Testosterone in males with COVID‐19: a 12‐month cohort study

Abstract

Background

Male patients with COVID‐19 have been found with reduced serum total testosterone (tT) levels and with more severe clinical outcomes.

Objectives

To assess total testosterone (tT) levels and the probability of recovering eugonadal tT levels during a minimum 12‐month timespan in a cohort of men who have been followed over time after the recovery from laboratory‐confirmed COVID‐19.

Materials and methods

Demographic, clinical and hormonal values were collected for the overall cohort. Hypogonadism was defined as tT ≤9.2 nmol/l. The Charlson Comorbidity Index was used to score health‐significant comorbidities. Descriptive statistics was used to compare hormonal levels at baseline versus 7‐month (FU1) versus 12‐month (FU2) follow‐up, respectively. Multivariate cox proportional hazards regression model was used to identify the potential predictors of eugonadism recovery over time among patients with hypogonadism at the time of infection.

Results

Of the original cohort of 286 patients, follow‐up data were available for 121 (42.3%) at FU1 and 63 (22%) patients at FU2, respectively. Higher median interquartile range (IQR) tT levels were detected at FU2 (13.8 (12.3–15.3) nmol/L) versus FU1 (10.2 [9.3–10.9] nmol/L) and versus baseline (3.6 [3.02–4.02] nmol/L) (all p < 0.0001), whilst both LH and E₂ levels significantly decreased over the same time frame (all p ≤ 0.01). Circulating IL‐6 levels further decreased at FU2 compared to FU1 levels (19.3 vs. 72.8 pg/ml) (p = 0.02). At multivariable cox regression analyses, baseline tT level (HR 1.19; p = 0.03 [1.02–1.4]) was independently associated with the probability of tT level normalization over time, after adjusting for potential confounders.

Conclusions

Circulating tT levels keep increasing over time in men after COVID‐19. Still, almost 30% of men who recovered from COVID‐19 had low circulating T levels suggestive for a condition of hypogonadism at a minimum 12‐month follow‐up.

1. INTRODUCTION

Host characteristics may favour more severe clinical outcomes in patients with the new severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2)‐induced disease (COVID‐19), thus including fatal events. Of those, sex‐related differences would prompt males to be in the least advantageous position. Numerous biopathology pathways associated with COVID‐19 were mostly observed with worse clinical outcomes in men. ¹

Numerous previous observations demonstrated that SARS‐CoV‐2 enters into the uro‐genital system, ² , ³ and eventually may damage both Sertoli and Leydig cells. ³ , ⁴ It has therefore been speculated that male hormonal milieu has a heavy pathophysiological role in association with SARS‐CoV‐2 infection, with endogenous testosterone leaving men more inclined to develop even serious complications as a consequence of COVID‐19. ⁵ In this context, an association between SARS‐CoV‐2 infection and secondary hypogonadism has been unveiled already at hospital admission in male patients, with lower total testosterone (tT) levels predicting the most severe clinical outcomes. ⁶ Findings from a recent meta‐analysis clearly confirmed that SARS‐Cov‐2‐infected men were characterized by reduced tT levels, whereas no difference in either follicle‐stimulating hormone (FSH) or luteinizing hormone (LH) levels was observed. ⁷ Similar findings were observed when patients with severe and mild forms of COVID‐19 were compared with each other. ⁷ Of clinical relevance, lower tT levels in men with COVID‐19 resulted in up to four‐ and five‐fold increased risk to be admitted to the Intensive Care Unit (ICU) or even to die. ⁷ As a further relevant observation, a secondary or at least a mixed hypogonadism was found during the acute phase of the disease, with a progressive change throughout the recovery phase. Indeed, the negative effects of SARS‐Cov‐2 infection on T levels were confirmed only in those studies including patients in the acute phase of COVID‐19 ⁷ and then disappearing in those considering subjects over the recovery phase. ⁸ However, even though tT levels eventually increased over time, Salonia et al. showed that more than 50% of men who recovered from COVID‐19 still presented with tT levels suggestive for biochemical hypogonadism at a 7‐month follow‐up investigation of their original data. ⁶ , ⁹

Therefore, based on the preclinical data suggesting a potential damage of the Leydig cells ³ , ⁴ , ⁷ and the clinical observation that circulating tT significantly ‐ and hyper‐acutely ‐ decreases after viral infection, ⁶ , ⁷ , ¹⁰ with a slow and perhaps incomplete recovery of predisease values over time, ⁹ we analysed tT levels and the proportion of patients with biochemical criteria compatible with hypogonadism still after 12 months minimum from healing in a cohort of men who eventually survived after COVID‐19 and had been comprehensively followed while recovering .

2. METHODS

Comprehensive data from a sub‐cohort of 121 patients belonging to the original cohort of men admitted to the Emergency or Clinical departments because of symptomatic confirmed SARS‐CoV‐2‐induced disease at a single academic hospital between February 29th and May 2nd 2020 and who eventually recovered from COVID‐19 were prospectively followed‐up until December 2021. ⁶ , ⁹

Patients’ enrollment followed the same criteria previously described in detail. ⁶ , ⁹ In summary, upon obtaining written individual patient's consent, patients’ clinical data were retrieved using a dedicated case report form, according to an institutional protocol (Covid‐BioB, ClinicalTrials.gov NCT04318366; Ethical Committee approval number 34/int/2020). ¹¹ Data collection followed the principles outlined in the Declaration of Helsinki. All methods were performed in accordance with the relevant guidelines and regulations.

As for previous studies, comorbidities were scored with the Charlson Comorbidity Index (CCI, which was categorised as 0 or 1 or  ≥ 2). ¹² Body mass index (BMI) (kg/m2) was measured for every patient. Moreover, at hospital admission, patients were subdivided into mild, moderate, and severe acute respiratory distress syndrome, according to standard definitions ¹³ and baseline chest radiography findings of SARS‐CoV‐2 pneumonia severity. ¹⁴ Thereafter, patients with COVID‐19 were divided into three groups according to the outcome after hospital admission: Group (1) patients in good clinical conditions and discharged home from the emergency department; Group (2) patients admitted in the internal medicine unit until possible discharge home; and, Group (3) patients invasively ventilated in the ICU, and subsequently successfully extubated and discharged either to the internal medicine unit or home. ⁶ , ⁹ Furthermore, a validated composite risk score based on the characteristics at the time of first hospital admission was calculated for every patient (i.e., Critical‐ill COVID‐19 score). ¹⁵

2.1. Biochemical measurements

Baseline and follow‐up venous blood samples were drawn in all patients between 7 AM and 11 AM, after an overnight fast, and kept at 4°C until serum or plasma was separated by centrifugation. ⁶ , ⁹ , ¹⁶ Serum and plasma aliquots were then stored at –80°C until assay. Venous blood samples and laboratory measurements have been performed in all patients according to the National Committee for Clinical Laboratory Standards guidelines. ¹⁷

For the specific purposes of this analysis, on each sample for every patient at baseline we measured (FSH; FSH: LIAISON FSH ([REF 312251)), LH (LH: LIAISON LH ([REF] 312201)), tT (testosterone: LIAISON Testosterone ([REF] 310410)) and 17β‐estradiol (E₂; E₂: LIAISON Estradiol II Gen ([REF] 310680); DiaSorin SpA, Saluggia, Italy). Furthermore, Interleukin‐6 (IL‐6) was measured by ECLIA (Elecsys IL‐6, COBAS ROCHE) in every patient. ⁶ , ⁹ Routine blood tests encompassed serum biochemistry (including complete blood count with differential and C‐reactive protein [CRP] as inflammation markers).

2.2. Post‐discharge follow‐up

Patients were followed‐up after hospital discharge and complete recovery from COVID‐19 (laboratory‐confirmed negativity for SARS‐CoV‐2) until December 2021. Each patient was recalled for a first follow‐up visit (FU1) after a median of 7 (range: 4–8) months (n = 121, 42.3%) from hospital discharge and a second follow‐up visit (FU2) after a minimum long‐term follow‐up of 12 (range: 12–18) months (n = 63, 22%) after discharge. A serum blood sample was drawn in all patients and stored in our institution biobank during the post‐discharge follow‐up visits. For every patient, we measured FSH, LH, tT, E₂ and IL‐6, as previously described. ⁹

2.3. Outcomes

Primary outcome was to assess the overall rate of patients with tT levels suggestive for hypogonadism (according to a tT threshold of 9.2 nmol/l ¹⁸ ) at follow‐up and to investigate the probability of recovering eugonadal tT levels over time. Secondary outcome was to assess clinical predictors of recovering normal tT levels over time. ¹⁶ , ¹⁹

2.4. Statistical methods

Distribution of data was tested with the Shapiro–Wilk test. Data are presented as medians IQR or frequencies (proportions). We used paired t‐Test or chi‐square test to compare hormonal levels between COVID‐19 patients at these follow‐up time‐marks: baseline versus FU1, versus FU2. Moreover, to test the hypothesis that COVID‐19 complete recovery is associated with gradual recovery of tT levels, we also used Kaplan–Meier analysis to estimate the probability of recovering eugonadal (normal tT levels) status over time among patients with hypogonadism at the time of infection.

Multivariate cox proportional hazards regression model was used to identify the potential predictors of eugonadal status recovery over time. The model was adjusted for baseline clinical confounders.

Statistical analyses were performed using Stata 14.0 (StataCorp, College Station, TX, USA). All tests were two sided, and statistical significance level was determined at p < 0.05.

3. RESULTS

Of the original cohort of 286 patients with COVID‐19, ⁶ follow‐up data were available for 121 patients at FU1 and 63 patients at FU2, respectively; the overall median IQR follow‐up was 7 (6–17) months. Descriptive statistics of the whole cohort at baseline has been previously detailed. ⁶ , ⁹ Median IQR age of patients was 57 (49–65) years with a median BMI of 28.4 (25–31) kg/m².

Table 1 details the hormonal levels at hospital admission as compared with FU1 and FUI versus FU2 assessments, respectively. At FU2, both LH (3.7 vs. 4.7 mU/ml) and E₂ (26.1 vs. 30.1 pg/ml) were lower compared to FU1 values (all p < 0.0001) (Figure 1). Conversely, tT levels were higher at FU2 compared to FU1 (13.8 vs. 10.2 nmol/L; p < 0.0001) (Figure 1). Circulating IL‐6 levels further decreased at FU2 compared to FU1 levels (19.3 vs. 72.8 pg/ml) (p = 0.02).

TABLE 1.

Hormone levels at admission and at 7‐month, 12‐month follow‐up assessments

Variable	Hospital admission	7‐month follow‐up	p‐Value*	12‐month follow‐up	p‐Value**
Total testosterone (ng/ml)	3.6 (3.02–4.02)	10.2 (9.3–10.9)	<0.0001	13.8 (12.3–15.3)	<0.0001
E2 (pg/ml)	34.8 (31.2–38.4)	30.1 (28.8–31.2)	0.006	26.1 (24.3–27.9)	<0.0001
FSH (mUI/L)	7.2 (5.4–9.1)	8.8 (7.9–9.7)	0.04	8.8 (6.7–10.9)	0.6
LH (mUI/L)	5.2 (4.7–5.7)	4.7 (4.2–5.2)	0.01	3.7 (3.1–4.2)	<0.0001
IL‐6 (mUI/L)	32.3 (9.5–76.5)	72.8 (25.5–120.2)	0.01	19.3 (3.5–35.1)	0.02
Hypogonadism (<9.2 nmol/l) (N [%])	115 (95%)	66 (55%)	0.01	19 (30.1%)	<0.0001
Hypogonadism (<12 nmol/l) (N [%])	116 (97%)	80 (66%)	0.04	21 (33.3%)	<0.0001

Abbreviations: E₂, 17β‐Estradiol; FSH, follicle‐stimulating hormone; LH, luteinizing hormone; tT, total testosterone.

^*Hospital admission versus 7‐month follow‐up.

^**7‐month follow‐up versus 12‐month follow‐up.

FIGURE 1.

Scatter plot, vertical. Sex‐related hormonal analyses at baseline compared to 6‐month follow‐up in COVID‐19 patients. (A and D) Total testosterone. (B and E) Follicule‐stimulating hormone (FSH). (C and F) Luteinizing hormone (LH). *p = 0.01; ****p < 0.0001

Of all, tT levels <9.2 nmol/L was still observed in 19 (30.1%) patients at FU2 compared to 66 (55%) patients at FU1 and to 115 (95%) patients at baseline, respectively. Likewise, according to a 12 nmol/L threshold, as many as 21 (33.3%) patients still had tT levels suggestive for hypogonadism at FU2 follow‐up compared to 116 (80%) patients at hospital admittance (p < 0.0001) (Table 2).

TABLE 2.

Estimated risk of tT levels normalisation according to Kaplan–Meier analysis (median follow‐up 7 [6–20] months)

Time	Kaplan–Meier estimate, 95% CI
7‐month FU	18% (12–26)
12‐month FU	47% (37–57)
18‐month FU	81% (69–91)

Figure 2 graphically depicts the Kaplan–Meier analysis estimating the probability of achieving eugonadal status. As represented in Table 2, the estimated probability of achieving eugonadal status was 18% (12, 26), 47% (37, 57) and 81% (69, 91) at 6‐month, 12‐month and 18‐month follow‐up dates, respectively.

FIGURE 2.

Kaplan–Meier analysis estimating the probability of recovering eugonadal status

At multivariable cox regression analyses baseline tT level (HR 1.19; p = 0.03 [1.02–1.4]) was independently associated with the probability of tT level normalization (i.e., eugonadal status) over time after adjusting for age, CCI, BMI and COVID‐19 clinical severity (i.e., Group 2 vs. Group 3) (Table 3).

TABLE 3.

Multivariate cox regression analysis testing the probability of recovering eugonadal status

	MVA
Variable	OR (95% CI)	p‐Value
Age	0.98 (0.95–1.01)	0.2
CCI	0.72 (0.46–1.14)	0.2
BMI	0.91 (0.82–1)	0.06
Baseline tT levels	1.19 (1.02–1.4)	0.03
Group 2 versus Group 3 †	1 (0.21–4.87)	1

Abbreviations: BMI, body mass index; CCI, Charlson Comorbidity Index; tT, total testosterone.

^† Groups were as follows: Group (1) patients in good clinical conditions and discharged home from the emergency department; Group (2) patients who have been admitted in the internal medicine unit until possible discharge at home; and Group (3) patients invasively ventilated in the intensive care unit, and subsequently successfully extubated and discharged either to the internal medicine unit or at home. .

4. DISCUSSION

We report the longest follow‐up ever published in terms of circulating testosterone milieu in men who have recovered from COVID‐19. Current novel findings show that tT levels further significantly increased after 12 months minimum from healing both compared to baseline values ⁶ and circulating levels at 7‐month investigation. ⁹ Clinically relevant, of 63 patients with complete data available, still 30% depicted tT levels suggestive for hypogonadism (i.e., tT < 9.2 nmol/L, as for Endocrine Society threshold criteria ¹⁸ ) even at the longest follow‐up date, despite the time elapsed and the patients being clinically healed. Nevertheless, our observations further stressed the relevance of the time frame during recovery from COVID‐19, since the estimated probability of achieving eugonadal status clearly has increased throughout follow‐up dates. Overall, the higher the baseline tT level, the greater the probability of tT levels recovery at subsequent assessments.

Major strength of the study is that this analysis is the first to investigate a relatively large group of men homogeneously studied during a quite long timeframe since hospital admittance ⁶ ; indeed, here we confirm our previous finding that tT slowly increases over time, but still one of three men with available tT assessment at a minimum 12‐month follow‐up assessment had median concentrations suggestive for hypogonadism. This finding is even of greater clinical importance since we adopted both a tT < 9.2 nmol/L ¹⁸ and a tT < 12 nmol/L ¹⁶ thresholds to define eugonadism. Overall, this observation further supports the plausible clinical impact of tT levels in patients with COVID‐19 ⁷ ; thereof, as previously hypothesized, ⁹ it could also be useful to examine whether applying T therapy in men with viral infections and baseline parameters suggestive of potential worse outcomes and more difficult tT recovery already at the time of hospital admittance may be effective. ⁷ , ²⁰

Here, we may discuss at least a few of our original biopathology hypotheses dealing with the potential importance of low T levels in terms of differences in COVID‐19 severity and clinical outcomes between sexes. ⁶ Indeed, on the one hand, current findings support the concept that tT levels may be a marker of illness severity, with a progressive notwithstanding slow recovery of circulating tT levels, also recapitulating a progressive recovery from COVID‐19‐associated severe multisystem impairment. ⁶ , ²¹ On the other hand, although probably still incomplete, this further recovery of circulating tT levels may only partially exclude a chronic low T pre‐existing condition in those men, which could have somehow facilitated the overall greater incidence, higher severity and greater probability of fatal events in males than in females. ²² In fact, even current findings can neither corroborate nor rebut the possible role of pre‐infection tT levels in terms of subsequent worse clinical outcomes in men with COVID‐19, as already previously extensively argued. ⁹ Nevertheless, although pre‐illness serum T values were not available for our cohort of patients—as likely for no other cohort of patients even published—with up to 70% of men having recovered tT levels suggestive for eugonadism over time, current data would confirm that serum tT had actually and dramatically dropped down after SARS‐CoV‐2 infection. In contrast, we cannot exclude that the most severe forms of COVID‐19 had been developed in those men with serum T concentrations chronically lower than the so‐called reference range even prior to their viral illness. ⁹ , ²³ Once more, here, we also confirmed the clinical importance that the impact of SARS‐CoV‐2 infection in the testes eventually promotes. ³ Interestingly enough, in their recent meta‐regression analysis, Corona et al. ⁷ reported that the presence of SARS‐CoV‐2 mRNA in male genital tract samples was neither influenced by patient age nor by disease severity and nor by comorbid conditions; in contrast, both studies in semen and analyses evaluating the presence of SARS‐CoV‐2 in testis autopsy demonstrated that the probability of having viral mRNA presence significantly decreased with longer time since diagnosis. Thus, considering a potential biological damage of the Leydig cells—both in terms of number and in overall functionality—this meta‐analytic result would support the possibility of a long‐term recovery of testis function, which biologically may explain current findings of a progressive although slow tT levels increase in men during their long‐term clinical recovery from COVID‐19. Lastly, we discussed the hypothesis that the virus‐host interaction mechanism may be associated with sex—that is, chromosomal sex—thus showing that SARS‐CoV‐2 infection per se could decrease serum tT levels in men up to a condition compatible with hypogonadism in 90% of cases already at hospital admission. ⁶ Accordingly, previous observations suggested that tT levels could significantly drop because of acute illnesses, ¹⁹ thus including acute viral infections. ¹⁶ Nevertheless, further studies are necessary to comprehensively support this speculation regarding a hyperacute androgenic collapse.

Our study has a number of limitations. First, although current findings report the longest follow‐up ever published in this setting, this was a single centre‐based observational study, therefore raising the possibility of selection biases and limiting the generalizability of the findings. Second, to reflect real‐life common practice, we elected to measure circulating hormones using commercially available analytical methods. In this context, it is necessary to underline that the use of an immunoassay may be unreliable for the measurement of circulating E2 values in the range of values reported in this study, and therefore current findings must be taken with adequate caution. Third, it has not been possible to follow the whole original group of patients who eventually survived the disease up to the latest assessment date. In this context, it is certainly possible to argue that sicker patients more likely to have hypothalamic–pituitary–gonadal axis abnormalities could be the ones to be eventually followed‐up over time; conversely, those without relevant problems could be the ones to skip subsequent follow‐up visits. However, here, we provide data from the largest cohort ever studied with a rigorous and homogenous approach at a long‐term follow‐up. Third, this study is part of an institutional protocol, which lacks a pre‐infection hormonal milieu assessment in men with COVID‐19. ⁶ , ⁹ Thereof, even if we cannot infer a final interpretation of causality, the observed specific tT levels behaviour over time may support the hypothesis of a direct impact of viral infection in terms of steroidogenesis impairment. Fourth, since signs and symptoms of hypogonadism have not been regularly collected at the baseline and throughout the follow‐up, according to circulating tT levels, we may only suggest a condition suggestive for biochemical hypogonadism. In this context, despite the fact that the definition of hypogonadism usually applied in the literature refers to an established testosterone deficiency condition (i.e., late onset hypogonadism considered as a chronic condition), using the wording of ‘suggestive for biochemical hypogonadism’ is more scientifically accurate. Fifth, our data did not allow to study yet a potential correlation between circulating tT over time and the so‐called long‐COVID syndrome. ²⁴ , ²⁵ Likewise, our data did not allow to investigate a potential correlation between circulating tT both at baseline and over time and the impact of the new variants of SARS‐CoV‐2, which had emerged over the last a few months. ²⁶

5. CONCLUSIONS

Circulating tT levels keep increasing over time in men healing from COVID‐19. Still, almost 30% of men presented with serum T levels suggestive for a condition of biochemical hypogonadism even after 12 months during the recovery period. Of clinical relevance, the lower the T al hospital admittance, the worse the clinical outcomes and the lower the probability of reaching a condition of eugonadism even at long‐term follow‐up.

AUTHOR CONTRIBUTIONS

AS designed and led the study. AS, PC, EP and IS wrote the report. MP, EP, AL, AMF, CA, FB, CC, CC, GAR, CT, ML, GC, LD, AC, AZ, MT, PR‐Q and FC took care of patients and acquired data. AS, PC, SG, IR and FM analysed data and drafted the report.

CONFLICT OF INTEREST

The authors have declared that no conflict of interest exists.

Salonia A, Pontillo M, Capogrosso P, et al. Testosterone in males with COVID‐19: a 12‐month cohort study. Andrology. 2023;11:17–23. 10.1111/andr.13322

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request. cd_value_code=text