Search Weight Loss Topics:




Feb 5

Systematic review and meta-analysis of serum total testosterone and luteinizing hormone variations across … – Nature.com

Testosterone Comments Off on Systematic review and meta-analysis of serum total testosterone and luteinizing hormone variations across … – Nature.com

Search results

The search strategy identified 214 studies and 6 additional papers were identified through other sources for a total of 220 studies. After removing the duplicates, 184 studies were screened of which 152 were excluded based on title and abstract. Full text of the 32 selected studies was obtained. 14 studies were excluded: 4 studies were conducted on animal models, 10 didnt evaluate TT and LH levels. The PRISMA flow diagram is presented in Fig.1. Complessively, 18 records fulfilled the inclusion criteria and were included in the final analysis14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31.

PRISMA flow diagram. TT: Testosterone; LH: Luteinizing Hormone.

Eighteen studies met the inclusion criteria and were included in this analysis, overall 1,575 male patients with Covid-19 infection and 886 male controls. According to the severity of the disease, 823 patients were classified as presenting moderate disease and 500 as having severe disease. Patients enrollment ranged between 2020 and 2022. 15 studies14,17,18,19,20,21,22,24,25,26,27,28,29,30,31 exhibited a prospective cohort design and 3 studies reported a retrospective cohort design15,16,23. The studies included were conducted in Turkey14,17,21,22,25,29 Italy15,16,26,27,28 China24,30,31 USA18, Greece20, Russia19, Austria23. Among the studies included, 11 were controlled16,17,19,20,21,24,25,28,29,30. Collected data included testosterone levels 14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31, LH levels14,17,19,22,23,24,28,29,30, IL-6 levels15,23,26,27,28,30,31, lymphocytes count14,15,16,23,24,26,27,28,30,31, D-dimer levels14,15,16,20,24,26,27,31. The characteristics of the 18 studies included are presented in Table1.

Out of n=1814,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 studies encompassed in the systematic review, n=1016,17,19,21,22,24,25,28,29,30 were further meta-analytically compared with regards to mean TT levels variations between Covid-19 and control populations. Of note, at this preliminary assessment, there was considerable heterogeneity across each single study with I2 98.45%, Q (9): 580.22, p<0.00. Publication bias was initially assessed by Galbright and Funnel plot (Suppl. Fig. 1). Inspection of both plots suggested that there was no small-study effect with the smaller studies tending to have higher SMD variability, suggesting absence of publication bias (Egger test, p=0.96). Additionally, the Trim and Fill method suggested that no studies would have needed to be included to remove residual asymmetry from the funnel plot. The main contribution to study heterogeneity was indeed identified by sub-group analysis summarized in Suppl. Fig. 2. Interestingly, year of publication which in this case is directly associated with the progression of the Covid-19 pandemic exhibited a significant ascending influence on SMD observed, while no further study design nor cumulative sample size of patients accrued was associated with the observed heterogeneity. The effect of publication year was further highlighted at cumulative meta-analysis sorted by year, where increasing variations in SMD of TT levels across the studies were observed (Suppl. Fig. 2). However, as per sensitivity analysis, the leave-one-out analysis suggested acceptable variation along the cumulative SMD observed when clustering the data by publication year. According to predefined random-effects model, we found a significant reduction in the SMD of testosterone levels in Covid-19 patients compared to controls (3.25, 95%CI 5.93, 0.57, p=0.00, I2=98.45%). The difference was even higher when examining men with severe Covid-19 (5.04, 95%CI 8.82, 1.26, p=0.00, I2=96.60%) (Fig.2).

Forest plot for Standardized Mean Difference (SMD) across studies presenting serum Total Testosterone (TT) in Covid-19 male patients vs. male control (A), clinically severe Covid-19 patients vs. control (B), clinically severe Covid-19 vs. clinically moderate Covid-19 patients (C) and, Covid-19 related deaths vs. Covid-19 survivor patients (D).

Additionally, within the Covid-19 population, we assessed patients stratified by severity of clinical manifestations. As further evidence, compared with patients exhibiting moderate Covid-19 symptoms, patients with more severe sequelae exhibited lower levels of testosterone showing a slight yet significant absolute higher SMD (3.53nmol/L, 95%CI 5.11, 1.95, p=0.00, I2=88.19%).

This was corroborated across the n=314,15,18 experiences which reported outcomes stratified according Covid-19 survivors and non-survivors, confirming a similar overall decrease in SMD serum TT levels (3.04, 95%CI 4.05, 2.04, p=0.42, I2=0.00%) (Fig.2).

Finally, in order to explore the remaining heterogeneity observed, we investigated the role of patient available comorbidities or inflammatory/haemato-chemical variables retrieved to the SMD estimates by meta-regression analysis. As expected, there was a direct correlation between increasing BMI of the Covid-19 population and the SMD observed across the studies (Coeff. 2.48, SE: 1.15; p=0.033; Suppl. Fig. 3). Moreover, the sole haemato-chemical confounder significantly associated with TT variation was the absolute lymphocyte count depicting an inverse trajectory with the SMD observed across the studies (Coeff. 7.37, SE: 2.19; p=0.001; Suppl. Fig. 3).

Regarding LH levels, only n=717,19,22,24,28,29,30 studies reported the needed information to further compare the overall populations and the sub-groups assessing the outcomes according to the severity of Covid-19 clinical manifestations. A considerable heterogeneity was documented also in this setting with I2 95.60%, Q(6): 136.32, p<0.02. Additionally, the inspection of both Galbright and Funnel plots suggested that there was a significant small-study effect with the smaller studies tending to have higher SMD variability, suggesting the existence of higher risk of bias among the publications assessed (Egger test, p=0.035). However, the Trim and Fill method suggested that only n=1 study would have needed to be included to remove residual asymmetry from the Funnel plot (Suppl. Fig. 4). This preliminary finding was further confirmed at sub-group analysis where the size of the sample analysed was significantly associated with greater variability in the pooled SMD yet not being influenced by the publication year like previously observed (Suppl. Fig.4).

In contrast to TT levels, the effect on cumulative LH variations was non-significant nor clinically relevant between Covid-19 cases and matched controls (SMD: 0.69, 95%CI 0.56, 1.94, p=0.00, I2=95.60%) ranging along the null-effect line from 3.55, 95%CI 2.64, 4.46, to 2.53, 95%CI 4.73, 0.33 in the study of Cinislioglu et al.17 and Xu et al.30 respectively. This was also true when comparing controls with the n=317,28,30 studies with more severe Covid-19 population (SMD: 0.41, 95%CI 2.37, 3.19, p=0.00, I2=95.54%) (Fig.3).

Forest plot for Standardized Mean Difference (SMD) across studies presenting serum Luteinizing Hormone (LH) levels in the Covid-19 patients vs. control (A), clinically severe Covid-19 patients vs. control (B), and clinically severe Covid-19 vs. clinically moderate Covid-19 patients (C).

Furthermore, at sensitivity analysis through leave-one-out assessment there was no single study effect size which would have significantly impaired the observed pooled results if omitted from the analysis (Suppl. Fig. 5).

At meta-regression on quantitative variables available, the relative percentage of Covid-19 patients with active smoking status was the sole factor influencing the pooled SMD estimate for serum LH levels (Coeff. 0.08, SE: 0.01; p=0.001; Suppl. Fig. 6).

Finally, when sub-analyzing within the Covid-19 severity of clinical manifestations, there was no significant or clinically relevant SMD across the n=5 studies included14,17,26,28,30 0.28 (95%CI 1.55, 1.00, p=0.00, I2=88.52%).

Here is the original post:
Systematic review and meta-analysis of serum total testosterone and luteinizing hormone variations across ... - Nature.com

Related Posts

    Your Full Name

    Your Email

    Your Phone Number

    Select your age (30+ only)

    Select Your US State

    Program Choice

    Confirm over 30 years old

    Yes

    Confirm that you resident in USA

    Yes

    This is a Serious Inquiry

    Yes

    Message:



    matomo tracker