- Email: [email protected]
Congestive Heart Failure and Noncardiac Operations: Risk of Serious Morbidity, Readmission, Reoperation, and Mortality
Congestive Heart Failure and Noncardiac Operations: Risk of Serious Morbidity, Readmission, Reoperation, and Mortality Florence E Turrentine,
PhD, RN,
Min-Woong Sohn,
PhD,
Rayford Scott Jones,
MD, FACS
Congestive heart failure (CHF) predicts surgical morbidity and mortality. However, few studies evaluate CHF’s impact on noncardiac operations. Because of CHFs serious threat to health and survival, surgeons must understand risks CHF poses to patients undergoing a diverse array of operations. STUDY DESIGN: We used 2009 to 2013 American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Participant Use Files to estimate the risk of serious morbidity, reoperation, readmission, mortality, and other postoperative complications associated with preoperative diagnosis of CHF. Multivariable logistic regression analysis provided odds ratios (OR) and 95% confidence intervals (CI) for outcomes in 34 ACS NSQIP procedure groups, controlling for age, sex, race, emergency surgery status, American Society of Anesthesiologists Classification, body mass index, and selected laboratory values. RESULTS: Unadjusted ORs indicate adverse effects of CHF on surgical outcomes for most procedures considered. When adjusted for age and other confounders, CHF persists with adverse effects on most outcomes, including serious morbidity (OR 1.52, 95% CI, 1.44 to 1.61; p < 0.001); reoperation (OR 1.29, 95% CI, 1.17 to 1.42; p < 0.001); readmission (OR 1.39, 95% CI, 1.29 to 1.50; p < 0.001); and 30-day mortality (OR 1.96, 95% CI 1.80 to 2.13; p < 0.001). The impact of CHF on morbidity and mortality substantially affected those undergoing carotid endarterectomy and lower extremity endovascular repair. Cardiac arrest, mortality, unplanned intubation, and ventilator > 48 hours were complications most affected by CHF. CONCLUSIONS: Congestive heart failure strongly predicts serious morbidity, unplanned reoperation, readmission, and surgical mortality for noncardiac operations. Surgeons must pay particular attention to recognizing CHF and optimizing perioperative management when considering surgery. (J Am Coll Surg 2016;222:1220e1229. 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.)
BACKGROUND:
Congestive heart failure (CHF), a cause of morbidity and mortality after noncardiac surgery,1 predicts mortality for Medicare beneficiaries2,3 and for patients undergoing a range of operations, from outpatient laparoscopic cholecystectomy4 Disclosure Information: Nothing to disclose. Disclaimer: American College of Surgeons (ACS) NSQIP and the hospitals participating in the ACS NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Presented at the American College of Surgeons 101st Annual Clinical Congress, Chicago, IL, October 2015. Received December 8, 2015; Revised February 2, 2016; Accepted February 29, 2016. From the Departments of Surgery (Turrentine, Jones) and Public Health Sciences (Sohn), University of Virginia, Charlottesville, VA. Correspondence address: Rayford Scott Jones, MD, FACS, Department of Surgery, University of Virginia, Box 800709, Charlottesville, VA 22908. email: [email protected]
ª 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.
to pancreaticoduodenectomy for cancer.5 Even though CHF increases the risk of a variety of surgical complications2,3,6 including postoperative delirium7 and anastomotic leaks,8 there is no systematic examination of the kinds of noncardiac operations and adverse outcomes it affects. In this study, we conducted a comprehensive analysis of the impact of CHF on serious morbidity, reoperation, readmission, mortality, and other postoperative complications for an extensive array of surgical procedures. Undergoing surgery with the comorbidity of CHF should alert clinicians to increased risk that may be mitigated through preoperative counseling and vigilant perioperative management.
METHODS We used the 2009 through 2013 American College of Surgeons National Quality Improvement Program
1220
http://dx.doi.org/10.1016/j.jamcollsurg.2016.02.025 ISSN 1072-7515/16
Vol. 222, No. 6, June 2016
Turrentine et al
(ACS NSQIP) Participant Use Files (PUF)9 to identify patients with CHF undergoing surgery. The PUF is an approved Public Data Set at the University of Virginia, so no additional institutional review or approval was required. The ACS NSQIP CHF variable includes patients with a new diagnosis of CHF or a recent exacerbation of chronic CHF. The ACS NSQIP defines CHF as “the inability of the heart to pump a sufficient quantity of blood to meet the metabolic needs of the body or a situation where the heart can do so only at increased ventricular filling pressure.” Common CHF manifestations include exercise intolerance due to dyspnea or fatigue; orthopnea; paroxysmal nocturnal dyspnea; increased jugular venous pressure; pulmonary rales; cardiomegaly; pulmonary vascular engorgement; and pulmonary edema. To meet ACS NSQIP criteria, CHF must be newly diagnosed, or a diagnosis of chronic CHF requires current signs or symptoms in the 30 days before the principal operation.10 We grouped operations into 34 targeted procedure categories as outlined by the ACS NSQIP.11 Congestive heart failure occurred rarely for many of these procedures. Therefore, we excluded groups with 30 or fewer patients with CHF or <0.5% of the total procedure volume from further analyses. These groups included hepatectomy, esophagectomy, carotid artery stenting, gynecologic reconstruction, bladder suspension, prostatectomy, cystectomy, breast reduction, breast reconstruction, and abdominoplasty. We used the ACS NSQIP definition of serious morbidity, which includes 1 or more of the following occurrences: cardiac arrest, myocardial infarction, pneumonia, progressive renal insufficiency, acute renal failure, venous thromboembolism, return to the operating room, deep incisional surgical site infection, organ space surgical site infection, systemic sepsis (sepsis or septic shock), unplanned intubation, urinary tract infection, or wound disruption. When present at the time of surgery, superficial, deep, and organ space surgical site infection, pneumonia, unplanned intubation, sepsis, progressive renal insufficiency, acute renal failure, and urinary tract infection were not considered as postoperative complications.12 Renal failure and dialysis were combined into 1 variable to indicate presence of chronic kidney disease. A priori we included independent variables, age, sex, ethnicity, emergent status, functional status, American Society of Anesthesiologists (ASA) classification, BMI, dyspnea, renal dysfunction, and laboratory values serum glutamic oxaloacetic transaminase (SGOT), blood urea nitrogen (BUN), creatinine, and international normalized ratio (INR).
Congestive Heart Failure and Surgical Risk
1221
Study outcomes included serious morbidity, readmission, reoperation, and mortality within 30 days of discharge. We analyzed these outcomes together as well as separately for each of the 34 ACS NSQIP targeted procedure groups using multivariable logistic regression. In adjusted models, we controlled for the following potential confounders: age, sex, race, emergency surgery status, ASA score (3, 4, 5 vs 1 or 2), BMI, and selected preoperative laboratory values. Sensitivity analyses evaluated how and whether the addition of preoperative cardiac variables (previous percutaneous coronary intervention, history of myocardial infarction, previous cardiac surgery, history of angina, revascularization for peripheral vascular disease) diminished the magnitude of association between CHF and study outcomes. We also included operative duration and outpatient surgery status as 2 other measures of patient severity in our sensitivity analysis models. Stata SE version 13 (Statacorp) was used to conduct the analysis.
RESULTS Of 2.3 million records in the 2009 through 2013 ACS NSQIP PUFs, we used data for 1,268,999 patients who met our inclusion criteria. Only 0.61% of patients who underwent surgery met the ACS NSQIP definition of CHF, with the highest rates occurring in patients having surgery for carotid artery stenting, hip fracture repair, and lower extremity endovascular revascularization, 3.59%, 3.27%, and 2.97%, respectively (Table 1). Patients undergoing breast reconstruction, breast reduction, and gynecologic reconstruction presented with the fewest cases of CHF: 0.02%, 0.04%, and 0.04%, respectively. Table 2 shows characteristics of patients who met inclusion criteria and presented with and without CHF (n ¼ 1,174,002). Patients were significantly different on all characteristics, with age and race/ethnicity being the most prominent characteristics that distinguished patients with CHF from those without the comorbid condition. Congestive heart failure increased exponentially with increasing age; of the patients with CHF, 46.54% were 75 years of age or older compared with 14.76% of those without CHF (p < 0.001). Thirteen percent of patients with CHF were non-Hispanic black compared with 9.16% of patients without the comorbidity (p < 0.001). We estimated a logistic regression for each adverse outcome with all patients combined, and Table 3 summarizes the association of CHF with adverse outcomes (n ¼ 1,172,632). These models were adjusted for patient age, race/ethnicity, and other confounders. For these analyses, we excluded an additional 1,370 patients due
1222
Turrentine et al
Table 1. Percent of Patients with Congestive Heart Failure by Procedure Procedure
Total Carotid artery stenting Hip fracture repair Lower extremity endovascular revascularization Lower extremity open bypass graft Endovascular aortoiliac repair Open aortoiliac repair Endovascular aneurysm repair Colectomy Abdominal aortic aneurysm repair Carotid endarterectomy Skin/deep tissue flaps and grafts Transurethral resection of the prostate Total hip arthroplasty Lung resection Cystectomy Nephrectomy Proctectomy Ventral hernia repair Spine repair Brain tumor resection Thyroidectomy Esophagectomy Pancreatectomy Abdominoplasty Total knee arthroplasty Hepatectomy Bariatric surgery Appendectomy Hysterectomy/myomectomy Prostatectomy Bladder suspension Gynecological reconstruction Breast reduction Breast reconstruction
J Am Coll Surg
Congestive Heart Failure and Surgical Risk
All patients n %
With CHF, %
1,268,999 807 21,405
100.00 0.06 1.69
0.61 3.59 3.27
9,756
0.77
2.97
17,186 3,678 13,098 20,222 143,166
1.35 0.29 1.03 1.59 11.28
2.62 2.31 2.21 1.67 1.43
7,507 47,397
0.59 3.73
1.40 1.19
6,252
0.49
0.96
19,837 63,222 13,692 4,355 18,529 19,902 159,175 94,224 12,232 58,024 4,074 23,536 3,910 84,129 12,969 94,205 121,451 102,177 23,217 12,726 5,712 9,490 17,737
1.56 4.98 1.08 0.34 1.46 1.57 12.54 7.43 0.96 4.57 0.32 1.85 0.31 6.63 1.02 7.42 9.57 8.05 1.83 1.00 0.45 0.75 1.40
0.88 0.72 0.65 0.64 0.61 0.51 0.34 0.29 0.29 0.27 0.27 0.26 0.26 0.25 0.20 0.18 0.13 0.10 0.10 0.07 0.04 0.04 0.02
CHF, congestive heart failure.
to missing values in one of the covariates. The risks of all but 6 of 23 postoperative complications we considered were significantly increased by CHF. Outcomes not affected by CHF included deep incisional surgical site infection, dehiscence, pulmonary embolism, cerebral vascular accident, deep vein thrombosis, and sepsis.
Table 2. Characteristics of Surgical Patients With and Without Congestive Heart Failure (n ¼ 1,174,002) CHF Characteristic
All patients Age categories, y <55 55e59 60e64 65e69 70e74 75 or older Male sex Race/ethnicity Non-Hispanic White Non-Hispanic Black Hispanic Other/unknown BMI categories, kg/m2 <25 25e29.9 30 Unknown Smoker (current within 1 y) Emergency procedure ASA Class III, IV, or V Functionally independent Dyspnea Ascites Renal failure/dialysis SGOT > 20 UL BUN > 18 mg/dL Creatinine > 1.2 mg/dL INR > 3
n
7,582 739 542 789 945 1,038 3,529 3,980
%
No CHF n %
p Value
0.65 1,166,420 99.35 9.75 7.15 10.41 12.46 13.69 46.54 52.49
519,510 122,427 127,819 123,894 100,591 172,179 495,846
<0.001 44.54 10.50 10.96 10.62 8.62 14.76 42.51 <0.001 <0.001
5,374 70.88
837,066 71.76
986 13.00 368 4.85 854 11.26
106,795 9.16 77,752 6.67 144,807 12.41
2,307 30.43 4,112 54.23 770 10.16 393 5.18
282,520 24.22 700,092 60.02 148,584 12.74 35,224 3.02
1,530 20.18 1,813 23.91
220,653 18.92 <0.001 143,931 12.34 <0.001
7,344 96.86
559,801 47.99 <0.001
<0.001
5,460 3,289 240 892 3,448 4,754
72.01 1,122,644 96.25 <0.001 43.38 94,961 8.14 <0.001 3.17 5,223 0.45 <0.001 11.76 11,420 0.98 <0.001 45.48 354,626 30.40 <0.001 62.70 275,627 23.63 <0.001
3,360 44.32 1,438 18.97
120,852 10.36 <0.001 591,570 50.72 <0.001
ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CHF, congestive heart failure; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase.
The magnitude of association between CHF and postoperative adverse outcomes was largest for cardiac arrest, followed by mortality, unplanned intubation, and ventilator dependence > 48 hours. Conversely, CHF was associated with lower risk of organ space surgical site infection and superficial surgical site infection. Full models for mortality and serious morbidity are shown in Table 4.
Turrentine et al
Vol. 222, No. 6, June 2016
Table 3.
Congestive Heart Failure and Surgical Risk
1223
Association of Congestive Heart Failure with Adverse Outcomes for All Procedures
Adverse outcome
Patients experiencing adverse outcomes All With CHF Without CHF (n ¼ 1,172,632) (n ¼ 7,544) (n ¼ 1,165,088) n % n % n %
Cardiac arrest requiring CPR 30-d mortality Unplanned intubation Ventilator > 48 h Acute renal failure Serious morbidity Any readmission Bleeding transfusions Septic shock Myocardial infarction Pneumonia Unplanned reoperation Renal insufficiency Urinary tract infection CVA with neurologic deficit Return to operating room Pulmonary embolism Dehiscence Sepsis DVT/thrombophlebitis Superficial SSI Deep incisional SSI Organ space SSI
3,854 12,502 12,150 14,240 4,002 101,696 53,666 88,731 7,170 5,245 13,420 26,687 3,717 18,860 2,778 38,229 4,548 5,736 15,109 7,904 25,204 6,745 13,237
0.33 1.07 1.04 1.21 0.34 8.67 4.58 7.57 0.61 0.45 1.14 2.28 0.32 1.61 0.24 3.26 0.39 0.49 1.29 0.67 2.15 0.58 1.13
262 933 553 780 233 2,248 858 1,906 322 197 444 500 113 316 77 756 64 123 265 137 234 69 92
3.47 12.37 7.33 10.34 3.09 29.80 11.37 25.27 4.27 2.61 5.89 6.63 1.50 4.19 1.02 10.02 0.85 1.63 3.51 1.82 3.10 0.91 1.22
3,592 11,569 11,597 13,460 3,769 99,448 52,808 86,825 6,848 5,048 12,976 26,187 3,604 18,544 2,701 37,473 4,484 5,613 14,844 7,767 24,970 6,676 13,145
0.31 0.99 1.00 1.16 0.32 8.54 4.53 7.45 0.59 0.43 1.11 2.25 0.31 1.59 0.23 3.22 0.38 0.48 1.27 0.67 2.14 0.57 1.13
Risk of adverse outcomes for CHF*
Odds ratio (95% CI)
2.003 1.958 1.690 1.610 1.599 1.521 1.394 1.388 1.366 1.357 1.318 1.293 1.277 1.248 1.233 1.231 1.185 1.139 1.083 0.991 0.873 0.786 0.700
(1.742e2.302) (1.800e2.129) (1.534e1.862) (1.468e1.765) (1.380e1.852) (1.439e1.607) (1.294e1.502) (1.308e1.473) (1.205e1.550) (1.167e1.578) (1.187e1.463) (1.174e1.423) (1.049e1.556) (1.109e1.405) (0.974e1.562) (1.135e1.335) (0.919e1.530) (0.943e1.376) (0.951e1.234) (0.830e1.183) (0.762e1.000) (0.615e1.003) (0.566e0.866)
p Value
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.015 <0.001 0.082 <0.001 0.191 0.178 0.228 0.919 0.049 0.053 <0.001
*Adjusted for procedure type, patient age, sex, race/ethnicity, emergency surgery status, ASA Class, BMI, preoperative risk factors (functional dependence, dyspnea, renal failure or dialysis), preoperative lab values (SGOT, BUN, creatinine, INR). ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CHF, congestive heart failure; CVA, cerebrovascular accident; DVT, deep venous thromboembolism; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase; SSI, surgical site infection.
To further investigate the effect of CHF on adverse outcomes for each procedure, we estimated logistic regressions for 4 adverse outcomes (mortality, readmission, reoperation, and serious morbidity) for each procedure separately. The results are summarized in Table 5. This table excluded pancreatectomy, proctectomy, brain tumor resection, and skin/deep tissue flaps and grafts because CHF was not significantly associated with any of the adverse outcomes. Congestive heart failure significantly increased the risk of mortality for 12 of the 19 procedures. Of all procedures, the effect of CHF on mortality risk was greatest for thyroidectomy and lung resection patients, who had a 5-fold increase. Similarly, CHF at least tripled the mortality risk for patients who underwent bariatric surgery, carotid endarterectomy, hysterectomy/myomectomy, and lower extremity endovascular surgery.
The risk of 30-day readmission was significantly increased by CHF for 10 of the 19 procedures, with CHF doubling the risk for patients with transurethral resection of the prostate (TURP), total knee arthroplasty, and carotid endarterectomy. Patients presenting with CHF have a significantly increased risk of unplanned reoperation for 6 of 19 procedures. The risk increased more than 4-fold for bariatric surgery and 2-fold for thyroidectomy, TURP, total knee arthroplasty, and carotid endarterectomy. Serious morbidity was the adverse outcome most affected by CHF, which significantly increased the risk for 18 of 19 procedures shown in Table 5. The exception was lung resection. Congestive heart failure tripled the risk of serious morbidity for thyroidectomy and doubled the risk for TURP, carotid endarterectomy, and bariatric surgery.
Variable
30-d Readmission OR (95% CI) p Value
Reoperation OR (95% CI)
p Value
Serious morbidity OR (95% CI) p Value
1.394 (1.294e1.502)
<0.001
1.293 (1.174e1.423)
<0.001
1.521 (1.439e1.607)
<0.001
1.399 1.619 1.965 2.250 3.342 1.133
(1.270e1.541) (1.482e1.769) (1.808e2.137) (2.071e2.445) (3.107e3.595) (1.088e1.179)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.953 0.979 0.981 1.010 1.073 0.979
(0.923e0.984) (0.949e1.010) (0.950e1.013) (0.976e1.045) (1.041e1.106) (0.961e0.998)
0.003 0.181 0.236 0.563 <0.001 0.032
1.023 1.005 1.003 0.982 0.928 1.140
(0.980e1.069) (0.963e1.050) (0.960e1.048) (0.937e1.029) (0.890e0.968) (1.110e1.171)
0.300 0.813 0.897 0.450 <0.001 <0.001
1.051 1.073 1.127 1.159 1.240 1.003
(1.025e1.077) (1.047e1.100) (1.100e1.155) (1.129e1.189) (1.212e1.268) (0.988e1.018)
<0.001 <0.001 <0.001 <0.001 <0.001 0.700
0.826 (0.769e0.888) 0.835 (0.755e0.924) 0.894 (0.843e0.948)
<0.001 <0.001 <0.001
1.141 (1.108e1.175) 1.004 (0.967e1.043) 0.896 (0.870e0.922)
<0.001 0.832 <0.001
1.147 (1.101e1.195) 0.971 (0.918e1.027) 1.003 (0.964e1.043)
<0.001 0.303 0.885
1.132 (1.106e1.158) 0.923 (0.895e0.952) 1.032 (1.011e1.054)
<0.001 <0.001 0.003
1.049 0.758 0.273 0.168 0.131 0.095 0.212
(0.947e1.162) (0.666e0.863) (0.249e0.300) (0.139e0.203) (0.098e0.175) (0.081e0.112) (0.190e0.238)
0.356 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.537 1.669 0.490 0.392 0.359 0.464 0.497
(1.468e1.610) (1.587e1.754) (0.473e0.509) (0.373e0.412) (0.338e0.382) (0.443e0.487) (0.473e0.522)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.889 1.290 0.355 0.315 0.349 0.167 0.428
(0.824e0.959) (1.199e1.388) (0.336e0.374) (0.293e0.339) (0.320e0.381) (0.154e0.181) (0.398e0.460)
0.002 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.696 1.540 0.297 0.219 0.195 0.185 0.264
(1.640e1.753) (1.483e1.599) (0.289e0.306) (0.211e0.228) (0.185e0.205) (0.178e0.192) (0.254e0.275)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.903 (1.741e2.080)
<0.001
0.433 (0.385e0.487)
<0.001
1.332 (1.211e1.465)
<0.001
1.402 (1.330e1.478)
<0.001
0.527 (0.478e0.581) 0.853 (0.769e0.946)
<0.001 0.003
0.575 (0.539e0.615) 0.776 (0.722e0.833)
<0.001 <0.001
0.669 (0.615e0.729) 1.655 (1.542e1.777)
<0.001 <0.001
0.410 (0.391e0.430) 1.010 (0.965e1.056)
<0.001 0.680
0.510 (0.393e0.662) 0.390 (0.344e0.441)
<0.001 <0.001
1.313 (1.178e1.463) 1.134 (1.074e1.198)
<0.001 <0.001
1.910 (1.687e2.162) 2.169 (2.044e2.302)
<0.001 <0.001
0.497 (0.448e0.551) 0.986 (0.947e1.027)
<0.001 0.499
0.379 (0.320e0.448)
<0.001
1.554 (1.458e1.656)
<0.001
2.405 (2.236e2.587)
<0.001
0.525 (0.494e0.559)
<0.001
0.300 0.320 1.254 0.189
<0.001 <0.001 0.001 <0.001
0.555 0.603 1.363 0.710
<0.001 <0.001 <0.001 <0.001
0.503 0.692 1.085 0.420
<0.001 <0.001 0.083 <0.001
0.504 0.378 0.657 0.380
(0.251e0.357) (0.287e0.356) (1.103e1.425) (0.155e0.232)
(0.531e0.580) (0.580e0.627) (1.280e1.453) (0.663e0.761)
(0.472e0.536) (0.657e0.728) (0.989e1.190) (0.374e0.472)
(0.488e0.520) <0.001 (0.367e0.390) <0.001 (0.621e0.695) <0.001 (0.359e0.402) <0.001 (Continued)
J Am Coll Surg
<0.001
Congestive Heart Failure and Surgical Risk
1.958 (1.800e2.129)
Turrentine et al
Congestive heart failure Age categories [<55 y] 55e59 60e 64 65e69 70e74 75 or older Male [female] Race/ethnicity [non-Hispanic White] Non-Hispanic Black Hispanic Other/unknown Surgical procedure [colectomy] Pancreatectomy Proctectomy Ventral hernia repair Bariatric surgery Thyroidectomy Appendectomy Carotid endarterectomy Abdominal aortic aneurysm repair Endovascular aneurysm repair Open aortoiliac repair Endovascular aortoiliac repair Lower extremity open Lower extremity endovascular Hysterectomy/ myomectomy Spine Brain tumor TURP
30-d Mortality OR (95% CI) p Value
1224
Table 4. Full Regression Models for Mortality and Serious Morbidity, All Procedures Combined (n ¼ 1,172,632)
Variable
30-d Readmission OR (95% CI) p Value
<0.001 <0.001 <0.001 <0.001 <0.001 0.002
0.693 0.456 0.576 0.826 1.105 0.805
(0.647e0.742) (0.437e0.477) (0.551e0.603) (0.781e0.875) (1.003e1.216) (0.749e0.864)
<0.001 <0.001 <0.001 <0.001 0.043 <0.001
3.717 (3.553e3.888)
<0.001
1.016 (0.982e1.050)
4.461 (4.117e4.833)
<0.001
1.424 0.967 1.843 0.350
(1.300e1.561) (0.884e1.057) (1.648e2.062) (0.334e0.366)
1.624 1.942 1.461 1.461 1.553 0.748
(1.548e1.703) (1.798e2.098) (1.405e1.519) (1.396e1.528) (1.480e1.629) (0.714e0.785)
p Value
Serious morbidity OR (95% CI) p Value
(0.450e0.558) (0.310e0.357) (0.578e0.653) (0.395e0.479) (1.820e2.229) (0.835e1.008)
<0.001 <0.001 <0.001 <0.001 <0.001 0.073
0.462 0.255 0.311 0.404 1.074 0.668
(0.438e0.487) (0.246e0.264) (0.300e0.322) (0.387e0.422) (1.002e1.150) (0.634e0.704)
<0.001 <0.001 <0.001 <0.001 0.044 <0.001
0.365
1.720 (1.650e1.792)
<0.001
1.945 (1.901e1.989)
<0.001
1.660 (1.625e1.696)
<0.001
1.668 (1.617e1.721)
<0.001
1.745(1.716e1.775)
<0.001
<0.001 0.460 <0.001 <0.001
0.920 0.901 0.846 0.933
(0.888e0.954) (0.872e0.931) (0.795e0.900) (0.898e0.969)
<0.001 <0.001 <0.001 <0.001
0.917 0.846 0.897 0.899
(0.872e0.965) (0.807e0.886) (0.825e0.976) (0.856e0.944)
0.001 <0.001 0.011 <0.001
0.861 0.854 0.984 0.552
(0.838e0.886) (0.833e0.876) (0.942e1.028) (0.538e0.566)
<0.001 <0.001 0.474 <0.001
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.092 1.330 1.056 1.064 1.165 0.832
(1.060e1.124) (1.251e1.415) (1.036e1.077) (1.040e1.088) (1.132e1.199) (0.816e0.849)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.063 1.512 1.076 1.065 1.098 0.882
(1.021e1.106) (1.405e1.628) (1.047e1.105) (1.032e1.099) (1.056e1.142) (0.858e0.907)
0.003 <0.001 <0.001 <0.001 <0.001 <0.001
1.345 1.458 1.110 1.115 1.293 0.809
(1.317e1.373) (1.392e1.528) (1.094e1.127) (1.096e1.134) (1.265e1.321) (0.797e0.821)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.501 0.333 0.614 0.435 2.014 0.918
Reference categories in brackets. ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CVA, cerebrovascular accident; DVT, deep venous thromboembolism; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase; SSI, surgical site infection; TURP, transurethral repair of the prostate.
Congestive Heart Failure and Surgical Risk
(0.331e0.473) (0.083e0.121) (0.276e0.338) (0.528e0.609) (0.245e0.464) (0.702e0.920)
0.396 0.101 0.305 0.567 0.337 0.804
Reoperation OR (95% CI)
Turrentine et al
Nephrectomy Total knee arthroplasty Total hip arthroplasty Hip fracture Flap Lung resection Emergency surgery [non-emergency] ASA Class III, IV, or V [I or II] BMI [<25 kg/m2] 25e30 >30 Unknown/unmeasured Functionally independent Dyspnea at rest or moderate exertion Renal failure or dialysis SGOT > 20 UL BUN > 18 mg/dL Creatinine > 1.2 mg/dL INR > 3
30-d Mortality OR (95% CI) p Value
Vol. 222, No. 6, June 2016
Table 4. Continued
1225
Procedure
30-d Mortality Odds ratio 95% CI
30-d Readmission Odds ratio 95% CI
Unplanned reoperation Odds ratio 95% CI
Serious morbidity Odds ratio 95% CI
3.869*
(2.490e6.013)
2.071*
(1.575e2.723)
2.225*
(1.493e3.314)
2.385*
(1.913e2.974)
9,745 59,006 94,193 102,144
3.583* 5.101* 4.249* 3.673*
(2.196e5.845) (2.028e12.828) (1.435e12.586) (1.233e10.946)
1.882* 1.216 1.611 1.931*
(1.423e2.488) (0.614e2.412) (0.867e2.994) (1.068e3.492)
1.443* 2.831* 4.368* 2.313
(1.043e1.997) (1.397 e5.737) (2.333e8.178) (0.927e5.771)
1.730* 3.090* 2.138* 1.946*
(1.304e2.295) (2.011e4.749) (1.345e3.397) (1.203e3.148)
19,836 84,098 21,386 16,804 63,160 121,443 94,151 143,016 12,904 159,148
0.389 2.503 2.102* 1.752* 2.682* 2.122* 1.564 1.788* 1.627* 1.529
(0.089e1.701) (0.747e8.385) (1.683e2.624) (1.107e2.774) (1.875e3.836) (1.058e4.255) (0.772e3.169) (1.578e2.025) (1.090e2.427) (0.987e2.370)
2.831* 2.276* 1.482* 1.375* 1.204 1.482 1.831* 1.009 1.344 1.686*
(1.887e4.246) (1.473e3.516) (1.174e1.869) (1.044e1.811) (0.877e1.653) (0.856e2.569) (1.284e2.609) (0.854e1.192) (0.907e1.993) (1.284e2.214)
2.729* 2.348* 1.042 1.324 1.066 0.870 1.219 1.133 1.355 1.485
(1.288e5.780) (1.185e4.649) (0.651e1.667) (0.992e1.769) (0.639 e1.777) (0.312e2.426) (0.704e2.112) (0.946e1.358) (0.934e1.966) (0.962e2.292)
2.471* 1.877* 1.525* 1.336* 1.308* 1.575* 1.648* 1.267* 1.778* 1.411*
(1.694e3.605) (1.243e2.833) (1.272e1.829) (1.083e1.648) (1.014e1.687) (1.069e2.319) (1.210e2.244) (1.151e1.395) (1.379e2.293) (1.126e1.768)
20,136 13,688 18,523
1.433 5.059* 1.976
(0.881e2.330) (2.657e9.632) (0.710e5.503)
1.615* 0.994 1.287
(1.110e2.350) (0.455 e2.171) (0.679e2.442)
1.133 0.950 1.191
(0.669e1.920) (0.344e2.628) (0.423e3.356)
1.554* 1.517 1.787*
(1.171e2.061) (0.930e2.476) (1.139e2.806)
7,450
1.235
(0.691e2.208)
1.134
(0.454e2.835)
1.596
(0.855e2.980)
1.639*
(1.087e2.47)
Congestive Heart Failure and Surgical Risk
47,336
Turrentine et al
Carotid endarterectomy Lower extremity endovascular Thyroidectomy Bariatric surgery Hysterectomy/myomectomy Transurethral repair of prostate Total knee arthroplasty Hip fracture repair Lower extremity open Total hip arthroplasty Appendectomy Spine repair Colectomy Open aortoiliac repair Ventral hernia repair Endovascular aneurysm repair Lung resection Nephrectomy Aortic abdominal aneurysm repair
n
1226
Table 5. Adjusted Odds Ratios and 95% Confidence Intervals of Patients With Preoperative Congestive Heart Failure Experiencing Adverse Outcomes Compared With Those Without, Selected Procedures
Adjusted for patient age, sex, race/ethnicity, emergency surgery status, American Society of Anesthesiologists Class, BMI, preoperative risk factors (functional dependence, dyspnea, renal failure, or dialysis), preoperative lab values (serum glutamic oxaloacetic transaminase, blood urea nitrogen, creatinine, international normalized ratio). All procedures that had at least 1 outcome with p < 0.05 were included in this table. *Odds ratios with p values < 0.05.
J Am Coll Surg
Vol. 222, No. 6, June 2016
Turrentine et al
The risks of all 4 adverse outcomes for carotid endarterectomy and lower extremity endovascular surgery were increased by CHF. Seven procedures were adversely affected by CHF in 3 of 4 outcomes, including bariatric and thyroidectomy surgery which more than doubled the risk of mortality, reoperation, and serious morbidity; and TURP and total knee arthroplasty, which significantly increased the risk of readmission, reoperation, and serious morbidity. Sensitivity analyses revealed the addition of operative duration and preoperative cardiac variables did not significantly change the association between CHF and adverse outcomes (not shown). However, the magnitude of association was always stronger with inclusion of operative duration, and it remained similar with or without preoperative cardiac variables in the models. Additional inclusion of severity indicators, such as operative duration or outpatient surgery status, did not change the association between the two.
DISCUSSION Congestive heart failure significantly increased the risk of serious morbidity, readmission, reoperation, and mortality in our data. The pervasive effect of CHF on surgical outcomes after noncardiac operations may not be fully appreciated by surgeons. Most procedures we examined had at least 1 of 23 postoperative complications significantly affected by CHF. Awareness of CHF’s impact on surgical outcomes is important to both surgeons and patients. For surgeons, it underscores the importance of optimal perioperative care, particularly fluid management, in this vulnerable population. Pulse oximeter plethysmography accurately measures volume and can be helpful in monitoring volume responsiveness and the need for fluid.13 Pleth Variability Index, an algorithm allowing continuous monitoring of variations in pulse oximetry,14 promotes goal-directed fluid management in Early Recovery After Surgery (ERAS) protocols15 and could mitigate risk of fluid overload in the CHF patient undergoing surgery. Patients, particularly those undergoing elective procedures, could potentially modify risks by avoiding conditions that contribute to or exacerbate CHF: smoking, obesity, diets high in saturated fats, sugar, and sodium. In addition, patients with heart disease or other chronic conditions can be encouraged to follow their treatment plans, stay active, limit alcohol, and get enough rest.16 Patients with CHF, compared with those without the condition, are 96% more likely to die in the 30 days after surgery. Patients undergoing thyroidectomy had the highest odds of mortality. Although at first this finding is
Congestive Heart Failure and Surgical Risk
1227
surprising, the fact that patients are undergoing surgery with acute heart failure suggests an urgent nature to the procedure. Further explanation may be that “Thyroid hormone influences every cell, tissue, and organ in the body, and its homeostasis is essential to the optimal functioning of the heart,”17 and thyroid hormone disarray is a risk factor for heart failure.18 Similarly, CHF was significantly associated with postoperative adverse events in thyroidectomy and parathyroidectomy patients in 2007/2008 ACS NSQIP PUF data.19 Patients with CHF undergoing bariatric procedures had a 4-fold increase in unplanned reoperation. Patients requiring dialysis were significantly more likely to require reoperation.20 This may explain the higher rate of reoperation in our cohort; our bariatric patients with CHF had significantly higher rates of renal failure compared with patients without CHF. This differs from work that did not identify CHF as a risk for return to the operating room in a subset of ACS NSQIP 2007 to 2009 bariatric PUF procedures.21 However, only laparoscopic gastric bypass and adjustable gastric band were analyzed; we also included data for open gastric bypass and sleeve gastrectomy. Patients with CHF undergoing more complex procedures may be at higher risk for complications. Patients with CHF who underwent lower extremity endovascular revascularization were more likely to experience all 4 adverse outcomes. This is unexpected because lower extremity bypass is recommended for patients who are not candidates for angioplasty or have a failed angioplasty, and one would think outcomes for open repair would be more adversely affected compared with endovascular repair. However, this finding may be explained in part by findings that CHF independently predicted patency loss after endovascular treatment.22 The subject of patient selection, evaluation, and indication for treatment of lower extremity endovascular surgery bears further investigation. Congestive heart failure was associated with serious morbidity collectively and the majority of postoperative complications individually. In our study population, CHF had the biggest increases on the risks of cardiac arrest, unplanned intubation, and mechanical ventilation > 48 hours. Excessive fluid volume administration and decreased cardiac output during the perioperative period may explain these complications.23 Similarly, CHF has been associated with severe perioperative cardiac complications, including cardiac arrest, in noncardiac surgical procedures in a Veterans Affairs hospital.24 A multifactorial risk index for predicting postoperative respiratory failure in men found CHF to be the only significant cardiac status predictor of respiratory failure in a study of
1228
Turrentine et al
Congestive Heart Failure and Surgical Risk
81,719 patients who underwent noncardiac procedures within the Department of Veterans Affairs NSQIP.25 Age and race/ethnicity were the most prominent characteristics distinguishing patients with CHF from those without the comorbid condition. This is not surprising because age is a risk factor for surgery26 and is a component of risk stratification in a number of indices27-29 and guidelines.30 Advanced age independently predicted morbidity31 and mortality32 in patients with CHF.33 African Americans with CHF, compared with whites with CHF, have higher rates of readmission, reoperation, and serious morbidity, but lower mortality rates. Nonwhites, compared with whites, have lower risk of mortality, but differed from our study in that nonwhites were less likely to experience readmission.34 Perhaps reduced mortality among African Americans, in patients hospitalized with CHF, is due to differences in disease severity or duration.35 In addition, differences in quality of care, type of hospital, cardiac testing, and concurrent ischemic heart disease between African Americans and Caucasians may provide an explanation.36 Limitations In our study, the diagnosis of CHF depended primarily on history and physical findings recorded in the medical record. Our ACS NSQIP definition limited us to events of acute CHF and therefore prevented evaluation of the influence of chronic CHF on surgical outcomes. In addition, the degree of CHF could not be ascertained, nor could preoperative management of the condition. Another limitation is that we used ACS NSQIP procedure-targeted groups, which precluded assignment of minor or major classifications to individual procedures. Procedure-targeted groups do not include data on some operations such as small bowel resections and cholecystectomy, which could have been affected by CHF. This large dataset contained few patients with CHF who underwent certain operations such as gynecologic reconstruction and carotid artery stenting.
CONCLUSIONS Although the prevalence of acute CHF is low among surgical patients, it significantly increases serious morbidity, readmission, reoperation, mortality, and most other postoperative complications. Surgeons should manage CHF carefully when considering surgery. Author Contributions Study conception and design: Turrentine, Jones Acquisition of data: Turrentine, Sohn, Jones
J Am Coll Surg
Analysis and interpretation of data: Turrentine, Sohn, Jones Drafting of manuscript: Turrentine, Sohn Critical revision: Jones REFERENCES 1. Grasso AW, Jaber WA. Cardiac risk stratification for noncardiac surgery. Available at: http://www.clevelandclinicmeded. com/medicalpubs/diseasemanagement/cardiology/cardiac-riskstratification-for-noncardiac-surgery/Default.htm. Accessed February 11, 2015. 2. Sheer AJ, Heckman JE, Schneider EB, et al. Congestive heart failure and chronic obstructive pulmonary disease predict poor surgical outcomes in older adults undergoing elective diverticulitis surgery. Dis Colon Rectum 2011;54:1430e1437. 3. Hernandez AF, Whellan DJ, Stroud S, et al. Outcomes in heart failure patients after major noncardiac surgery. J Am Coll Cardiol 2004;44:1446e1453. 4. Rao A, Polanco A, Qui S, et al. Safety of outpatient laparoscopic cholecystectomy in the elderly: Analysis of 15,248 patients using the NSQIP database. J Am Coll Surg 2013; 217:1038e1043. 5. Annamalai A, Kakarla VR, Nandipati K. Predictors of mortality following pancreaticoduodenectomy for periampullary cancer. OA Surg 2014;2:2e7. 6. Maile MD, Engoren MC, Tremper KK, et al. Worsening preoperative heart failure is associated with mortality and noncardiac complications, but not myocardial infarction after noncardiac surgery; A retrospective cohort study. Soc Cardiovasc Anesthesiol 2014;119:522e532. 7. Parante D, Luis C, Veiga D, et al. Congestive heart failure as a determinant of postoperative delirium. Portuguese J Cardiol 2013;32:665e671. 8. Turrentine FE, Denlinger CE, Simpson VB, et al. Morbidity, mortality, cost, and survival estimates of gastrointestinal anastomotic leaks. J Am Coll Surg 2015;220:195e206. 9. American College of Surgeons. ACS NSQIP Participant Use Data File. Available at: http://site.acsnsqip.org/participantuse-data-file/. Accessed February 11, 2015. 10. ACS NSQIP. ACS NSQIP variables and definitions. In: ACS NSQIP operations manual. Chicago: American College of Surgeons; 2014:47. 11. ACS NSQIP. ACS NSQIP: Procedure targeted algorithm worksheet. Available at: http://reports.nsqip.facs.org/acsmain/private/ documents/operationsmanual/index.jsp. Accessed March 20, 2015. 12. ACS NSQIP. Definitions. In: ACS NSQIP semiannual report: July1, 2013 through June 30, 2015. Chicago, Illinois: ACS NSQIP; 2015:32. 13. Marik P, Monnet X, Teboul J. Hemodynamic parameters to guide fluid therapy. Ann Intensive Care 2011;1:1e9. 14. Cannesson M, Desebbe O, Rosamel P, et al. Pleth variability index to monitor the respiratory variations in the pulse oximeter plethysmographic waveform amplitude and predict fluid responsiveness in the operating theatre. Br J Anaesthesia 2008;101:200e206. 15. Thiele R, Rea K, Turrentine F, et al. Standardization of care: Impact of an enhanced recovery protocol on length of stay, complications, and direct costs after colorectal surgery. J Am Coll Surg 2015;220:430e443.
Vol. 222, No. 6, June 2016
Turrentine et al
16. Eckel R, Jakicic J, Ard J, et al. 2013 AHA/ACC guidelines on lifestyle management to reduce cardiovascular risk: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2960e2984. 17. Mitchell JE, Hellkamp AS, Mark DB, et al. Thyroid function in heart failure and impact on mortality. J Am Coll Cardiol 2013;1:48e55. 18. Schmidt-Ott UM, Ascheim DD. Thyroid hormone and heart failure. Current Heart Failure Reports 2006;3:114e119. 19. Gupta PK, Smith RB, Gupta H, et al. Outcomes after thyroidectomy and parathyroidectomy. Head Neck 2012;34:477e484. 20. Gajdos C, Hawn M, Kile D, et al. Risk of major nonemergent inpatient general surgical procedures in patients on long-term dialysis. JAMA Surg 2013;148:137e143. 21. Nandipati K, Lin E, Husain F, et al. Factors predicting the increased risk for return to the operating room in bariatric patients: A NSQIP database study. Surg Endosc 2013;27: 1172e1177. 22. Meltzer AJ, Shrikhande G, Gallagher KA, et al. Heart failure is associated with reduced patency after endovascular intervention for symptomatic peripheral arterial disease. J Vasc Surg 2012;55:353e362. 23. Pradeep A, Rajagopalam S, Kolli HK, et al. High volumes of intravenous fluid during cardiac surgery are associated with increased mortality. HSR Proc Intensive Care Cardiovasc Anesthesia 2010;2:287e296. 24. Kumar R, McKinney WP, Raj G, et al. Adverse cardiac events after surgery: Assessing risk in a veteran population. J Gen Intern Med 2001;16:507e518. 25. Arozullah AM, Daley J, Henderson WG, Khuri SF. Multifactorial risk index for predicting postoperative respiratory failure in men after major noncardiac surgery. Ann Surg 2000;2:242e253. 26. Turrentine FE, Wang H, Simpson VB, Jones RS. Surgical risk factors, morbidity, and mortality in elderly patients. J Am Coll Surg 2006;203:865e877. 27. Goldman L, Caldera DL, Nussbauam SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845e850.
Congestive Heart Failure and Surgical Risk
1229
28. Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery. Report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. Committee on Perioperative Cardiovascular Evaluation for Noncardiac Surgery. Circulation 1996;93: 1278e1317. 29. Detsky AS, Abrams HB, Frobath N, et al. Cardiac assessment for patients undergoing noncardiac surgery. A multifactorial clinical risk index. Arch Intern Med 1986;146: 2131e2134. 30. The Expert Work Group Members. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;129: S49eS73. 31. Polanczyk CA, Marcantonio E, Goldman L, et al. Impact of age on perioperative complications and length of stay in patients undergoing noncardiac surgery. Ann Intern Med 2001;134:637e643. 32. Feringa HH, Bax JJ, Karagiannis SE, et al. Elderly patients undergoing major vascular surgery: Risk factors and medication associated with risk reduction. Arch Gerontol Geriatrics 2009;48:116e120. 33. Healy KO, Waksmonski CA, Altman RK, et al. Perioperative outcome and long term mortality for heart failure patients undergoing intermediate and high risk noncardiac surgery: Impact of left ventricular ejection fraction. Congestive Heart Failure 2010;16:45e49. 34. Hammill BG, Curtis LH, Bennett-Guerrero E, et al. Impact of heart failure on patients undergoing major noncardiac surgery. Anesthesiology 2008;108:559e567. 35. Joshi AV, D’Souza AO, Madhavan S. Differences in hospital length-of-stay, charges, and mortality in congestive heart failure patients. Congestive Heart Failure 2004;10: 76e84. 36. Philbin EF, DiSalvo TG. Influence of race and gender on care processes, resource use, and hospital-based outcomes in congestive heart failure. Am J Cardiol 1998;82:76e81.
PhD, RN,
Min-Woong Sohn,
PhD,
Rayford Scott Jones,
MD, FACS
Congestive heart failure (CHF) predicts surgical morbidity and mortality. However, few studies evaluate CHF’s impact on noncardiac operations. Because of CHFs serious threat to health and survival, surgeons must understand risks CHF poses to patients undergoing a diverse array of operations. STUDY DESIGN: We used 2009 to 2013 American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Participant Use Files to estimate the risk of serious morbidity, reoperation, readmission, mortality, and other postoperative complications associated with preoperative diagnosis of CHF. Multivariable logistic regression analysis provided odds ratios (OR) and 95% confidence intervals (CI) for outcomes in 34 ACS NSQIP procedure groups, controlling for age, sex, race, emergency surgery status, American Society of Anesthesiologists Classification, body mass index, and selected laboratory values. RESULTS: Unadjusted ORs indicate adverse effects of CHF on surgical outcomes for most procedures considered. When adjusted for age and other confounders, CHF persists with adverse effects on most outcomes, including serious morbidity (OR 1.52, 95% CI, 1.44 to 1.61; p < 0.001); reoperation (OR 1.29, 95% CI, 1.17 to 1.42; p < 0.001); readmission (OR 1.39, 95% CI, 1.29 to 1.50; p < 0.001); and 30-day mortality (OR 1.96, 95% CI 1.80 to 2.13; p < 0.001). The impact of CHF on morbidity and mortality substantially affected those undergoing carotid endarterectomy and lower extremity endovascular repair. Cardiac arrest, mortality, unplanned intubation, and ventilator > 48 hours were complications most affected by CHF. CONCLUSIONS: Congestive heart failure strongly predicts serious morbidity, unplanned reoperation, readmission, and surgical mortality for noncardiac operations. Surgeons must pay particular attention to recognizing CHF and optimizing perioperative management when considering surgery. (J Am Coll Surg 2016;222:1220e1229. 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.)
BACKGROUND:
Congestive heart failure (CHF), a cause of morbidity and mortality after noncardiac surgery,1 predicts mortality for Medicare beneficiaries2,3 and for patients undergoing a range of operations, from outpatient laparoscopic cholecystectomy4 Disclosure Information: Nothing to disclose. Disclaimer: American College of Surgeons (ACS) NSQIP and the hospitals participating in the ACS NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Presented at the American College of Surgeons 101st Annual Clinical Congress, Chicago, IL, October 2015. Received December 8, 2015; Revised February 2, 2016; Accepted February 29, 2016. From the Departments of Surgery (Turrentine, Jones) and Public Health Sciences (Sohn), University of Virginia, Charlottesville, VA. Correspondence address: Rayford Scott Jones, MD, FACS, Department of Surgery, University of Virginia, Box 800709, Charlottesville, VA 22908. email: [email protected]
ª 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.
to pancreaticoduodenectomy for cancer.5 Even though CHF increases the risk of a variety of surgical complications2,3,6 including postoperative delirium7 and anastomotic leaks,8 there is no systematic examination of the kinds of noncardiac operations and adverse outcomes it affects. In this study, we conducted a comprehensive analysis of the impact of CHF on serious morbidity, reoperation, readmission, mortality, and other postoperative complications for an extensive array of surgical procedures. Undergoing surgery with the comorbidity of CHF should alert clinicians to increased risk that may be mitigated through preoperative counseling and vigilant perioperative management.
METHODS We used the 2009 through 2013 American College of Surgeons National Quality Improvement Program
1220
http://dx.doi.org/10.1016/j.jamcollsurg.2016.02.025 ISSN 1072-7515/16
Vol. 222, No. 6, June 2016
Turrentine et al
(ACS NSQIP) Participant Use Files (PUF)9 to identify patients with CHF undergoing surgery. The PUF is an approved Public Data Set at the University of Virginia, so no additional institutional review or approval was required. The ACS NSQIP CHF variable includes patients with a new diagnosis of CHF or a recent exacerbation of chronic CHF. The ACS NSQIP defines CHF as “the inability of the heart to pump a sufficient quantity of blood to meet the metabolic needs of the body or a situation where the heart can do so only at increased ventricular filling pressure.” Common CHF manifestations include exercise intolerance due to dyspnea or fatigue; orthopnea; paroxysmal nocturnal dyspnea; increased jugular venous pressure; pulmonary rales; cardiomegaly; pulmonary vascular engorgement; and pulmonary edema. To meet ACS NSQIP criteria, CHF must be newly diagnosed, or a diagnosis of chronic CHF requires current signs or symptoms in the 30 days before the principal operation.10 We grouped operations into 34 targeted procedure categories as outlined by the ACS NSQIP.11 Congestive heart failure occurred rarely for many of these procedures. Therefore, we excluded groups with 30 or fewer patients with CHF or <0.5% of the total procedure volume from further analyses. These groups included hepatectomy, esophagectomy, carotid artery stenting, gynecologic reconstruction, bladder suspension, prostatectomy, cystectomy, breast reduction, breast reconstruction, and abdominoplasty. We used the ACS NSQIP definition of serious morbidity, which includes 1 or more of the following occurrences: cardiac arrest, myocardial infarction, pneumonia, progressive renal insufficiency, acute renal failure, venous thromboembolism, return to the operating room, deep incisional surgical site infection, organ space surgical site infection, systemic sepsis (sepsis or septic shock), unplanned intubation, urinary tract infection, or wound disruption. When present at the time of surgery, superficial, deep, and organ space surgical site infection, pneumonia, unplanned intubation, sepsis, progressive renal insufficiency, acute renal failure, and urinary tract infection were not considered as postoperative complications.12 Renal failure and dialysis were combined into 1 variable to indicate presence of chronic kidney disease. A priori we included independent variables, age, sex, ethnicity, emergent status, functional status, American Society of Anesthesiologists (ASA) classification, BMI, dyspnea, renal dysfunction, and laboratory values serum glutamic oxaloacetic transaminase (SGOT), blood urea nitrogen (BUN), creatinine, and international normalized ratio (INR).
Congestive Heart Failure and Surgical Risk
1221
Study outcomes included serious morbidity, readmission, reoperation, and mortality within 30 days of discharge. We analyzed these outcomes together as well as separately for each of the 34 ACS NSQIP targeted procedure groups using multivariable logistic regression. In adjusted models, we controlled for the following potential confounders: age, sex, race, emergency surgery status, ASA score (3, 4, 5 vs 1 or 2), BMI, and selected preoperative laboratory values. Sensitivity analyses evaluated how and whether the addition of preoperative cardiac variables (previous percutaneous coronary intervention, history of myocardial infarction, previous cardiac surgery, history of angina, revascularization for peripheral vascular disease) diminished the magnitude of association between CHF and study outcomes. We also included operative duration and outpatient surgery status as 2 other measures of patient severity in our sensitivity analysis models. Stata SE version 13 (Statacorp) was used to conduct the analysis.
RESULTS Of 2.3 million records in the 2009 through 2013 ACS NSQIP PUFs, we used data for 1,268,999 patients who met our inclusion criteria. Only 0.61% of patients who underwent surgery met the ACS NSQIP definition of CHF, with the highest rates occurring in patients having surgery for carotid artery stenting, hip fracture repair, and lower extremity endovascular revascularization, 3.59%, 3.27%, and 2.97%, respectively (Table 1). Patients undergoing breast reconstruction, breast reduction, and gynecologic reconstruction presented with the fewest cases of CHF: 0.02%, 0.04%, and 0.04%, respectively. Table 2 shows characteristics of patients who met inclusion criteria and presented with and without CHF (n ¼ 1,174,002). Patients were significantly different on all characteristics, with age and race/ethnicity being the most prominent characteristics that distinguished patients with CHF from those without the comorbid condition. Congestive heart failure increased exponentially with increasing age; of the patients with CHF, 46.54% were 75 years of age or older compared with 14.76% of those without CHF (p < 0.001). Thirteen percent of patients with CHF were non-Hispanic black compared with 9.16% of patients without the comorbidity (p < 0.001). We estimated a logistic regression for each adverse outcome with all patients combined, and Table 3 summarizes the association of CHF with adverse outcomes (n ¼ 1,172,632). These models were adjusted for patient age, race/ethnicity, and other confounders. For these analyses, we excluded an additional 1,370 patients due
1222
Turrentine et al
Table 1. Percent of Patients with Congestive Heart Failure by Procedure Procedure
Total Carotid artery stenting Hip fracture repair Lower extremity endovascular revascularization Lower extremity open bypass graft Endovascular aortoiliac repair Open aortoiliac repair Endovascular aneurysm repair Colectomy Abdominal aortic aneurysm repair Carotid endarterectomy Skin/deep tissue flaps and grafts Transurethral resection of the prostate Total hip arthroplasty Lung resection Cystectomy Nephrectomy Proctectomy Ventral hernia repair Spine repair Brain tumor resection Thyroidectomy Esophagectomy Pancreatectomy Abdominoplasty Total knee arthroplasty Hepatectomy Bariatric surgery Appendectomy Hysterectomy/myomectomy Prostatectomy Bladder suspension Gynecological reconstruction Breast reduction Breast reconstruction
J Am Coll Surg
Congestive Heart Failure and Surgical Risk
All patients n %
With CHF, %
1,268,999 807 21,405
100.00 0.06 1.69
0.61 3.59 3.27
9,756
0.77
2.97
17,186 3,678 13,098 20,222 143,166
1.35 0.29 1.03 1.59 11.28
2.62 2.31 2.21 1.67 1.43
7,507 47,397
0.59 3.73
1.40 1.19
6,252
0.49
0.96
19,837 63,222 13,692 4,355 18,529 19,902 159,175 94,224 12,232 58,024 4,074 23,536 3,910 84,129 12,969 94,205 121,451 102,177 23,217 12,726 5,712 9,490 17,737
1.56 4.98 1.08 0.34 1.46 1.57 12.54 7.43 0.96 4.57 0.32 1.85 0.31 6.63 1.02 7.42 9.57 8.05 1.83 1.00 0.45 0.75 1.40
0.88 0.72 0.65 0.64 0.61 0.51 0.34 0.29 0.29 0.27 0.27 0.26 0.26 0.25 0.20 0.18 0.13 0.10 0.10 0.07 0.04 0.04 0.02
CHF, congestive heart failure.
to missing values in one of the covariates. The risks of all but 6 of 23 postoperative complications we considered were significantly increased by CHF. Outcomes not affected by CHF included deep incisional surgical site infection, dehiscence, pulmonary embolism, cerebral vascular accident, deep vein thrombosis, and sepsis.
Table 2. Characteristics of Surgical Patients With and Without Congestive Heart Failure (n ¼ 1,174,002) CHF Characteristic
All patients Age categories, y <55 55e59 60e64 65e69 70e74 75 or older Male sex Race/ethnicity Non-Hispanic White Non-Hispanic Black Hispanic Other/unknown BMI categories, kg/m2 <25 25e29.9 30 Unknown Smoker (current within 1 y) Emergency procedure ASA Class III, IV, or V Functionally independent Dyspnea Ascites Renal failure/dialysis SGOT > 20 UL BUN > 18 mg/dL Creatinine > 1.2 mg/dL INR > 3
n
7,582 739 542 789 945 1,038 3,529 3,980
%
No CHF n %
p Value
0.65 1,166,420 99.35 9.75 7.15 10.41 12.46 13.69 46.54 52.49
519,510 122,427 127,819 123,894 100,591 172,179 495,846
<0.001 44.54 10.50 10.96 10.62 8.62 14.76 42.51 <0.001 <0.001
5,374 70.88
837,066 71.76
986 13.00 368 4.85 854 11.26
106,795 9.16 77,752 6.67 144,807 12.41
2,307 30.43 4,112 54.23 770 10.16 393 5.18
282,520 24.22 700,092 60.02 148,584 12.74 35,224 3.02
1,530 20.18 1,813 23.91
220,653 18.92 <0.001 143,931 12.34 <0.001
7,344 96.86
559,801 47.99 <0.001
<0.001
5,460 3,289 240 892 3,448 4,754
72.01 1,122,644 96.25 <0.001 43.38 94,961 8.14 <0.001 3.17 5,223 0.45 <0.001 11.76 11,420 0.98 <0.001 45.48 354,626 30.40 <0.001 62.70 275,627 23.63 <0.001
3,360 44.32 1,438 18.97
120,852 10.36 <0.001 591,570 50.72 <0.001
ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CHF, congestive heart failure; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase.
The magnitude of association between CHF and postoperative adverse outcomes was largest for cardiac arrest, followed by mortality, unplanned intubation, and ventilator dependence > 48 hours. Conversely, CHF was associated with lower risk of organ space surgical site infection and superficial surgical site infection. Full models for mortality and serious morbidity are shown in Table 4.
Turrentine et al
Vol. 222, No. 6, June 2016
Table 3.
Congestive Heart Failure and Surgical Risk
1223
Association of Congestive Heart Failure with Adverse Outcomes for All Procedures
Adverse outcome
Patients experiencing adverse outcomes All With CHF Without CHF (n ¼ 1,172,632) (n ¼ 7,544) (n ¼ 1,165,088) n % n % n %
Cardiac arrest requiring CPR 30-d mortality Unplanned intubation Ventilator > 48 h Acute renal failure Serious morbidity Any readmission Bleeding transfusions Septic shock Myocardial infarction Pneumonia Unplanned reoperation Renal insufficiency Urinary tract infection CVA with neurologic deficit Return to operating room Pulmonary embolism Dehiscence Sepsis DVT/thrombophlebitis Superficial SSI Deep incisional SSI Organ space SSI
3,854 12,502 12,150 14,240 4,002 101,696 53,666 88,731 7,170 5,245 13,420 26,687 3,717 18,860 2,778 38,229 4,548 5,736 15,109 7,904 25,204 6,745 13,237
0.33 1.07 1.04 1.21 0.34 8.67 4.58 7.57 0.61 0.45 1.14 2.28 0.32 1.61 0.24 3.26 0.39 0.49 1.29 0.67 2.15 0.58 1.13
262 933 553 780 233 2,248 858 1,906 322 197 444 500 113 316 77 756 64 123 265 137 234 69 92
3.47 12.37 7.33 10.34 3.09 29.80 11.37 25.27 4.27 2.61 5.89 6.63 1.50 4.19 1.02 10.02 0.85 1.63 3.51 1.82 3.10 0.91 1.22
3,592 11,569 11,597 13,460 3,769 99,448 52,808 86,825 6,848 5,048 12,976 26,187 3,604 18,544 2,701 37,473 4,484 5,613 14,844 7,767 24,970 6,676 13,145
0.31 0.99 1.00 1.16 0.32 8.54 4.53 7.45 0.59 0.43 1.11 2.25 0.31 1.59 0.23 3.22 0.38 0.48 1.27 0.67 2.14 0.57 1.13
Risk of adverse outcomes for CHF*
Odds ratio (95% CI)
2.003 1.958 1.690 1.610 1.599 1.521 1.394 1.388 1.366 1.357 1.318 1.293 1.277 1.248 1.233 1.231 1.185 1.139 1.083 0.991 0.873 0.786 0.700
(1.742e2.302) (1.800e2.129) (1.534e1.862) (1.468e1.765) (1.380e1.852) (1.439e1.607) (1.294e1.502) (1.308e1.473) (1.205e1.550) (1.167e1.578) (1.187e1.463) (1.174e1.423) (1.049e1.556) (1.109e1.405) (0.974e1.562) (1.135e1.335) (0.919e1.530) (0.943e1.376) (0.951e1.234) (0.830e1.183) (0.762e1.000) (0.615e1.003) (0.566e0.866)
p Value
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.015 <0.001 0.082 <0.001 0.191 0.178 0.228 0.919 0.049 0.053 <0.001
*Adjusted for procedure type, patient age, sex, race/ethnicity, emergency surgery status, ASA Class, BMI, preoperative risk factors (functional dependence, dyspnea, renal failure or dialysis), preoperative lab values (SGOT, BUN, creatinine, INR). ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CHF, congestive heart failure; CVA, cerebrovascular accident; DVT, deep venous thromboembolism; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase; SSI, surgical site infection.
To further investigate the effect of CHF on adverse outcomes for each procedure, we estimated logistic regressions for 4 adverse outcomes (mortality, readmission, reoperation, and serious morbidity) for each procedure separately. The results are summarized in Table 5. This table excluded pancreatectomy, proctectomy, brain tumor resection, and skin/deep tissue flaps and grafts because CHF was not significantly associated with any of the adverse outcomes. Congestive heart failure significantly increased the risk of mortality for 12 of the 19 procedures. Of all procedures, the effect of CHF on mortality risk was greatest for thyroidectomy and lung resection patients, who had a 5-fold increase. Similarly, CHF at least tripled the mortality risk for patients who underwent bariatric surgery, carotid endarterectomy, hysterectomy/myomectomy, and lower extremity endovascular surgery.
The risk of 30-day readmission was significantly increased by CHF for 10 of the 19 procedures, with CHF doubling the risk for patients with transurethral resection of the prostate (TURP), total knee arthroplasty, and carotid endarterectomy. Patients presenting with CHF have a significantly increased risk of unplanned reoperation for 6 of 19 procedures. The risk increased more than 4-fold for bariatric surgery and 2-fold for thyroidectomy, TURP, total knee arthroplasty, and carotid endarterectomy. Serious morbidity was the adverse outcome most affected by CHF, which significantly increased the risk for 18 of 19 procedures shown in Table 5. The exception was lung resection. Congestive heart failure tripled the risk of serious morbidity for thyroidectomy and doubled the risk for TURP, carotid endarterectomy, and bariatric surgery.
Variable
30-d Readmission OR (95% CI) p Value
Reoperation OR (95% CI)
p Value
Serious morbidity OR (95% CI) p Value
1.394 (1.294e1.502)
<0.001
1.293 (1.174e1.423)
<0.001
1.521 (1.439e1.607)
<0.001
1.399 1.619 1.965 2.250 3.342 1.133
(1.270e1.541) (1.482e1.769) (1.808e2.137) (2.071e2.445) (3.107e3.595) (1.088e1.179)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.953 0.979 0.981 1.010 1.073 0.979
(0.923e0.984) (0.949e1.010) (0.950e1.013) (0.976e1.045) (1.041e1.106) (0.961e0.998)
0.003 0.181 0.236 0.563 <0.001 0.032
1.023 1.005 1.003 0.982 0.928 1.140
(0.980e1.069) (0.963e1.050) (0.960e1.048) (0.937e1.029) (0.890e0.968) (1.110e1.171)
0.300 0.813 0.897 0.450 <0.001 <0.001
1.051 1.073 1.127 1.159 1.240 1.003
(1.025e1.077) (1.047e1.100) (1.100e1.155) (1.129e1.189) (1.212e1.268) (0.988e1.018)
<0.001 <0.001 <0.001 <0.001 <0.001 0.700
0.826 (0.769e0.888) 0.835 (0.755e0.924) 0.894 (0.843e0.948)
<0.001 <0.001 <0.001
1.141 (1.108e1.175) 1.004 (0.967e1.043) 0.896 (0.870e0.922)
<0.001 0.832 <0.001
1.147 (1.101e1.195) 0.971 (0.918e1.027) 1.003 (0.964e1.043)
<0.001 0.303 0.885
1.132 (1.106e1.158) 0.923 (0.895e0.952) 1.032 (1.011e1.054)
<0.001 <0.001 0.003
1.049 0.758 0.273 0.168 0.131 0.095 0.212
(0.947e1.162) (0.666e0.863) (0.249e0.300) (0.139e0.203) (0.098e0.175) (0.081e0.112) (0.190e0.238)
0.356 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.537 1.669 0.490 0.392 0.359 0.464 0.497
(1.468e1.610) (1.587e1.754) (0.473e0.509) (0.373e0.412) (0.338e0.382) (0.443e0.487) (0.473e0.522)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.889 1.290 0.355 0.315 0.349 0.167 0.428
(0.824e0.959) (1.199e1.388) (0.336e0.374) (0.293e0.339) (0.320e0.381) (0.154e0.181) (0.398e0.460)
0.002 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.696 1.540 0.297 0.219 0.195 0.185 0.264
(1.640e1.753) (1.483e1.599) (0.289e0.306) (0.211e0.228) (0.185e0.205) (0.178e0.192) (0.254e0.275)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.903 (1.741e2.080)
<0.001
0.433 (0.385e0.487)
<0.001
1.332 (1.211e1.465)
<0.001
1.402 (1.330e1.478)
<0.001
0.527 (0.478e0.581) 0.853 (0.769e0.946)
<0.001 0.003
0.575 (0.539e0.615) 0.776 (0.722e0.833)
<0.001 <0.001
0.669 (0.615e0.729) 1.655 (1.542e1.777)
<0.001 <0.001
0.410 (0.391e0.430) 1.010 (0.965e1.056)
<0.001 0.680
0.510 (0.393e0.662) 0.390 (0.344e0.441)
<0.001 <0.001
1.313 (1.178e1.463) 1.134 (1.074e1.198)
<0.001 <0.001
1.910 (1.687e2.162) 2.169 (2.044e2.302)
<0.001 <0.001
0.497 (0.448e0.551) 0.986 (0.947e1.027)
<0.001 0.499
0.379 (0.320e0.448)
<0.001
1.554 (1.458e1.656)
<0.001
2.405 (2.236e2.587)
<0.001
0.525 (0.494e0.559)
<0.001
0.300 0.320 1.254 0.189
<0.001 <0.001 0.001 <0.001
0.555 0.603 1.363 0.710
<0.001 <0.001 <0.001 <0.001
0.503 0.692 1.085 0.420
<0.001 <0.001 0.083 <0.001
0.504 0.378 0.657 0.380
(0.251e0.357) (0.287e0.356) (1.103e1.425) (0.155e0.232)
(0.531e0.580) (0.580e0.627) (1.280e1.453) (0.663e0.761)
(0.472e0.536) (0.657e0.728) (0.989e1.190) (0.374e0.472)
(0.488e0.520) <0.001 (0.367e0.390) <0.001 (0.621e0.695) <0.001 (0.359e0.402) <0.001 (Continued)
J Am Coll Surg
<0.001
Congestive Heart Failure and Surgical Risk
1.958 (1.800e2.129)
Turrentine et al
Congestive heart failure Age categories [<55 y] 55e59 60e 64 65e69 70e74 75 or older Male [female] Race/ethnicity [non-Hispanic White] Non-Hispanic Black Hispanic Other/unknown Surgical procedure [colectomy] Pancreatectomy Proctectomy Ventral hernia repair Bariatric surgery Thyroidectomy Appendectomy Carotid endarterectomy Abdominal aortic aneurysm repair Endovascular aneurysm repair Open aortoiliac repair Endovascular aortoiliac repair Lower extremity open Lower extremity endovascular Hysterectomy/ myomectomy Spine Brain tumor TURP
30-d Mortality OR (95% CI) p Value
1224
Table 4. Full Regression Models for Mortality and Serious Morbidity, All Procedures Combined (n ¼ 1,172,632)
Variable
30-d Readmission OR (95% CI) p Value
<0.001 <0.001 <0.001 <0.001 <0.001 0.002
0.693 0.456 0.576 0.826 1.105 0.805
(0.647e0.742) (0.437e0.477) (0.551e0.603) (0.781e0.875) (1.003e1.216) (0.749e0.864)
<0.001 <0.001 <0.001 <0.001 0.043 <0.001
3.717 (3.553e3.888)
<0.001
1.016 (0.982e1.050)
4.461 (4.117e4.833)
<0.001
1.424 0.967 1.843 0.350
(1.300e1.561) (0.884e1.057) (1.648e2.062) (0.334e0.366)
1.624 1.942 1.461 1.461 1.553 0.748
(1.548e1.703) (1.798e2.098) (1.405e1.519) (1.396e1.528) (1.480e1.629) (0.714e0.785)
p Value
Serious morbidity OR (95% CI) p Value
(0.450e0.558) (0.310e0.357) (0.578e0.653) (0.395e0.479) (1.820e2.229) (0.835e1.008)
<0.001 <0.001 <0.001 <0.001 <0.001 0.073
0.462 0.255 0.311 0.404 1.074 0.668
(0.438e0.487) (0.246e0.264) (0.300e0.322) (0.387e0.422) (1.002e1.150) (0.634e0.704)
<0.001 <0.001 <0.001 <0.001 0.044 <0.001
0.365
1.720 (1.650e1.792)
<0.001
1.945 (1.901e1.989)
<0.001
1.660 (1.625e1.696)
<0.001
1.668 (1.617e1.721)
<0.001
1.745(1.716e1.775)
<0.001
<0.001 0.460 <0.001 <0.001
0.920 0.901 0.846 0.933
(0.888e0.954) (0.872e0.931) (0.795e0.900) (0.898e0.969)
<0.001 <0.001 <0.001 <0.001
0.917 0.846 0.897 0.899
(0.872e0.965) (0.807e0.886) (0.825e0.976) (0.856e0.944)
0.001 <0.001 0.011 <0.001
0.861 0.854 0.984 0.552
(0.838e0.886) (0.833e0.876) (0.942e1.028) (0.538e0.566)
<0.001 <0.001 0.474 <0.001
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.092 1.330 1.056 1.064 1.165 0.832
(1.060e1.124) (1.251e1.415) (1.036e1.077) (1.040e1.088) (1.132e1.199) (0.816e0.849)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
1.063 1.512 1.076 1.065 1.098 0.882
(1.021e1.106) (1.405e1.628) (1.047e1.105) (1.032e1.099) (1.056e1.142) (0.858e0.907)
0.003 <0.001 <0.001 <0.001 <0.001 <0.001
1.345 1.458 1.110 1.115 1.293 0.809
(1.317e1.373) (1.392e1.528) (1.094e1.127) (1.096e1.134) (1.265e1.321) (0.797e0.821)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.501 0.333 0.614 0.435 2.014 0.918
Reference categories in brackets. ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; CVA, cerebrovascular accident; DVT, deep venous thromboembolism; INR, international normalized ratio; SGOT, serum glutamic oxaloacetic transaminase; SSI, surgical site infection; TURP, transurethral repair of the prostate.
Congestive Heart Failure and Surgical Risk
(0.331e0.473) (0.083e0.121) (0.276e0.338) (0.528e0.609) (0.245e0.464) (0.702e0.920)
0.396 0.101 0.305 0.567 0.337 0.804
Reoperation OR (95% CI)
Turrentine et al
Nephrectomy Total knee arthroplasty Total hip arthroplasty Hip fracture Flap Lung resection Emergency surgery [non-emergency] ASA Class III, IV, or V [I or II] BMI [<25 kg/m2] 25e30 >30 Unknown/unmeasured Functionally independent Dyspnea at rest or moderate exertion Renal failure or dialysis SGOT > 20 UL BUN > 18 mg/dL Creatinine > 1.2 mg/dL INR > 3
30-d Mortality OR (95% CI) p Value
Vol. 222, No. 6, June 2016
Table 4. Continued
1225
Procedure
30-d Mortality Odds ratio 95% CI
30-d Readmission Odds ratio 95% CI
Unplanned reoperation Odds ratio 95% CI
Serious morbidity Odds ratio 95% CI
3.869*
(2.490e6.013)
2.071*
(1.575e2.723)
2.225*
(1.493e3.314)
2.385*
(1.913e2.974)
9,745 59,006 94,193 102,144
3.583* 5.101* 4.249* 3.673*
(2.196e5.845) (2.028e12.828) (1.435e12.586) (1.233e10.946)
1.882* 1.216 1.611 1.931*
(1.423e2.488) (0.614e2.412) (0.867e2.994) (1.068e3.492)
1.443* 2.831* 4.368* 2.313
(1.043e1.997) (1.397 e5.737) (2.333e8.178) (0.927e5.771)
1.730* 3.090* 2.138* 1.946*
(1.304e2.295) (2.011e4.749) (1.345e3.397) (1.203e3.148)
19,836 84,098 21,386 16,804 63,160 121,443 94,151 143,016 12,904 159,148
0.389 2.503 2.102* 1.752* 2.682* 2.122* 1.564 1.788* 1.627* 1.529
(0.089e1.701) (0.747e8.385) (1.683e2.624) (1.107e2.774) (1.875e3.836) (1.058e4.255) (0.772e3.169) (1.578e2.025) (1.090e2.427) (0.987e2.370)
2.831* 2.276* 1.482* 1.375* 1.204 1.482 1.831* 1.009 1.344 1.686*
(1.887e4.246) (1.473e3.516) (1.174e1.869) (1.044e1.811) (0.877e1.653) (0.856e2.569) (1.284e2.609) (0.854e1.192) (0.907e1.993) (1.284e2.214)
2.729* 2.348* 1.042 1.324 1.066 0.870 1.219 1.133 1.355 1.485
(1.288e5.780) (1.185e4.649) (0.651e1.667) (0.992e1.769) (0.639 e1.777) (0.312e2.426) (0.704e2.112) (0.946e1.358) (0.934e1.966) (0.962e2.292)
2.471* 1.877* 1.525* 1.336* 1.308* 1.575* 1.648* 1.267* 1.778* 1.411*
(1.694e3.605) (1.243e2.833) (1.272e1.829) (1.083e1.648) (1.014e1.687) (1.069e2.319) (1.210e2.244) (1.151e1.395) (1.379e2.293) (1.126e1.768)
20,136 13,688 18,523
1.433 5.059* 1.976
(0.881e2.330) (2.657e9.632) (0.710e5.503)
1.615* 0.994 1.287
(1.110e2.350) (0.455 e2.171) (0.679e2.442)
1.133 0.950 1.191
(0.669e1.920) (0.344e2.628) (0.423e3.356)
1.554* 1.517 1.787*
(1.171e2.061) (0.930e2.476) (1.139e2.806)
7,450
1.235
(0.691e2.208)
1.134
(0.454e2.835)
1.596
(0.855e2.980)
1.639*
(1.087e2.47)
Congestive Heart Failure and Surgical Risk
47,336
Turrentine et al
Carotid endarterectomy Lower extremity endovascular Thyroidectomy Bariatric surgery Hysterectomy/myomectomy Transurethral repair of prostate Total knee arthroplasty Hip fracture repair Lower extremity open Total hip arthroplasty Appendectomy Spine repair Colectomy Open aortoiliac repair Ventral hernia repair Endovascular aneurysm repair Lung resection Nephrectomy Aortic abdominal aneurysm repair
n
1226
Table 5. Adjusted Odds Ratios and 95% Confidence Intervals of Patients With Preoperative Congestive Heart Failure Experiencing Adverse Outcomes Compared With Those Without, Selected Procedures
Adjusted for patient age, sex, race/ethnicity, emergency surgery status, American Society of Anesthesiologists Class, BMI, preoperative risk factors (functional dependence, dyspnea, renal failure, or dialysis), preoperative lab values (serum glutamic oxaloacetic transaminase, blood urea nitrogen, creatinine, international normalized ratio). All procedures that had at least 1 outcome with p < 0.05 were included in this table. *Odds ratios with p values < 0.05.
J Am Coll Surg
Vol. 222, No. 6, June 2016
Turrentine et al
The risks of all 4 adverse outcomes for carotid endarterectomy and lower extremity endovascular surgery were increased by CHF. Seven procedures were adversely affected by CHF in 3 of 4 outcomes, including bariatric and thyroidectomy surgery which more than doubled the risk of mortality, reoperation, and serious morbidity; and TURP and total knee arthroplasty, which significantly increased the risk of readmission, reoperation, and serious morbidity. Sensitivity analyses revealed the addition of operative duration and preoperative cardiac variables did not significantly change the association between CHF and adverse outcomes (not shown). However, the magnitude of association was always stronger with inclusion of operative duration, and it remained similar with or without preoperative cardiac variables in the models. Additional inclusion of severity indicators, such as operative duration or outpatient surgery status, did not change the association between the two.
DISCUSSION Congestive heart failure significantly increased the risk of serious morbidity, readmission, reoperation, and mortality in our data. The pervasive effect of CHF on surgical outcomes after noncardiac operations may not be fully appreciated by surgeons. Most procedures we examined had at least 1 of 23 postoperative complications significantly affected by CHF. Awareness of CHF’s impact on surgical outcomes is important to both surgeons and patients. For surgeons, it underscores the importance of optimal perioperative care, particularly fluid management, in this vulnerable population. Pulse oximeter plethysmography accurately measures volume and can be helpful in monitoring volume responsiveness and the need for fluid.13 Pleth Variability Index, an algorithm allowing continuous monitoring of variations in pulse oximetry,14 promotes goal-directed fluid management in Early Recovery After Surgery (ERAS) protocols15 and could mitigate risk of fluid overload in the CHF patient undergoing surgery. Patients, particularly those undergoing elective procedures, could potentially modify risks by avoiding conditions that contribute to or exacerbate CHF: smoking, obesity, diets high in saturated fats, sugar, and sodium. In addition, patients with heart disease or other chronic conditions can be encouraged to follow their treatment plans, stay active, limit alcohol, and get enough rest.16 Patients with CHF, compared with those without the condition, are 96% more likely to die in the 30 days after surgery. Patients undergoing thyroidectomy had the highest odds of mortality. Although at first this finding is
Congestive Heart Failure and Surgical Risk
1227
surprising, the fact that patients are undergoing surgery with acute heart failure suggests an urgent nature to the procedure. Further explanation may be that “Thyroid hormone influences every cell, tissue, and organ in the body, and its homeostasis is essential to the optimal functioning of the heart,”17 and thyroid hormone disarray is a risk factor for heart failure.18 Similarly, CHF was significantly associated with postoperative adverse events in thyroidectomy and parathyroidectomy patients in 2007/2008 ACS NSQIP PUF data.19 Patients with CHF undergoing bariatric procedures had a 4-fold increase in unplanned reoperation. Patients requiring dialysis were significantly more likely to require reoperation.20 This may explain the higher rate of reoperation in our cohort; our bariatric patients with CHF had significantly higher rates of renal failure compared with patients without CHF. This differs from work that did not identify CHF as a risk for return to the operating room in a subset of ACS NSQIP 2007 to 2009 bariatric PUF procedures.21 However, only laparoscopic gastric bypass and adjustable gastric band were analyzed; we also included data for open gastric bypass and sleeve gastrectomy. Patients with CHF undergoing more complex procedures may be at higher risk for complications. Patients with CHF who underwent lower extremity endovascular revascularization were more likely to experience all 4 adverse outcomes. This is unexpected because lower extremity bypass is recommended for patients who are not candidates for angioplasty or have a failed angioplasty, and one would think outcomes for open repair would be more adversely affected compared with endovascular repair. However, this finding may be explained in part by findings that CHF independently predicted patency loss after endovascular treatment.22 The subject of patient selection, evaluation, and indication for treatment of lower extremity endovascular surgery bears further investigation. Congestive heart failure was associated with serious morbidity collectively and the majority of postoperative complications individually. In our study population, CHF had the biggest increases on the risks of cardiac arrest, unplanned intubation, and mechanical ventilation > 48 hours. Excessive fluid volume administration and decreased cardiac output during the perioperative period may explain these complications.23 Similarly, CHF has been associated with severe perioperative cardiac complications, including cardiac arrest, in noncardiac surgical procedures in a Veterans Affairs hospital.24 A multifactorial risk index for predicting postoperative respiratory failure in men found CHF to be the only significant cardiac status predictor of respiratory failure in a study of
1228
Turrentine et al
Congestive Heart Failure and Surgical Risk
81,719 patients who underwent noncardiac procedures within the Department of Veterans Affairs NSQIP.25 Age and race/ethnicity were the most prominent characteristics distinguishing patients with CHF from those without the comorbid condition. This is not surprising because age is a risk factor for surgery26 and is a component of risk stratification in a number of indices27-29 and guidelines.30 Advanced age independently predicted morbidity31 and mortality32 in patients with CHF.33 African Americans with CHF, compared with whites with CHF, have higher rates of readmission, reoperation, and serious morbidity, but lower mortality rates. Nonwhites, compared with whites, have lower risk of mortality, but differed from our study in that nonwhites were less likely to experience readmission.34 Perhaps reduced mortality among African Americans, in patients hospitalized with CHF, is due to differences in disease severity or duration.35 In addition, differences in quality of care, type of hospital, cardiac testing, and concurrent ischemic heart disease between African Americans and Caucasians may provide an explanation.36 Limitations In our study, the diagnosis of CHF depended primarily on history and physical findings recorded in the medical record. Our ACS NSQIP definition limited us to events of acute CHF and therefore prevented evaluation of the influence of chronic CHF on surgical outcomes. In addition, the degree of CHF could not be ascertained, nor could preoperative management of the condition. Another limitation is that we used ACS NSQIP procedure-targeted groups, which precluded assignment of minor or major classifications to individual procedures. Procedure-targeted groups do not include data on some operations such as small bowel resections and cholecystectomy, which could have been affected by CHF. This large dataset contained few patients with CHF who underwent certain operations such as gynecologic reconstruction and carotid artery stenting.
CONCLUSIONS Although the prevalence of acute CHF is low among surgical patients, it significantly increases serious morbidity, readmission, reoperation, mortality, and most other postoperative complications. Surgeons should manage CHF carefully when considering surgery. Author Contributions Study conception and design: Turrentine, Jones Acquisition of data: Turrentine, Sohn, Jones
J Am Coll Surg
Analysis and interpretation of data: Turrentine, Sohn, Jones Drafting of manuscript: Turrentine, Sohn Critical revision: Jones REFERENCES 1. Grasso AW, Jaber WA. Cardiac risk stratification for noncardiac surgery. Available at: http://www.clevelandclinicmeded. com/medicalpubs/diseasemanagement/cardiology/cardiac-riskstratification-for-noncardiac-surgery/Default.htm. Accessed February 11, 2015. 2. Sheer AJ, Heckman JE, Schneider EB, et al. Congestive heart failure and chronic obstructive pulmonary disease predict poor surgical outcomes in older adults undergoing elective diverticulitis surgery. Dis Colon Rectum 2011;54:1430e1437. 3. Hernandez AF, Whellan DJ, Stroud S, et al. Outcomes in heart failure patients after major noncardiac surgery. J Am Coll Cardiol 2004;44:1446e1453. 4. Rao A, Polanco A, Qui S, et al. Safety of outpatient laparoscopic cholecystectomy in the elderly: Analysis of 15,248 patients using the NSQIP database. J Am Coll Surg 2013; 217:1038e1043. 5. Annamalai A, Kakarla VR, Nandipati K. Predictors of mortality following pancreaticoduodenectomy for periampullary cancer. OA Surg 2014;2:2e7. 6. Maile MD, Engoren MC, Tremper KK, et al. Worsening preoperative heart failure is associated with mortality and noncardiac complications, but not myocardial infarction after noncardiac surgery; A retrospective cohort study. Soc Cardiovasc Anesthesiol 2014;119:522e532. 7. Parante D, Luis C, Veiga D, et al. Congestive heart failure as a determinant of postoperative delirium. Portuguese J Cardiol 2013;32:665e671. 8. Turrentine FE, Denlinger CE, Simpson VB, et al. Morbidity, mortality, cost, and survival estimates of gastrointestinal anastomotic leaks. J Am Coll Surg 2015;220:195e206. 9. American College of Surgeons. ACS NSQIP Participant Use Data File. Available at: http://site.acsnsqip.org/participantuse-data-file/. Accessed February 11, 2015. 10. ACS NSQIP. ACS NSQIP variables and definitions. In: ACS NSQIP operations manual. Chicago: American College of Surgeons; 2014:47. 11. ACS NSQIP. ACS NSQIP: Procedure targeted algorithm worksheet. Available at: http://reports.nsqip.facs.org/acsmain/private/ documents/operationsmanual/index.jsp. Accessed March 20, 2015. 12. ACS NSQIP. Definitions. In: ACS NSQIP semiannual report: July1, 2013 through June 30, 2015. Chicago, Illinois: ACS NSQIP; 2015:32. 13. Marik P, Monnet X, Teboul J. Hemodynamic parameters to guide fluid therapy. Ann Intensive Care 2011;1:1e9. 14. Cannesson M, Desebbe O, Rosamel P, et al. Pleth variability index to monitor the respiratory variations in the pulse oximeter plethysmographic waveform amplitude and predict fluid responsiveness in the operating theatre. Br J Anaesthesia 2008;101:200e206. 15. Thiele R, Rea K, Turrentine F, et al. Standardization of care: Impact of an enhanced recovery protocol on length of stay, complications, and direct costs after colorectal surgery. J Am Coll Surg 2015;220:430e443.
Vol. 222, No. 6, June 2016
Turrentine et al
16. Eckel R, Jakicic J, Ard J, et al. 2013 AHA/ACC guidelines on lifestyle management to reduce cardiovascular risk: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2960e2984. 17. Mitchell JE, Hellkamp AS, Mark DB, et al. Thyroid function in heart failure and impact on mortality. J Am Coll Cardiol 2013;1:48e55. 18. Schmidt-Ott UM, Ascheim DD. Thyroid hormone and heart failure. Current Heart Failure Reports 2006;3:114e119. 19. Gupta PK, Smith RB, Gupta H, et al. Outcomes after thyroidectomy and parathyroidectomy. Head Neck 2012;34:477e484. 20. Gajdos C, Hawn M, Kile D, et al. Risk of major nonemergent inpatient general surgical procedures in patients on long-term dialysis. JAMA Surg 2013;148:137e143. 21. Nandipati K, Lin E, Husain F, et al. Factors predicting the increased risk for return to the operating room in bariatric patients: A NSQIP database study. Surg Endosc 2013;27: 1172e1177. 22. Meltzer AJ, Shrikhande G, Gallagher KA, et al. Heart failure is associated with reduced patency after endovascular intervention for symptomatic peripheral arterial disease. J Vasc Surg 2012;55:353e362. 23. Pradeep A, Rajagopalam S, Kolli HK, et al. High volumes of intravenous fluid during cardiac surgery are associated with increased mortality. HSR Proc Intensive Care Cardiovasc Anesthesia 2010;2:287e296. 24. Kumar R, McKinney WP, Raj G, et al. Adverse cardiac events after surgery: Assessing risk in a veteran population. J Gen Intern Med 2001;16:507e518. 25. Arozullah AM, Daley J, Henderson WG, Khuri SF. Multifactorial risk index for predicting postoperative respiratory failure in men after major noncardiac surgery. Ann Surg 2000;2:242e253. 26. Turrentine FE, Wang H, Simpson VB, Jones RS. Surgical risk factors, morbidity, and mortality in elderly patients. J Am Coll Surg 2006;203:865e877. 27. Goldman L, Caldera DL, Nussbauam SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845e850.
Congestive Heart Failure and Surgical Risk
1229
28. Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery. Report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. Committee on Perioperative Cardiovascular Evaluation for Noncardiac Surgery. Circulation 1996;93: 1278e1317. 29. Detsky AS, Abrams HB, Frobath N, et al. Cardiac assessment for patients undergoing noncardiac surgery. A multifactorial clinical risk index. Arch Intern Med 1986;146: 2131e2134. 30. The Expert Work Group Members. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;129: S49eS73. 31. Polanczyk CA, Marcantonio E, Goldman L, et al. Impact of age on perioperative complications and length of stay in patients undergoing noncardiac surgery. Ann Intern Med 2001;134:637e643. 32. Feringa HH, Bax JJ, Karagiannis SE, et al. Elderly patients undergoing major vascular surgery: Risk factors and medication associated with risk reduction. Arch Gerontol Geriatrics 2009;48:116e120. 33. Healy KO, Waksmonski CA, Altman RK, et al. Perioperative outcome and long term mortality for heart failure patients undergoing intermediate and high risk noncardiac surgery: Impact of left ventricular ejection fraction. Congestive Heart Failure 2010;16:45e49. 34. Hammill BG, Curtis LH, Bennett-Guerrero E, et al. Impact of heart failure on patients undergoing major noncardiac surgery. Anesthesiology 2008;108:559e567. 35. Joshi AV, D’Souza AO, Madhavan S. Differences in hospital length-of-stay, charges, and mortality in congestive heart failure patients. Congestive Heart Failure 2004;10: 76e84. 36. Philbin EF, DiSalvo TG. Influence of race and gender on care processes, resource use, and hospital-based outcomes in congestive heart failure. Am J Cardiol 1998;82:76e81.