Long COVID and risk of erectile dysfunction in recovered patients from mild to moderate COVID-19

Patients with coronavirus disease 2019 (COVID-19) were shown to have reduced serum testosterone levels compared to healthy individuals. Low testosterone levels are linked with the development of erectile dysfunction (ED). In this case-controlled study, 20 healthy controls and 39 patients with ED 3 months after recovering from mild-to-moderate COVID-19 pneumonia were studied. The patients ranged in age from 31 to 47 years. To identify early and late COVID-19 infections, real-time polymerase-chain reaction (RT-PCR) and COVID-19 antibody testing were done. The levels of luteinizing hormone (LH), follicular stimulating hormone (FSH), total testosterone (TT), free testosterone (FT), free androgenic index (FAI), and sex hormone-binding globulin (SHBG) were measured. The sexual health inventory for patients (SHIM) score was used to measure the erectile function of the patients and controls. When compared to the controls, the TT serum level in long COVID-19 (LC) patients with ED was low (p = 0.01). In contrast to controls, FT and FAI were both lower in LC patients with ED. (p = 0.001). FSH serum levels did not significantly differ (p = 0.07), but in ED patients, LH serum levels were elevated. SHIM scores were associated with low TT (p = 0.30), FT (p = 0.09), and high LH (p = 0.76) in LC patients with ED. Male patients with decreased serum levels of LH and testosterone may have hypothalamic-pituitary–gonadal axis dysfunction, which could lead to the development of LC-induced ED. Therefore, an in-depth research is necessary to confirm the causal link between COVID-19 and ED in LC patients.

www.nature.com/scientificreports/ Most COVID-19 patients recover with low mortality; however, some patients experience long-term symptoms, described as Long COVID (LC) 6 . LC was originally recognized as COVID-19 patients who had persistent symptoms for weeks after acute SARS-CoV-2 infection. These symptoms included dyspnea, fatigue, myalgia, insomnia, and cognitive and olfactory disorders 7 . However, these symptoms may last for months in some patients and negatively impact their daily activities 8 . LC can be classified as post-acute COVID-19 with symptoms that last longer than 3 weeks after the onset of the illness, or as chronic COVID-19 with symptoms that last more than 12 weeks 9 . However, previous studies have reported that patients with LC had persistent symptoms in varying frequencies and durations among COVID-19 survivor 10,11 . Endocrine disruption in LC or testicular damage may be caused by hypothalamic-pituitary-gonadal (HPG) axis dysfunction brought on by SARS-CoV-2 infection 12 . A case-control study by Okcelik, 2021 found that compared to healthy controls, COVID-19 patients have lower testosterone serum levels 13 . Other research has shown how testosterone regulates human erectile function and male sexual responsiveness 14 . Additionally, erectile dysfunction (ED) has been connected to low testosterone levels 14 . A systematic review and meta-analysis of 204 published articles comprising 2092 COVID-19 patients and 1138 controls showed that SARS-CoV-2 infection is associated with a reduction of total testosterone (TT), and impaired sperm production 15 .
As an end, endothelial dysfunction, low serum TT levels, and SARS-CoV-2 infection and subsequent hypercytokinemia interact to cause ED in patients with LC. A SARS-CoV-2 infection may also result in long-term complications such cardiovascular problems, endothelial dysfunction, and endocrine dysfunction 9,12 . Therefore, early hyper-inflammatory status and associated multi-systemic injuries associated with acute SARS-CoV-2 infection may extend into COVID-19 survivors, causing LC and subsequently ED 8 . The present study hypothesizes that COVID-19 patients may experience long-term cardio metabolic, endocrine and neurological complications which affect the normal physiology of erectile function could result in ED 16 . Different studies revealed that LC patients were associated with development of ED 12,17 . However, the underlying mechanism for development of ED in LC patients was not fully elucidated. Therefore, this study aimed to elucidate the link between ED and testosterone in patients with LC.

Materials and methods
The present case-controlled study involved 39 patients, aged 31-47 years, three months after overcoming mildto-moderate COVID-19 pneumonia with the ED. They were recruited randomly from Al-Shiffa Medical Center, Baghdad, Iraq from June to October 2020, and were compared with matched 20 healthy controls that were recruited among relatives of the patients and outpatient visitors. The selection of patients with ED was performed according to the specific diagnostic criteria reported by Segraves, 2010 18 . The sample data were selected and drawn from a normally distributed population, and the sample size calculation was determined using baseline incidence, population variance, and treatment effect size with alpha = 0.05 and beta as power = 1 − β. The Scientific Jury and Ethical Committee of Mustansiriyah University in Iraq granted permission and gave their seal of approval for the current study (ethical approval number 45QTR/2020) in accordance with the Helsinki Declaration. All participants provided knowledgeable consent for their participation in the study. Following detailed medical histories were collected and physical examinations, the control participants and selected patients were divided up into the following categories. Group A: LC patients with ED (n = 39), Group B: healthy controls (n = 20). The selection of patients with LC was accomplished according to established diagnostic criteria for LC 19 . Inclusion and exclusion criteria. Inclusion criteria. Patients who were male and had healed from COV-ID-19 but still had at least one lasting symptom, who were also diagnosed with ED (3 months after acute COV-ID-19 recovery), and were participating in this research and ranged in age from 31 to 47 years without previous history of ED prior COVID-19.
Exclusion criteria. The following patients were excluded from the study: those with a history of severe COVID-19, diabetes mellitus, hypothyroidism, congenital hypogonadism, chronic liver diseases, acute and chronic kidney diseases, acute and chronic bacterial infections, psychological and mental disorders, history of pelvic trauma, and those receiving treatment with diuretics or β-blockers.
Anthropometric parameters. All participants had their systolic (SBP) and diastolic (DBP) blood pressures taken, and a specific equation was used to compute the pulse pressure (PP) and mean arterial pressure (MAP). PP = SBP-DBP, MAP = DBP + 1/3 PP 20 . Additionally, the explicit formula BMI = body weight (kg)/height (m 2 ) was used to calculate body mass index (BMI) 21 .
Biochemical parameters. All selected patients and controls had an overnight fast before having 10 mL of blood taken. The blood samples were centrifuged for 5 min at 3000 rpm that stored till time of analysis. Early and late SARS-CoV-2 infections were identified using real-time polymerase-chain reaction (RT-PCR) and COVID-19 antibody testing. For both IgG and IgM, the cut-off level was 0.00-0.04 mLU/mL, respectively.
Regarding hormonal assay, in this study, sex hormone-binding globulin (SHBG) and TT were evaluated using ELISA methods (Human testosterone and SHBG, ELISA Kit; MGC126834, Elabscience, Wuhan, China). Free testosterone (FT) was measured according to a previous study 22 . Free androgenic index (FAI) was calculated using the equation FAI = 100 × [TT/SHBG] 23 . Additionally, ELISA techniques were used to measure luteinizing hormone (LH) and follicular stimulating hormone (FSH). Erectile function assessment. LC patients and controls' erectile functioning were calculated using Sexual Health Inventory for Men (SHIM) score 24 . SHIM which is a patient self-administrated questionnaire for assessment of male sexual dysfunction, scoring from 1 to 21 scores, with 98% sensitivity and 88% specificity 24 . The observed SHIM scores in the LC patients ranged from 1 to 21, with the following characterizations: normal = > 21, mild ED is 17-21, moderate ED is 8-16, and severe ED is 1-7. The only observed scores in the LC/ ED group ranged from 1 to 21. However, baseline psychiatric evaluations of the recruited patients were not performed.
Statistical analysis. The mean ± standard deviation (SD), number (n), and percentages (%) used in this study were used to express and present the data. Data analysis was carried out utilizing SPSS, version 24. (IBM Corp., Armonk, NY, USA). To determine the significance of the difference between the LC patients and controls, an unpaired Student's t-test was used. p values < 0.05 were regarded as significant.
Ethics approval. The Scientific Jury and Ethical Committee of Mustansiriyah University in Iraq granted permission and gave their seal of approval for the current study (Ethical Approval Number 45QTR/2020) in accordance with the Helsinki Declaration. All individuals gave their informed consent to participate in the study. Consent to participate. All individuals taking part in the study gave their informed consent.

Results
A total of 65 eligible patients and healthy controls were enrolled in the current study. Six patients were excluded due to diabetes mellitus (n = 3), severe hypertension (n = 2), and unidentified reasons (n = 1). Therefore, a total of n = 59 were recruited and continued the study, with n = 39 post-COVID-19 patients, and n = 20 healthy controls (Fig. 2). In the current investigation, LC patients with ED had blood pressure profiles that were higher than those of the controls (p < 0.01), with the exception of pulse pressure (PP) where the difference was not statistically significant (p = 0.06). Patients with LC have anti-SARS-CoV-2 IgG antibodies, and were associated with other comorbidities, including hypertension (23.07%) and asthma (7.69%). Regarding the patients' response histories  (Table 3).

Discussion
Acute COVID-19 can certainly cause ED due to cytokine storm, endothelial dysfunction, cardiovascular instability, psychological impact, and hypoxia, as well as due to fear of sexual contact 25,26 . This novel study demonstrated the potential link between COVID-19 and ED in individuals who had recovered from mild-to-moderate COVID-19. Here, LC patients with ED were observed to have higher blood pressure profiles compared with healthy controls, with 23% patients being hypertensive and regular users of antihypertensive medications. Nonetheless, and based on previous studies, long-term cardiovascular complications are also linked to SARS-CoV-2 infection 27,28 . Additionally, LC may exacerbate underlying hypertension and other cardiovascular complications   Table 2. Hormone differences between erectile dysfunction patients and controls. # p < 0.05, *p < 0.01. TT total testosterone, FT free testosterone, SHBG sex hormone-binding globulin, FAI free androgenic index, FSH follicular stimulating hormone, LH luteinizing hormone.  www.nature.com/scientificreports/ due to immunological and inflammatory instability 29,30 . It is noteworthy that the effects of LC appear to be very similar to other systemic diseases, such as diabetes mellitus, which cause microvascular dysfunction and autonomic neuropathy 31,32 .

Parameters Controls (mean ± SD) Post-COVID-19 patients (mean ± SD) 95% CI
In the present study, patients with ED have low TT and FT serum levels, and high LH serum levels compared with controls. Okcelik, 2021 confirmed that COVID-19 was associated with low TT serum levels due to suspected testicular injury induced by SARS-CoV-2 13 . It has been shown that Leydig cells and seminiferous tubule damage due to severe SARS-CoV-2 infection may impede spermatogenesis 33 . Therefore, the net effect of testicular injury in COVID-19 may adversely impact the androgenic profile and sex drive in patients with LC. It has been reported that testosterone is the primary regulator of sex desire, motivation, and penile erection in both humans and animals, thus, a reduction in testosterone may lead to the development of ED 34 .
Of note, thrombosis, cytokine storm, and hyperinflammation, SARS-CoV-2 infection may impair the HPG axis 35,36 . This may extend for a period and cause ED and male infertility. Due to a decrease in the production of LH and FSH from the anterior pituitary gland and gonadotropin-releasing factor (GnRF) from the hypothalamus, these pathological alterations may potentially result in central hypogonadism 37 . In the current investigation, LC patients' LH serum levels were lower than those of the controls, indicating dysfunction of the HPG axis. However, FSH serum level was insignificantly higher compared with the controls, which could be due to Leydig and Sertoli cell dysfunction 38 39 . Moreover, other endocrinopathies in LC patients, such as thyroid dysfunction and insulin resistance, may cause functional hypogonadotropic hypogonadism 12,40 . These findings suggest that dysfunction of the HPG axis in LC patients could be the possible mechanism for the development of ED.

. Tabar et al. shown how the ACE2-mediated pathway may be used by SARS-CoV-2 infection to harm Sertoli cells and trigger the onset of male infertility
Notably, down-regulation of ACE2 receptors in the testis and penile cavernous tissue leads to a reduction in testosterone secretion from Leydig cells 41 , and induction of AngII dependent-cavernous vasoconstriction due to reduction of Ang1-7/Mas bioactivity 42 . This is because testicular Ang1-7/Mas complex, which is expressed in Leydig and Sertoli cells, are distorted by high local or circulating AngII 43 . Thus, low testosterone-dependent ED could be caused by RAS dysregulation in COVID-19 44 .
Since high testosterone serum levels increase expressions of the angiotensin-converting enzyme-2/serine transmembrane protease serine 2 (ACE2/TRMPSS2) axes, which is essential for the entry and pathogenesis of SARS-CoV-2, it has been demonstrated that a reduction in testosterone levels in COVID-19 and LC patients may be a counter-balancing protective mechanism against SARS-CoV-2 propagation and induced complications 45 . Additionally, high testosterone serum levels increase the permissive effect of AngII through the expression of angiotensin type 1 receptors 46 . Similar to this, experimental study investigation from Mishra et al., 2016 showed that testosterone injection decreases the expression of the vasodilator angiotensin type 2 receptors (AT2R) in rats 47 . According to past and present experiences, testosterone therapy for the treatment of hypogonadism and ED in LC patients should be avoided because hyperandrogenic status is linked to the severity of COVID-19 48 . Of note, animal and human studies have reported that hypogonadism is linked with pro-inflammatory cytokines which may adversely affect erectile function mainly in aging men because testosterone blocks a variety of cytokines, including TNF-α, IL-1β, and IL-6 49 . Therefore, down-regulation of ACE2 by pro-inflammatory cytokines in hypogonadism may augment the seriousness of COVID-19 in men because of the negative correlation between COVID-19 mortality and the expression of ACE2 50 . However, testosterone exerts dimorphic roles in COVID-19 infections. At optimal levels, it is protective against the severe form for males but otherwise for females. It has been reported that testosterone serum level is reduced by aging and cardio-metabolic diseases including T2DM, obesity, dyslipidemia, heart failure, and atherosclerosis, which are common risk factors for the development of COVID-19 severity 51 . In addition, Pro-inflammatory cytokines, mainly IL-6, are involved in the pathogenesis of acute lung injury (ALI), acute respiratory distress syndrome (ARDS), and cytokine storm-induced multi-organ damage in COVID-19 52 . Of note, testosterone inhibits the synthesis and release of IL-6 and downregulates the expression of IL-6 receptors 51 . Therefore, IL-6 serum level is augmented in hypogonadism patients that increase their susceptibility for COVID-19 severity 53 . Therefore, COVID-19-induced reduction in circulating testosterone may induce ALI due to increase of pro-inflammatory.
Lower SHIM scores were also observed in patients with ED in comparison with the controls. These scores were associated with low TT, FT, FAI, and high LH, as reported by Cappelleri and Rosen, 2005 54 . Therefore, ED in LC patients has a complicated etiology and not simply attributable to low testosterone serum levels, as endothelial dysfunction, oxidative stress, pro-inflammatory cytokine hyper-activation, and high AngII serum levels are also Table 3. Association of sexual health inventory for men in post-COVID-19 patients with erectile dysfunction. TT total testosterone, FT free testosterone, SHBG sex hormone-binding globulin, FAI free androgenic index, FSH follicular stimulating hormone, LH luteinizing hormone. www.nature.com/scientificreports/ interrelated in the onset of ED 55,56 . Furthermore, the response to the effect of phosphodiesterase type 5 inhibitors during sexual excitation could be a possible diagnostic tool for the assessment of ED 57 . Notably, the autonomic nervous system is crucial for the regulation of male erection and ejaculation 58 , and autonomic dysfunction is associated with the development of ED 59 . Interestingly, different studies have revealed that LC patients experience dysautonomic features because auto-antibodies against muscarinic receptors and β adrenoceptors have developed 60,61 . Unfortunately, in the present study, autonomic functions were not assessed precisely. However, the blood pressure of LC patients was higher compared to the controls that might due to dysregulation of RAS by SARS-CoV-2 infection with increasing of pro-inflammatory and vasoconstrictor AngII, and reduction of anti-inflammatory and vasodilator Ang1-7 3,62,63 . Chen et al. 64 from a cohort study confirmed that dysregulation of RAS in acute SARS-CoV-2 infection with increasing AngII could continue for long time and predispose for development of hypertension in LC patients. These verdicts proposed development of autonomic dysfunction in LC patients.
There underlying mechanism of LC-induced ED could be related to the development long-lasting endothelial dysfunction, autonomic neuropathy, endocrine dysfunction, testicular injury, and neuropsychiatric disorders 12,65 . However, the precise cause of ED in LC patients was not fully elucidated that needs further studies.
The present case-controlled study has a variety of limitations, including a limited sample size and a casecontrolled study design that may affect the identification of causality and outcomes. Additionally, biomarkers of endothelial and oxidative stress and the ACE2/TRMPSS2 axis were not evaluated. Evaluation of anxiety, depression, stress levels, and psychological profile of the LC patients were also not conducted, although these factors may all interfere with erectile function. Furthermore, the overnight erection test, injection test, penile colored Doppler test, and ultrasound for ED that is more specific were not performed in the present study. Inflammatory biomarkers were also not estimated although they may reflect inflammatory status in LC patients. Given that ED, regardless of covid-19, increases with age, particularly after the age of 40, though we not have considered stratifying by age. In particular, LC patients with ED were not compared with LC patients without ED that is an integral point to show effect of LC on the hormonal changes and their relation in the development of ED. Nonetheless, despite of these limitations, strength of this case-controlled study it was regarded as an initial step toward authenticating the underlying causes of ED in LC patients. The present study confirmed that LC patients may experience a rate of ED due to reduction of testosterone levels.

Conclusions
LC-induced ED may develop due to dysfunction of the HPA axis in male patients, as evidenced by reduced testosterone and LH serum levels. Therefore, in-depth research is required to confirm the causal link between COVID-19 and ED in LC patients.
Clinical significance. The COVID-19 pandemic is brought on by SARS-CoV-2. There have been claims that COVID-19 patients have lower testosterone serum levels when compared with healthy controls. Additionally, low testosterone levels are linked with ED, so testosterone therapy may improve sexual function and prevent or reduce ED.

Data availability
All data generated or analyzed during this study are included in this published article.