Abstract
Abstract
Background:
Although laparoscopy is becoming the standard of care for the treatment of colorectal disease, its application in case of postoperative peritonitis is still not widespread. The objective of this article is to evaluate the role of laparoscopy in the management of postoperative peritonitis after elective minimally invasive colorectal resection for malignant and benign diseases.
Materials and Methods:
Between April 2010 and May 2016, 536 patients received primary minimally invasive colorectal surgery at our Department. Among this series, we carried out a retrospective study of those patients who, having developed signs of peritonitis, were treated with a laparoscopic reintervention. Patient demographics, type of complication and of the main relaparoscopic treatment, and main outcomes of reoperation were recorded.
Results:
A total of 20 patients (3.7%) underwent relaparoscopy for the management of postoperative peritonitis, of which exact causes were detected by laparoscopy in 75% as follows: anastomotic leakage (n = 8, 40%), colonic ischemia (n = 2, 10%), iatrogenic bowel tear (n = 4, 20%), and other (n = 1, 5%). The median time between operations was 3.5 days (range, 2–8). The laparoscopic reintervention was tailored case by case and ranged from lavage and drainage to redo anastomosis with ostomy fashioning. Conversion rate was 10% and overall morbidity was 50%. No cases required additional surgery and 30-day mortality was nil. Three patients (15%) were admitted to intensive care unit for 24-hour surveillance.
Conclusion:
Our experience suggests that in experienced hands and in hemodynamically stable patients, a prompt laparoscopic reoperation appears as an accurate diagnostic tool and an effective and safe option for the treatment of postoperative peritonitis after primary colorectal minimally invasive surgery.
Introduction
L
Despite the abovementioned benefits, the rate of anastomotic leakage (AL) after the implementation of laparoscopic colorectal (CR) surgery remains similar to that after open surgery, ranging from 1% to 39%.4–6 This complication contributes not only to postoperative morbidity and mortality but also to local recurrence and poor prognosis in case of malignancy.7,8 Moreover, AL together with other complications after CR surgery, such as iatrogenic bowel perforation or colonic ischemia, may lead to peritonitis, a worrying condition that must be promptly recognized and treated.
On this basis, to date, some studies have shown that laparoscopy for postoperative peritonitis may be feasible and safe,9–11 especially if the reintervention occurs after a primary minimally invasive (MI) colorectal surgical procedure.12–19 However, this approach might be challenging and requires well-versed laparoscopic surgeons. For these reasons, laparoscopy for the treatment of peritonitis after CR surgery is not commonly applied.
The aim of this study is to assess safety, feasibility, and efficacy of relaparoscopy (RL) in case of postoperative peritonitis after elective MI colorectal resection.
Materials and Methods
MI colorectal surgery is the standard of care at our Department and includes the laparoscopic and robotic approaches (da Vinci® Si™; Intuitive Surgical, Inc.): the choice of the robotic technique is based solely on the availability of the robotic system and of the surgeon. For right colectomy or colonic flexure resection, an intracorporeal isoperistaltic side-to-side mechanical anastomosis is usually performed; for left-sided or rectal resections, after splenic flexure takedown, a standard Knight–Griffin anastomosis is created and is always tested for air tightness. We routinely place a drain only for rectal resection, and an ileostomy is usually fashioned in case of low rectal anastomosis, especially after neoadjuvant radiochemotherapy. Moreover, an Enhanced Recovery After Surgery (ERAS) protocol is carefully followed. 20
From April 2010 to May 2016, a total of 691 patients underwent elective resections with primary anastomosis for benign or malignant CR disease within ERAS protocol: 536 patients (77.5%) received MI surgery (laparoscopic n = 426, robotic n = 109). Among this series, we carried out a retrospective study on those patients who, having developed clinical and radiological signs of peritonitis (i.e., acute abdominal pain, leucocytosis, elevated C-reactive protein, and computed tomography findings of intra-abdominal air or liquid), were treated with a laparoscopic reintervention by 4 CR surgeons with extensive laparoscopic experience. All included patients were hemodynamically stable.
Candidates were identified in a prospectively collected database where all clinical information on patients, who gave their consent, is entered using predefined parameters, including all patient characteristics, clinical outcomes, and follow-up.
Data on patient demographics and findings of the primary surgery were retrospectively reviewed. In all reoperated patients, time elapsed from primary surgery and from the onset of peritonitis/symptoms to RL, type and duration of RL, morbidity, length of hospital stay, and readmission were all considered topics. Conversion, defined as laparotomy or extension of the extraction site incision, was also recorded.
When AL was intraoperatively detected, it was considered as minor if the anastomotic suture integrity was interrupted <30% of its length. Morbidity was defined as any complication occurring within 30 days from the day of reintervention. Long-term outcomes were achieved at regular outpatient appointments.
Results
Patient characteristics
A total of 20 patients (3.7%) underwent RL for postoperative peritonitis after elective MI colorectal surgery (laparoscopic n = 19, robotic n = 1) at our Department between April 2010 and May 2016. Patients' data are presented in Table 1. Indication to RL was established on the clinical findings and preoperative investigations; the interval until reintervention from primary MI surgery was a median of 3.5 days (range, 2–8) and it was performed after discharge from the primary surgery in 2 of 20 patients (10%). Median time elapsed from the onset of symptoms and RL was 8 hours (range, 1–12).
ASA, American Society of Anesthesiologists; BMI, body mass index.
Characteristics of reintervention
The diagnosis of peritonitis was confirmed during laparoscopy in all patients (100%) by the finding of puruloid or fecal fluid in the peritoneal cavity. The exact cause of peritonitis was established in 75% of patients (15 of 20) and the intraoperative findings are reported according to the type of primary surgery in Table 2. In the majority of cases (n = 8, 40%), RL was required following anterior resection (AR) for cancer. AL represents the most common cause of reintervention (n = 8, 40%) and has been mainly detected after primary left-sided resection (n = 4, 50%).
Leakage of ileal anastomosis performed during low anterior rectal resection for iatrogenic small bowel tear.
AL, anastomotic leak; MIS, minimally invasive surgery; RL, relaparoscopy.
All reoperations were fully laparoscopic. Under general anesthesia, access to the abdominal cavity was achieved by open introduction of Hasson cannula through the previous incision site, usually the periumbilical one. After pneumoperitoneum was established, other trocars were inserted through the previous port sites under direct vision.
In all patients, a careful abdominal exploration and adhesiolysis were carried out by means of atraumatic graspers. Peritoneal collections were promptly aspired and a sample of the material was sent to the laboratory for testing. A peritoneal lavage was implemented extensively with warm saline solution and at least one drain was placed. To check the integrity of CR anastomosis, an air test was always performed. 21
Main surgical treatment varied according to intraoperative findings, the degree of peritoneal contamination, and the type of primary MI surgery (Table 3):
• AL (n = 8): after right colectomy, anastomotic repair represented the treatment of choice (n = 1). In case of left colectomy or AR (n = 7), an ostomy was always fashioned (except in 1 case) and represented the only procedure in case of minor breakdown (n = 4); • Colonic ischemia (n = 2): a redo anastomosis was carried out intracorporeally in both cases together with a diverting ileostomy when this complication occurred following primary AR; • Iatrogenic bowel tears (n = 4): a direct repair with interrupted sutures was implemented in all cases; • No findings (n = 5): 3 of 5 patients (60%) were treated with peritoneal lavage and drainage. In the remaining 2 patients, an extracorporeal redo ileocolic anastomosis and a loop ileostomy were attempted, respectively: in the first case, it occurred because of surgeon's preference, while in the second case, it was due to the type of primary surgery (AR). • Other (n = 1): a robotic AR with loop ileostomy was performed in a patient with rectal cancer after neoadjuvant therapy. During the primary surgery, a manual end-to-end ileal anastomosis was carried out up to the ileostomy due to an iatrogenic small bowel tear. This patient presented peritonitis due to the leakage of the ileal anastomosis; only in this case, after a careful laparoscopic abdominal exploration, the Pfannenstiel incision, used for specimen retrieval at the first surgery, was reopened to perform a redo ileal anastomosis.
Converted patient.
Leakage of ileal anastomosis performed during low anterior rectal resection for iatrogenic small bowel tear.
AL, anastomotic leak; MIS, minimally invasive surgery.
Reoperations were completed laparoscopically in 18 of 20 cases (90%). Causes of conversion to open were an unclear reason of peritoneal contamination associated with severe adhesions, following a right colectomy, and an additional small bowel injury during bowel mobilization after primary left colectomy (where a small bowel tear was already identified as the cause of peritonitis), respectively.
Clinical outcomes
RL outcomes are summarized in Table 4. Intraoperative complications occurred in 2 patients (10%) and consisted in two iatrogenic small bowel perforations: in 1 case, it was repaired with laparoscopic suture, while in the other patient, it required conversion to laparotomy because a primary small bowel tear was already identified as the cause of peritonitis. Sepsis was treated according to Rivers' Protocol, 22 and 3 patients (15%) were admitted to intensive care unit (ICU) for 24-hour surveillance.
Leakage of ileal anastomosis performed during low anterior rectal resection for iatrogenic small bowel tear.
Following reintervention.
AL, anastomotic leak; ICU, intensive care unit; L, liters; LOS, length of stay; OT, operative time; NGT, nasogastric tube.
Overall morbidity was 50%: in particular, 1 patient presented an abdominal collection treated with a percutaneous drainage, and 2 patients, after an ileostomy fashioning for AL after rectal resection, presented a small presacral sinus and were managed conservatively. No cases required additional surgery and 30-day mortality was nil. Thirty-day hospital readmission occurred in 2 patients (10%) because of dehydration secondary to high output stoma and pneumonia, respectively.
Bowel continuity was restored in 7 of 8 patients carrying diverting stomas after a median of 10 months (range, 3–15) following RL (stoma closure rate 90%); in 1 patient, who is still receiving adjuvant chemotherapy, stoma will be closed at the end of the treatment. At a median follow-up of 41 months (range, 8–74), 1 of 2 converted patients presented a midline incisional hernia and underwent surgery. No patient was lost to follow-up.
Discussion
In this retrospective study of prospectively collected data, we present our monocentric experience of RL for postoperative peritonitis following elective MI colorectal surgery. The feasibility and safety of this approach are demonstrated by the satisfying intraoperative outcomes in our series. In addition, short- and mid-term outcomes are acceptable.
Laparoscopy in the last decade has gained widespread acceptance as a diagnostic and therapeutic tool.9,23,24 It has been shown to be safe and effective for the management of postoperative complications after open and laparoscopic surgery, avoiding diagnostic delay and unnecessary laparotomy. 25 Moreover, in case of peritonitis, as shown by some reports,9,11,26,27 this approach provided favorable outcomes, especially in terms of hospital stays and postoperative complications; compared with open reintervention, laparoscopy might reduce the additional trauma, systemic stress response,25,28 and postoperative septic complications.29,30
In the field of CR surgery, although laparoscopy has been increasingly adopted for the treatment of benign and malignant diseases, to the best of our knowledge, few promising experiences have been reported in literature about its applications for the management of postoperative complications.31–33 Even fewer articles have been published at present about the use of RL for peritonitis after a primary MI colorectal surgery either caused only by AL12,13,17–19 or by mixed complications.14–16 In experienced hands, laparoscopy is an effective tool and can guarantee all the well-known advantages of MI surgery in selected patients who develop peritonitis: this is realized especially in terms of reduced abdominal wall or wound complications and a tendency to faster recovery than with relaparotomy.12,13,17 Moreover, an early laparoscopy plays a pivotal role in the diagnosis of the complication, especially in case of unclear radiological evaluation, and thus might reduce peritonitis-related mortality. 16
In our series, peritonitis was confirmed in all 20 patients (100%), but we identified the exact cause of peritonitis in a slightly lower percentage (75%) in comparison with other studies where the diagnostic accuracy of RL is reported to range from 84.8% to 98.6%.9,34,35 However, unlike other experiences, we considered the laparoscopic diagnosis to be correct only if the exact origin of pathology causing peritonitis was identified. In our study, a proper diagnosis was not established by laparoscopy mainly after right colectomy (n = 4, 80%); in those patients, who belong to our early experience, we believe that peritoneal contamination during intracorporeal anastomosis fashioning might have been the cause of the peritonitis.
Median time to RL from primary surgery and from the onset of acute abdominal pain was 3.5 days (range, 2–8) and 8 hours (range, 1–12), respectively. Our preliminary experience has taught us that the sooner the surgery is performed, the greater the chances of finding a less distended bowel, low inflammatory adhesions, and a localized peritonitis in a patient who is still likely to be hemodynamically stable.
We report an ICU admission rate of 15% with a maximum of 24-hour surveillance; considering the limited available literature reporting these outcomes, our data are consistent with the reduced rate and shorter length of ICU stay described by three experiences comparing laparoscopic versus open reoperations.13–15 In our series, 30-day complication rate was 50%. A literature comparison has some strong limitations because there are little available detailed data and different sample size populations; however, against this background, our rate is slightly higher than that reported by Rotholtz's and Cuccurullo's experiences (6% 15 and 25%, 16 respectively). On the other hand, contrary to other series,14–16 in our experience, no wound complication was reported, especially no patient required an additional reintervention and 30-day mortality was nil.
As other authors have reported,12–14,16 we have always established the pneumoperitoneum through previous incision sites by an open entry technique to minimize the risk of bowel injury, keeping the benefits of an MI approach, and to reduce wound problems and incisional hernia. Irrespective of treatment strategy, an abundant peritoneal lavage with a warm isotonic saline solution was always carried out. Actually, this issue is controversial; indeed, a lavage with <10 L of saline solution was shown to decrease bacterial counts in diluted peritoneal fluid, but did not reduce postoperative infectious complications 36 that (on the contrary) were decreased with a mean lavage volume of 25 L. 37
In our series, RL has been successfully carried out in 18 of 20 patients (90%), and only in 1 of 2 cases, conversion to laparotomy was due to severe adhesions. As already reported,38,39 patients undergoing MI surgery have a lower risk of developing clinically significant adhesions and might have a lower tendency to suffer a small bowel obstruction at a longer follow-up. This issue plays an important role, especially when RL is planned.
The first most common cause for RL in our series was AL (n = 8, 40%), and in 6 cases, it was treated with a temporary stoma that so far was closed in 5 patients since 1 patient is still receiving adjuvant chemotherapy. Contrary to other published series,16,17 no Hartman's procedure or end ileostomy has been performed for AL. Our total stoma closure rate (90%) is consistent with data reported by other experiences.12,13 As recently reported by Lee et al., laparoscopy was shown to potentially reduce the risk of having a permanent stoma. 12
Conclusion
A limitation of our study is its monocentric, retrospective, not comparative, and not randomized design; in addition, our group size is pretty small. Nonetheless, our preliminary data suggest that in a center of high MI experience, RL represents an accurate diagnostic tool and a safe approach for the management of postoperative peritonitis after primary MI colorectal surgery, with acceptable morbidity. Further high quality studies are needed to confirm our findings.
Footnotes
Disclosure Statement
No competing financial interests exist.
