- Email: [email protected]
Effect of Body Mass Index on Outcomes After Surgery for Perforated Diverticulitis
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Effect of Body Mass Index on Outcomes After Surgery for Perforated Diverticulitis Kathryn T. Weber, MD,* Paul J. Chung, MD, Nicholas La Gamma, MD, John A. Procaccino, MD, Antonio E. Alfonso, MD, Gene Coppa, MD, and Gainosuke Sugiyama, MD Department of Surgery, Zucker School of Medicine at Hosftra Northwell, New Hyde Park, New York
article info
abstract
Article history:
Background: Despite the increased adoption of minimally invasive techniques in colorectal
Received 8 July 2019
surgery, an open resection with ostomy creation remains an accepted operation for
Received in revised form
perforated diverticulitis. In the United States, there is an increase in the rates of both
15 September 2019
morbid obesity and diverticular disease. Therefore, we wanted to explore whether out-
Accepted 1 October 2019
comes for morbidly obese patients with diverticulitis are worse than nonmorbidly obese
Available online xxx
patients after open colectomy for diverticulitis. Materials and methods: Using the American College of Surgeons National Surgical Quality
Keywords:
Improvement Program database from 2005 to 2015, we identified adults with emergent
Perforated diverticulitis
admission for diverticulitis (International Classification of Diseases, Ninth Revision, code
Morbid obesity
562.11) with evidence of preoperative sepsis and intraoperative contaminated/dirty wound
NSQIP
classification, in which a resection with ostomy (Current Procedural Terminology codes 44141, 44143, or 44144) was performed. We excluded cases with age >90 y, ventilator dependence, evidence of disseminated cancer and missing sex, race, body mass index, functional status, American Society of Anesthesiologists class, length of stay (LOS), or operative time data. Morbid obesity was defined as body mass index >35 kg/m2. Risk variables of interest included age, sex, race, medical comorbidities, requirement for preoperative transfusion, preoperative sepsis, and operative time. Outcomes of interest included LOS, 30-d postoperative complications, and mortality. Univariate and propensity scores with postmatching analyses were performed. Results: A total of 2019 patients met inclusion and exclusion criteria, of which 413 (20.5%) were morbidly obese. Morbidly obese patients tended to be younger (mean 57.2 versus 62.6 y) and female (54.5% versus 45.5%). Morbidly obese patients also had higher rates of insulin-dependent diabetes (8.0% versus 4.2%), hypertension (60.1% versus 51.3%), renal failure (3.4% versus 1.5%), and higher American Society of Anesthesiologists class (class 4: 23.5% versus 19.6% and class 5: 1.45% versus 0.87%). Morbidly obese patient had no increase in 30-d mortality or LOS, but they had higher rates of superficial wound infection (9.0% versus 5.8%; P ¼ 0.0259), deep wound infection (4.4% versus 1.9%; P ¼ 0.0073), acute renal failure (4.8% versus 2.4%; P ¼ 0.0189), postoperative septic shock (17.7% versus 12.1%; P ¼ 0.0040), and return to the operating room (11.1% versus 6.4%; P ¼ 0.0015). We identified 397 morbidly obese patients well matched by propensity score to 397 nonmorbidly obese patients. Conditional logistic regression showed no difference in LOS (median 12.9 versus 12.4 d; P ¼ 0.4648) and no increased risk of 30-d mortality (P ¼ 0.947), but morbid obesity
* Corresponding author. Department of Surgery, 300 Community Drive, Manhasset, NY, 11030. Tel.: 516-562-0100; fax: 516-562-1521. E-mail address: [email protected] (K.T. Weber). 0022-4804/$ e see front matter ª 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jss.2019.10.020
2
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
was an independent predictor for return to the operating room (adjusted odds ratio: 27.09 [95% confidence interval: 2.68-274.20]; P ¼ 0.005). Conclusions: This analysis of a large national clinical database demonstrates that morbidly obese patients presenting with perforated diverticulitis undergoing a Hartmann’s procedure do not have increased mortality or LOS compared with nonobese patients. After adjusting for the effects of morbid obesity, morbidly obese patients had increased risk of return to operating room. Despite literature describing the many perioperative risks of obesity, our analysis showed only increased reoperation for obese patients with diverticulitis. ª 2019 Elsevier Inc. All rights reserved.
Introduction Diverticulitis is commonly encountered in the United States, and its incidence is increasing both nationally and globally.1,2 Hospitalization rates for diverticulitis have increased,2-4 with an elevated incidence of diverticulitis among younger patients and higher rates of elective surgery in all age groups.5 The etiology of diverticulitis is thought to be multifactorial, with proposed risk factors including diet, poor fiber intake, inflammatory conditions, genetics, race, geography, seasonality, and age.6,7 Diverticulitis has previously been associated with dietary habits, geographic variation, and weight change.8 The standard of care for the treatment of both uncomplicated and perforated diverticulitis encompasses a broad spectrum because of the variety of disease severity at the time of presentation, and distinct clinical patterns of disease severity have emerged.9 Among patients who are acutely ill with peritonitis, the Hartmann’s procedure has endured as a widely accepted treatment.5,10 This extensive and comparatively morbid operation has persisted as the operation of choice, resisting contemporary trends toward preferential utilization of minimally invasive techniques, because it is fast, simple, and effective.11 However, the morbidity of the Hartmann’s procedure remains significant, specifically with the creation of an ostomy, which is intended to be temporary, but often remains permanent as other clinical considerations prevent the reestablishment of bowel continuity.12 Despite the theoretical advantages a minimally invasive approach may offer, particularly among obese patients, in regards to wound infection, length of stay (LOS), and maintenance of bowel continuity, current practice guidelines continue to support management of complicated diverticulitis with an operative resection with sigmoidectomy and end ostomy creation (Hartmann’s procedure).13 Laparoscopic washout has declined in favor, with evidence suggesting that resection is preferable, even in acute diverticulitis.14 Morbid obesity has implications in the management of both diverticulitis as a possible risk factor for poor outcomes following surgical management of diverticulitis. Obesity has been linked to poorer operative outcomes for both elective and emergent surgery, specifically with increased rates of wound infection and longer operative times.15-20 Many of these complications are mitigated with minimally invasive approaches, particularly among obese patients, as outcomes after open surgery tend to be worse for obese patients; current literature demonstrates longer hospital stays and increased
incidence of wound infection in obese patients undergoing open surgery compared with laparoscopic approaches.16,20 Therefore, the purpose of this study was to investigate if morbid obesity is associated with worse outcomes after Hartmann’s procedure for diverticulitis. With rising rates of both morbid obesity and diverticular disease in the United States, optimizing the standard of care for these complicated cases becomes increasingly important. We hypothesized that morbidly obese patients would have worse perioperative outcomes following open resection and end ostomy creation for perforated diverticulitis.
Methods Patient selection and data Data were obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database from 2005 to 2015. We identified cases in which an adult (aged 18 y) underwent emergent admission for diverticulitis (International Classification of Diseases, Ninth Revision, code 562.11) with evidence of preoperative sepsis and intraoperative contaminated/dirty wound classification, in which a resection with ostomy (Current Procedural Terminology codes 44141, 44143, or 44144) was performed. The combination of codes allowed us to restrict our analysis to patients with acute perforated diverticulitis who underwent a Hartmann’s. Morbid obesity was defined as body mass index (BMI) >35 kg/m2 for the purposes of this study, as most of the patients had obesity-related comorbidities such as hypertension or diabetes and therefore met the definition at a lower BMI. We excluded patients with age >90 y, cases performed by nongeneral surgeons (as characterized by NSQIP), patients who had ventilator dependence, evidence of disseminated cancer, or missing sex, race, BMI, functional status, American Society of Anesthesiologists (ASA) class, or LOS.
Risk variables and outcomes Risk variables of interest included age, sex, race, presence of diabetes, smoking status, presence of dyspnea, hypertension, functional status, history of chronic obstructive pulmonary disease (COPD), history of congestive heart failure (CHF), acute renal failure, end-stage renal disease (ESRD), steroid use, bleeding disorders, morbid obesity, preoperative transfusion of 1 unit of packed red blood cells within 72 h of surgery,
3
weber et al effect of bmi on surgery for diverticulitis
Table 1 e Patient characteristics of obese and control cohorts before and after propensity score matching. Characteristic
Overall cohorts
Matched cohorts
Nonobese (n ¼ 1606)
Morbidly obese (n ¼ 413)
Absolute standardized difference
Nonobese (n ¼ 397)
Morbidly obese (n ¼ 397)
Absolute standardized difference
62.6 (13.6)
57.2 (14.1)
0.3842
58.3 (14.36)
57.72 (13.93)
0.0415
Female
806 (50.2)
225 (54.5)
0.0429
219 (55.2)
214 (53.9)
0.0126
Male
800 (49.8)
188 (45.5)
0.0429
178 (44.8)
183 (46.1)
0.0126
Age, y, mean (SD) Sex, n (%)
BMI Race, n (%) 13 (0.8)
2 (0.5)
0.0033
1 (0.3)
2 (0.5)
0.0025
African American, n (%)
Asian, n (%)
109 (6.8)
49 (11.9)
0.0508
45 (11.3)
45 (11.3)
0.0050
Native Hawaiian/Pacific Islander, n (%)
1 (0.1)
0 (0.0)
0.0006
0 (0.0)
0 (0.0)
0.0000
American Indian or Alaskan, n (%)
17 (1.1)
3 (0.7)
0.0033
2 (0.5)
3 (0.8)
0.0025
White, n (%)
1466 (91.3)
359 (86.9)
0.0436
349 (87.9)
347 (87.4)
0.0050
Smoker, n (%)
395 (24.6)
90 (21.8)
0.0280
85 (21.4)
87 (21.9)
0.0050
0.0151
Dyspnea, n (%) None
1464 (91.2)
365 (88.4)
0.0278
359 (90.4)
353 (88.9)
Moderate exertion
97 (6.0)
34 (8.2)
0.0219
27 (6.8)
32 (8.1)
0.0126
At rest
45 (2.8)
14 (3.4)
0.0059
11 (2.8)
12 (3.0)
0.0025
1434 (89.3)
312 (75.5)
0.01375
309 (77.1)
306 (77.1)
0.0076
104 (6.5)
68 (16.5)
0.0999
59 (14.9)
61 (15.4)
0.0050
68 (4.2)
33 (8.0)
0.0376
29 (7.3)
30 (7.6)
0.0025
103 (6.4)
22 (5.3)
0.0109
25 (6.3)
22 (5.5)
0.0076
16 (1.0)
4 (1.0)
0.0003
2 (0.5)
3 (0.8)
0.0025
Diabetes, n (%) None Noninsulin dependent Insulin dependent Functional status, n (%) Partially dependent Totally dependent Independent Hypertension, n (%) History of CHF, n (%) History of COPD, n (%) Renal failure, n (%) ESRD, n (%)
1487 (92.6)
387 (93.7)
0.0111
370 (93.2)
372 (93.7)
0.0050
824 (51.3)
248 (60.1)
0.0874
241 (60.7)
235 (59.2)
0.0151 0.0025
46 (2.9)
13 (3.2)
0.0028
13 (3.27)
12 (3.0)
160 (10.0)
43 (10.4)
0.0045
43 (10.4)
43 (10.4)
0.0000
24 (1.5)
14 (3.4)
0.0190
11 (2.8)
10 (2.5)
0.0025
32 (2.0)
4 (1.0)
0.0102
4 (1.0)
4 (1.0)
0.0000
Steroid use, n (%)
236 (14.7)
52 (12.6)
0.0210
45 (11.3)
50 (12.6)
0.0126
Bleeding disorder, n (%)
171 (11.7)
50 (12.1)
0.0146
41 (10.3)
47 (11.8)
0.0151
16 (1.0)
3 (0.7)
0.0027
4 (1.0)
3 (0.8)
0.0025
Preoperative transfusion, n (%) Preoperative sepsis, n (%) SIRS Sepsis Septic shock
204 (12.7)
50 (12.1)
0.0060
45 (11.3)
50 (12.6)
0.0126
1286 (80.1)
326 (78.9)
0.0114
315 (79.4)
313 (78.8)
0.0050
116 (7.2)
37 (9.0)
0.0174
37 (9.3)
34 (8.6)
0.0076
ASA class, n (%) Class I
30 (1.9)
3 (0.7)
0.0114
6 (1.5)
3 (0.8)
0.0076
Class II
497 (31.0)
83 (20.1)
0.1085
84 (21.2)
82 (20.7)
0.0050
Class III
751 (46.8)
224 (54.2)
0.0748
219 (55.2)
220 (55.4)
0.0025
Class IV
315 (19.6)
97 (23.5)
0.0394
84 (21.2)
89 (22.4)
0.0126
Class V
14 (0.9)
6 (1.5)
0.0053
4 (1.0)
3 (0.76)
0.0025
Contaminated
129 (8.0)
31 (7.5)
0.0053
33 (8.3)
31 (7.8)
0.0050
Dirty/infected
1477 (92.0)
382 (92.5)
0.0053
364 (91.7)
366 (92.2)
0.0050
Wound class, n (%)
SD ¼ standard deviation; SIRS ¼ systemic inflammatory response syndrome.
4
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
Table 2 e Univariate analysis of unmatched morbidly obese and control cohorts. Outcome
Overall cohorts Non obese (n ¼ 1606)
Matched cohorts
Morbidly obese (n ¼ 413)
P value
Non obese (n ¼ 397)
Morbidly obese (n ¼ 397)
12.42 (8.36)
12.88 (9.18)
LOS, d, median, (SD)
10 (8.38)
10 (9.17)
0.104
Superficial wound infection, n (%)
93 (5.8)
37 (8.9)
0.026
Deep wound infection, n (%)
31 (1.9)
18 (4.4)
Organ space infection, n (%)
146 (9.1)
46 (11)
Dehiscence, n (%)
48 (2.9)
Pneumonia, n (%)
P value 0.465
21 (5.3)
33 (8.3)
0.121
0.007
8 (2.0)
18 (4.5)
0.071
0.242
38 (9.6)
43 (10.8)
0.639
13 (3.1)
0.994
10 (2.5)
13 (3.3)
0.673
121 (7.5)
36 (8.7)
0.486
31 (7.8)
36 (9.1)
0.610
Unplanned intubation, n (%)
109 (6.7)
21 (5.1)
0.252
35 (8.8)
21 (5.3)
0.072
Failure to wean ventilator, n (%)
176 (11)
59 (14)
0.073
49 (12.3)
54 (13.6)
0.673
Pulmonary embolism, n (%)
19 (12)
6 (1.4)
0.621
4 (1.0)
6 (1.5)
0.752
Urinary tract infection, n (%)
35 (2.2)
9 (2.3)
1
10 (2.5)
8 (2.0)
0.812
Progressive renal insufficiency, n (%)
15 (<1)
5 (1.2)
0.581
4 (1.0)
3 (0.8)
1
Acute renal failure, n (%)
40 (2.49)
20 (4.84)
0.019
14 (3.5)
19 (4.8)
0.477
Urinary tract infection, n (%)
35 (2.2)
9 (2.3)
1
10 (2.5)
8 (2.0)
0.812
VTE/thrombophlebitis, n (%)
59 (3.6)
15 (3.6)
1
12 (3.0)
15 (3.8)
0.695
Myocardial infarct, n (%)
23 (1.4)
3 (<1)
0.333
7 (1.8)
3 (0.8)
0.341
Cardiac arrest, n (%)
20 (1.2)
6 (1.4)
0.806
6 (1.5)
5 (1.3)
1
Postoperative transfusion, n (%)
228 (14)
52 (13)
0.446
66 (16.6)
47 (11.8)
0.068
Postoperative sepsis, n (%)
306 (19)
90 (22)
0.23
74 (18.6)
89 (22.4)
0.219
Postoperative septic shock, n (%)
195 (12)
73 (18)
0.004
61 (15.4)
66 (16.6)
0.699
Return to operating room, n (%)
103 (6.4)
46 (11)
0.001
21 (5.3)
42 (10.6)
0.009
83 (5.2)
21 (5.1)
1
21 (5.3)
20 (5.0)
1
Death, n (%)
Bold values are reached statistical significance where P < 0.05. SD ¼ standard deviation; VTE ¼ venous thromboembolism.
systemic sepsis within 48 h of surgery, operative wound classification, and ASA classification. Primary outcomes of interest were 30-d mortality and total LOS. Secondary outcomes of interest included superficial, deep, and organ-space wound infections, wound dehiscence, pneumonia, failure to extubate, unplanned intubation, pulmonary embolism, venous thromboembolism, urinary tract infection, progressive renal insufficiency, acute renal failure, myocardial infarction, cardiac arrest, bleeding requiring transfusion, sepsis, septic shock, and return to the operating room.
transfusions, preoperative sepsis, ASA classification, and wound class. Quality of match was assessed using the absolute standardized mean difference with a goal of 0.2. For categorical outcomes, we conducted conditional logistic regression adjusting for all the matched variables, and for continuous variables, we used the Wilcoxon rank-sum test. We then performed sensitivity analysis by conducting propensity score matching using the same parameters with the exception of using a caliper of 0.1 and then performed postmatch analysis. The R programming language version 3.4 was used for statistical analysis. Institutional Review Board exemption was obtained for this study as it used deidentified data from a publicly available source.
Statistical analysis Univariate analysis and propensity score matching with postmatching analysis were performed. We used propensity score matching using the so-called nearest neighbor method with a caliper of 0.25 to create a 1:1 match of morbidly obese to control cases over patient characteristics including age, sex, race, diabetes, smoking status, dyspnea, hypertension, functional status, history of COPD, history of CHF, renal failure, ESRD, steroid use, bleeding disorder, preoperative blood
Results Patient characteristics A total of 2019 patients met inclusion and exclusion criteria, of which 413 (20.5%) were morbidly obese. Compared with control, morbidly obese patients tended to be younger (mean 57.2
5
weber et al effect of bmi on surgery for diverticulitis
versus 62.6 y) and had a higher proportion of female compared with male patients (54.5% versus 45.5%). Morbidly obese patients had a median BMI of 39.3, compared with 27.4 in the nonobese group. Morbidly obese patients also had higher rates of insulin-dependent diabetes (8.0% versus 4.2%), hypertension (60.1% versus 51.3%), renal failure (3.4% versus 1.5%), and higher ASA class (class 4: 23.5% versus 19.6% and class 5: 1.45% versus 0.87%; Table 1).
Univariate analysis Unadjusted comparison of morbidly obese patients to control demonstrated higher rates superficial wound infection (9.0% versus 5.8%; P ¼ 0.0259), deep wound infection (4.4% versus 1.9%; P ¼ 0.0073), acute renal failure (4.8% versus 2.4%; P ¼ 0.0189), postoperative septic shock (17.7% versus 12.1%; P ¼ 0.0040), and return to the operating room (11.1% versus 6.4%; P ¼ 0.0015). However, there were no differences found in LOS between the morbidly obese group and control (median 10 versus 10 d; P ¼ 0.1038; Table 2).
of variables was nonsignificant after propensity matching the patients. Conditional logistic regression showed that morbid obesity was an independent predictor for return to the operating room (adjusted odds ratio [OR] 27.09 [95% confidence interval (CI) 2.68-274.20]; P ¼ 0.005; Table 3).
Sensitivity analysis Sensitivity analysis was performed by setting the caliper to 0.1, which resulted in 398 morbidly obese patients being matched to 398 controls. The postmatch maximum absolute standardized difference between covariates was 0.0231, signifying good match. There was no significant difference in the LOS between the matched morbidly obese group and control (median 10 versus 10 d; P ¼ 0.1969). Conditional logistic regression again showed that morbid obesity was still an independent predictor for return to the operating room (adjusted OR 2.72 [95% CI 1.06-6.97; P ¼ 0.038; Table 4).
Propensity score match and analysis
Discussion
Propensity matched score with a caliper of 0.25 resulted in 397 morbidly obese patients being matched to 397 controls. Postmatch, the maximum absolute standardized difference was 0.0415, which confirmed good match. There was no difference in the LOS between the matched morbidly obese cohort and control (median 12.9 versus 12.4 d; P ¼ 0.4648). The remainder
This analysis of a large national clinical database demonstrated that morbidly obese patients compared with nonobese patients do not have increased 30-d mortality or longer LOS after an emergent Hartmann’s procedure for perforated diverticulitis. However, morbidly obese patients did have
Table 3 e Outcomes of matched morbidly obese and control cohorts. Outcome
Table 4 e Outcomes of matched morbidly obese and control cohorts and sensitivity analysis. Outcome
Adjusted OR
95% CI
P value
Adjusted OR
95% CI
P value
0.87-6.40
0.090
0.8479.73
0.071
2.37
Superficial wound infection
8.16
Superficial wound infection
0.83-1.12
0.621
0.7001.83
0.740
0.96
Deep space wound infection
0.96
Deep space wound infection
0.56-2.03
0.842
1.42
0.64-3.18
0.390
Organ space wound infection
1.07
Organ space wound infection
Pneumonia
2.93
1.05
0.43-2.57
0.907
0.7910.86
0.108
Pneumonia Unplanned intubation
0.97
0.29-3.26
0.962
Unplanned intubation
0.91
0.08-1.03
0.053
Failure to wean ventilator
1.78
0.88-3.61
0.110
Failure to wean ventilator
1.16
0.63-2.13
0.629
Acute renal failure
0.98
0.84-1.14
0.823
Acute renal failure
1.00
0.86-1.16
0.961
VTE/thrombophlebitis
0.99
0.86-1.15
0.923
VTE/thrombophlebitis
1.00
0.87-1.17
0.954
Myocardial infarct
1.01
0.87-1.17
0.889
Myocardial infarct
1.01
0.87-1.17
0.938
Cardiac arrest
1.00
0.86-1.15
0.971
Cardiac arrest
1.01
0.86-1.17
0.931
Postoperative transfusion
0.55
0.27-1.10
0.091
Postoperative transfusion
0.65
0.38-1.11
0.115
Postoperative sepsis
1.02
0.68-1.52
0.932
Postoperative sepsis
1.11
0.77-1.61
0.580
Postoperative septic shock
1.44
0.81-2.57
0.215
Postoperative septic shock
1.40
0.86-2.29
0.177
Return to operating room
27.09
2.68274.2
0.005
Return to operating room
2.72
1.06-6.97
0.038
Dehiscence, pulmonary embolism, progressive renal insufficiency, and death were nonsignificant. Urinary tract infection did not converge. Bold values are reached statistical significance where P < 0.05. VTE ¼ venous thromboembolism.
Pulmonary embolism, urinary tract infection, and death were nonsignificant. Dehiscence and progressive renal insufficiency did not converge. Bold values are reached statistical significance where P < 0.05. VTE ¼ venous thromboembolism.
6
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
increased morbidity as characterized by increased rates of wound infection, postoperative septic shock, acute renal failure, and return to the operating room. Despite these worse perioperative outcomes, after performing sensitivity analysis, morbidly obese patients were consistently at risk only for return to the operating room compared with matched nonobese control. Obesity is implicated in perioperative complications in elective colorectal surgeries, with increased rates of wound infection and prolonged operative times.21 Although emergent operations have increased morbidity compared with elective surgeries, obese patients are particularly vulnerable to worse outcomes.15 We observed that morbidly obese patients have increased rates of postoperative complications, consistent with what literature describes for elective cases. However, following sensitivity analysis, these results were no longer statistically significant. Acute diverticulitis is a common disease process with varying levels of severity, which can cause critical illness, sepsis, and even death at its most grave iterations.22 There are many therapeutic options and ongoing discussion about the value of resection with end colostomy, resection with primary anastomosis, and creation of diverting ostomy, in comparison with laparoscopic lavage, with limited evidence to direct surgical decisions.23,24 Although meta-analyses and prospective studies are required to better describe the role of these procedures in optimal patient outcomes, few studies have examined patient factors with regards to outcomes from diverticulitis. This study reiterated the multifactorial nature of perioperative patient outcomes, and further work is necessary to determine which patient population would benefit from an alternative to the Hartmann’s procedure. This study has several limitations. First, it is an observational study, restricting our analysis to the available data. Second, the designation of morbid obesity was made by BMI >35 kg/m2. Although this number was selected to include the patients who meet the definition of morbid obesity with a BMI >35 kg/m2 and concurrent obesity-related conditions, it is possible a stricter criteria of BMI >40 kg/m2 would have made outcomes that were not found to be significant after sensitivity analysis would have stayed significant. Third, as we used ACS NSQIP data, we only looked at short-term outcomes, as ACS NSQIP only provides data collected within 30 d of a procedure. The data include return to OR for any reason within 30 d but does not specify what additional surgical intervention was performed. Strengths of the study include the fact that we used a large national, prospectively collected database, which should generalize to the American population. We also performed a more conservative analysis. In conclusion, morbid obesity does not independently increase the risk of poor outcomes following a Hartmann’s procedure for perforated diverticulitis with the exception of increased risk of return to the operating room. The overall morbidity of obesity remains a significant factor in the perioperative management of patients with diverticulitis; however, many compromised outcomes may be related to comorbidities rather than an intrinsic function of the elevated BMI. As the rates of morbid obesity and diverticulitis continue to rise, further prospective studies are warranted to identify
populations that may benefit from less invasive treatment options for complicated diverticulitis requiring surgical intervention.
Acknowledgment Authors’ contributions: K.T.W. contributed to data acquisition and analysis, article preparation, final and review. P.C. contributed to study design, data acquisition and analysis, article preparation, and final review. N.L.G., J.A.P., A.E.A., and G.C. contributed to study design and article review. G.S. contributed to study design, data acquisition and analysis, article preparation, and final review. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Disclosure The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.
references
1. Matrana MR, Margolin DA. Epidemiology and pathophysiology of diverticular disease. Clin Colon Rectal Surg. 2009;22:141e146. 2. Greenwood-Ericksen MB, Havens JM, Ma J, Weissman JS, Schuur JD. Trends in hospital admission and surgical procedures following ED visits for diverticulitis. West J Emerg Med. 2016;17:409e417. 3. Etzioni DA, Mack TM, Beart RW, Kaiser AM. Diverticulitis in the United States: 1998e2005. Ann Surg. 2009;249:210e217. 4. Wheat CL, Strate LL. Trends in hospitalization for diverticulitis and diverticular bleeding in the United States from 2000 to 2010. Clin Gastroenterol Hepatol. 2016;14:96e103.e1. 5. Weizman AV, Nguyen GC. Diverticular disease: epidemiology and management. Can J Gastroenterol. 2011;25:385e389. 6. Maguire LH, Song M, Strate LL, Giovannucci EL, Chan AT. Association of geographic and seasonal variation with diverticulitis admissions. JAMA Surg. 2015;150:74e77. 7. Commane DM, Arasaradnam RP, Mills S, Mathers JC, Bradburn M. Diet, ageing and genetic factors in the pathogenesis of diverticular disease. World J Gastroenterol. 2009;15:2479e2488. 8. Ma W, Jovani M, Liu PH, et al. Association between obesity and weight change and risk of diverticulitis in women. Gastroenterology. 2018;155:58e66.e4. 9. Lembcke B, Kruis W. DivertikelkrankheiteKlinisches Bild und Therapie. Deut Med Wochenschr. 2015;140:1353e1359. 10. Siddiqui J, Zahid A, Hong J, Young CJ. Colorectal surgeon consensus with diverticulitis clinical practice guidelines. World J Gastrointest Surg. 2017;9:224e232. 11. Ronel DN, Hardy MA. Henri Albert Hartmann: labor and discipline. Curr Surg. 2002;59:59e64. 12. Salem L, Anaya DA, Roberts KE, Flum DR. Hartmann’s colectomy and reversal in diverticulitis: a population-level assessment. Dis Colon Rectum. 2005;48:988e995.
weber et al effect of bmi on surgery for diverticulitis
13. Feingold D, Steele SR, Lee S, et al. Practice parameters for the treatment of sigmoid diverticulitis. Dis Colon Rectum. 2014;57:284e294. 14. Penna M, Markar SR, Mackenzie H, Hompes R, Cunningham C. Laparoscopic lavage versus primary resection for acute perforated diverticulitis. Ann Surg. 2018;267:252e258. 15. Hourigan JS. Impact of obesity on surgical site infection in colon and rectal surgery. Clin Colon Rectal Surg. 2011;24:283e290. 16. Gendall KA, Raniga S, Kennedy R, Frizelle FA. The impact of obesity on outcome after major colorectal surgery. Dis Colon Rectum. 2007;50:2223e2237. 17. He Y, Wang J, Bian H, Deng X, Wang Z. BMI as a predictor for perioperative outcome of laparoscopic colorectal surgery: a pooled analysis of comparative studies. Dis Colon Rectum. 2017;60:433e445. 18. Mustain WC, Davenport DL, Hourigan JS, Vargas HD. Obesity and laparoscopic colectomy. Dis Colon Rectum. 2012;55:429e435. 19. Kamoun S, Alves A, Bretagnol F, Lefevre JH, Valleur P, Panis Y. Outcomes of laparoscopic colorectal surgery in obese and
20.
21.
22.
23.
24.
7
nonobese patients: a case-matched study of 180 patients. Am J Surg. 2009;198:450e455. Leroy J, Ananian P, Rubino F, Claudon B, Mutter D, Marescaux J. The impact of obesity on technical feasibility and postoperative outcomes of laparoscopic left colectomy. Ann Surg. 2005;241:69e76. Schwandner O, Farke S, Schiedeck TH, Bruch HP. Laparoscopic colorectal surgery in obese and nonobese patients: do differences in body mass indices lead to different outcomes? Surg Endosc. 2004;18:1452e1456. Tan JP, Barazanchi AW, Singh PP, Hill AG, Maccormick AD. Predictors of acute diverticulitis severity: a systematic review. Int J Surg. 2016;26:43e52. Cirocchi R, Afshar S, Di Saverio S, et al. A historical review of surgery for peritonitis secondary to acute colonic diverticulitis: from Lockhart-Mummery to evidence-based medicine. World J Emerg Surg. 2017;12:14. Barbieux J, Plumereau F, Hamy A. Current indications for the Hartmann procedure. J Visc Surg. 2016;153:31e38.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Effect of Body Mass Index on Outcomes After Surgery for Perforated Diverticulitis Kathryn T. Weber, MD,* Paul J. Chung, MD, Nicholas La Gamma, MD, John A. Procaccino, MD, Antonio E. Alfonso, MD, Gene Coppa, MD, and Gainosuke Sugiyama, MD Department of Surgery, Zucker School of Medicine at Hosftra Northwell, New Hyde Park, New York
article info
abstract
Article history:
Background: Despite the increased adoption of minimally invasive techniques in colorectal
Received 8 July 2019
surgery, an open resection with ostomy creation remains an accepted operation for
Received in revised form
perforated diverticulitis. In the United States, there is an increase in the rates of both
15 September 2019
morbid obesity and diverticular disease. Therefore, we wanted to explore whether out-
Accepted 1 October 2019
comes for morbidly obese patients with diverticulitis are worse than nonmorbidly obese
Available online xxx
patients after open colectomy for diverticulitis. Materials and methods: Using the American College of Surgeons National Surgical Quality
Keywords:
Improvement Program database from 2005 to 2015, we identified adults with emergent
Perforated diverticulitis
admission for diverticulitis (International Classification of Diseases, Ninth Revision, code
Morbid obesity
562.11) with evidence of preoperative sepsis and intraoperative contaminated/dirty wound
NSQIP
classification, in which a resection with ostomy (Current Procedural Terminology codes 44141, 44143, or 44144) was performed. We excluded cases with age >90 y, ventilator dependence, evidence of disseminated cancer and missing sex, race, body mass index, functional status, American Society of Anesthesiologists class, length of stay (LOS), or operative time data. Morbid obesity was defined as body mass index >35 kg/m2. Risk variables of interest included age, sex, race, medical comorbidities, requirement for preoperative transfusion, preoperative sepsis, and operative time. Outcomes of interest included LOS, 30-d postoperative complications, and mortality. Univariate and propensity scores with postmatching analyses were performed. Results: A total of 2019 patients met inclusion and exclusion criteria, of which 413 (20.5%) were morbidly obese. Morbidly obese patients tended to be younger (mean 57.2 versus 62.6 y) and female (54.5% versus 45.5%). Morbidly obese patients also had higher rates of insulin-dependent diabetes (8.0% versus 4.2%), hypertension (60.1% versus 51.3%), renal failure (3.4% versus 1.5%), and higher American Society of Anesthesiologists class (class 4: 23.5% versus 19.6% and class 5: 1.45% versus 0.87%). Morbidly obese patient had no increase in 30-d mortality or LOS, but they had higher rates of superficial wound infection (9.0% versus 5.8%; P ¼ 0.0259), deep wound infection (4.4% versus 1.9%; P ¼ 0.0073), acute renal failure (4.8% versus 2.4%; P ¼ 0.0189), postoperative septic shock (17.7% versus 12.1%; P ¼ 0.0040), and return to the operating room (11.1% versus 6.4%; P ¼ 0.0015). We identified 397 morbidly obese patients well matched by propensity score to 397 nonmorbidly obese patients. Conditional logistic regression showed no difference in LOS (median 12.9 versus 12.4 d; P ¼ 0.4648) and no increased risk of 30-d mortality (P ¼ 0.947), but morbid obesity
* Corresponding author. Department of Surgery, 300 Community Drive, Manhasset, NY, 11030. Tel.: 516-562-0100; fax: 516-562-1521. E-mail address: [email protected] (K.T. Weber). 0022-4804/$ e see front matter ª 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jss.2019.10.020
2
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
was an independent predictor for return to the operating room (adjusted odds ratio: 27.09 [95% confidence interval: 2.68-274.20]; P ¼ 0.005). Conclusions: This analysis of a large national clinical database demonstrates that morbidly obese patients presenting with perforated diverticulitis undergoing a Hartmann’s procedure do not have increased mortality or LOS compared with nonobese patients. After adjusting for the effects of morbid obesity, morbidly obese patients had increased risk of return to operating room. Despite literature describing the many perioperative risks of obesity, our analysis showed only increased reoperation for obese patients with diverticulitis. ª 2019 Elsevier Inc. All rights reserved.
Introduction Diverticulitis is commonly encountered in the United States, and its incidence is increasing both nationally and globally.1,2 Hospitalization rates for diverticulitis have increased,2-4 with an elevated incidence of diverticulitis among younger patients and higher rates of elective surgery in all age groups.5 The etiology of diverticulitis is thought to be multifactorial, with proposed risk factors including diet, poor fiber intake, inflammatory conditions, genetics, race, geography, seasonality, and age.6,7 Diverticulitis has previously been associated with dietary habits, geographic variation, and weight change.8 The standard of care for the treatment of both uncomplicated and perforated diverticulitis encompasses a broad spectrum because of the variety of disease severity at the time of presentation, and distinct clinical patterns of disease severity have emerged.9 Among patients who are acutely ill with peritonitis, the Hartmann’s procedure has endured as a widely accepted treatment.5,10 This extensive and comparatively morbid operation has persisted as the operation of choice, resisting contemporary trends toward preferential utilization of minimally invasive techniques, because it is fast, simple, and effective.11 However, the morbidity of the Hartmann’s procedure remains significant, specifically with the creation of an ostomy, which is intended to be temporary, but often remains permanent as other clinical considerations prevent the reestablishment of bowel continuity.12 Despite the theoretical advantages a minimally invasive approach may offer, particularly among obese patients, in regards to wound infection, length of stay (LOS), and maintenance of bowel continuity, current practice guidelines continue to support management of complicated diverticulitis with an operative resection with sigmoidectomy and end ostomy creation (Hartmann’s procedure).13 Laparoscopic washout has declined in favor, with evidence suggesting that resection is preferable, even in acute diverticulitis.14 Morbid obesity has implications in the management of both diverticulitis as a possible risk factor for poor outcomes following surgical management of diverticulitis. Obesity has been linked to poorer operative outcomes for both elective and emergent surgery, specifically with increased rates of wound infection and longer operative times.15-20 Many of these complications are mitigated with minimally invasive approaches, particularly among obese patients, as outcomes after open surgery tend to be worse for obese patients; current literature demonstrates longer hospital stays and increased
incidence of wound infection in obese patients undergoing open surgery compared with laparoscopic approaches.16,20 Therefore, the purpose of this study was to investigate if morbid obesity is associated with worse outcomes after Hartmann’s procedure for diverticulitis. With rising rates of both morbid obesity and diverticular disease in the United States, optimizing the standard of care for these complicated cases becomes increasingly important. We hypothesized that morbidly obese patients would have worse perioperative outcomes following open resection and end ostomy creation for perforated diverticulitis.
Methods Patient selection and data Data were obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database from 2005 to 2015. We identified cases in which an adult (aged 18 y) underwent emergent admission for diverticulitis (International Classification of Diseases, Ninth Revision, code 562.11) with evidence of preoperative sepsis and intraoperative contaminated/dirty wound classification, in which a resection with ostomy (Current Procedural Terminology codes 44141, 44143, or 44144) was performed. The combination of codes allowed us to restrict our analysis to patients with acute perforated diverticulitis who underwent a Hartmann’s. Morbid obesity was defined as body mass index (BMI) >35 kg/m2 for the purposes of this study, as most of the patients had obesity-related comorbidities such as hypertension or diabetes and therefore met the definition at a lower BMI. We excluded patients with age >90 y, cases performed by nongeneral surgeons (as characterized by NSQIP), patients who had ventilator dependence, evidence of disseminated cancer, or missing sex, race, BMI, functional status, American Society of Anesthesiologists (ASA) class, or LOS.
Risk variables and outcomes Risk variables of interest included age, sex, race, presence of diabetes, smoking status, presence of dyspnea, hypertension, functional status, history of chronic obstructive pulmonary disease (COPD), history of congestive heart failure (CHF), acute renal failure, end-stage renal disease (ESRD), steroid use, bleeding disorders, morbid obesity, preoperative transfusion of 1 unit of packed red blood cells within 72 h of surgery,
3
weber et al effect of bmi on surgery for diverticulitis
Table 1 e Patient characteristics of obese and control cohorts before and after propensity score matching. Characteristic
Overall cohorts
Matched cohorts
Nonobese (n ¼ 1606)
Morbidly obese (n ¼ 413)
Absolute standardized difference
Nonobese (n ¼ 397)
Morbidly obese (n ¼ 397)
Absolute standardized difference
62.6 (13.6)
57.2 (14.1)
0.3842
58.3 (14.36)
57.72 (13.93)
0.0415
Female
806 (50.2)
225 (54.5)
0.0429
219 (55.2)
214 (53.9)
0.0126
Male
800 (49.8)
188 (45.5)
0.0429
178 (44.8)
183 (46.1)
0.0126
Age, y, mean (SD) Sex, n (%)
BMI Race, n (%) 13 (0.8)
2 (0.5)
0.0033
1 (0.3)
2 (0.5)
0.0025
African American, n (%)
Asian, n (%)
109 (6.8)
49 (11.9)
0.0508
45 (11.3)
45 (11.3)
0.0050
Native Hawaiian/Pacific Islander, n (%)
1 (0.1)
0 (0.0)
0.0006
0 (0.0)
0 (0.0)
0.0000
American Indian or Alaskan, n (%)
17 (1.1)
3 (0.7)
0.0033
2 (0.5)
3 (0.8)
0.0025
White, n (%)
1466 (91.3)
359 (86.9)
0.0436
349 (87.9)
347 (87.4)
0.0050
Smoker, n (%)
395 (24.6)
90 (21.8)
0.0280
85 (21.4)
87 (21.9)
0.0050
0.0151
Dyspnea, n (%) None
1464 (91.2)
365 (88.4)
0.0278
359 (90.4)
353 (88.9)
Moderate exertion
97 (6.0)
34 (8.2)
0.0219
27 (6.8)
32 (8.1)
0.0126
At rest
45 (2.8)
14 (3.4)
0.0059
11 (2.8)
12 (3.0)
0.0025
1434 (89.3)
312 (75.5)
0.01375
309 (77.1)
306 (77.1)
0.0076
104 (6.5)
68 (16.5)
0.0999
59 (14.9)
61 (15.4)
0.0050
68 (4.2)
33 (8.0)
0.0376
29 (7.3)
30 (7.6)
0.0025
103 (6.4)
22 (5.3)
0.0109
25 (6.3)
22 (5.5)
0.0076
16 (1.0)
4 (1.0)
0.0003
2 (0.5)
3 (0.8)
0.0025
Diabetes, n (%) None Noninsulin dependent Insulin dependent Functional status, n (%) Partially dependent Totally dependent Independent Hypertension, n (%) History of CHF, n (%) History of COPD, n (%) Renal failure, n (%) ESRD, n (%)
1487 (92.6)
387 (93.7)
0.0111
370 (93.2)
372 (93.7)
0.0050
824 (51.3)
248 (60.1)
0.0874
241 (60.7)
235 (59.2)
0.0151 0.0025
46 (2.9)
13 (3.2)
0.0028
13 (3.27)
12 (3.0)
160 (10.0)
43 (10.4)
0.0045
43 (10.4)
43 (10.4)
0.0000
24 (1.5)
14 (3.4)
0.0190
11 (2.8)
10 (2.5)
0.0025
32 (2.0)
4 (1.0)
0.0102
4 (1.0)
4 (1.0)
0.0000
Steroid use, n (%)
236 (14.7)
52 (12.6)
0.0210
45 (11.3)
50 (12.6)
0.0126
Bleeding disorder, n (%)
171 (11.7)
50 (12.1)
0.0146
41 (10.3)
47 (11.8)
0.0151
16 (1.0)
3 (0.7)
0.0027
4 (1.0)
3 (0.8)
0.0025
Preoperative transfusion, n (%) Preoperative sepsis, n (%) SIRS Sepsis Septic shock
204 (12.7)
50 (12.1)
0.0060
45 (11.3)
50 (12.6)
0.0126
1286 (80.1)
326 (78.9)
0.0114
315 (79.4)
313 (78.8)
0.0050
116 (7.2)
37 (9.0)
0.0174
37 (9.3)
34 (8.6)
0.0076
ASA class, n (%) Class I
30 (1.9)
3 (0.7)
0.0114
6 (1.5)
3 (0.8)
0.0076
Class II
497 (31.0)
83 (20.1)
0.1085
84 (21.2)
82 (20.7)
0.0050
Class III
751 (46.8)
224 (54.2)
0.0748
219 (55.2)
220 (55.4)
0.0025
Class IV
315 (19.6)
97 (23.5)
0.0394
84 (21.2)
89 (22.4)
0.0126
Class V
14 (0.9)
6 (1.5)
0.0053
4 (1.0)
3 (0.76)
0.0025
Contaminated
129 (8.0)
31 (7.5)
0.0053
33 (8.3)
31 (7.8)
0.0050
Dirty/infected
1477 (92.0)
382 (92.5)
0.0053
364 (91.7)
366 (92.2)
0.0050
Wound class, n (%)
SD ¼ standard deviation; SIRS ¼ systemic inflammatory response syndrome.
4
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
Table 2 e Univariate analysis of unmatched morbidly obese and control cohorts. Outcome
Overall cohorts Non obese (n ¼ 1606)
Matched cohorts
Morbidly obese (n ¼ 413)
P value
Non obese (n ¼ 397)
Morbidly obese (n ¼ 397)
12.42 (8.36)
12.88 (9.18)
LOS, d, median, (SD)
10 (8.38)
10 (9.17)
0.104
Superficial wound infection, n (%)
93 (5.8)
37 (8.9)
0.026
Deep wound infection, n (%)
31 (1.9)
18 (4.4)
Organ space infection, n (%)
146 (9.1)
46 (11)
Dehiscence, n (%)
48 (2.9)
Pneumonia, n (%)
P value 0.465
21 (5.3)
33 (8.3)
0.121
0.007
8 (2.0)
18 (4.5)
0.071
0.242
38 (9.6)
43 (10.8)
0.639
13 (3.1)
0.994
10 (2.5)
13 (3.3)
0.673
121 (7.5)
36 (8.7)
0.486
31 (7.8)
36 (9.1)
0.610
Unplanned intubation, n (%)
109 (6.7)
21 (5.1)
0.252
35 (8.8)
21 (5.3)
0.072
Failure to wean ventilator, n (%)
176 (11)
59 (14)
0.073
49 (12.3)
54 (13.6)
0.673
Pulmonary embolism, n (%)
19 (12)
6 (1.4)
0.621
4 (1.0)
6 (1.5)
0.752
Urinary tract infection, n (%)
35 (2.2)
9 (2.3)
1
10 (2.5)
8 (2.0)
0.812
Progressive renal insufficiency, n (%)
15 (<1)
5 (1.2)
0.581
4 (1.0)
3 (0.8)
1
Acute renal failure, n (%)
40 (2.49)
20 (4.84)
0.019
14 (3.5)
19 (4.8)
0.477
Urinary tract infection, n (%)
35 (2.2)
9 (2.3)
1
10 (2.5)
8 (2.0)
0.812
VTE/thrombophlebitis, n (%)
59 (3.6)
15 (3.6)
1
12 (3.0)
15 (3.8)
0.695
Myocardial infarct, n (%)
23 (1.4)
3 (<1)
0.333
7 (1.8)
3 (0.8)
0.341
Cardiac arrest, n (%)
20 (1.2)
6 (1.4)
0.806
6 (1.5)
5 (1.3)
1
Postoperative transfusion, n (%)
228 (14)
52 (13)
0.446
66 (16.6)
47 (11.8)
0.068
Postoperative sepsis, n (%)
306 (19)
90 (22)
0.23
74 (18.6)
89 (22.4)
0.219
Postoperative septic shock, n (%)
195 (12)
73 (18)
0.004
61 (15.4)
66 (16.6)
0.699
Return to operating room, n (%)
103 (6.4)
46 (11)
0.001
21 (5.3)
42 (10.6)
0.009
83 (5.2)
21 (5.1)
1
21 (5.3)
20 (5.0)
1
Death, n (%)
Bold values are reached statistical significance where P < 0.05. SD ¼ standard deviation; VTE ¼ venous thromboembolism.
systemic sepsis within 48 h of surgery, operative wound classification, and ASA classification. Primary outcomes of interest were 30-d mortality and total LOS. Secondary outcomes of interest included superficial, deep, and organ-space wound infections, wound dehiscence, pneumonia, failure to extubate, unplanned intubation, pulmonary embolism, venous thromboembolism, urinary tract infection, progressive renal insufficiency, acute renal failure, myocardial infarction, cardiac arrest, bleeding requiring transfusion, sepsis, septic shock, and return to the operating room.
transfusions, preoperative sepsis, ASA classification, and wound class. Quality of match was assessed using the absolute standardized mean difference with a goal of 0.2. For categorical outcomes, we conducted conditional logistic regression adjusting for all the matched variables, and for continuous variables, we used the Wilcoxon rank-sum test. We then performed sensitivity analysis by conducting propensity score matching using the same parameters with the exception of using a caliper of 0.1 and then performed postmatch analysis. The R programming language version 3.4 was used for statistical analysis. Institutional Review Board exemption was obtained for this study as it used deidentified data from a publicly available source.
Statistical analysis Univariate analysis and propensity score matching with postmatching analysis were performed. We used propensity score matching using the so-called nearest neighbor method with a caliper of 0.25 to create a 1:1 match of morbidly obese to control cases over patient characteristics including age, sex, race, diabetes, smoking status, dyspnea, hypertension, functional status, history of COPD, history of CHF, renal failure, ESRD, steroid use, bleeding disorder, preoperative blood
Results Patient characteristics A total of 2019 patients met inclusion and exclusion criteria, of which 413 (20.5%) were morbidly obese. Compared with control, morbidly obese patients tended to be younger (mean 57.2
5
weber et al effect of bmi on surgery for diverticulitis
versus 62.6 y) and had a higher proportion of female compared with male patients (54.5% versus 45.5%). Morbidly obese patients had a median BMI of 39.3, compared with 27.4 in the nonobese group. Morbidly obese patients also had higher rates of insulin-dependent diabetes (8.0% versus 4.2%), hypertension (60.1% versus 51.3%), renal failure (3.4% versus 1.5%), and higher ASA class (class 4: 23.5% versus 19.6% and class 5: 1.45% versus 0.87%; Table 1).
Univariate analysis Unadjusted comparison of morbidly obese patients to control demonstrated higher rates superficial wound infection (9.0% versus 5.8%; P ¼ 0.0259), deep wound infection (4.4% versus 1.9%; P ¼ 0.0073), acute renal failure (4.8% versus 2.4%; P ¼ 0.0189), postoperative septic shock (17.7% versus 12.1%; P ¼ 0.0040), and return to the operating room (11.1% versus 6.4%; P ¼ 0.0015). However, there were no differences found in LOS between the morbidly obese group and control (median 10 versus 10 d; P ¼ 0.1038; Table 2).
of variables was nonsignificant after propensity matching the patients. Conditional logistic regression showed that morbid obesity was an independent predictor for return to the operating room (adjusted odds ratio [OR] 27.09 [95% confidence interval (CI) 2.68-274.20]; P ¼ 0.005; Table 3).
Sensitivity analysis Sensitivity analysis was performed by setting the caliper to 0.1, which resulted in 398 morbidly obese patients being matched to 398 controls. The postmatch maximum absolute standardized difference between covariates was 0.0231, signifying good match. There was no significant difference in the LOS between the matched morbidly obese group and control (median 10 versus 10 d; P ¼ 0.1969). Conditional logistic regression again showed that morbid obesity was still an independent predictor for return to the operating room (adjusted OR 2.72 [95% CI 1.06-6.97; P ¼ 0.038; Table 4).
Propensity score match and analysis
Discussion
Propensity matched score with a caliper of 0.25 resulted in 397 morbidly obese patients being matched to 397 controls. Postmatch, the maximum absolute standardized difference was 0.0415, which confirmed good match. There was no difference in the LOS between the matched morbidly obese cohort and control (median 12.9 versus 12.4 d; P ¼ 0.4648). The remainder
This analysis of a large national clinical database demonstrated that morbidly obese patients compared with nonobese patients do not have increased 30-d mortality or longer LOS after an emergent Hartmann’s procedure for perforated diverticulitis. However, morbidly obese patients did have
Table 3 e Outcomes of matched morbidly obese and control cohorts. Outcome
Table 4 e Outcomes of matched morbidly obese and control cohorts and sensitivity analysis. Outcome
Adjusted OR
95% CI
P value
Adjusted OR
95% CI
P value
0.87-6.40
0.090
0.8479.73
0.071
2.37
Superficial wound infection
8.16
Superficial wound infection
0.83-1.12
0.621
0.7001.83
0.740
0.96
Deep space wound infection
0.96
Deep space wound infection
0.56-2.03
0.842
1.42
0.64-3.18
0.390
Organ space wound infection
1.07
Organ space wound infection
Pneumonia
2.93
1.05
0.43-2.57
0.907
0.7910.86
0.108
Pneumonia Unplanned intubation
0.97
0.29-3.26
0.962
Unplanned intubation
0.91
0.08-1.03
0.053
Failure to wean ventilator
1.78
0.88-3.61
0.110
Failure to wean ventilator
1.16
0.63-2.13
0.629
Acute renal failure
0.98
0.84-1.14
0.823
Acute renal failure
1.00
0.86-1.16
0.961
VTE/thrombophlebitis
0.99
0.86-1.15
0.923
VTE/thrombophlebitis
1.00
0.87-1.17
0.954
Myocardial infarct
1.01
0.87-1.17
0.889
Myocardial infarct
1.01
0.87-1.17
0.938
Cardiac arrest
1.00
0.86-1.15
0.971
Cardiac arrest
1.01
0.86-1.17
0.931
Postoperative transfusion
0.55
0.27-1.10
0.091
Postoperative transfusion
0.65
0.38-1.11
0.115
Postoperative sepsis
1.02
0.68-1.52
0.932
Postoperative sepsis
1.11
0.77-1.61
0.580
Postoperative septic shock
1.44
0.81-2.57
0.215
Postoperative septic shock
1.40
0.86-2.29
0.177
Return to operating room
27.09
2.68274.2
0.005
Return to operating room
2.72
1.06-6.97
0.038
Dehiscence, pulmonary embolism, progressive renal insufficiency, and death were nonsignificant. Urinary tract infection did not converge. Bold values are reached statistical significance where P < 0.05. VTE ¼ venous thromboembolism.
Pulmonary embolism, urinary tract infection, and death were nonsignificant. Dehiscence and progressive renal insufficiency did not converge. Bold values are reached statistical significance where P < 0.05. VTE ¼ venous thromboembolism.
6
j o u r n a l o f s u r g i c a l r e s e a r c h - 2 0 1 9 ( - ) 1 e7
increased morbidity as characterized by increased rates of wound infection, postoperative septic shock, acute renal failure, and return to the operating room. Despite these worse perioperative outcomes, after performing sensitivity analysis, morbidly obese patients were consistently at risk only for return to the operating room compared with matched nonobese control. Obesity is implicated in perioperative complications in elective colorectal surgeries, with increased rates of wound infection and prolonged operative times.21 Although emergent operations have increased morbidity compared with elective surgeries, obese patients are particularly vulnerable to worse outcomes.15 We observed that morbidly obese patients have increased rates of postoperative complications, consistent with what literature describes for elective cases. However, following sensitivity analysis, these results were no longer statistically significant. Acute diverticulitis is a common disease process with varying levels of severity, which can cause critical illness, sepsis, and even death at its most grave iterations.22 There are many therapeutic options and ongoing discussion about the value of resection with end colostomy, resection with primary anastomosis, and creation of diverting ostomy, in comparison with laparoscopic lavage, with limited evidence to direct surgical decisions.23,24 Although meta-analyses and prospective studies are required to better describe the role of these procedures in optimal patient outcomes, few studies have examined patient factors with regards to outcomes from diverticulitis. This study reiterated the multifactorial nature of perioperative patient outcomes, and further work is necessary to determine which patient population would benefit from an alternative to the Hartmann’s procedure. This study has several limitations. First, it is an observational study, restricting our analysis to the available data. Second, the designation of morbid obesity was made by BMI >35 kg/m2. Although this number was selected to include the patients who meet the definition of morbid obesity with a BMI >35 kg/m2 and concurrent obesity-related conditions, it is possible a stricter criteria of BMI >40 kg/m2 would have made outcomes that were not found to be significant after sensitivity analysis would have stayed significant. Third, as we used ACS NSQIP data, we only looked at short-term outcomes, as ACS NSQIP only provides data collected within 30 d of a procedure. The data include return to OR for any reason within 30 d but does not specify what additional surgical intervention was performed. Strengths of the study include the fact that we used a large national, prospectively collected database, which should generalize to the American population. We also performed a more conservative analysis. In conclusion, morbid obesity does not independently increase the risk of poor outcomes following a Hartmann’s procedure for perforated diverticulitis with the exception of increased risk of return to the operating room. The overall morbidity of obesity remains a significant factor in the perioperative management of patients with diverticulitis; however, many compromised outcomes may be related to comorbidities rather than an intrinsic function of the elevated BMI. As the rates of morbid obesity and diverticulitis continue to rise, further prospective studies are warranted to identify
populations that may benefit from less invasive treatment options for complicated diverticulitis requiring surgical intervention.
Acknowledgment Authors’ contributions: K.T.W. contributed to data acquisition and analysis, article preparation, final and review. P.C. contributed to study design, data acquisition and analysis, article preparation, and final review. N.L.G., J.A.P., A.E.A., and G.C. contributed to study design and article review. G.S. contributed to study design, data acquisition and analysis, article preparation, and final review. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Disclosure The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.
references
1. Matrana MR, Margolin DA. Epidemiology and pathophysiology of diverticular disease. Clin Colon Rectal Surg. 2009;22:141e146. 2. Greenwood-Ericksen MB, Havens JM, Ma J, Weissman JS, Schuur JD. Trends in hospital admission and surgical procedures following ED visits for diverticulitis. West J Emerg Med. 2016;17:409e417. 3. Etzioni DA, Mack TM, Beart RW, Kaiser AM. Diverticulitis in the United States: 1998e2005. Ann Surg. 2009;249:210e217. 4. Wheat CL, Strate LL. Trends in hospitalization for diverticulitis and diverticular bleeding in the United States from 2000 to 2010. Clin Gastroenterol Hepatol. 2016;14:96e103.e1. 5. Weizman AV, Nguyen GC. Diverticular disease: epidemiology and management. Can J Gastroenterol. 2011;25:385e389. 6. Maguire LH, Song M, Strate LL, Giovannucci EL, Chan AT. Association of geographic and seasonal variation with diverticulitis admissions. JAMA Surg. 2015;150:74e77. 7. Commane DM, Arasaradnam RP, Mills S, Mathers JC, Bradburn M. Diet, ageing and genetic factors in the pathogenesis of diverticular disease. World J Gastroenterol. 2009;15:2479e2488. 8. Ma W, Jovani M, Liu PH, et al. Association between obesity and weight change and risk of diverticulitis in women. Gastroenterology. 2018;155:58e66.e4. 9. Lembcke B, Kruis W. DivertikelkrankheiteKlinisches Bild und Therapie. Deut Med Wochenschr. 2015;140:1353e1359. 10. Siddiqui J, Zahid A, Hong J, Young CJ. Colorectal surgeon consensus with diverticulitis clinical practice guidelines. World J Gastrointest Surg. 2017;9:224e232. 11. Ronel DN, Hardy MA. Henri Albert Hartmann: labor and discipline. Curr Surg. 2002;59:59e64. 12. Salem L, Anaya DA, Roberts KE, Flum DR. Hartmann’s colectomy and reversal in diverticulitis: a population-level assessment. Dis Colon Rectum. 2005;48:988e995.
weber et al effect of bmi on surgery for diverticulitis
13. Feingold D, Steele SR, Lee S, et al. Practice parameters for the treatment of sigmoid diverticulitis. Dis Colon Rectum. 2014;57:284e294. 14. Penna M, Markar SR, Mackenzie H, Hompes R, Cunningham C. Laparoscopic lavage versus primary resection for acute perforated diverticulitis. Ann Surg. 2018;267:252e258. 15. Hourigan JS. Impact of obesity on surgical site infection in colon and rectal surgery. Clin Colon Rectal Surg. 2011;24:283e290. 16. Gendall KA, Raniga S, Kennedy R, Frizelle FA. The impact of obesity on outcome after major colorectal surgery. Dis Colon Rectum. 2007;50:2223e2237. 17. He Y, Wang J, Bian H, Deng X, Wang Z. BMI as a predictor for perioperative outcome of laparoscopic colorectal surgery: a pooled analysis of comparative studies. Dis Colon Rectum. 2017;60:433e445. 18. Mustain WC, Davenport DL, Hourigan JS, Vargas HD. Obesity and laparoscopic colectomy. Dis Colon Rectum. 2012;55:429e435. 19. Kamoun S, Alves A, Bretagnol F, Lefevre JH, Valleur P, Panis Y. Outcomes of laparoscopic colorectal surgery in obese and
20.
21.
22.
23.
24.
7
nonobese patients: a case-matched study of 180 patients. Am J Surg. 2009;198:450e455. Leroy J, Ananian P, Rubino F, Claudon B, Mutter D, Marescaux J. The impact of obesity on technical feasibility and postoperative outcomes of laparoscopic left colectomy. Ann Surg. 2005;241:69e76. Schwandner O, Farke S, Schiedeck TH, Bruch HP. Laparoscopic colorectal surgery in obese and nonobese patients: do differences in body mass indices lead to different outcomes? Surg Endosc. 2004;18:1452e1456. Tan JP, Barazanchi AW, Singh PP, Hill AG, Maccormick AD. Predictors of acute diverticulitis severity: a systematic review. Int J Surg. 2016;26:43e52. Cirocchi R, Afshar S, Di Saverio S, et al. A historical review of surgery for peritonitis secondary to acute colonic diverticulitis: from Lockhart-Mummery to evidence-based medicine. World J Emerg Surg. 2017;12:14. Barbieux J, Plumereau F, Hamy A. Current indications for the Hartmann procedure. J Visc Surg. 2016;153:31e38.