Table 1 Statistical analyzes and characteristics of the participants in the studies
Author | Year | Study design | Sample size | N of in tramadol group | Outcome | Mean Age (Year) | Sex (Female) | Statistically analysis |
|---|---|---|---|---|---|---|---|---|
Zeng et al. [17] | 2019 | Prospective cohort | 88,902 | 16,922 | ACM&CVD | 70.1 | 61.2% | Compared cause-specific mortality in the tramadol cohort with codeine, it matched the comparator cohort using a cause-specific Cox-proportional hazard model. |
Wei et al. [19] | 2020 | Prospective cohort | 66,048 | 42,722 | CVD | 68.3 | 61.8% | Comparison risk of CVD in the tramadol cohort with codeine matched comparator cohort using a cause-specific Cox-proportional hazard model adjusted for other variables. |
Li et al. [18] | 2020 | Prospective cohort | 56,325 | 7813 | ACM&CVD | 68 | 62.7% | Sequential propensity score-matched cohort study. Comparing the risk of ACM and CVD in the tramadol cohort with codeine, it matched the comparator cohort using a cause-specific Cox-proportional hazard model adjusted for other variables. |
Xie et al. [15] | 2021 | Prospective cohort | 3,330,572 | 28,206 | ACM&CVD | 53.1 | 51.7% | The possibility of confounding by indication was controlled by using propensity score matching. In addition, sensitivity analysis was performed to control for confounding variables, and the effect size was reported as HR in 95 CI. |
Sørensen et al. [23] | 2022 | Prospective cohort | 56,325 | 7813 | ACM&CVD | 68 | 62.7% | A pre-post design was used to evaluate changes in ACM and CVD after use of tramadol vs. Codeine. |
Gau et al. [22] | 2022 | Prospective cohort | 1320 | 880 | ACM&CVD | NA | NA | Matching for age, gender, and year of diagnosis of rheumatoid arthritis was used for two groups. A Cox proportional regression model estimated future CVD development’s adjusted hazard ratio (aHR). Kaplan-Meier and log-rank tests were used to estimate the cumulative incidence of EC and CVD. |
Li et al. [16] | 2022 | Prospective cohort | 56,325 | 14,622 | ACM&CVD | 68 | 62.7% | Logistic regression was used to estimate propensity scores for the initial tramadol prescription. The basic characteristics of the tramadol group were compared with the pumpkin group before and after adjusting the propensity score. SAS software was used for statistical analyses, and hazard ratios (HRs) with 95% confidence intervals (95% CI) for the risk of tramadol versus codeine use with outcomes were calculated and reported. |