A confidence evaluation helps to make informed decisions about the results of meta-analyses. The goal of this work is to perform a confidence evaluation of results of a network meta-analysis (NMA) on the digestive side effects of tapentadol in patients with chronic pain. An updated search in PubMed/Medline and Web of Science search until March 2020 was done to perform pairwise meta-analyses with NMA using random-effect models and confidence in network meta-analysis (CiNeMA) for the confidence analysis. Twenty-five studies were included in the final analyses. Pairwise and indirect comparisons showed a reduced risk of constipation with tapentadol compared to oxycodone. The confidence evaluation did not raise any concerns in terms of confidence for the oxycodone versus tapentadol comparisons. The oxycodone-naloxone versus tapentadol comparisons showed some concerns, particularly in terms of imprecision and incoherence. Regarding the overall risk of any side effects, the confidence evaluation showed a major concern regarding imprecision, but not for the comparison between tapentadol and oxycodone. However, this comparison showed a major heterogeneity. A confidence evaluation in meta-analysis on the effect of tapentadol compared to other opioids in chronic pain showed possible imprecision, heterogeneity and/or incoherence. However, with a high level of confidence, tapentadol was associated with a lower incidence of constipation than oxycodone. Confidence analyses can help to get more information from meta-analyses.
Read full abstract