Erectile dysfunction (ED) is defined as the inability to achieve or maintain a satisfactory penile erection during sexual intercourse (1, 2). The main causes of ED are categorized as either psychogenic (depression, stress and anxiety) or organic (hormonal, neurological, vascular and tissue problems) (3). Hypogonadism is one of the cause of ED, so ED can be caused by any category of hypogonadism including that which occur due to aging (4, 5), non-infectious conditions (obesity, diabetes, hypertension, cancers, malnutrition among others) (4, 6–9), anemia (4, 7), common acute and chronic illnesses (10), weight loss (11), invasion of the glands by pathogens (Hepatitis virus, Cytomegalovirus, HIV among others) (2, 12), cigarette smoking, using drugs, such as opiates, megestrol acetate, methadone (4, 13, 14) and steroids (15), hyperprolactinemia as well as primary and secondary hypogonadism (12, 16, 17). Some of the herbal medicines are known to possess antifertility properties through various mechanisms including inhibiting 5-alpha reductase, a factor that converts testosterone into dihydrotestosterone, reducing gonadotropins and testosterone secretion, increasing the testosterone affinity for sex specific proteins among others (18). Since traditional herbal medicine and complementary alternative medicine are commonly used by treatment naïve HIV patients (19, 20), they can therefore be one of the cofactors for hypogonadism.

The prevalence of ED is higher in men living with human immunodeficiency virus (HIV) (MLWH) than in the general population reaching to 74% in some studies (21). The pathogenesis of ED in these patient is not clear. Hypogonadism is one of the most frequent endocrine disorders in HIV infected men. Low levels of testosterone leads to reduced sexual desire and ED, however some studies have reported poor relationship of testosterone to these sexual function parameters (22, 23). Further, antiretroviral therapy (ART) had been implicated as the cause of sexual problems including erectile dysfunction (24–26) but with contradicting results in other studies (21, 27, 28). In addition, psychological (depression) or neurological (infection, dementia) problems often cause ED (3, 29–33).

Even though ED is more prevalent in MLWH than in the general population, little consideration has been given worldwide to the diagnosis and management of ED among MLWH, and Tanzania inclusive. There is paucity of evidence on the magnitude and factors associated with ED in MLWH in Tanzania. This study is aimed to determine the prevalence of ED and its associated factors among newly diagnosed MLWH, in Mwanza, Northwestern Tanzania.

A total of 388 participants were enrolled in this study, however, the final analysis included 373 newly diagnosed ART naïve MLWH with a median age [IQR] of 40 [33–46] years. Fifteen participants were excluded from the final analysis as they did not complete the IIEF-5 questionnaire for ED assessment. Of the 373 participants, 35% (132/373) were from NDH, 29% (110/373), 28% (105/373) and 7.0% (26/373) from BMC, STRR and MDH respectively. The majority (58.7%, 219/373) of participants were in the age range 31–45 years, were married (64.3%, 240/373) and reported to have not used herbal medicine in the past six months (66.5%, 248/373). The median BMI and WC of the study participants were 21.1 [19.4–23.5] kg/m² and 80 [76–83] cm respectively. About three-quarter of participants had BMI between 18.5 and 24.9 (normal), more than fifty percent had a WC of less than 81 cm (lower) and three-quarters, 75.1% (280/373) were self-employed ( Table 1 ).

Erectile dysfunction (ED) is a global public health concern owing to its negative impact on the well-being of individuals, families, and communities. In this study, the prevalence of ED among ART naïve MLWH was found to be 56.3% (95% CI 52.1%–61.3%). This prevalence is notably higher than the prevalence rate (29.7%) among adult men in the general population (44) but lower than the prevalence rate (74.6%) reported among MLWH from Northern Tanzania (45). This finding is similar to the report by Falade et al. in Nigeria (46), Moreno-perez et al. in Spain (2) and Crum cianflone et al. in the US (28) which reported the prevalence of 57%, 53.4% and 61.2% respectively. On the other hand, our finding was higher than 37.8% and 21.6% rates reported by Adebimpe et al. in Ogbomoso, Southwest Nigeria (35) and Gomes and Brites in Brazil (3) respectively. The variation in these prevalence rates could be attributed to the differences in the research instruments used in assessing ED, characteristics of the study populations as well as sociocultural and geographical differences. Hypogonadism is one of the major cause of erectile dysfunction and report from previous studies have shown the existence of racial/ethnic variations in sex/gonadal hormone which may be due to genetic factors, socio-demographic factors, cultural factors, and/or environmental factors/dietary variations (5, 47–50). Also different tools may be defining ED differently and this might affect the results by either under reporting or over reporting ED cases leading to variations in the finding between different studies (51). For example in a current study, International Index of Erectile Function-5 (IIEF-5) questionnaire with five scored questions (ED: total score< 22). was used but the previous study by Gomes and Brites (3) in Brazil used IIEF-15 questionnaire which have 15 questions (ED: total score< 26) and Adebimpe et al. (35) in Nigeria used just a pre-tested semi structured questionnaire with unscored questions about erectile function.

Regarding categories of ED according to severity among participants, of the 210 subjects with ED in our study, 95 (45.2%) were mild ED, 84 (40.0%) were mild to moderate ED, 24 (11.4%) were moderate ED and 7 (3.3%) had severe ED. A closely similar pattern was reported by Mbwambo et al. in Northern Tanzania (45) and Falade et al. in North central Nigeria (46) where mild, mild-moderate, moderate and severe ED in Tanzania were found in 37.7%, 26.2%, 5.7% and 4.9% patients respectively and in 64.9%, 24%, 8.2% and 2.9% patients respectively in Nigeria. However, in Brazil, Gomes and Brites (3) found that only 13.7% of participants had a mild presentation of ED, while the remaining 86.3% of the study subjects presented with severe ED. Also, in a study done by Enoma et al. in Malaysia (52), ED was almost evenly distributed across all severity grades with severe (24.1%), moderate (19.1%), mild to moderate (20.9%), and mild (18.3%). This variation in the pattern of ED could be due to the use of different patient group, with distinct epidemiological, social and clinical characteristics as different patient group may differ in terms of various factors such as age, comorbid conditions (such as Diabetes mellitus and cardiovascular diseases), ART and other drug status, and life style habits of which have been also shown to have effect on gonadal hormone and erectile function (53–58). However, this requires further research for confirmation.

The risk factors for ED include vascular diseases, hypertension and obesity which are typically more common in the elderly population (59). Several studies have found a significant association between older age and the prevalence of ED (2, 16, 28, 35, 45), however, that was not detected in the current study. The study sample was relatively young (median age 40 [33–46] years) with the majority (58.7%) in the age range of 31 to 45 years. Furthermore, as most of the studies that report a significant association with age, involved HIV patients who were on ART, our study consisted of newly diagnosed ART naïve HIV males. Thus the previously reported relation of ED with young age and the association of ART with ED especially when the regimens contain protease inhibitors (21, 60, 61), could explain the observed discrepancy in this study. However, studies by Gomes and colleague (3) and Shindel et al. (30) among young HIV patients on ART (mean age 44 and 42 years, respectively) showed similar results but both studies included patients with comorbid conditions (Diabetes mellitus, Chronic pulmonary diseases, Hypertension) while Shindel et al. also included HIV-negative subjects in their study.

In our study, no significant association was found between marital status and ED. This finding corresponds to the reports by Gomez et al. (3) in Brazil, Mbwambo et al. (45) in Northern Tanzania, and Falade et al. (46) in Nigeria, and although is in contrast of those other studies which showed significant association between marital status and ED (58, 62–64). The finding from most previous studies with a significant association between marital status and ED indicated that single or divorced men were significantly associated with severe to moderate ED (62–64). Divorce/separation can involve significant changes in a person’s lifestyle, financial status, social connections and intimacy, which can negatively impact physical health including sexual health (62, 65). On the other hand, a study done in Thailand found that married men rate their sexual abilities better than single, divorced, separated, and widowed males (66). The variations in the findings could be due to differences in sociodemographic and clinical characteristics between study groups, as our study involved men living with HIV which is similar for most of previous studies reporting no significant associations and there were a low number of divorced/separated men among our study participants.

In this study, an association between herbal medicine use and ED was not observed. In a previous report, herbal medicine use was found to be significantly associated with hypogonadism in newly diagnosed ART naïve HIV-infected males (67). As hypogonadism is reported to be one of the factors causing ED (22), a positive association between herbal medicine use and ED was anticipated. The lack of such an association between herbal medicine use and ED in this study could be attributed to the small number of participants who reported to use herbal medicine. The higher rate of participants not using herbal medicine recorded in our study could be due to regional variations and the fact that patients with debilitating/systemic diseases were excluded from the study.

Findings of the present study show that employment status is significantly associated with ED and this is in support of the findings of previous studies (3, 58). This finding suggests that the development of ED is prone to be affected by employment status. Particularly, non-self-employed men were 3.7 times more likely to develop ED as compared to self-employed men. This finding could be attributed to financial instability (financial insecurity) leading to stress and anxiety related disorders causing ED.

A statistically significant association was also found between WHO clinical stage (that determines the clinical progression of the disease) and ED whereby MLWH in WHO clinical stage 4 for HIV infection were three times more likely to develop ED as compared to those in stage 1. HIV infection causes impaired immunity with an associated decrease in CD4+ count leading to comorbid conditions, hence poor health status. This observation may be explained by the link between poor health status and impaired gonadal function (6, 68). This finding is in support of the reports by Crum-Cianflone et al., which showed that a higher CD4 count was protective against ED but contrary to the findings by Mbwambo et al. which showed a lack of association between CD4 and ED. This difference could be attributed to the variation in the clinical characteristics of study participants, such as ethnicity, ART status and the prevalence of hypogonadism. Studies have shown racial/ethnic variation in infection acquisition and CD4 cell count as well as viral load and therefore these might reflect the WHO clinical stage for HIV infection in different populations (69–72).

Furthermore, this study shows that nearly half of the study participants had hypogonadism. Several studies have also demonstrated an association between HIV and low serum testosterone (28, 73, 74). Contrary to the previous studies by De Vincentis et al. (23) and Mbwambo et al. (45), the current study revealed a significant association between ED and low testosterone. This discrepancy could be due to variation in the number of participants with low testosterone levels. HIV can cause low testosterone, although currently, the relationship between HIV and testosterone is a controversial one, as other studies have found normal testosterone levels in MLWH (68, 75).

Slightly more than half of ART naïve MLWH were found to have ED with the majority presenting with a mild to mild-moderate form of ED. There was a significant association of ED with WHO clinical stage 4 for HIV infection, lower testosterone level, and being non self-employed. The results of this study emphasizes the need to include ED screening as part of the overall care of MLWH in CTC clinics as well as to strengthen the HIV screening program for early diagnosis and treatment to mitigate most of the observed risk factors for ED.