Excessive Body Weight and Diverticular Disease

Abstract

Background

The worldwide proportion of overweight adults almost doubled from 22% in 1975 to 39% in 2016. Comparably, for the USA and Germany in 2016, the proportion was 68 and 56.8%, respectively. In Olmsted County, Minnesota, the prevalence of diverticulitis also doubled between 1980 and 2007, from 19 to 40%. Obesity substantially increases the risk of multiple gastrointestinal (GI) diseases and non-GI diseases. In a narrative review, we examined the evidence on whether obesity also increases the risk for the development of diverticulosis or diverticular disease and its outcome.

Summary

Evidence suggests that being overweight (body mass index ≥25 kg/m²) or obese (≥30 kg/m²), especially viscerally obese, is a risk factor for diverticulosis, diverticular disease and diverticulitis, diverticular bleeding, more severe or complicated disease, recurrent disease, and for worse outcomes after surgery.

Key Messages

There is a well-founded association between overweight and diverticular disease as well as diverticulosis. It is not clear whether overweight per se or confounders linked with it are responsible for the association. However, means to fight the overweight and obesity epidemic might also help to reduce the prevalence of morbidity and mortality from diverticular disease.

Introduction

Diverticulosis is one of the most frequent benign conditions of the colon − affecting <10% of people below 40 years of age, but mounts to about 30% up to age 60, and 50–70% in people beyond 80 years. Among patients with diverticulosis, about 20% will develop diverticular disease in their lifetime. Therefore, diverticular disease is among the most significant clinically and economically diseases in gastroenterology [1]. In the USA alone, diverticular disease accounts for 1,920,970 ambulatory visits, and 371,742 emergency department visits − up to 45% from 2006 to 208,015 hospital admissions, and 4,602 deaths every year [2]. The health-care expenditures attributable to colonic diverticulosis and diverticular disease are with USD 5.5 billion annually comparable to the USD 4.1 billion for colon cancer or USD 6.1 billion for cancer of other gastrointestinal (GI) organs [2].

Twin studies estimate that genetic factors contribute between 53 and 60% of susceptibility to diverticular disease [3, 4]. Genome-wide association studies point to diverticulosis primarily as a disorder of intestinal neuromuscular function and of impaired connective fiber support, while an additional diverticulitis risk might be conferred by epithelial dysfunction [5]. Nevertheless, around 40% of susceptibility for diverticular disease is due to environmental factors, including lifestyle and diet [6].

Obesity, defined as body mass index (BMI) of 30 kg/m² or higher, is one of those possible environmental factors. The worldwide prevalence of obesity has increased 27.5% for adults in the last 3 decades. Obesity substantially increases the risk of non-GI diseases, such as type 2 diabetes mellitus, hypertension, myocardial infarction, stroke, dementia, osteoarthritis, obstructive sleep apnea, and also a host of GI diseases, such as gastroesophageal reflux disease, erosive esophagitis, Barrett's esophagus, esophageal adenocarcinoma, erosive gastritis, gastric cancer, diarrhea, colonic polyps and cancer, liver disease, including nonalcoholic fatty liver disease, cirrhosis, hepatocellular carcinoma, gallstones, acute pancreatitis, and pancreatic cancer [7, 8, 9]. Our article reviews evidence on whether obesity also increases the risk of diverticulosis or diverticular disease and its outcome, respectively.

Obesity as Risk Factor for Diverticulosis

The first prospective cohort study (HPFS) examining the relationship between obesity and asymptomatic diverticulosis found no association in 47,228 male health professionals [10]. Comparing the group with the lowest (<21) and highest (≥30) BMI, the multivariate RR of 1.16 remained not statistically significant. Other anthropometric measures indicative of abdominal obesity, like waist circumference and waist-to-hip ratio, was also not significantly associated with asymptomatic diverticulosis in this cohort [10].

Wijarnpreecha and colleagues [11] undertook a meta-analysis on the topic including 10 studies with 53,520 participants published between 2011 and 2016. Meeting the eligibility criteria, 9 cross-sectional studies diagnosed diverticulosis by colonoscopy, and 1 prospective cohort study derived the presence of diverticulosis from diagnostic codes of the national hospitalization registry. Five studies came from countries of the Eastern [12, 13, 14, 15, 16] and Western world [17, 18, 19, 20, 21]. Results from the male health professionals' cohort [10] were not considered in the meta-analysis. When using a BMI ≥25 kg/m² as cutoff for obesity, the result is a significantly increased risk of colonic diverticulosis among obese subjects, with a pooled odds ratio (OR) of 1.41 (95% confidence interval [CI], 1.20–1.65). The OR in Western countries was higher (pooled OR 1.72 [95% CI, 1.30–2.27]) than in Eastern countries (pooled OR 1.11 [95% CI, 1.03–1.20]). Using the standard cutoff BMI ≥30 kg/m² in all Western world studies (except for the study of Peery 2012, which provided only data for BMI ≥25 kg/m²), resulted in a pooled OR of 1.43 (95% CI, 1.20–1.71). Nevertheless, the meta-analysis had several limitations, including: high statistical heterogeneity between studies; many studies could not establish a temporal relationship between obesity an diverticulosis due to the cross-sectional methodology; selection bias could favor inclusion of only positive studies [11]. Meanwhile, several other studies corroborate the general finding of an association between overweight/obesity and diverticulosis.

Peery and colleagues [22] reported in a cross-sectional study (N = 2,108) an increased odds of diverticulosis in overweight participants (BMI ≥25 kg/m²) compared to those with normal BMI (OR 1.65; 95% CI, 1.27–2.15). A positive association between BMI and diverticulosis was also reported in a study from Austria with 976 subjects [23]. Another cross-sectional study from the Massachusetts General Hospital (N = 243) found higher odds of diverticulosis among participants with BMI 25.0–29.9 kg/m² (AOR 3.02, 95% CI, 1.33–6.88) and BMI ≥30.0 kg/m² (AOR 4.43, 95% CI, 1.88–10.49), than individuals with normal BMI [24].

A recent study from the Chapel Hill group claims that obesity is a risk factor for diverticulosis only in women and not men. Furthermore, only postmenopausal women are at risk. The authors speculate that premenopausal ovarian steroid hormones may protect women from developing diverticulosis [25]. The study included 274 male and 349 female participants. Before 51 years of age, colonic diverticulosis was less prevalent in women (29%) than men (45%) (p = 0.06). With advancing age, this hiatus diminishes to 37% in women and 43% in men (p = 0.31) in the group 51–60 years old, and disappears after 60 years of age with 55% in women and 57% in men (p = 0.90). In the multivariate analysis, women with an obese BMI (≥30 kg/m²) had an increased risk of diverticulosis (PR 1.48; 95% CI, 1.08–2.04) compared to women with normal BMI (18.5–24.9 kg/m²). The strength of association increased with severity of obesity, and was strongest for women with BMI >40 kg/m² (PR 1.92; 95% CI, 1.92–2.84). Stratified by age category, women ≤50 of age (PR 1.99; 95% CI, 0.79–5.05) and 51–60 years of age (PR 1.63; 95% CI, 1.07–2.47) with an obese BMI (≥30 kg/m²) had an increased risk of diverticulosis compared to woman with a normal BMI. There was no association between BMI and risk of diverticulosis among women >60 years of age (PR 1.00; 95% CI, 0.57–1.74). There was no association between measures of central obesity, like waist circumference, waist-to-hip ratio, or waist-to-height ratio and diverticulosis, in either women or men [25]. The observation of a sex-specific association of BMI and diverticulosis is supported by a study from Germany and Lithuania with 707 subjects [26].

In contrast, there are several studies claiming that central obesity and not general obesity is associated with diverticulosis. In a study from South Korea with 133 subjects, fat was quantified with a single cross-sectional computed tomography (CT)-scan at the level of the umbilicus and the total fat area divided into visceral and subcutaneous fractions. The total, subcutaneous, and visceral fat areas (VFAs) but not the BMI was associated with diverticulosis [27]. The much larger study from Japan with 1,445 participants was similarly designed. Again, the subcutaneous and VFAs but not the BMI correlated with diverticulosis. Even in normal-weight subjects (BMI <25), diverticulosis was independently associated with subcutaneous and VFAs[15]. This study is included in the meta-analysis of Wijarnpreecha et al. [11] discussed above, which demonstrated a weaker association of BMI with diverticulosis in studies from Eastern countries than those from the Western world. Up to this point, there is only 1 report from Spain supporting the concept of central obesity being of more importance for the development of diverticulosis than general obesity in Western countries. The cross-sectional study with 62 individuals with diverticulosis and 141 controls found that general obesity (BMI ≥30 kg/m²) was not a risk factor for diverticulosis, rather, an increased waist circumference leads to a significantly increased OR in the multivariate analysis (OR 2.129 [95% CI, 1.005–4.510]) [28].

A relation between body weight and the quantity of diverticula was demonstrated in a prospective study of 624 patients with an about even sex distribution undergoing screening colonoscopy. Participants who were overweight (adjusted OR 2.0, 95% CI, 1.0–3.8) or obese (adjusted OR 3.1, 95% CI, 1.7–5.8) also had increased odds of having >10 diverticula [21].

Possible mechanisms for the association between obesity and the formation of colonic diverticula may be: (a) higher concentrations of methane in obese subjects resulting from an altered gut microbial flora and leading to increased intraluminal pressure; (b) higher intra-abdominal pressure caused by the accumulation of visceral fat (VF) in the abdominal cavity; (c) common confounders, like poor eating habits, including low fiber consumption and little physical activity (Wijampreecha et al. [11]). In conclusion, the current evidence on a 2a level suggests an association between obesity and diverticulosis with special emphasis on premenopausal women (Table 1).

Table 1.

Grade of evidence for the relationship of overweight/obesity with outcomes in diverticulosis/diverticular disease

Obesity/overweight as risk factor/predictor of	Evidence	References
Diverticulosis	2a	[10, 11, 12–16, 17, 18–21, 22, 23, 24, 25, 26, 27]
Diverticular bleeding	2b	[10, 30]
Diverticular disease/diverticulitis	2a	[10, 30, 31–36, 37, 38, 39–44, 45, 62, 63, 64]
Disease severity/complicated disease	2a	[36, 38, 66, 67, 70, 71, 72, 73, 74, 75, 77, 78, 79]
Recurrent disease	2b	[30, 75, 80, 81, 82]
Worse outcome after surgery	2b	[55, 79, 83, 84, 85, 86, 87, 88, 89, 90]

Bold face = systematic review/meta-analysis, italics = cohort study.

Obesity as Risk Factor for Diverticular Bleeding

The first study to investigate an association of obesity with diverticular bleeding is the analysis of the male health professionals cohort published by Strate et al. [10] in 2009. Out of 47,228 subjects, they identified 383 incident cases of diverticular bleeding during 18 years of follow-up. After adjustment for other risk factors, men with a BMI ≥30 kg/m² had a relative risk (RR) of 3.19 (95% CI, 1.45–7.00) for diverticular bleeding, compared to men with a BMI of <21 kg/m². Anthropometric measures indicative of central obesity were also associated with an increased risk for diverticular bleeding. Men in the highest quintile of waist circumference and waist-to-hip ratio, compared with those in the lowest, had a multivariable RR of 1.96 (95% CI, 1.30–2.97) and RR, 1.91 (95% CI, 1.26–2.90), respectively, for diverticular bleeding. BMI at age 21 years was not significantly associated with diverticular bleeding in the multivariable analysis; however, weight gain of >45 pounds since age 21 years had a RR of 2.44 (95% CI, 1.62–3.66; p for trend <0.0001) for diverticular bleeding when compared to men who gained <5 pounds [10].

An association of BMI and diverticular bleeding was also recently described among a female cohort in the Nurses' Health Study with 46,079 participants. Compared to women with a BMI <22.5 kg/m², the hazard ratios (HRs) (95% CIs) were 1.98 (1.32–2.98), 1.97 (1.24–3.13), and 1.56 (1.01–2.41; p-trend = 0.02) for those with a BMI 25.0–27.4, 27.5–29.9, and ≥30.0 kg/m², respectively [29] (summarized in Table 1).

The pathogenetic pathway leading to diverticular hemorrhage is thought to be triggered by a vascular injury resulting in segmental intimal proliferation, breaks and/or duplication of the internal elastic lamina, and atrophy of the media. Experimentally, very similar changes can be induced through mechanical, thermal, chemical or immunological injury. The mechanism by which obesity may affect vascular integrity remains unexplored [30].

Obesity as Risk Factor for Diverticular Disease and Diverticulitis

Smaller case series and case-control studies have documented a relationship between being overweight, particularly for younger people, and a frequent occurrence of diverticulitis [31, 32, 33, 34, 35, 36]. On the other hand, in the Los Angeles Vetarans Affairs cohort, BMI was not a risk factor for progression from diverticulosis to diverticulitis [37].

Meanwhile the association between BMI and the risk for diverticular disease, diverticulitis, and complications, like perforation, abscess, and bleeding, was investigated in a systematic review and meta-analysis of prospective studies [38]. For the analysis of BMI and diverticular disease risk, 6 studies with 28,915 cases and 1,636,777 participants were included [39, 40, 41, 42, 43, 44]. The summary RR for the highest versus the lowest BMI was 1.78 (95% CI, 1.48–2.14). There was a linear dose-response relationship even within the normal BMI range. The summary RR for a 5 kg/m² increase in BMI was 1.28 (95% CI, 1.18–1.40) [38].

Two studies were included in the analysis of BMI and diverticulitis [10, 45], including 1,159 cases and 89,798 participants. The summary RR for the highest versus lowest BMI category was 2.09 (95% CI, 1.63–2.68). In the dose-response analysis, there was a linear relationship without threshold with a summary RR for a 5 kg/m² increase in BMI of 1.31 (95% CI, 1.09–1.56) [38].

Three studies, [10, 17, 46] representing both sexes were included in the analysis of BMI and risk of diverticular disease complications (bleeding, perforation, or abscess), including 2,326 cases and 93,699 participants. The summary RR for a 5 kg/m² increase in BMI was 1.20 (95% CI, 1.04–1.40). There was evidence of a nonlinear association between BMI and diverticular disease complications with the lowest risk observed at a BMI of 22 [38]. Because only 1 study [10] reported data regarding waist circumference, waist-to-hip ratio, and diverticular disease incidence, a meta-analysis of these measures could not be conducted [38].

In the health professionals follow-up study (HPFS) with 47,228 men participants, the highest quintile for waist circumference had a multivariate RR of 1.56 (95% CI, 1.18–2.07) for developing diverticulitis compared to the lowest quintile [10]. In addition, waist-to-hip ratio was also associated with the risk of diverticulitis, when the highest and lowest quintiles were compared: with a multivariable RR 1.62 (95% CI, 1.23–2.14). Moreover, after correcting for BMI, the waist-to-hip ratio remained an independent risk factor for complications. Visceral fat is metabolically more active than subcutaneous fat. This has led to the hypothesis that central adiposity is particularly relevant for the occurrence of diverticular disease, due to the release of pro-inflammatory cytokines and adipokines from the VF [10, 47, 48, 49, 50].

Another mechanism potentially contributing to the deleterious effects of VF is its association with impaired intestinal permeability. The impaired gut barrier function may lead to the “leakage” of gut-derived antigens to the adjacent visceral adipose tissue and induce hyperplasia, inflammation and other metabolic dysfunctions of the fat [51]. The Berlin IBD group around Siegmund expands those results and draws parallels between the role of VF in Crohn's disease and diverticulitis. In Crohn's disease, inflamed intestinal segments are covered by hyperplastic mesenteric fat. This phenomenon is known as “creeping fat” and has been shown to actively regulate intestinal inflammation [52]. Similarly, in diverticulitis, locally accumulated VF surrounds the inflamed diverticula resembling the “creeping fat” in Crohn's disease. In this respect, the data of Gummesson could be alternatively explained by mediators, like adipokine leptin being released from VF − increasing the permeability of the mucosal barrier, thereby facilitating bacterial translocation and promoting inflammatory processes [52]. Remarkably, an increased ratio of visceral to total fat in Crohn's disease patients has been associated with a more pronounced disease activity, fistulising disease, and higher levels of systemic interleukin-6 [53]. Recent evidence also suggests that in diverticular disease, the pivotal aspect of overweight is visceral obesity. As already mentioned, general obesity has been linked in large cohort studies to a higher risk for diverticular complications [10, 17, 46]. Meanwhile, in addition to the results of the HPFS cohort [10], several studies postulate that measures for visceral obesity, like VFA or visceral to subcutaneous fat ratio (V/S fat ratio) are predictors for more complicated disease [54, 55, 56, 57]. This is discussed in more detail below.

Pathophysiologically, another intensely debated subject includes the potential role of microbiota in obesity and diverticular disease. Evidence suggests that microbiome alterations in obese subjects presents by a reduced proportion of Bacteroidetes and higher levels of Firmicutes [58]. However, existing studies profiling fecal and mucosa-associated microbiota in diverticular disease are conflicting, incoherent, and not indicative of alterations associated with disease progression [59, 60, 61].

The analysis of the Nurses' Health Study cohort with 46,079 participants and 1,084 cases of diverticulitis published in 2018 was not part of the meta-analysis by Aune et al. [30]. After adjustment for other risk factors, women with a BMI ≥35.0 kg/m² had a HR for diverticulitis of 1.42 (95% CI, 1.08–1.85) compared to women with a BMI <22.5 kg/m². Compared to women in the lowest quintile, the multivariable HRs among women in the highest quintile were 1.35 (95% CI, 1.02–1.78) for waist circumference and 1.40 (95% CI, 1.07–1.84) for waist-to-hip ratio; these associations were, other than in the male HPFS cohort, attenuated with further adjustment for BMI. Still, when BMI and waist-to-hip ratio were jointly assessed, waist-to-hip ratio appeared to play a role in determining diverticulitis among women who were overweight or obese. The data suggest that the effect of a slight increase in waist circumference and waist-to-hip ratio over time may be lower in women than men [30].

The study also looked at the effect of weight change over time. Compared to women maintaining weight from age 18 years to the present (over roughly 50 years), those who gained 10.0–19.9 kg had a 62% increased risk of diverticulitis (95% CI, 19–119%), and those who gained 20 kg or more had a 73% increased risk (95% CI, 27%–136%). In contrast, a 4-year weight gain or weight loss was not significantly associated with risk of diverticulitis [30].

A group from the Mayo Clinic in Rochester looked at the contribution of BMI, subcutaneous, and VF to diverticulitis risk in a case-control study with about even sex distribution. 381 diverticulitis cases and 381 age- and sex-matched controls were selected. In the univariate analysis, mean BMI was greater in cases (29.8 ± 6.3 kg/m²) than controls (28.3 ± 5.3 kg/m²) (p = 0.001). However, after adjusting for visceral and subcutaneous fat content in the multivariable model, the normalized indices of visceral (OR, 2.4; 95% CI, 1.6–3.7) and subcutaneous fat (OR, 2.9; 95% CI, 1.7–5.2) were independently associated with an increased risk of diverticulitis. A higher BMI was even associated with a lower risk of diverticulitis (OR, 0.8; 95% CI, 0.7–0.8) [62].

A retrospective analysis of data from 43,772 men age 18–20 years, conscripted to military service in Sweden from 1969 to 1970, and followed up for a period of 39 years allows a unique insight into risk factors for diverticular disease. All conscripts underwent an extensive physical and mental examination and answered questionnaires on many lifestyle aspects. With the help of the National Patient Register, 444 men were identified, who required hospitalization for a diagnosis of symptomatic diverticular disease in the following 39 years. This allowed investigation into adolescent lifestyle factors that affect risk of diverticular disease later in life. BMI was associated with an increased risk of diverticular disease both as a continuous variable and a categorical variable. In the multivariate model, the risk increased by 9% per unit increment in BMI (HR, 1.09; p < 0.001). Conscripts in the combined overweight and obese category (BMI ≥25) had a 2-fold risk of developing symptomatic diverticular disease compared with those with a low normal weight (BMI, 18.5–22.5; HR, 2.02; p < 0.001) [63].

The review from Pilgrim and others published in 2013 focuses on diverticular disease in relation to younger age and obesity and presents a rather comprehensive list of publications on those topics up to this point in time [64]. Although not each of the 39 articles listed in this review is mentioned in the above section, they would not add to the conclusion that there is firm 2a-evidence that obesity is a risk factor for development of diverticular disease and diverticulitis (Table 1). Being overweight or obese starts to act as risk factor already in late adolescence.

Obesity as Predictor for Disease Severity or Complicated Disease

Recent advances in the understanding of diverticular disease led to relevant changes in patient management, including a shift to outpatient treatment, omission of antibiotic treatment, and narrower indications for surgery [65]. In the Netherlands, the National Guideline for general practitioners (NHG standard) considers diverticulitis, a clinical diagnosis; and an ultrasound, or CT scan is deemed unnecessary in the primary care setting when typical clinical signs are present. The NHG standard advises only to refer patients with a suspicion of complicated diverticulitis to secondary care. The difficulty to estimate the risk of developing complicated diverticulitis makes a prognostic model that can predict the severity of diverticulitis desirable [66]. Bolkenstein and colleagues [66] undertook a systematic review and meta-analysis, including only articles comparing patients with uncomplicated to complicated diverticulitis with the aim to develop such algorithms. They found only 1 study on BMI as a risk factor for complicated diverticulitis [66]. Longstreth et al. [67] found no significant difference between patients with a BMI greater or smaller than 25 (OR 1.00 [CI, 0.96–1.04]). In further work, the Dutch group arrived at a final validated diagnostic model, including abdominal guarding, C-reactive protein, and leukocytosis (AUC 0.79 [95% CI, 0.73–0.84]). At a complicated diverticulitis risk threshold of ≤7.5%, this model had a negative predictive value of 96% [68]. The Charlson score, a widely used score to assess comorbidity and mortality includes diabetes mellitus but not obesity [69].

Another review from New Zealand selected 21 articles that evaluated factors predictive of severe diverticulitis. Severe diverticulitis (SD) was defined as complicated diverticulitis (associated with hemorrhage, abscess, phlegmon, perforation, purulent/fecal peritonitis, stricture, fistula, or small-bowel obstruction) or diverticulitis that resulted in prolonged hospital admission, surgical intervention, or death. Predictors for SD included first episode of diverticulitis, comorbidities (Charlson score 3), nonsteroidal anti-inflammatory drug use, steroid use, a high C-reactive protein (>175 mg/L) on admission and severe disease on radiological imaging. BMI or obesity was not mentioned [70].

Nevertheless, there are a number of publications claiming that BMI, general or visceral obesities predicts more severe, complicated, or recurrent diverticular disease. We will look at the Asian studies with their predominance of right-sided diverticulosis first.

A small South Korean case series from 2008 looked at clinical characteristics of acute diverticulitis with respect to the relationship between disease location and age, obesity, and complications. 53 (76%) patients with right-sided and 17 (24%) patients with left-sided diverticulitis were enrolled. Right-sided diverticulitis was more common than left-sided diverticulitis in obese patients (p < 0.05); however, the overall complication rate was higher in left-sided than in right-sided diverticulitis (p < 0.05). By multivariate analysis, age <40 years (p = 0.025), BMI ≥25 kg/m² (p = 0.045), and abdominal rebound tenderness (p = 0.011) were risk factors for acute right-sided rather than left-sided diverticulitis [71].

Another study from South Korea focused on risk factors associated with SD. SD was defined as at least one of the following: perforation, abscess, fistula, obstruction, sepsis, and peritonitis requiring urgent operation. In the retrospective analysis, 24 of 190 patients hospitalized for acute diverticulitis met the requirements for SD 175 (92%) patients suffered from right-sided and only 15 from left-sided diverticulitis. The proportion of SD was (18/175 [10%]) for right-sided and (6/15 [40%]) for left-sided diverticulitis. Cases of both locations were analyzed together to determine risk factors for SD. They were of older age, had a fever over 38°C, and presented with changes in bowel habits and a high visceral adipose tissue/total adipose tissue (TAT) ratio. In the multivariate analysis, age of 40 years or more (OR, 3.2; p = 0.032), male gender (OR, 4.0; p = 0.021) and left-sided diverticulitis (OR, 6.2; p = 0.017) remained significant risk factors for SD [72].

A third study from South Korea investigated the relationship between complicated diverticulitis and obesity, particularly visceral obesity. Complicated diverticulitis was defined as having perforation, abscess, systemic inflammatory response syndrome, or recurrent diverticulitis. For the retrospective case series, 140 cases were selected; of which 87 (62%) were assigned to the simple diverticulitis and 53 (38%) to the complicated diverticulitis group. In both groups, only 5 patients had left-sided diverticulitis. The groups did not differ significantly in BMI (p = 0.229) but in several measures for visceral obesity. In the multivariate analysis, VFA remained the only significant factor with a linear risk increase from a VFA <66.32cm² (OR set as 1) to VFA >137.35 (OR 3.21 [95% CI 2.77–3.71]) [73].

For the sake of comprehensiveness, the Japanese study is added; although it does not explore obesity as a risk factor for more severe or complicated disease. However, it sheds light on the relation of right- and left-sided diverticulosis and diverticulitis in different Asian societies. There has been a relative increase in the number of patients with left-sided diverticulosis in Japan. In contrast to the South Korean studies of Kim et al. [72] and Jeong et al. [73], which included <10% patients with left-sided diverticulitis, Yamada et al. [74] included 23% patients with left-sided diverticulitis in their study. They found an association of obesity, particularly visceral obesity, with left-sided diverticulitis [74].

In summary, of the South Korean studies, 2 found that complications are more common in left-sided than in right-sided diverticulitis, and 2 studies presented data that a BMI ≥25 is associated with right-sided diverticulitis and visceral obesity with complicated right-sided diverticulitis, respectively. The Japanese study suggests an association of visceral obesity with left-sided diverticulitis.

In the “Western world,” Dobbins et al. [75] from Australia started out examining the relationship of obesity to the complications of diverticular disease in a small retrospective case-control study published in 2005. Of the 61 patients included, 16 patients had perforated diverticular disease, 11 patients recurrent diverticulitis and 16 patients a single episode of diverticulitis compared to 18 controls. The control group had a significantly lower BMI than patients presenting with perforation (p = 0.001) or recurrent diverticulitis (p = 0.002) [75].

A study from Baltimore, US, hypothesized that acute diverticulitis occurs more frequently in young adult patients (age ≤50 years) than previously recognized, and correlated abdominal obesity as determined by measurement of the sagittal abdominal diameter. The case series included 104 patients, 56 (54%) patients were aged ≤50 years, 38 (36%) suffered from complications. The mean sagittal abdominal diameter for patients ≤50 years old (27.0 cm) was greater than that for patients >50 years old (25.6 cm) (p = 0.05); but severe disease requiring hospital admission, surgery, or percutaneous drainage (or both surgery and percutaneous drainage) was common in all age-groups [36]. Another retrospective chart review form the US included 104 controls and 614 patients hospitalized with complicated diverticulitis. This study was unable to demonstrate any correlation between BMI and the incidence of complicated diverticulitis [76].

A study from New York, US, investigated the postulated link between visceral obesity and more complicated diverticular disease by reviewing retrospectively all adult patients who underwent emergent or elective surgery at our institution for diverticulitis from 2010 to 2014. Thirty-four patients underwent emergent, and 32 patients underwent elective surgery. VFAs and subcutaneous fat areas were measured, and the V/S ratio was calculated. The difference between the V/S ratio for each group was significant (p = 0.0238). In addition, the complication rate was significantly higher (p = 0.024) in the emergent group (n = 12, 35.2%) than the elective group (n = 4, 12.5%) [77].

Investigators from New York, US, aimed to identify preoperative characteristics that may determine the need for emergency surgery for diverticulitis and assess postoperative outcomes for these patients when compared with elective surgery. From the 8,708 patients who underwent colectomy with an underlying diagnosis of diverticulitis from 2012 to 2013, 28.1% had an emergent/urgent colectomy. After performing multivariable analysis, preoperative steroid use, weight loss >10%, BMI <18 kg/m², smoking, age >65, and comorbid conditions but not obesity were associated with a higher rate of emergent/urgent surgery. Mortality (5.2% vs. 0.2%) and infectious and noninfectious complications were higher after nonelective colectomy [78].

As outlined above, the meta-analysis of 3 prospective cohort studies found a positive association of obesity with diverticular disease complications as perforation, abscess or bleeding with a summary RR for a 5 kg/m² increase in BMI of 1.20 (95% CI, 1.04–1.40) [38]. Stressing the importance of visceral obesity, the prospective study of the HPFS cohort described that the waist-to-hip ratio is associated with risk of diverticular complications. When the highest and lowest quintiles were compared, the multivariable RR was 1.62 (95% CI, 1.23–2.14) for diverticulitis. Adjustment for BMI did not change the associations seen for waist-to-hip ratio [10]. Some of the studies discussed above have already been reviewed in the article by Johnson and Champagne [79]. Although the evidence for the Western world presented in case series and case-control studies is not uniform, there is also 2a-evidence for the association of obesity with more severe or complicated diverticular disease for both sexes (Table 1).

Obesity as Predictor for Recurrent Disease

Dobbins et al. [75] found in a small retrospective case-control study with sixty-one patients (including 11 patients with recurrent diverticulitis) and 18 controls that recurrent diverticulitis was significantly associated with higher BMI (p = 0.002). In a retrospective case series with 347 patients of different ethnicities from New York, US, obese patients (BMI >30) were more likely than nonobese patients to experience a recurrent episode of diverticulitis (OR, 1.69; 95% CI, 1.08–2.64; p = 0.02) [80]. Another group from New York identified 265,724 patients with diverticulitis in a database and aimed to identify factors leading to multiple admissions for repeated attacks. 16% of patients required 2 or more admissions. Obesity was associated with a 11% increased risk for ≥2 admissions (p < 0.0001) [81]. A similar retrospective population-based cohort study was presented by a group from Birmingham, UK 65,162 patients were identified with a first episode of acute diverticulitis. The rate of hospital admission for recurrent acute diverticulitis was 11.2%. A logistic regression model identified obesity as a factor associated with recurrent acute diverticulitis (OR 1.38 [95% CI, 1.26–1.52]) [82].

Finally, the analysis of the Nurses' Health Study from Ma et al. [30] also reported on the association of obesity with recurrent diverticulitis. Among the 1,084 diverticulitis cases occurring during 2008–2014, 240 recurrent cases (22%) with >1 episode were documented. The association of BMI with an episode of recurrent diverticulitis seemed to be slightly stronger than that with nonrecurrent diverticulitis cases. As compared to women with a BMI <22.5 kg/m², those with a BMI ≥30.0 kg/m² had a multivariable HR of 1.66 (95% CI, 1.09–2.51; p-trend = 0.002) for recurrent diverticulitis and 1.44 (95% CI, 1.16–1.79; p-trend <0.001) for nonrecurrent diverticulitis [30]. In summary, there is 2b-evidence that obesity is also associated with recurrent diverticulitis.

Obesity as Risk Factor for Worse Outcome after Surgery

Obesity is widely regarded as a factor that increases a surgical procedure's difficulty and occurrence of postoperative morbidity. Johnson and Champagne [79] have reviewed the literature on a possible link between obesity and worse surgical outcome in patients with diverticular disease in 2011.

The review discusses that an increasing BMI in colorectal surgery is generally a risk factor for surgical site infection (SSI), and malfunction of a GI stoma is frequently required in emergency surgery for complicated diverticular disease. Furthermore, parastomal infection is also related to obesity [79].

In the surgical management of diverticulitis, the laparoscopic approach is preferred, yet it can be technically difficult in the abdominally obese. Bouillot and colleagues [83] reported a conversion rate of 13.9% in 179 cases of laparoscopic sigmoidectomy performed for diverticular disease. 20% of conversions were directly related to abdominal obesity making it the second most common reason after inflammation. The conversion of laparoscopic colectomy to an open procedure results in increased morbidity, particularly wound complications and increased lengths of hospital stay in the converted cases [84]. Tuech et al. [85] studied outcomes in normal, overweight, and obese patients and reported no difference in conversion rates, which ranged from 14.8 to 19%. They did, however, demonstrate significantly longer operative times when comparing the normal group to the obese group [79, 85]. A 2003 retrospective database analysis of 582 elective sigmoid resections for diverticular disease found a mortality rate of 1.2% and a morbidity rate of 24.9%. The multivariate analysis revealed 2 statistically significant independent risk factors of mortality: age >75 and obesity (OR = 5.2; 95% CI, 1.1–27.9; p = 0.04) [79, 86]. The review finally provides an example of potential intraoperative complications in obese patients undergoing laparoscopic sigmoid colectomy for diverticulitis injury to the ureter, and discusses preoperative placement of urethral catheters as a prophylactic measure [79]. The studies on the topic published after the review from Johnson and Champagne [79] are summarized here:

A study on 10,952 patients from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database undergoing surgery for diverticulitis examined the relationship between morbid obesity and perioperative outcomes. There were 592 (5.7%) morbidly obese (BMI ≥40 kg/m²) and 2,530 (24.2%) normal-weight (BMI 18.5–25 kg/m²) patients in the cohort. Morbidly obese patients underwent emergency surgery more frequently than normal-weight patients (19.3% vs. 15.4%; p = 0.025). Multivariable regression identified morbid obesity as an independent risk factor for emergency surgery (OR 1.75, 95% CI, 1.37–2.24, p < 0.001), ostomy creation (OR 1.67, 95% CI, 1.34–2.08, p < 0.001), undergoing procedures without an anastomosis (OR 1.78, 95% CI, 1.42–2.24, p < 0.001), and open surgery (OR 2.09, 95% CI, 1.72–2.53, p < 0.001). Morbidly obese patients undergoing emergency surgery had more preoperative systemic inflammatory response syndrome/sepsis/septic shock than normal-weight patients (72.8% vs. 57.7%, p = 0.004) [87].

In 211 diverticulitis patients, volumetric fat parameters were measured on preoperative CT scans and correlated with the frequency of serious postoperative complications (Clavien-Dindo grades 2–4). The serious postoperative complication rate was 12.7%. On univariate analysis, several factors including older age (p = 0.0001), ostomy creation (p = 0.02), higher VF (p = 0.01), emergent surgery (p = 0.05), and higher American Society of Anesthesiology score (p = 0.05) were associated with complications. On multivariate regression analysis, the only factor independently associated with complications was higher VF [55].

Another study based on the ACS NSQIP database tried to further define the association BMI and 30-day postoperative outcomes, including SSI, among patients undergoing colorectal surgery. The cohort comprises 74,891 patients collected from 2011 to 2013 including 21.7% who underwent surgery for diverticulitis. 4.4% were underweight (BMI <18.5), 29.0% had normal weight (BMI 18.5–24.9), 33.0% overweight (BMI 25.0–29.9), 19.8% obesity class I (BMI 30.0–34.9), 8.4% obesity class II (BMI 35.0–39.9), and 5.5% obesity class III (BMI ≥40.0). Compared with normal-weight patients, obese patients experienced incremental odds of SSI from class I to class III (I: OR = 1.5 [95% CI, 1.4–1.6]; II: OR = 1.9 [95% CI, 1.7–2.0]; III: OR = 2.1 [95% CI, 1.9–2.3]). Obesity class III patients were most likely to experience wound disruption, sepsis, respiratory or renal complication, and urinary tract infection. Mortality was highest among underweight patients (OR = 1.3 [95% CI, 1.0–1.8]) and lowest among overweight (OR = 0.8 [95% CI, 0.6–0.9]) and obesity class I patients (OR = 0.8 [95% CI, 0.6–1.0]) [88].

The study from Beresneva and Hall [89] with a similar approach was also based on the ACS NSQIP database. They focused on patients undergoing surgery for diverticular disease and were able to include 52,196 patients from 2005 to 2015, and stratified according to BMI into 9 groups. Increasing BMI was associated with worse outcomes including superficial SSI, deep incisional SSI, organ space SSI, wound disruption complications, ventilator dependence >48 h, acute renal failure, and return to operating room. Risk of developing pneumonia did not have similar correlation with BMI. Overweight status had protective effect on mortality [89].

The most recently published study from the US aimed to explore whether outcomes for morbidly obese patients with diverticulitis are worse than for nonmorbidly obese patients after open colectomy for diverticulitis. A total of 2,019 adults with emergent admission for diverticulitis with evidence of preoperative sepsis and intraoperative contaminated/dirty wound classification, in which a resection with ostomy was performed, were identified from the ACS NSQIP database. 413 patients (20.5%) were morbidly obese (≥35 kg/m²). Morbidly obese patient had no increase in 30-day mortality or length of stay, but they had higher rates of superficial wound infection (9.0% vs. 5.8%; p = 0.0259), deep wound infection (4.4% vs. 1.9%; p = 0.0073), acute renal failure (4.8% vs. 2.4%; p = 0.0189), and postoperative septic shock (17.7% vs. 12.1%; p = 0.0040), and return to the operating room (11.1% vs. 6.4%; p = 0.0015). When 397 morbidly obese patients were propensity matched to 397 nonmorbidly obese patients, conditional logistic regression showed no difference in length of stay (median 12.9 vs. 12.4 day; p = 0.4648) and no increased risk of 30-day mortality (p = 0.947). Nevertheless, morbid obesity was an independent predictor for return to the operating room (adjusted OR: 27.09 [95% CI: 2.68–274.20]; p = 0.005). In conclusion, this study analysis showed only increased reoperation for obese patients with diverticulitis, in conflict with the majority of literature describing the many perioperative risks of obesity [90]. In summary, there is ample evidence up to the 2b-level that obesity in patients undergoing surgery for diverticular disease is associated with multiple forms of postoperative morbidity, although there is no evidence for excess mortality (Table 1).

Obesity is associated with the development of diverticula, the progress to diverticular disease, including diverticular hemorrhage, diverticulitis, more severe or complicated disease, and recurrent diverticulitis as well as increased postoperative morbidity in the case of surgical therapy. Levels of visceral adiposity have been shown to be effectively reduced following bariatric surgery [91, 92]. There are no reports yet on a possible effect of weight loss surgery on diverticular disease.

Conflict of Interest Statement

The author has no conflicts of interest to declare.

Funding Sources

There was no funding of any research relevant to the article.