Data Source and Study Population

Data was obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) targeted colectomy database. As the ACS-NSQIP database is de-identified and publicly available, Institutional Review Board review was not required. The database was queried for patients with a Body Mass Index (BMI) less than 30 (non-obese) or between 35-40 (obese) undergoing elective, non-emergent, minimally invasive partial colectomy with or without primary anastomosis for an indication of diverticular disease from 2012-2019. Included CPT codes were 44140, 44141, 44143-44146, 44204, and 44206-44208. Patients with a BMI over 40 were excluded as these patients would have already met criteria for bariatric surgery. Patients with diabetes, ascites, a history of congestive heart failure or myocardial infarction, and hypertension requiring medical management were excluded as these would be considered a weight related comorbidity that would meet criteria for bariatric surgery. Patients with disseminated cancer, ventilator support, pre-operative sepsis, Association of Anesthesia Classification 5, or incomplete data were also excluded.

Outcomes

The outcomes of interest included anastomotic leak, conversion to open, colostomy or ileostomy construction, overall surgical site infections (SSI), superficial SSI, deep SSI, organ-space SSI, readmission, reoperation, operative time, and length of stay.

Statistical Analysis

Computations were performed in R (R Core Team, 2020). A 2:1 propensity scored nearest neighbor match was performed. Patients with a BMI less than 30 were matched against patients with a BMI between 35 and 40. Patients were matched for age, sex, smoking status, functional status, history of COPD, history of steroid use, dialysis requirements, pre-operative weight loss greater than or equal to 10%, bleeding disorders, and ASA class. A covariate balance assessment was performed using the standardized mean difference (SMD) with an SMD greater than or equal to 0.1 considered unbalanced. A post-weighting balance assessment was performed to ensure an optimal balance. Outcomes were compared both before and after matching. Binomial outcomes were reported as rates and hypothesis testing conducted with a standard test for binomial proportions. Continuous outcomes were reported as means with hypothesis testing conducted with a two-sample t-test. A p-value of 0.05 or less was considered significant.

Pre-match balance assessment and outcomes

Prior to matching, the groups were relatively well balanced with significant differences in the SMD of age, pre-operative weight loss >10%, and ASA class. The BMI 35-40 group tended to be slightly younger, with less reported pre-operative weight loss, and with a higher proportion of ASA class 3 patients.

	BMI <30	BMI 35-40	SMD
n	1954	263
sex = TRUE (%)	931 ( 47.6)	120 ( 45.6)	0.040
age (mean (SD))	56.12 (11.54)	47.60 (11.37)	0.743
smoke = TRUE (%)	374 ( 19.1)	50 ( 19.0)	0.003
fnstatus2 (%)			0.056
Independent	1934 ( 99.8)	262 (100.0)
Partially dependent	2 ( 0.1)	0 ( 0.0)
Totally dependent	1 ( 0.1)	0 ( 0.0)
hxcopd = TRUE (%)	40 ( 2.0)	3 ( 1.1)	0.072
dialysis = FALSE (%)	1954 (100.0)	263 (100.0)	<0.001
steroid = TRUE (%)	51 ( 2.6)	8 ( 3.0)	0.026
wtloss = TRUE (%)	48 ( 2.5)	0 ( 0.0)	0.224
bleeddis = TRUE (%)	13 ( 0.7)	2 ( 0.8)	0.011
transfus = TRUE (%)	1 ( 0.1)	0 ( 0.0)	0.032
asaclas (%)			0.439
1	133 ( 6.8)	5 ( 1.9)
2	1456 ( 74.5)	164 ( 62.4)
3	360 ( 18.4)	93 ( 35.4)
4	5 ( 0.3)	1 ( 0.4)
5	0 ( 0.0)	0 ( 0.0)

A significant difference in outcomes existed only for operative time.

Outcome	Group	N	Missing	Events	Rate (%)	Difference (%)	95% CI	p
Anastomotic Leak
	BMI <30	1954	5	54	2.77
	BMI 35-40	263	1	13	4.96	-2.19	-4.92 to 0.54	0.052
Conversion
	BMI <30	1954	0	173	8.85
	BMI 35-40	263	0	30	11.41	-2.56	-6.6 to 1.49	0.178
Ostomy
	BMI <30	1954	0	61	3.12
	BMI 35-40	263	0	8	3.04	0.08	-2.13 to 2.29	0.944
Overall SSI
	BMI <30	1954	0	110	5.63
	BMI 35-40	263	0	18	6.84	-1.21	-4.43 to 2	0.428
Superficial SSI
	BMI <30	1954	0	35	1.79
	BMI 35-40	263	0	4	1.52	0.27	-1.32 to 1.86	0.754
Deep SSI
	BMI <30	1954	0	12	0.61
	BMI 35-40	263	0	2	0.76	-0.15	-1.25 to 0.96	0.779
Organ-space SSI
	BMI <30	1954	0	69	3.53
	BMI 35-40	263	0	14	5.32	-1.79	-4.63 to 1.04	0.151
Readmission
	BMI <30	1954	0	100	5.12
	BMI 35-40	263	0	19	7.22	-2.1	-5.38 to 1.17	0.155
Reoperation
	BMI <30	1954	0	57	2.92
	BMI 35-40	263	0	12	4.56	-1.64	-4.28 to 0.98	0.149

Outcome	Group	N	Missing	Mean	Lower SD	Upper SD	Mean difference	95% CI	p
Operative time (min)
	BMI <30	1954	0	166.3	94.4	238.2
	BMI 35-40	263	0	178.4	102.8	254.0	-12.1	-21.8 to -2.4	0.015
LOS (days)
	BMI <30	1954	3	3.7	1.0	6.4
	BMI 35-40	263	0	3.7	1.1	6.3	0	-0.3 to 0.4	0.855

Post-match balance assessment

	BMI <30	BMI 35-40	SMD
n	522	261
sex = TRUE (%)	238 ( 45.6)	118 ( 45.2)	0.008
age (mean (SD))	47.94 (11.25)	47.48 (11.29)	0.041
smoke = TRUE (%)	99 ( 19.0)	50 ( 19.2)	0.005
fnstatus2 = Independent (%)	522 (100.0)	261 (100.0)	<0.001
hxcopd = TRUE (%)	5 ( 1.0)	3 ( 1.1)	0.019
dialysis = FALSE (%)	522 (100.0)	261 (100.0)	<0.001
steroid = TRUE (%)	8 ( 1.5)	6 ( 2.3)	0.056
wtloss = FALSE (%)	522 (100.0)	261 (100.0)	<0.001
bleeddis = TRUE (%)	3 ( 0.6)	2 ( 0.8)	0.023
transfus = FALSE (%)	522 (100.0)	261 (100.0)	<0.001
asaclas (%)			0.075
1	9 ( 1.7)	5 ( 1.9)
2	341 ( 65.3)	162 ( 62.1)
3	171 ( 32.8)	93 ( 35.6)
4	1 ( 0.2)	1 ( 0.4)
5	0 ( 0.0)	0 ( 0.0)