A total of 893 studies were found in the electronic searches. Of these, 88 duplicates were identified and removed, resulting in 805 studies. After title and abstract review, 27 studies were included for full reading. Of these, 9 studies remained in this review, the others related to reviews, posters, and proceedings (n = 12), studies including women (n = 2), studies in other languages (n = 2), and older studies (n = 2, although this was not an inclusion criterion two studies were published in the year 1980 and were therefore excluded).
A total of 8,076 men participated in the selected studies. Of these studies, all were published in English. They were published in countries such as Japan, USA, Italy, Brazil, Korea and Switzerland (Table 1).
The sample was composed of adult men aged 18 to 82 years with BMI ranging from 17.3 to 34.6 kg/m2. The main dysfunction evaluated was erectile dysfunction, followed by premature ejaculation. There was a higher prevalence of sexually active men and with comorbidities in the samples, as presented in table 2.
The time of physical activity practice was analyzed and it can be observed in table 3 that the time the participants have practiced physical activity varies from 8 weeks to more than 10 years, showing great variability in relation to the time and type of activity practiced.
Regarding the type of physical activity, walking and cycling were identified as the most frequent, since other activities were not specified, or were assessed as "level of physical activity/physical exercise).
The studies are controversial regarding the practice of cycling, since the articles by Balasubramanian, et al. and Kim; Kim; Kwan, evaluated this practice, and the first concluded that some specific characteristics can influence comfort and sexual function, generating complaints such as pelvic pain and numbness, predisposing practitioners to the onset of ED. The study by Kim; Kim; Kwan concluded that the practice does not offer any harm to the urinary or sexual function in men who cycle recreationally, as observed in table 2.
Two studies did not specify the type of exercise evaluated, because they performed a retrospective evaluation. The study by Mialon used a physical activity questionnaire classifying the participants in low, moderate or high activity level, while Parazzini used a physical activity questionnaire where participants classified self-reported physical activity in none, low, moderate or intense, therefore, in some studies in the next table, variables such as frequency, load and duration were not clarified and were replaced by N/I (not informed). There was a predominance of cross-sectional studies, and in these, the evaluation was done retrospectively using a questionnaire to identify the level of physical activity of the participants, where characteristics such as load or frequency were not assessed.