Mechanical bowel preparations not supported in elective colo-rectal surgeries with anastomosis: A retrospective study

Abstract

Objective

To analyze the risk and benefit of bowel preparations in elective colo-rectal surgery.

Background

Mechanical bowel preparations (MBPs) have been popularized in colo-rectal surgery since studies in the 1970s, but recent data has called their use into question and examined complication rates between patients with and without bowel preparations.

Methods

A retrospective case-review was performed consisting of 1237 elective colo-rectal surgeries performed by two surgeons between 2008 and 2021. Patients received either a MBP, a mechanical bowel preparation with oral antibiotics (OAMBP), oral antibiotics alone (OA), or no bowel preparation; some patients across all categories received an enema.

Results

Bowel preparations combined (MBP and OAMBP) totaled 436 patients and showed no statistically significant difference (P > .05) in primary outcomes of wound infection and anastomotic leak when compared to the 636 patients without a bowel preparation and 165 patients with OA. The analysis controlled for comorbidities and presence of enema. Of secondary outcomes, urinary tract infections (UTIs) were significantly more common in patients who received a bowel preparation (P = .047). All other outcomes showed no significant difference between groups, including complications on day of surgery; complications, readmission with and without surgery, and ileus formation within 30 days of surgery; sepsis; pneumonia; and length of stay (LOS). The presence of enemas did not have a statistically significant effect on outcomes.

Conclusions

This study’s data does not support the routine use of MBPs in elective colo-rectal surgery and draws into further question whether MBPs should remain standard of care.

Keywords

bowel preparation mechanical bowel preparation rural surgery colorectal

Key Takeaways

• Bowel preparations, despite being standard of care in elective colo-rectal surgeries with anastomoses, did not show any benefit to use and were associated with a significantly higher rate of urinary tract infections in n = 1237 patients.

• The use of enemas and oral antibiotics had no significant effect on patient outcomes, but enemas trended towards significance with increased rates of post-operative urinary tract infections and oral antibiotics trended towards significance with increased rates of post-operative pneumonia.

• The use of routine bowel preparations is not supported by this data.

Introduction

Mechanical bowel preparations (MBPs), which facilitate the mechanical removal of intestinal contents, were first suggested by Sir William Halstead in the 1880s to address high infection rates in colo-rectal surgery.¹ Even today, literature finds a weighted mean of 11.4% of colo-rectal surgeries result in surgical site infection (SSIs),² along with 8.1% of anastomoses resulting in anastomotic leak.³ Historically, these issues were addressed in the 1970s by two landmark studies that examined oral antibiotics (OA) alone and mechanical bowel preparations with oral antibiotics (OAMBPs). In 1973, Nichols’ study indicated that pre-operative neomycin and erythromycin reduced surgical site infections (SSIs).⁴ Four years later, Clarke combined the erythromycin-neomycin with MBPs and found reduced SSIs and anastomotic leak.⁵ Following these publications, BPs became standard of care and have since been supported by a significant body of work.^6–11 The most recent (2019) American Society of Colon and Rectal Surgeons’ clinical practice guidelines for bowel preparation in elective colo-rectal surgeries still recommends OAMBPs as standard of care.²

Though some early studies questioned the colo-rectal surgical field’s adherence to bowel preps—notably Hughes’ 1972 analysis¹² which concluded that the “omission of enemas and bowel washes from the pre-operative procedures will be welcomed by both patients and nursing staff”—the bulk of the data questioning MBPs has been in the last fifteen years.^13–20 Of particular note, the most recent Cochrane review examined 18 studies totaling over 5000 patients and found no benefit to MBPs or enemas.¹³ Another meta-analysis of seven studies totaling 1454 patients found a significantly higher rate of anastomotic leak in patients that had received MBPs.¹⁴

Given the conflicting literature, in addition to poor patient tolerance and adherence, we aimed to evaluate the use of BPs in elective colo-rectal surgery and its effect on several post-surgical outcomes following a retrospective review of patients at one center over 13 years.

Methods

This study was reviewed and approved by the Institutional Review Board at Essentia Health Duluth with no reported considerations. We performed a retrospective review of 1237 elective colo-rectal surgeries with anastomoses performed by two surgeons at the same center using similar techniques between 2008 and 2021. The same surgical techniques were used between surgeons and the post-operative protocols were identical. We examined outcomes between three primary variables: patients who received a bowel preparation (either an MBP or an OAMBP), patients without a bowel preparation, and patients who received OA only. Some patients from all categories, though primarily rectal and sigmoid resections, also received an enema, which became a secondary variable of analysis along with comparing open and minimally invasive surgeries (MIS).

The surgeries we examined included partial colectomy with primary colocolonic, ileocolic, rectal, or low colo-rectal anastomoses; total proctocolectomy with ileoanal pouch; total abdominal colectomy with ileorectal anastomosis; subtotal colectomy with anastomosis; and colostomy reversals with anastomosis. Enemas given to patients examined in this study were self-administered fleets enemas, in the pre-operative area the morning of surgery. Data involving the methods of surgery, both open and MIS (laparoscopic and robotic combined) were collected.

The primary outcomes measured were SSI and anastomotic leak. Secondary outcomes included urinary tract infection (UTI), Clostridium difficile (C. difficile) infection, ileus formation, complications within 30 days, readmission for any reason, sepsis, pneumonia, and length of stay (LOS).

The data were analyzed using a general estimating equation (GEE), with a GEE mean for the quantitative variable (LOS) and a GEE binary logistic model for the remaining outcome variables. Length of stay was analyzed by ranking, with median considered to be a ranked score of 619; for example, scores above 619 indicate a cohort of patients that stayed longer than the mean of 5.9 days (with two significant outliers excluded). A standard 95% confidence interval (CI) was considered significant. A multivariate analysis was conducted and we controlled for influential covariates including age, gender, and patient comorbidities. Comorbidities included congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), and renal disease (any stage of CKD).

Results

The cohort was 51.1% female; 43.6% of patients had renal disease; 22.6% had CAD; and 13.5% had CHF. Specific data were unable to be collected on race, but the patient cohort had an overwhelming majority of Caucasian patients. For context, per the US census, St. Louis County, where Duluth, MN is located, is demographically 88.0% White, 4.8% two or more races, 2.2% Native American, 2.0% Black or African American, 1.8% Hispanic or Latino, .9% Asian, and .3% other.²¹ The study population came from a majority rural area, with a catchment area that extends from the upper peninsula of Michigan to North Dakota, covering primarily rural northern Minnesota and Wisconsin.²²

Retrospective data acquisition was not always able to distinguish OAMBPs from MBPs, so both were combined into one category of bowel preparations (preps). Of the 1237 patients, 436 (35.2%) had a bowel preparation, of which 76 (25%) had an enema; 165 (13.3%) had only OA, of which 48 (15.8%) had an enema; and 636 (51.4%) had no bowel prep, of which 180 (59.2%) had an enema. In total, 304 patients (24.6%) had an enema. Of the 369 (29.8%) patients with open surgery, 196 (53.1%) had no bowel prep, 56 (15.2%) had OA only, and 117 (31.7%) had a bowel prep. Comparatively, of the 868 patients with MIS, 440 (50.7%) had no bowel prep, 109 (12.6%) had OA only, and 319 (36.8%) had a bowel preparation. Enemas were more common in patients without a prep than patients with a prep or patients with OA (Table 1).

Table 1.

Bowel preparation type cohorts by method of surgery and percent with enema.

Category	Total (N, % pts)	With Enema (N, % pts)
No BP	636 (51.4)	180 (59.2)
MIS	440 (69.2)	151 (83.9)
Open	196 (30.8)	29 (16.1)
OA only	165 (13.3)	48 (15.8)
MIS	109 (66.1)	36 (75.0)
Open	56 (33.9)	12 (25.0)
Preps	436 (35.2)	76 (25.0)
MIS	319 (73.2)	64 (84.2)
Open	117 (26.8)	12 (15.8)

A majority of patients in this study cohort did not receive any bowel preparation. Between all groups, similar percentages (about three-quarters to four-fifths) of patients had MIS. Most patients without a bowel preparation got an enema, and patients without any bowel prep were much more likely to get enemas than patients with OA only or preps.

Comparing bowel preparations to patients with OA to patients without preps, no benefit to preps, enemas, or OA was seen. The only significant outcome (P = .047) was an association between patients who had bowel preparations and post-operative UTIs (Table 2). This may be influenced by the relatively low number of actual incidences, and would be found not to be significant if there were greater power to the study.

Table 2.

Multivariate analysis: Significance of bowel preparations, OA, Enemas, and Open surgery by secondary outcomes.

Variable	Category	No	Yes	P-value	OR	95% CI
Anastomotic leak	Preps	.80%	1.00%	.94	.957	.308, 2.978
	OA only	.90%	.80%	.617	1.309	.456, 3.761
	Enemas	.60%	1.30%	.204	2.016	.683, 5.949
	Open	.90%	.80%	.754	1.166	.447, 3.044
SSI	Preps	.70%	1.10%	.533	1.454	.448, 4.723
	OA only	1.00%	.70%	.91	1.08	.285, 4.085
	Enemas	.60%	1.30%	.107	3.127	.781, 12.518
	Open	.40%	1.60%	.004	5.938	1.784,19.764
UTI	Preps	.10%	.50%	.047	4.273	1.023, 17.854
	OA only	.20%	.20%	.71	1.466	.195, 10.988
	Enemas	.10%	.60%	.052	7.088	.983, 51.104
	Open	.10%	.40%	.06	3.506	.947, 12.985
Sepsis	Preps	.80%	.90%	.924	.944	.285, 3.120
	OA only	.70%	1.10%	.144	2.396	.741, 7.744
	Enemas	.90%	.90%	.956	1.038	.277, 3.883
	Open	.50%	1.40%	.062	2.733	.952, 7.848
Pneumonia	Preps	.40%	1.40%	.226	4.749	.381, 59.196
	OA only	.90%	.30%	.056	.281	.077, 1.031
	Enemas	.50%	1.10%	.715	1.369	.253, 7.408
	Open	.40%	1.20%	.159	4.656	.547, 39.616
LOS - Rank	Preps	731.6	727.9	.854	.023	7.010E-20, 7.383 E+15
	OA only	735.9	723.6	.575	<.001	9.570E-25, 2.185 E+13
	Enemas	731.1	728.4	.905	.063	1.119E-21, 3.519 E+18
	Open	618.8	840.7	<.001	2.49 E+96	2.430 E+79, 2.544 E+113
Complications within 30 days	Preps	2.90%	3.50%	.747	.952	.705, 1.285
	OA only	3.00%	3.50%	.966	1.007	.740, 1.370
	Enemas	2.60%	4.00%	.238	1.216	.879, 1.683
	Open	2.40%	3.90%	<.001	2.105	1.580, 2.805
Ileus within 30 days	Preps	1.50%	1.90%	.765	.935	.604, 1.449
	OA only	1.60%	1.80%	.68	1.095	.711, 1.689
	Enemas	1.40%	2.10%	.906	1.029	.643, 1.644
	Open	1.20%	2.30%	.07	1.486	.969, 2.280
Readmission within 30 days	Preps	2.10%	2.50%	.887	.971	.652, 1.468
	OA only	2.40%	2.30%	.903	.974	.638, 1.487
	Enemas	2.10%	2.60%	.46	.845	.540, 1.322
	Open	1.70%	3.10%	.006	1.758	1.177, 2.625
Readmission within 30 days with surgery	Preps	1.00%	1.30%	.924	.959	.407, 2.258
	OA only	1.50%	.90%	.297	.631	.265, 1.501
	Enemas	1.00%	1.30%	.78	.875	.343, 2.231
	Open	.70%	1.70%	.023	2.248	1.121, 4.508

Significant P-values bolded. OR compare against no preparation. Open surgery was significant for several outcomes. Bowel preparations were significantly more likely to be associated with post-operative UTIs.

Multivariate analysis indicated six covariates that held significance with regards to different variables: gender, renal disease, CAD, CHF, COPD, and patient age. These covariates were controlled for during data analysis, and their level of significance is recorded in Table 3.

Table 3.

Statistical significance of influence of covariates on secondary outcomes.

Variable	Gender	Renal Disease	CAD	CHF	COPD	Age
LOS - Rank	.026	< .001	.025	< .001	.186	.980
Complications within 30 days	.005	< .001	.254	.089	.849	.739
Ileus within 30 days	.232	.019	.555	.851	.899	.911
Readmission within 30 days	.906	< .001	.022	.775	.446	.023
Readmission within 30 days with surgery	.404	.015	.239	.685	.957	.16
Anastomotic leak	.142	.025	.718	.934	.633	.514
SSI	.12	.664	.262	.412	.213	.542
UTI	.044	.035	.982	.256	.006	.348
Sepsis	.136	.02	.914	.936	.034	.873
Pneumonia	.961	.91	.501	.164	.012	.003

Significant P-values bolded. Gender as a significant covariate indicates that women were associated with longer LOS, higher complication rates within 30 days of the surgery date, and post-surgical UTIs. Renal disease, in particular, has a significant effect on all variables except SSIs and post-operative pneumonia.

The incidence of C. difficile infection was too low during the study period to analyze C. difficile infection with regards to bowel preparations.

Discussion

Our study data did not show any significant benefit to MBPs, OAMBPs, or OA use and showed a statistically significant association between bowel preparations and increased post-operative UTIs (Table 2). Additionally, there were no associations with negative outcomes found in not performing bowel preparations prior to these elective colo-rectal procedures.

Six significant covariates were found in multivariate analysis: gender, renal disease, CAD, CHF, COPD, and patient age. The latter five are known risk factors, but further research is warranted into why gender was a significant covariate for LOS, complication rate within 30 days of surgery, and post-operative UTIs. One potential explanation is that women are more likely to develop UTIs, thus raising the rate of post-operative complications and potentially increasing LOS. The effect of gender may therefore also influence the association noted between bowel preparations and increased post-operative UTIs, though gender did not affect when Foley catheters were removed. However, other confounding factors may influence the gender differences found in this data. While the other covariates are unsurprising in their influence, it is noteworthy that renal disease was found to be a significant association with so many negative outcomes, especially compared to the other covariates.

The data showed several correlations between open surgeries and negative outcomes, along with a few trends towards significance. While there is likely merit to the correlations, the confounding variable is that the cases that were performed open were likely done due to increased complexity and were more likely to develop complications. Though we have attempted to control for this in our analysis, it is possible that this confounder is still influencing the data.

There are several limitations to this study. Our data is retrospective and does not differentiate between MBPs and OAMBPs. Further, incidence of the outcomes of interest was low enough that many of the secondary outcomes were relatively underpowered, with large confidence intervals, despite a large sample size (N = 1237). This meant that some of our data trended towards significance, rather than being more definitively significant or not significant (P-values between .05 and .10). Open surgery trended towards significance in several outcomes. Additionally, enemas trended towards a significant association with increased rates of post-operative UTIs and oral antibiotics trended towards a significant association with increased incidence of post-operative pneumonia. These issues warrant further research to distinguish between MBPs and OAMBPs along with greater study size values to better examine the data trends and the effect of preps and OA on C. difficile infection.

Sir William Halsted, the original proponent of bowel preparations nearly a century and a half ago, attributed infection as “manifestly from the contents of the intestine.”¹ Despite Dr. Halsted’s warning against anastomotic leak, we found that anastomotic leak and SSI rates, our primary outcomes, were not statistically significant with regards to bowel preparations. Though much of Sir Halstead’s work has become foundational material to the surgical field, and bowel preparations are currently regarded as a standard of care, the null hypothesis that outcomes were not influenced by bowel preparations was upheld and our data does not support the use of bowel preparations in elective colo-rectal surgery.

Footnotes

Acknowledgments

Special thanks to Colleen Renier, biostatistician at Essentia Health, for performing the statistical analyses, and to the Northern Rural Plains Surgical Society for allowing this study to be presented at the annual meeting and supporting this paper.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Halsted

. Circular suture of the intestine: An experimental study. Am J Med Sci. (1827-1924). 1887;(188):436-460. Accessed March 8, 2023. https://libproxy.lib.unc.edu/login?url=https://www.proquest.com/scholarly-journals/circular-suture-intestine-experimental-study-5/docview/125247179/se-2

Migaly

Bafford

Francone

, et al. The American society of colon and rectal surgeons clinical practice guidelines for the use of bowel preparation in elective colon and rectal surgery. Dis Colon Rectum. 2019;62(1):3-8. doi:10.1097/dcr.0000000000001238

The 2015 European Society of Coloproctology collaborating group . Relationship between method of anastomosis and anastomotic failure after right hemicolectomy and Ileo‐Caecal resection: An international snapshot auditt [published online ahead of print, 2017 Mar 6]. Colorectal Dis. 2017. doi:10.1111/codi.13646

Nichols

Broido

Condon

Gorbach

Nyhus

. Effect of preoperative neomycin-erythromycin intestinal preparation on the incidence of infectious complications following colon surgery. Ann Surg. 1973;178(4):453-462. doi:10.1097/00000658-197310000-00008

Clarke

Condon

Bartlett

Gorbach

Nichols

Ochi

. Preoperative oral antibiotics reduce septic complications of colon operations: Results of prospective, randomized, double-blind clinical study. Ann Surg. 1977;186:251-259. doi:10.1097/00000658-197709000-00003

Dahabreh

Steele

Shah

Trikalinos

. Oral mechanical bowel preparation for colorectal surgery: Systematic review and meta-analysis. Dis Colon Rectum. 2015;58:698-707.

Kiran

Murray

ACA

Chiuzan

Estrada

Forde

. Combined preoperative mechanical bowel preparation with oral antibiotics significantly reduces surgical site infection, anastomotic leak, and ileus after colorectal surgery. Ann Surg. 2015;262(3):416-425. doi:10.1097/SLA.0000000000001416

Klinger

Green

Monlezun

, et al. The role of bowel preparation in colorectal surgery: Results of the 2012-2015 ACS-NSQIP data. Ann Surg. 2019;269(4):671-677. doi:10.1097/SLA.0000000000002568

Matsuda

Colvin

Adachi

. Combination of oral antibiotics and mechanical bowel preparation for colorectal surgery. Ann Gastroenterol Surg. 2018;2(2):162-162. doi:10.1002/ags3.12064

10.

Morris

Graham

Chu

Cannon

Hawn

. Oral antibiotic bowel preparation significantly reduces surgical site infection rates and readmission rates in elective colorectal surgery. Ann Surg. 2015;261(6):1034-1040. doi:10.1097/SLA.0000000000001125

11.

Chen

Song

Chen

Lin

Zhang

. Comparing mechanical bowel preparation with both oral and systemic antibiotics versus mechanical bowel preparation and systemic antibiotics alone for the prevention of surgical site infection after elective colorectal surgery: A meta-analysis of randomized controlled clinical trials. Dis Colon Rectum. 2016;59(1):70-78. doi:10.1097/DCR.0000000000000524

12.

Hughes

ESR

. Asepsis in large-bowel surgery. Ann R Coll Surg Engl. 1972;51:347-356. Accessed March 11, 2023:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2388132/

13.

Güenaga

Matos

Wille-Jørgensen

. Mechanical bowel preparation for elective colorectal surgery. Cochrane Database Syst Rev. 2011;2011(9):CD001544. Published 2011 Sep 7. doi:10.1002/14651858.CD001544.pub4

14.

Slim

Vicaut

Panis

Chipponi

. Meta-analysis of randomized clinical trials of colorectal surgery with or without mechanical bowel preparation. Br J Surg. 2004;91(9):1125-1130. doi:10.1002/bjs.4651

15.

Leenen

Hentzen

Ockhuijsen

. Effectiveness of mechanical bowel preparation versus no preparation on anastomotic leakage in colorectal surgery: a systematic review and meta-analysis. Updates Surg. 2019;71(2):227-236. doi:10.1007/s13304-018-0526-4

16.

Rollins

Javanmard-Emamghissi

Lobo

. Impact of mechanical bowel preparation in elective colorectal surgery: A meta-analysis. World J Gastroenterol. 2018;24(4):519-536. doi:10.3748/wjg.v24.i4.519

17.

Zwart

Van Ginkel

Hulsker

CCC

Witvliet

Van Herwaarden-Lindeboom

MYA

. Does mechanical bowel preparation reduce the risk of developing infectious complications in pediatric colorectal surgery? A systematic review and meta-analysis. J Pediatr. 2018;203:288-293. doi:10.1016/j.jpeds.2018.07.057

18.

Haskins

Fleshman

Amdur

Agarwal

. The impact of bowel preparation on the severity of anastomotic leak in colon cancer patients. J Surg Oncol. 2016;114:810-813. doi:10.1002/jso.24426

19.

Pittet

Nocito

Balke

, et al. Rectal enema is an alternative to full mechanical bowel preparation for primary rectal cancer surgery. Colorectal Dis. 2015;17:1007-1010. doi:10.1111/codi.12974

20.

Koskenvuo

Lehtonen

Koskensalo

, et al. Mechanical and oral antibiotic bowel preparation versus no bowel preparation for elective colectomy (MOBILE): a multicentre, randomised, parallel, single-blinded trial. Lancet. 2019;394(10201):840-848. doi:10.1016/S0140-6736(19)31269-3

21.

United States Census Bureau . Minnesota’s Population at 5,706,494 in 2020, Up 7.6% Since 2010. 2021; https://www.census.gov/library/stories/state-by-state/minnesota-population-change-between-census-decade.html#:∼:text=Race_and_ethnicity_(White_alone,or_More_Races_10.2%25)

22.

Bret

EBF

. Essentia Health Community Cancer Research Program – NCORP. National Cancer Institute: Division of Cancer Prevention; 2023. https://prevention.cancer.gov/funding-and-grants/funded-grants/UG1CA189812