PET/CT for differentiating between tuberculous peritonitis and peritoneal carcinomatosis the parietal peritoneum

Abstract

Objectives: Tuberculous peritonitis (TBP) mimics peritoneal carcinomatosis (PC). We aimed to investigate the discriminative use of PET/CT findings in the parietal peritoneum. Materials and Methods: Parietal peritoneal PET/CT findings from 76 patients with TBP (n=25) and PC (n=51) were retrospectively reviewed. The lesion locations were noted as right subdiaphragmatic, left subdiaphragmatic, right paracolic gutters, left paracolic gutters, and pelvic regions. The distribution characteristic consisted of a dominant distribution in the pelvic and/or right subdiaphragmatic region (susceptible area for peritoneal implantation, SAPI) (SAPI distribution), a dominant distribution in the remaining regions (less-susceptible area for peritoneal implantation, LSAPI) (LSAPI distribution), or a uniform distribution. PET morphological patterns were classified as F18-fluorodeoxyglucose (18F-FDG) uptake in a long beaded line (string-of-beads 18F-FDG uptake) or in a cluster (clustered 18F-FDG uptake) or focal 18F-FDG uptake. CT patterns included smooth uniform thickening, irregular thickening, or nodules. Results: More common findings in the parietal peritoneum corresponding to TBP as opposed to PC were (a) ≥4 involved regions (80.0% vs 19.6%), (b) uniform distribution (72.0% vs 5.9%), (c) string-of-beads 18F-FDG uptake (76.0% vs 7.8%), and (d) smooth uniform thickening (60.0% vs 7.8%) (all P<0.001), whereas more frequent findings in PC compared with TBP were (a) SAPI distribution (78.4% vs 28.0%), (b) clustered 18F-FDG uptake (56.9% vs 20.0%), (c) focal 18F-FDG uptake (21.6% vs 4.0%), (d) irregular thickening (51.0% vs 12.0%), and (e) nodules (21.6% vs 4.0%) (P<0.001, P<0.05, P>0.05, P<0.05, P>0.05, respectively). Conclusion: Our data show that PET/CT findings in the parietal peritoneum are useful for differentiating between TBP and PC.