Abstract
Anastomotic leakage (AL) after resection for rectal carcinoma accelerates morbidity and mortality rates, extends hospital stay, and increases treatment costs, particularly when requiring laparotomy. The role of a protective diverting stoma (DS) in avoiding leakage has repeatedly been discussed, but prospective randomized studies on this subject are rare and their results contradictory. The MEDLINE database was searched for studies of AL requiring laparotomy and of the associated rate of protective DSs in initial anterior resection (AR) to review these studies systematically. The collected data were used to determine the average rate of AL requiring laparotomy after rectal cancer surgery in the DS group compared with that in the non-DS group. A total of 930 abstracts were retrieved from MEDLINE; 15 articles on AR and 22 on low/ultralow AR (LAR) were included in the review and analysis. The overall rate of AL requiring laparotomy was 6.57% (813/12, 376) in the AR studies and 4.13% (157/3, 802) in the LAR studies. In the AR studies, the pooled AL rate in the DS group was higher than that in the non-DS group (12.30% vs. 9.16%, P < 0.001). However, the pooled rate of AL requiring laparotomy in the DS group was lower than that in the non-DS group (3.69% vs. 7.42%, P < 0.001). In the LAR studies, the pooled AL rate in the DS group was lower than that in the non-DS group (7.74% vs. 9.64%, P = 0.045). The pooled rate of AL requiring laparotomy in the DS group was also lower than that in the non-DS group (2.67% vs. 5.21%, P < 0.001). By contrast, the pooled rate of definitive stomas and mortality caused by AL did not have any statistical difference between the DS and non-DS groups in both AR studies (definitive stomas: 0% vs. 0.65%; mortality: 0.95% vs. 1.19%) and LAR studies (definitive stomas: 1.03% vs. 1.01%; mortality: 0.35% vs. 0.36%). Protective DSs significantly decrease the rate of AL in LAR. AL requiring surgical correction was significantly reduced in the DS group in both AR and LAR studies. Protective DSs did not affect the definitive stomas and mortality rate; this lack of an effect warrants further high-quality clinical trials.
Keywords: Anastomotic leakage, diverting stoma, anterior resection, rectal cancer, complication
Introduction
Advances in surgical procedures and concepts, such as total mesorectal excision (TME), have dramatically increased the proportion of sphincter-saving procedures as the treatment of choice for rectal cancer patients. Simple and easy reconstruction has been facilitated by circular stapling devices, even in low-level anastomosis within a narrow pelvis. However, increased risk of anastomotic leakage (AL) is associated with sphincter-saving procedures. Clinically manifest anastomotic leaks are observed after 1-21 percent of resections for rectal carcinoma [1,2]. The mortality rate asso-ciated with symptomatic anastomotic leaks varies between 6 and 22 percent [3]. The role of a protective diverting stoma (DS) in avoiding this serious complication has repeatedly been discussed, but prospective randomized studies on this subject are rare and their results contradictory. Several authors have also argued that the stoma only mitigates the consequences of a leakage but does not lower the leakage rate itself [4].
This study aimed to systematically review studies of AL requiring laparotomy and the associated rate of DSs in initial anterior resection (AR) for rectal cancer. We gathered relevant data from these studies and determined the average rate of ALs requiring laparotomy after rectal cancer surgery in the DS group compared with that in the non-DS group. Given the impact of TME and stapling devices on AL, only studies using these techniques after the year 2000 were included in this review.
Methods
Definition of AL
The International Study Group of Rectal Cancer defines AL as a defect in intestinal wall integrity at the colorectal or coloanal anastomotic site, including the suture and staple lines of neorectal reservoirs. This defect leads to communication between the intra- and extra-luminal compartments. A pelvic abscess near the anastomosis is also considered AL [5]. In the systematic review, we used the number of ALs as defined in the study. Regarding the difference in impact and consequences for a patient, clinically relevant AL should be distinguished from radiologic leakage. In this review, we assessed clinical anastomotic leakage.
Literature search and selection strategy
Relevant studies published between January 2000 and December 2012 were identified from the MEDLINE search results. The following search terms were used: (rectum OR rectal OR proctectomy) AND (leakage OR failure OR integrity OR insufficiency OR breakdown OR defect OR separation OR dehiscence). Additional relevant articles were obtained from the citations in the publications identified by the initial search. Publications in English that met the following criteria were included in the review: availability of data on (i) laparotomic or laparoscopic sphincter-saving resection for rectal cancer, (ii) DS and non-DS groups, and (iii) the subsequent management of AL, including conservative treatment or laparotomy. Studies of preoperative chemoradiation therapy were excluded from the analysis, as were those that used single-access laparoscopic or robot-assisted surgery. Two authors (C. ZJ and H. LH) independently reviewed each of the included studies and extracted data from them. Any discrepancies were resolved by discussion. To increase the sensitivity of the search strategy, the reference lists of the retrieved literature were manually cross-searched for additional relevant publications.
Data extraction and analysis strategy
We used the number of ALs according to their definition in the studies. The selected studies were grouped into two: AR and low/ultralow AR (LAR). The pooled AL rates of the DS and non-DS groups in each group of studies were compared. Similarly, the data on AL requiring laparotomy for the DS and non-DS groups in each group were compared. When available, data on patients with AL managed by permanent colostomy or Hartmann’s procedure and on the mortality rate associated with AL were also extracted and compared between the two groups.
Statistical analysis
The relative frequencies were statistically analyzed through the chi-squared test in SPSS 13.0 for Windows (SPSS, Chicago, Illinois, USA). A two-sided P value of < 0.05 was considered statistically significant.
Results
Bibliometrics
A total of 930 abstracts were retrieved from MEDLINE from January 1, 2000, to December 5, 2012. Among these articles, 106 non-English articles and 289 non-relevant English articles with no or minimal association with AL were excluded. More articles (201) were excluded after the downloaded abstracts were examined according to the criteria in Figure 1. A total of 334 full papers were examined, 291 of which were rejected because of irrelevance. The final analysis included 37 studies: 15 AR studies [4,6-19] and 22 LAR studies [20-41].
Figure 1.
Literature search, review, and analysis (From January 2000 to December 2012).
The selected studies
Thirty-seven studies of the rate of DSs and AL requiring laparotomy after AR or LAR were analyzed. The included studies had a total population of 16,178 patients. The sample sizes of the studies varied from 27 to 2,729 patients. No clear, applicable criteria for DS construction were stipulated in any of the included studies, and the DS construction decision was made by the surgeon in each study except in three randomized controlled trials [8,26,27]. Twenty studies reported a total of 2,069 DS cases: loop ileostomy (1628, 78.69%), loop transverse colostomy (366, 17.69%), and percutaneous ileostomy (75, 3.62%).
Fourteen studies reported a total of 124 AL cases cured by conservative treatment: 23 (18.54%) by transanal drainage, 21 (16.94%) by endoscopic drainage, 14 (11.29%) by computed tomography (CT)-guided/percutaneous drainage, 44 (35.48%) by drain irrigation, and 22 (17.74%) by conservative antibiotherapy.
Twenty-six studies reported a total of 434 AL patients who underwent reoperation. The detail of reoperation was described in Table 1. The rate of temporary stomas was 77.65%, and definitive stomas, such as permanent stomas, Hartmann’s procedure, and abdominoperineal resection, made up only 14.52% of all ALs requiring surgical re-intervention.
Table 1.
Detail of reoperation from 26 studies with 434 patients of AL
| No (%) | Reoperation |
|---|---|
| 29 (6.68%) | Only drainage |
| 266 (61.29%) | Only diverting stomas* |
| 3 (0.69%) | Drainage and repair anastomosis |
| 61 (14.06%) | Drainage with stoma† |
| 2 (0.46%) | Repair anastomosis via the anus |
| 2 (0.46%) | Repair anastomosis via the anus with stoma |
| 2 (0.46%) | Reconstruct anastomosis |
| 6 (1.38%) | Reconstruct anastomosis with stoma |
| 21 (4.84%) | Permanent end colostomy |
| 38 (8.76%) | Hartmann’s procedure |
| 4 (0.92%) | Abdominoperineal resection |
Including 139 for loop ileostomy, 44 for loop colostomy, and 83 unspecified.
Including 6 for loop ileostomy, 7 for loop colostomy, and 48 unspecified.
Incidence of ALs and ALs requiring laparotomy
The 15 AR studies had a total of 12,376 patients. The number of patients per study confirmed to have anastomotic leaks ranged from 6 to 390 (1,222 in total). The overall rate of AL was 9.87%. The highest AL rate reported was 19.23%, whereas the lowest was 2.52% (Figure 2). The number of patients per study confirmed to have AL requiring laparotomy ranged from 3 to 218 (813 in total). The overall rate of AL requiring laparotomy was 6.57%; in other words, 66.53% (813/1222) of AL patients required surgical correction. The highest rate of AL requiring laparotomy was 11.97%, whereas the lowest was 2.52% (Figure 2).
Figure 2.
Rates of anastomotic leakage and anastomotic leakage requiring laparotomy in AR studies.
The 22 LAR studies had a total of 3,802 patients. The number of patients per study confirmed to have anastomotic leaks ranged from 2 to 51 (338 in total). The overall rate of AL was 8.89%. The highest AL rate reported was 20.59%, whereas the lowest was 1.89% (Figure 3). The number of patients per study confirmed to have AL requiring laparotomy ranged from 0 to 26 (157 in total). The overall rate of AL requiring laparotomy was 4.13%; in other words, 46.45% (157/338) of AL patients required surgical correction. The highest rate of AL requiring laparotomy was 17.65%, whereas the lowest was 0% (Figure 3).
Figure 3.
Rates of anastomotic leakage and anastomotic leakage requiring laparotomy in LAR studies.
DS vs. non-DS group in AR
AL rate
The 15 AR studies had a total of 2820 patients with a DS (ranging from 0 to 881). The overall rate of DS was 22.79%. The highest DS rate reported was 56.60%, whereas the lowest was 0%. Interestingly, the pooled AL rate in the DS group was 12.30% (347/2820), higher than that in the non-DS group (9.16%, 875/9556). The difference between these rates was statistically significant (P < 0.001).
Rate of AL requiring laparotomy
The pooled rate of AL requiring laparotomy in the DS group was 3.69% (104/2820), significantly lower than that in the non-DS group (7.42%, 709/9556) (P < 0.001). Only 29.97% (104/347) of the AL cases in the DS group needed surgical re-intervention, whereas the proportion in the non-DS group was as high as 81.03% (709/875). The AL rates in patients with or without a DS in the 15 AR studies are described and compared in Table 2.
Table 2.
Rate of anastomotic leakage in patients with or without diverting stoma in AR studies
| Author | Year | Study Design | Patients (n) | Anastomotic Leakage | AL Requiring Laparotomy | ||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
| DS (%) | Non-DS (%) | P | DS (%) | Non-DS (%) | P | ||||
| Bittorf B | 2003 | - | 150 | 14 (18.18%) | 8 (10.96%) | .211 | 1 (1.30%) | 6 (8.22%) | .044 |
| Matthiessen P | 2004 | Retrospective | 432 | 11 (15.28%) | 42 (11.67%) | .393 | 2 (2.78%) | 37 (10.28%) | .042 |
| Gastinger I | 2005 | Prospective | 2729 | 128 (14.53%) | 262 (14.18%) | .806 | 32 (3.63%) | 186 (10.06%) | < .001 |
| Vlot EA | 2005 | - | 144 | - | 7 (4.86%) | - | 7 (4.86%) | ||
| Peeters KC | 2005 | Retrospective | 924 | 43 (8.22%) | 64 (15.96%) | < .001 | 26 (4.97%) | 60 (14.96%) | < .001 |
| Ptok H | 2007 | - | 2044 | 125 (15.78%) | 178 (14.22%) | .331 | 39 (4.92%) | 118 (9.42%) | < .001 |
| Matthiessen P | 2007 | RCT | 234 | 12 (10.34%) | 33 (27.97%) | < .001 | 0 (0%) | 28 (23.73%) | < .001 |
| Eberl T | 2008 | Retrospective | 472 | 4 (3.64%) | 45 (12.43%) | .008 | 2 (1.82%) | 44 (12.15%) | .001 |
| Jung SH | 2008 | Retrospective | 1391 | 0 (0%) | 35 (2.62%) | .228 | 0 (0%) | 35 (2.62%) | .228 |
| Choi DH | 2010 | Prospective | 178 | 1 (4.55%) | 16 (10.26%) | .393 | 0 (0%) | 16 (10.26%) | .115 |
| Shin US | 2010 | - | 1838 | - | 79 (4.30%) | - | 79 (4.30%) | ||
| Lin JK | 2011 | Retrospective | 821 | 8 (5.52%) | 35 (5.18%) | .867 | 2 (1.38%) | 35 (5.18%) | .045 |
| Chen W | 2011 | - | 750 | 0 (0%) | 57 (7.65%) | .519 | 0 (0%) | 46 (6.17%) | .566 |
| Zhao WT | 2012 | - | 158 | - | 9 (5.70%) | - | 9 (5.70%) | ||
| Yamamoto S | 2012 | Prospective | 111 | 1 (4.35%) | 5 (5.68%) | .801 | 0 (0%) | 3 (3.41%) | .369 |
| Total | 12,376 | 347 (12.30%) | 875 (9.16%) | < .001 | 104 (3.69%) | 709 (7.42%) | < .001 | ||
AL, anastomotic leakage; DS, diverting stoma; RCT, randomized controlled trial.
Definitive stoma after AL
For patients with AL requiring re-operation, a permanent stoma, Hartmann’s procedure, or abdominoperineal resection was usually considered the possible definitive stomas that led poor quality of life. In eight AR studies that reported the re-operation procedures for AL between the groups, the overall number of patients in the DS group was 315, and AL requiring re-operation occurred in three patients, none of whom developed a definitive stoma. In the non-DS group (3536 patients), AL requiring re-operation occurred in 223 patients, 23 of which (0.65%) developed a definitive stoma. However, the difference was not significant (P = 0.151). The proportion of definitive stomas in patients with AL requiring re-operation was zero in the DS group and 10.31% (23/223) in the non-DS group.
Mortality after AL
We also extracted the number of postoperative mortalities caused by AL as defined in the studies between the two groups. In 10 AR studies that met the criteria, the overall number of patients in the DS group was 1680; AL occurred in 314 patients, 16 of which died because of AL. The number of patients in the non-DS group was 5,272, and AL occurred in 676 patients, 63 of which died. Therefore, the rate of mortality caused by AL in the DS and non-DS groups was 0.95% (16/1680) and 1.19% (63/5272), respectively, with no significant difference (P = 0.413).
DS vs. non-DS group in LAR
AL rate
The 22 LAR studies had a total of 1498 patients with DS (ranging from 0 to 148). The overall rate of DS was 39.4%. The highest DS rate reported was 100%, whereas the lowest was 0%. The pooled AL rate in the DS group was 7.74% (116/1498), lower than that in the non-DS group (9.64%, 222/2304); the difference was statistically significant (P = 0.045).
Rate of AL requiring laparotomy
As to AL requiring laparotomy, the pooled rate in the DS group (2.67%, 40/1498) was also much lower than that in the non-DS group (5.21%, 120/2304), with a significant difference (P < 0.001). Only 34.48% (40/116) of those with AL in the DS group required surgical re-intervention, whereas the proportion in the non-DS group was as high as 54.05% (120/222). The AL rates in patients with or without a DS in the 22 LAR studies are described and compared in Table 3.
Table 3.
Rate of anastomotic leakage in patients with or without diverting stoma in LAR studies
| Author | Year | Study Design | Patients (n) | Anastomotic Leakage | AL Requiring Laparotomy | ||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
| DS (%) | Non-DS (%) | P | DS (%) | Non-DS (%) | P | ||||
| Law WI | 2000 | Prospective | 196 | 5 (4.85%) | 15 (16.13%) | .009 | 0 (0%) | 10 (10.75%) | < .001 |
| Nesbakken A | 2001 | - | 92 | 5 (17.24%) | 12 (19.05%) | .835 | 1 (3.45%) | 11 (17.46%) | .063 |
| Z’graggen K | 2001 | - | 41 | 3 (7.32%) | - | 1 (2.44%) | - | ||
| Ho YH | 2002 | - | 88 | 7 (7.95%) | - | 1 (1.14%) | - | ||
| Leester B | 2002 | Retrospective | 249 | 7 (9.46%) | 9 (5.14%) | .204 | 0 (0%) | 8 (4.57%) | .061 |
| Marusch F | 2002 | Prospective | 482 | 16 (10.81%) | 35 (10.48%) | .913 | 3 (2.03%) | 23 (6.89%) | .029 |
| Kanellos I | 2002 | - | 82 | - | 8 (9.76%) | - | 4 (4.88%) | ||
| Eckmann C | 2004 | - | 306 | - | 30 (9.80%) | - | 12 (3.92%) | ||
| Vorobiev GI | 2004 | - | 27 | 2 (7.41%) | - | 0 (0%) | - | ||
| Chamlou R | 2007 | - | 90 | 8 (8.89%) | - | 3 (3.33%) | - | ||
| Chude GG | 2008 | RCT | 256 | 3 (2.21%) | 12 (10.00%) | .008 | 0 (0%) | 4 (3.33%) | .031 |
| Lefebure B | 2008 | Retrospective | 132 | 3 (7.14%) | 10 (11.11%) | .476 | 0 (0%) | 5 (5.56%) | .119 |
| Ulrich AB | 2009 | RCT | 34 | 1 (5.56%) | 6 (37.50%) | .021 | 0 (0%) | 6 (37.50%) | .004 |
| Peng J | 2010 | - | 639 | - | 45 (7.04%) | - | 12 (1.88%) | ||
| Akasu T | 2010 | Retrospective | 120 | 14 (13.21%) | 1 (7.14%) | .519 | 6 (5.66%) | 0 (0%) | .361 |
| Kruschewski M | 2011 | - | 128 | 15 (11.72%) | - | 12 (9.38%) | - | ||
| Shiomi A | 2011 | Retrospective | 329 | 9 (7.50%) | 24 (11.48%) | .246 | 1 (0.83%) | 13 (6.22%) | .019 |
| Fouda E | 2011 | - | 56 | 1 (5.88%) | 7 (17.95%) | .235 | 0 (0%) | 7 (17.95%) | .061 |
| Glancy DG | 2012 | - | 144 | 5 (8.93%) | 8 (9.09%) | .973 | 5 (8.93%) | 5 (5.68%) | .454 |
| Biondo S | 2012 | - | 106 | 2 (1.89%) | - | 1 (0.94%) | - | ||
| Rondelli F | 2012 | - | 143 | 5 (3.50%) | - | 4 (2.80%) | - | ||
| Gong H | 2012 | Retrospective | 62 | 5 (19.23%) | 0 (0%) | .006 | 2 (7.69%) | 0 (0%) | .090 |
| Total | 3,802 | 116 (7.74%) | 222 (9.64%) | .045 | 40 (2.67%) | 120 (5.21%) | < .001 | ||
AL, anastomotic leakage; DS, diverting stoma; RCT, randomized controlled trial.
Definitive stoma after AL
In 13 LAR studies that reported the re-operation procedures for AL between the groups, the overall number of patients in the DS group was 679; AL requiring re-operation occurred in 21 patients, 7 of which (1.03%) developed a definitive stoma. In the non-DS group, which had a total of 1683 patients, AL requiring re-operation occurred in 79 patients, 17 of which (1.01%) developed a definitive stoma. The difference in the incidence of definitive stomas was not significant (P = 0.963). However, the proportion of definitive stomas in patients with AL requiring re-operation was higher in the DS group (33.33%, 7/21) than in the non-DS group (21.52%, 17/79).
Mortality after AL
Twenty-one LAR studies obtained the rate of mortalities caused by AL between the two groups. The overall number of patients in the DS group was 1442; AL occurred in 111 patients, 5 of which died because of AL. The number of patients in the non-DS group was 2226; AL occurred in 214 patients, 8 of which died. Therefore, the rate of mortality caused by AL in the DS and non-DS groups was 0.35% (5/1442) and 0.36% (8/2226), respectively, with no significant difference (P = 0.949).
Discussion
Leakage accelerates the morbidity and mortality rates, extends hospital stay, and increases treatment costs, particularly when requiring laparotomy. Leakage is also associated with postoperative local recurrence and functional results and influences long-term outcomes [19,38,42-44]. Therefore, AL has become an urgent problem among colorectal surgeons.
Patients with AL requiring laparotomy are often in critical condition, presenting most often with purulent/fecal drains and markedly increased parameters of infection (e.g., leukocytosis and C-reactive proteins). These patients usually have abdominal pain and fever and develop signs of peritonitis (e.g., tenderness to palpation, abdominal wall rigidity, and tachycardia). Imaging studies, such as of CT with transrectal instillation of contrast, reveal considerable leakage at the anastomotic site with fluid collection in the pelvis. With significant clinical sepsis after rectal anastomosis, infection control should meet three requirements: drainage of the infected material, eradication of the source of infection, and prevention of recurrent sepsis [45,46]. The source of infection may be removed from AL by diverting colostomy or Hartmann’s procedure. If such operative re-intervention is delayed or not performed, the clinical conditions of the patient deteriorate and ultimately result in sepsis with clinical signs of hypothermia, leukopenia, and organ failure.
AL development depends on numerous factors. AL occurs in medically fragile patients, after a technically difficult operation, or with intraoperative adverse events. However, AL also occurs in patients with no obvious risk factors [47]. The difficulty in predicting AL, including in patients considered at low risk, has generated several studies aiming to identify risk factors [8,48-50]. One risk factor found by retrospective studies with multivariable analysis is the absence of a DS [16,51].
Fecal diversion in rectal cancer surgery is an old concept. To minimize the risk of clinical leakage, the construction of a DS seems useful for patients with distal rectal cancer. As a result, routine DS is usually recommended when risk factors for AL are present. However, this recommendation is not universally accepted [3,47,52,53] because closing a protective DS indicates additional surgery, admission to a hospital, and a risk of complications and death [54,55]. However, many surgeons feel that they would harm their patients by abandoning a protective DS.
The role of a protective DS in avoiding AL and its related complications has repeatedly been discussed. Theoretically, a DS is constructed to divert the fecal stream from a healing anastomotic site and protect fragile such sites. However, whether diverting the fecal stream in itself directly prevents AL remains unconfirmed, and the necessity of a protective DS remains controversial because of the lack of data from large randomized controlled trials. The results of rat studies on this subject are similarly contradictory [56-58]. Other prospective and retrospective studies also had different conclusions. Some studies found that the absence of a DS was a risk factor for leakage in LAR. In Peeters et al. [16], for instance, 9% of defunctioned patients leaked, compared with the 24% of those not defunctioned. Other studies found that the stoma does not lower the leakage rate but only mitigates the consequences of a leakage [18-21,38]. The rate of AL requiring surgical intervention is significantly reduced by a protective stoma provided after LAR [20]. The overall rate of AL is also not influenced by the presence of a DS, although patients with a stoma developed significantly fewer leaks that required surgical correction [4]. Thus, several authors suggest that a protective DS does not prevent AL itself but only mitigates the consequence of the AL.
Our review revealed that a DS can reduce the rate of AL requiring laparotomy. The pooled proportion of ALs requiring laparotomy was lower in the LAR studies (46.45%) than in the AR studies (66.53%) probably because of the higher DS rate in the former (39.40%) than in the latter (22.79%). The relationship was more obvious when the rate of AL requiring laparotomy in the DS group was compared with that in the non-DS group. In the AR studies, the proportion of re-laparotomy in AL patients with a DS (29.97%) was much lower than that in AL patients without a DS (81.03%). The same observation was made in the LAR studies (34.48% in DS vs. 54.05% in non-DS). The rate of AL requiring laparotomy in the DS group was significantly lower than that in the non-DS group, in both the AR studies (3.69% in DS vs. 7.42% in non-DS, P < 0.001) and the LAR studies (2.67% in DS vs. 5.21% in non-DS, P < 0.001).
As to the protective effect of a DS on reducing the overall AL rate, our findings revealed that the AL rate in the LAR studies was significantly reduced (7.74% in DS vs. 9.64% in non-DS, P = 0.045). Interestingly, in the AR studies, the overall rate of AL in the DS group was unexpectedly higher than that in the non-DS group (12.30% in DS vs. 9.16% in non-DS, P < 0.001). However, because of the general selection bias of most of the non-randomized studies in our review, we cannot conclude whether a DS can prevent overall leakage. This selection bias resulted from the selective creation of a protective DS based on the subjective judgment of surgeons for predicting “risky” anastomoses to minimize potential consequences. Therefore, the protective effect of a DS might even be greater than what our results imply. Another bias is the incidence of asymptomatic ALs that might have been missed in either group because anastomoses in clinically stable patients were not systematically assessed in nearly all the studies. Therefore, estimating the overall percentage of AL patients that benefit from a DS is difficult, but ALs requiring laparotomy can be estimated without this bias.
The protective effect of a DS on reducing AL requiring laparotomy is confirmed by our review. However, can this evidence justify the creation of a routine DS despite associated additional surgery, admission to hospital, risk of complications and death, and additional costs? We propose that reducing AL requiring laparotomy is not enough to prove the value of a DS. Our results show that 33.47% of ALs in the AR studies and 53.55% of those in the LAR studies were cured by conservative treatment, such as drainage placement and irrigation. Even for AL requiring laparotomy, most patients may be treated with only an ileostomy or colostomy. Only 14.52% of patients required an emergency operation to sever the anastomosis and create a terminal stoma and a Hartmann’s pouch because of post-AL pelvic sepsis after AL, which may affect the quality of life of the patient in the future. Therefore, we should focus on the protective effect of a DS on reducing definitive stomas and AL-associated mortality.
We extracted and compared data on the definitive stomas and AL-associated mortality of the DS and non-DS groups. No cases of definitive stoma were found in the DS group in the eight AR studies, whereas the non-DS group had a rate of 0.65%; however, the difference was not statistically significant (P = 0.151). The same result was obtained in 22 LAR studies (1.03% vs. 1.01%, P = 0.963). The rate of AL-associated mortality in the 10 AR studies did not have significant difference between the DS and non-DS groups (0.95% vs. 1.19%, P = 0.413) either. The same result was obtained in 21 LAR studies (0.35% vs. 0.36%, P = 0.963). However, few studies have focused on these endpoints to evaluate the value of a DS; therefore, the pooled data were too limited to allow a definitive conclusion. More studies about the DS in relation to this topic are recommended for future research.
This article is limited by the source of the publications reviewed. All abstracts were retrieved from MEDLINE, and non-English language papers were excluded from the final analysis. A formal meta-analysis will provide more powerful evidence; the methodology used in this study was not as powerful as a meta-analysis. However, our systematic review provides data consolidated directly from original publications on the rate of AL requiring laparotomy between DS and non-DS groups.
In conclusion, DSs, as a mode of fecal diversion, significantly decreased the rate of AL in LAR studies. AL requiring surgical correction was significantly reduced in the DS group in both AR and LAR studies. However, a protective DS did not affect the definitive stoma and mortality rates; this lack of an effect warrants further high-quality trials because the data were too limited to produce powerful evidence on the subject. Considering the morbidity and mortality associated with DSs and the uncertain effect of DSs on reducing definitive stomas and mortality, DSs should be used only in situations of intra-operative difficulty, for lower rectal carcinomas, and in patients with poor general health conditions.
Disclosure of conflict of interest
None.
References
- 1.Akagi T, Inomata M, Etoh T, Moriyama H, Yasuda K, Shiraishi N, Eshima N, Kitano S. Multivariate evaluation of the technical difficulties in performing laparoscopic anterior resection for rectal cancer. Surg Laparosc Endosc Percutan Tech. 2012;22:52–7. doi: 10.1097/SLE.0b013e31824019fc. [DOI] [PubMed] [Google Scholar]
- 2.Maggiori L, Bretagnol F, Lefèvre JH, Ferron M, Vicaut E, Panis Y. Conservative management is associated with a decreased risk of definitive stoma after anastomotic leakage complicating sphincter-saving resection for rectal cancer. Colorectal Dis. 2011;13:632–7. doi: 10.1111/j.1463-1318.2010.02252.x. [DOI] [PubMed] [Google Scholar]
- 3.Ajani JA. In rectal carcinoma, colostomy or no colostomy: is this the question? J. Clin. Oncol. 1993;11:193–4. doi: 10.1200/JCO.1993.11.1.193. [DOI] [PubMed] [Google Scholar]
- 4.Gastinger I, Marusch F, Steinert R, Wolff S, Koeckerling F, Lippert H. Protective defunctioning stoma in low anterior resection for rectal carcinoma. Br J Surg. 2005;92:1137–42. doi: 10.1002/bjs.5045. [DOI] [PubMed] [Google Scholar]
- 5.Rahbari NN, Weitz J, Hohenberger W, Heald RJ, Moran B, Ulrich A, Holm T, Wong WD, Tiret E, Moriya Y, Laurberg S, den Dulk M, van de Velde C, Büchler MW. Definition and grading of anastomotic leakage following anterior resection of the rectum: a proposal by the International Study Group of Rectal Cancer. Surgery. 2010;147:339–51. doi: 10.1016/j.surg.2009.10.012. [DOI] [PubMed] [Google Scholar]
- 6.Matthiessen P, Hallböök O, Andersson M, Rutegård J, Sjödahl R. Risk factors for anastomotic leakage after anterior resection of the rectum. Colorectal Dis. 2004;6:462–9. doi: 10.1111/j.1463-1318.2004.00657.x. [DOI] [PubMed] [Google Scholar]
- 7.Vlot EA, Zeebregts CJ, Gerritsen JJ, Mulder HJ, Mastboom WJ, Klaase JM. Anterior resection of rectal cancer without bowel preparation and diverting stoma. Surg Today. 2005;35:629–33. doi: 10.1007/s00595-005-2999-2. [DOI] [PubMed] [Google Scholar]
- 8.Matthiessen P, Hallböök O, Rutegård J, Simert G, Sjödahl R. Defunctioning stoma reduces symptomatic anastomotic leakage after low anterior resection of the rectum for cancer: a randomized multicenter trial. Ann Surg. 2007;246:207–14. doi: 10.1097/SLA.0b013e3180603024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Eberl T, Jagoditsch M, Klingler A, Tschmelitsch J. Risk factors for anastomotic leakage after resection for rectal cancer. Am J Surg. 2008;196:592–8. doi: 10.1016/j.amjsurg.2007.10.023. [DOI] [PubMed] [Google Scholar]
- 10.Jung SH, Yu CS, Choi PW, Kim DD, Park IJ, Kim HC. Risk factors and oncologic impact of anastomotic leakage after rectal cancer surgery. Dis Colon Rectum. 2008;51:902–8. doi: 10.1007/s10350-008-9272-x. [DOI] [PubMed] [Google Scholar]
- 11.Choi DH, Hwang JK, Ko YT, Jang HJ, Shin HK, Lee YC. Risk factors for anastomotic leakage after laparoscopic rectal resection. J Korean Soc Coloproctol. 2010;26:265–73. doi: 10.3393/jksc.2010.26.4.265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shin US, Kim CW, Yu CS, Kim JC. Delayed anastomotic leakage following sphincter-preserving surgery for rectal cancer. Int J Colorectal Dis. 2010;25:843–9. doi: 10.1007/s00384-010-0938-1. [DOI] [PubMed] [Google Scholar]
- 13.Lin JK, Yueh TC, Chang SC, Lin CC, Lan YT, Wang HS, Yang SH, Jiang JK, Chen WS, Lin TC. The influence of fecal diversion and anastomotic leakage on survival after resection of rectal cancer. J Gastrointest Surg. 2011;15:2251–61. doi: 10.1007/s11605-011-1721-5. [DOI] [PubMed] [Google Scholar]
- 14.Chen W, Li Y, Liao Z, Lin G, Cai G, Lin K. Active lymphangiogenesis is a major risk factor for anastomotic leakage following sphincter-sparing resection of rectal cancer. J Surg Oncol. 2011;104:493–8. doi: 10.1002/jso.21965. [DOI] [PubMed] [Google Scholar]
- 15.Zhao WT, Hu FL, Li YY, Li HJ, Luo WM, Sun F. Use of a transanal drainage tube for prevention of anastomotic leakage and bleeding after anterior resection for rectal cancer. World J Surg. 2013;37:227–32. doi: 10.1007/s00268-012-1812-9. [DOI] [PubMed] [Google Scholar]
- 16.Peeters KC, Tollenaar RA, Marijnen CA, Klein Kranenbarg E, Steup WH, Wiggers T, Rutten HJ, van de Velde CJ Dutch Colorectal Cancer Group. Risk factors for anastomotic failure after total mesorectal excision of rectal cancer. Br J Surg. 2005;92:211–6. doi: 10.1002/bjs.4806. [DOI] [PubMed] [Google Scholar]
- 17.Yamamoto S, Fujita S, Akasu T, Inada R, Moriya Y, Yamamoto S. Risk factors for anastomotic leakage after laparoscopic surgery for rectal cancer using a stapling technique. Surg Laparosc Endosc Percutan Tech. 2012;22:239–43. doi: 10.1097/SLE.0b013e31824fbb56. [DOI] [PubMed] [Google Scholar]
- 18.Bittorf B, Stadelmaier U, Merkel S, Hohenberger W, Matzel KE. Does anastomotic leakage affect functional outcome after rectal resection for cancer? Langenbecks Arch Surg. 2003;387:406–10. doi: 10.1007/s00423-003-0349-7. [DOI] [PubMed] [Google Scholar]
- 19.Ptok H, Marusch F, Meyer F, Schubert D, Gastinger I, Lippert H. Impact of anastomotic leakage on oncological outcome after rectal cancer resection. Br J Surg. 2007;94:1548–54. doi: 10.1002/bjs.5707. [DOI] [PubMed] [Google Scholar]
- 20.Marusch F, Koch A, Schmidt U, Geibetaler S, Dralle H, Saeger HD, Wolff S, Nestler G, Pross M, Gastinger I, Lippert H. Value of a protective stoma in low anterior resections for rectal cancer. Dis Colon Rectum. 2002;45:1164–71. doi: 10.1007/s10350-004-6384-9. [DOI] [PubMed] [Google Scholar]
- 21.Leester B, Asztalos I, Polnyib C. Septic complications after low anterior rectal resection--is diverting stoma still justified? Acta Chir Iugosl. 2002;49:67–71. doi: 10.2298/aci0202067l. [DOI] [PubMed] [Google Scholar]
- 22.Law WI, Chu KW, Ho JW, Chan CW. Risk factors for anastomotic leakage after low anterior resection with total mesorectal excision. Am J Surg. 2000;179:92–6. doi: 10.1016/s0002-9610(00)00252-x. [DOI] [PubMed] [Google Scholar]
- 23.Eckmann C, Kujath P, Schiedeck TH, Shekarriz H, Bruch HP. Anastomotic leakage following low anterior resection: results of a standardized diagnostic and therapeutic approach. Int J Colorectal Dis. 2004;19:128–33. doi: 10.1007/s00384-003-0498-8. [DOI] [PubMed] [Google Scholar]
- 24.Vorobiev GI, Odaryuk TS, Tsarkov PV, Talalakin AI, Rybakov EG. Resection of the rectum and total excision of the internal anal sphincter with smooth muscle plasty and colonic pouch for treatment of ultralow rectal carcinoma. Br J Surg. 2004;91:1506–12. doi: 10.1002/bjs.4330. [DOI] [PubMed] [Google Scholar]
- 25.Chamlou R, Parc Y, Simon T, Bennis M, Dehni N, Parc R, Tiret E. Long-term results of intersphincteric resection for low rectal cancer. Ann Surg. 2007;246:916–21. doi: 10.1097/SLA.0b013e31815c29ff. [DOI] [PubMed] [Google Scholar]
- 26.Chude GG, Rayate NV, Patris V, Koshariya M, Jagad R, Kawamoto J. Defunctioning loop ileostomy with low anterior resection for distal rectal cancer: should we make an ileostomy as a routine procedure? A prospective randomized study. Hepatogastroenterology. 2008;55:1562–7. [PubMed] [Google Scholar]
- 27.Ulrich AB, Seiler C, Rahbari N, Weitz J, Büchler MW. Diverting stoma after low anterior resection: more arguments in favor. Dis Colon Rectum. 2009;52:412–8. doi: 10.1007/DCR.0b013e318197e1b1. [DOI] [PubMed] [Google Scholar]
- 28.Peng J, Lu J, Xu Y, Guan Z, Wang M, Cai G, Cai S. Standardized pelvic drainage of anastomotic leaks following anterior resection without diversional stomas. Am J Surg. 2010;199:753–8. doi: 10.1016/j.amjsurg.2009.03.026. [DOI] [PubMed] [Google Scholar]
- 29.Akasu T, Takawa M, Yamamoto S, Yamaguchi T, Fujita S, Moriya Y. Risk factors for anastomotic leakage following intersphincteric resection for very low rectal adenocarcinoma. J Gastrointest Surg. 2010;14:104–11. doi: 10.1007/s11605-009-1067-4. [DOI] [PubMed] [Google Scholar]
- 30.Biondo S, Frago R, Codina Cazador A, Farres R, Olivet F, Golda T, Miguel B, Kreisler E. Long-term functional results from a randomized clinical study of transverse coloplasty compared with colon J-pouch after low anterior resection for rectal cancer. Surgery. 2013;153:383–92. doi: 10.1016/j.surg.2012.08.012. [DOI] [PubMed] [Google Scholar]
- 31.Rondelli F, Balzarotti R, Bugiantella W, Mariani L, Pugliese R, Mariani E. Temporary percutaneous ileostomy versus conventional loop ileostomy in mechanical extraperitoneal colorectalanastomosis: a retrospective study. Eur J Surg Oncol. 2012;38:1065–70. doi: 10.1016/j.ejso.2012.07.110. [DOI] [PubMed] [Google Scholar]
- 32.Gong H, Yu Y, Yao Y. Clinical Value of Preventative Ileostomy Following Ultra-Low Anterior Rectal Resection. Cell Biochem Biophys. 2013;65:491–3. doi: 10.1007/s12013-012-9445-z. [DOI] [PubMed] [Google Scholar]
- 33.Kanellos I, Zacharakis E, Christoforidis E, Demetriades H, Betsis D. Low anterior resection without defunctioning stoma. Tech Coloproctol. 2002;6:153–6. doi: 10.1007/s101510200035. [DOI] [PubMed] [Google Scholar]
- 34.Lefebure B, Tuech JJ, Bridoux V, Costaglioli B, Scotte M, Teniere P, Michot F. Evaluation of selective defunctioning stoma after low anterior resection for rectal cancer. Int J Colorectal Dis. 2008;23:283–8. doi: 10.1007/s00384-007-0380-1. [DOI] [PubMed] [Google Scholar]
- 35.Glancy DG, Chaudhray BN, Greenslade GL, Dixon AR. Laparoscopic total mesorectal excision can be performed on a nonselective basis in patients with rectal cancer with excellent medium-term results. Colorectal Dis. 2012;14:453–7. doi: 10.1111/j.1463-1318.2011.02682.x. [DOI] [PubMed] [Google Scholar]
- 36.Kruschewski M, Gröne J, Vogel N, Zimmermann M, Buhr HJ. Management and results of complications after anterior resection with colonic pouch reconstruction for rectal cancer. Colorectal Dis. 2011;13:284–9. doi: 10.1111/j.1463-1318.2009.02140.x. [DOI] [PubMed] [Google Scholar]
- 37.Shiomi A, Ito M, Saito N, Ohue M, Hirai T, Kubo Y, Moriya Y. Diverting stoma in rectal cancer surgery. A retrospective study of 329 patients from Japanese cancer centers. Int J Colorectal Dis. 2011;26:79–87. doi: 10.1007/s00384-010-1036-0. [DOI] [PubMed] [Google Scholar]
- 38.Nesbakken A, Nygaard K, Lunde OC. Outcome and late functional results after anastomotic leakage following mesorectal excision for rectal cancer. Br J Surg. 2001;88:400–4. doi: 10.1046/j.1365-2168.2001.01719.x. [DOI] [PubMed] [Google Scholar]
- 39.Fouda E, El Nakeeb A, Magdy A, Hammad EA, Othman G, Farid M. Early detection of anastomotic leakage after elective low anterior resection. J Gastrointest Surg. 2011;15:137–44. doi: 10.1007/s11605-010-1364-y. [DOI] [PubMed] [Google Scholar]
- 40.Z’graggen K, Maurer CA, Birrer S, Giachino D, Kern B, Büchler MW. A new surgical concept for rectal replacement after low anterior resection: the transverse coloplasty pouch. Ann Surg. 2001;234:780–5. doi: 10.1097/00000658-200112000-00009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Ho YH, Brown S, Heah SM, Tsang C, Seow-Choen F, Eu KW, Tang CL. Comparison of J-pouch and coloplasty pouch for low rectal cancers: a randomized, controlled trial investigating functional results and comparative anastomotic leak rates. Ann Surg. 2002;236:49–55. doi: 10.1097/00000658-200207000-00009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Hallbook O, Sjodahl R. Anastomotic leakage and functional outcome after anterior resection of the rectum. Br J Surg. 1996;83:60–2. doi: 10.1002/bjs.1800830119. [DOI] [PubMed] [Google Scholar]
- 43.Bell SW, Walker KG, Rickard MJ, Sinclair G, Dent OF, Chapuis PH, Bokey EL. Anastomotic leakage after curative anterior resection results in a higher prevalence of local recurrence. Br J Surg. 2003;90:1261–6. doi: 10.1002/bjs.4219. [DOI] [PubMed] [Google Scholar]
- 44.Merkel S, Wang WY, Schmidt O, Dworak O, Wittekind C, Hohenberger W, Hermanek P. Locoregional recurrence in patients with anastomotic leakage after anterior resection for rectal carcinoma. Colorectal Dis. 2001;3:154–60. doi: 10.1046/j.1463-1318.2001.00232.x. [DOI] [PubMed] [Google Scholar]
- 45.Wittmann DH, Schein M, Condon RE. Management of secondary peritonitis. Ann Surg. 1996;224:10–8. doi: 10.1097/00000658-199607000-00003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Frileux P, Quilichini MA, Cugnenc PH, Parc R, Levy E, Loygue J. Postoperative peritonitis of colonic origin. Apropos of 155 cases. Ann Chir. 1985;39:649–59. [PubMed] [Google Scholar]
- 47.Poon RT, Chu KW, Ho JW, Chan CW, Law WL, Wong J. Prospective evaluation of selective defunctioning stoma for low anterior resection with total mesorectal excision. World J Surg. 1999;23:463–467. doi: 10.1007/pl00012331. discussion 467-8. [DOI] [PubMed] [Google Scholar]
- 48.Law WL, Chu KW. Anterior resection for rectal cancer with mesorectal excision: a prospective evaluation of 622 patients. Ann Surg. 2004;240:260–268. doi: 10.1097/01.sla.0000133185.23514.32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Rullier E, Laurent C, Garrelon JL, Michel P, Saric J, Parneix M. Risk factors for anastomotic leakage after resection of rectal cancer. Br J Surg. 1998;85:355–8. doi: 10.1046/j.1365-2168.1998.00615.x. [DOI] [PubMed] [Google Scholar]
- 50.Eriksen MT, Wibe A, Norstein J, Haffner J, Wiig JN Norwegian Rectal Cancer Group. Anastomotic leakage following routine mesorectal excision for rectal cancer in a national cohort of patients. Colorectal Dis. 2005;7:51–7. doi: 10.1111/j.1463-1318.2004.00700.x. [DOI] [PubMed] [Google Scholar]
- 51.Dehni N, Schlegel RD, Cunningham C, Guiguet M, Tiret E, Parc R. Influence of a defunctioning stoma on leakage rates after low colorectal anastomosis and colonic J pouch-anal anastomosis. Br J Surg. 1998;85:1114–7. doi: 10.1046/j.1365-2168.1998.00790.x. [DOI] [PubMed] [Google Scholar]
- 52.Guivarc’h M, Mosnier H, Roulett-Audy JC. Protective transverse loop colostomy associated with low colo-rectal anastomoses. Int J Colorectal Dis. 1997;12:340–1. doi: 10.1007/s003840050120. [DOI] [PubMed] [Google Scholar]
- 53.Wang JY, You YT, Chen HH, Chiang JM, Yeh CY, Tang R. Stapled colonic J-pouch-anal anastomosis without a diverting colostomy for rectal carcinoma. Dis Colon Rectum. 1997;40:30–4. doi: 10.1007/BF02055678. [DOI] [PubMed] [Google Scholar]
- 54.Camilleri-Brennan J, Steele RJ. Prospective analysis of quality of life after reversal of a defunctioning loop ileostomy. Colorectal Dis. 2002;4:167–71. doi: 10.1046/j.1463-1318.2002.00352.x. [DOI] [PubMed] [Google Scholar]
- 55.O’Leary DP, Fide CJ, Foy C, Lucarotti ME. Quality of life after low anterior resection with total mesorectal excision and temporary loop ileostomy for rectal carcinoma. Br J Surg. 2001;88:1216–20. doi: 10.1046/j.0007-1323.2001.01862.x. [DOI] [PubMed] [Google Scholar]
- 56.Tornqvist A, Blomquist P, Jiborn H, Zederfeldt B. The effect of diverting colostomy on anastomotic healing after resection of left colon obstruction. An experimental study in the rat. Int J Colorectal Dis. 1990;5:167–9. doi: 10.1007/BF00300412. [DOI] [PubMed] [Google Scholar]
- 57.Mansson P, Fork T, Blomqvist P, Jeppsson B, Thorlacius H. Diverting colostomy increases anastomotic leakage in the rat colon. Eur Surg Res. 2000;32:246–50. doi: 10.1159/000008771. [DOI] [PubMed] [Google Scholar]
- 58.Bielecki K, Grotowski M, Kalczak M. Influence of proximal end diverting colostomy on the healing of left-sided colonic anastomosis: an experimental study in rats. Int J Colorectal Dis. 1995;10:193–6. doi: 10.1007/BF00346217. [DOI] [PubMed] [Google Scholar]