Home > DARE Reviews > Systematic review of tumour number and...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

Systematic review of tumour number and outcome after radical treatment of colorectal liver metastases

MD Smith and JL McCall.

Review published: 2009.

Link to full article: [Journal publisher]

CRD summary

The authors concluded that radical treatment of more than three colorectal liver metastases resulted in poorer overall survival. However, five-year survival was achievable. The authors' conclusions appeared to reflect the evidence presented, but limitations highlighted by the authors and their warning to heed caution when interpreting the findings should be borne in mind.

Authors' objectives

To determine the effectiveness of radical treatment in patients with one to three or four or more colorectal liver metastases (CLMs).

Searching

MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL) were searched from January 2000 to June 2008. Search terms were reported. Reference lists of relevant articles were searched. Authors' personal libraries were searched. No search was made for unpublished material.

Study selection

Eligible studies were published in English and described the outcome of patients who underwent radical resection and/or ablation of CLMs. Studies need to separate outcome data according to the number of tumours in the liver. Cointerventions of neo-adjuvant or adjuvant chemotherapy were allowed and recorded. Studies that described en bloc resection of structures adjacent to the liver, resection of extrahepatic disease, repeat hepatectomies and two-stage hepatectomies were included if the treatment intent was radical. Studies that included patients with CLMs and other types of liver tumour were included if it was possible to separate data that pertained to patients with CLMs. Studies of ablation only were eligible for inclusion if ablation was considered a second-line treatment and used for multiple, unresectable or recurrent CLMs. Studies that were excluded from the review were: case reports with 10 or fewer patients; studies that selected patients according to their outcomes; studies that did not provide at least one-year outcomes; and studies of liver transplantation.

Approximately half of the studies were performed in Europe; others in North America or South-East Asia. Median start date for patient accrual was 1988 (range 1960 to 2001). Median closing date was 2001 (range 1992 to 2007). Treatments included in the review were resection and/or ablation (cryoablation, laser-induced thermotherapy, radiofrequency ablation). Most studies used resection only. Most studies did not report on use of chemotherapy either before or after radical treatment. Where reported, chemotherapy regimens varied widely. Outcomes included: overall, median, three- and five-year survival; median time to recurrence; three- and five-year disease-free survival; and postoperative mortality.

Studies were selected for inclusion by one reviewer; any uncertainties were discussed with a second reviewer.

Assessment of study quality

Study quality was assessed using eight criteria graded yes or no: prospective data collection; participation of all consecutive patients; complete description of participants; attrition (>24 months follow-up); prognostic factor measured; outcome measured and described; confounding measured and reported; analysis appropriate and reported.

One reviewer performed study quality assessment; any uncertainties were discussed with a second reviewer.

Data extraction

Hazard ratios (HRs) and standard errors were extracted from individual studies into a standard form. Where these data were not reported, values were estimated from relevant data.

One reviewer performed data extraction; any uncertainties were discussed with a second reviewer.

Methods of synthesis

Hazard ratios were combined in a meta-analysis using a random-effects (DerSimonian and Laird) model. Heterogeneity was assessed using X2 and I2 tests. Subgroup analyses included: studies of liver resection only; studies of tumour ablation only; prospective studies with median follow-up longer than 24 months; and an estimated power greater than 80% (as defined in the review). Publication bias was assessed using funnel plots and Egger regression.

Results of the review

Forty-six studies (n=9,934 patients) were included in the review. Follow-up ranged from 17 to 99 months. Sample sizes ranged from 33 to 929 patients. Thirty-nine studies were of hepatic resection (n=8,485 patients). Seven studies were of radical tumour ablation with or without concomitant resection (n=1,449 patients). The highest-quality study met six of the eight criteria. Four studies met five criteria. Most studies were low quality in terms of definition and measurement of the number of CLMs (reported adequately in six studies), recurrence of the disease (reported adequately in five studies) and all defined confounding factors (reported adequately in 13 studies). Fifteen studies were included in the meta-analysis: 12 that reported hazard ratios and standard errors and three that provided sufficient data from which hazard ratios could be calculated.

Five-year survival (42 papers): For all patients, five-year overall survival ranged from 7% to 58%. Median reported survival was 36% (24 studies). Reported five-year overall survival rates in patients with one to three CLMs ranged from 25.9% to 50%, with a median of 39%. For patients with four or more CLMs, the reported five-year overall survival rate ranged from 0% to 53% (median 17.1%). For patients who underwent resection only, five-year overall survival rate ranged from 25.9% to 50% for those with one to three CLMs (19 studies) and 0% to 53% for those with four or more CLMs (25 studies). For patients who underwent ablation only, reported five-year overall survival rates ranged from 7% to 33% and median overall survival rates ranged from 16% to 35%. Median overall survival in patients with fewer CLMs ranged from 18 to 32 months and ranged from 14 to 34 months in those with more numerous CLMs.

Overall survival was statistically significantly poorer in patients with four or more CLMs compared to those with three or less after undergoing either resection and/or ablation (HR 1.67,95% CI 1.43 to 1.95; n= 15 studies) or resection only (HR 1.74, 95% CI 1.51 to 1.99; n=14 studies). However, there was evidence of moderate statistical heterogeneity for resection and/or ablation only (I2=34. 4%). There was no statistically significant difference in overall survival after ablation only (HR 1.18, 95% CI 0.99 to 1.40; n=one study).

Findings for disease-free survival and median time to relapse were reported in the review.

There was evidence of publication bias for resection and/or ablation.

Authors' conclusions

Radical treatment of more than three CLMs resulted in poorer overall survival. However, five-year survival was achievable and the number of lesions should not in itself be used to exclude patients from surgery.

CRD commentary

The review addressed a clear research question. Inclusion criteria were adequate but broad. Only published studies in English were included in the review, so there were risks of language and publication biases. There was evidence of publication bias, which the authors acknowledged. Review processes (study selection, data extraction and study quality assessment) were performed by one reviewer only, except in cases of uncertainty when the help of a second reviewer was sought; this left the review processes at risk of reviewer error and bias. The included studies were generally of low quality and most were underpowered. There was substantial clinical heterogeneity among the studies and fewer than half were included in the meta-analysis, which the authors acknowledged.

The authors' conclusions appeared to reflect the evidence presented, but limitations highlighted by the authors and their warning to heed caution when interpreting the findings should be borne in mind.

Implications of the review for practice and research

The authors did not state any implications for practice and research.

Funding

Not stated.

Bibliographic details

Smith MD, McCall JL. Systematic review of tumour number and outcome after radical treatment of colorectal liver metastases. British Journal of Surgery 2009; 96(10): 1101-1113. [PubMed: 19787755]

Indexing Status

Subject indexing assigned by NLM

MeSH

Bias (Epidemiology); Colorectal Neoplasms /mortality; Hepatectomy /mortality; Humans; Liver Neoplasms /mortality /secondary /surgery; Prognosis; Proportional Hazards Models; Survival Analysis

AccessionNumber

12009109590

Database entry date

26/01/2011

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 19787755