Background: Evaluation of peritoneal metastases by computed tomography (CT) scans is challenging and has been reported to be inaccurate.
Methods: A multi-institutional prospective observational registry study of patients with peritoneal carcinomatosis from colorectal cancer was conducted and a subset analysis was performed to examine peritoneal cancer index (PCI) based on CT and intraoperative exploration.
Results: Fifty-two patients (mean age 52.6 ± 12.4 years) from 16 institutions were included in this study. Inaccuracies of CT-based assessment of lesion sizes were observed in the RUQ (P = 0.004), LLQ (P < 0.0005), RLQ (P = 0.003), distal jejunum (P = 0.004), and distal ileum (P < 0.0005). When CT-PCI was classified based on the extent of carcinomatosis, 17 cases (33%) were underestimations, of which, 11 cases (21%) were upstaged from low to moderate, 4 cases (8%) were upstaged from low to severe, and 2 cases (4%) were upstaged from moderate to severe. Relevant clinical discordance where an upstage occurred to severe carcinomatosis constituted a true inaccuracy and was observed in six cases (12%).
Conclusions: The actual clinical impact of inaccuracies of CT-PCI was modest. CT-PCI will remain as a mandatory imaging tool and may be supplemented with other tools including positron emission tomography scan or diagnostic laparoscopy, in the patient selection for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy.