Efficacy of colonoscopy after an episode of acute diverticulitis and risk of colorectal cancer

Introduction

Colonic diverticular disease is a pathology with a high prevalence throughout the world, being most prevalent in developed countries, and whose incidence is increasing. It is estimated that 5-10% of the population >45 years old and approximately 80% of those >85 years old suffer this pathology [1-4]. The economic impact due to episodes of acute diverticulitis (AD) has been estimated at approximately $3,500-12,800 for each patient admitted, depending on the length of stay and the type of treatment performed [5].

Its pathogenesis is still not fully known, but could be related to changes in the structure and resistance of the colonic wall, due to alterations in intestinal motility or diets lacking in fiber, which cause herniation of the mucosa and submucosa leading to diverticula formation [1]. Most patients with this pathology are asymptomatic, but approximately 25% of them may suffer an episode of AD, which can be complicated by the formation of abscesses, fistulas or colonic perforation [2,4].

Historically and currently, the guidelines recommend performing a routine colonoscopy 4-6 weeks after the acute episode of AD to confirm the diagnosis and to exclude the presence of advanced adenomas (AA) and colorectal cancer (CRC). However, these recommendations are based on old studies, when computed tomography (CT) was not used to diagnose AD [6]. In the last few years, some studies have attempted to demonstrate that it is not necessary to perform a routine colonoscopy in all patients with AD, because the incidence of AA and CRC in uncomplicated AD patients (UAD) is low and almost similar to that found in CRC population screening, whereas it would be necessary in patients with complicated AD (CAD) [6,7].

Given the controversy between the guidelines and the latest published studies, we decided to perform this study to assess the real incidence of CRC after an episode of AD, both complicated and uncomplicated, found by colonoscopy, and to determine whether a colonoscopy should really be necessary for all patients with AD.

Patients and methods

We conducted a retrospective cohort study that included adult patients treated in the Surgery Department at the Valladolid University Clinic Hospital, Spain, between July 2016 and December 2017, with a diagnosis based on clinical (abdominal pain, fever) and analytical (leukocytosis, elevation of C-reactive protein [CRP]) data, and confirmed by CT during a first episode of AD. We excluded patients with a history of CRC, as well as those who underwent colonoscopy before the episode of AD and patients who had no follow-up colonoscopy. UAD was defined as a thickening of the colonic wall and/or increased density of pericolonic fat, ≤Ia according to the modified Hinchey classification (mHinchey), whereas CAD was associated with the presence of abscesses, fistula, intestinal occlusion and/or pneumoperitoneum, mHinchey ≥Ib.

The Hinchey classification has traditionally been used to distinguish 4 stages of perforated diverticular colonic disease; Hinchey et al published it in 1978 [8]. Since the introduction of CT, the Hinchey classification has been modified based on radiological findings. The modification by Kayser et al [9] was used in this study. The Hinchey classification as modified by Kayser et al includes 6 stages based on radiological findings: stage 0, diverticulitis ± colonic wall thickening; stage Ia: colonic wall thickening with pericolic soft tissue changes; stage Ib, Ia changes + pericolic or mesocolic abscess; stage II, Ia changes + distant abscess (generally deep in the pelvis or interloop regions); stage III, free perforation with purulent peritonitis; and stage IV, the same findings as stage III plus fecal peritonitis [10,11].

The colonoscopy was performed during follow up between 6 and 12 months. The diagnosis of CRC was based on the pathological result of the colonoscopy biopsy or the surgical specimen. The study was approved by the local institutional review board, clinical research ethics committee, Valladolid University Clinic Hospital, Spain.

Results

A total of 174 patients were diagnosed clinically and radiologically with AD. Forty-six patients were excluded because colonoscopy was not performed during follow up. Finally, we analyzed a total of 128 patients. The median age of the patients in our study was 67.5 years, and 72 (56.3%) patients were women. A total of 90 (70.3%) patients had UAD and 38 (29.7%) CAD. The location of the AD was 89.08% in the sigmoid colon, 9.36% in the descending colon and 1.56% in the ascending colon. Twenty-seven (21.1%) patients required surgery, performed through a laparoscopic approach in 10 (37%) patients (Table 1).

Table 1 Demographic, clinical and analytical characteristics of acute diverticulitis patients

The colonoscopy performed after the episode of AD showed lesions in 18 (14.06%) patients. Eight (6.25%) patients in the CAD group had 3 (2.34%) non-advanced adenomas (NAA) and 5 (3.9%) CRC. Ten (7.81%) patients with UAD had NAA and none had CRC (Fig. 1). No adverse events secondary to colonoscopy were evidenced.

Figure 1 Flow chart of patients through the study

The incidence of CRC, length of stay and CRP values were higher in patients with CAD (P=0.0001, P=0.002 and P=0.008 respectively) (Table 2). Patients with CRC were older and had lower hemoglobin values (P=0.002 and P=0.009 respectively). In addition, patients older than 70 had a higher incidence of CRC (P=0.009) (Table 3).

Table 2 Comparative outcomes of the main characteristics between complicated and uncomplicated acute diverticulitis

Table 3 Comparative outcomes of the main characteristics between group with and without colorectal cancer

Discussion

Colorectal cancer and colonic diverticular disease share multiple epidemiological characteristics. Both are very frequent pathologies in the Western world, with an incidence that increases with age, and a higher prevalence in men. In the west, the most frequent location of diverticular disease and CRC is the left colon [3,12].

Currently, most international clinical guidelines, as well as the American Gastroenterological Association [13] and the Society of American Gastrointestinal and Endoscopic Surgeons [14], recommend performing a routine colonoscopy or CT colonography after an episode of AD to exclude the presence of CRC. However, nowadays these recommendations are controversial, because they were based on old studies in which the diagnosis of AD was determined by clinical, analytical and imaging tests—such as barium enema, abdominal ultrasound, and CT—that were of inferior quality and resolution compared to current modalities [5,15-17]. In addition, invasive procedures are not exempt from complications and they entail a risk of morbidity and mortality. Colonoscopy is associated with a risk of colonic perforation in between 0.1-0.2% of patients and also a risk of producing iatrogenic diverticulitis [6,16].

Nowadays, CT has revolutionized the diagnosis and management of colonic diverticular disease. Multidetector CT can obtain thinner sections and has a better resolution and images of higher quality [6,17]. It is presently considered the best imaging technique for the diagnosis of AD and its complications, with a sensitivity and specificity of 94% and 99% respectively [4,18,19].

There are some studies that do not recommend routine colonoscopy in all patients with AD. Sharma et al performed a meta-analysis in which they included 11 studies, with a total of 1970 patients diagnosed with AD by CT; they found a higher CRC incidence in patients with CAD compared to those with UAD (10.8% vs. 0.7%, P≤0.001), with an overall CRC incidence of 1.6% [6]. Another systematic review performed by De Vries et al, in which 2490 patients with UAD were included, found that 11% of the patients had CRC and 2.2% AA; they concluded that performing a routine colonoscopy after an episode of UAD is not necessary unless there are signs of CRC [7]. There are also other more recent retrospective studies, such as those performed by Daniels, Horesh, Suhardja and Khoury, which also concluded that the incidence of CRC is not higher in patients with AD, so a routine colonoscopy should not be necessary [16,20-22].

A retrospective study carried out by Ramphal et al found that the CRC risk in patients with UAD without warning symptoms was lower than in those who present such symptoms (0.2% vs. 4.4%, P=0.0002), so they recommend that colonoscopy be performed only in patients with warning symptoms of CRC, such as lower gastrointestinal bleeding, constitutional syndrome and/or persistent abdominal pain [12].

However, there are also some studies that continue to recommend performing a routine colonoscopy. Meireles et al performed a retrospective study with 427 patients, and through colonoscopies they found lesions compatible with CRC in 20 (4.7%) patients. This finding was greater in complicated AD (10% vs. 3.5%, P=0.021), so they recommend that colonoscopies should continue to be performed after an episode of AD, especially in patients older than 50 years [1]. Soh et al also carried out a retrospective study in Asia that included 135 patients with UAD. Through colonoscopies they found 2 (1.5%) patients with AA and 2 (1.5%) with CRC. They advised that a follow-up colonoscopy should be performed in UAD patients [19].

In our study we found that the overall incidence of CRC was 3.9% and that of NAA was 10.15%, being slightly higher than in the previously mentioned studies, but not higher than in the colonoscopies of CRC population screening. Another significant finding was that all patients with CRC presented CAD and were older than 70 years (P=0.009).

The findings of this study have to be seen in light of some limitations. First, it was performed in a single center, it was a retrospective study and only a small number of patients were included. Despite these limitations and based on our findings, in agreement with the few existing studies, we can say that a routine colonoscopy should not be performed in all patients with AD. However, patients with CAD (mHinchey ≥Ib) and those older than 70 could benefit from undergoing a colonoscopy after an episode of AD, because the CRC incidence is higher in these cases.

To conclude, in our environment, routine colonoscopy in patients with UAD (mHinchey ≤Ia) has few benefits because the incidence of CRC is similar to that found through CRC population screening. However, the efficacy of colonoscopy is higher if it is performed in patients with CAD (mHinchey ≥Ib) and in those older than 70, so in these patients it would be advisable to perform it routinely.