Format

Send to

Choose Destination
World J Surg. 2001 May;25(5):638-44.

Patterns of recurrence after resection of colorectal liver metastases: prediction by models of outcome analysis.

Author information

1
Dipartimento di Scienze Oncologiche e Chirurgiche, Sezione di Clinica Chirurgica, University of Padova, Via Giustiniani 2, 35128 Padova, Italy. lisem@ux1.unipd.it

Abstract

Various series have reported similar survival and recurrence rates after resection of colorectal liver metastases (CRLM). If outcomes were predictable, indications for surgery could be improved. This hypothesis was tested in 135 consecutive patients with CRLM who underwent "curative" resection from 1977 to 1997. Among the 132 patients available for follow-up, three groups were identified on the basis of outcome: (1) survival of more than 5 years disease-free (n = 32; 24%); (2) diffuse recurrences within the first 6 months (n = 24; 18%); and (3) discrete recurrences for which reresection was performed (n = 16; 12%). As our results are similar to those reported in the literature, we assumed that about 50% of patients with resectable lesions have recognizable patterns of recurrence. At multivariate analysis, factors significant for disease-free survival (DFS) were the percentage of liver invasion, metastases to lymph nodes at the primary site, number of metastases, preoperative glutamic pyruvic transaminase (GPT) level, and type of liver resection. On the basis of the relative risk (RR) expressed by significant prognostic factors, a score model was developed, and three prognostic groups were defined: Group A, with the best prognostic score, included 23 of 32 (72%) patients who survived more than 5 years, and that with the worst prognostic score (group C) included 22 of 24 (92%) patients with early diffuse recurrences. Extreme (especially unfavorable) outcomes can therefore be predicted. By using improved models of outcome analysis, many patients could be spared surgery as first-line treatment, and stratification criteria could be worked out for future trials.

PMID:
11369992
DOI:
10.1007/s002680020138
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center