Dirk R. Larson, MS

As the population ages, hip fractures and heart failure increase in prevalence.1, 2 Heart failure prevalence is also increasing in hospitalized patients.3 Indeed, hospitalizations involving heart failure as an active issue tripled in the last 30 years.4 Heart failure has been associated with an increased risk for hip fracture,5, 6 and previous studies report a 6%20% prevalence of preoperative heart failure in hip fracture patients.710 While exacerbation of heart failure increases the mortality risk in patients admitted for hip fractures,8 the incidence of new heart failure, as well as the preoperative factors that predict postoperative heart failure in this patient population remain unclear.

American College of Cardiology/American Heart Association (ACC/AHA) perioperative guidelines identify orthopedic surgeries, including hip fracture repair, as intermediate risk procedures.11 Compared to other intermediate risk operations, however, postoperative outcomes following hip fracture repair differ significantly.1216 Overall mortality in hip fracture patients has been reported at 29% at one year,8 with the excess mortality from hip fracture alone at nearly 20%.10, 13 However, the exact factors that contribute to this excess mortality, particularly with regard to heart failure, remain unclear.

To examine the preoperative prevalence, subsequent incidence, and predictors of heart failure in patients undergoing hip fracture repair operations, this study used an established, population‐based database to compare the postoperative consequences in hip fracture repair patients with and without preexisting heart failure. We hypothesized that preoperative heart failure worsens postoperative outcomes in hip fracture patients.

METHODS

RESULTS

From among 1327 potential hip fracture repairs, we excluded 115 cases involving multiple injuries or operations (19), pathological fractures (20), in‐hospital fractures (3), or an operation >72 hours after the initial fracture (5). Three patients under 65 years of age were also excluded, as were cases with missing information (9) or cases managed nonoperatively (56). The final analysis included 1212 surgical cases in 1116 subjects. No subjects were lost to surveillance for 1 year following their hip fracture repair.

Table 1 summarizes the baseline characteristics of the study population. The overall prevalence of preoperative heart failure was 27.0% (327 of 1212). Those with preoperative heart failure were older, heavier, more likely male and white, and less likely to live independently preoperatively. They were also more likely to suffer from preexisting cardiovascular comorbidities.

Table 1 also summarizes the main outcome characteristics of the study population. Those with preoperative heart failure had longer mean lengths of stay (LOS), were more often discharged to a skilled facility, and demonstrated higher inpatient mortality rates.

Table 2 summarizes the outcomes associated with preoperative heart failure. The overall rate of postoperative heart failure was 6.7% within 7 postoperative days and 21.3% within 1 postoperative year. Postoperative heart failure was significantly more common among those with preoperative heart failure (hazard ratio [HR], 3.0; 95% confidence interval [CI], 2.3 to 3.9; P < 0.001). Among those without preoperative heart failure, rates of postoperative incident heart failure were 4.8% at 7 days and 15.0% at 1 year. Compared to patients without preoperative heart failure, those with preoperative heart failure demonstrated higher one year mortality rates and higher rates of postoperative heart failure at 7 days and 1 year.

Figure 1 displays the outcomes to 1 year of surveillance. Rates of postoperative heart failure and postoperative mortality were consistently higher among those with, versus without, preoperative heart failure. Figure 2 displays similar data stratified by gender. Postoperative heart failure rates did not differ significantly between genders (HR, 1.0; 95% CI, 0.8 to 1.4), but postoperative mortality rates were significantly higher among males than females (HR, 1.9; 95% CI, 1.5 to 2.5; P < 0.001).

Figure 1

Figure 2

Figure 3 displays survival rates to 1 year based on the occurrence of incident or recurrent heart failure within the first 7 postoperative days. Survival rates were lowest among patients with recurrent heart failure in the first 7 postoperative days and highest among those with no preoperative or postoperative heart failure. Subjects with incident heart failure in the first postoperative week, and those with preoperative heart failure who did not suffer a recurrence, demonstrated intermediate survival rates (P < 0.001 for trend across all four groups).

Figure 3

DISCUSSION

This population‐based study found that heart failure represents a highly prevalent condition in elderly patients undergoing hip fracture repairs. It demonstrates that those with preoperative heart failure typically suffer from more cardiovascular comorbidities and carry a higher risk of postoperative heart failure and postoperative mortality.

While many studies have focused on the epidemiology of hip fractures,21 population‐based data on cardiac complications following hip fracture repair are significantly less common. The ACC/AHA preoperative cardiac evaluation guidelines classify orthopedic procedures, including hip fracture repair, as intermediate risk.11 Consequently, some may assume that all orthopedic patients will have a mortality rate less than 5%. Indeed, the 30‐day postoperative mortality rate published from our institution's Total Joint Registry was 0.6% following elective total hip arthroplasty.22 However, the present study demonstrates that current ACC/AHA preoperative cardiac evaluation guidelines may not apply to the population of frail patients undergoing hip fracture repair. Particularly among those who experience new heart failure within the first seven days following surgery, outcomes are substantially worse than the ACC/AHA perioperative guidelines may suggest.11

Preoperative heart failure has been associated with adverse risk for postoperative mortality after hip fracture.9, 10, 12 However, these studies did not report heart failure as a complication of hip fracture repair. A prospective cohort study of 2448 hip fracture patients at an academic hospital in Great Britain found a 5% rate of inpatient heart failure as a postoperative complication.23 The hazard ratio for one‐year mortality was 11.3 with postoperative heart failure.23 However, the British study did not distinguish heart failure from other cardiovascular diseases as a preoperative comorbidity or stratify the risk for postoperative mortality by preoperative heart failure status.23 Our findings add to previous literature by measuring heart failure as a specific complication of hip fracture repair and examining the association of preoperative heart failure with postoperative heart failure and mortality.

Length of stay after hip fracture repair varies in the literature, but previous work has not clearly associated heart failure with length of hospitalization in the setting of hip fracture repair.24, 25 Our study found a significantly higher mean length of stay among those with preoperative heart failure. This adds to previous work by delineating an association between heart failure and increased length of stay after hip fracture repair.

We found a higher rate of postoperative mortality among men compared to women. Rates of postoperative heart failure, however, were more similar (Figure 2). Previous studies have found a consistently higher mortality rate among men versus women after hip fracture.9, 23, 2529 Possible explanations for these findings include the overall increased burden of cardiovascular disease among men, lower treatment rates of osteoporosis in men,30 and increased susceptibility to other postoperative complications, such as infection.25

The findings of this study carry important clinical implications for the perioperative care of hip fracture patients with, or at risk for, heart failure. They suggest that current risk stratification guidelines classifying orthopedic operations as intermediate risk procedures do not reflect the high risk for morbidity that hip fracture patients face.11 The association of heart failure with adverse outcomes implies the need for heightened surveillance in the perioperative period, particularly with regard to volume status and medication reconciliation. Hip fracture patients and their families must be counseled about the ramifications of perioperative heart failure, including higher rates of postoperative heart failure, longer hospitalizations, and ultimate mortality.

This research carries several limitations and remains subject to biases inherent in retrospective cohort studies. The reported effects of heart failure on outcomes after hip fracture repair may be due to confounding from age, functional status, and other comorbidities. We attempted to minimize sampling bias through complete enumeration of hip fracture surgeries among Olmsted County residents. Completeness of follow‐up (100% at one year) was possible given the availability of documentation of all inpatient and outpatient medical care in the community.17 We used objectively defined outcomes to minimize measurement bias. Applicability to a more diverse population may be limited because >95% of the research population was from a single, predominantly white community. However, prior studies have documented that hip fracture incidence rates31 and socioeconomic factors17 in Olmsted County are similar to those for other white residents of the United States. Heart failure rates were determined clinically according to the Framingham criteria. However, the Framingham criteria may inappropriately diagnose individuals with heart failure32 and falsely elevate the prevalence of heart failure as a preoperative comorbidity or postoperative complication.

The statistical analysis included patients counted multiple times if they underwent subsequent hip fracture repair during the study period. Including these patients may inaccurately inflate event rates or contribute to incorrect estimates of standard error. However, we felt it was appropriate to include recurrent hip fracture repair cases in the analysis because they represent a clinically distinct patient from both a medical and functional perspective. We used a robust variance estimator in the Cox proportional hazards models to provide an accurate calculation of the standard error given the possibility for correlation within subjects.20 Finally, the proportion of these patients was low (94 of 1116 unique patients; 8.4%).

Future work must involve further risk stratification and therapeutic interventions in perioperative hip fracture patients. A more robust analysis of heart failure, with differentiation between systolic and diastolic dysfunction, may facilitate risk stratification. Assessment of compliance with standard preoperative heart failure medications and the impact of heightened clinical vigilance may enlighten means to improve postoperative outcomes. Studies on risk stratification and therapeutic interventions may then inform policy regarding length of stay and reimbursement in hip fracture patients.

CONCLUSION

In summary, our population‐based findings reveal that heart failure represents a prevalent and serious comorbidity in patients undergoing hip fracture repair. Clinicians caring for perioperative hip fracture patients must pay particular attention to risk for, and implications of, new or recurrent heart failure.

As the population ages, hip fractures and heart failure increase in prevalence.1, 2 Heart failure prevalence is also increasing in hospitalized patients.3 Indeed, hospitalizations involving heart failure as an active issue tripled in the last 30 years.4 Heart failure has been associated with an increased risk for hip fracture,5, 6 and previous studies report a 6%20% prevalence of preoperative heart failure in hip fracture patients.710 While exacerbation of heart failure increases the mortality risk in patients admitted for hip fractures,8 the incidence of new heart failure, as well as the preoperative factors that predict postoperative heart failure in this patient population remain unclear.

American College of Cardiology/American Heart Association (ACC/AHA) perioperative guidelines identify orthopedic surgeries, including hip fracture repair, as intermediate risk procedures.11 Compared to other intermediate risk operations, however, postoperative outcomes following hip fracture repair differ significantly.1216 Overall mortality in hip fracture patients has been reported at 29% at one year,8 with the excess mortality from hip fracture alone at nearly 20%.10, 13 However, the exact factors that contribute to this excess mortality, particularly with regard to heart failure, remain unclear.

To examine the preoperative prevalence, subsequent incidence, and predictors of heart failure in patients undergoing hip fracture repair operations, this study used an established, population‐based database to compare the postoperative consequences in hip fracture repair patients with and without preexisting heart failure. We hypothesized that preoperative heart failure worsens postoperative outcomes in hip fracture patients.

METHODS

RESULTS

From among 1327 potential hip fracture repairs, we excluded 115 cases involving multiple injuries or operations (19), pathological fractures (20), in‐hospital fractures (3), or an operation >72 hours after the initial fracture (5). Three patients under 65 years of age were also excluded, as were cases with missing information (9) or cases managed nonoperatively (56). The final analysis included 1212 surgical cases in 1116 subjects. No subjects were lost to surveillance for 1 year following their hip fracture repair.

Table 1 summarizes the baseline characteristics of the study population. The overall prevalence of preoperative heart failure was 27.0% (327 of 1212). Those with preoperative heart failure were older, heavier, more likely male and white, and less likely to live independently preoperatively. They were also more likely to suffer from preexisting cardiovascular comorbidities.

Table 1 also summarizes the main outcome characteristics of the study population. Those with preoperative heart failure had longer mean lengths of stay (LOS), were more often discharged to a skilled facility, and demonstrated higher inpatient mortality rates.

Table 2 summarizes the outcomes associated with preoperative heart failure. The overall rate of postoperative heart failure was 6.7% within 7 postoperative days and 21.3% within 1 postoperative year. Postoperative heart failure was significantly more common among those with preoperative heart failure (hazard ratio [HR], 3.0; 95% confidence interval [CI], 2.3 to 3.9; P < 0.001). Among those without preoperative heart failure, rates of postoperative incident heart failure were 4.8% at 7 days and 15.0% at 1 year. Compared to patients without preoperative heart failure, those with preoperative heart failure demonstrated higher one year mortality rates and higher rates of postoperative heart failure at 7 days and 1 year.

Figure 1 displays the outcomes to 1 year of surveillance. Rates of postoperative heart failure and postoperative mortality were consistently higher among those with, versus without, preoperative heart failure. Figure 2 displays similar data stratified by gender. Postoperative heart failure rates did not differ significantly between genders (HR, 1.0; 95% CI, 0.8 to 1.4), but postoperative mortality rates were significantly higher among males than females (HR, 1.9; 95% CI, 1.5 to 2.5; P < 0.001).

Figure 1

Figure 2

Figure 3 displays survival rates to 1 year based on the occurrence of incident or recurrent heart failure within the first 7 postoperative days. Survival rates were lowest among patients with recurrent heart failure in the first 7 postoperative days and highest among those with no preoperative or postoperative heart failure. Subjects with incident heart failure in the first postoperative week, and those with preoperative heart failure who did not suffer a recurrence, demonstrated intermediate survival rates (P < 0.001 for trend across all four groups).

Figure 3

DISCUSSION

This population‐based study found that heart failure represents a highly prevalent condition in elderly patients undergoing hip fracture repairs. It demonstrates that those with preoperative heart failure typically suffer from more cardiovascular comorbidities and carry a higher risk of postoperative heart failure and postoperative mortality.

While many studies have focused on the epidemiology of hip fractures,21 population‐based data on cardiac complications following hip fracture repair are significantly less common. The ACC/AHA preoperative cardiac evaluation guidelines classify orthopedic procedures, including hip fracture repair, as intermediate risk.11 Consequently, some may assume that all orthopedic patients will have a mortality rate less than 5%. Indeed, the 30‐day postoperative mortality rate published from our institution's Total Joint Registry was 0.6% following elective total hip arthroplasty.22 However, the present study demonstrates that current ACC/AHA preoperative cardiac evaluation guidelines may not apply to the population of frail patients undergoing hip fracture repair. Particularly among those who experience new heart failure within the first seven days following surgery, outcomes are substantially worse than the ACC/AHA perioperative guidelines may suggest.11

Preoperative heart failure has been associated with adverse risk for postoperative mortality after hip fracture.9, 10, 12 However, these studies did not report heart failure as a complication of hip fracture repair. A prospective cohort study of 2448 hip fracture patients at an academic hospital in Great Britain found a 5% rate of inpatient heart failure as a postoperative complication.23 The hazard ratio for one‐year mortality was 11.3 with postoperative heart failure.23 However, the British study did not distinguish heart failure from other cardiovascular diseases as a preoperative comorbidity or stratify the risk for postoperative mortality by preoperative heart failure status.23 Our findings add to previous literature by measuring heart failure as a specific complication of hip fracture repair and examining the association of preoperative heart failure with postoperative heart failure and mortality.

Length of stay after hip fracture repair varies in the literature, but previous work has not clearly associated heart failure with length of hospitalization in the setting of hip fracture repair.24, 25 Our study found a significantly higher mean length of stay among those with preoperative heart failure. This adds to previous work by delineating an association between heart failure and increased length of stay after hip fracture repair.

We found a higher rate of postoperative mortality among men compared to women. Rates of postoperative heart failure, however, were more similar (Figure 2). Previous studies have found a consistently higher mortality rate among men versus women after hip fracture.9, 23, 2529 Possible explanations for these findings include the overall increased burden of cardiovascular disease among men, lower treatment rates of osteoporosis in men,30 and increased susceptibility to other postoperative complications, such as infection.25

The findings of this study carry important clinical implications for the perioperative care of hip fracture patients with, or at risk for, heart failure. They suggest that current risk stratification guidelines classifying orthopedic operations as intermediate risk procedures do not reflect the high risk for morbidity that hip fracture patients face.11 The association of heart failure with adverse outcomes implies the need for heightened surveillance in the perioperative period, particularly with regard to volume status and medication reconciliation. Hip fracture patients and their families must be counseled about the ramifications of perioperative heart failure, including higher rates of postoperative heart failure, longer hospitalizations, and ultimate mortality.

This research carries several limitations and remains subject to biases inherent in retrospective cohort studies. The reported effects of heart failure on outcomes after hip fracture repair may be due to confounding from age, functional status, and other comorbidities. We attempted to minimize sampling bias through complete enumeration of hip fracture surgeries among Olmsted County residents. Completeness of follow‐up (100% at one year) was possible given the availability of documentation of all inpatient and outpatient medical care in the community.17 We used objectively defined outcomes to minimize measurement bias. Applicability to a more diverse population may be limited because >95% of the research population was from a single, predominantly white community. However, prior studies have documented that hip fracture incidence rates31 and socioeconomic factors17 in Olmsted County are similar to those for other white residents of the United States. Heart failure rates were determined clinically according to the Framingham criteria. However, the Framingham criteria may inappropriately diagnose individuals with heart failure32 and falsely elevate the prevalence of heart failure as a preoperative comorbidity or postoperative complication.

The statistical analysis included patients counted multiple times if they underwent subsequent hip fracture repair during the study period. Including these patients may inaccurately inflate event rates or contribute to incorrect estimates of standard error. However, we felt it was appropriate to include recurrent hip fracture repair cases in the analysis because they represent a clinically distinct patient from both a medical and functional perspective. We used a robust variance estimator in the Cox proportional hazards models to provide an accurate calculation of the standard error given the possibility for correlation within subjects.20 Finally, the proportion of these patients was low (94 of 1116 unique patients; 8.4%).

Future work must involve further risk stratification and therapeutic interventions in perioperative hip fracture patients. A more robust analysis of heart failure, with differentiation between systolic and diastolic dysfunction, may facilitate risk stratification. Assessment of compliance with standard preoperative heart failure medications and the impact of heightened clinical vigilance may enlighten means to improve postoperative outcomes. Studies on risk stratification and therapeutic interventions may then inform policy regarding length of stay and reimbursement in hip fracture patients.

CONCLUSION

In summary, our population‐based findings reveal that heart failure represents a prevalent and serious comorbidity in patients undergoing hip fracture repair. Clinicians caring for perioperative hip fracture patients must pay particular attention to risk for, and implications of, new or recurrent heart failure.

As the population ages, hip fractures and heart failure increase in prevalence.1, 2 Heart failure prevalence is also increasing in hospitalized patients.3 Indeed, hospitalizations involving heart failure as an active issue tripled in the last 30 years.4 Heart failure has been associated with an increased risk for hip fracture,5, 6 and previous studies report a 6%20% prevalence of preoperative heart failure in hip fracture patients.710 While exacerbation of heart failure increases the mortality risk in patients admitted for hip fractures,8 the incidence of new heart failure, as well as the preoperative factors that predict postoperative heart failure in this patient population remain unclear.

American College of Cardiology/American Heart Association (ACC/AHA) perioperative guidelines identify orthopedic surgeries, including hip fracture repair, as intermediate risk procedures.11 Compared to other intermediate risk operations, however, postoperative outcomes following hip fracture repair differ significantly.1216 Overall mortality in hip fracture patients has been reported at 29% at one year,8 with the excess mortality from hip fracture alone at nearly 20%.10, 13 However, the exact factors that contribute to this excess mortality, particularly with regard to heart failure, remain unclear.

To examine the preoperative prevalence, subsequent incidence, and predictors of heart failure in patients undergoing hip fracture repair operations, this study used an established, population‐based database to compare the postoperative consequences in hip fracture repair patients with and without preexisting heart failure. We hypothesized that preoperative heart failure worsens postoperative outcomes in hip fracture patients.

METHODS

RESULTS

From among 1327 potential hip fracture repairs, we excluded 115 cases involving multiple injuries or operations (19), pathological fractures (20), in‐hospital fractures (3), or an operation >72 hours after the initial fracture (5). Three patients under 65 years of age were also excluded, as were cases with missing information (9) or cases managed nonoperatively (56). The final analysis included 1212 surgical cases in 1116 subjects. No subjects were lost to surveillance for 1 year following their hip fracture repair.

Table 1 summarizes the baseline characteristics of the study population. The overall prevalence of preoperative heart failure was 27.0% (327 of 1212). Those with preoperative heart failure were older, heavier, more likely male and white, and less likely to live independently preoperatively. They were also more likely to suffer from preexisting cardiovascular comorbidities.

Table 1 also summarizes the main outcome characteristics of the study population. Those with preoperative heart failure had longer mean lengths of stay (LOS), were more often discharged to a skilled facility, and demonstrated higher inpatient mortality rates.

Table 2 summarizes the outcomes associated with preoperative heart failure. The overall rate of postoperative heart failure was 6.7% within 7 postoperative days and 21.3% within 1 postoperative year. Postoperative heart failure was significantly more common among those with preoperative heart failure (hazard ratio [HR], 3.0; 95% confidence interval [CI], 2.3 to 3.9; P < 0.001). Among those without preoperative heart failure, rates of postoperative incident heart failure were 4.8% at 7 days and 15.0% at 1 year. Compared to patients without preoperative heart failure, those with preoperative heart failure demonstrated higher one year mortality rates and higher rates of postoperative heart failure at 7 days and 1 year.

Figure 1 displays the outcomes to 1 year of surveillance. Rates of postoperative heart failure and postoperative mortality were consistently higher among those with, versus without, preoperative heart failure. Figure 2 displays similar data stratified by gender. Postoperative heart failure rates did not differ significantly between genders (HR, 1.0; 95% CI, 0.8 to 1.4), but postoperative mortality rates were significantly higher among males than females (HR, 1.9; 95% CI, 1.5 to 2.5; P < 0.001).

Figure 1

Figure 2

Figure 3 displays survival rates to 1 year based on the occurrence of incident or recurrent heart failure within the first 7 postoperative days. Survival rates were lowest among patients with recurrent heart failure in the first 7 postoperative days and highest among those with no preoperative or postoperative heart failure. Subjects with incident heart failure in the first postoperative week, and those with preoperative heart failure who did not suffer a recurrence, demonstrated intermediate survival rates (P < 0.001 for trend across all four groups).

Figure 3

DISCUSSION

This population‐based study found that heart failure represents a highly prevalent condition in elderly patients undergoing hip fracture repairs. It demonstrates that those with preoperative heart failure typically suffer from more cardiovascular comorbidities and carry a higher risk of postoperative heart failure and postoperative mortality.

While many studies have focused on the epidemiology of hip fractures,21 population‐based data on cardiac complications following hip fracture repair are significantly less common. The ACC/AHA preoperative cardiac evaluation guidelines classify orthopedic procedures, including hip fracture repair, as intermediate risk.11 Consequently, some may assume that all orthopedic patients will have a mortality rate less than 5%. Indeed, the 30‐day postoperative mortality rate published from our institution's Total Joint Registry was 0.6% following elective total hip arthroplasty.22 However, the present study demonstrates that current ACC/AHA preoperative cardiac evaluation guidelines may not apply to the population of frail patients undergoing hip fracture repair. Particularly among those who experience new heart failure within the first seven days following surgery, outcomes are substantially worse than the ACC/AHA perioperative guidelines may suggest.11

Preoperative heart failure has been associated with adverse risk for postoperative mortality after hip fracture.9, 10, 12 However, these studies did not report heart failure as a complication of hip fracture repair. A prospective cohort study of 2448 hip fracture patients at an academic hospital in Great Britain found a 5% rate of inpatient heart failure as a postoperative complication.23 The hazard ratio for one‐year mortality was 11.3 with postoperative heart failure.23 However, the British study did not distinguish heart failure from other cardiovascular diseases as a preoperative comorbidity or stratify the risk for postoperative mortality by preoperative heart failure status.23 Our findings add to previous literature by measuring heart failure as a specific complication of hip fracture repair and examining the association of preoperative heart failure with postoperative heart failure and mortality.

Length of stay after hip fracture repair varies in the literature, but previous work has not clearly associated heart failure with length of hospitalization in the setting of hip fracture repair.24, 25 Our study found a significantly higher mean length of stay among those with preoperative heart failure. This adds to previous work by delineating an association between heart failure and increased length of stay after hip fracture repair.

We found a higher rate of postoperative mortality among men compared to women. Rates of postoperative heart failure, however, were more similar (Figure 2). Previous studies have found a consistently higher mortality rate among men versus women after hip fracture.9, 23, 2529 Possible explanations for these findings include the overall increased burden of cardiovascular disease among men, lower treatment rates of osteoporosis in men,30 and increased susceptibility to other postoperative complications, such as infection.25

The findings of this study carry important clinical implications for the perioperative care of hip fracture patients with, or at risk for, heart failure. They suggest that current risk stratification guidelines classifying orthopedic operations as intermediate risk procedures do not reflect the high risk for morbidity that hip fracture patients face.11 The association of heart failure with adverse outcomes implies the need for heightened surveillance in the perioperative period, particularly with regard to volume status and medication reconciliation. Hip fracture patients and their families must be counseled about the ramifications of perioperative heart failure, including higher rates of postoperative heart failure, longer hospitalizations, and ultimate mortality.

This research carries several limitations and remains subject to biases inherent in retrospective cohort studies. The reported effects of heart failure on outcomes after hip fracture repair may be due to confounding from age, functional status, and other comorbidities. We attempted to minimize sampling bias through complete enumeration of hip fracture surgeries among Olmsted County residents. Completeness of follow‐up (100% at one year) was possible given the availability of documentation of all inpatient and outpatient medical care in the community.17 We used objectively defined outcomes to minimize measurement bias. Applicability to a more diverse population may be limited because >95% of the research population was from a single, predominantly white community. However, prior studies have documented that hip fracture incidence rates31 and socioeconomic factors17 in Olmsted County are similar to those for other white residents of the United States. Heart failure rates were determined clinically according to the Framingham criteria. However, the Framingham criteria may inappropriately diagnose individuals with heart failure32 and falsely elevate the prevalence of heart failure as a preoperative comorbidity or postoperative complication.

The statistical analysis included patients counted multiple times if they underwent subsequent hip fracture repair during the study period. Including these patients may inaccurately inflate event rates or contribute to incorrect estimates of standard error. However, we felt it was appropriate to include recurrent hip fracture repair cases in the analysis because they represent a clinically distinct patient from both a medical and functional perspective. We used a robust variance estimator in the Cox proportional hazards models to provide an accurate calculation of the standard error given the possibility for correlation within subjects.20 Finally, the proportion of these patients was low (94 of 1116 unique patients; 8.4%).

Future work must involve further risk stratification and therapeutic interventions in perioperative hip fracture patients. A more robust analysis of heart failure, with differentiation between systolic and diastolic dysfunction, may facilitate risk stratification. Assessment of compliance with standard preoperative heart failure medications and the impact of heightened clinical vigilance may enlighten means to improve postoperative outcomes. Studies on risk stratification and therapeutic interventions may then inform policy regarding length of stay and reimbursement in hip fracture patients.

CONCLUSION

In summary, our population‐based findings reveal that heart failure represents a prevalent and serious comorbidity in patients undergoing hip fracture repair. Clinicians caring for perioperative hip fracture patients must pay particular attention to risk for, and implications of, new or recurrent heart failure.

Public health concerns such as the aging population¹ and the increasing prevalence of obesity² are also important issues to hospitals. However, little attention has been given to the interface of obesity and the elderly, largely due to the dearth of studies that include elderly patients. An aging population leads to an increase in geriatric syndromes, such as osteoporosis³ and its most devastating complication, hip fracture.⁴ These frail, hip‐fracture patients pose management challenges to practicing geriatricians and hospitalists.^5,6 Furthermore, although fracture risk is inversely correlated to body mass index (BMI),⁷ this relationship has yet to be fully examined in the postoperative hip‐fracture population. In other surgical settings, there is disagreement as to whether underweight or obese patients are at higher risk of developing medical complications,^8‐11 but for orthopedic patients, data have been limited to elective orthopedic populations.^12‐14 We previously demonstrated that underweight hip‐fracture patients are at higher risk of postoperative cardiac complications at 1 year,¹⁵ consistent with studies of cardiac risk indices determining long‐term events. Because of different pathophysiologic mechanisms, the purpose of this study was to ascertain the influence of BMI on inpatient postoperative noncardiac medical complications and to assess predictors of such complications following urgent hip fracture repair.

Patients and Methods

All Olmsted County, Minnesota, residents undergoing urgent hip repair due to fracture were identified using the Rochester Epidemiology Project, a medical‐record linkage system funded by the Federal government since 1966 to support disease‐related epidemiology studies.¹⁶ All patient medical care is indexed, and both inpatient and outpatient visits are captured and available for review, allowing for complete case ascertainment. Medical care in Olmsted County is primarily provided by Mayo Clinic with its affiliated hospitals (St. Mary's and Rochester Methodist) and the Olmsted Medical Center, in addition to a few individual providers. Over 95% of all Olmsted County hip fracture surgeries are ultimately managed at St. Mary's Hospital.

Following approval by the Institutional Review Board we used this unique data resource to identify all residents with an International Classification of Diseases, 9th edition (ICD‐9) diagnosis code of 820 to 829 for hip fracture (n = 1310). Both sexes were included, and all patients included in the study provided research authorization for use of their medical records for research purposes.¹⁷ We excluded patients who were managed conservatively (n = 56), had a pathological fracture (n = 20), had multiple injuries (n = 19), were operated on >72 hours after fracture (n = 5), were aged <65 years (n = 2), or were admitted for reasons other than a fracture and experienced an in‐hospital fracture (n = 3). We subsequently excluded patients with missing information (n = 10). World Health Organization (WHO) criteria were used for classifying BMI: underweight (BMI < 18.5); normal (BMI = 18.5‐24.9); overweight (BMI = 25.0‐29.9); and obese (BMI 30.0).¹⁸

All data were abstracted using standardized collection forms by trained nurse abstractors blinded to the study hypothesis. Patients' admission height and weight were documented; if unavailable, the nearest data within 2 months prior to surgery were recorded. Patients' preadmission residence, functional status, baseline comorbidities, admission medications, discharge destination, as well as whether patients had an intensive care unit stay or any major surgeries in the past 90 days were abstracted. In addition, American Society of Anesthesia (ASA) class, type of anesthesia, and length of stay were also obtained. Inpatient complications that had been identified by the treating physicians and documented in the medical record or identified on imaging studies were assessed from the time of hip fracture repair to the time of discharge using standardized clinical criteria (Table 1). For criteria that were based on either objective findings or clinical documentation/suspicion, the patient was considered to meet the criteria of having a complication if they fulfilled either one. We did not include any cardiac outcomes, including congestive heart failure, angina, myocardial infarction, or arrhythmias that had been previously reported.¹⁵ Noncardiac complications were classified broadly: respiratory (respiratory failure, respiratory depression, or pulmonary hypoxemia); neurologic (any cerebral event including hemorrhagic or ischemic stroke, transient ischemic attack, or delirium); gastrointestinal (ileus or gastrointestinal bleeding); vascular (pulmonary embolus, or deep vein thrombosis); infectious (pneumonia, sepsis, urinary tract, wound, or cellulitis); renal/metabolic (acute renal failure, dehydration, or electrolyte abnormalities); or other (fractures or falls).

Continuous data are presented as means standard deviation and categorical data as counts and percentages. In testing for differences in patient demographics, past medical history, and baseline clinical data among BMI groups, Kruskal‐Wallis tests were performed for continuous variables and Fisher's Exact or Cochran‐Mantel‐Haenszel tests were used for discrete variables. Bonferroni adjustments were performed where appropriate. The primary outcome was the risk of any noncardiac medical complication during the postoperative hospitalization, based on patients with complications. Incidence rates were calculated for the overall group as well as for each BMI category. BMI was evaluated categorically according to the WHO criteria, as a continuous variable dichotomized as a BMI 18.5 kg/m² to 24.9 kg/m² (normal) vs. all others, and above/below 25.0 kg/m². The effect of BMI and other potential risk factors on the complication rate was evaluated using logistic regression. The effect of BMI category on the overall complication rate was adjusted for the a priori risk factors of age, sex, surgical year, and ASA class both univariately (Model 1) and multivariately (Model 2). In addition to these variables, we also evaluated other potential risk factors, including baseline demographic and baseline clinical variables that were significant (P < 0.05) univariately using a stepwise selection; first forcing in BMI as a categorical variable (Model 3), then repeating the stepwise selection process without forcing in BMI (Model 4). Using data from Lawrence et al.,¹⁹ we estimated that we would have 80% power to detect differences in rates of inpatient noncardiac complications equal to an odds ratio (OR) = 2.2 (normal vs. underweight), OR = 2.0 (normal vs. overweight), and OR = 2.4 (normal vs. obese). Finally, because of power considerations, as an exploratory analysis, we additionally identified predictors of inpatient complications within each BMI category using stepwise selection. All statistical tests were 2‐sided, and P values <0.05 were considered significant. All analyses were performed using SAS for UNIX (version 9.1.3; SAS Institute, Inc., Cary, NC).

Results

Between 1988 and 2002, 1195 urgent repairs for hip fracture met our inclusion/exclusion criteria. We subsequently excluded 15 repairs with missing BMI data, and, of the 7 patients with >1 repair, we included only their first fracture episode in our analysis. Two were subsequently excluded due to an administrative error. Ultimately, 1180 hip fracture repairs were included in the analysis cohort. There were 184 (15.6%) patients in the underweight group, 640 (54.2%) with normal BMI, 251 (21.3%) with a BMI 25.0 to 29.9 kg/m², and 105 (8.9%) with a BMI 30 kg/m². Baseline characteristics are otherwise shown in Table 2. Normal BMI patients were significantly older than the other groups, and underweight patients were less likely to be admitted from home. Past history of having a cardiovascular risk factor or a cardiovascular diagnosis appeared to increase with increasing BMI. Underweight patients were more likely to have chronic obstructive pulmonary disease (COPD) than patients with normal BMI (P = 0.03) or overweight patients (P = 0.009), but not more than obese patients (P = 0.21). There were no differences across BMI groups in ASA class, type of anesthesia, intensive care unit stay, or length of stay.

There were 77 (41.8%) postoperative inpatient noncardiac complications in the underweight group, 234 (36.6%) in the normal BMI group, 90 (35.9%) in the overweight group, and 42 (40.0%) in the obese group (P = 0.49). Figure 1 demonstrates the main subcategory complication rates by BMI group, and Table 3 outlines the univariate unadjusted complication rates. Other than gastrointestinal complications being more prevalent as BMI increases (P = 0.005), there were no significant differences in crude complication rates across BMI categories (all P > 0.05) for the other complication subcategories. A multiple comparisons analysis did not demonstrate any differences between normal and any of the other BMI categories for ileus. Normal BMI patients were more likely to be discharged to a nursing facility than overweight or obese patients (85.5% vs. 79.3%, P = 0.03; and 85.5% vs. 79.0%, P = 0.03, respectively). The proportion of in‐hospital deaths among underweight patients was significantly higher than in any of the other groups (9.3% vs. 4.4%; P = 0.01), but mean length of stay was not significantly different.

Figure 1

Significant univariate predictors of the composite outcome of any noncardiac complication included: age (OR, 1.04 95% confidence interval [CI>], 1.02‐1.06; P < 0.001), age 75 years (OR, 2.25; 95% CI, 1.52‐3.33; P < 0.001), age 85 years (OR, 1.49; 95% CI, 1.17‐1.89; P < 0.001), male sex (OR, 1.41; 95% CI, 1.05‐1.90; P = 0.02), admission from home (OR, 0.77; 95% CI, 0.61‐0.98; P = 0.03), a history of cerebrovascular disease (OR, 1.41; 95% CI, 1.08‐1.83; P = 0.01), myocardial infarction (OR, 1.41; 95% CI, 1.07‐1.86; P = 0.02), angina (OR, 1.32; 95% CI, 1.03‐1.69; P = 0.03), congestive heart failure (OR, 1.45; 95% CI, 1.11‐1.89; P = 0.006), dementia (OR, 1.39; 95% CI, 1.08‐1.78; P = 0.01), peripheral vascular disease (OR, 1.47; 95% CI, 1.06‐2.03; P = 0.02), COPD/asthma (OR, 1.56; 95% CI, 1.18‐2.08; P = 0.002), osteoarthritis (OR, 1.29; 95% CI, 1.01‐1.65; P = 0.04), code status as Do Not Resuscitate (OR, 0.74; 95% CI, 0.58‐0.94; P = 0.015), or ASA class III‐V (OR, 2.24; 95% CI, 1.53‐3.29; P < 0.001). Results were no different after using the Charlson comorbidity index in place of ASA class (data not shown). No significant differences in overall noncardiac complications were observed when examining BMI as a continuous variable, as a categorical variable, as 25 kg/m² vs. <25 kg/m², or as 18.5 kg/m² to 24.9 kg/m² vs. all others. Examining renal, respiratory, peripheral vascular, or neurologic complications univariately within these aforementioned strata also did not demonstrate any significant differences among BMI categories (data not shown).

Multivariable analyses (Models 1‐4) are shown for any overall noncardiac inpatient medical complication in Table 4. BMI was not a significant predictor in any of our models, specifically in our main model that examined the effect of BMI adjusting for a priori variables (Model 2). However, older age, male sex, and ASA class were highly significant predictors of complications in all four models; however, surgical year was nonsignificant. Notably, after stepwise selection for other demographic and premorbid variables, a history of COPD or asthma was found to be an additional significant factor both in Model 3 (forcing BMI in the model) and Model 4 (without BMI in the model). Exploratory analysis of individual predictors of inpatient noncardiac complications within each BMI category demonstrated that, in underweight patients, admission use of ‐blockers was a significant predictor of having any medical complication (OR, 3.1; 95% CI, 1.1‐8.60; P = 0.03). In normal BMI patients, age 75 years (OR, 2.6; 95% CI, 1.4‐4.9; P = 0.003), ASA class III‐V (OR, 2.3; 95% CI, 1.3‐3.9; P = 0.003), and a history of cerebrovascular disease (OR, 1.5; 95%CI, 1.04‐2.1; P = 0.03) were predictors; and, in obese patients, only age (OR, 1.1; 95% CI, 1.00‐1.12; P = 0.05) was significant. There were no significant predictors of having a medical complication in the overweight group.

Discussion

Most research describing the association of BMI with postoperative outcomes has concentrated on cardiac surgery, general surgical procedures, and intensive care unit utilization.^8‐11,20 In the orthopedic literature, an elevated BMI has been associated with a higher number of short‐term complications, but this was limited to elective knee arthroplasty and spine surgery populations.^12,13,21 Conversely, no differences were observed in obese patients undergoing hip arthroplasties.^14,22 To the best of our knowledge, this study may be the first to examine the impact of BMI on inpatient hospital outcomes following urgent hip fracture repair. Our results suggest the risk of developing a noncardiac medical complication is the same regardless of BMI.

Our overall complication rate was higher (38%) than previous reports by others.^19,23‐26 Thus, Lawrence et al.,¹⁹ in their retrospective study of 20 facilities, demonstrated an overall complication rate of 17%, even though they also included postoperative cardiac complications. Although their study period overlapped our own (1982‐1993), they additionally included patients aged 60 to 65 years, a population known to have fewer comorbidities and fewer postoperative complications than the elderly hip‐fracture patients studied here. In addition, their population may have been healthier at baseline, in that a higher proportion lived at home (73%) and a lower percentage were ASA class III‐V (71%) than our cohort. These differences in baseline characteristics may explain the higher complication rates observed in our study.

Our findings did not suggest any relationship of BMI with noncardiac postoperative medical complications in any of the 4 methods we used to stratify BMI (continuous, categorical, normal vs. abnormal, and 25 kg/m²). Evidence is contradictory as to what the effect of BMI has on postoperative complications. An elevated BMI (30 kg/m²) has been shown to lead to increased sternal wound infection and saphenous vein harvest infection in a cardiac surgery population,²⁷ but other studies^10,28,29 have demonstrated the opposite effect. Among 6336 patients undergoing elective general surgery procedures, the incidence of complications were similar by body mass.³⁰ A matched study design that included urgent and emergent surgeries also did not find any appreciable increased perioperative risk in noncardiac surgery.²⁸ Whether this may be due to the elective nature of the surgeries in these studies, hence leading to selection bias, is unknown.

In geriatric patients, multiple baseline comorbid conditions often are reflected in a higher ASA class, which increases the risk of significant perioperative complications. Our multivariate modeling showed that a high ASA class strongly predicts morbidity and mortality following hip fracture repair, in line with other studies.^19,31,32 Although the Charlson comorbidity index could alternatively been used, we elected to adjust for ASA class as it is more commonly used and is simple to use. Interestingly, surgical year did not significantly predict any complication, which can suggest that practice changes play a minimal impact on patient outcomes. However, we caution that because the individual event rates, particularly vascular, were low, we were unable to fully determine whether changes in practice management, such as improved thromboprophylaxis, would impact event rates over time. Finally, other predictors such as older age³³ and a concomitant history of either COPD or asthma,³⁴ are well‐accepted predictors of inpatient complications. Our attempt to examine specific predictors of complications in each BMI category revealed differing results, making interpretations difficult. Because of power considerations, this was meant solely as an exploratory analysis, and larger cohorts are needed to further ascertain whether predictors are different in these groups. Such a study may in fact identify perioperative issues that allow practitioners caring for this population to modify these factors.

One of the major limitations in our study was our inability to adjust for individual complications using multivariable models, such as deep vein thrombosis or delirium, within each BMI stratum, because of statistical power issues. Such a study would require large numbers of individual complications or events to allow for appropriate adjustments. The authors acknowledge that such individual complication rates may vary dramatically. We were aware of this potential problem, and therefore a priori ascertained a composite outcome of any noncardiac medical complication. However, our results do provide preliminary information regarding the impact of BMI on noncardiac medical complications. Further studies would be needed, though, to fully determine the effect of BMI on the number of cases with each complication.

Obesity (or BMI) is a known cardiovascular risk factor, and our previous study's aim was to determine cardiovascular events in a comparable manner to the way risk indices, such as the Goldman, Lee, or the AHA preoperative algorithm function. The surgical literature often presents noncardiac complications separately, allowing us to directly compare our own data to other published studies. We used 2 separate approaches, focusing on the inpatient stay (ascertaining noncardiac complications) and 1‐year cardiac outcomes (cardiac complications), as these are mediated by different mechanisms and factors. Furthermore, the intent of both studies was to dispel any concerns that an elevated BMI would in fact lead to an increased number of complications. Whether cardiac complications, though, would impact noncardiac complications, or vice‐versa, is unknown, and would require further investigation.

Although we relied on well‐established definitions for body mass, they have often been challenged, as they may underestimate adiposity in the elderly population due to age‐related reductions in lean mass.^35,36 Studies have demonstrated a poor correlation between percent body fat and BMI in the >65 year age group,³⁷ which could impact our results and outcomes by misclassifying patients. Yet, as an anthropometric measurement, BMI is easily obtainable and its variables are routinely documented in patients' medical records, as compared to other anthropometric measurements. Other means of estimating adiposity, such as densitometry or computed tomography (CT) scanning, are impractical, expensive, and not used clinically but routinely in research settings. The lack of standardization in obtaining height and weight, despite nurse‐initiated protocols for bed calibration, may have introduced a degree of measurement bias. Furthermore, the extent of lean mass lost and volume status changes lead to further challenges of using BMI in hospital settings. Whether other anthropometric measurements, including hip circumference, waist circumference, or waist‐hip ratio, should be used in this group of patients requires further examination. However, despite its shortcomings in elderly patients, BMI is still deemed an appropriate surrogate for obesity.

Our main strength was the use of the Rochester Epidemiology Project medical record linkage system to ascertain all patient data. This focuses on patients from a single geographically‐defined community minimizing referral biases often observed in studies originating from a tertiary care referral center. Previous disease‐related epidemiology studies using the Olmsted County population have demonstrated excellent external validity to the U.S. white population.¹⁶ We relied on the medical documentation of the treating clinician for many diagnoses in our data abstraction. Although we attempted to use standardized definitions, clinicians may have inadvertently forgotten to document subjective signs or symptoms that would assist in the categorization of these complications. Hence, added inpatient complications may have been overlooked, suggesting that our results may slightly underestimate the true incidence in this population. Additionally, certain complications may overlap categories, such as pneumonia and infections. We agree with Lawrence et al.¹⁹ that long periods of time are necessary to accumulate data of this kind in an effort to describe complication rates epidemiologically.

Despite no difference in outcomes among BMI categories, our results have striking implications for the hospitalized patient. Thus, underweight elderly patients, often considered frail with minimal functional reserve, are at no higher risk for developing inpatient medical complications than patients with higher BMIs. This is contrary to our study focusing on cardiac complications, where underweight patients were at higher risk.¹⁵ Conversely, obese patients, who have been demonstrated to be at higher risk of medical complications (particularly pulmonary), had no greater risk than patients with normal BMI. To the practicing geriatrician and hospitalist, this information provides important prognostication regarding additional perioperative measures that need to be implemented in these different groups. Based on our results, BMI does not play a particular role in noncardiac medical complications, dispelling any myths of the added burden of excess weight on surgical outcomes in this population. From a hospital perspective, this may be important since additional testing or preventative management in these patients may lead to additional resource use. However, in‐hospital deaths were higher in underweight patients than in patients with a normal BMI. Although we were underpowered to detect any differences in mortality between groups and could therefore not adjust for additional variables, it is unknown whether cardiac or noncardiac complications may be a stronger predictor of death in the underweight patient population. Further studies would be needed to better ascertain this relationship.

Conclusions

In elderly patients undergoing urgent hip fracture repair, BMI does not appear to lead to an excess rate of inpatient noncardiac complications. Our results are the first to demonstrate that BMI has no impact on morbidity in this patient population. Further research on the influence of body composition on inpatient complications in this population is needed to accurately allow for appropriate perioperative prophylaxis. Whether BMI impacts specific complications or in‐patient mortality in this population still requires investigation.

Public health concerns such as the aging population¹ and the increasing prevalence of obesity² are also important issues to hospitals. However, little attention has been given to the interface of obesity and the elderly, largely due to the dearth of studies that include elderly patients. An aging population leads to an increase in geriatric syndromes, such as osteoporosis³ and its most devastating complication, hip fracture.⁴ These frail, hip‐fracture patients pose management challenges to practicing geriatricians and hospitalists.^5,6 Furthermore, although fracture risk is inversely correlated to body mass index (BMI),⁷ this relationship has yet to be fully examined in the postoperative hip‐fracture population. In other surgical settings, there is disagreement as to whether underweight or obese patients are at higher risk of developing medical complications,^8‐11 but for orthopedic patients, data have been limited to elective orthopedic populations.^12‐14 We previously demonstrated that underweight hip‐fracture patients are at higher risk of postoperative cardiac complications at 1 year,¹⁵ consistent with studies of cardiac risk indices determining long‐term events. Because of different pathophysiologic mechanisms, the purpose of this study was to ascertain the influence of BMI on inpatient postoperative noncardiac medical complications and to assess predictors of such complications following urgent hip fracture repair.

Patients and Methods

All Olmsted County, Minnesota, residents undergoing urgent hip repair due to fracture were identified using the Rochester Epidemiology Project, a medical‐record linkage system funded by the Federal government since 1966 to support disease‐related epidemiology studies.¹⁶ All patient medical care is indexed, and both inpatient and outpatient visits are captured and available for review, allowing for complete case ascertainment. Medical care in Olmsted County is primarily provided by Mayo Clinic with its affiliated hospitals (St. Mary's and Rochester Methodist) and the Olmsted Medical Center, in addition to a few individual providers. Over 95% of all Olmsted County hip fracture surgeries are ultimately managed at St. Mary's Hospital.

Following approval by the Institutional Review Board we used this unique data resource to identify all residents with an International Classification of Diseases, 9th edition (ICD‐9) diagnosis code of 820 to 829 for hip fracture (n = 1310). Both sexes were included, and all patients included in the study provided research authorization for use of their medical records for research purposes.¹⁷ We excluded patients who were managed conservatively (n = 56), had a pathological fracture (n = 20), had multiple injuries (n = 19), were operated on >72 hours after fracture (n = 5), were aged <65 years (n = 2), or were admitted for reasons other than a fracture and experienced an in‐hospital fracture (n = 3). We subsequently excluded patients with missing information (n = 10). World Health Organization (WHO) criteria were used for classifying BMI: underweight (BMI < 18.5); normal (BMI = 18.5‐24.9); overweight (BMI = 25.0‐29.9); and obese (BMI 30.0).¹⁸

All data were abstracted using standardized collection forms by trained nurse abstractors blinded to the study hypothesis. Patients' admission height and weight were documented; if unavailable, the nearest data within 2 months prior to surgery were recorded. Patients' preadmission residence, functional status, baseline comorbidities, admission medications, discharge destination, as well as whether patients had an intensive care unit stay or any major surgeries in the past 90 days were abstracted. In addition, American Society of Anesthesia (ASA) class, type of anesthesia, and length of stay were also obtained. Inpatient complications that had been identified by the treating physicians and documented in the medical record or identified on imaging studies were assessed from the time of hip fracture repair to the time of discharge using standardized clinical criteria (Table 1). For criteria that were based on either objective findings or clinical documentation/suspicion, the patient was considered to meet the criteria of having a complication if they fulfilled either one. We did not include any cardiac outcomes, including congestive heart failure, angina, myocardial infarction, or arrhythmias that had been previously reported.¹⁵ Noncardiac complications were classified broadly: respiratory (respiratory failure, respiratory depression, or pulmonary hypoxemia); neurologic (any cerebral event including hemorrhagic or ischemic stroke, transient ischemic attack, or delirium); gastrointestinal (ileus or gastrointestinal bleeding); vascular (pulmonary embolus, or deep vein thrombosis); infectious (pneumonia, sepsis, urinary tract, wound, or cellulitis); renal/metabolic (acute renal failure, dehydration, or electrolyte abnormalities); or other (fractures or falls).

Continuous data are presented as means standard deviation and categorical data as counts and percentages. In testing for differences in patient demographics, past medical history, and baseline clinical data among BMI groups, Kruskal‐Wallis tests were performed for continuous variables and Fisher's Exact or Cochran‐Mantel‐Haenszel tests were used for discrete variables. Bonferroni adjustments were performed where appropriate. The primary outcome was the risk of any noncardiac medical complication during the postoperative hospitalization, based on patients with complications. Incidence rates were calculated for the overall group as well as for each BMI category. BMI was evaluated categorically according to the WHO criteria, as a continuous variable dichotomized as a BMI 18.5 kg/m² to 24.9 kg/m² (normal) vs. all others, and above/below 25.0 kg/m². The effect of BMI and other potential risk factors on the complication rate was evaluated using logistic regression. The effect of BMI category on the overall complication rate was adjusted for the a priori risk factors of age, sex, surgical year, and ASA class both univariately (Model 1) and multivariately (Model 2). In addition to these variables, we also evaluated other potential risk factors, including baseline demographic and baseline clinical variables that were significant (P < 0.05) univariately using a stepwise selection; first forcing in BMI as a categorical variable (Model 3), then repeating the stepwise selection process without forcing in BMI (Model 4). Using data from Lawrence et al.,¹⁹ we estimated that we would have 80% power to detect differences in rates of inpatient noncardiac complications equal to an odds ratio (OR) = 2.2 (normal vs. underweight), OR = 2.0 (normal vs. overweight), and OR = 2.4 (normal vs. obese). Finally, because of power considerations, as an exploratory analysis, we additionally identified predictors of inpatient complications within each BMI category using stepwise selection. All statistical tests were 2‐sided, and P values <0.05 were considered significant. All analyses were performed using SAS for UNIX (version 9.1.3; SAS Institute, Inc., Cary, NC).

Results

Between 1988 and 2002, 1195 urgent repairs for hip fracture met our inclusion/exclusion criteria. We subsequently excluded 15 repairs with missing BMI data, and, of the 7 patients with >1 repair, we included only their first fracture episode in our analysis. Two were subsequently excluded due to an administrative error. Ultimately, 1180 hip fracture repairs were included in the analysis cohort. There were 184 (15.6%) patients in the underweight group, 640 (54.2%) with normal BMI, 251 (21.3%) with a BMI 25.0 to 29.9 kg/m², and 105 (8.9%) with a BMI 30 kg/m². Baseline characteristics are otherwise shown in Table 2. Normal BMI patients were significantly older than the other groups, and underweight patients were less likely to be admitted from home. Past history of having a cardiovascular risk factor or a cardiovascular diagnosis appeared to increase with increasing BMI. Underweight patients were more likely to have chronic obstructive pulmonary disease (COPD) than patients with normal BMI (P = 0.03) or overweight patients (P = 0.009), but not more than obese patients (P = 0.21). There were no differences across BMI groups in ASA class, type of anesthesia, intensive care unit stay, or length of stay.

There were 77 (41.8%) postoperative inpatient noncardiac complications in the underweight group, 234 (36.6%) in the normal BMI group, 90 (35.9%) in the overweight group, and 42 (40.0%) in the obese group (P = 0.49). Figure 1 demonstrates the main subcategory complication rates by BMI group, and Table 3 outlines the univariate unadjusted complication rates. Other than gastrointestinal complications being more prevalent as BMI increases (P = 0.005), there were no significant differences in crude complication rates across BMI categories (all P > 0.05) for the other complication subcategories. A multiple comparisons analysis did not demonstrate any differences between normal and any of the other BMI categories for ileus. Normal BMI patients were more likely to be discharged to a nursing facility than overweight or obese patients (85.5% vs. 79.3%, P = 0.03; and 85.5% vs. 79.0%, P = 0.03, respectively). The proportion of in‐hospital deaths among underweight patients was significantly higher than in any of the other groups (9.3% vs. 4.4%; P = 0.01), but mean length of stay was not significantly different.

Figure 1

Significant univariate predictors of the composite outcome of any noncardiac complication included: age (OR, 1.04 95% confidence interval [CI>], 1.02‐1.06; P < 0.001), age 75 years (OR, 2.25; 95% CI, 1.52‐3.33; P < 0.001), age 85 years (OR, 1.49; 95% CI, 1.17‐1.89; P < 0.001), male sex (OR, 1.41; 95% CI, 1.05‐1.90; P = 0.02), admission from home (OR, 0.77; 95% CI, 0.61‐0.98; P = 0.03), a history of cerebrovascular disease (OR, 1.41; 95% CI, 1.08‐1.83; P = 0.01), myocardial infarction (OR, 1.41; 95% CI, 1.07‐1.86; P = 0.02), angina (OR, 1.32; 95% CI, 1.03‐1.69; P = 0.03), congestive heart failure (OR, 1.45; 95% CI, 1.11‐1.89; P = 0.006), dementia (OR, 1.39; 95% CI, 1.08‐1.78; P = 0.01), peripheral vascular disease (OR, 1.47; 95% CI, 1.06‐2.03; P = 0.02), COPD/asthma (OR, 1.56; 95% CI, 1.18‐2.08; P = 0.002), osteoarthritis (OR, 1.29; 95% CI, 1.01‐1.65; P = 0.04), code status as Do Not Resuscitate (OR, 0.74; 95% CI, 0.58‐0.94; P = 0.015), or ASA class III‐V (OR, 2.24; 95% CI, 1.53‐3.29; P < 0.001). Results were no different after using the Charlson comorbidity index in place of ASA class (data not shown). No significant differences in overall noncardiac complications were observed when examining BMI as a continuous variable, as a categorical variable, as 25 kg/m² vs. <25 kg/m², or as 18.5 kg/m² to 24.9 kg/m² vs. all others. Examining renal, respiratory, peripheral vascular, or neurologic complications univariately within these aforementioned strata also did not demonstrate any significant differences among BMI categories (data not shown).

Multivariable analyses (Models 1‐4) are shown for any overall noncardiac inpatient medical complication in Table 4. BMI was not a significant predictor in any of our models, specifically in our main model that examined the effect of BMI adjusting for a priori variables (Model 2). However, older age, male sex, and ASA class were highly significant predictors of complications in all four models; however, surgical year was nonsignificant. Notably, after stepwise selection for other demographic and premorbid variables, a history of COPD or asthma was found to be an additional significant factor both in Model 3 (forcing BMI in the model) and Model 4 (without BMI in the model). Exploratory analysis of individual predictors of inpatient noncardiac complications within each BMI category demonstrated that, in underweight patients, admission use of ‐blockers was a significant predictor of having any medical complication (OR, 3.1; 95% CI, 1.1‐8.60; P = 0.03). In normal BMI patients, age 75 years (OR, 2.6; 95% CI, 1.4‐4.9; P = 0.003), ASA class III‐V (OR, 2.3; 95% CI, 1.3‐3.9; P = 0.003), and a history of cerebrovascular disease (OR, 1.5; 95%CI, 1.04‐2.1; P = 0.03) were predictors; and, in obese patients, only age (OR, 1.1; 95% CI, 1.00‐1.12; P = 0.05) was significant. There were no significant predictors of having a medical complication in the overweight group.

Discussion

Most research describing the association of BMI with postoperative outcomes has concentrated on cardiac surgery, general surgical procedures, and intensive care unit utilization.^8‐11,20 In the orthopedic literature, an elevated BMI has been associated with a higher number of short‐term complications, but this was limited to elective knee arthroplasty and spine surgery populations.^12,13,21 Conversely, no differences were observed in obese patients undergoing hip arthroplasties.^14,22 To the best of our knowledge, this study may be the first to examine the impact of BMI on inpatient hospital outcomes following urgent hip fracture repair. Our results suggest the risk of developing a noncardiac medical complication is the same regardless of BMI.

Our overall complication rate was higher (38%) than previous reports by others.^19,23‐26 Thus, Lawrence et al.,¹⁹ in their retrospective study of 20 facilities, demonstrated an overall complication rate of 17%, even though they also included postoperative cardiac complications. Although their study period overlapped our own (1982‐1993), they additionally included patients aged 60 to 65 years, a population known to have fewer comorbidities and fewer postoperative complications than the elderly hip‐fracture patients studied here. In addition, their population may have been healthier at baseline, in that a higher proportion lived at home (73%) and a lower percentage were ASA class III‐V (71%) than our cohort. These differences in baseline characteristics may explain the higher complication rates observed in our study.

Our findings did not suggest any relationship of BMI with noncardiac postoperative medical complications in any of the 4 methods we used to stratify BMI (continuous, categorical, normal vs. abnormal, and 25 kg/m²). Evidence is contradictory as to what the effect of BMI has on postoperative complications. An elevated BMI (30 kg/m²) has been shown to lead to increased sternal wound infection and saphenous vein harvest infection in a cardiac surgery population,²⁷ but other studies^10,28,29 have demonstrated the opposite effect. Among 6336 patients undergoing elective general surgery procedures, the incidence of complications were similar by body mass.³⁰ A matched study design that included urgent and emergent surgeries also did not find any appreciable increased perioperative risk in noncardiac surgery.²⁸ Whether this may be due to the elective nature of the surgeries in these studies, hence leading to selection bias, is unknown.

In geriatric patients, multiple baseline comorbid conditions often are reflected in a higher ASA class, which increases the risk of significant perioperative complications. Our multivariate modeling showed that a high ASA class strongly predicts morbidity and mortality following hip fracture repair, in line with other studies.^19,31,32 Although the Charlson comorbidity index could alternatively been used, we elected to adjust for ASA class as it is more commonly used and is simple to use. Interestingly, surgical year did not significantly predict any complication, which can suggest that practice changes play a minimal impact on patient outcomes. However, we caution that because the individual event rates, particularly vascular, were low, we were unable to fully determine whether changes in practice management, such as improved thromboprophylaxis, would impact event rates over time. Finally, other predictors such as older age³³ and a concomitant history of either COPD or asthma,³⁴ are well‐accepted predictors of inpatient complications. Our attempt to examine specific predictors of complications in each BMI category revealed differing results, making interpretations difficult. Because of power considerations, this was meant solely as an exploratory analysis, and larger cohorts are needed to further ascertain whether predictors are different in these groups. Such a study may in fact identify perioperative issues that allow practitioners caring for this population to modify these factors.

One of the major limitations in our study was our inability to adjust for individual complications using multivariable models, such as deep vein thrombosis or delirium, within each BMI stratum, because of statistical power issues. Such a study would require large numbers of individual complications or events to allow for appropriate adjustments. The authors acknowledge that such individual complication rates may vary dramatically. We were aware of this potential problem, and therefore a priori ascertained a composite outcome of any noncardiac medical complication. However, our results do provide preliminary information regarding the impact of BMI on noncardiac medical complications. Further studies would be needed, though, to fully determine the effect of BMI on the number of cases with each complication.

Obesity (or BMI) is a known cardiovascular risk factor, and our previous study's aim was to determine cardiovascular events in a comparable manner to the way risk indices, such as the Goldman, Lee, or the AHA preoperative algorithm function. The surgical literature often presents noncardiac complications separately, allowing us to directly compare our own data to other published studies. We used 2 separate approaches, focusing on the inpatient stay (ascertaining noncardiac complications) and 1‐year cardiac outcomes (cardiac complications), as these are mediated by different mechanisms and factors. Furthermore, the intent of both studies was to dispel any concerns that an elevated BMI would in fact lead to an increased number of complications. Whether cardiac complications, though, would impact noncardiac complications, or vice‐versa, is unknown, and would require further investigation.

Although we relied on well‐established definitions for body mass, they have often been challenged, as they may underestimate adiposity in the elderly population due to age‐related reductions in lean mass.^35,36 Studies have demonstrated a poor correlation between percent body fat and BMI in the >65 year age group,³⁷ which could impact our results and outcomes by misclassifying patients. Yet, as an anthropometric measurement, BMI is easily obtainable and its variables are routinely documented in patients' medical records, as compared to other anthropometric measurements. Other means of estimating adiposity, such as densitometry or computed tomography (CT) scanning, are impractical, expensive, and not used clinically but routinely in research settings. The lack of standardization in obtaining height and weight, despite nurse‐initiated protocols for bed calibration, may have introduced a degree of measurement bias. Furthermore, the extent of lean mass lost and volume status changes lead to further challenges of using BMI in hospital settings. Whether other anthropometric measurements, including hip circumference, waist circumference, or waist‐hip ratio, should be used in this group of patients requires further examination. However, despite its shortcomings in elderly patients, BMI is still deemed an appropriate surrogate for obesity.

Our main strength was the use of the Rochester Epidemiology Project medical record linkage system to ascertain all patient data. This focuses on patients from a single geographically‐defined community minimizing referral biases often observed in studies originating from a tertiary care referral center. Previous disease‐related epidemiology studies using the Olmsted County population have demonstrated excellent external validity to the U.S. white population.¹⁶ We relied on the medical documentation of the treating clinician for many diagnoses in our data abstraction. Although we attempted to use standardized definitions, clinicians may have inadvertently forgotten to document subjective signs or symptoms that would assist in the categorization of these complications. Hence, added inpatient complications may have been overlooked, suggesting that our results may slightly underestimate the true incidence in this population. Additionally, certain complications may overlap categories, such as pneumonia and infections. We agree with Lawrence et al.¹⁹ that long periods of time are necessary to accumulate data of this kind in an effort to describe complication rates epidemiologically.

Despite no difference in outcomes among BMI categories, our results have striking implications for the hospitalized patient. Thus, underweight elderly patients, often considered frail with minimal functional reserve, are at no higher risk for developing inpatient medical complications than patients with higher BMIs. This is contrary to our study focusing on cardiac complications, where underweight patients were at higher risk.¹⁵ Conversely, obese patients, who have been demonstrated to be at higher risk of medical complications (particularly pulmonary), had no greater risk than patients with normal BMI. To the practicing geriatrician and hospitalist, this information provides important prognostication regarding additional perioperative measures that need to be implemented in these different groups. Based on our results, BMI does not play a particular role in noncardiac medical complications, dispelling any myths of the added burden of excess weight on surgical outcomes in this population. From a hospital perspective, this may be important since additional testing or preventative management in these patients may lead to additional resource use. However, in‐hospital deaths were higher in underweight patients than in patients with a normal BMI. Although we were underpowered to detect any differences in mortality between groups and could therefore not adjust for additional variables, it is unknown whether cardiac or noncardiac complications may be a stronger predictor of death in the underweight patient population. Further studies would be needed to better ascertain this relationship.

Conclusions

In elderly patients undergoing urgent hip fracture repair, BMI does not appear to lead to an excess rate of inpatient noncardiac complications. Our results are the first to demonstrate that BMI has no impact on morbidity in this patient population. Further research on the influence of body composition on inpatient complications in this population is needed to accurately allow for appropriate perioperative prophylaxis. Whether BMI impacts specific complications or in‐patient mortality in this population still requires investigation.

Public health concerns such as the aging population¹ and the increasing prevalence of obesity² are also important issues to hospitals. However, little attention has been given to the interface of obesity and the elderly, largely due to the dearth of studies that include elderly patients. An aging population leads to an increase in geriatric syndromes, such as osteoporosis³ and its most devastating complication, hip fracture.⁴ These frail, hip‐fracture patients pose management challenges to practicing geriatricians and hospitalists.^5,6 Furthermore, although fracture risk is inversely correlated to body mass index (BMI),⁷ this relationship has yet to be fully examined in the postoperative hip‐fracture population. In other surgical settings, there is disagreement as to whether underweight or obese patients are at higher risk of developing medical complications,^8‐11 but for orthopedic patients, data have been limited to elective orthopedic populations.^12‐14 We previously demonstrated that underweight hip‐fracture patients are at higher risk of postoperative cardiac complications at 1 year,¹⁵ consistent with studies of cardiac risk indices determining long‐term events. Because of different pathophysiologic mechanisms, the purpose of this study was to ascertain the influence of BMI on inpatient postoperative noncardiac medical complications and to assess predictors of such complications following urgent hip fracture repair.

Patients and Methods

All Olmsted County, Minnesota, residents undergoing urgent hip repair due to fracture were identified using the Rochester Epidemiology Project, a medical‐record linkage system funded by the Federal government since 1966 to support disease‐related epidemiology studies.¹⁶ All patient medical care is indexed, and both inpatient and outpatient visits are captured and available for review, allowing for complete case ascertainment. Medical care in Olmsted County is primarily provided by Mayo Clinic with its affiliated hospitals (St. Mary's and Rochester Methodist) and the Olmsted Medical Center, in addition to a few individual providers. Over 95% of all Olmsted County hip fracture surgeries are ultimately managed at St. Mary's Hospital.

Following approval by the Institutional Review Board we used this unique data resource to identify all residents with an International Classification of Diseases, 9th edition (ICD‐9) diagnosis code of 820 to 829 for hip fracture (n = 1310). Both sexes were included, and all patients included in the study provided research authorization for use of their medical records for research purposes.¹⁷ We excluded patients who were managed conservatively (n = 56), had a pathological fracture (n = 20), had multiple injuries (n = 19), were operated on >72 hours after fracture (n = 5), were aged <65 years (n = 2), or were admitted for reasons other than a fracture and experienced an in‐hospital fracture (n = 3). We subsequently excluded patients with missing information (n = 10). World Health Organization (WHO) criteria were used for classifying BMI: underweight (BMI < 18.5); normal (BMI = 18.5‐24.9); overweight (BMI = 25.0‐29.9); and obese (BMI 30.0).¹⁸

All data were abstracted using standardized collection forms by trained nurse abstractors blinded to the study hypothesis. Patients' admission height and weight were documented; if unavailable, the nearest data within 2 months prior to surgery were recorded. Patients' preadmission residence, functional status, baseline comorbidities, admission medications, discharge destination, as well as whether patients had an intensive care unit stay or any major surgeries in the past 90 days were abstracted. In addition, American Society of Anesthesia (ASA) class, type of anesthesia, and length of stay were also obtained. Inpatient complications that had been identified by the treating physicians and documented in the medical record or identified on imaging studies were assessed from the time of hip fracture repair to the time of discharge using standardized clinical criteria (Table 1). For criteria that were based on either objective findings or clinical documentation/suspicion, the patient was considered to meet the criteria of having a complication if they fulfilled either one. We did not include any cardiac outcomes, including congestive heart failure, angina, myocardial infarction, or arrhythmias that had been previously reported.¹⁵ Noncardiac complications were classified broadly: respiratory (respiratory failure, respiratory depression, or pulmonary hypoxemia); neurologic (any cerebral event including hemorrhagic or ischemic stroke, transient ischemic attack, or delirium); gastrointestinal (ileus or gastrointestinal bleeding); vascular (pulmonary embolus, or deep vein thrombosis); infectious (pneumonia, sepsis, urinary tract, wound, or cellulitis); renal/metabolic (acute renal failure, dehydration, or electrolyte abnormalities); or other (fractures or falls).

Continuous data are presented as means standard deviation and categorical data as counts and percentages. In testing for differences in patient demographics, past medical history, and baseline clinical data among BMI groups, Kruskal‐Wallis tests were performed for continuous variables and Fisher's Exact or Cochran‐Mantel‐Haenszel tests were used for discrete variables. Bonferroni adjustments were performed where appropriate. The primary outcome was the risk of any noncardiac medical complication during the postoperative hospitalization, based on patients with complications. Incidence rates were calculated for the overall group as well as for each BMI category. BMI was evaluated categorically according to the WHO criteria, as a continuous variable dichotomized as a BMI 18.5 kg/m² to 24.9 kg/m² (normal) vs. all others, and above/below 25.0 kg/m². The effect of BMI and other potential risk factors on the complication rate was evaluated using logistic regression. The effect of BMI category on the overall complication rate was adjusted for the a priori risk factors of age, sex, surgical year, and ASA class both univariately (Model 1) and multivariately (Model 2). In addition to these variables, we also evaluated other potential risk factors, including baseline demographic and baseline clinical variables that were significant (P < 0.05) univariately using a stepwise selection; first forcing in BMI as a categorical variable (Model 3), then repeating the stepwise selection process without forcing in BMI (Model 4). Using data from Lawrence et al.,¹⁹ we estimated that we would have 80% power to detect differences in rates of inpatient noncardiac complications equal to an odds ratio (OR) = 2.2 (normal vs. underweight), OR = 2.0 (normal vs. overweight), and OR = 2.4 (normal vs. obese). Finally, because of power considerations, as an exploratory analysis, we additionally identified predictors of inpatient complications within each BMI category using stepwise selection. All statistical tests were 2‐sided, and P values <0.05 were considered significant. All analyses were performed using SAS for UNIX (version 9.1.3; SAS Institute, Inc., Cary, NC).

Results

Between 1988 and 2002, 1195 urgent repairs for hip fracture met our inclusion/exclusion criteria. We subsequently excluded 15 repairs with missing BMI data, and, of the 7 patients with >1 repair, we included only their first fracture episode in our analysis. Two were subsequently excluded due to an administrative error. Ultimately, 1180 hip fracture repairs were included in the analysis cohort. There were 184 (15.6%) patients in the underweight group, 640 (54.2%) with normal BMI, 251 (21.3%) with a BMI 25.0 to 29.9 kg/m², and 105 (8.9%) with a BMI 30 kg/m². Baseline characteristics are otherwise shown in Table 2. Normal BMI patients were significantly older than the other groups, and underweight patients were less likely to be admitted from home. Past history of having a cardiovascular risk factor or a cardiovascular diagnosis appeared to increase with increasing BMI. Underweight patients were more likely to have chronic obstructive pulmonary disease (COPD) than patients with normal BMI (P = 0.03) or overweight patients (P = 0.009), but not more than obese patients (P = 0.21). There were no differences across BMI groups in ASA class, type of anesthesia, intensive care unit stay, or length of stay.

There were 77 (41.8%) postoperative inpatient noncardiac complications in the underweight group, 234 (36.6%) in the normal BMI group, 90 (35.9%) in the overweight group, and 42 (40.0%) in the obese group (P = 0.49). Figure 1 demonstrates the main subcategory complication rates by BMI group, and Table 3 outlines the univariate unadjusted complication rates. Other than gastrointestinal complications being more prevalent as BMI increases (P = 0.005), there were no significant differences in crude complication rates across BMI categories (all P > 0.05) for the other complication subcategories. A multiple comparisons analysis did not demonstrate any differences between normal and any of the other BMI categories for ileus. Normal BMI patients were more likely to be discharged to a nursing facility than overweight or obese patients (85.5% vs. 79.3%, P = 0.03; and 85.5% vs. 79.0%, P = 0.03, respectively). The proportion of in‐hospital deaths among underweight patients was significantly higher than in any of the other groups (9.3% vs. 4.4%; P = 0.01), but mean length of stay was not significantly different.

Figure 1

Significant univariate predictors of the composite outcome of any noncardiac complication included: age (OR, 1.04 95% confidence interval [CI>], 1.02‐1.06; P < 0.001), age 75 years (OR, 2.25; 95% CI, 1.52‐3.33; P < 0.001), age 85 years (OR, 1.49; 95% CI, 1.17‐1.89; P < 0.001), male sex (OR, 1.41; 95% CI, 1.05‐1.90; P = 0.02), admission from home (OR, 0.77; 95% CI, 0.61‐0.98; P = 0.03), a history of cerebrovascular disease (OR, 1.41; 95% CI, 1.08‐1.83; P = 0.01), myocardial infarction (OR, 1.41; 95% CI, 1.07‐1.86; P = 0.02), angina (OR, 1.32; 95% CI, 1.03‐1.69; P = 0.03), congestive heart failure (OR, 1.45; 95% CI, 1.11‐1.89; P = 0.006), dementia (OR, 1.39; 95% CI, 1.08‐1.78; P = 0.01), peripheral vascular disease (OR, 1.47; 95% CI, 1.06‐2.03; P = 0.02), COPD/asthma (OR, 1.56; 95% CI, 1.18‐2.08; P = 0.002), osteoarthritis (OR, 1.29; 95% CI, 1.01‐1.65; P = 0.04), code status as Do Not Resuscitate (OR, 0.74; 95% CI, 0.58‐0.94; P = 0.015), or ASA class III‐V (OR, 2.24; 95% CI, 1.53‐3.29; P < 0.001). Results were no different after using the Charlson comorbidity index in place of ASA class (data not shown). No significant differences in overall noncardiac complications were observed when examining BMI as a continuous variable, as a categorical variable, as 25 kg/m² vs. <25 kg/m², or as 18.5 kg/m² to 24.9 kg/m² vs. all others. Examining renal, respiratory, peripheral vascular, or neurologic complications univariately within these aforementioned strata also did not demonstrate any significant differences among BMI categories (data not shown).

Multivariable analyses (Models 1‐4) are shown for any overall noncardiac inpatient medical complication in Table 4. BMI was not a significant predictor in any of our models, specifically in our main model that examined the effect of BMI adjusting for a priori variables (Model 2). However, older age, male sex, and ASA class were highly significant predictors of complications in all four models; however, surgical year was nonsignificant. Notably, after stepwise selection for other demographic and premorbid variables, a history of COPD or asthma was found to be an additional significant factor both in Model 3 (forcing BMI in the model) and Model 4 (without BMI in the model). Exploratory analysis of individual predictors of inpatient noncardiac complications within each BMI category demonstrated that, in underweight patients, admission use of ‐blockers was a significant predictor of having any medical complication (OR, 3.1; 95% CI, 1.1‐8.60; P = 0.03). In normal BMI patients, age 75 years (OR, 2.6; 95% CI, 1.4‐4.9; P = 0.003), ASA class III‐V (OR, 2.3; 95% CI, 1.3‐3.9; P = 0.003), and a history of cerebrovascular disease (OR, 1.5; 95%CI, 1.04‐2.1; P = 0.03) were predictors; and, in obese patients, only age (OR, 1.1; 95% CI, 1.00‐1.12; P = 0.05) was significant. There were no significant predictors of having a medical complication in the overweight group.

Discussion

Most research describing the association of BMI with postoperative outcomes has concentrated on cardiac surgery, general surgical procedures, and intensive care unit utilization.^8‐11,20 In the orthopedic literature, an elevated BMI has been associated with a higher number of short‐term complications, but this was limited to elective knee arthroplasty and spine surgery populations.^12,13,21 Conversely, no differences were observed in obese patients undergoing hip arthroplasties.^14,22 To the best of our knowledge, this study may be the first to examine the impact of BMI on inpatient hospital outcomes following urgent hip fracture repair. Our results suggest the risk of developing a noncardiac medical complication is the same regardless of BMI.

Our overall complication rate was higher (38%) than previous reports by others.^19,23‐26 Thus, Lawrence et al.,¹⁹ in their retrospective study of 20 facilities, demonstrated an overall complication rate of 17%, even though they also included postoperative cardiac complications. Although their study period overlapped our own (1982‐1993), they additionally included patients aged 60 to 65 years, a population known to have fewer comorbidities and fewer postoperative complications than the elderly hip‐fracture patients studied here. In addition, their population may have been healthier at baseline, in that a higher proportion lived at home (73%) and a lower percentage were ASA class III‐V (71%) than our cohort. These differences in baseline characteristics may explain the higher complication rates observed in our study.

Our findings did not suggest any relationship of BMI with noncardiac postoperative medical complications in any of the 4 methods we used to stratify BMI (continuous, categorical, normal vs. abnormal, and 25 kg/m²). Evidence is contradictory as to what the effect of BMI has on postoperative complications. An elevated BMI (30 kg/m²) has been shown to lead to increased sternal wound infection and saphenous vein harvest infection in a cardiac surgery population,²⁷ but other studies^10,28,29 have demonstrated the opposite effect. Among 6336 patients undergoing elective general surgery procedures, the incidence of complications were similar by body mass.³⁰ A matched study design that included urgent and emergent surgeries also did not find any appreciable increased perioperative risk in noncardiac surgery.²⁸ Whether this may be due to the elective nature of the surgeries in these studies, hence leading to selection bias, is unknown.

In geriatric patients, multiple baseline comorbid conditions often are reflected in a higher ASA class, which increases the risk of significant perioperative complications. Our multivariate modeling showed that a high ASA class strongly predicts morbidity and mortality following hip fracture repair, in line with other studies.^19,31,32 Although the Charlson comorbidity index could alternatively been used, we elected to adjust for ASA class as it is more commonly used and is simple to use. Interestingly, surgical year did not significantly predict any complication, which can suggest that practice changes play a minimal impact on patient outcomes. However, we caution that because the individual event rates, particularly vascular, were low, we were unable to fully determine whether changes in practice management, such as improved thromboprophylaxis, would impact event rates over time. Finally, other predictors such as older age³³ and a concomitant history of either COPD or asthma,³⁴ are well‐accepted predictors of inpatient complications. Our attempt to examine specific predictors of complications in each BMI category revealed differing results, making interpretations difficult. Because of power considerations, this was meant solely as an exploratory analysis, and larger cohorts are needed to further ascertain whether predictors are different in these groups. Such a study may in fact identify perioperative issues that allow practitioners caring for this population to modify these factors.

One of the major limitations in our study was our inability to adjust for individual complications using multivariable models, such as deep vein thrombosis or delirium, within each BMI stratum, because of statistical power issues. Such a study would require large numbers of individual complications or events to allow for appropriate adjustments. The authors acknowledge that such individual complication rates may vary dramatically. We were aware of this potential problem, and therefore a priori ascertained a composite outcome of any noncardiac medical complication. However, our results do provide preliminary information regarding the impact of BMI on noncardiac medical complications. Further studies would be needed, though, to fully determine the effect of BMI on the number of cases with each complication.

Obesity (or BMI) is a known cardiovascular risk factor, and our previous study's aim was to determine cardiovascular events in a comparable manner to the way risk indices, such as the Goldman, Lee, or the AHA preoperative algorithm function. The surgical literature often presents noncardiac complications separately, allowing us to directly compare our own data to other published studies. We used 2 separate approaches, focusing on the inpatient stay (ascertaining noncardiac complications) and 1‐year cardiac outcomes (cardiac complications), as these are mediated by different mechanisms and factors. Furthermore, the intent of both studies was to dispel any concerns that an elevated BMI would in fact lead to an increased number of complications. Whether cardiac complications, though, would impact noncardiac complications, or vice‐versa, is unknown, and would require further investigation.

Although we relied on well‐established definitions for body mass, they have often been challenged, as they may underestimate adiposity in the elderly population due to age‐related reductions in lean mass.^35,36 Studies have demonstrated a poor correlation between percent body fat and BMI in the >65 year age group,³⁷ which could impact our results and outcomes by misclassifying patients. Yet, as an anthropometric measurement, BMI is easily obtainable and its variables are routinely documented in patients' medical records, as compared to other anthropometric measurements. Other means of estimating adiposity, such as densitometry or computed tomography (CT) scanning, are impractical, expensive, and not used clinically but routinely in research settings. The lack of standardization in obtaining height and weight, despite nurse‐initiated protocols for bed calibration, may have introduced a degree of measurement bias. Furthermore, the extent of lean mass lost and volume status changes lead to further challenges of using BMI in hospital settings. Whether other anthropometric measurements, including hip circumference, waist circumference, or waist‐hip ratio, should be used in this group of patients requires further examination. However, despite its shortcomings in elderly patients, BMI is still deemed an appropriate surrogate for obesity.

Our main strength was the use of the Rochester Epidemiology Project medical record linkage system to ascertain all patient data. This focuses on patients from a single geographically‐defined community minimizing referral biases often observed in studies originating from a tertiary care referral center. Previous disease‐related epidemiology studies using the Olmsted County population have demonstrated excellent external validity to the U.S. white population.¹⁶ We relied on the medical documentation of the treating clinician for many diagnoses in our data abstraction. Although we attempted to use standardized definitions, clinicians may have inadvertently forgotten to document subjective signs or symptoms that would assist in the categorization of these complications. Hence, added inpatient complications may have been overlooked, suggesting that our results may slightly underestimate the true incidence in this population. Additionally, certain complications may overlap categories, such as pneumonia and infections. We agree with Lawrence et al.¹⁹ that long periods of time are necessary to accumulate data of this kind in an effort to describe complication rates epidemiologically.

Despite no difference in outcomes among BMI categories, our results have striking implications for the hospitalized patient. Thus, underweight elderly patients, often considered frail with minimal functional reserve, are at no higher risk for developing inpatient medical complications than patients with higher BMIs. This is contrary to our study focusing on cardiac complications, where underweight patients were at higher risk.¹⁵ Conversely, obese patients, who have been demonstrated to be at higher risk of medical complications (particularly pulmonary), had no greater risk than patients with normal BMI. To the practicing geriatrician and hospitalist, this information provides important prognostication regarding additional perioperative measures that need to be implemented in these different groups. Based on our results, BMI does not play a particular role in noncardiac medical complications, dispelling any myths of the added burden of excess weight on surgical outcomes in this population. From a hospital perspective, this may be important since additional testing or preventative management in these patients may lead to additional resource use. However, in‐hospital deaths were higher in underweight patients than in patients with a normal BMI. Although we were underpowered to detect any differences in mortality between groups and could therefore not adjust for additional variables, it is unknown whether cardiac or noncardiac complications may be a stronger predictor of death in the underweight patient population. Further studies would be needed to better ascertain this relationship.

Conclusions

In elderly patients undergoing urgent hip fracture repair, BMI does not appear to lead to an excess rate of inpatient noncardiac complications. Our results are the first to demonstrate that BMI has no impact on morbidity in this patient population. Further research on the influence of body composition on inpatient complications in this population is needed to accurately allow for appropriate perioperative prophylaxis. Whether BMI impacts specific complications or in‐patient mortality in this population still requires investigation.

In the United States, hip operations (internal fixation of fracture or total hip arthroplasty [THA]) are the most common noncardiac major surgical procedures performed in patients age 65 years and older (45.2 procedures per 100,000 persons per year).1 This number of procedures is projected to increase substantially in the coming decades.

Little is known about the clinical predictors of postoperative stroke in patients undergoing hip surgical procedures. Further, recent results of the Perioperative Ischemic Evaluation (POISE) trial have shown that measures taken to reduce cardiac complications postoperatively may adversely affect the risk of stroke.2 The POISE study showed decreases in myocardial infarction and coronary revascularization but accompanying increases in stroke and death with use of ‐blockers in patients undergoing noncardiac surgery.

Prevention of adverse events is one of the top priorities of the U.S. health care system today.35 Risk stratification and therapeutic optimization of underlying chronic diseases may be important in decreasing perioperative risk and improving postoperative outcomes.

Our objective was to determine the rate of postoperative ischemic stroke in all residents of Olmsted County, MN, who underwent hip operation between 1988 and 2002 and to identify clinical predictors of postoperative stroke.

Subjects and Methods

Olmsted County is one of the few places in the world where comprehensive population‐based studies of disease etiology and outcomes are feasible. This feasibility is due to the Rochester Epidemiology Project, a medical records linkage system that provides access to the records of all medical care in the community.1 All medical diagnoses made for a resident of Olmsted County are entered on a master sheet in the patient's medical record, which is then entered into a central computer index.

Hip operations were identified using the Surgical Information Recording System data warehouse, where detailed data are stored as International Classification of Diseases, 9th edition (ICD‐9) codes for all surgical procedures performed from January 1, 1988, forward. A total of 2028 THAs and hip fracture repairs (ICD‐9 codes 81.51, 81.52, 81.53, 79.15, and 79.25) performed between 1988 and 2002 in Olmsted County were identified. Of the hip procedures, 142 were excluded (Figure 1). The final analysis cohort contained 1886 hip operations1195 hip fracture repairs and 691 THAs.

Figure 1

The population‐based cohort was assembled and the data were abstracted from complete inpatient and outpatient records from admission for surgical treatment up to 1 year after surgery. Only those patients who had given prior authorization for research were included in the study cohort. The Mayo Clinic Institutional Review Board approved the study.

Results

Among the patients with the 1886 hip procedures, 67 ischemic strokes were identified within 1 year after the index surgical procedure10 (1.4%) among the 691 THAs and 57 (4.8%) among the 1195 hip fracture repairs. Baseline characteristics are summarized in Table 1. Compared with the THA group, patients in the hip fracture repair group were more likely to be older and female. Additionally, such comorbid conditions as a history of stroke, diabetes mellitus, congestive heart failure, atrial fibrillation, or dementia were more prevalent in the hip fracture repair group.

Univariate analyses assessing the rate and risk of postoperative ischemic stroke are shown in Table 2. The rate of stroke was significantly greater among hip fracture repairs than THAs 30 days postoperatively and 1 year postoperatively (1.5% vs. 0.6% and 5.5% vs. 1.5%, respectively; P < 0.001) (Figure 2). In our study we found an annual incidence rate of ischemic stroke of 4093 per 100,000 person‐years (95% confidence interval [CI], 3172‐5198 per 100,000 person‐years). Accounting for death as a competing risk for stroke had little impact on the rate of stroke overall or within the 2 surgical groups (results not shown). Univariate Cox proportional hazards models showed that neither sex nor history of hypertension, diabetes mellitus, COPD, chronic renal insufficiency, or CAD or use of HMG‐CoA reductase inhibitors or ‐blockers were significant predictors of ischemic stroke. However, other clinical risk factors, such as a history of atrial fibrillation (hazard ratio [HR], 2.16; P = 0.005), hip fracture repair vs. THA (HR, 3.80; P < 0.001), increased age (HR, 2.20; P = 0.017), aspirin use (HR, 1.8; P = 0.014), and history of previous stroke (HR, 4.18; P < 0.001), were significantly associated with an increased risk of stroke (Table 2).

Figure 2

Because age was associated with the type of surgical procedure (87% of hip fracture repair patients were 75 years or older compared with 45% of THA patients), the effect of hip fracture repair on ischemic stroke was adjusted for age. For similar reasons, sex was also examined as an adjusting factor. Adjustment for age and sex resulted in only a slight attenuation of the HR for hip fracture repair vs. THA, from 3.8 to 3.4. A further analysis also adjusted for history of hypertension and history of atrial fibrillation, both comorbidities commonly associated with ischemic stroke. After adjustment for age, sex, history of hypertension, and history of atrial fibrillation, the risk of ischemic stroke was still significantly greater in the hip fracture repair group than in the THA group (HR, 2.8; 95% CI, 1.4‐5.7; P = 0.005).

To determine the most important predictors of postoperative ischemic stroke, multivariable analysis was conducted with stepwise selection. Potential risk factors included the following: operative procedure type (hip fracture repair vs. THA), age, sex, and history of stroke, hypertension, atrial fibrillation, CAD, COPD, diabetes mellitus, and chronic renal insufficiency, as well as use of ‐blockers, HMG‐CoA reductase inhibitors, and aspirin on hospital admission. Among all these factors, history of stroke (HR, 3.27; P < 0.001) and hip fracture repair vs. THA (HR, 2.74; P = 0.004) were confirmed to be significant predictors of postoperative ischemic stroke; the other factors did not significantly affect the model (Figure 2).

Comment

Our findings contrast those of previous studies that focused on perioperative ischemic stroke rates for specific surgical procedures,2, 8, 9 but do seem concordant with published results for early event rates of cerebrovascular accident or transient ischemic attack (1%) following hip fracture.10 The data from our study suggest that perioperative stroke cumulative probability is relatively high for hip procedures at both 30 days (1.2%) and 1 year (3.9%) after the index surgical procedure compared with general procedures. Subjects with a history of stroke who were undergoing hip operation had a postoperative stroke risk of 3.0% at 30 days and 9.9% at 1 year.

The incidence of stroke was greater in the hip fracture repair group (1.5% at 30 days and 5.5% at 1 year) than in the elective THA group (0.6% at 30 days and 1.5% at 1 year). The increased 1‐year mortality for patients undergoing hip surgery compared with the general population is in part due to cerebrovascular disease,10 and, therefore, the 1‐year stroke incidence is important.

After adjustment for age, sex, and comorbidities (hypertension and atrial fibrillation), the risk of postoperative ischemic stroke was 2.71 times greater in the hip fracture repair group than in the THA group (P = 0.006). These data are important in counseling and caring for patients undergoing different types of hip procedures.

From 1985 through 1989, for the age group (75‐84 years old) that best fits the demographics of our cohort, both men and women had limited variation over time in annual incidence rates of stroke (2149‐1074 strokes per 100,000 population per year) for Olmsted County, MN.11 In our study we found an annual incidence rate of ischemic stroke of 4,093 per 100,000 person‐years (95% CI, 3172‐5198 per 100,000 person‐years). The lower limit of the 95% CI is higher than the rates reported for Olmsted County, suggesting that having hip surgery increases the 1‐year risk of ischemic stroke.

Previous studies have shown that the risk factor most consistently correlated to perioperative ischemic stroke is a history of stroke.9 In our study, history of stroke and type of hip fracture surgery were confirmed to be the strongest predictors of postoperative stroke. History of hypertension, atrial fibrillation, CAD, COPD, diabetes, or chronic renal insufficiency was not correlated to perioperative ischemic stroke.

Nonmodifiable risk factors, such as advanced age, serve as markers of stroke risk and help identify high‐risk populations that may require aggressive intervention. After age adjustment of hip fracture repair, age was no longer significantly associated with postoperative stroke.

Cerebrovascular disease appears to be a marker for CAD, and, therefore, patients with a history of stroke usually have a Revised Cardiac Risk Index that may suggest the use of ‐blockers. According to the recent results of the POISE trial, use of ‐blockers could lead to increased stroke incidence.2 Our results showed no significant correlation between stroke risk and ‐blocker use, but our study period was from 1988 to 2002, when titration of ‐blocker dose to heart rates of 55 to 60 beats per minute was not common practice.

Several studies have confirmed the value of aspirin in decreasing the rate of vascular outcomes after diagnosis of transient ischemic attack or stroke.12 In our study, aspirin use on hospital admission was found in the univariate analysis to be associated with an increased risk of stroke, but this finding was not confirmed after adjustments for age, sex, and comorbid conditions. Aspirin use on admission was not a significant predictor of postoperative stroke, most likely because aspirin use can be considered a marker of increased cardiovascular risk and we adjusted for these comorbid conditions.

The limitations of this study are inherent in its retrospective design. First, we identified all incident cases of stroke after hip operation by reviewing medical records and then abstracting data from those records. We may have missed some mild strokes if they were misclassified as peripheral vestibular neuropathy, migraine, or even seizure. Less likely is that we missed strokes within the first 30 days after the procedure because that is the period in which patients with hip operation are either hospitalized or sent for rehabilitation in skilled nursing facilities. It is known that institutionalization leads to better surveillance and more complete ascertainment of any medical event.

The event rate of postoperative stroke at 30 days after hip operation was low. Therefore, we did not have the statistical power to comment meaningfully on predictors of stroke at 30 days after the hip procedure. Any nonrespondent or volunteer bias was addressed by using data from the Rochester Epidemiology Project, which allowed us to identify all Olmsted County residents who underwent hip operation between 1988 and 2002. The diagnostic suspicion bias was also accounted for in our study design because different physicians provided care and outcome measurement.

Our results apply for the patients who underwent hip operation between 1988 and 2002. The noncardiac surgery guidelines have been revised between 1988 and 2002, and we did not perform a stratified analysis by index year. The next step in our study will be to extend our data collection to 2008 and look at time trends.

Conclusion

In this population‐based historical cohort study, patients undergoing hip operation had a 3.9% cumulative probability of ischemic stroke during the first postoperative year. History of stroke and type of hip procedure (ie, hip fracture repair) were the strongest predictors of this complication. Because history of stroke is such a strong predictor of postoperative stroke, the perioperative management of these patients should probably be tailored, with closely observed blood pressure management and antihypertensive medication adjustment, to avoid compromising cerebral perfusion. Also, to avoid postoperative hypercoagulability that increases the risk of stroke, these patients may need to begin receiving antiplatelets as soon as is surgically acceptable.1315