Repeat Curative Intent Liver Surgery Is Safe and Effective for Recurrent Colorectal Liver Metastases: Results from An International Multi-Institutional Analysis
Mechteld C. De Jong*1, Carlo Pulitano2, Alessandro Ferrero3, Jennifer Strub4, Michael a. Choti1, Richard D. Schulick1, Gilles Mentha4, Lorenzo Capussotti3, Timothy M. Pawlik1
1Department of Surgery, The Johns Hopkins University School of Medicine, Baltimore, MD; 2Vita-Salute San Raffaele University, Milan, Italy; 3Unit of Hepato-Biliary-Pancreatic Surgery, A.O. Ordine Mauriziano, Torino, Italy; 4University Hospitals of Geneva, Geneva, Switzerland
Introduction: Although 5-yr survival approaches 55% following resection of colorectal liver metastases (CRLM), most patients (pts) develop recurrent disease that is often isolated to the liver. Although repeat curative intent surgery (CIS) is increasingly performed for recurrent CRLM, only small series have been reported. We sought to determine safety and efficacy of repeat CIS for recurrent CRLM, as well as determine factors predictive of survival in a large multi-center cohort of pts.Methods: Between 1985-2008, 1512 pts who underwent CIS-defined as curative intent hepatic resection/radiofrequency ablation (RFA)-for CRLM were identified from an international multi-institutional database. 209(14%) pts underwent 251 repeat CIS. Data on clinicopathologic factors, morbidity, mortality were collected and analyzed.Results: Following initial CIS, 594(64%) pts recurred within the liver. Of these, 209 pts underwent repeat CIS for recurrent disease. The majority had hepatic resection alone as initial therapy (n=188;90%). A subset of pts underwent 3rd (n=35) or 4th(n=7) repeat CIS. Mean interval between surgeries was similar (1st→2nd: 20 mon; 2nd→3rd: 22 mon; 3rd→4th: 12 mon; P=0.20). Tumor characteristics changed with each subsequent CIS (Table). Extent of hepatic resection decreased with subsequent CIS (≥ hemi-hepatectomy: 1st CIS: 70% vs 2nd CIS: 37% vs 3rd/4th CIS: 19%; P<0.05). RFA was utilized in roughly 1/4 of pts undergoing repeat CIS (2nd CIS: 22% vs 3rd/4th CIS: 23%). Mortality and morbidity were similar following 2nd, 3rd, 4th CIS, respectively (all P>0.05)(Table). 5-yr survival was 36%, 40%, and 44% following 2nd, 3rd, 4th CIS, respectively (P>0.05). Factors predictive of survival included tumor size (HR=1.8;P=0.04), tumor number (HR=4.5;P=0.03) and presence of extrahepatic disease (HR=3.1;P=0.008). Conclusion: Repeat CIS for recurrent CRLM can be performed with low morbidity and near zero mortality. Pts with low tumor burden and no extrahepatic disease are best candidates for repeat CIS. In these pts, repeat CIS can provide long-term survival rates similar to those of 1st CIS.
| Initial CIS | Second CIS | Third+Fourth CIS | |
| Time between CIS | N/A | 20 months | 18 months |
| Solitary Hepatic Lesion | 45% | 59% | 73%* |
| Largest Lesion > 5 cm | 22% | 11% | 7%* |
| Mean CEA Level | 92 ng/ml | 45 ng/ml | 26 ng/ml* |
| Presence of Extrahepatic Disease | 9% | 17% | 24%* |
| ≥ Hemi-hepatectomy | 70% | 37% | 19%* |
| Use of RFA | 10% | 22% | 23% |
| Peri-Op Morbidity | 23% | 25% | 20% |
| Peri-op Mortality | 0 | 0.5 | 0 |
| 5-Year Survival | 47% | 36% | 44% |
* P<0.05
Back to Program | 2009 Program and Abstracts | 2009 Posters
