Forty-three studies were included in the review; 12 RCTs, 30 comparative studies and one case series. Five RCTs (701 patients) were included in the meta-analysis. Sequence generation, allocation concealment, and follow-up were generally adequate in all five trials; blinding was not possible. Four trials scored 3, and one trial scored 2 on the Jadad scale. Average follow-up ranged from 22.4 to 37.2 months.
Patients with unresectable hepatocellular carcinoma treated with radiofrequency ablation had significantly higher rates of survival compared with percutaneous ethanol injection at one year (RR 0.62, 95% CI 0.41 to 0.94; five trials; Ι² = 0%) and three years (RR 0.79, 95% CI 0.65 to 0.96; four trials; Ι²=57%). Disease-free recurrence rates were significantly lower with radiofrequency ablation (two to 14%) than percutaneous ethanol injection (11 to 35%).
Results for radiofrequency ablation compared with laser-induced thermal therapy (one study) and microwave coagulation therapy (three studies) were mixed; a combination of radiofrequency ablation with transarterial chemoembolisation had better outcomes than radiofrequency ablation alone (five studies); and radiofrequency ablation compared with resection for hepatocellular carcinoma outside the Milan criteria had significantly lower overall and disease-free survival rates (three studies). There were no differences for disease-free or overall survival when different radiofrequency ablation electrodes and generators were compared.