To Register      SMDM Homepage

Monday, 18 October 2004

This presentation is part of: Poster Session - CEA: Methods and Applications; Health Services Research

BIOPSY OR RESECTION FOR SINGLE SMALL LIVER NODULES IN PATIENTS WITH COMPENSATED CIRRHOSIS- A DECISION ANALYSIS

KE Bremner, BSc1, AM Bayoumi, MD, MSc2, M Sherman, MBBCH, PHd3, and M Krahn, MD, MSc1. (1) University Health Network, Department of Medicine, Toronto, ON, Canada, (2) St. Michael's Hospital, University of Toronto, Inner City Health Research Unit, Toronto, ON, Canada, (3) University Health Network, Department of Hepatology, Toronto, ON, Canada

Purpose: To determine the optimal management of small (1-2 cm) liver nodules detected during routine screening for hepatocellular carcinoma (HCC) in cirrhotic patients. These nodules may be, but are not invariably, malignant. Biopsy is often recommended, but its limited sensitivity results in false negative findings. Hepatic resection offers potential cure, but a risk of liver failure and unnecessary surgery.

Methods: We compared two strategies in a Markov decision model: immediate resection, or liver biopsy and resection if positive. The patient cohort, men and women, aged 55-70 years, had compensated cirrhosis, no comorbidities, and a single 1-2 cm liver nodule identified as probable HCC with routine ultrasound screening and CT confirmation. Patients who had resection (immediately or after biopsy) faced the risks of surgery (unnecessary if not HCC), recurrent HCC, and liver decompensation, but had the advantage of early treatment (if HCC-positive). Biopsied patients had the potential benefit of confirmatory diagnosis, but the risks of biopsy, and the chance of missed HCC, and potential progression to an unresectable state before re-screening. Probabilities and utilities were obtained from a comprehensive literature review and local data.

Results: The baseline analysis favoured initial hepatic resection, but the gain was small (4.72 quality-adjusted life months, or 5.49 life months). The model was robust to most assumptions; the decision was not sensitive to probabilites for transitions to, or death from, biopsy, resection, liver decompensation, or HCC recurrence, or the utility values. Only the sensitivity of biopsy changed the decision; if 95% or higher, biopsy was preferred. Patients with false negative biopsies had the poorest prognosis, as they had the risks of biopsy (needle-track seeding, death), and the chance of decompensation or developing other contraindications to resection prior to accurate HCC diagnosis by imaging in follow-up screening.

Conclusions: Our model suggests that resection of all suspicious, 1-2 cm liver nodules offers longer survival and better quality of life than selection of patients for resection based on liver biopsy. Diagnostic test modalities for HCC must attain high sensitivity (>95%), thus assuring fewer missed HCC, to offset the benefits of early treatment.


See more of Poster Session - CEA: Methods and Applications; Health Services Research
See more of The 26th Annual Meeting of the Society for Medical Decision Making (October 17-20, 2004)