Abstract
Background
Intraoperative pelvic drains are often placed during low anterior resection (LAR) to evacuate postoperative fluid collections and identify/control potential anastomotic leaks. Our aim was to assess the validity of this practice.
Methods
Patients from the US Rectal Cancer Consortium (2007–2017) who underwent curative-intent LAR for a primary rectal cancer were included. Patients were categorized as receiving a closed suction drain intraoperatively or not. Primary outcomes were superficial surgical site infection (SSI), deep SSI, intraabdominal abscess, anastomotic leak, and need for secondary drain placement. Three subgroup analyses were conducted in patients who received neoadjuvant chemoradiation, had a diverting loop ileostomy (DLI), and had low anastomoses < 6 cm from the anal verge.
Results
Of 996 patients 67% (n = 551) received a drain. Drain patients were more likely to be male (64 vs 54%), have a smoking history (25 vs 19%), have received neoadjuvant chemoradiation (73 vs 61%), have low tumors (56 vs 36%), and have received a DLI (80 vs 71%) (all p < 0.05). Drains were associated with an increased anastomotic leak rate (14 vs 8%, p = 0.041), although there was no difference in the need for a secondary drainage procedure to control the leak (82 vs 88%, p = 0.924). These findings persisted in all subset analyses. Drains were not associated with increased superficial SSI, deep SSI, or intraabdominal abscess in the entire cohort or each subset analysis. Reoperation (12 vs 10%, p = 0.478) and readmission rates (28 vs 31%, p = 0.511) were similar.
Conclusions
Although not associated with increased infectious complications, intraoperatively placed pelvic drains after low anterior resection for rectal cancer are associated with an increase in anastomotic leak rate and no reduction in the need for secondary drain placement or reoperation. Routine drainage appears to be unnecessary.
Keywords: Rectal cancer, Low anterior resection, Closed suction drain
Introduction
Anastomotic leak is a serious complication following low anterior resection (LAR) for rectal cancer. Reported incidence ranges from 3 to 25% of patients, with subclinical leaks occurring in up to 50%.1–6 Resulting morbidity is considerable, including delayed wound healing, prolonged hospital stay, and decreased quality of life; in addition, anastomotic leak is associated with increased postoperative mortality.2, 6, 7 Consensus regarding risk factors and the ability to predict leakage remain elusive, and use of methods to prevent or reduce the impact of an anastomotic leak, including intraoperative pelvic drain placement, vary by institution and individual physician practice.2,6, 8
Proponents of prophylactic pelvic drain placement routinely use drains in order to evacuate postoperative fluid collections, potentially preventing anastomotic leakage due to infected hematoma or seroma eroding anastomotic integrity.2, 6 Additionally, drains are thought to provide early identification of leaks when they occur, and to mitigate the consequences, including obviating the need for a secondary drain placement.2, 6 Most secondary drains require transgluteal placement which are painful and more limiting for patients. However, randomized controlled trials and meta-analyses have failed to consistently show a benefit to routine drain placement.9–11 Urbach and colleagues conducted a metaanalysis of four randomized controlled trials and found no difference in anastomotic leak rates, both clinical and radiological, wound infections, or overall mortality.11 Two additional large meta-analyses published similar results.9, 10 In contrast, there is some evidence that drains, rather than providing benefit, may cause harm. In a prospective study of 978 patients undergoing LAR for rectal cancer, Yeh and colleagues found that intraoperative drain placement was an independent predictor of clinical anastomotic leak, with the caveat that patients were not randomized; thus, these results are subject to selection bias and there is no long-term outcome or other complication data available.12
Given these conflicting results and the recent formation of the United States Rectal Cancer Consortium, our aim was to determine the association of intraoperative closed suction drain placement with postoperative infectious complications, including superficial and deep surgical site infection and intraabdominal abscess, anastomotic leak, and need for secondary drain placement using a large, multi-institutional database from the United States Rectal Cancer Consortium.
Materials and Methods
The United States Rectal Cancer Consortium (USRCC) represents a collaboration of six tertiary and quaternary referral institutions: Emory University, University of Michigan, University of Pittsburgh Medical Center, The Ohio State University, Vanderbilt University Medical Center, and Washington University School of Medicine in St Louis. Institutional Review Board (IRB) approval was obtained at each individual institution prior to data collection. The USRCC prospectively determined a series of objectives and designed a standardized data form along with a data dictionary. This study was a result of one of these objectives. A master database was created with variables of interest and distributed to all sites. Patients who underwent surgical intervention for rectal cancer of any histology between 2007 and 2017 were evaluated, and electronic medical records were reviewed thoroughly. Pertinent baseline intraoperative, pathologic, and postoperative outcome data were collected. Staging was based on the American Committee on Cancer (AJCC) 8th edition guidelines. Data regarding neoadjuvant and adjuvant therapy, disease recurrence, and survival were also recorded. Data was de-identified and returned to the lead institution, then combined to form the final master dataset.
Patients who underwent curative intent LAR for a primary rectal cancer were included. Patients with recurrent rectal cancer, those who underwent abdominoperineal resection or any operation with palliative intent, those with metastatic disease, and those without data regarding intraoperative closed suction drain use were excluded. Patients were categorized as receiving a closed suction intraoperative drain (drain) or not (no drain). Primary outcomes were superficial surgical site infection (SSI), deep SSI, intra-abdominal abscess, anastomotic leak, and need for secondary, postoperative drain placement. Presence of an anastomotic leak was defined as documentation of a clinical leak in the medical record with method of diagnosis left to the discretion of the individual institution. Subclinical anastomotic leaks were not included in our analysis. Three subgroup analyses were conducted in patients who received neoadjuvant chemoradiation, had a diverting loop ileostomy (DLI), and had low anastomoses < 6 cm from the anal verge.
Statistical Analysis
Statistical analysis was conducted using SPSS 25.0 software (IBM Inc., Armonk, NY). Descriptive analyses were performed for the entire cohort. Chi-squared analysis was used to compare categorical variables, and Student’s t test or oneway ANOVA was used for continuous variables, where indicated. The univariate and multivariable associations between each covariate including study cohorts and study outcomes (superficial SSI, deep SSI, intra-abdominal abscess, anastomotic leak, secondary drain placement) were assessed using binary logistic regression. Statistical significance was predefined as p < 0.05.
Results
Patient Characteristics
Of 1881 patients in the database, 996 met inclusion criteria. Sixty-seven percent (n = 551) received an intraoperative drain. Seventy-one percent (n = 708) received neoadjuvant chemoradiation, 78% (n = 772) had a diverting loop ileostomy, and 47% (n = 168) had low anastomoses, < 6 cm from the anal verge. Five percent (n = 45) of patients had a superficial SSI, 3% (n = 30) a deep SSI, and 8% (n = 76) an intra-abdominal infection. Eleven percent (n = 57) had a postoperative anastomotic leak and 8% (n = 73) received a secondary drain (Table 1).
Table 1.
Baseline demographic and clinicopathologic data
| Variable | All patients, n = 996 N (%) |
Drain, n = 551 N (%) |
No drain, n = 445 N (%) |
p value |
|---|---|---|---|---|
| Age | 58.2 ± 12.1 | 58.8 ± 11.7 | 58.0 ± 12.9 | 0.364 |
| Gender | ||||
| Male | 503 (60.7) | 352 (64.0) | 151 (54.3) | 0.009 |
| Female | 325 (39.3) | 198 (36.0) | 127 (45.7) | |
| Missing | 168 | |||
| BMI | 28.8 ±6.6 | 29.1 ± 6.6 | 28.7 ± 7.0 | 0.450 |
| Diabetes | 128 (15.4) | 87 (15.8) | 41 (14.7) | 0.139 |
| Missing | 167 | |||
| Smoking history | 190 (22.9) | 136 (24.7) | 54 (19.4) | 0.025 |
| Missing | 167 | |||
| Neoadjuvant chemoXRT | 569 (68.8) | 399 (72.7) | 170 (61.2) | 0.001 |
| Missing | 169 | |||
| Diverting loop ileostomy | 631 (76.8) | 435 (79.5) | 196 (71.3) | 0.011 |
| Missing | 174 | |||
| Tumor distance from anal verge | ||||
| 0–6 cm | 163 (47.5) | 109 (55.9) | 54 (36.5) | 0.001 |
| 7–11 cm | 109 (31.8) | 57 (29.2) | 52 (35.1) | |
| 12–15 cm | 71 (20.7) | 29 (14.9) | 42 (28.4) | |
| Missing | 653 | |||
| Operative approach | ||||
| Open | 246 (29.7) | 170 (30.9) | 76 (27.4) | 0.008 |
| MIS | 368 (44.4) | 257 (46.6) | 111 (40.1) | |
| Hand assist | 214 (25.8) | 124 (22.5) | 90 (32.5) | |
| Missing | 168 | |||
| Anastomosis type | ||||
| Stapled | 716 (91.6) | 480 (91.4) | 236 (91.8) | 0.958 |
| Hand-sewn | 66 (8.4) | 45 (8.6) | 21 (8.2) | |
| Missing | 214 | |||
| Stage | ||||
| T0N0 | 142 (17.8) | 96 (18.1) | 46 (17.2) | 0.841 |
| 1 | 230 (28.9) | 150 (28.3) | 80 (30.0) | |
| 2 | 176 (22.1) | 114 (21.5) | 62 (23.2) | |
| 3 | 249 (31.2) | 170 (32.1) | 79 (29.6) | |
| Missing | 199 | |||
| Tumor size | 3.05 ± 2.1 | 3.00 ± 2.1 | 2.82 ± 2.2 | 0.281 |
| Intraoperative drain placement | 551 (66.8) | |||
| Superficial SSI | 38 (5.0) | 30 (5.9) | 8 (3.2) | 0.163 |
| Missing | 238 | |||
| Deep SSI | 19 (2.5) | 17 (3.3) | 2 (0.8) | 0.066 |
| Missing | 240 | |||
| Intra-abdominal abscess | 61 (8.5) | 42 (8.5) | 19 (8.4) | 1.000 |
| Missing | 275 | |||
| Secondary drainage procedure | 71 (9.7) | 43 (8.5) | 28 (12.3) | 0.144 |
| Missing | 263 | |||
| Anastomotic leak | 56 (11.6) | 39 (14.4) | 17 (8.0) | 0.041 |
| Missing | 514 | |||
| Intervention for leak | 46 (83.6) | 32 (82.1) | 14 (87.5) | 0.924 |
| Missing | 941 | |||
| Reoperation | 85 (11.5) | 62 (12.1) | 23 (10.0) | 0.478 |
| Missing | 254 | |||
| Readmission | 224 (29.2) | 148 (28.4) | 76 (31.0) | 0.511 |
| Missing | 230 |
Italics denote statistical significance
Average age was 58 years and 61% (n = 608) were male. Patients who received drains were more likely to be male (64% vs 54%, p = 0.009), to have a smoking history (25% vs 19%, p = 0.007), to have received neoadjuvant chemoradiation (73% vs 61%, p = 0.001), to have a DLI (80% vs 71%, p = 0.011), and to have low anastomoses (56% vs 37%, p = 0.001). Patients who received drains were more likely to be diagnosed with an anastomotic leak (14% vs 8%, p = 0.041); however, there was no difference in the rate of secondary drainage procedures between the drain and no drain groups (9% vs 12%, p = 0.144). Reoperation (12% vs 10%, p = 0.478) and readmission rates (28% vs 31%, p = 0.511) were also similar between the two groups (Table 1).
Intraoperative drain placement was not associated with superficial SSI (OR 1.875, 95% CI 0.847–4.153, p = 0.121), intra-abdominal abscess (OR 1.017, 95% CI 0.577–1.792, p = 0.953), or secondary drain placement on univariate analysis (OR 0.665, 95% CI 0.402–1.101, p = 0.112) (Table 2). Intra-operative drain placement was associated with increased odds of an anastomotic leak (OR 1.937, 95% CI 1.062–3.531, p = 0.031) and trended toward association with odds of a deep SSI (OR 4.233, 95% CI 0.970–18.467, p = 0.055) on univariate analysis; however, drain placement was not associated with either outcome on multivariable analysis (anastomotic leak: OR 1.722, 95% CI 0.910–3.256, p = 0.095; deep SSI: OR 10.292, 95% CI 0.975–108.680, p = 0.053) (Table 2). Factors independently associated with increased odds of superficial SSI were higher BMI and increased operative time. Higher BMI was also associated with increased odds of deep SSI, while minimally invasive operative technique was associated with decreased odds of deep SSI compared to open technique. Male gender, history of smoking, diabetes, and receipt of total neoadjuvant therapy were independently associated with increased odds of anastomotic leak, and anastomotic leak was the only factor independently associated with both intra-abdominal abscess and need for secondary drainage procedure (Table 2).
Table 2.
Univariate and multivariable association with superficial surgical site infection, deep surgical site infection, intra-abdominal abscess, anastomotic leak, and secondary drainage procedure for all patients
| Univariate analysis | Multivariable analysis | |||
|---|---|---|---|---|
| OR (95% CI) | p value | OR (95% CI) | p value | |
| Superficial surgical site infection | ||||
| Age | 1.012 (0.987–1.037) | 0.362 | ||
| Gender | ||||
| Female | Reference | 0.781 | ||
| Male | 1.093 (0.585–2.044) | |||
| BMI | 1.096 (1.055–1.138) | < 0.001 | 1.088 (1.046–1.133) | < 0.001 |
| Diabetes | 1.446 (0.680–3.076) | 0.338 | ||
| Smoking history | 0.628 (0.288–1.369) | 0.242 | ||
| Neoadjuvant chemoradiation | 0.931 (0.480–1.804) | 0.831 | ||
| Total neoadjuvant therapy | 0.935 (0.474–1.846) | 0.847 | ||
| Operative approach | ||||
| Open | Reference | 0.304 | ||
| Minimally invasive | 0.706 (0.364–1.370) | 0.206 | ||
| Hand assisted Operative time |
0.579 (0.249–1.350) | 0.206 0.002 |
||
| Operative time | 1.004 (1.001–1.006) | 0.002 | 1.003 (1.000–1.006) | 0.024 |
| Intraoperative drain | 1.875 (0.847–4.153) | 0.121 | ||
| Deep surgical site infection | ||||
| Age | 1.008 (0.978–1.039) | 0.611 | ||
| Gender | ||||
| Female | Reference | 0.780 | ||
| Male BMI |
0.899 (0.428–1.891) 1.059 (1.016–1.104) |
0.007 | 1.088 (1.032–1.147) | 0.002 |
| Diabetes | 1.441 (0.578–3.596) | 0.433 | ||
| Smoking history | 1.753 (0.821–3.742) | 0.147 | ||
| Neoadjuvant chemoradiation | 5.100 (1.201–21.650) | 0.027 | 5.695 (0.734–44.197) | 0.096 |
| Total neoadjuvant therapy | 1.792 (0.843–3.809) | 0.129 | ||
| Operative approach | ||||
| Open | Reference | < 0.001 | Reference | 0.045 |
| Minimally invasive | 0.153 (0.58–0.408) | 0.004 | 0.289 (0.086–0.974) | 0.145 |
| Hand assisted | 0.114 (0.027–0.490) | 0.313 (0.065–1.495) | ||
| Operative time | 1.000 (0.996–1.003) | 0.821 | ||
| Intraoperative drain | 4.233 (0.970–18.467) | 0.055 | 10.292 (0.975–108.680) | 0.053 |
| Intra-abdominal abscess | ||||
| Age | 0.992 (0.973–1.012) | 0.425 | ||
| Gender | ||||
| Female | Reference | 0.192 | ||
| Male | 1.404 (0.843–2.339) | |||
| BMI | 0.978 (0.940–1.017) | 0.263 | ||
| Diabetes | 1.278 (0.693–2.358) | 0.432 | ||
| Smoking history | 1.527 (0.921–2.532) | 0.101 | ||
| Neoadjuvant chemoradiation | 1.356 (0.763–2.411) | 0.299 | ||
| Total neoadjuvant therapy | 1.306 (0.788–2.163) | 0.301 | ||
| Operative approach | ||||
| Open | Reference | 0.965 | ||
| Minimally invasive | 0.988 (0.583–1.675) | 0.365 | ||
| Hand assisted | 0.733 (0.374–1.435) | |||
| Operative time | 1.001 (0.999–1.004) | 0.179 | ||
| Intraoperative drain | 1.017 (0.577–1.792) | 0.953 | ||
| Diverting loop ileostomy | 0.782 (0.452–1.351) | 0.378 | ||
| Anastomotic leak | 19.092 (9.719–37.505) | < 0.001 | ||
| Age | 1.002 (0.980–1.024) | 0.876 | ||
| Gender | ||||
| Female | Reference | 0.001 | Reference | 0.004 |
| Male | 3.157 (1.593–6.254) | 2.582 (1.402–5.803) | ||
| BMI | 1.000 (0.961–1.041) | 0.994 | ||
| Diabetes | 2.842 (1.531–5.277) | 0.001 | 3.076 (1.559–6.067) | 0.001 |
| Smoking history | 2.917 (1.643–5.180) | < 0.001 | 3.100 (1.666–5.768) | < 0.001 |
| Neoadjuvant chemoradiation | 1.410 (0.767–2.593) | 0.269 | ||
| Total neoadjuvant therapy | 1.998 (1.095–3.643) | 0.024 | 2.087 (1.095–3.978) | 0.025 |
| Operative approach | Reference | 0.667 | ||
| Open | 0.874 (0.474–1.612) | 0.120 | ||
| Minimally invasive | 0.539 (0.247–1.176) | |||
| Hand assisted | ||||
| Anastomosis distance from anal verge | Reference | 0.842 | ||
| 0–6 cm | 1.095 (0.449–2.671) | 0.711 | ||
| 7–11 cm | 1.201 (0.456–3.158) | |||
| 12–15 cm | ||||
| Anastomosis type | Reference | 0.241 | ||
| Stapled | 0.486 (0.146–1.621) | |||
| Hand sewn | ||||
| Intraoperative drain | 1.937 (1.062–3.531) | 0.031 | 1.722(0.910–3.256) | 0.095 |
| Diverting loop ileostomy | 0.679 (0.382–1.209) | 0.188 | ||
| Secondary drainage procedure | ||||
| Age | 0.994 (0.974–1.014) | 0.530 | ||
| Gender | ||||
| Female | Reference | 0.071 | ||
| Male | 1.637 (0.959–2.792) | |||
| BMI | 1.010 (0.975–1.047) | 0.583 | ||
| Diabetes | 1.586 (0.881–2.856) | 0.124 | ||
| Smoking history | 1.321 (0.781–2.235) | 0.299 | ||
| Operative approach | ||||
| Open | Reference | 0.769 | ||
| Minimally invasive | 0.920 (0.528–1.602) | 0.769 | ||
| Hand assisted | 1.035 (0.528–1.602) | |||
| Operative time | 1.003 (1.001–1.005) | 0.001 | 1.001 (0.999–1.004) | 0.236 |
| Intraoperative drain | 0.665 (0.402–1.101) | 0.112 | ||
| Diverting loop ileostomy | 1.013 (0.559–1.835) | 0.996 | ||
| Anastomotic leak | 14.227 (7.513–26.942) | < 0.001 | 14.875 (7.712–28.691) | < 0.001 |
Italics denote statistical significance
Patients Who Received Neoadjuvant Chemoradiation
Of 996 patients in our study cohort, 708 received neoadjuvant chemoradiation. Average age was 58 years and 63% (n = 445) were male. Of patients for whom intraoperative drain placement data was available, 76% (n = 399) received a drain. Five percent (n = 32) of patients had a superficial SSI, 4% (n = 26) a deep SSI, and 9% (n = 58) had an intra-abdominal infection. Twelve percent (n = 41) had an anastomotic leak and 9% (n = 57) had a secondary drainage procedure (Table 3).
Table 3.
Baseline demographic and clinicopathologic data for patients who received neoadjuvant chemoradiation, had a diverting loop ileostomy, and with low anastomoses
| Neoadjuvant chemoradiation | Diverting loop ileostomy | Low anastomosis | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All | Drain | No drain | p value | All | Drain | No drain | p value | All | Drain | No drain | p value | |
| Age | 57.7 ±11.7 | 58.0 ± 11.1 | 57.7 ± 12.8 | 0.848 | 57.5 ± 11.7 | 58.3 ±11.8 | 56.7 ± 12.1 | 0.109 | 57.9 ± 12.2 | 57.9 ± 12.4 | 57.4 ±11.9 | 0.779 |
| Gender | ||||||||||||
| Male | 445 (62.9) | 259 (65.1) | 94 (55.3) | 0.035 | 484 (62.8) | 280 (64.5) | 111 (56.6) | 0.072 | 106 (63.1) | 73 (67.6) | 28 (51.9) | 0.076 |
| Female | 262 (37.1) | 139 (34.9) | 76 (44.7) | 287 (37.2) | 154 (35.5) | 85 (43.4) | 61 (36.3) | 35 (32.4) | 26 (48.1) | |||
| BMI | 28.7 ± 6.6 | 28.9 ±6.4 | 28.5 ± 7.5 | 0.474 | 28.8 ± 6.5 | 29.0 ± 6.3 | 28.6 ±7.1 | 0.493 | 28.3 ±5.9 | 28.5 ±6.1 | 28.0 ± 5.6 | 0.626 |
| Diabetes | 99 (14.0) | 63 (15.8) | 21 (12.4) | 0.014 | 101 (13.1) | 63 (14.5) | 24 (12.2) | 0.146 | 22 (13.1) | 17 (15.6) | 5 (9.3) | 0.408 |
| Smoking history | 185 (26.1) | 106 (26.6) | 41 (24.1) | 0.191 | 199 (25.8) | 114 (26.2) | 44 (22.4) | 0.039 | 35 (20.8) | 23 (21.1) | 9 (16.7) | 0.611 |
| Neoadjuvant chemoXRT | 627 (81.2) | 354 (81.4) | 150 (76.5) | 0.194 | 139 (82.7) | 97 (89.8) | 37 (68.5) | 0.002 | ||||
| Diverting loop ileostomy | 627 (89.3) | 354 (89.4) | 150 (89.3) | 1.000 | 148 (88.1) | 97 (89.0) | 46 (85.2) | 0.657 | ||||
| Distance of tumor from anal verge | 139 (56.3) | 97 (65.1) | 37 (43.0) | 0.004 | 148 (60.2) | 97 (68.8) | 46 (49.5) | 0.012 | ||||
| 0–6 cm | 85 (34.4) | 42 (28.2) | 37 (43.0) | 73 (29.7) | 33 (23.4) | 36 (38.7) | ||||||
| 7–11 cm | 23 (9.3) | 10 (6.7) | 12 (14.0) | 25 (3.2) | 11 (7.8) | 11 (11.8) | ||||||
| 12–15 cm | 461 | 526 | ||||||||||
| Missing | ||||||||||||
| Operative approach | ||||||||||||
| Open | 251 (35.6) | 128 (32.1) | 47 (27.6) | 0.202 | 234 (30.4) | 115 (26.4) | 52 (26.7) | 0.269 | 60 (35.9) | 33 (30.3) | 23 (43.4) | 0.014 |
| Minimally invasive | 299 (42.4) | 181 (45.4) | 73 (42.9) | 347 (45.1) | 213 (49.0) | 84 (43.1) | 66 (39.5) | 53 (48.6) | 13 (24.5) | |||
| Hand assisted | 156 (22.1) | 90 (22.6) | 50 (29.4) | 189 (24.5) | 107 (24.6) | 59 (30.3) | 41 (24.6) | 23 (21.1) | 17 (32.1) | |||
| Anastomosis type | ||||||||||||
| Stapled | 602 (90.8) | 341 (89.5) | 142 (92.8) | 0.311 | 679 (90.1) | 386 (89.8) | 168 (89.8) | 1.000 | 121 (75.6) | 83 (78.3) | 35 (71.4) | 0.465 |
| Hand-sewn | 61 (9.2) | 40 (10.5) | 11 (7.2) | 75 (9.9) | 44 (10.2) | 19 (10.9) | 39 (24.4) | 23 (21.7) | 14 (28.6) | |||
| Stage | ||||||||||||
| T0N0 | 117 (17.2) | 68 (17.7) | 29 (17.8) | 0.730 | 144 (18.7) | 81 (19.3) | 40 (21.2) | 0.553 | 30 (18.8) | 18 (17.5) | 11 (21.2) | 0.478 |
| 1 | 192 (28.2) | 109 (28.4) | 53 (32.5) | 223 (28.9) | 125 (29.8) | 64 (33.9) | 53 (33.1) | 37 (35.9) | 15 (28.8) | |||
| 2 | 145 (21.3) | 86 (22.4) | 31 (19.0) | 145 (19.5) | 82 (19.5) | 35 (18.5) | 25 (15.6) | 18 (17.5) | 6 (11.5) | |||
| 3 | 226 (33.2) | 121 (31.5) | 50 (30.7) | 2333 (31.3) | 132 (31.4) | 50 (26.5) | 52 (32.5) | 30 (29.1) | 20 (38.5) | |||
| Tumor size | 2.83 ±1.9 | 2.75 ± 1.7 | 2.26 ± 1.9 | 0.008 | 2.82 ±2.1 | 2.8 ±2.1 | 2.3 ±2.1 | 0.015 | 2.55 ±1.9 | 2.67 ± 2.0 | 2.4 ± 1.8 | 0.414 |
| Superficial SSI | 32 (4.9) | 20 (5.4) | 6 (3.9) | 0.638 | 37 (5.3) | 24 (6.0) | 7 (4.0) | 0.426 | 16 (9.5) | 14 (13.6) | 1 (2.1) | 0.060 |
| Deep SSI | 26 (4.0) | 14 (3.8) | 2 (1.3) | 0.227 | 22 (3.1) | 13 (3.3) | 1 (0.6) | 0.101 | 8 (4.8) | 6 (5.9) | 1 (2.1) | 0.555 |
| Intraabdominal abscess | 58 (9.2) | 32 (8.8) | 13 (9.2) | 1.000 | 57 (8.4) | 30 (7.8) | 14 (8.6) | 0.859 | 17 (10.1) | 12 (11.8) | 3 (6.4) | 0.471 |
| Secondary drainage procedure | 57 (9.1) | 36 (9.8) | 20 (14.1) | 0.213 | 57 (8.4) | 34 (8.5) | 22 (13.5) | 0.105 | 19 (11.3) | 13 (13.0) | 5 (10.6) | 0.891 |
| Anastomotic leak | 41 (12.4) | 31 (16.8) | 10 (8.1) | 0.044 | 36 (4.7) | 26 (13.8) | 10 (6.9) | 0.065 | 13 (7.7) | 11 (11.3) | 2 (4.3) | 0.280 |
| Intervention for leak | 34 (85.0) | 25 (80.6) | 9(100) | 0.376 | 29 (82.9) | 21 (80.8) | 8 (88.9) | 0.965 | 11 (6.5) | 9 (81.9) | 2 (100) | 1.000 |
| Reoperation | 66 (9.3) | 17 (12.0) | 43 (11.6) | 1.000 | 63 (9.1) | 40 (9.9) | 16 (9.7) | 1.000 | 24 (14.3) | 18 (18.0) | 6 (13.0) | 0.610 |
| Readmission | 192 (30.1) | 54 (36.0) | 106 (28.2) | 0.098 | 207 (29.7) | 117 (28.5) | 58 (33.3) | 0.290 | 54 (32.1) | 36 (35.6) | 16 (34.0) | 0.996 |
Italics denote statistical significance
Patients who received drains were more likely to be male (65% vs 55%, p = 0.035), to have a history of diabetes (16% vs 12%, p = 0.014), to have larger tumors (2.8 vs 2.3 cm, p = 0.008), and to have low anastomoses < 6 cm from the anal verge (65% vs 43%, p = 0.004) (Table 3). Intraoperative drain placement was not associated with superficial SSI (OR 1.388, 95% CI 0.546–3.527, p = 0.491), deep SSI (OR 2.953, 95% CI 0.663–13.150, p = 0.155), intra-abdominal abscess (OR 0.956, 95% CI 0.487–1.880, p = 0.897), or secondary drainage procedure (OR 0.659, 95%CI 0.368–1.183, p = 0.163) on univariate analysis. Drain placement was associated with anastomotic leak on univariate (OR 2.167, 95% CI 1.009–4.653, p = 0.047) but not multivariable analysis (OR 1.968, 95% CI 0.894–4.333, p = 0.093) (Table 4).
Table 4.
Association of intraoperative drain placement with superficial surgical site infection, deep surgical site infection, intra-abdominal abscess, anastomotic leak, and secondary drainage procedure
| Univariate analysis | Multivariable analysis | |||
|---|---|---|---|---|
| Cohort | OR (95% CI) | p value | OR (95% CI) | p value |
| Superficial SSI | ||||
| All patients | 1.875 (0.847–4.153) | 0.121 | ||
| Neoadjuvant chemoradiation | 1.388 (0.546–3.527) | 0.491 | ||
| Diverting loop ileostomy | 1.545 (0.653–3.656) | 0.322 | ||
| Low anastomoses (0–6 cm from anal verge) | 7.236 (0.922–56.758) | 0.060 | ||
| Deep SSI | ||||
| All patients | 4.233 (0.970–18.467) | 0.055 | ||
| Neoadjuvant chemoradiation | 2.953 (0.663–13.150) | 0.155 | ||
| Diverting loop ileostomy | 5.909 (0.767–45.526) | 0.088 | ||
| Low anastomoses (0–6 cm from anal verge) | 2.875 (0.336–24.582) | 0.335 | ||
| Intra-abdominal abscess | ||||
| All patients | 1.017 (0.577–1.792) | 0.953 | ||
| Neoadjuvant chemoradiation | 0.956 (0.487–1.880) | 0.897 | ||
| Diverting loop ileostomy | 0.888 (0.458–1.723) | 0.726 | ||
| Low anastomoses (0–6 cm from anal verge) | 1.956 (0.525–7.288) | 0.318 | ||
| Anastomotic leak | ||||
| All patients* | 1.937 (1.062–3.531) | 0.031 | 1.722 (0.910–3.256) | 0.095 |
| Neoadjuvant chemoradiation** | 2.167 (1.009–4.653) | 0.047 | 1.968 (0.894–4.333) | 0.093 |
| Diverting loop ileostomy** | 2.275 (1.071–4.830) | 0.032 | 2.003 (0.900–4.455) | 0.089 |
| Low anastomoses (0–6 cm from anal verge) | 2.878 (0.611–13.548) | 0.181 | ||
| Secondary drainage procedure | ||||
| All patients | 0.665 (0.402–1.101) | 0.112 | ||
| Neoadjuvant chemoradiation | 0.659 (0.368–1.183) | 0.163 | ||
| Diverting loop ileostomy | 0.599 (0.338–1.059) | 0.078 | ||
| Low anastomoses (0–6 cm from anal verge) | 1.255 (0.420–3.753) | 0.684 | ||
Italics denote statistical significance
Variables included in MVA: gender, diabetes, smoking history, total neoadjuvant therapy
Variables included in MVA: gender, diabetes, smoking history
Patients Who Received a Diverting Loop Ileostomy
A total of 772 patients had a DLI. Average age was 58 years and 63% (n = 484) were male. Sixty eight percent (n = 435) received a drain. Five percent (n = 37) of patients had a superficial SSI, 3% (n = 22) a deep SSI, and 8% (n = 57) an intra-abdominal abscess. Five percent (n = 36) of patients had an anastomotic leak, and 8% (n = 57) had a secondary drainage procedure (Table 3).
Patients who received drains were more likely to have a smoking history (26% vs 22%, p = 0.039), to have larger tumors (2.8 vs 2.3 cm, p = 0.015), and to have low anastomoses, < 6 cm from the anal verge (69% vs 50%, p = 0.012) (Table 3). Intraoperative drain placement was not associated with superficial SSI (OR 1.545, 95% CI 0.653–3.656, p = 0.322), deep SSI (OR 5.909, 95% CI 0.767–45.536, p = 0.088), intra-abdominal abscess (OR 0.888, 95% CI 0.458–1.723, p = 0.726), or secondary drainage procedure (OR 0.599, 95% CI 0.338–1.059, p = 0.078) on univariate analysis. Drain placement was associated with anastomotic leak on univariate (OR 2.275, 95% CI 1.071–4.830, p = 0.032) but not multivariable analysis (OR 2.003, 95% CI 0.900–4.455, p = 0.089) (Table 4).
Patients with Anastomoses < 6 cm from the Anal Verge
A total of 168 patients had low anastomoses. Average age was 58 years and 63% (n = 103) were male. Fifty eight percent (n = 97) received a drain. Ten percent (n = 16) of patients had a superficial SSI, 5% (n = 8) a deep SSI, and 10% (n = 17) had an intra-abdominal abscess. Eight percent (n = 13) of patients had an anastomotic leak, and 11% (n =19) had a secondary drainage procedure (Table 3).
Patients who received a drain were more likely to have received neoadjuvant chemoradiation (90% vs 69%, p = 0.002) and to have had a minimally invasive surgical approach (49% vs 25%, p = 0.014) (Table 3). Intraoperative drain placement was not associated with any study outcomes on univariate analysis (superficial SSI: OR 7.236, 95% CI 0.922–56.758, p = 0.060; deep SSI: OR 2.875, 95% CI 0.336–24.582, p = 0.335; intra-abdominal abscess: OR 1.956, 95% CI 0.525–7.288, p = 0.318; anastomotic leak: OR 2.878, 95% CI 0.611–13.548, p = 0.181; secondary drainage procedure: OR 1.255, 95% CI 0.420–3.752, p = 0.684) (Table 4).
Patients with an anastomotic leak
Among patients with an anastomotic leak, there were no significant differences in baseline demographic characteristics, tumor characteristics, superficial, deep, or intra-abdominal infections, secondary drainage procedures for any reason, intervention specifically for anastomotic leak, ICU admission, reoperation, readmission, or length of stay between patients who did and did not receive a drain (Table 5).
Table 5.
Demographic, clinicopathologic, complication, and outcome data for patients with an anastomotic leak with and without intraoperative drain placement
| Variable | Drain, n = 39 N (%) |
No drain, n = 17 N (%) |
p value |
|---|---|---|---|
| Age (mean ± SD) | 58.9 ±9.2 | 58.4 ± 10.7 | 0.863 |
| Gender | |||
| Male | 32 (82.1) | 13 (76.5) | 0.906 |
| Female | 7 (17.9) | 4 (23.5) | |
| BMI (mean ± SD) | 28.6 ±6.2 | 29.8 ± 6.3 | 0.483 |
| Neoadjuvant chemoXRT | 31 (79.5) | 10 (58.8) | 0.201 |
| Diverting loop ileostomy | 26 (66.7) | 10 (58.8) | 0.795 |
| Anastomosis distance from anal verge | |||
| 0–6 cm | 11 (57.9) | 2 (20.0) | 0.130 |
| 7–11 cm | 5 (26.3) | 4 (40.0) | |
| 12–15 cm | 3 (15.8) | 4 (40.0) | |
| Missing | 20 | 7 | |
| Stage | |||
| T0N0 | 7 (17.9) | 4 (23.5) | 0.733 |
| 1 | 5 (12.8) | 5 (29.4) | |
| 2 | 10 (25.6) | 3 (17.6) | |
| 3 | 14(35.9) | 4 (23.5) | |
| Tumor size (cm) (mean ± SD) | 3.3 ±1.5 | 3.1 ± 1.6 | 0.591 |
| Superficial SSI | 4 (10.3) | 1 (6.3) | 1.000 |
| Deep SSI | 9 (23.1) | 0 (0.0) | 0.089 |
| Intraabdominal abscess | 18(46.2) | 10 (71.4) | 0.189 |
| Secondary drainage procedure | 19 (48.7) | 11 (78.6) | 0.105 |
| Intervention for leak | 32 (82.1) | 14 (87.5) | 0.924 |
| Intervention type | |||
| Percutaneous drain | 10 (38.5) | 8 (72.7) | 0.237 |
| Colostomy | 3 (11.5) | 2 (18.2) | |
| Ileostomy | 8 (30.5) | 0 (0.0) | |
| Pelvic washout | 2 (7.7) | 0 (0.0) | |
| Anastomotic revision | 2 (7.7) | 1 (9.1) | |
| Completion proctectomy | 1 (3.8) | 0 (0.0) | |
| Reoperation | 27 (69.2) | 6 (49.2) | 0.154 |
| ICU admission | 6 (16.7) | 1 (6.7) | 0.618 |
| Length of stay (mean ± SD) | 10.2 ±7.0 | 7.5 ± 5.4 | 0.157 |
| Readmission | 30 (78.9) | 14 (82.4) | 1.000 |
| Death | 6 (17.6) | 3 (21.4) | 1.000 |
Discussion
Ultimately, definitive conclusions regarding risk factors for anastomotic leak remain elusive, which contributes to the debate regarding drain use. Despite the differences between patients in our cohort who did and did not receive an intraoperative drain, there were no differences in rates of postoperative infectious outcomes. On chi-squared analysis, patients who received a drain had a higher incidence of anastomotic leak compared to those who did not (14% vs 8%, p = 0.041); however, there were no differences in either intervention specifically for the leak (82% vs 88% of patients with an identified leak, p = 0.924), or secondary drainage procedures (9% vs 12% of all patients, p = 0.144) (Table 1). This raises the question as to whether the identification of these additional, presumably subclinical, leaks is helpful. Furthermore, though there were no significant differences between patients who received a drain and those who did not among only patients who experienced an anastomotic leak, there were trends toward higher leak severity in patients who received a drain, as evidenced by higher proportions of re-operations, more aggressive interventions to address the leak including ostomy creation, anastomotic revision, and pelvic washout, and higher rates of ICU admission. These data should certainly be taken in the context of relatively small sample size, with the trend toward more advanced cancers in the drain group, and with the inherent selection bias present with drain placement at the discretion of the individual surgeon.
Furthermore, if the vast majority of patients in both groups who experience a leak will undergo additional intervention to address these clinically apparent leaks, is the initial drain valuable? In their systematic review of four randomized controlled trials, Urbach and colleagues noted that in only one case of twenty anastomotic leaks did purulent or feculent material drain from the intraoperatively placed drain, further calling into question whether prophylactic drain placement will effectively control anastomotic leaks that occur.11
In our multi-institutional cohort of patients undergoing low anterior resection for rectal cancer, intraoperative, prophylactic closed suction drain placement was not associated with either increased or decreased odds of superficial or deep surgical site infection, intra-abdominal abscess, or decreased odds of receiving a secondary drainage procedure. This was consistent in the entire cohort and in three subgroup analyses: patients who received chemoradiotherapy and those with low anastomoses (0–6 cm from the anal verge), two groups considered high risk of anastomotic leak, and in patients who received a diverting loop ileostomy, which is considered protective and often performed in patients deemed high risk. The overall anastomotic leak rate was 11.2%, falling in the middle of published estimates and only slightly higher than the accepted average of approximately 10%.6 Leak rates were lowest in patients who received a diverting loop ileostomy (4.7%) and highest in those who received neoadjuvant chemoradiation (12.4%); however, rates between groups were not directly compared.
As this is a retrospective study, drains were placed at the discretion of the surgeon. This is reflected in the differences between the patients who did and did not receive a drain and currently accepted risk factors for anastomotic leak. Patients who received a drain were more likely to be male, and male gender has been shown to be independently associated with anastomotic leak, presumably due to the increased difficulty of operating in the narrower male pelvis.6, 13, 14 Smoking has also been found to be associated with anastomotic leak, due to tissue ischemia and impaired wound healing. 6, 15 Tissue damage from neoadjuvant chemoradiotherapy has also been implicated; initial data in published studies reveal conflicting results. Though a recent systematic review and metaanalysis by Hu and colleagues found no association between neoadjuvant therapy and anastomotic leak, these patients remain considered high risk by many physicians.15–19 Finally, distance of the anastomosis from the anal verge has been associated with increased risk of anastomotic leak, with more distal anastomoses associated with higher risk. However, again, data yield conflicting results.20–23
Intraoperative prophylactic closed suction drain placement was not associated with worse infectious outcomes, and thus was not associated with harm to the patient. However, it is important to note that the odds ratio point estimates for the association of drain with superficial SSI, deep SSI, intra-abdominal abscess, and anastomotic leak are all > 1 (Table 2) favoring improved outcomes for the “No Drain” group, although these analyses did not reach statistical significance. Though this is a large dataset, we are hindered by the number of events for each postoperative complication, all of which are quite small, leaving our analysis potentially underpowered to detect significant differences or associations. This has doubtless contributed to the equivocal nature of the literature on the subject of anastomotic leaks and prophylactic drainage and very large, multi-center trials are likely needed for definitive answers to these questions. The French Research Group of Rectal Cancer Surgery (GRECCAR) conducted the largest randomized clinical trial to date regarding drain placement after low anterior resection for rectal cancer in France.24 Four hundred ninety-four patients were randomized to drain and no drain groups, and the authors found no difference in postoperative rates of pelvic sepsis, surgical morbidity, reoperation, or length of hospital stay between the groups and suggested that use of intraoperative pelvic drains were not beneficial to the patient.24 Our data support these results and provide analysis of the additional endpoints of surgical site and deep space infections and anastomotic leak, as well as subset analyses in classically highrisk patient groups, and in a North American cohort.
The results for the entire cohort were mirrored in all three subset analyses: in patients who received neoadjuvant chemotherapy, those with diverting loop ileostomies, and with low anastomoses 0–6 cm from the anal verge, three groups classically considered high risk for postoperative complications, specifically anastomotic leak. However, the lack of statistical significance should be considered in the context of small numbers of events, as reflected in the large confidence intervals seen in these analyses, though in all three groups the point estimates and confidence interval ranges favor the no drain group. It should also be acknowledged that there is a large amount of overlap between the groups. For example, 89% of patients who received neoadjuvant chemotherapy also received a diverting loop ileostomy, and 56% had low anastomoses (Table 3). This further highlights the need for randomized controlled trials to determine not only the definitive role of drain placement but also to more clearly define risk factors for postoperative complications.
There are several limitations to this study. Our database, while large and multi-institutional, was retrospective and thus subject to selection bias, especially regarding which patients were selected to receive drains. As previously stated, drain placement was entirely at the discretion of the surgeon. Therefore, drains were more commonly placed in patients with risk factors traditionally associated with anastomotic leak. This likely contributed to higher leak rates and more severe leaks seen in the drained group. The institutions involved are also all high volume, tertiary or quaternary referral centers, which may limit the generalizability of the data, as practice patterns and complication rates may vary from lower volume facilities. And finally, as previously mentioned, there were low rates of study outcomes, making it difficult to derive definitive conclusions regarding the help or harm associated with prophylactic drain placement.
Conclusion
The routine use of intraoperative prophylactic closed suction drain placement did not benefit patient outcome, since it was not associated with decreased odds of anastomotic leak or decreased odds of undergoing a secondary drainage procedure, both in the entire cohort and in subset analyses of classically considered high-risk groups. Furthermore, routine drainage was not significantly associated with decreased odds of postoperative infectious complications. These data indicate that routine drainage during low anterior resection for rectal cancer may not be warranted. Large randomized controlled trials are needed to clarify risk factors for anastomotic leaks and the definitive role of prophylactic drainage in these patients.
Footnotes
Conflict of Interest The authors declare that they have no conflicts of interest.
Compliance with Ethical Standards
References
- 1.Bruce J, Krukowski ZH, Al-Khairy G, Russell EM, Park KG. Systematic review of the definition and measurement of anastomotic leak after gastrointestinal surgery. Br J Surg. 2001;88(9):1157–68. 10.1046/j.0007-1323.2001.01829.x. [DOI] [PubMed] [Google Scholar]
- 2.Cavaliere D, Popivanov G, Cassini D, Cirocchi R, Henry BM, Vettoretto N, et al. Is a drain necessary after anterior resection of the rectum? A systematic review and meta-analysis. Int J Colorectal Dis. 2019;34(6):973–81. 10.1007/s00384-019-03276-4. [DOI] [PubMed] [Google Scholar]
- 3.Goligher JC, Graham NG, De Dombal FT. Anastomotic dehiscence after anterior resection of rectum and sigmoid. Br J Surg. 1970;57(2):109–18. 10.1002/bjs.1800570208. [DOI] [PubMed] [Google Scholar]
- 4.Nikolian VC, Kamdar NS, Regenbogen SE, Morris AM, Byrn JC, Suwanabol PA, et al. Anastomotic leak after colorectal resection: A population-based study of risk factors and hospital variation. Surgery. 2017;161(6):1619–27. 10.1016/j.surg.2016.12.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Platell C, Barwood N, Dorfmann G, Makin G. The incidence of anastomotic leaks in patients undergoing colorectal surgery. Colorectal Dis. 2007;9(1):71–9. 10.1111/j.1463-1318.2006.01002.x. [DOI] [PubMed] [Google Scholar]
- 6.Taflampas P, Christodoulakis M, Tsiftsis DD. Anastomotic leakage after low anterior resection for rectal cancer: facts, obscurity, and fiction. Surg Today. 2009;39(3):183–8. 10.1007/s00595-008-3835-2. [DOI] [PubMed] [Google Scholar]
- 7.Wu SW, Ma CC, Yang Y. Role of protective stoma in low anterior resection for rectal cancer: a meta-analysis. World J Gastroenterol. 2014;20(47):18031–7. 10.3748/wjg.v20.i47.18031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Matsuda K, Yokoyama S, Hotta T, Watanabe T, Tamura K, Iwamoto H, et al. Pelvic Drain After Laparoscopic Low Anterior Resection for Rectal Cancer in Patients With Diverting Stoma. Surg Laparosc Endosc Percutan Tech. 2018;28(2):82–5. 10.1097/SLE.0000000000000517. [DOI] [PubMed] [Google Scholar]
- 9.Karliczek A, Jesus EC, Matos D, Castro AA, Atallah AN, Wiggers T. Drainage or nondrainage in elective colorectal anastomosis: a systematic review and meta-analysis. Colorectal Dis. 2006;8(4): 259–65. 10.1111/j.1463-1318.2006.00999.x. [DOI] [PubMed] [Google Scholar]
- 10.Petrowsky H, Demartines N, Rousson V, Clavien PA. Evidence-based value of prophylactic drainage in gastrointestinal surgery: a systematic review and meta-analyses. Ann Surg. 2004;240(6): 1074–84; discussion 84–5. 10.1097/01.sla.0000146149.17411.c5.. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Urbach DR, Kennedy ED, Cohen MM. Colon and rectal anastomoses do not require routine drainage: a systematic review and meta-analysis. Ann Surg. 1999;229(2):174–80. 10.1097/00000658-199902000-00003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yeh CY, Changchien CR, Wang JY, Chen JS, Chen HH, Chiang JM, et al. Pelvic drainage and other risk factors for leakage after elective anterior resection in rectal cancer patients: a prospective study of 978 patients. Ann Surg. 2005;241(1):9–13. 10.1097/01.sla.0000150067.99651.6a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Trencheva K, Morrissey KP, Wells M, Mancuso CA, Lee SW, Sonoda T, et al. Identifying important predictors for anastomotic leak after colon and rectal resection: prospective study on 616 patients. Ann Surg. 2013;257(1):108–13. 10.1097/SLA.0b013e318262a6cd. [DOI] [PubMed] [Google Scholar]
- 14.Vasiliu EC, Zarnescu NO, Costea R, Neagu S. Review of Risk Factors for Anastomotic Leakage in Colorectal Surgery. Chirurgia (Bucur). 2015;110(4):319–26. [PubMed] [Google Scholar]
- 15.Sorensen LT, Jorgensen T, Kirkeby LT, Skovdal J, Vennits B, Wille-Jorgensen P. Smoking and alcohol abuse are major risk factors for anastomotic leakage in colorectal surgery. Br J Surg. 1999;86(7):927–31. 10.1046/j.1365-2168.1999.01165.x.. [DOI] [PubMed] [Google Scholar]
- 16.Randomised trial of surgery alone versus surgery followed by radiotherapy for mobile cancer of the rectum. Medical Research Council Rectal Cancer Working Party. Lancet. 1996;348(9042): 1610–4. [PubMed] [Google Scholar]
- 17.Hu MH, Huang RK, Zhao RS, Yang KL, Wang H. Does neoadjuvant therapy increase the incidence of anastomotic leakage after anterior resection for mid and low rectal cancer? A systematic review and meta-analysis. Colorectal Dis. 2017;19(1):16–26. 10.1111/codi.13424. [DOI] [PubMed] [Google Scholar]
- 18.Lele S, Radstone D, Eremin J, Kendal R, Hosie KB. Prospective audit following the introduction of short-course preoperative radiotherapy for rectal cancer. Br J Surg. 2000;87(1):97–9. 10.1046/j.1365-2168.2000.01301.x. [DOI] [PubMed] [Google Scholar]
- 19.Marijnen CA, Kapiteijn E, van de Velde CJ, Martijn H, Steup WH, Wiggers T, et al. Acute side effects and complications after shortterm preoperative radiotherapy combined with total mesorectal excision in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol. 2002;20(3):817–25. 10.1200/JCO.2002.20.3.817. [DOI] [PubMed] [Google Scholar]
- 20.Hayden DM, Mora Pinzon MC, Francescatti AB, Saclarides TJ. Patient factors may predict anastomotic complications after rectal cancer surgery: Anastomotic complications in rectal cancer. Ann Med Surg (Lond). 2015. ;4(1): 11–6. 10.1016/j.amsu.2014.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Karanjia ND, Corder AP, Bearn P, Heald RJ. Leakage from stapled low anastomosis after total mesorectal excision for carcinoma of the rectum. Br J Surg. 1994;81 (8):1224–6. 10.1002/bjs.1800810850. [DOI] [PubMed] [Google Scholar]
- 22.Makela JT, Kiviniemi H, Laitinen S. Risk factors for anastomotic leakage after left-sided colorectal resection with rectal anastomosis. Dis Colon Rectum. 2003;46(5):653–60. 10.1007/s10350-004-6627-9. [DOI] [PubMed] [Google Scholar]
- 23.Vignali A, Fazio VW, Lavery IC, Milsom JW, Church JM, Hull TL, et al. Factors associated with the occurrence of leaks in stapled rectal anastomoses: a review of 1,014 patients. J Am Coll Surg. 1997;185(2):105–13. 10.1016/s1072-7515(97)00018-5. [DOI] [PubMed] [Google Scholar]
- 24.Denost Q, Rouanet P, Faucheron JL, Panis Y, Meunier B, Cotte E, et al. To Drain or Not to Drain Infraperitoneal Anastomosis After Rectal Excision for Cancer: The GRECCAR 5 Randomized Trial. Ann Surg. 2017;265(3):474–80. 10.1097/SLA.0000000000001991. [DOI] [PubMed] [Google Scholar]