Longitudinal studies demonstrate a gradual decline in total serum testosterone levels in men from the age of 40, 20% and 30% of men meet LOH criteria in their 60s and 70s (5, 6). However, no overall association was observed between cardiovascular events and either free or total serum testosterone at baseline or during testosterone treatment. Within our IPD analysis, two masked clinical investigators identified a total of 342 cardiovascular events from the included IPD, which to date is the highest published rate of cardiovascular events. Eligible studies included prospective cohort studies with plasma or serum testosterone concentrations measured for adult men using mass spectrometry with at least 5 years of follow-up data and one of the specified outcome measures recorded. GS designed the study, participated in quality check of studies included, in the data extraction and wrote the manuscript. In conclusion, we describe here a temporal trend in testosterone serum levels in men over the last 50 years that could be interpreted as an adaptation of the hypothalamic-pituitary-gonadal axis. However, these data were available for only a proportion of the studies included, limiting the statistical power of the correlation analyses. When heterogeneity is found, we will attempt to determine potential reasons for it by examining individual study and subgroup characteristics. We will investigate attrition rates, e.g. drop-outs, losses to follow-up and withdrawals, and critically appraise issues of missing data and imputation methods (e.g. last observation carried forward). Evaluation of the important numerical data such as screened, randomised participants as well as intention-to-treat and as-treated will be undertaken. A ‘risk of bias’ graph and a ‘risk of bias summary’ figure will be generated. We will judge ‘risk of bias criteria’ as ‘low risk’, ‘high risk’ or ‘unclear risk’ and evaluate individual bias items as described in the Cochrane Handbook for Systematic Reviews of Interventions . We used the Cochrane risk of bias tool to assess the risk of bias of the retrieved RCTs (16). In instances where there were two distinct androgen administration approaches within the same RCTs, two sets of data were collected. Search terms include "androgen or testosterone", "late-onset hypogonadism" and related expressions (Supplementary Data Sheet 1). The aging process directly correlates with LOH, potentially influencing the function or production of sex steroid hormones such as testosterone, follicular stimulating hormone, luteinizing hormon, and gonadotropinreleasing hormone (3, 4). SBhas, PJS, SSE, MG, TGT, EJGil, YTvderS, MHE-V, EJGia, GH, SR, JS, KLH, KGA, GBB, JLT, HMT, CHCK, WST, LSM, RJR, RSS, SR, MSA, and LVM were involved in trial data collection and data transfer. MB, CNJ, KG, LA, WSD, NO, RQ, and SBhat were involved in conceptualisation and study design. The authors are grateful to Prakash Abraham, Alison Avenell, Craig Ramsay, Graham Scotland, Neil Scott, and Finlay MacKenzie for their advice; and to the many individuals from academia and industry who helped in the conduct of this study. At DW analyses, LH and years of blood collection showed a slight, positive autocorrelation, confirming the potential existence of a trend across the years. The first considered study groups in which blood samples were collected before 2000 and the second after 2000. However, among the 117 papers using MS, only five (comprising eight study groups) were conducted before 2000, specifically between 1991 and 1998. The years of sample collection were considered as a covariate, while the number of subjects enrolled, subjects’ age and BMI as factors affecting the regression. However, a meta-analysis reported that TRT had no effect on the cardiovascular events (64). Pearl et al. (52) performed a retrospective study including 120 patients and reported that IPSS decreased by an average of 7.42 for patients with severe IPSS after treatment of TRT. The systematic review and meta-analysis included 28 RCTs (3253 patients) published since 1990. However, some studies revealed that symptoms related to LOH may not be correlated with serum total testosterone levels (9, 10). Nevertheless, we cannot exclude that a high number of unreported cardiovascular events in the non-IPD studies could ultimately change the conclusions of our analysis. Despite these risk factors, the overall incidence of cardiovascular events was not significantly higher during testosterone treatment than for placebo. The databases also do not include testosterone formulations administered in a clinic setting, so our results underrepresent the scope of the observed TT prescribing practices. Our study’s main strength is the utilization of state PDMP databases with mandatory inclusion of all dispensed prescriptions regardless of payor, in contrast to commercial databases and is the first to report new TT prescribing data since 2018. While TT did not show an increased risk of major adverse cardiac events (MACE) in a high-risk population , most research has been limited to topical gels and not the more widely used injectable formulations that have shown a higher incidence of MACE, hospitalizations, and death .
