Visceral adiposity and inflammatory bowel disease.

BACKGROUND: Rates of obesity are increasing worldwide, as is the incidence of inflammatory bowel disease (IBD). Obesity is now considered an inflammatory state. Visceral adiposity in particular may be associated with a more severe inflammatory phenotype in IBD. AIM: The aim of this review article is to summarise the current literature on the association between visceral adiposity and outcomes in inflammatory bowel disease METHODS: To collect relevant articles, PubMed/MEDLINE and Embase searches were performed using Boolean search phrases. Grey literature and manual searches were also performed. Abstracts were selected by two independent reviewers based on pre-determined criteria. Full text articles were reviewed, and data extracted and assessed. RESULTS: One hundred twenty-seven abstracts were obtained through the initial search, with 85 abstracts reviewed and 22 full text articles included. Characteristics are included in Table 1. Most of these were retrospective studies and of moderate or weak quality. Studies suggested visceral fat content is higher in Crohn's disease than in healthy controls. Visceral adiposity was associated with an increased risk of complex Crohn's disease phenotype (OR 26.1 95% CI 2-75.4; p = 0.02). Post-operative recurrence was higher in patients with higher visceral fat indices (RR 2.1; CI 1.5-3; p = 0.012). There were conflicting data regarding the effect of visceral adiposity on post-operative complications and the efficacy of medical therapy. Table 1 Study characteristics Author Year Country Study type Study numbers Control group Disease type Methodology e.g. CT Body composition measurements Results Argeny [24] 2018 Austria Retrospective cohort N = 95 N/A Crohn's disease CT; L3 level Visceral fat area (cm2) Visceral fat index (VFA/m2) No association between VFA or VFI and short-term post-operative outcomes Bryant [30] 2018 Australia Prospective cohort N = 110 N/A Crohn's disease and UC DXA Visceral adipose tissue (VAT) (cm3) Visceral adipose tissue (grams) VAT/height index (cm3/m2) VAT:subcutaneous adipose tissue ratio Fat mass index (kg/m2) VAT and VHI increased significantly over 24 months Bryant [13] 2018 Australia Prospective cohort N = 72 N/A Crohn's disease; female DXA Visceral adipose tissue (VAT) (cm3) Visceral adipose tissue (grams) VAT/height index (cm3/m2) VAT:subcutaneous adipose tissue ratio VAT:SAT positively associated with stricturing disease Adiposity not associated with fistulising disease phenotype VAT:SAT significantly associated with faecal calprotectin in L3 phenotype VAT:SAT significantly negatively associated with VHI and QoL over 24 months Buning [25] 2015 Germany Case control N = 50 N = 19 healthy controls Crohn's disease MRI US VAT Thickness of abdominal fat Distance to posterior wall of aorta Area of inferior part of perirenal fat VAT accumulation was higher in CD patients vs healthy controls VAT and VAT/fat mass ratio higher in patients in short-term remission vs long-term remission VAT/FM higher in stricturing/fistulising disease vs inflammatory subtype No association between VAT/FM and CDAI, HBI or anti-TNF treatment Connolly [26] 2014 US Retrospective cohort N = 143 N/A Crohn's disease CT (L1-L5 level) Visceral/intra-abdominal adiposity (VA) Subcutaneous adiposity (SA) VA not associated with post-operative morbidity Decreased SA and increased visceral/subcutaneous ratio were predictive of post-op complications. (p = 0.02; p < 0.001) Cravo [27] 2017 Portugal Retrospective cohort N = 71 N/A Crohn's disease CT (L3 level) Smooth muscle area (cm2) Visceral fat area (cm2) Subcutaneous fat area (cm2) Visceral fat index Muscle radiation attenuation L2 phenotype associated with lower muscle attenuation and higher visceral fat index (non-significant) B2/B3/surgery - significantly lower muscle attenuation. VFI associated with increased risk of complicated phenotype. (OR 26.1; 95% CI 1-75; p = 0.02) Ding [17] 2016 US Retrospective cohort N = 164 N/A Crohn's disease CT (L3 level) Visceral fat area (cm2) Subcutaneous fat area Total fat area Visceral obesity associated with longer duration of surgery, increased intra-operative blood loss and longer length of bowel resected Higher complication rates in patients with visceral obesity (p < 0.001) VFA independent risk factor of adverse post-op outcomes Ding [14] 2017 Retrospective cohort N = 106 N/A Crohn's disease CT (L3 level) Visceral fat area Subcutaneous fat area Skeletal muscle area Skeletal muscle index Visceral obesity and myopenic obesity not significantly associated with risk of primary non-response Body composition factors not associated with secondary loss of response Erhayiem [18] 2011 UK Retrospective cohort N = 50 N/A Crohn's disease CT (L4 level) Mesenteric fat index (visceral:subcutaneous area ratio)N = 50 Mesenteric fat index was significantly higher in complicated Crohn's disease. ROC analysis for MFI in identifying complicated Crohn's disease: AUC = 0.95 (95% CI 0.89-1.0) Feng [28] 2018 China Retrospective cohort N = 80 Non-IBD GI patients Crohn's disease CT-energy spectral Visceral fat area (cm2) Subcutaneous fat area (cm2) Mesenteric fat index No significant difference in VFA between Crohn's disease cohort and control group. (p = 0.669). ROC analysis: detection of disease based on VFA and MFI: AUC 0.776 Sensitivity 77.5% Specificity 67.5% Hafraoui [16] 1998 France/Belgium Prospective N = 43 Healthy volunteers n = 13 Intestinal resection n = 9 Crohn's disease MRI (umbilicus) Total abdominal fat (cm2) Intra-abdominal fat (cm2) Subcutaneous fat (cm2) Ratio of intra-abdominal:total fat area was significantly higher in patients with Crohn's vs controls. (p = 0.012) No correlation between abdominal fat tissue and disease activity, duration or steroid therapy Holt [29] 2017 Australia/New Zealand RCT N = 44 N = 11 placebo group Crohn's disease CT/MRI (L3, L4-5 levels) Visceral adipose tissue area Subcutaneous adipose tissue area Skeletal muscle area Visceral adipose tissue/height index VHI > 1.5 times gender mean was specific for endoscopic recurrence (100%) with sensitivity of 29%. PPV = 1 (0.59-1.00) There was no significant difference in disease activity at 18 months post-resection based on VHI > 1.5 gender mean Li [31] 2015 China Retrospective cohort N = 72 N/A Crohn's disease CT (umbilicus) Visceral fat area (cm2) Subcutaneous fat area (cm2) Mesenteric fat index Post-op recurrence was more frequent with high VFA values. (p = 0.019) VFA and MFI were independent risk factors for post-operative recurrence. (p = 0.013 and p = 0.028, respectively) High VFA and high MFI were significantly higher in patients with endoscopic activity (p = 0.023) Liu [32] 2016 Retrospective case-control N = 59 N = 30 (< 15% increase VFA) IBD with IPAA CT (L3) Visceral fat area Subcutaneous fat area No difference in pouchitis, pouch sinus formation and composite adverse pouch outcomes between the 2 groups with and without VFA increase > 15%. Excessive VAT gain was an independent risk factor for the composite adverse pouch outcomes. (OR 12.6 (95% CI 1.19-133.5) Magro [33] 2018 Brazil Cross-sectional study N = 78 N = 28 Health control Crohn's disease DEXA Fat and lean masses Visceral fat (kg) Visceral fat/BMI Visceral fat per %body fat VF was higher in Crohn's disease group (p = 0.004) compared to controls Parmentier-Decrucq [34] 2009 Prospective study N = 132 N/A Crohn's disease MRI Subcutaneous fat Visceral fat Total abdominal fat increased 18% in Crohn's disease patients treated with infliximab induction therapy Shen [35] 2018 China Retrospective N = 97 N/A Crohn's disease CT (umbilicus) Subcutaneous fat area Visceral fat area Mesenteric fat index VFA and MFI were significantly lower in patients with mucosal healing (post-infliximab). (p < 0.0001) SFA was not significantly different VFA correlated with CDAI (p < 0.001) and was an independent predictive factor for mucosal healing Stidham [15] 2015 Retrospective N = 269 N/A Crohn's disease CT(T10-L5) Subcutaneous fat volume Visceral fat volume No significant difference in visceral fat volume between patients with surgical complications Thiberge [36] 2018 France Retrospective N = 149 N/A Crohn's disease CT (L3 level) Skeletal muscle index Visceral adiposity index Subcutaneous adiposity index SAI and VAI were significantly lower in patients who underwent surgery or who died in 6 months post-CT(p = 0.009 and p < 0.001) VanDerSloot [37] 2017 Cohort study N/A Crohn's disease CT (T11-S5) Visceral adipose tissue volume Non-significant trend toward increased risk of surgery and penetrating disease with increasing VAT Wei [38] 2018 China Retrospective N = 86 N/A IBD post-resection CT (L3 level) Visceral adipose volume Subcutaneous adipose volume Increased visceral:subcutaneous fat ratio was associated with increased procalcitonin levels on post-op days 1, 3 and 5 Yadav [39] 2017 India Prospective N = 97 N/A IBD CT (L4 level) Visceral fat area Subcutaneous fat area No statistically significant correlation between visceral fat and disease behaviour in Crohn's disease N/A not applicable, VFA visceral fat area, VFI visceral fat index, VAT visceral adipose tissue, VHI visceral adipose tissue to height index, SAT subcutaneous adipose tissue, DXA dual-energy X-ray absorptiometry, CT computer tomography, MRI magnetic resonance imaging, US ultrasound, CDAI Crohn's disease activity index, HBI Harvey-Bradshaw Index, anti-TNF anti-tumour necrosis factor, SA subcutaneous adiposity, ROC receiver operating curve, AUC area under the curve, MFI mesenteric fat index, SAI subcutaneous adiposity index, PPV positive predictive value CONCLUSION: Visceral adiposity appears to be increased in Crohn's disease with some evidence that it is also associated with more complex disease phenotypes. There is also a signal that post-operative recurrence rates are affected by increasing mesenteric adiposity. There is a relative lack of data in UC patients and further high-quality studies are necessary to elucidate the relationship between visceral adiposity and IBD and the implications for patient outcomes.