Abstract
OBJECTIVES: Systematic review to determine any association between imaging features of idiopathic mesenteric panniculitis (MP) and subsequent malignancy. METHODS: Two researchers searched primary literature independently for imaging studies of MP. They extracted data focusing on methodology for unbiased patient accrual and capability to determine a link between MP and subsequent malignancy. They noted imaging features of MP. Data were accrued and meta-analysis intended. RESULTS: Fourteen of 675 articles were eligible; 1,226 patients. Only three (21 %) accrued patients prospectively. Twelve (86 %) studies described CT features. Follow-up varied widely; 1 month to 8 years. Prevalence of MP was influenced by accrual: 0.2 % for keyword search versus 1.7 % for consecutive series. Accrual bias affected nine (64 %) studies. 458 (38 %) of 1,209 patients had malignancy at accrual but varied widely (8-89 %), preventing meta-analysis. Sixty (6.4 %) of 933 patients developed new malignancy subsequently, also varying widely (0-11 %). Of just four studies that determined the proportion of unselected, consecutive patients with MP developing subsequent malignancy, three were retrospective and the fourth excluded patients with lymphadenopathy, likely excluding patients with MP. CONCLUSION: Studies were heterogeneous, with biased accrual. No available study can determine an association between MP and subsequent malignancy with certainty. KEY POINTS: • Our systematic review of mesenteric panniculitis found that imaging studies were biased. • Spectrum and recruitment bias was largely due to retrospective study designs. • No study could confirm a certain link between mesenteric panniculitis and subsequent malignancy. • Excessive methodological heterogeneity precluded meaningful meta-analysis. • High-quality research linking mesenteric panniculitis imaging features and subsequent malignancy is needed.