CT indicates computed tomography; PRA, primary resection and anastomosis; SCANDIV, Scandinavian Diverticulitis.
aIncludes all patients undergoing operations for perforated diverticulitis, continuously reported data by all participating centers, and patients identified through diagnostic code search and review of patient files at the end of the study.
bPerforated diverticulitis diagnosed at surgery but not suspected preoperatively.
cSee Table 1 for explanation of Hinchey classification stages.
dPrimary outcome assessment only, otherwise excluded from analysis
eTable 1. Participating Hospitals
eBox 1. Items Covered by CRF for the Index Admission
eBox 2. Items Covered by CRF for 3-Month and 1-Year Follow-up
eTable 2. Patients With False Preoperative Diagnosis
eTable 3. Diverticulitis Patients Deceased Within 90 Days of Surgery
eTable 4. Mixed-Effects Model Analysis
eTable 5. Logistic Regression of Possible Confounding Factors
Johannes Kurt Schultz, Sheraz Yaqub, Conny Wallon, Ljiljana Blecic, Håvard Mjørud Forsmo, Joakim Folkesson, Pamela Buchwald, Hartwig Körner, Fredrik A Dahl, Tom Øresland, . Laparoscopic Lavage vs Primary Resection for Acute Perforated DiverticulitisThe SCANDIV Randomized Clinical Trial. JAMA. 2015;314(13):1364–1375. doi:10.1001/jama.2015.12076
Perforated colonic diverticulitis usually requires surgical resection, which is associated with significant morbidity. Cohort studies have suggested that laparoscopic lavage may treat perforated diverticulitis with less morbidity than resection procedures.
To compare the outcomes from laparoscopic lavage with those for colon resection for perforated diverticulitis.
Design, Setting, and Participants
Multicenter, randomized clinical superiority trial recruiting participants from 21 centers in Sweden and Norway from February 2010 to June 2014. The last patient follow-up was in December 2014 and final review and verification of the medical records was assessed in March 2015. Patients with suspected perforated diverticulitis, a clinical indication for emergency surgery, and free air on an abdominal computed tomography scan were eligible. Of 509 patients screened, 415 were eligible and 199 were enrolled.
Patients were assigned to undergo laparoscopic peritoneal lavage (n = 101) or colon resection (n = 98) based on a computer-generated, center-stratified block randomization. All patients with fecal peritonitis (15 patients in the laparoscopic peritoneal lavage group vs 13 in the colon resection group) underwent colon resection. Patients with a pathology requiring treatment beyond that necessary for perforated diverticulitis (12 in the laparoscopic lavage group vs 13 in the colon resection group) were also excluded from the protocol operations and treated as required for the pathology encountered.
Main Outcomes and Measures
The primary outcome was severe postoperative complications (Clavien-Dindo score >IIIa) within 90 days. Secondary outcomes included other postoperative complications, reoperations, length of operating time, length of postoperative hospital stay, and quality of life.
The primary outcome was observed in 31 of 101 patients (30.7%) in the laparoscopic lavage group and 25 of 96 patients (26.0%) in the colon resection group (difference, 4.7% [95% CI, −7.9% to 17.0%]; P = .53). Mortality at 90 days did not significantly differ between the laparoscopic lavage group (14 patients [13.9%]) and the colon resection group (11 patients [11.5%]; difference, 2.4% [95% CI, −7.2% to 11.9%]; P = .67). The reoperation rate was significantly higher in the laparoscopic lavage group (15 of 74 patients [20.3%]) than in the colon resection group (4 of 70 patients [5.7%]; difference, 14.6% [95% CI, 3.5% to 25.6%]; P = .01) for patients who did not have fecal peritonitis. The length of operating time was significantly shorter in the laparoscopic lavage group; whereas, length of postoperative hospital stay and quality of life did not differ significantly between groups. Four sigmoid carcinomas were missed with laparoscopic lavage.
Conclusions and Relevance
Among patients with likely perforated diverticulitis and undergoing emergency surgery, the use of laparoscopic lavage vs primary resection did not reduce severe postoperative complications and led to worse outcomes in secondary end points. These findings do not support laparoscopic lavage for treatment of perforated diverticulitis.
clinicaltrials.gov Identifier: NCT01047462
Perforated colonic diverticulitis is common, having an incidence of 3 to 4/100 000 people per year, is highly morbid, and can be lethal.1,2 The severity of perforated colonic diverticulitis is proportional to the degree of abdominal contamination (Table 1),3 and the extent of disease can be assessed by computed tomographic (CT) imaging.4,5 Depending on what is found as a result of the CT imaging, treatments will differ. If the perforation results in a contained abscess, it can be managed with percutaneous drainage and intravenous antibiotics. If there is perforation with uncontained purulent or fecal contamination, surgery is required.6,7 The standard operative approach is to resect the perforated segment of colon along with as much diverticula-diseased colon as possible. In general, the colon is not anastomosed back together and a colostomy is created. During the last 2 decades, surgeons have been restoring colonic continuity with a primary anastomosis when the peritonitis is not too severe.7,8
Resection for perforated diverticulitis has a high morbidity rate (30%-50% complications) and a mortality rate of 10% to 20%.9- 11 Laparoscopic approaches to treating surgical conditions are generally associated with lower morbidity than open approaches and have been tried for the treatment of perforated colonic diverticulitis.12,13 Laparoscopic lavage without resection seems feasible because on some occasions, perforations are already sealed when surgery is performed and the only intervention that seems necessary is to lavage the abdominal cavity to remove debris. This approach, when performed laparoscopically, was reported to have lower rates of morbidity and mortality. However, the series showing this were likely affected by selection bias, ie, with optimal patients selected for the new procedure and more severely ill patients being treated with open colon resections, making the laparoscopic approach appear to have better outcomes.14
Only a randomized clinical trial (RCT) can overcome selection bias and definitively answer the question of whether laparoscopic lavage is an adequate operation for perforated diverticulitis. To our knowledge, only 1 such trial has been performed. However, it involved a small number of patients (N = 90), was terminated early, and may have been underpowered to definitively answer this question.15,16 In the present RCT, we tested the hypothesis that laparoscopic lavage was superior to colon resection in patients with perforated diverticulitis and purulent peritonitis, with respect to severe postoperative complications within 90 days of surgery.
The Scandinavian Diverticulitis trial (SCANDIV) was designed as a 2-group, open-labeled, pragmatic, superiority, multicenter RCT. A total of 21 surgical units were included in the trial (9 in Sweden and 12 in Norway), ranging from small community hospitals to tertiary referral centers. These units covered a catchment population of approximately 5 million people (eTable 1 in Supplement 1). A protocol draft was written by the principal investigators and distributed to all potential participant hospitals in Scandinavia. After a thorough review, all participating hospitals consented to the final protocol (Supplement 2), which was made accessible at the study web page. The trial was approved by the regional ethics committee in southeastern Norway (reference number 2009/177) and by the regional ethical review board in Stockholm, Sweden (protocol 2010/3:2, number 2010/113-31/3).
Between February 5, 2010, and June 28, 2014, we enrolled patients older than 18 years who were suspected clinically of having perforated diverticulitis requiring urgent surgery. Participating hospitals joined the study at different times (eTable 1 in Supplement 1). Inclusion criteria included patient ability to tolerate general anesthesia and diagnostic imaging results (via an abdominal CT scan) showing free air and findings compatible with perforated diverticulitis (usually including colonic wall thickening and pericolic inflammation). The indication for surgery was presence of clinical peritonitis. The decision to operate was made by the surgeon in charge, a position that varied between hospitals from senior surgical residents to colorectal attending surgeons. Exclusion criteria were bowel obstruction and pregnancy. Written informed consent was obtained by the surgeon before study enrollment. In Sweden, consent to undergo surgery and for study enrollment was provided by the next of kin for some patients who could not self-provide consent (approved by the Swedish ethical board). In Norway, patients not capable of providing consent were not enrolled.
Patients were randomly assigned to undergo either laparoscopic lavage or colon resection (Figure). Patients were not informed about the randomization result until after surgery. A center-stratified block randomization was achieved with a web-based randomization and data collection system, developed and administered by the Unit of Applied Clinical Research, Institute of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway. The first block per center was 2 + 2 and the subsequent blocks were randomly assigned 1 + 1 or 2 + 2. Neither participants nor investigators were informed about the block sizes. A web-based case report form (CRF) was completed for all participants during the hospital stay and at 3 and 12 months postoperatively. The items covered by the CRF are accessible online (eBox 1 and eBox 2 in Supplement 1).
All patients were administered intravenous antibiotics, according to local practices, after a diagnosis of peritonitis was established. Because the trial design was pragmatic, the intention was that the surgeons who normally operate on these patients would do so in the study. There was no requirement for minimum number of cases performed before joining the trial. All surgeons had basic laparoscopic skills. Surgical techniques were not monitored other than the monitoring surgeon reviewing the operation report. All patients were marked preoperatively for potential stoma placement on the left and right lower abdominal quadrants. Preexisting comorbidities were assessed and scores for an American Society of Anesthesiologists (ASA) physical status and Charlson Comorbidity index were determined.17 A Hinchey grade was assigned after the abdomen was entered. In this study, an assessment of a Hinchey grade III was defined as free air detected on a preoperative CT scan and an intraoperative finding of localized or generalized pus in the abdominal cavity. Irrespective of preoperative randomization, the Hartmann procedure was performed on all patients with feculent peritonitis (Hinchey grade IV), including patients with a visible defect in the colon wall. If conditions other than perforated diverticulitis were found during abdominal exploration, the surgeons, using their clinical judgement, treated for that condition.
In the resection group, the choices of laparoscopic vs open resection, and also of Hartmann procedure vs primary resection and anastomosis (PRA) were determined by surgeon preference and local practices. The diseased colon segment was resected to the level of the rectum (determined as that part of the colon not having taenia coli), with or without mobilization of the splenic flexure. The Hartmann procedure was performed by closing the rectum with a stapling device and marking the blind rectal pouch with a nonabsorbable suture. A colostomy was created through the stoma site, which was marked preoperatively. For PRA, the choice of whether to protect the anastomosis with a diverting stoma or not was determined by the surgeon’s discretion. In all cases, at least 1 drain was placed in the pelvis (either suction or nonsuction per the surgeon’s discretion).
In laparoscopic lavage, pneumoperitoneum was preferably obtained by an open transumbilical technique with a 12-mm trocar, using at least 2 additional 5-mm trocars for abdominal access. All quadrants were rinsed before placing a nonsuction drain on each side of the pelvis. Adhesions to the sigmoid were not to be dissected.
In both groups, the time to drain removal was determined by the surgeon. According to the protocol, the abdominal cavity in all patients was rinsed with at least 4 L of saline or until drainage was clear. All resected specimens were examined by a pathologist.
Follow-up at 3 months and 1 year postoperatively was performed at an outpatient clinic, by a telephone interview, or with a form sent by mail. The protocol required that a colonoscopy be performed within 3 months after surgery for patients treated with laparoscopic lavage, and before stoma reversal for patients operated with the Hartmann procedure. Patients with incorrect preoperative diagnosis were included in the follow-up for the primary outcome, but otherwise excluded from the study.
The primary outcome variable for this study was having severe complications at 90 days after surgery. All complications classified greater than Clavien-Dindo IIIa were defined as severe; ie, any complications that led to a reintervention requiring general anesthesia (IIIb), a life-threatening organ dysfunction (IV), or death (V).18 Secondary outcomes were the length of operating time; length of postoperative hospital stay; postoperative complications individually, including those resulting in reoperations; and 90-day postoperative quality of life (measured using the Cleveland Global Quality of Life instrument).19 The Cleveland Global Quality of Life questionnaire includes 3 items (current quality of life, current quality of health, and current energy level) each scaled from 0 (worst) to 10 (best). The total calculated score ranges from 0 (worst) to 1 (best). We considered a change of 0.1 as clinically important.
To ensure the accuracy of the collected data, a colorectal surgeon from a hospital different than the one in which the study procedure was performed reviewed patient files and completed a monitor CRF. In case of discrepancies between the original CRF and the monitor CRF, the main study coordinator requested a settlement between the monitoring surgeon and the local study coordinator. The final decision was made by the monitor. To ensure conformity between groups, the evaluation of Hinchey grading was a substantial part of the monitoring process. Evaluations were based on criteria including free air on CT and free pus in the abdomen (Hinchey III), and fecal contamination or visible hole in the colon (Hinchey IV) (Table 1). An independent radiologist, who was blinded to patients’ randomization, reevaluated all CT scans obtained during each patient’s enrollment in the study.
As part of a local quality control effort, participating hospitals registered data for all patients who underwent operations for perforated diverticulitis during the study period. These data were reviewed, but only at the center level, because they included patients who were not in the trial and who had not consented.
A safety committee was appointed, consisting of 3 Scandinavian professors (in colorectal surgery) who were not involved in the study. The safety committee, which completed a safety analysis after 75 patients were enrolled, was supplied with all patient data including patient characteristics and primary and secondary outcomes. The trial was to be terminated if there was a statistically significant difference in primary outcome between the study groups with a P value less than .01.
The sample size was based on 80% statistical power and a total 2-tailed type 1 error α = .05 distributed to α = .01 for the safety analysis and α = .045 for the final analysis. Previous meta-analyses for resection surgery in perforated diverticulitis have shown mortality rates9 ranging between 10% and 20%, and morbidity rates 10 between 20% and 30%, while Myers et al12 reported a 3% mortality rate and a 4% morbidity rate following laparoscopic lavage. Being aware of the probable selection bias in earlier trials, we deliberately estimated a smaller difference in primary outcome than could be expected. We postulated that the proportion of severe complications within 90 days would be 30% in the resection group vs 10% in the laparoscopic lavage group and calculated a sample size of 130 patients (65 in each group). To account for patients with free fecal contamination of the peritoneal space (Hinchey grade IV) who could not be treated with the trial intervention, we planned to include 150 patients. In May 2013, the SCANDIV Study Group agreed to slightly increase the sample size because of the unexpected high number of included patients with incorrect preoperative diagnoses. Inclusion was to be continued until at least 65 patients with Hinchey grades I, II, or III had undergone treatment per protocol in each study group.
Statistical analysis was performed on an intention-to-treat basis using IBM SPSS Statistics version 21. We included the 2-sided Fisher exact test for proportions and a nonparametric test (Mann-Whitney) for continuous variables. The hypothesis testing was 2-sided with a significance level of .045, and for all other analyses, it was .05. The 95% CIs for differences in proportions for the primary outcome were calculated with the Wilson score method without continuity correction, according to Newcomb.20 Primary outcome analysis was performed according to intention-to-treat principles. A modified intention-to-treat analysis was performed, which included patients who at the time of surgery were believed to have perforated diverticulitis and no other diseases such as cancer (macroscopically evaluated). Secondary outcomes were assessed without correction for multiple comparison. All participants, including crossover patients and those with protocol violations, were analyzed strictly in the groups to which they were randomized. An analysis for possible site effects was performed using mixed-effects models. Additionally, logistic regression was used to explore the relationship between various clinical features and outcomes. Patients who were lost to follow-up were excluded from all analyses. Patients with missing data were excluded only from analysis of the particular item.
The 90-day follow-up data on the last enrolled patient were obtained on December 12, 2014, and data monitoring was completed in March 2015. A total of 199 patients were randomly assigned to undergo laparoscopic lavage (n = 101) or primary resection (n = 98). Participating hospitals enrolled between 1 and 27 patients (Table 2). In the resection group, 1 patient was lost to follow-up and 1 was randomized without consent (Figure). Both patients were excluded from the intention-to-treat analysis because of lack of primary outcome data. For 12 patients in the lavage group and 13 patients in the resection group, perforated diverticulitis was not found at the time of surgery (eTable 2 in Supplement 1).
Of 415 eligible patients, 216 were not enrolled. The main reasons for patients not enrolling were because of the on-call surgeon not asking (n = 162), previously diagnosed advanced malignancies or severe comorbidities (n = 35), and patients declining to participate (n = 13). Of the 94 patients who did not meet the inclusion criteria but who underwent operations for perforated diverticulitis during the study period, most were preoperatively thought to have another cause of peritonitis (n = 43) or were unable to provide informed consent (n = 24), thus they were not enrolled (Figure and Table 2). Demographic characteristics of these patients are reported in Table 3. The center-specific operation rate for perforated diverticulitis varied between 1.2 and 3.8 patients per 100 000 inhabitants per year (Table 2). The monitor CRF was completed for 97% of the patients with perforated diverticulitis. Discrepancies from the original CRF, mostly of minor importance, were found in approximately 18% of the listed items. The major disagreements between monitors and investigators were on Hinchey grading; as a result, the grade was changed in 17% of patients. The radiologist’s review of the CT scans, which formed the basis for inclusion, showed extraluminal air in all patients. In 16% of the cases, there were only air bubbles in close proximity of the diseased colon. Baseline characteristics were similar between the groups (Table 4). The protocol allowed for a laparoscopic approach to resection in the resection group, which was performed in 3 patients.
The primary outcome of severe complications within 90 days after surgery did not statistically differ between patients in the laparoscopic lavage group (30.7% [31 of 101]) and patients in the resection group (26.0% [25 of 96]) (difference, 4.7% [95% CI, −7.9% to 17.0%]; P = .53; Table 5). The rate of death was 13.9% (14 patients) in the laparoscopic lavage group and 11.5% (11 patients) in the resection group (difference, 2.4% [95% CI, −7.2% to 11.9%]; P = .67). In contrast to our hypothesis, the observed severe complication and mortality rates were lower in the resection group than in the laparoscopic lavage group. In the modified intention-to-treat analysis of patients with macroscopic findings of perforated diverticulitis, the rate of severe complications in the laparoscopic lavage group was 31.5% (28 of 89 patients) vs 21.7% (18 of 83 patients) in the resection group (difference, 9.8% [95% CI, −3.5% to 22.5%]; P = .17), and 90-day mortality in the laparoscopic lavage group was 13.5% (12 of 89 patients) vs 8.4% (7 of 83 patients) in the resection group (difference, 5.1% [95% CI, −4.7% to 14.7%]; P = .34; Table 5 and eTable 3 in Supplement 1). Among patients without fecal peritonitis, the target population of this trial, statistically significant differences were only observed in the secondary outcomes (Tables 5 and 6). Considerably more patients in the laparoscopic lavage group (15 of 74 [20%]) than in the resection group (4 of 70 [6%]) required a secondary surgical procedure because of complications (difference, 14.6% [95% CI, 3.5% to 25.6%]; P = .01; Table 6). The main reasons for reoperation included secondary peritonitis in the lavage group and wound dehiscence in the resection group. Furthermore, intraabdominal infections (secondary peritonitis and abscesses) occurred more frequently in the laparoscopic lavage group. Primary resection was associated with more superficial wound infections, longer operating times, and a higher proportion of patients with a stoma 90 days after primary surgery.
Analysis of the primary outcome with mixed-effects models, including random effects for centers, showed an estimated center-level variance of 0 (eTable 4 in Supplement 1). With different combinations of fixed effects (age, sex, ASA score, body mass index [BMI, calculated as weight in kilograms divided by height in meters squared]), the estimated center-level random effect was still 0 in all cases. Logistic regression of other possible confounding factors, including age, sex, ASA score, BMI, center catchment population, center size (number of operations), and whether the most experienced surgeon on the team was a specialist, did not affect outcomes as shown by logistic regression analysis. The only statistically significant predictor was the ASA score, with higher scores predicting worse outcome, as would be expected (eTable 5 in Supplement 1).
Three patients in each group underwent oncological resection rather than a diverticulitis procedure when colon cancers were found at the time of the study procedure. These patients were excluded from the modified intention-to-treat analysis. Six patients with sigmoid cancers that were not detected during the study procedure were retained in the modified intention-to-treat analysis because cancer diagnosis was not known at the time of surgery (4 patients in the laparoscopic lavage group [diagnosed within 90 days after surgery by reoperation or colonoscopy] and 2 patients in the resection group [1 diagnosed when specimen was opened and 1 diagnosed during histopathological examination of the specimen]). In 48 of 63 patients (76%) treated with lavage who were alive but did not undergo reoperation after 3 months, a postoperative colonoscopy or CT colonography was performed (missing data n = 3). In 21 of these patients (33%), the colon was examined within 3 months and in another 15 of these patients (24%) within 6 months after surgery. A specialist in colorectal surgery was involved in more operations in the resection group.
Some patients with signs of generalized peritonitis and free air on the CT scan had minimal contamination of the abdominal cavity (Table 6). Patients with Hinchey I and Hinchey II grades were all treated according to randomization. One of these in the lavage group underwent reoperation because of a missed sigmoid carcinoma. In the lavage group, an abscess was drained in the patient with a Hinchey II grade; the patients with Hinchey I grades (free air and localized inflammation only) were treated with lavage only. In all patients with fecal peritonitis (Hinchey grade IV), irrespective of randomization, the diseased bowel was resected according to the study protocol. Procedures and outcomes for these patients are displayed in Tables 5 and 7 After reviewing the data of the first 75 included patients, the safety committee concluded that it was safe to continue the trial.
No statistically significant difference was found between the lavage and resection groups for the 90-day primary outcome of having severe complications. Statistically significant differences between the 2 study groups were only observed for secondary outcomes. Laparoscopic lavage was associated with a greater need for reoperations. However, patients in the laparoscopic lavage group had shorter operating times, less blood loss, and fewer patients had a stoma after 3 months. Because these findings were not adjusted for multiple comparisons, they should be interpreted as exploratory.
This did not confirm findings from previous non RCTs suggesting that laparoscopic lavage was associated with better outcomes than colon resection procedures.13,21 Results similar to ours were recently published in a preliminary report from a small Swedish-Danish study.22 Our results are similar to those of the LOLA (Laparoscopic Peritoneal Lavage or Resection for Generalised Peritonitis for Perforated Diverticulitus [Ladies]) trial, which was terminated early and only enrolled a small number of patients. Both the SCANDIV and LOLA trials found the need for more reoperations in the laparoscopic lavage than the resection group.15 SCANDIV was completed, reached the intended sample size, and included considerably more patients in fewer centers than LOLA.
Our methodology differed from other lavage studies that randomized patients after laparoscopic Hinchey staging.15,22 The preoperative allocation in the present trial resulted in the inclusion of patients with Hinchey grade IV and patients with diagnoses other than perforated diverticulitis. However, making the decision to randomize after inspecting the abdominal cavity risks creating selection bias because clinicians might deem some patients unsuitable for the study. Preoperative randomization in our current study resulted in similar numbers of patients with feculent peritonitis and number of incorrect preoperative diagnoses between the 2 study groups.
Our findings do not exclude the possibility that there are harms associated with laparoscopic lavage for the treatment of purulent perforated diverticulitis. When the modified intention-to-treat analysis was restricted only to patients with this clinical problem, the difference in serious complication rates between the lavage and resection procedures was 11.4% (95% CI, −1.8% to 24.1%; P = .10; Table 5). With the lower bound of the CI being reasonably close to zero, it is possible that a study with more patients would have concluded that more harm than good could result from the lavage of purulent perforated diverticulitis as compared with treatment by colon resection. Based on the results of the current study, approximately 300 patients in each group are needed to adequately study this possibility.
To date, 6 RCTs have been published on surgical approaches for perforated diverticulitis,15,22- 26 2 of which investigated the role of laparoscopic lavage.15,22 The largest trial before SCANDIV included only 105 patients.25 Four of these trials were prematurely terminated,15,23,24,26 limiting the conclusions that can be drawn from them.16
Strengths of SCANDIV include its generalizability, because many surgeons were involved and they operated at 21 different hospitals ranging from community hospitals to academic medical centers. We also had a high degree of enrollment, with nearly 50% of all eligible patients enrolled in the study. We were able to obtain clinical information from the patients with perforated diverticulitis treated in the participating hospitals who were not enrolled in the study.
A limitation of SCANDIV is that patients not enrolled in the study had a numerically higher ASA score, reflective of having more severe disease and worse postoperative outcomes. Thus, SCANDIV’s results may not pertain to patients with perforated diverticulitis who are very ill.
Despite the strict inclusion criteria, the spectrum of abdominal contamination varied greatly within the groups. However, the prerequisite for inclusion was a need for urgent surgery based on clinical findings, which usually included clinical signs of generalized peritonitis. The determination of the Hinchey grade was not as straightforward as expected. We could not rule out the possibility of some discrepancies in grading. This was the main reason patients with all Hinchey grades were included in the modified intention-to-treat analysis of the primary outcome.
The study design was pragmatic and compared a standard, well-established treatment for perforated diverticulitis (resection) to a newer approach, laparoscopic lavage. Both of these treatments are reasonably common in clinical practice and most surgeons are very familiar with technical aspects of these operations. The study protocol for SCANDIV precluded dissection of inflammatory adhesions to the sigmoid colon during the laparoscopic lavage procedure. These adhesions might have influenced the study’s results in 2 ways. The adhesions might have served to seal small perforations or to conceal larger defects in the colon that would never close and, therefore, require resection. Missed malignancies were another major concern associated with laparoscopic lavage. In SCANDIV, 12 patients had a perforated colon carcinoma. Leaving 4 carcinomas in the laparoscopic lavage group unresected may have impaired the prognosis for these patients. Distinguishing between diverticulitis and cancer was not always possible when performing urgent operations for what was thought to be diverticulitis. Because of the relatively high rate of missed colon carcinomas in the lavage group, it was essential to perform a colonoscopy after a patient recovered from the perforation.
Among patients with likely perforated diverticulitis undergoing emergency surgery, the use of laparoscopic lavage compared with primary resection did not reduce the rate of severe postoperative complications and led to worse outcomes in secondary end points. These findings do not support laparoscopic lavage for treatment of perforated diverticulitis.
Corresponding Author: Johannes Kurt Schultz, MD, Department of Gastrointestinal Surgery, Akershus University Hospital, PB 1000, 1478 Lørenskog, Norway (email@example.com).
Author Contributions: Dr Schultz had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Schultz, Yaqub, Wallon, Forsmo, Folkesson, Buchwald, Körner, Dahl, Øresland.
Acquisition, analysis, or interpretation of data: Schultz, Yaqub, Wallon, Blecic, Forsmo, Folkesson, Buchwald, Körner, Dahl, Øresland.
Drafting of the manuscript: Schultz, Yaqub, Øresland.
Critical revision of the manuscript for important intellectual content: Schultz, Yaqub, Wallon, Blecic, Forsmo, Folkesson, Buchwald, Körner, Dahl, Øresland.
Statistical analysis: Schultz, Dahl.
Obtained funding: Schultz, Yaqub, Øresland.
Administrative, technical, or material support: Schultz, Yaqub, Wallon, Blecic, Forsmo, Folkesson, Buchwald, Körner, Dahl, Øresland.
Study supervision: Schultz, Yaqub, Wallon, Forsmo, Folkesson, Buchwald, Körner, Dahl, Øresland.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
SCANDIV Study Group Collaborators, Sweden: András Papp, MD, Urban Ersson, MD, Tilman Zittel, MD, Niklas Fagerström, MD, Dan Gustafsson, PhD (Hudiksvalls Hospital, Hudiksvall); George Dafnis, PhD (Eskilstuna County Hospital, Eskilstuna); Maria Cornelius, MD (Helsingborg Hospital, Helsingborg); Monika Egenvall, PhD, Per Olof Nyström, PhD (Karolinska University Hospital, Stockholm); Ingvar Syk, PhD, Dadi Vilhjalmsson, MD (Skåne University Hospital, Malmö); Lars Påhlman, PhD (Uppsala University, Uppsala); Gunnar Arbman, PhD (Vrinnevis Hospital, Linköping University, Norrköping); and Abbas Chabok, PhD (Västmanland Hospital, Västerås). SCANDIV Study Group Collaborators, Norway: Merete Helgeland, BSc (Akershus University Hospital, Lørenskog); Johan Bondi, PhD (Bærum Hospital, Vestre viken HF); Anders Husby, MD (Diakonhjemmet Hospital, Oslo); Ronny Helander, MD (Drammen Hospital, Vestre viken HF, Drammen); Arnulf Kjos, MD, Hilde Gregussen, MD (Innlandet Hospital, Hamar); Aras Jamal Talabani, MD, Gerd Tranø, PhD (Levanger Hospital, North-Trøndelag Hospital Trust, Levanger); Inge Holm Nygaard, MD (Molde Hospital, Helse Møre og Romsdal, Molde); Gro Wiedswang, PhD, Ole Helmer Sjo, PhD (Oslo University Hospital, Oslo); Kari Festøy Desserud, MD (Stavanger University Hospital, Stavanger); Stig Norderval, PhD, Mads Vikhammer Gran, MD (University Hospital of North Norway, Tromsø); Torkil Pettersen, MD, and Arild Sæther, MD (Østfold Hospital, Fredrikstad).
Funding/Support: The study was funded by the South-Eastern Norway Regional Health Authority with a fellowship awarded to Johannes Kurt Schultz (PNR 2719011). The study was also sponsored with research funding by Akershus University Hospital, which covered running expenses (PNR 2619041, 2629038, and 2649054).
Role of Funder/Sponsor: The funders of the study had no part in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We would like to thank the following individuals for their kind assistance: Berit Marianne Bjelkåsen, Cand Mag, Unit of Applied Clinical Research, Institute of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway; Claes Johansson, PhD, Danderyd Hospital, Stockholm, Sweden. Mitra Rashidi, MD, Oslo University Hospital, Oslo, Norway. Frode Reier-Nilsen, MD, Akershus University Hospital, Lørenskog, Norway. The Safety Committee: Rune Sjödahl, PhD, University of Linköping, Sweden; Arild Nesbakken, PhD, University of Oslo, Norway; and Søren Laurberg, PhD, University of Århus, Denmark. Computed tomography monitoring: Peter Mæhre Lauritzen, MD, Radiologist, Department of Clinical Research, Akershus University Hospital, Norway. Website support: André Øien, BSc, University of Oslo, Oslo, Norway. Valuable advice on the manuscript: Michael Bretthauer, PhD, University of Oslo, Oslo, Norway. None of these individuals received compensation in association with their work on this article.