Methods to decrease blood loss during liver resection: a network meta-analysis

Liver resection is a major surgery with significant mortality and morbidity. Various methods have been attempted to decrease blood loss and morbidity during elective liver resection. These methods include different methods of vascular occlusion, parenchymal transection, and management of the cut surface of the liver. A surgeon typically uses only one of the methods from each of these three categories. Together, one can consider this combination as a treatment strategy. The optimal treatment strategy for liver resection is unknown.

To assess the comparative benefits and harms of different treatment strategies that aim to decrease blood loss during elective liver resection.

We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, and Science Citation Index Expanded to July 2012 to identify randomised clinical trials. We also handsearched the references lists of identified trials.

We included only randomised clinical trials (irrespective of language, blinding, or publication status) where the method of vascular occlusion, parenchymal transection, and management of the cut surface were clearly reported, and where people were randomly assigned to different treatment strategies based on different combinations of the three categories (vascular occlusion, parenchymal transection, cut surface).

Two review authors identified trials and collected data independently. We assessed the risk of bias using The Cochrane Collaboration's methodology. We conducted a Bayesian network meta-analysis using the Markov chain Monte Carlo method in WinBUGS 1.4 following the guidelines of the National Institute for Health and Care Excellence Decision Support Unit guidance documents. We calculated the odds ratios (OR) with 95% credible intervals (CrI) (which are similar to confidence intervals in the frequentist approach for meta-analysis) for the binary outcomes and mean differences (MD) with 95% CrI for continuous outcomes using a fixed-effect model or random-effects model according to model-fit.

We identified nine trials with 617 participants that met our inclusion criteria. Interventions in the trials included three different options for vascular occlusion, four for parenchymal transection, and two for management of the cut liver surface. These interventions were combined in different ways in the trials giving 11 different treatment strategies. However, we were only able to include 496 participants randomised to seven different treatment strategies from seven trials in our network meta-analysis, because the treatment strategies from the trials that used fibrin sealant for management of the raw liver surface could not be connected to the network for any outcomes. Thus, the trials included in the network meta-analysis varied only in their approaches to vascular exclusion and parenchymal transection and none used fibrin sealant. All the trials were of high risk of bias and the quality of evidence was very low for all the outcomes. The differences in mortality between the different strategies was imprecise (seven trials; seven treatment strategies; 496 participants). Five trials (six strategies; 406 participants) reported serious adverse events. There was an increase in the proportion of people with serious adverse events when surgery was performed using radiofrequency dissecting sealer compared with the standard clamp-crush method in the absence of vascular occlusion and fibrin sealant. The OR for the difference in proportion was 7.13 (95% CrI 1.77 to 28.65; 15/49 (adjusted proportion 24.9%) in radiofrequency dissecting sealer group compared with 6/89 (6.7%) in the clamp-crush method). The differences in serious adverse events between the other groups were imprecise. There was a high probability that 'no vascular occlusion with clamp-crush method and no fibrin' and 'intermittent vascular occlusion with Cavitron ultrasonic surgical aspirator and no fibrin' are better than other treatments with regards to serious adverse events. Quality of life was not reported in any of the trials.The differences in the proportion of people requiring blood transfusion was imprecise (six trials; seven treatments; 446 participants). Two trials (three treatments; 155 participants) provided data for quantity of blood transfused. People undergoing liver resection by intermittent vascular occlusion had higher amounts of blood transfused than people with continuous vascular occlusion when the parenchymal transection was carried out with the clamp-crush method and no fibrin sealant was used for the cut surface (MD 1.2 units; 95% CrI 0.08 to 2.32). The differences in the other comparisons were imprecise (very low quality evidence). Three trials (four treatments; 281 participants) provided data for operative blood loss. People undergoing liver resection using continuous vascular occlusion had lower blood loss than people with no vascular occlusion when the parenchymal transection was carried out with clamp-crush method and no fibrin sealant was used for the cut surface (MD -130.9 mL; 95% CrI -255.9 to -5.9). None of the trials reported the proportion of people with major blood loss.The differences in the length of hospital stay (six trials; seven treatments; 446 participants) and intensive therapy unit stay (four trials; six treatments; 261 participants) were imprecise. Four trials (four treatments; 245 participants) provided data for operating time. Liver resection by intermittent vascular occlusion took longer than liver resection performed with no vascular occlusion when the parenchymal transection was carried out with Cavitron ultrasonic surgical aspirator and no fibrin sealant was used for the cut surface (MD 49.6 minutes; 95% CrI 29.8 to 69.4). The differences in the operating time between the other comparisons were imprecise. None of the trials reported the time needed to return to work.

Very low quality evidence suggested that liver resection using a radiofrequency dissecting sealer without vascular occlusion or fibrin sealant may increase serious adverse events and this should be evaluated in further randomised clinical trials. The risk of serious adverse events with liver resection using no special equipment compared with more complex methods requiring special equipment was uncertain due to the very low quality of the evidence. The credible intervals were wide and considerable benefit or harm with a specific method of liver resection cannot be ruled out.