Ranking of mechanical thromboprophylaxis methods in surgery: a network meta-analysis

Objective. To obtain a ranking scale for effectiveness of active mechanical thromboprophylaxis methods: 1) muscle pump activator (MPA), 2) graduated compression stockings (GCS), 3) intermittent pneumatic compression (IPC), 4) electrical muscle stimulation (EMS), 5) IPC+GCS, 6) basic thromboprophylaxis. Material and methods. Selection of articles for network meta-analysis was carried out in three stages. At the first stage, we found systematic reviews and meta-analyses of randomized clinical trials of known methods for mechanical thromboprophylaxis in surgery. All these studies were performed in accordance with PRISMA principles while RCT quality was assessed using Cochrane Collaboration “Risk of Bias” tool. At the second stage, we extracted RCTs that met inclusion criteria (adults in surgical departments of any profile, designs comparing various interventions, cases of deep vein thrombosis). In the third stage, we searched for RCTs for the period not covered by the authors of systematic reviews and meta-analyses found at the first stage. Confidence of evidence was assessed using the CINeMA approach, and interventions were categorized using the minimally contextualized GRADE approach. Results. We formed an evidence base of 32 RCTs selected from 3 meta-analyses and 1 systematic report, as well as tested its quality. A rank scale of effectiveness of interventions was obtained: AMP> (IPC+GCS, EMS, IPC)> GCS> basic thromboprophylaxis. Sensitivity analysis does not change the rank for AMP. Among 15 designs, 2 ones had “medium”, 7 — “low”, 6 — “very low” evidence base. Categorization suggested that three methods (GCS, EMS and IPC) crossed the threshold of clinical significance as better ones than basic thromboprophylaxis. AMP was recognized as the best. Conclusion. The ranking scale of effectiveness may be useful in re-evaluating preferences in surgical practice. Further high-quality RCTs are needed to improve its significance.