enabled us to retain more observations). We also did not find any interactions between SSE and noise in our regression models. Further work is warranted with larger sample sizes to better understand the role of SSE in the context of sleep and noise levels. In addition, females also received more sleep than males in our study.

There are several limitations to this study. This study was carried out at a single service at a single institution, limiting the ability to generalize the findings to other hospital settings. This study had a relatively high rate of patients who were unable to complete at least 1 night of data collection (42%), often due to watch removal for imaging or procedures, which may also affect the representativeness of our sample. Moreover, we can only examine associations and not causal relationships. The SSE scale has never been used in hospitalized patients, making comparisons between scores from hospitalized patients and population controls difficult. In addition, the SSE scale also has not been dichotomized in previous studies into high and low SSE. However, a sensitivity analysis with raw SSE scores did not change the results of our study. It can be difficult to perform actigraphy measurements in the hospital because many patients spend most of their time in bed. Because we chose a relatively healthy cohort of patients without significant limitations in mobility, actigraphy could still be used to differentiate time spent awake from time spent sleeping. Because we did not perform polysomnography, we cannot explore the role of sleep architecture which is an important component of sleep quality. Although the use of pharmacologic sleep aids is a potential confounding factor, the rate of use was very low in our cohort and unlikely to significantly affect our results. Continued study of this patient population is warranted to further develop the findings.

In conclusion, patients with high SSE sleep better in the hospital, tend to be in quieter rooms, and may report fewer noise complaints. Our findings suggest that a greater confidence in the ability to sleep may be beneficial in hospitalized adults. In addition to noise control, hospitals should also consider targeting patients with low SSE when designing novel interventions to improve in‐hospital sleep.

Disclosures

This work was supported by funding from the National Institute on Aging through a Short‐Term Aging‐Related Research Program (1 T35 AG029795), National Institute on Aging career development award (K23AG033763), a midcareer career development award (1K24AG031326), a program project (P01AG‐11412), an Agency for Healthcare Research and Quality Centers for Education and Research on Therapeutics grant (1U18HS016967), and a National Institute on Aging Clinical Translational Sciences award (UL1 RR024999). Dr. Arora had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the statistical analysis. The funding agencies had no role in the design of the study; the collection, analysis, and interpretation of the data; or the decision to approve publication of the finished manuscript. The authors report no conflicts of interest.

A 51‐year‐old man presented with severe pain and swelling in the lower anterior right thigh. He stated that the symptoms limited his movement, and began 4 days prior to this presentation. He rated the pain severity a 10 on a 10‐point scale. He denied fevers, chills, or history of trauma or weight loss.

Cellulitis of the lower extremity is the most likely possibility, but the presence of severe pain and swelling of an extremity in the absence of trauma should always make the clinician consider deep‐seated infections such as myositis or necrotizing fasciitis. An early clue for necrotizing fasciitis is severe pain that is disproportionate to the physical examination findings. Erythema, bullous lesions, or crepitus can develop later in the course. The absence of fever and chills also raises the possibility of noninfectious causes such as unrecognized trauma, deep vein thrombosis, or tumor.

The patient had a 15‐year history of type 2 diabetes complicated by end‐stage renal disease secondary to diabetic nephropathy for which he had been on hemodialysis for 5 months, proliferative diabetic retinopathy that rendered him legally blind, hypertension, and anemia. He stated that his diabetes had been poorly controlled, especially after he started dialysis.

A history of poorly controlled diabetes mellitus certainly increases the risk of the infectious disorders mentioned above. The patient's long‐standing history of diabetes mellitus with secondary nephropathy and retinopathy puts him at higher risk of atherosclerosis and vascular insufficiency, which consequently increase his risk for ischemic myonecrosis. Diabetic amyotrophy (diabetic lumbosacral plexopathy) is also a possibility, as it usually manifests with acute, unilateral, and focal tenderness followed by weakness involving a proximal leg. However, it typically occurs in patients who have been recently diagnosed with type 2 diabetes mellitus or whose disease has been under fairly good control and usually is associated with significant weight loss.

The patient was on oral medications for his diabetes until 1year before his presentation, at which point he was switched to insulin therapy. His other medications were amlodipine, lisinopril, aspirin, sevelamer, calcitriol, and calcium and iron supplements. He denied using alcohol, tobacco, or illicit drugs. He lives in Chicago and denies a recent travel history. His family history was significant for type 2 diabetes in multiple family members.

The absence of drugs, tobacco, and alcohol lowers the risk of some infectious and ischemic conditions. Patients with alcoholic liver disease who live in the southern United States are predisposed to developing Vibrio vulnificus myositis and fasciitis after ingesting contaminated oysters during the summer months. However, the clinical presentation of Vibrio usually includes septic shock and bullous lesions on the lower extremity. Also, the patient denies any recent travel to the southern United States, which makes Vibrio myositis and fasciitis less likely. Tobacco abuse increases the risk of atherosclerosis, peripheral vascular insufficiency, and ischemic myonecrosis.

The patient had a temperature of 99.1F, blood pressure of 139/85 mm Hg, pulse of 97 beats/minute, and respiratory rate of 18 breaths/minute. His body mass index was 31 kg/m². Physical examination revealed a firm, warm, severely tender area of swelling in the inferomedial aspect of the right thigh. The knee was also swollen, and effusion could not be ruled out. The range of motion of the knee was markedly limited by pain. The skin overlying the swelling was erythematous but not broken. No crepitus was noted. The strength of the right lower extremity muscles could not be accurately assessed because of the patient's excruciating pain, but the patient was able to move his foot and toes against gravity. Sensation was absent in most of the tested points in the feet but was normal in the legs. The deep tendon reflexes in both ankles were normal. The pedal pulses were mildly decreased in both feet. He also had extremely decreased visual acuity, which has been chronic. The rest of the physical examination was unremarkable.

The absence of fever does not rule out a serious infection in a diabetic patient but does raise the possibility of a noninfectious cause. Also, over‐the‐counter acetaminophen or nonsteroidal anti‐inflammatory drugs could mask a fever. The patient's physical examination was significant for obesity, a risk factor for developing deep‐seated infections, and a firm and severely tender area of swelling near the right knee that limited range of motion. Septic arthritis of the knee is one possibility; arthrocentesis should be performed as soon as possible. The absence of crepitus, because it is a late physical examination finding, does not rule out myositis or necrotizing fasciitis. The presence of unilateral lower extremity swelling also raises the suspicion for a deep vein thrombosis, which warrants compression ultrasonography. The localized tenderness and the lack of dermatological manifestations, such as Gottron's papules, makes an inflammatory myositis such as dermatomyositis much less likely.

Laboratory studies demonstrated a hemoglobin A1C of 13.0% (reference range, 4.36.1%), fasting blood glucose level of 224 mg/dL (reference range, 7099 mg/dL), white blood cell count of 8300 cells/mm³ (reference range, 450011,000 cells/mm³) without band forms, erythrocyte sedimentation rate of 81 mm/hr (reference range, <14 mm/hr), and creatinine kinase level of 582 IU/L (reference range, 30200 IU/L). Routine chemistries were normal otherwise. An x‐ray of the right knee revealed soft tissue edema. The right knee was aspirated, and fluid analysis revealed a white blood cell count of 106 cells/mm³ (reference range, <200 cell/mm³). Compression ultrasonography of the right lower extremity did not reveal thrombosis.

Poor glycemic control, as evidenced by a high hemoglobin A1C level, is associated with a higher probability of infectious complications. An elevated sedimentation rate is compatible with an infection, and an increased creatinine kinase intensifies suspicion of myositis or myonecrosis. A normal white blood cell count decreases, but does not eliminate, the likelihood of a serious bacterial infection. The fluid analysis rules out septic arthritis, and the compression ultrasonography findings make deep vein thrombosis very unlikely. However, the differential diagnosis still includes myositis, clostridial myonecrosis, cellulitis, and necrotizing fasciitis. The patient should undergo magnetic resonance imaging (MRI) of the lower extremity, and a surgical consultation should be obtained to consider the possibility of surgical exploration.

Blood and the aspirated fluids were sent for culturing, and the patient was started on empiric antibiotics. MRI of his right thigh revealed extensive edema involving the vastus medialis and lateralis of the quadriceps as well as subcutaneous edema without fascial enhancement or gas (Figure 1).

Figure 1

The absence of gas and fascial enhancement makes clostridial myonecrosis or necrotizing fasciitis less likely. The absence of a fluid collection in the muscle makes pyomyositis due to Staphylococci unlikely. Broad‐spectrum antibiotic coverage (usually vancomycin and either piperacillin/tazobactam or a carbapenem) targeting methicillin‐resistant Staphylococcus aureus, anaerobes, Streptococci, and Enterobacteriaceae should be empirically started as soon as cultures are obtained. Clindamycin should be part of the empiric antibiotic regimen to block toxin production in the event that Streptococcus pyogenes is responsible.

Surgical biopsy of the right vastus medialis muscle was performed, and tissue was sent for Gram staining, culture, and routine histopathological analysis. Gram staining was negative, and histopathological analysis revealed ischemic skeletal muscle fibers with areas of necrosis (Figure 2). Cultures from blood, fluid from the right knee, and muscular tissue samples did not grow any bacteria.

Figure 2

The muscle biopsy results are consistent with myonecrosis. Clostridial myonecrosis is possible but usually is associated with gas in tissues or occurs in the setting of intra‐abdominal pathology or severe trauma, and tissue culture was negative. Ischemic myonecrosis due to severe vascular insufficiency would be unlikely given the presence of pedal pulses and the absence of toes or forefoot cyanosis. A vasculitis syndrome is also unlikely because of the focal nature of the findings and the absence of weight loss, muscle weakness, and chronic joint pain in the patient's history. Calciphylaxis (calcific uremic arteriolopathy) might be considered in a patient with end‐stage renal disease who presents with a thigh pain; however, this condition is usually characterized by areas of ischemic necrosis that develop in the dermis and/or subcutaneous fat and infrequently involve muscles. The absence of the painful subcutaneous nodules typical of calciphylaxis makes it an unlikely diagnosis.

A diagnosis of diabetic myonecrosis was made. Antibiotics were discontinued, and the patient was treated symptomatically. His symptoms improved during the next few days. The patient was discharged from the hospital, and conservative management with bed rest and analgesics for 4 weeks was prescribed. Four months later, however, the patient returned with similar symptoms in the contralateral thigh. The patient was diagnosed with recurrent diabetic myonecrosis by MRI and muscle biopsy findings. Conservative management was advised, and the patient became pain‐free in a few weeks.

DISCUSSION

Diabetic myonecrosis (also known as diabetic muscle infarction) is a rare disorder initially described in 1965[1] that typically presents spontaneously as an acute, localized, severely painful swelling that limits the mobility of the affected extremity, usually without systemic signs of infection. It affects the thighs in 83% of patients and the calves in 17% of patients.[2, 3] Bilateral involvement, which is usually asynchronous, occurs in one‐third of patients.[4] The upper limbs are rarely involved. Diabetic myonecrosis affects patients who have a relatively longstanding history of diabetes. It is commonly associated with the microvascular complications of diabetes, including nephropathy (80% of patients), retinopathy (60% of patents), and/or neuropathy (64% of patients).[3, 5] The pathogenesis of diabetic myonecrosis is unclear, but the disease is likely due to a diffuse microangiopathy and atherosclerosis.[2, 5] Some authors have suggested that abnormalities in the clotting or fibrinolytic pathways play a role in the etiology of the disorder.[6]

Clinical and MRI findings can be used to make the diagnosis with reasonable certainty.[3, 5] Although both ultrasonography and MRI have been used to assess patients with diabetic myonecrosis, MRI with intravenous contrast enhancement appears to be the most useful diagnostic technique. It demonstrates extensive edema within the muscle(s), muscle enlargement, subcutaneous and interfascial edema, a patchwork pattern of involvement, and a high signal intensity on T2‐weighted images.[4, 7] Gadolinium enhancement may reveal an enhanced margin of the infarcted muscle with a central nonenhancing area of necrotic tissue.[4, 5] Muscle biopsy is not typically indicated because it may prolong recovery time and lead to infections.[8, 9, 10, 11] When performed, however, muscle biopsy reveals ischemic muscle fibers in different stages of degeneration and regeneration, with areas of necrosis and edema. Occlusion of arterioles and capillaries by fibrin could also be seen.[1] Although the patient underwent a muscle biopsy because infection could not be excluded definitively on clinical grounds, we believe that repeating the biopsy 4 months later was inappropriate.

Diabetic myonecrosis should be considered in a diabetic patient who presents with severe localized muscle pain and swelling of an extremity, especially if the clinical features favoring infection are absent. The differential diagnosis should include infection (eg, clostridial myonecrosis, myositis, cellulitis, abscess, necrotizing fasciitis, osteomyelitis), trauma (eg, hematoma, muscle rupture, myositis ossificans), peripheral neuropathy (particularly lumbosacral plexopathy), vascular disorders (deep vein thrombosis, and compartment syndrome), tumors, inflammatory muscle diseases, and drug‐related myositis.

No evidence‐based recommendations regarding the management of diabetic myonecrosis are available, although the findings of one retrospective analysis support conservative management with bed rest, leg elevation, and analgesics.[12] Physiotherapy may cause the condition to worsen,[13, 14] but routine daily activity, although often painful, is not harmful.[14] Some authors suggest a cautious use of antiplatelet or anti‐inflammatory medications.[12] We would also recommend achieving good glycemic control during the illness. Owing to the rarity of the disease, however, no studies have definitively shown that this hastens recovery or prevents recurrent diabetic myonecrosis. Surgery may prolong the recovery period; one study found that the recovery period of patients with diabetic myonecrosis who underwent surgery was longer than that of those who were treated conservatively (13 weeks vs 5.5 weeks).[12] Patients with diabetic myonecrosis have a good short‐term prognosis. Longer‐term, however, they have a poor prognosis; their recurrence rate is as high as 40%, and their 2‐year mortality rate is 10%, even after one episode of the disease. Death in these patients is mainly due to macrovascular events.[12]

TEACHING POINTS

1. Diabetic myonecrosis is a rare complication of longstanding and poorly controlled diabetes. It usually presents with acute localized muscular pain in the lower extremities.
2. Although a definitive diagnosis of diabetic myonecrosis is histopathologic, a clinical diagnosis can be made with reasonable certainty for patients with compatible MRI findings and no clinical or laboratory features suggesting infection.
3. Conservative management with bed rest, analgesics, and antiplatelets is recommended. Surgery should be avoided, as it may prolong recovery.

Disclosure

Nothing to report.

A 51‐year‐old man presented with severe pain and swelling in the lower anterior right thigh. He stated that the symptoms limited his movement, and began 4 days prior to this presentation. He rated the pain severity a 10 on a 10‐point scale. He denied fevers, chills, or history of trauma or weight loss.

Cellulitis of the lower extremity is the most likely possibility, but the presence of severe pain and swelling of an extremity in the absence of trauma should always make the clinician consider deep‐seated infections such as myositis or necrotizing fasciitis. An early clue for necrotizing fasciitis is severe pain that is disproportionate to the physical examination findings. Erythema, bullous lesions, or crepitus can develop later in the course. The absence of fever and chills also raises the possibility of noninfectious causes such as unrecognized trauma, deep vein thrombosis, or tumor.

The patient had a 15‐year history of type 2 diabetes complicated by end‐stage renal disease secondary to diabetic nephropathy for which he had been on hemodialysis for 5 months, proliferative diabetic retinopathy that rendered him legally blind, hypertension, and anemia. He stated that his diabetes had been poorly controlled, especially after he started dialysis.

A history of poorly controlled diabetes mellitus certainly increases the risk of the infectious disorders mentioned above. The patient's long‐standing history of diabetes mellitus with secondary nephropathy and retinopathy puts him at higher risk of atherosclerosis and vascular insufficiency, which consequently increase his risk for ischemic myonecrosis. Diabetic amyotrophy (diabetic lumbosacral plexopathy) is also a possibility, as it usually manifests with acute, unilateral, and focal tenderness followed by weakness involving a proximal leg. However, it typically occurs in patients who have been recently diagnosed with type 2 diabetes mellitus or whose disease has been under fairly good control and usually is associated with significant weight loss.

The patient was on oral medications for his diabetes until 1year before his presentation, at which point he was switched to insulin therapy. His other medications were amlodipine, lisinopril, aspirin, sevelamer, calcitriol, and calcium and iron supplements. He denied using alcohol, tobacco, or illicit drugs. He lives in Chicago and denies a recent travel history. His family history was significant for type 2 diabetes in multiple family members.

The absence of drugs, tobacco, and alcohol lowers the risk of some infectious and ischemic conditions. Patients with alcoholic liver disease who live in the southern United States are predisposed to developing Vibrio vulnificus myositis and fasciitis after ingesting contaminated oysters during the summer months. However, the clinical presentation of Vibrio usually includes septic shock and bullous lesions on the lower extremity. Also, the patient denies any recent travel to the southern United States, which makes Vibrio myositis and fasciitis less likely. Tobacco abuse increases the risk of atherosclerosis, peripheral vascular insufficiency, and ischemic myonecrosis.

The patient had a temperature of 99.1F, blood pressure of 139/85 mm Hg, pulse of 97 beats/minute, and respiratory rate of 18 breaths/minute. His body mass index was 31 kg/m². Physical examination revealed a firm, warm, severely tender area of swelling in the inferomedial aspect of the right thigh. The knee was also swollen, and effusion could not be ruled out. The range of motion of the knee was markedly limited by pain. The skin overlying the swelling was erythematous but not broken. No crepitus was noted. The strength of the right lower extremity muscles could not be accurately assessed because of the patient's excruciating pain, but the patient was able to move his foot and toes against gravity. Sensation was absent in most of the tested points in the feet but was normal in the legs. The deep tendon reflexes in both ankles were normal. The pedal pulses were mildly decreased in both feet. He also had extremely decreased visual acuity, which has been chronic. The rest of the physical examination was unremarkable.

The absence of fever does not rule out a serious infection in a diabetic patient but does raise the possibility of a noninfectious cause. Also, over‐the‐counter acetaminophen or nonsteroidal anti‐inflammatory drugs could mask a fever. The patient's physical examination was significant for obesity, a risk factor for developing deep‐seated infections, and a firm and severely tender area of swelling near the right knee that limited range of motion. Septic arthritis of the knee is one possibility; arthrocentesis should be performed as soon as possible. The absence of crepitus, because it is a late physical examination finding, does not rule out myositis or necrotizing fasciitis. The presence of unilateral lower extremity swelling also raises the suspicion for a deep vein thrombosis, which warrants compression ultrasonography. The localized tenderness and the lack of dermatological manifestations, such as Gottron's papules, makes an inflammatory myositis such as dermatomyositis much less likely.

Laboratory studies demonstrated a hemoglobin A1C of 13.0% (reference range, 4.36.1%), fasting blood glucose level of 224 mg/dL (reference range, 7099 mg/dL), white blood cell count of 8300 cells/mm³ (reference range, 450011,000 cells/mm³) without band forms, erythrocyte sedimentation rate of 81 mm/hr (reference range, <14 mm/hr), and creatinine kinase level of 582 IU/L (reference range, 30200 IU/L). Routine chemistries were normal otherwise. An x‐ray of the right knee revealed soft tissue edema. The right knee was aspirated, and fluid analysis revealed a white blood cell count of 106 cells/mm³ (reference range, <200 cell/mm³). Compression ultrasonography of the right lower extremity did not reveal thrombosis.

Poor glycemic control, as evidenced by a high hemoglobin A1C level, is associated with a higher probability of infectious complications. An elevated sedimentation rate is compatible with an infection, and an increased creatinine kinase intensifies suspicion of myositis or myonecrosis. A normal white blood cell count decreases, but does not eliminate, the likelihood of a serious bacterial infection. The fluid analysis rules out septic arthritis, and the compression ultrasonography findings make deep vein thrombosis very unlikely. However, the differential diagnosis still includes myositis, clostridial myonecrosis, cellulitis, and necrotizing fasciitis. The patient should undergo magnetic resonance imaging (MRI) of the lower extremity, and a surgical consultation should be obtained to consider the possibility of surgical exploration.

Blood and the aspirated fluids were sent for culturing, and the patient was started on empiric antibiotics. MRI of his right thigh revealed extensive edema involving the vastus medialis and lateralis of the quadriceps as well as subcutaneous edema without fascial enhancement or gas (Figure 1).

Figure 1

The absence of gas and fascial enhancement makes clostridial myonecrosis or necrotizing fasciitis less likely. The absence of a fluid collection in the muscle makes pyomyositis due to Staphylococci unlikely. Broad‐spectrum antibiotic coverage (usually vancomycin and either piperacillin/tazobactam or a carbapenem) targeting methicillin‐resistant Staphylococcus aureus, anaerobes, Streptococci, and Enterobacteriaceae should be empirically started as soon as cultures are obtained. Clindamycin should be part of the empiric antibiotic regimen to block toxin production in the event that Streptococcus pyogenes is responsible.

Surgical biopsy of the right vastus medialis muscle was performed, and tissue was sent for Gram staining, culture, and routine histopathological analysis. Gram staining was negative, and histopathological analysis revealed ischemic skeletal muscle fibers with areas of necrosis (Figure 2). Cultures from blood, fluid from the right knee, and muscular tissue samples did not grow any bacteria.

Figure 2

The muscle biopsy results are consistent with myonecrosis. Clostridial myonecrosis is possible but usually is associated with gas in tissues or occurs in the setting of intra‐abdominal pathology or severe trauma, and tissue culture was negative. Ischemic myonecrosis due to severe vascular insufficiency would be unlikely given the presence of pedal pulses and the absence of toes or forefoot cyanosis. A vasculitis syndrome is also unlikely because of the focal nature of the findings and the absence of weight loss, muscle weakness, and chronic joint pain in the patient's history. Calciphylaxis (calcific uremic arteriolopathy) might be considered in a patient with end‐stage renal disease who presents with a thigh pain; however, this condition is usually characterized by areas of ischemic necrosis that develop in the dermis and/or subcutaneous fat and infrequently involve muscles. The absence of the painful subcutaneous nodules typical of calciphylaxis makes it an unlikely diagnosis.

A diagnosis of diabetic myonecrosis was made. Antibiotics were discontinued, and the patient was treated symptomatically. His symptoms improved during the next few days. The patient was discharged from the hospital, and conservative management with bed rest and analgesics for 4 weeks was prescribed. Four months later, however, the patient returned with similar symptoms in the contralateral thigh. The patient was diagnosed with recurrent diabetic myonecrosis by MRI and muscle biopsy findings. Conservative management was advised, and the patient became pain‐free in a few weeks.

DISCUSSION

Diabetic myonecrosis (also known as diabetic muscle infarction) is a rare disorder initially described in 1965[1] that typically presents spontaneously as an acute, localized, severely painful swelling that limits the mobility of the affected extremity, usually without systemic signs of infection. It affects the thighs in 83% of patients and the calves in 17% of patients.[2, 3] Bilateral involvement, which is usually asynchronous, occurs in one‐third of patients.[4] The upper limbs are rarely involved. Diabetic myonecrosis affects patients who have a relatively longstanding history of diabetes. It is commonly associated with the microvascular complications of diabetes, including nephropathy (80% of patients), retinopathy (60% of patents), and/or neuropathy (64% of patients).[3, 5] The pathogenesis of diabetic myonecrosis is unclear, but the disease is likely due to a diffuse microangiopathy and atherosclerosis.[2, 5] Some authors have suggested that abnormalities in the clotting or fibrinolytic pathways play a role in the etiology of the disorder.[6]

Clinical and MRI findings can be used to make the diagnosis with reasonable certainty.[3, 5] Although both ultrasonography and MRI have been used to assess patients with diabetic myonecrosis, MRI with intravenous contrast enhancement appears to be the most useful diagnostic technique. It demonstrates extensive edema within the muscle(s), muscle enlargement, subcutaneous and interfascial edema, a patchwork pattern of involvement, and a high signal intensity on T2‐weighted images.[4, 7] Gadolinium enhancement may reveal an enhanced margin of the infarcted muscle with a central nonenhancing area of necrotic tissue.[4, 5] Muscle biopsy is not typically indicated because it may prolong recovery time and lead to infections.[8, 9, 10, 11] When performed, however, muscle biopsy reveals ischemic muscle fibers in different stages of degeneration and regeneration, with areas of necrosis and edema. Occlusion of arterioles and capillaries by fibrin could also be seen.[1] Although the patient underwent a muscle biopsy because infection could not be excluded definitively on clinical grounds, we believe that repeating the biopsy 4 months later was inappropriate.

Diabetic myonecrosis should be considered in a diabetic patient who presents with severe localized muscle pain and swelling of an extremity, especially if the clinical features favoring infection are absent. The differential diagnosis should include infection (eg, clostridial myonecrosis, myositis, cellulitis, abscess, necrotizing fasciitis, osteomyelitis), trauma (eg, hematoma, muscle rupture, myositis ossificans), peripheral neuropathy (particularly lumbosacral plexopathy), vascular disorders (deep vein thrombosis, and compartment syndrome), tumors, inflammatory muscle diseases, and drug‐related myositis.

No evidence‐based recommendations regarding the management of diabetic myonecrosis are available, although the findings of one retrospective analysis support conservative management with bed rest, leg elevation, and analgesics.[12] Physiotherapy may cause the condition to worsen,[13, 14] but routine daily activity, although often painful, is not harmful.[14] Some authors suggest a cautious use of antiplatelet or anti‐inflammatory medications.[12] We would also recommend achieving good glycemic control during the illness. Owing to the rarity of the disease, however, no studies have definitively shown that this hastens recovery or prevents recurrent diabetic myonecrosis. Surgery may prolong the recovery period; one study found that the recovery period of patients with diabetic myonecrosis who underwent surgery was longer than that of those who were treated conservatively (13 weeks vs 5.5 weeks).[12] Patients with diabetic myonecrosis have a good short‐term prognosis. Longer‐term, however, they have a poor prognosis; their recurrence rate is as high as 40%, and their 2‐year mortality rate is 10%, even after one episode of the disease. Death in these patients is mainly due to macrovascular events.[12]

TEACHING POINTS

1. Diabetic myonecrosis is a rare complication of longstanding and poorly controlled diabetes. It usually presents with acute localized muscular pain in the lower extremities.
2. Although a definitive diagnosis of diabetic myonecrosis is histopathologic, a clinical diagnosis can be made with reasonable certainty for patients with compatible MRI findings and no clinical or laboratory features suggesting infection.
3. Conservative management with bed rest, analgesics, and antiplatelets is recommended. Surgery should be avoided, as it may prolong recovery.

Disclosure

Nothing to report.

A 51‐year‐old man presented with severe pain and swelling in the lower anterior right thigh. He stated that the symptoms limited his movement, and began 4 days prior to this presentation. He rated the pain severity a 10 on a 10‐point scale. He denied fevers, chills, or history of trauma or weight loss.

Cellulitis of the lower extremity is the most likely possibility, but the presence of severe pain and swelling of an extremity in the absence of trauma should always make the clinician consider deep‐seated infections such as myositis or necrotizing fasciitis. An early clue for necrotizing fasciitis is severe pain that is disproportionate to the physical examination findings. Erythema, bullous lesions, or crepitus can develop later in the course. The absence of fever and chills also raises the possibility of noninfectious causes such as unrecognized trauma, deep vein thrombosis, or tumor.

The patient had a 15‐year history of type 2 diabetes complicated by end‐stage renal disease secondary to diabetic nephropathy for which he had been on hemodialysis for 5 months, proliferative diabetic retinopathy that rendered him legally blind, hypertension, and anemia. He stated that his diabetes had been poorly controlled, especially after he started dialysis.

A history of poorly controlled diabetes mellitus certainly increases the risk of the infectious disorders mentioned above. The patient's long‐standing history of diabetes mellitus with secondary nephropathy and retinopathy puts him at higher risk of atherosclerosis and vascular insufficiency, which consequently increase his risk for ischemic myonecrosis. Diabetic amyotrophy (diabetic lumbosacral plexopathy) is also a possibility, as it usually manifests with acute, unilateral, and focal tenderness followed by weakness involving a proximal leg. However, it typically occurs in patients who have been recently diagnosed with type 2 diabetes mellitus or whose disease has been under fairly good control and usually is associated with significant weight loss.

The patient was on oral medications for his diabetes until 1year before his presentation, at which point he was switched to insulin therapy. His other medications were amlodipine, lisinopril, aspirin, sevelamer, calcitriol, and calcium and iron supplements. He denied using alcohol, tobacco, or illicit drugs. He lives in Chicago and denies a recent travel history. His family history was significant for type 2 diabetes in multiple family members.

The absence of drugs, tobacco, and alcohol lowers the risk of some infectious and ischemic conditions. Patients with alcoholic liver disease who live in the southern United States are predisposed to developing Vibrio vulnificus myositis and fasciitis after ingesting contaminated oysters during the summer months. However, the clinical presentation of Vibrio usually includes septic shock and bullous lesions on the lower extremity. Also, the patient denies any recent travel to the southern United States, which makes Vibrio myositis and fasciitis less likely. Tobacco abuse increases the risk of atherosclerosis, peripheral vascular insufficiency, and ischemic myonecrosis.

The patient had a temperature of 99.1F, blood pressure of 139/85 mm Hg, pulse of 97 beats/minute, and respiratory rate of 18 breaths/minute. His body mass index was 31 kg/m². Physical examination revealed a firm, warm, severely tender area of swelling in the inferomedial aspect of the right thigh. The knee was also swollen, and effusion could not be ruled out. The range of motion of the knee was markedly limited by pain. The skin overlying the swelling was erythematous but not broken. No crepitus was noted. The strength of the right lower extremity muscles could not be accurately assessed because of the patient's excruciating pain, but the patient was able to move his foot and toes against gravity. Sensation was absent in most of the tested points in the feet but was normal in the legs. The deep tendon reflexes in both ankles were normal. The pedal pulses were mildly decreased in both feet. He also had extremely decreased visual acuity, which has been chronic. The rest of the physical examination was unremarkable.

The absence of fever does not rule out a serious infection in a diabetic patient but does raise the possibility of a noninfectious cause. Also, over‐the‐counter acetaminophen or nonsteroidal anti‐inflammatory drugs could mask a fever. The patient's physical examination was significant for obesity, a risk factor for developing deep‐seated infections, and a firm and severely tender area of swelling near the right knee that limited range of motion. Septic arthritis of the knee is one possibility; arthrocentesis should be performed as soon as possible. The absence of crepitus, because it is a late physical examination finding, does not rule out myositis or necrotizing fasciitis. The presence of unilateral lower extremity swelling also raises the suspicion for a deep vein thrombosis, which warrants compression ultrasonography. The localized tenderness and the lack of dermatological manifestations, such as Gottron's papules, makes an inflammatory myositis such as dermatomyositis much less likely.

Laboratory studies demonstrated a hemoglobin A1C of 13.0% (reference range, 4.36.1%), fasting blood glucose level of 224 mg/dL (reference range, 7099 mg/dL), white blood cell count of 8300 cells/mm³ (reference range, 450011,000 cells/mm³) without band forms, erythrocyte sedimentation rate of 81 mm/hr (reference range, <14 mm/hr), and creatinine kinase level of 582 IU/L (reference range, 30200 IU/L). Routine chemistries were normal otherwise. An x‐ray of the right knee revealed soft tissue edema. The right knee was aspirated, and fluid analysis revealed a white blood cell count of 106 cells/mm³ (reference range, <200 cell/mm³). Compression ultrasonography of the right lower extremity did not reveal thrombosis.

Poor glycemic control, as evidenced by a high hemoglobin A1C level, is associated with a higher probability of infectious complications. An elevated sedimentation rate is compatible with an infection, and an increased creatinine kinase intensifies suspicion of myositis or myonecrosis. A normal white blood cell count decreases, but does not eliminate, the likelihood of a serious bacterial infection. The fluid analysis rules out septic arthritis, and the compression ultrasonography findings make deep vein thrombosis very unlikely. However, the differential diagnosis still includes myositis, clostridial myonecrosis, cellulitis, and necrotizing fasciitis. The patient should undergo magnetic resonance imaging (MRI) of the lower extremity, and a surgical consultation should be obtained to consider the possibility of surgical exploration.

Blood and the aspirated fluids were sent for culturing, and the patient was started on empiric antibiotics. MRI of his right thigh revealed extensive edema involving the vastus medialis and lateralis of the quadriceps as well as subcutaneous edema without fascial enhancement or gas (Figure 1).

Figure 1

The absence of gas and fascial enhancement makes clostridial myonecrosis or necrotizing fasciitis less likely. The absence of a fluid collection in the muscle makes pyomyositis due to Staphylococci unlikely. Broad‐spectrum antibiotic coverage (usually vancomycin and either piperacillin/tazobactam or a carbapenem) targeting methicillin‐resistant Staphylococcus aureus, anaerobes, Streptococci, and Enterobacteriaceae should be empirically started as soon as cultures are obtained. Clindamycin should be part of the empiric antibiotic regimen to block toxin production in the event that Streptococcus pyogenes is responsible.

Surgical biopsy of the right vastus medialis muscle was performed, and tissue was sent for Gram staining, culture, and routine histopathological analysis. Gram staining was negative, and histopathological analysis revealed ischemic skeletal muscle fibers with areas of necrosis (Figure 2). Cultures from blood, fluid from the right knee, and muscular tissue samples did not grow any bacteria.

Figure 2

The muscle biopsy results are consistent with myonecrosis. Clostridial myonecrosis is possible but usually is associated with gas in tissues or occurs in the setting of intra‐abdominal pathology or severe trauma, and tissue culture was negative. Ischemic myonecrosis due to severe vascular insufficiency would be unlikely given the presence of pedal pulses and the absence of toes or forefoot cyanosis. A vasculitis syndrome is also unlikely because of the focal nature of the findings and the absence of weight loss, muscle weakness, and chronic joint pain in the patient's history. Calciphylaxis (calcific uremic arteriolopathy) might be considered in a patient with end‐stage renal disease who presents with a thigh pain; however, this condition is usually characterized by areas of ischemic necrosis that develop in the dermis and/or subcutaneous fat and infrequently involve muscles. The absence of the painful subcutaneous nodules typical of calciphylaxis makes it an unlikely diagnosis.

A diagnosis of diabetic myonecrosis was made. Antibiotics were discontinued, and the patient was treated symptomatically. His symptoms improved during the next few days. The patient was discharged from the hospital, and conservative management with bed rest and analgesics for 4 weeks was prescribed. Four months later, however, the patient returned with similar symptoms in the contralateral thigh. The patient was diagnosed with recurrent diabetic myonecrosis by MRI and muscle biopsy findings. Conservative management was advised, and the patient became pain‐free in a few weeks.

DISCUSSION

Diabetic myonecrosis (also known as diabetic muscle infarction) is a rare disorder initially described in 1965[1] that typically presents spontaneously as an acute, localized, severely painful swelling that limits the mobility of the affected extremity, usually without systemic signs of infection. It affects the thighs in 83% of patients and the calves in 17% of patients.[2, 3] Bilateral involvement, which is usually asynchronous, occurs in one‐third of patients.[4] The upper limbs are rarely involved. Diabetic myonecrosis affects patients who have a relatively longstanding history of diabetes. It is commonly associated with the microvascular complications of diabetes, including nephropathy (80% of patients), retinopathy (60% of patents), and/or neuropathy (64% of patients).[3, 5] The pathogenesis of diabetic myonecrosis is unclear, but the disease is likely due to a diffuse microangiopathy and atherosclerosis.[2, 5] Some authors have suggested that abnormalities in the clotting or fibrinolytic pathways play a role in the etiology of the disorder.[6]

Clinical and MRI findings can be used to make the diagnosis with reasonable certainty.[3, 5] Although both ultrasonography and MRI have been used to assess patients with diabetic myonecrosis, MRI with intravenous contrast enhancement appears to be the most useful diagnostic technique. It demonstrates extensive edema within the muscle(s), muscle enlargement, subcutaneous and interfascial edema, a patchwork pattern of involvement, and a high signal intensity on T2‐weighted images.[4, 7] Gadolinium enhancement may reveal an enhanced margin of the infarcted muscle with a central nonenhancing area of necrotic tissue.[4, 5] Muscle biopsy is not typically indicated because it may prolong recovery time and lead to infections.[8, 9, 10, 11] When performed, however, muscle biopsy reveals ischemic muscle fibers in different stages of degeneration and regeneration, with areas of necrosis and edema. Occlusion of arterioles and capillaries by fibrin could also be seen.[1] Although the patient underwent a muscle biopsy because infection could not be excluded definitively on clinical grounds, we believe that repeating the biopsy 4 months later was inappropriate.

Diabetic myonecrosis should be considered in a diabetic patient who presents with severe localized muscle pain and swelling of an extremity, especially if the clinical features favoring infection are absent. The differential diagnosis should include infection (eg, clostridial myonecrosis, myositis, cellulitis, abscess, necrotizing fasciitis, osteomyelitis), trauma (eg, hematoma, muscle rupture, myositis ossificans), peripheral neuropathy (particularly lumbosacral plexopathy), vascular disorders (deep vein thrombosis, and compartment syndrome), tumors, inflammatory muscle diseases, and drug‐related myositis.

No evidence‐based recommendations regarding the management of diabetic myonecrosis are available, although the findings of one retrospective analysis support conservative management with bed rest, leg elevation, and analgesics.[12] Physiotherapy may cause the condition to worsen,[13, 14] but routine daily activity, although often painful, is not harmful.[14] Some authors suggest a cautious use of antiplatelet or anti‐inflammatory medications.[12] We would also recommend achieving good glycemic control during the illness. Owing to the rarity of the disease, however, no studies have definitively shown that this hastens recovery or prevents recurrent diabetic myonecrosis. Surgery may prolong the recovery period; one study found that the recovery period of patients with diabetic myonecrosis who underwent surgery was longer than that of those who were treated conservatively (13 weeks vs 5.5 weeks).[12] Patients with diabetic myonecrosis have a good short‐term prognosis. Longer‐term, however, they have a poor prognosis; their recurrence rate is as high as 40%, and their 2‐year mortality rate is 10%, even after one episode of the disease. Death in these patients is mainly due to macrovascular events.[12]

TEACHING POINTS

1. Diabetic myonecrosis is a rare complication of longstanding and poorly controlled diabetes. It usually presents with acute localized muscular pain in the lower extremities.
2. Although a definitive diagnosis of diabetic myonecrosis is histopathologic, a clinical diagnosis can be made with reasonable certainty for patients with compatible MRI findings and no clinical or laboratory features suggesting infection.
3. Conservative management with bed rest, analgesics, and antiplatelets is recommended. Surgery should be avoided, as it may prolong recovery.

Disclosure

Nothing to report.

In‐hospital cardiac arrest (IHCA) research often relies on the first documented cardiac rhythm (FDR) on resuscitation records at the time of cardiopulmonary resuscitation (CPR) initiation as a surrogate for arrest etiology.[1] Over 1000 hospitals report the FDR and associated cardiac arrest data to national registries annually.[2, 3] These data are subsequently used to report national IHCA epidemiology, as well as to develop and refine guidelines for in‐hospital resuscitation.[4]

Suspecting that the FDR might represent the later stage of a progressive cardiopulmonary process rather than a sudden dysrhythmia, we sought to compare the first rhythm documented on resuscitation records at the time of CPR initiation with the telemetry rhythm at the time of the code blue call. We hypothesized that the agreement between FDR and telemetry rhythm would be <80% beyond that predicted by chance (kappa<0.8).[5]

METHODS

RESULTS

During the study period, there were 135 code blue calls for urgent assistance among telemetry‐monitored non‐ICU patients. Of the 135 calls, we excluded 4 events (3%) that did not meet the definition of IHCA, 9 events (7%) with missing or uninterpretable data, and 53 events (39%) with unobtainable data due to automatic purging from the telemetry server. Therefore, 69 events in 69 different patients remained for analysis. Twelve of the 69 included arrests that occurred at Wilmington Hospital and 57 at Christiana Hospital. The characteristics of the patients are shown in Table 2.

Of the 69 arrests, we used the telemetry rhythm at minute 0 in 42 patients (61%), minute 1 in 22 patients (32%), and minute 2 in 5 patients (7%). Agreement between telemetry and FDR was 65% (kappa=0.37, 95% confidence interval: 0.17‐0.56) (Table 3). Agreement did not vary significantly by sex, race, hospital, weekday, time of day, or minute used in the analysis. Agreement was not associated with survival to hospital discharge.

Of the 69 IHCA events, the FDRs vs telemetry rhythms at the time of IHCA were: asystole 17% vs 7%, VTA 25% vs 31%, and other organized rhythms 58% vs 62%. Among the 12 events with FDR recorded as asystole, telemetry at the time of the code call was asystole in 3 (25%), VTA in 1 (8%), and other organized rhythms in 8 (67%). Among the 17 events with FDR recorded as VTA, telemetry at the time of the code call was VTA in 12 (71%) and other organized rhythms in 5 (29%). Among the 40 events with FDR recorded as other organized rhythms, telemetry at the time of the code call was asystole in 2 (5%), VTA in 8 (20%), and other organized rhythms in 30 (75%). Among the 8 patients with VTA on telemetry and other organized rhythms on the resuscitation record, the other organized rhythms were documented as PEA (n=6), sinus (n=1), and bradycardia (n=1). Of the 12 patients with VTA on telemetry and on the resuscitation record, 8 (67%) had ventricular tachycardia on telemetry. Four of the 8 (50%) who had ventricular tachycardia on telemetry had deteriorated into ventricular fibrillation by the time the FDR was recorded. Of the 4 who had ventricular fibrillation on telemetry, all had ventricular fibrillation as the FDR on the resuscitation record.

DISCUSSION

These results establish that FDRs often differ from the telemetry rhythms at the time of the code blue call. This is important because national registries such as the American Heart Association's Get with the GuidelinesResuscitation[2] database use the FDR as a surrogate for arrest etiology, and use their findings to report national IHCA outcomes as well as to develop and refine evidence‐based guidelines for in‐hospital resuscitation. Our findings suggest that using the FDR may be an oversimplification of the complex progression of cardiac rhythms that occurs in the periarrest period. Adding preceding telemetry rhythms to the data elements collected may shed additional light on etiology. Furthermore, our results demonstrate that, among adults with VTA or asystole documented upon arrival of the code blue team, other organized rhythms are often present at the time the staff recognized a life‐threatening condition and called for immediate assistance. This suggests that the VTA and asystole FDRs may represent the later stages of progressive cardiopulmonary processes. This is in contrast to out‐of‐hospital cardiac arrests typically attributed to sudden catastrophic dysrhythmias that often progress to asystole unless rapidly defibrillated.[6, 7, 8] Out‐of‐hospital and in‐hospital arrests are likely different (but overlapping) entities that might benefit from different resuscitation strategies.[9, 10] We hypothesize that, for a subset of these patients, progressive respiratory insufficiency and circulatory shockconditions classically associated more strongly with pediatric than adult IHCAmay have been directly responsible for the event.[1] If future research supports the concept that progressive respiratory insufficiency and circulatory shock are responsible for more adult IHCA than previously recognized, more robust monitoring may be indicated for a larger subset of adult patients hospitalized on general wards. This could include pulse oximetry (wave form can be a surrogate for perfusion), respiratory rate, and/or end‐tidal CO₂ monitoring. In addition, if future research confirms that there is a greater distinction between in‐hospital and out‐of‐hospital cardiac arrest etiology, the expert panels that develop resuscitation guidelines should consider including setting of resuscitation as a branch point in future algorithms.

Our study had several limitations. First, the sample size was small due to uninterpretable rhythm strips, and for 39% of the total code events, the telemetry data had already been purged from the system by the time research staff attempted to retrieve it. Although we do not believe that there was any systematic bias to the data analyzed, the possibility cannot be completely excluded. Second, we were constrained by the inability to retrospectively ascertain the presence of pulses to determine if an organized rhythm identified on telemetry tracings was PEA. Thus, we categorized rhythms into large groups. Although this limited the granularity of the rhythm groups, it was a conservative approach that likely biased toward agreement (the opposite direction of our hypothesis). Third, the lack of perfect time synchronization between the telemetry system, wall clocks in the hospital, and wrist watches that may be referenced when documenting resuscitative efforts on the resuscitation record means that the rhythms we used may have reflected physiology after interventions had already commenced. Thus, in some situations, minute 1, 2, or earlier minutes may more accurately reflect the preintervention rhythm. Highly accurate time synchronization should be a central component of future prospective work in this area.

CONCLUSIONS

The FDR had only fair agreement with the telemetry rhythm at the time of the code blue call. Among those with VTA or asystole documented on CPR initiation, telemetry often revealed other organized rhythms present at the time hospital staff recognized a life‐threatening condition. In contrast to out‐of‐hospital cardiac arrest, FDR of asystole was only rarely preceded by VTA, and FDR of VTA was often preceded by an organized rhythm.[8, 11] Future studies should examine antecedent rhythms in combination with respiratory and perfusion status to more precisely determine arrest etiology.

Disclosures

Dr. Zubrow had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The authors report no conflicts of interest.

In‐hospital cardiac arrest (IHCA) research often relies on the first documented cardiac rhythm (FDR) on resuscitation records at the time of cardiopulmonary resuscitation (CPR) initiation as a surrogate for arrest etiology.[1] Over 1000 hospitals report the FDR and associated cardiac arrest data to national registries annually.[2, 3] These data are subsequently used to report national IHCA epidemiology, as well as to develop and refine guidelines for in‐hospital resuscitation.[4]

Suspecting that the FDR might represent the later stage of a progressive cardiopulmonary process rather than a sudden dysrhythmia, we sought to compare the first rhythm documented on resuscitation records at the time of CPR initiation with the telemetry rhythm at the time of the code blue call. We hypothesized that the agreement between FDR and telemetry rhythm would be <80% beyond that predicted by chance (kappa<0.8).[5]

METHODS

RESULTS

During the study period, there were 135 code blue calls for urgent assistance among telemetry‐monitored non‐ICU patients. Of the 135 calls, we excluded 4 events (3%) that did not meet the definition of IHCA, 9 events (7%) with missing or uninterpretable data, and 53 events (39%) with unobtainable data due to automatic purging from the telemetry server. Therefore, 69 events in 69 different patients remained for analysis. Twelve of the 69 included arrests that occurred at Wilmington Hospital and 57 at Christiana Hospital. The characteristics of the patients are shown in Table 2.

Of the 69 arrests, we used the telemetry rhythm at minute 0 in 42 patients (61%), minute 1 in 22 patients (32%), and minute 2 in 5 patients (7%). Agreement between telemetry and FDR was 65% (kappa=0.37, 95% confidence interval: 0.17‐0.56) (Table 3). Agreement did not vary significantly by sex, race, hospital, weekday, time of day, or minute used in the analysis. Agreement was not associated with survival to hospital discharge.

Of the 69 IHCA events, the FDRs vs telemetry rhythms at the time of IHCA were: asystole 17% vs 7%, VTA 25% vs 31%, and other organized rhythms 58% vs 62%. Among the 12 events with FDR recorded as asystole, telemetry at the time of the code call was asystole in 3 (25%), VTA in 1 (8%), and other organized rhythms in 8 (67%). Among the 17 events with FDR recorded as VTA, telemetry at the time of the code call was VTA in 12 (71%) and other organized rhythms in 5 (29%). Among the 40 events with FDR recorded as other organized rhythms, telemetry at the time of the code call was asystole in 2 (5%), VTA in 8 (20%), and other organized rhythms in 30 (75%). Among the 8 patients with VTA on telemetry and other organized rhythms on the resuscitation record, the other organized rhythms were documented as PEA (n=6), sinus (n=1), and bradycardia (n=1). Of the 12 patients with VTA on telemetry and on the resuscitation record, 8 (67%) had ventricular tachycardia on telemetry. Four of the 8 (50%) who had ventricular tachycardia on telemetry had deteriorated into ventricular fibrillation by the time the FDR was recorded. Of the 4 who had ventricular fibrillation on telemetry, all had ventricular fibrillation as the FDR on the resuscitation record.

DISCUSSION

These results establish that FDRs often differ from the telemetry rhythms at the time of the code blue call. This is important because national registries such as the American Heart Association's Get with the GuidelinesResuscitation[2] database use the FDR as a surrogate for arrest etiology, and use their findings to report national IHCA outcomes as well as to develop and refine evidence‐based guidelines for in‐hospital resuscitation. Our findings suggest that using the FDR may be an oversimplification of the complex progression of cardiac rhythms that occurs in the periarrest period. Adding preceding telemetry rhythms to the data elements collected may shed additional light on etiology. Furthermore, our results demonstrate that, among adults with VTA or asystole documented upon arrival of the code blue team, other organized rhythms are often present at the time the staff recognized a life‐threatening condition and called for immediate assistance. This suggests that the VTA and asystole FDRs may represent the later stages of progressive cardiopulmonary processes. This is in contrast to out‐of‐hospital cardiac arrests typically attributed to sudden catastrophic dysrhythmias that often progress to asystole unless rapidly defibrillated.[6, 7, 8] Out‐of‐hospital and in‐hospital arrests are likely different (but overlapping) entities that might benefit from different resuscitation strategies.[9, 10] We hypothesize that, for a subset of these patients, progressive respiratory insufficiency and circulatory shockconditions classically associated more strongly with pediatric than adult IHCAmay have been directly responsible for the event.[1] If future research supports the concept that progressive respiratory insufficiency and circulatory shock are responsible for more adult IHCA than previously recognized, more robust monitoring may be indicated for a larger subset of adult patients hospitalized on general wards. This could include pulse oximetry (wave form can be a surrogate for perfusion), respiratory rate, and/or end‐tidal CO₂ monitoring. In addition, if future research confirms that there is a greater distinction between in‐hospital and out‐of‐hospital cardiac arrest etiology, the expert panels that develop resuscitation guidelines should consider including setting of resuscitation as a branch point in future algorithms.

Our study had several limitations. First, the sample size was small due to uninterpretable rhythm strips, and for 39% of the total code events, the telemetry data had already been purged from the system by the time research staff attempted to retrieve it. Although we do not believe that there was any systematic bias to the data analyzed, the possibility cannot be completely excluded. Second, we were constrained by the inability to retrospectively ascertain the presence of pulses to determine if an organized rhythm identified on telemetry tracings was PEA. Thus, we categorized rhythms into large groups. Although this limited the granularity of the rhythm groups, it was a conservative approach that likely biased toward agreement (the opposite direction of our hypothesis). Third, the lack of perfect time synchronization between the telemetry system, wall clocks in the hospital, and wrist watches that may be referenced when documenting resuscitative efforts on the resuscitation record means that the rhythms we used may have reflected physiology after interventions had already commenced. Thus, in some situations, minute 1, 2, or earlier minutes may more accurately reflect the preintervention rhythm. Highly accurate time synchronization should be a central component of future prospective work in this area.

CONCLUSIONS

The FDR had only fair agreement with the telemetry rhythm at the time of the code blue call. Among those with VTA or asystole documented on CPR initiation, telemetry often revealed other organized rhythms present at the time hospital staff recognized a life‐threatening condition. In contrast to out‐of‐hospital cardiac arrest, FDR of asystole was only rarely preceded by VTA, and FDR of VTA was often preceded by an organized rhythm.[8, 11] Future studies should examine antecedent rhythms in combination with respiratory and perfusion status to more precisely determine arrest etiology.

Disclosures

Dr. Zubrow had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The authors report no conflicts of interest.

In‐hospital cardiac arrest (IHCA) research often relies on the first documented cardiac rhythm (FDR) on resuscitation records at the time of cardiopulmonary resuscitation (CPR) initiation as a surrogate for arrest etiology.[1] Over 1000 hospitals report the FDR and associated cardiac arrest data to national registries annually.[2, 3] These data are subsequently used to report national IHCA epidemiology, as well as to develop and refine guidelines for in‐hospital resuscitation.[4]

Suspecting that the FDR might represent the later stage of a progressive cardiopulmonary process rather than a sudden dysrhythmia, we sought to compare the first rhythm documented on resuscitation records at the time of CPR initiation with the telemetry rhythm at the time of the code blue call. We hypothesized that the agreement between FDR and telemetry rhythm would be <80% beyond that predicted by chance (kappa<0.8).[5]

METHODS

RESULTS

During the study period, there were 135 code blue calls for urgent assistance among telemetry‐monitored non‐ICU patients. Of the 135 calls, we excluded 4 events (3%) that did not meet the definition of IHCA, 9 events (7%) with missing or uninterpretable data, and 53 events (39%) with unobtainable data due to automatic purging from the telemetry server. Therefore, 69 events in 69 different patients remained for analysis. Twelve of the 69 included arrests that occurred at Wilmington Hospital and 57 at Christiana Hospital. The characteristics of the patients are shown in Table 2.

Of the 69 arrests, we used the telemetry rhythm at minute 0 in 42 patients (61%), minute 1 in 22 patients (32%), and minute 2 in 5 patients (7%). Agreement between telemetry and FDR was 65% (kappa=0.37, 95% confidence interval: 0.17‐0.56) (Table 3). Agreement did not vary significantly by sex, race, hospital, weekday, time of day, or minute used in the analysis. Agreement was not associated with survival to hospital discharge.

Of the 69 IHCA events, the FDRs vs telemetry rhythms at the time of IHCA were: asystole 17% vs 7%, VTA 25% vs 31%, and other organized rhythms 58% vs 62%. Among the 12 events with FDR recorded as asystole, telemetry at the time of the code call was asystole in 3 (25%), VTA in 1 (8%), and other organized rhythms in 8 (67%). Among the 17 events with FDR recorded as VTA, telemetry at the time of the code call was VTA in 12 (71%) and other organized rhythms in 5 (29%). Among the 40 events with FDR recorded as other organized rhythms, telemetry at the time of the code call was asystole in 2 (5%), VTA in 8 (20%), and other organized rhythms in 30 (75%). Among the 8 patients with VTA on telemetry and other organized rhythms on the resuscitation record, the other organized rhythms were documented as PEA (n=6), sinus (n=1), and bradycardia (n=1). Of the 12 patients with VTA on telemetry and on the resuscitation record, 8 (67%) had ventricular tachycardia on telemetry. Four of the 8 (50%) who had ventricular tachycardia on telemetry had deteriorated into ventricular fibrillation by the time the FDR was recorded. Of the 4 who had ventricular fibrillation on telemetry, all had ventricular fibrillation as the FDR on the resuscitation record.

DISCUSSION

These results establish that FDRs often differ from the telemetry rhythms at the time of the code blue call. This is important because national registries such as the American Heart Association's Get with the GuidelinesResuscitation[2] database use the FDR as a surrogate for arrest etiology, and use their findings to report national IHCA outcomes as well as to develop and refine evidence‐based guidelines for in‐hospital resuscitation. Our findings suggest that using the FDR may be an oversimplification of the complex progression of cardiac rhythms that occurs in the periarrest period. Adding preceding telemetry rhythms to the data elements collected may shed additional light on etiology. Furthermore, our results demonstrate that, among adults with VTA or asystole documented upon arrival of the code blue team, other organized rhythms are often present at the time the staff recognized a life‐threatening condition and called for immediate assistance. This suggests that the VTA and asystole FDRs may represent the later stages of progressive cardiopulmonary processes. This is in contrast to out‐of‐hospital cardiac arrests typically attributed to sudden catastrophic dysrhythmias that often progress to asystole unless rapidly defibrillated.[6, 7, 8] Out‐of‐hospital and in‐hospital arrests are likely different (but overlapping) entities that might benefit from different resuscitation strategies.[9, 10] We hypothesize that, for a subset of these patients, progressive respiratory insufficiency and circulatory shockconditions classically associated more strongly with pediatric than adult IHCAmay have been directly responsible for the event.[1] If future research supports the concept that progressive respiratory insufficiency and circulatory shock are responsible for more adult IHCA than previously recognized, more robust monitoring may be indicated for a larger subset of adult patients hospitalized on general wards. This could include pulse oximetry (wave form can be a surrogate for perfusion), respiratory rate, and/or end‐tidal CO₂ monitoring. In addition, if future research confirms that there is a greater distinction between in‐hospital and out‐of‐hospital cardiac arrest etiology, the expert panels that develop resuscitation guidelines should consider including setting of resuscitation as a branch point in future algorithms.

Our study had several limitations. First, the sample size was small due to uninterpretable rhythm strips, and for 39% of the total code events, the telemetry data had already been purged from the system by the time research staff attempted to retrieve it. Although we do not believe that there was any systematic bias to the data analyzed, the possibility cannot be completely excluded. Second, we were constrained by the inability to retrospectively ascertain the presence of pulses to determine if an organized rhythm identified on telemetry tracings was PEA. Thus, we categorized rhythms into large groups. Although this limited the granularity of the rhythm groups, it was a conservative approach that likely biased toward agreement (the opposite direction of our hypothesis). Third, the lack of perfect time synchronization between the telemetry system, wall clocks in the hospital, and wrist watches that may be referenced when documenting resuscitative efforts on the resuscitation record means that the rhythms we used may have reflected physiology after interventions had already commenced. Thus, in some situations, minute 1, 2, or earlier minutes may more accurately reflect the preintervention rhythm. Highly accurate time synchronization should be a central component of future prospective work in this area.

CONCLUSIONS

The FDR had only fair agreement with the telemetry rhythm at the time of the code blue call. Among those with VTA or asystole documented on CPR initiation, telemetry often revealed other organized rhythms present at the time hospital staff recognized a life‐threatening condition. In contrast to out‐of‐hospital cardiac arrest, FDR of asystole was only rarely preceded by VTA, and FDR of VTA was often preceded by an organized rhythm.[8, 11] Future studies should examine antecedent rhythms in combination with respiratory and perfusion status to more precisely determine arrest etiology.

Disclosures

Dr. Zubrow had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The authors report no conflicts of interest.

Thousands of hospitals have recently implemented rapid response systems (RRSs), attempting to reduce mortality outside of intensive care units (ICUs).[1, 2] These systems have 2 clinical components, a response (efferent) arm and an identification (afferent) arm.[3] The response arm is usually composed of a medical emergency team (MET) that responds to calls for urgent assistance. The identification arm includes tools to help clinicians recognize patients who require assistance from the MET. In many hospitals, the identification arm includes an early warning score (EWS). In pediatric patients, EWSs assign point values to vital signs that fall outside of age‐based ranges, among other clinical observations. They then generate a total score intended to help clinicians identify patients exhibiting early signs of deterioration.[4, 5, 6, 7, 8, 9, 10, 11]

When experimentally applied to vital sign datasets, the test characteristics of pediatric EWSs in detecting clinical deterioration are highly variable across studies, with major tradeoffs between sensitivity, specificity, and predictive values that differ by outcome, score, and cut‐point (Table 1). This reflects the difficulty of identifying deteriorating patients using only objective measures. However, in real‐world settings, EWSs are used by clinicians in conjunction with their clinical judgment. We hypothesized that EWSs have benefits that extend beyond their ability to predict deterioration, and thus have value not demonstrated by test characteristics alone. In order to further explore this issue, we aimed to qualitatively evaluate mechanisms beyond their statistical ability to predict deterioration by which physicians and nurses use EWSs to support their decision making.

METHODS

RESULTS

DISCUSSION

This study is the first to analyze viewpoints on the mechanisms by which EWSs impact decision making among physicians and nurses who had recently experienced score failures. Our study, performed in a children's hospital, builds upon the findings of related studies performed in hospitals that care primarily for adults.[23, 24, 25, 26, 27, 28] Andrews and Waterman found that nurses consider the utility of EWSs to extend beyond detecting deterioration by providing quantifiable evidence, packaged in the form of a score that improves communication between nurses and physicians.[23] Mackintosh and colleagues found that a RRS that included an EWS helped to formalize the way nurses and physicians understand deterioration, enable them to overcome hierarchical boundaries through structured discussions, and empower them to call for help.[24] In a quasi‐experimental study, McDonnell and colleagues found that an EWS improved self‐assessed knowledge, skills, and confidence of nursing staff to detect and manage deteriorating patients.[25] In addition, we describe novel findings, including the use of EWS parameters as reference ranges independent of the score, and specific situations when the EWS fails to support decision making. The weaknesses we identified could be used to drive EWS optimization for low‐risk patients who are stable as well as higher‐risk patients with abnormal baseline physiology and those at risk of neurologic deterioration.

This study has several limitations. Although the interviewers were not involved in RRS operations, it is possible that social desirability bias influenced responses. Next, we identified a knowledge gap among surgeons, and they contributed minimally to our findings. This is most likely because (1) surgical patients deteriorate on the wards less often than medical patients in our hospital, so surgeons are rarely presented with EWSs; (2) surgeons spend less time on the wards compared with medical physicians; and (3) surgical residents rotate in short blocks interspersed with rotations at other hospitals and may be less engaged in hospital safety initiatives.

CONCLUSIONS

Although EWSs perform only marginally well as statistical tools to predict clinical deterioration, nurses and physicians who recently experienced score failures described substantial benefits in using them to help identify deteriorating patients and transcend barriers to escalation of care by serving as objective communication tools. Combining an EWS with a clinician's judgment may result in a system better equipped to respond to deterioration than previous EWS studies focused on their test characteristics alone suggest. Future research should seek to compare and prospectively evaluate the clinical effectiveness of EWSs in real‐world settings.

Thousands of hospitals have recently implemented rapid response systems (RRSs), attempting to reduce mortality outside of intensive care units (ICUs).[1, 2] These systems have 2 clinical components, a response (efferent) arm and an identification (afferent) arm.[3] The response arm is usually composed of a medical emergency team (MET) that responds to calls for urgent assistance. The identification arm includes tools to help clinicians recognize patients who require assistance from the MET. In many hospitals, the identification arm includes an early warning score (EWS). In pediatric patients, EWSs assign point values to vital signs that fall outside of age‐based ranges, among other clinical observations. They then generate a total score intended to help clinicians identify patients exhibiting early signs of deterioration.[4, 5, 6, 7, 8, 9, 10, 11]

When experimentally applied to vital sign datasets, the test characteristics of pediatric EWSs in detecting clinical deterioration are highly variable across studies, with major tradeoffs between sensitivity, specificity, and predictive values that differ by outcome, score, and cut‐point (Table 1). This reflects the difficulty of identifying deteriorating patients using only objective measures. However, in real‐world settings, EWSs are used by clinicians in conjunction with their clinical judgment. We hypothesized that EWSs have benefits that extend beyond their ability to predict deterioration, and thus have value not demonstrated by test characteristics alone. In order to further explore this issue, we aimed to qualitatively evaluate mechanisms beyond their statistical ability to predict deterioration by which physicians and nurses use EWSs to support their decision making.

METHODS

RESULTS

DISCUSSION

This study is the first to analyze viewpoints on the mechanisms by which EWSs impact decision making among physicians and nurses who had recently experienced score failures. Our study, performed in a children's hospital, builds upon the findings of related studies performed in hospitals that care primarily for adults.[23, 24, 25, 26, 27, 28] Andrews and Waterman found that nurses consider the utility of EWSs to extend beyond detecting deterioration by providing quantifiable evidence, packaged in the form of a score that improves communication between nurses and physicians.[23] Mackintosh and colleagues found that a RRS that included an EWS helped to formalize the way nurses and physicians understand deterioration, enable them to overcome hierarchical boundaries through structured discussions, and empower them to call for help.[24] In a quasi‐experimental study, McDonnell and colleagues found that an EWS improved self‐assessed knowledge, skills, and confidence of nursing staff to detect and manage deteriorating patients.[25] In addition, we describe novel findings, including the use of EWS parameters as reference ranges independent of the score, and specific situations when the EWS fails to support decision making. The weaknesses we identified could be used to drive EWS optimization for low‐risk patients who are stable as well as higher‐risk patients with abnormal baseline physiology and those at risk of neurologic deterioration.

This study has several limitations. Although the interviewers were not involved in RRS operations, it is possible that social desirability bias influenced responses. Next, we identified a knowledge gap among surgeons, and they contributed minimally to our findings. This is most likely because (1) surgical patients deteriorate on the wards less often than medical patients in our hospital, so surgeons are rarely presented with EWSs; (2) surgeons spend less time on the wards compared with medical physicians; and (3) surgical residents rotate in short blocks interspersed with rotations at other hospitals and may be less engaged in hospital safety initiatives.

CONCLUSIONS

Although EWSs perform only marginally well as statistical tools to predict clinical deterioration, nurses and physicians who recently experienced score failures described substantial benefits in using them to help identify deteriorating patients and transcend barriers to escalation of care by serving as objective communication tools. Combining an EWS with a clinician's judgment may result in a system better equipped to respond to deterioration than previous EWS studies focused on their test characteristics alone suggest. Future research should seek to compare and prospectively evaluate the clinical effectiveness of EWSs in real‐world settings.

Thousands of hospitals have recently implemented rapid response systems (RRSs), attempting to reduce mortality outside of intensive care units (ICUs).[1, 2] These systems have 2 clinical components, a response (efferent) arm and an identification (afferent) arm.[3] The response arm is usually composed of a medical emergency team (MET) that responds to calls for urgent assistance. The identification arm includes tools to help clinicians recognize patients who require assistance from the MET. In many hospitals, the identification arm includes an early warning score (EWS). In pediatric patients, EWSs assign point values to vital signs that fall outside of age‐based ranges, among other clinical observations. They then generate a total score intended to help clinicians identify patients exhibiting early signs of deterioration.[4, 5, 6, 7, 8, 9, 10, 11]

When experimentally applied to vital sign datasets, the test characteristics of pediatric EWSs in detecting clinical deterioration are highly variable across studies, with major tradeoffs between sensitivity, specificity, and predictive values that differ by outcome, score, and cut‐point (Table 1). This reflects the difficulty of identifying deteriorating patients using only objective measures. However, in real‐world settings, EWSs are used by clinicians in conjunction with their clinical judgment. We hypothesized that EWSs have benefits that extend beyond their ability to predict deterioration, and thus have value not demonstrated by test characteristics alone. In order to further explore this issue, we aimed to qualitatively evaluate mechanisms beyond their statistical ability to predict deterioration by which physicians and nurses use EWSs to support their decision making.