2017 WSES guidelines for the management of iatrogenic colonoscopy perforation

There are a number of risk factors that have been related to ICP in the literature (Table 3). Older patients are more vulnerable to ICP, and the ages of 65, 75, and 80 years have been shown to be independent risk factors for ICPs [23, 26, 27]. Female gender [28, 29], low BMI [28, 30], low albumin level, the presence of comorbidities, diverticulosis, Crohn’s disease, and admission to an ICU are also acknowledged to be risk factors in several studies [20, 23, 26, 28]. The endoscopist’s level of experience may also be considered a risk indicator, as higher incidences of ICP have been reported for non-gastroenterologist endoscopists and low-volume endoscopy centers [31‐33]. Finally, anesthesia during colonoscopy has been associated with an increased risk of ICP, in relation to the worsening of patient’s comorbidities and the increasing technical complexity of these procedures [34, 35].

In a recent study of 56,882 colonoscopies, full-thickness large bowel perforation occurred in forty patients, corresponding to an incidence rate of 0.07% (0.05% in diagnostic/screening procedures and 0.17% in therapeutic colonoscopies) [18]. A greater risk of ICP was associated with low-volume practices, female gender (due to greater colonic length and a more mobile transverse colon), advanced age (reduced wall strength), history of diverticular disease, previous abdominal surgery (especially pelvic), and colonic obstruction (risk of over-insufflation).

In a Spanish study of 16,285 colonoscopies, ICPs were reported in 0.09% of cases [16]. Colonic obstruction, prior abdominal surgery, and sigmoid diverticular disease were indicated as potential risk factors.

A review from the Netherlands including 30,366 endoscopic procedures found that ICP occurred in 35 patients (0.12%) [5]. The authors described a 4-fold higher risk of ICP in colonoscopies compared with sigmoidoscopies and a 5-fold greater risk of ICP in therapeutic compared with diagnostic procedures.

A review of 10,486 colonoscopies performed in a single institution included 20 ICPs over a period of 10 years (corresponding to an incidence rate of 0.19%) [29]. During the same time interval, 46,501 flexible sigmoidoscopies were performed and only two ICPs occurred (0.004%). Female patients had significantly more ICPs compared with males and, although not statistically significant, the risk of ICP was numerically higher for endoscopists in training than experienced endoscopists [29].

In a review of studies published between 2001 and 2009 analyzing 969,913 colonoscopies [36], the incidence of ICP ranged from 0.032 to 0.14%. The risk factors for ICP included age over 75 years (4- to 6-fold increase), colonoscopy instead of sigmoidoscopy (2–4 times greater), female gender, diverticular disease, previous abdominal surgery, and multiple comorbidities, including diabetes mellitus, chronic pulmonary disease, congestive heart failure, myocardial infarction, cerebrovascular disease, peripheral vascular disease, renal insufficiency, liver disease, and dementia.

Therapeutic colonoscopies generally involved a higher risk for ICP, particularly the following procedures: polypectomy for large polyps, multiple polypectomies, pneumatic dilatation for Crohn’s stricture [37], the use of argon plasma coagulation, and endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) for colorectal neoplasia [38]. For endoscopic polypectomies, the related perforation risk has been related to the size of the polyp (larger than 10 mm in the right colon or 20 mm in the left colon) and a sessile morphology [38], and it is considered to be less than 1%, even when more challenging polypectomy techniques such as EMR are performed [39]. Complex procedures such as EMR and ESD are associated with a higher perforation incidence and should be considered to have a high risk of ICP. In 2014, a meta-analysis by Wang et al. comparing procedure-related complications in EMR and ESD for colorectal tumors (including 4 retrospective case-control studies) reported ESD-related perforations in 31/347 cases and EMR-related perforations in 33/566 cases [40]. The current literature demonstrates that the perforation risk for ESD is decreasing in higher volume centers to less than 5% [41, 42].

Perforation in colorectal stenting is the main early adverse event [43]. Use of a self-expandable metal stent (SEMS) has been associated with an overall perforation rate of 7–8% [10, 44]. In cases of acute malignant colonic obstruction, retrospective studies have shown an SEMS-related perforation risk of 5–9% [45]. Stenting of either benign or neoplastic strictures has been associated with a 7.4% incidence of ICP in a recent meta-analysis [43]; the type of stent, benign etiology, bevacizumab therapy, and the need for re-dilation have been identified as risk factors for ICP [44, 46, 47].

Endoscopic balloon dilation may entail perforation rates up to 11%, even though the rate of iatrogenic perforation for Crohn’s disease stricture treatment is less than 5% in the majority of retrospective studies [37, 45, 48]. Balloon dilation of rectal anastomotic strictures has been associated with a 1.1% rate of ICP [49].

The most common site of perforation is the sigmoid colon (53–65%), followed by the cecum, the ascending colon, the transverse colon, the descending colon, and the rectum [6, 13, 15, 29, 50] (Fig. 1). ICPs are generally intra-peritoneal perforations; extra-peritoneal perforations may manifest as pneumoretroperitoneum, pneumomediastinum, or subcutaneous emphysema. Combined intra- and extra-peritoneal perforations have been reported anecdotally [51].

There is only one randomized study concerning the risk factors and preventive measures for ICP, whereas several reviews of large clinical series and meta-analyses to define the incidence and risk factors for ICP have been published [52, 53]. Recommendations for preventive measures derive from these studies and expert opinions [54].

Colonoscopy has been demonstrated to be the most cost-effective method for colorectal cancer screening. As the number of procedures performed worldwide is increasing, gastrointestinal professional societies have adopted strict safety standards for endoscopic practice, including the monitoring and auditing of complications to detect performance gaps and continuously improve the safety of colonoscopy [55]. The American Society for Gastrointestinal Endoscopy (ASGE)/American College of Gastroenterology (ACG) Task Force on Quality in Endoscopy recommends that post-colonoscopy perforation rates should be maintained at ≤ 1 per 500 colonoscopies (≤ 1/1000 in screening healthy subjects) [56]. For screening colonoscopies, the European Society of Gastrointestinal Endoscopy (ESGE) proposes that perforation should require surgery in ≤ 1/1000 [57]. In an audit of post-colonoscopy complications before starting national colorectal cancer screening, the British Society of Gastroenterology (BSG) reported post-colonoscopy perforation rates of 1/769 over a total of 9223 colonoscopies [58].

Perforation during diagnostic or screening endoscopic procedures may occur from one of these two main pathways: (a) direct mechanical damage to the colonic wall by the tip or the side of the endoscope as it is pushed forward or (b) a pneumatic distension due to barotrauma (Table 4). Direct mechanical trauma is the most frequent etiology of ICP, and perforations originating from mechanical trauma are commonly large and located in the sigmoid region. The injury is usually produced by direct trauma due to an inaccurate instrumental insertion, colonoscope movements toward the mucosal surface, retro-flexion maneuvers, or excessive torsion. Indirect injuries can also occur as the consequence of bowing or stretching the distal part of the colon. The presence of redundant colon diverticula or adhesions from previous surgeries can increase the risk of mechanical trauma during colonoscopy [16]. Barotrauma is instead produced by the excessive distension of the bowel due to over-insufflation, which produces linear lacerations at the colonic wall that may evolve into full-thickness defects. This type of perforation is more frequently located at the cecal region, where the thinner muscular layer and the larger lumen diameter make this region more vulnerable to pressure-related injuries [6, 16, 59, 60]. For interventional endoscopies, the mechanism of perforation can be the same as those occurring during diagnostic endoscopy, or they may be due to thermal/electrical injury of the colonic wall, manifesting as a wall ischemia. In this latter case, the perforation can occur with a delay of 24–72 h [18, 54]. Wall damage can be incomplete and the perforation concealed as it is confined by the surrounding tissues. During the following days or weeks, an abscess may develop that may delay the diagnosis.

Up to 60% of ICPs are detected by the endoscopist while performing the procedure [14, 16, 18, 60‐62]. In a retrospective evaluation of a single institution, 68% of ICPs were identified on the day of endoscopy, 23% on day 1 or 2 after the endoscopy, and 9% were identified at least 2 weeks after the procedure [29]. The results of a survey of 30,336 colonoscopies showed a mean delay of 0.36 days for the diagnosis of ICP after diagnostic endoscopies and 1.5 days after therapeutic procedures [5].

A delay in the diagnosis of ICP is a critical issue for therapeutic outcomes; when the diagnosis is delayed more than 24 h, the chance increases that more invasive treatments (e.g., surgery) will be required [2, 63]. Physicians should therefore search for this potentially life-threatening complication and run clinical and biochemical tests if an ICP is suspected.

An ICP can be appreciated by direct visualization of the parietal defect or the view of intra-abdominal tissues through the colonic wall during the endoscopy [15]. Otherwise, the diagnosis of ICP is based on clinical, laboratory, and radiologic findings [64]. The clinical presentation of an ICP can vary widely, depending on the size of the perforation, the type of etiologic agent, the affected colonic location, the degree of intra-peritoneal contamination, and the patient’s general status. In the majority of patients (91–92%), symptoms develop within the first 48 h following the completion of the endoscopy [14, 29]. The most common symptom is abdominal pain associated with distension, although painless cases of ICP or cases with severe cramp-like pain have been described [13, 16, 18]. In two large clinical series, the most consistent symptoms were abdominal pain (from 74 to 95%), guarding/rebound tenderness (82.5) with diffuse peritonitis, tachycardia (62.5%), leukocytosis (40%), fever (38%), rectal bleeding (15%), and isolated abdominal distension (6.6%) [16, 18]. Only a small number of patients with ICP (5%) remained asymptomatic [52, 59]. An unusual clinical sign (1/55 patients with ICPs) was a delayed subcutaneous emphysema and an ongoing necrotizing infection of the abdominal wall [16, 18]. It is a common belief that patients with diffuse peritonitis can be diagnosed and treated for perforation on a clinical basis, but peritonitis-like clinical scenarios can also occur in the absence of perforation. For instance, a transmural thermal injury after polypectomy with serosal irritation without any obvious perforation produces localized peritonitis that is amenable to non-operative management. Thus, biochemical and imaging studies are always indicated when an ICP is suspected.

Laboratory tests should be run for inflammatory markers that can reveal severe bacterial infections associated with the perforation [65], such as white blood cell count (WBC) and C-reactive protein (CRP) [66, 67]. In case of delayed presentation (> 12 h), the pro-calcitonin level (PCT) can be useful for ICP diagnosis.

Perforations of intra-peritoneal segments of the colon (e.g., the cecum, transverse colon, or sigmoid colon) more often lead to free intra-peritoneal fluid and air (large amounts in cases of barotrauma from insufflation), whereas perforations of the ascending and descending colon and rectum or wall injuries contained in the supplying mesentery result mainly in extra-peritoneal air. Mixed situations are possible if the perforation is in the middle between an intra- and extra-peritoneal portion [68]. Upright or decubitus abdominal radiographs can detect small amounts of free peritoneal air, but they are insensitive to the presence of fluid. Plain thoracic and abdominal radiographs have a positive predictive value (PPV) of 92% for ICPs [13]. Of note, the PPV has been shown to be higher for ICPs occurring during diagnostic procedures (PPV 100%) than for ICPs occurring during therapeutic procedures (PPV 45%) [2]. Alternatively, an ultrasound may be useful in cases in which the radiation burden should be limited, notably in children and pregnant women. However, this method should not be considered definitive in excluding a pneumoperitoneum [69].

If the clinical suspicion of ICP persists after a normal plain radiograph, a computed tomography (CT) scan with contrast enhancement should be requested, as this imaging tool can easily detect small amounts of both free intra-peritoneal air and fluids, in some cases with the foci of the gas congregating near the perforation site [68]. Air trapped in the mesenteric folds is found in perforation of the colon. A pneumoretroperitoneum is caused by extraperitoneal perforations such as perforations of the descending colon and rectum. Gas in the right anterior pararenal space indicates ascending colon perforation, whereas gas in the left pararenal space indicates descending or sigmoid colon perforations. Generally, rectal perforation causes bilateral pneumoretroperitoneum [70]. For extra-peritoneal perforations, the CT scan can show air tracking along the mesenteric and fascial planes, even in the mediastinum and abdominal, and chest and neck walls. Of note, the retro-peritoneal air dissecting the mediastinum and the retropharyngeal tissues can cause a change in the tone of the larynx, resulting in voice change [71].

Colonoscopy may also dissect within the wall of the colon with pneumatosis. Moreover, mucosal injury and intraluminal pressure may dissect air inside the mesenteric and portal venous system. For all these reasons, CT is much more effective in the diagnosis of extraluminal air compared to conventional radiography [15]. Double contrast CT (intravenous and rectal) is increasingly used in patients with clinical suspicion of ICP and without diffuse peritonitis. This diagnostic tool may be useful for detecting concealed or sealed perforations that are eligible for non-operative management [72]. Multi-detector CT (MDCT) is superior to single helical or conventional CT because it can provide rapid, high-volume coverage, and diagnostic images, even in patients who are unable to perform prolonged breath holds. One study showed that MDCT was 86% accurate in predicting the site of perforation [69].

The following recommendations were developed using a large clinical series and expert opinions, since randomized studies on this topic are lacking.

Once the diagnosis of perforation is confirmed by clinical and radiological examinations, the decision between surgical and non-operative treatments will depend on the type of injury, the quality of the bowel preparation, the underlying colonic pathology, and the clinical stability of the patient [6]. However, a surgical consultation should be obtained in all cases of perforation [73].

Whenever the risk of a large perforation is present and the patient presents with signs and symptoms of peritonitis, the emergency surgery approach is reasonable and safe [6]. Surgical management is also recommended in patients with concomitant colonic diseases requiring surgery, transplanted patients, and immunosuppressed patients [36, 74]. In selected patients with localized pain, free air without diffuse free fluids in radiographs, hemodynamic stability, and an absence of fever, non-operative management (conservative) may be appropriate [61, 68, 75‐78] and is associated with low morbidity, low mortality, and short hospital stays. Conservative management is usually suitable for small, sealed-off perforations that occurred during a therapeutic colonoscopy in patients with an optimal bowel preparation [8, 23, 24].

Conservative treatment consists of serial clinical and imaging monitoring (every 3–6 h) with absolute bowel rest, intravenous fluids for hydration, intravenous administration of broad-spectrum antibiotics, and a close multidisciplinary team follow-up to promptly detect the development of sepsis and peritoneal signs [6, 78, 79]. Drainage of the peritoneal air through a Veress needle punction may be useful in relieving abdominal pain, improving respiratory function, and facilitating the closure of the perforation site [80]. The overall success rate of conservative treatments for colonic perforation varies from 33 to 90% [36].

An early success with non-surgical treatment does not rule out the potential need for surgery [52]. If the conservative treatment is successful, clinical improvement will gradually occur within 24 h, but a continuous and strict clinical and biochemical follow-up is recommended. In cases of clinical deterioration or progression to a septic condition or peritonitis, surgical treatment should not be delayed. The sole presence of subdiaphragmatic free air does not constitute an indication for urgent surgery. Of note, complication rates and lengths of hospital stay are significantly higher in patients who have undergone surgery after conservative management than in patients who were initially treated with surgery [81]. Indeed, when surgical treatment is delayed, the peritonitis and colonic wall inflammation could worsen, requiring a more invasive surgery that is associated with a poorer prognosis [13, 82]. Ideally, the decision to pursue surgery should be made as early as possible after the endoscopy [2].

Endoscopic treatment is possible when the perforation site is recognized intra-procedurally or within 4 h following the procedure and the bowel preparation is still adequate [45]. Urgent endoscopic therapy with clip placement and the use of CO₂ may limit the volume of extraluminal insufflation and subsequently the need for surgery [83‐85]. Endoscopic clip closure of ICP was first reported in the literature in 1997 [86]. Today, it should be considered a valuable non-invasive method for ICP that is recognized during a colonoscopy. It has been shown to be effective in sealing and healing the perforation and avoiding surgery in most cases [2]. The decision to perform the endoscopic closure of colonic perforation depends on the size and the cause of the iatrogenic damage as well as the endoscopist’s experience and the availability of appropriate endoscopic devices [45]. Clipping closure of ICP is recommended for small perforations (less than 1 cm) originating from either diagnostic or therapeutic colonoscopies [2, 24, 87], with a success rate of 59–100% [2, 4, 88, 89]. In larger or difficult perforations, a combination of endoclips and endoloops might be used. There are also few reports in the literature about closure with conventional clips for perforations larger than 1 cm [90‐92]. A limitation of the endoscopic closure is the difficulty of evaluating the completeness of the colonic closure after the clip application. This might result in delayed complications such as intra-abdominal abscesses, which can occur due to the persistence of intestinal fluids in the peritoneal cavity or an intermittent leakage [2].

Over the last several years, new devices have been introduced to widen the spectrum of possibilities of performing an endoscopic closure of a gastrointestinal perforation. Through-the-scope (TTS) clips and over-the-scope clips (OTSC) are both effective for the early closure of defects smaller than 2 cm, with overall technical and clinical success rates of 93 and 89%, respectively [88, 93‐95]. TTS clips are more suitable for closure of small therapeutic perforations (less than 1 cm), whereas OTSC may be used for larger defects. The OTSC is a nitinol clip shaped to mimic a trap that allows for the inclusion of more tissue and consequently closure of larger perforations than the conventional clips [96]. Recent studies focusing on the outcomes after OTSC placement revealed a rate of procedural success of 80–100% and clinical success rates of 57–100% [96‐98].

The overstitch endoscopic suturing device (Apollo Endosurgery, Austin, TX, USA) was recently developed and might play a role in the future ICP closures [99]. Partially or totally covered stenting could potentially allow closing the perforation, but data supporting its clinical application are still lacking. A clear indication for surgery in the setting of endoscopic treatment of an ICP consists of a complicated procedure or a failed endoscopic closure with an ongoing leak that is causing fecal peritonitis [45].

After a successful endoscopic closure, it is advisable that a multidisciplinary team, including abdominal surgeons, endoscopists, gastroenterologists, and anesthesiologists, are involved in the patient’s follow-up [52]. Fasting, broad-spectrum antibiotic therapy and intravenous hydration are the basis of treatment [3, 88, 100]. Close observation for signs of peritoneal irritation and monitoring of biochemical inflammatory parameters are crucial. When pain disappears and the inflammatory parameters and bowel function return to normal, oral intake can be resumed [100]. The duration of observation is subjective but obviously related to the patient’s status and the response to the conservative (non-operative) or endoscopic treatment. The mean duration of hospital stay after non-surgical ICP management ranges from 9 to 13 days [88].

There are no studies in the literature focusing specifically on the clinical and biochemical follow-up of patients who have undergone endoscopic closure or conservative management of ICP.

The available evidence is mainly supported by retrospective series. During the observation period, the patient treated for ICP should be monitored clinically as well as through laboratory values and imaging. Clinically, peritoneal signs such as tenderness, rebound tenderness, and muscle guarding, as well as signs of infection, such as fever, nausea, vomiting, abdominal distension, and diarrhea, should be recorded [36, 69]. Frequent assessment of the physical status and vital signs should be completed by laboratory tests for WBC, CRP, Hb, blood urea nitrogen, PCT, and electrolytes [66]. As an imaging technique, the CT scan remains the most accurate tool to be performed in case of clinical deterioration, especially when the need for surgery is in question and before discharge for non-operative treatments.

In patients who have undergone endoscopic repair of ICP, infection control is usually attained with a short-term course of antibiotic therapy (3–5 days). Antibiotics should be stopped if there are no signs of systemic inflammation and/or peritonitis after the short-term treatment. Considering the composition of the intestinal microbiota in the large bowel, patients with ICP require antimicrobial coverage for Gram-negative bacteria as well as for anaerobes. The potential infecting organisms in colorectal procedures are derived from the bowel lumen, where Bacteroides fragilis and other obligate anaerobes as well as Enterobacteriaceae such as Escherichia coli are the most common bacteria [101]. If there is any sign of an ongoing infectious process, antibiotics should be continued. An abdominal CT is recommended after 5–7 days to exclude residual signs of peritonitis or abscess formation and to exclude the possible need for a surgical intervention.

The duration of antimicrobial therapy in patients with complicated intra-abdominal infections has been debated. Antibiotic therapy should be shortened in those patients demonstrating a positive response to treatment. A prospective trial published recently by Sawyer et al. demonstrated that, in patients with complicated intra-abdominal infections undergoing an adequate source-control procedure, the outcomes after approximately 4 days of fixed-duration antibiotic therapy were similar to those after a longer course of antibiotics that extended until after the resolution of physiological abnormalities [102].

Sepsis is associated with activation of blood coagulation (hypercoagulability) contributing to venous thromboembolism (VTE) [103‐105]. Patients with abdominal sepsis may be at increased risk of VTE due to their premorbid conditions, surgical intervention, admitting diagnosis of sepsis, and events and exposures such as central venous catheterization, invasive tests and procedures, and drugs that potentiate immobility. A prospective cohort study using the National Surgical Quality Improvement Program database of the American College of Surgeons (ACS-NSQIP) was designed to evaluate the impact of preoperative sepsis on the risk of postoperative arterial and venous thrombosis. The study included 2,305,380 adults who underwent a range of surgical procedures [106]. The systemic inflammatory response syndrome was defined by the presence of two or more of the following: temperature > 38 or < 36 °C; heart rate > 90 beats/min; respiratory rate > 20 breaths/min or a PaCO₂ < 32 mmHg (< 4.3 kPa); white blood cell count > 12,000 cells/mm³, < 4000 cells/mm³, or > 10% immature band forms; or anion gap acidosis (> 12 mEq/L). Among all surgical procedures, patients with preoperative systemic inflammatory response syndrome or any sepsis had three times the odds of having an arterial or venous postoperative thrombosis. The risk of thrombosis increased with the severity of the inflammatory response and was higher in both emergent and elective surgical procedures. Thus, patients with ICP should be considered at risk, and thromboprophylaxis should be recommended.

There are no prospective clinical trials assessing the necessary duration of fasting following non-operative management or endoscopic repair of ICP. In the setting of conservative treatment, the general recommendations called for “bowel rest,” but the duration is unclear. Retrospective studies reported fasting durations of between 2 and 6 days. In one of the largest series, 24 patients with ICP were managed with conservative treatment, which failed in 3 patients; 31 patients were initially clipped, of which 22 procedures were successful. Poor outcomes were related to patient age, ASA status, and failure of conservative treatment. The only significant predictor of failure of the conservative treatment was the perforation size. Fasting duration did not appear to impact the outcomes [81].

Park et al. [69] reported a single-center series on ICP including 15 patients managed with either conservative treatment (n = 4) or endoscopic repair (n = 11) and compared these patients with 35 patients managed surgically. The duration of fasting was significantly shorter in the non-surgery group than in the surgery group (3.8 vs. 5.6 days). The mean fasting time was also 1 day shorter for patients treated by endoscopic repair versus surgery in the study by Kim et al. [4]. Moreover, the fasting duration was not related to ICP treatment failure.

It has been suggested that a clear liquid diet can begin immediately after the endoscopic repair of ICP; the evidence is not strong, but there are no data to indicate that this practice is not feasible or unsafe [36]. Following open or laparoscopic repair of ICP, there is no restriction on oral intake, as supported by numerous studies that provided enteral nutrition in the early period after colorectal surgery [107].

Surgery is indicated as the first treatment in patients with ongoing sepsis, signs of diffuse peritonitis, large perforations, and failure of conservative management and in the presence of certain concomitant pathologies, such as unresected polyps with high suspicion of being a carcinoma [6, 60, 78].

The peri-operative morbidity and mortality related to surgery for ICP are considerable, with rates of 21–44% and 7–25%, respectively [5, 13‐18]. Particularly frail patients, such as older patients and patients with low preoperative blood pressure, can have higher risks of mortality associated with colorectal perforation [108]. Thus, appropriate patient selection and surgical procedures are crucial in limiting the morbidity and mortality related to surgery for ICP.

In general, intraoperative findings determine the best technique to apply according to the different scenarios. Surgical procedures for the management of ICP include colorraphy, wedge resection, colostomy by exteriorization of the perforation, and colonic resection with or without primary anastomosis or stoma. The decision regarding the type of surgical procedure depends on (a) the size, location, and etiology of the ICP; (b) the viability of the surrounding colon and mesocolon; (c) the degree of and time from the development of peritonitis; (d) the patient’s general status and the presence of comorbidities; (e) the quality of the colonic preparation; and (f) the presence of residual lesions not resected during the colonoscopy procedure [2, 8, 24, 60, 82, 109, 110].

The decision of which procedure to perform, therefore, depends on many variables, and it must be made after a careful inspection of the whole colon and peritoneal cavity. Explorative laparoscopy should be considered a minimally invasive technique useful for performing both diagnostic and potentially therapeutic procedures. A timely application of explorative laparoscopy may prevent ongoing inflammation and injury that would necessitate more invasive measures, such as open laparotomy and/or colonic diversion [82]. The use of laparoscopy allows for visualizing the parietal defect and its size and specific location, as well as for identifying the potential cause of the perforation (e.g., perforation caused by the shaft of the endoscope, cautery, presence of mesenteric hematomas, emphysema, or effusions), which, as previously stated, are the main factors influencing the choice of treatment option. Early diagnosis is mandatory, and when timely management is ensured, laparoscopy can be the best option, offering reduced morbidity and length of stay and faster postoperative recovery. If no underlying lesion requiring surgical resection is seen during the endoscopy, the size of the tear is small, and the colon is healthy and well perfused, then a laparoscopic primary repair can be safely performed [52, 111].

Moreover, laparoscopic exploration allows the presence of potential signs of peritonitis to be evaluated and eventually allows aspiration, culture, and irrigation of the peritoneal cavity to be performed. Indeed, peritoneal washout and drainage have gained acceptance in the treatment of more advanced cases of colonic infection, such as Hinchey grade 2–3 diverticulitis [112]. Accordingly, the treatment of less advanced inflammatory processes, such as ICP, seems reasonable and indicated.

To summarize, explorative laparoscopy is indicated:

Explorative laparoscopy has a significantly lower morbidity and mortality compared with explorative laparotomy in the emergency setting [121]: specifically, the reported postoperative complication rate is 18.2% for laparoscopy vs. 53.5% for laparotomy. The postoperative mortality rate is 1.11% for laparoscopy vs. 4.22% for laparotomy; and the need for further procedures is significantly lower for laparoscopy (1.11%) than for laparotomy (8.45%).

Explorative laparoscopy may not be indicated when there is:

The potential diagnostic/therapeutic value of explorative laparoscopy should also be compared with the role of a CT scan in the evaluation of ICP. There is no study in the literature focusing on whether explorative laparoscopy should be performed instead of CT scans in patients with highly suspected ICP. However, when comparing these two modalities for penetrating abdominal trauma, CT scans have a sensitivity/specificity rate of 95%/95%, whereas explorative laparoscopy can achieve a 67–100% sensitivity and 50–100% specificity [121]. Thus, a CT scan should be performed in all cases before contemplating explorative laparoscopy, with the only obvious impediment being hemodynamic instability.

Thanks to the improvements in minimally invasive surgery, the laparoscopic approach has been increasingly used in recent years, and it should currently be considered a safe and feasible technique for the management of ICP [9, 24, 82, 113, 124‐126]. Current literature comparing outcomes of laparoscopy versus laparotomy for the treatment of ICP is scarce and consists mainly of small retrospective studies. The first relevant study was published in 2008 [110] and compared the perioperative outcomes between laparoscopic and open procedures for ICP by including only primary colonic closures without diversion. The authors found fewer complications and a shorter length of hospital stay for the patients in the laparoscopic group [110]. Other studies by Rothold et al. [125] and Schloricke et al. [127] also observed fewer postoperative complications and significantly shorter hospital stays when utilizing the laparoscopic approach. Similar studies with similar results were published by Coimbra et al. [124] and Kim et al. [128], although in these studies, delayed surgeries (> 24 h) and ostomy formation rates were more frequently observed in the open groups, with higher primary repair rates in the laparoscopic groups.

Due to its favorable short-term outcomes, laparoscopy should be considered the preferred approach for both exploration and repair of ICPs that are not manageable with medical treatments. However, the surgeon’s experience and skills are the key factors limiting the applicability and feasibility of laparoscopic ICP management. Conversion from laparoscopy to laparotomy should be considered whenever necessary. The most frequent reasons for conversion are the inability of the surgeon to complete the procedure laparoscopically, the large size of the ICP defect, the extensive peritoneal contamination, the highly inflammatory or neoplastic conditions of the colon, and the patient’s hemodynamic instability.

The choice of the surgical approach and technique mainly depends on the underlying pathology (e.g., colon cancer, diverticulitis) and the size of the ICP. Primary surgical repair can be used if the colonic tissue appears healthy and well vascularized and if suturing the perforation edges could be performed without tension [24, 113]. Wedge resection is feasible if it does not imply an excessive narrowing of the colonic lumen (e.g., cecum) [108]. Whenever the perforation is too large, the edges appear devitalized, or an avulsion of the adjacent mesocolon is seen, colonic resection might constitute the best option. Generally, patients who undergo surgery within 24 h are more appropriate candidates for less invasive techniques, such as primary suturing of the defect or linear wedge resection. In cases of delayed surgery (> 24 h from the colonoscopy), extensive peritoneal contamination, important comorbidities, or a deterioration of the general status of the patient (i.e., sepsis), a staged repair or colostomy by exteriorization of the perforation (e.g., double-barreled colostomy) must be considered [36, 52].

Currently, there are no prospective or retrospective studies in the English literature comparing the different types of repair (primary suture or wedge resection vs. segmental resection). Therefore, the choice of the surgical technique appears to be mainly empirical, and it is left to the surgeon’s discretion according to the intraoperative findings. Independent of the surgical approach, the main goal of the therapy is the rapid diagnosis, repair, and prevention of abdominal sepsis. If an ICP is to be repaired laparoscopically, the operating surgeon and the surgical team should be comfortable with the laparoscopic techniques, such as mobilization of the colon and intracorporeal suturing. A clinical algorithm mainly based on the size of the perforation and the necrotic area was proposed in 1999 to assist in choosing which type of repair to perform [8]. The maximal size for sutured repair was set at 1 cm. Between 1 and 2.5 cm, a transverse tangential stapled resection was recommended, whereas above 2.5 cm, a segmental resection was indicated [8, 129]. The condition of the bowel to be repaired and the level of contamination and inflammation are the most important factors in determining whether the laparoscopic approach is safe [109]. Both sutured and stapled repair techniques seem to be safe and feasible to repair defects of up to 4 cm [82].

In case of perforated colon cancer, surgery must follow the oncologic principles of cancer resection.

The formation of a stoma is often included in the overall surgical strategy for the management of ICP. However, no randomized controlled trials or other high-level evidence trials exist to guide this operative decision in this specific indication. Case series of ICP report variable rates of stoma formation (up to 59.7%) [59, 114, 116, 126, 130]. As such, the formation of a stoma forms an adjunct to the overall treatment strategy for these patients.

The precise clinical or operative reasons for stoma formation are incompletely reported in the case series on ICP. Furthermore, these reports are generally limited by their largely retrospective study designs and low event numbers, complicating subgroup analyses. Notwithstanding these limitations, some authors have established increased stoma formation rates in patients with delayed diagnoses, significant peritonitis, and patients with left-sided perforations [114, 126]. Apart from these observations, the limited publications in this area infer that surgical judgment remains essential in the decision-making surrounding the formation of a stoma. Finally, no data exist to specifically address the type of stoma formation in ICP.

The placement of an intra-abdominal drainage after surgical management of an ICP can be justified by either the presence of peritoneal contamination or the early diagnosis of a potential bleeding or leakage of the repair used for the perforation (i.e., colorraphy, wedge resection, colonic resection) [131‐133]. There are no studies available in the literature focusing on the indications of abdominal drainage after successful surgical treatment of ICP. The decision is left to the discretion of the surgeon according to the ICP setting, the intraoperative findings, the type of surgical procedure performed, the adequateness of infection source control, and the patient’s general status [5, 14, 108].

At present, no study concerning ICP and damage control surgery (DCS) is available in the literature. However, once colonic perforation has occurred, the course of sepsis will develop independent of the underlying disease. Thus, to evaluate the use of DCS in cases of ICP, we could analyze the experience in similar settings, such as in perforated diverticulitis (PD), equating ICP to PD [134, 135].

Damage control is a surgical technique originally used in trauma surgery consisting of three stages: (1) an abbreviated initial laparotomy with the aim of controlling hemorrhage and contamination with temporary abdominal closure (TAC); (2) resuscitation until normal physiology is improved; and (3) return to the operating room after 24–72 h for definitive injury repair and abdominal wall closure [136‐138].

Untreated or misdiagnosed ICP can progress to peritonitis and sepsis, resulting in serious morbidity and a very poor prognosis. Notably, morbidity rates as high as 43% and mortality rates as high as 25% have been reported [17, 20, 36, 50, 60, 139]. Nearly one quarter of patients will receive a delayed diagnosis, with a 45% incidence of fecal peritonitis [140]. The resultant inflammatory process associated with peritonitis clearly limits the operative options, precluding a single-stage procedure and resulting in fecal diversion in 38% of patients with fecal peritonitis. Several studies reported that age > 67 years, ASA score, blunt injuries, poor bowel preparation, and steroids are risk factors for increased postoperative morbidity (Table 5) [20, 123, 141, 142].

Over the last decade, DCS has become a valuable technique in unstable patients with fecal peritonitis [36, 136, 143]. The potential progression of ICP in fecal peritonitis is as probable as it is in perforated diverticulitis. In accordance with the WSES guidelines for the management of acute left-sided colonic diverticulitis, DCS may be suggested for clinically unstable patients (severe sepsis/septic shock) [135]. Critically ill patients with severe sepsis, hemodynamically unstable patients with hypotension, and patients with myocardial depression combined with coagulopathy are not candidates for endoscopic treatment or immediate complex operative interventions. In such patients, DCS allows rapid source control, enhances physiologic optimization, improves primary anastomosis rates, and decreases the need for stoma formation [144]. Therefore, in patients with abdominal sepsis, the application of DCS is individualized but not routinely used, as suggested by current clinical guidelines [145], stressing the importance of a careful assessment by the surgeons. Clearly, an individual approach tailored to each patient’s clinical status might be the most appropriate. In cases of ICP, DCS should be performed in combination with the resection of the perforated colonic segment to bridge the patient to the definitive injury and colonic continuity repair. DCS can represent a very resource-heavy procedure for institutions, however, because of the requirements for access to facilities (operating rooms and intensive care units) and committed staff.

At present, there are no studies in the literature focusing on the indications and timing for surveillance endoscopy after successful ICP treatment. However, based on the available evidence and clinical experience, a surveillance colonoscopy may be performed based on the initial indication (e.g., benign or malignant pathology) and type (e.g., screening or interventional) of the primary colonoscopy (during which the ICP occurred) and considering the risk-benefit ratio of performing an endoscopic exam [146, 147].

Colonoscopy is specifically contraindicated in cases of known or suspected perforation [148]. Consequently, any endoscopy after ICP treatment should be performed once the colonic wall has completely healed. Assuming that the healing time after ICP treatment is comparable to that after surgical sutures or anastomosis, a surveillance endoscopy may be indicated after approximately 3 months from the successful ICP treatment, depending on the size of the perforation and the type of repair [149].

In general, prior to any surveillance colonoscopy, it is necessary to carefully re-evaluate the presence of specific conditions favoring perforation, including increasing age, female gender, low BMI, intensive care unit stay, inpatient setting, diverticular disease [150], Crohn’s disease [30], obstruction as an indication for the primary colonoscopy, and invasive interventional colonoscopy [26]. Indeed, colonoscopy is contraindicated whenever the risks for the patient’s health or life are judged to outweigh the most favorable benefits of the procedure [148].