Effect of cooling methods and target temperature on outcomes in comatose patients resuscitated from cardiac arrest: Systematic review and network meta-analysis of randomized trials

Background: Targeted temperature management (TTM) has been recommended after cardiac arrest (CA), however the specific temperature targets and cooling methods (intravascular cooling (IVC) versus surface cooling (SC)) remain uncertain. Methods: PUBMED and EMBASE were searched until October 8, 2022 for randomized clinical trials (RCTs) investigating the efficacy of TTM after CA. The randomized treatment arms were categorized into the following 6 groups: 31.C to 33.C IVC, 31.C to 33.C SC, 34.C to 36.C IVC, 34.C to 36.C SC, strict normothermia or fever prevention (Strict NT or FP), and standard of care without TTM (No-TTM). The primary outcome was neurological recovery. P-score was used to rank the treatments, where a larger value indicates better performance. Results: We identified 15 RCTs, involving 5,218 patients with CA. Compared to No-TTM as the reference, the other therapeutic options significantly improved neurological outcomes (vs No-TTM; 31.C to 33.C IVC: RR = 0.67, 95% CI 0.54 to 0.83; 31.C to 33.C SC RR = 0.73, 95% CI 0.61 to 0.87; 34.C to 36.C IVC: RR = 0.66, 95% CI 0.51 to 0.86; 34.C to 36.C SC: RR = 0.73, 0.59 to 0.90; Strict NT or FP: RR = 0.75, 95% CI 0.62 to 0.90). Overall, 31-33.C IVC had the highest probability to be the best therapeutic option to improve outcomes (the ranking P-score of 0.836). As a subgroup analysis, the ranking P-score showed that IVC might be a better cooling method compared to SC (IVC vs SC P-score: 0.960 vs 0.670). Conclusions: Hypothermia (31.C to 36.C IVC and SC) and active normothermia (Strict-NT and Strict-FP) were associated with better neurological outcomes compared to No-TTM, with IVC having a greater probability of being the better cooling method than SC.