Abstract
Erectile dysfunction (ED) is a common condition affecting men worldwide. The U.S. Food and Drug Administration (FDA) has approved phosphodiesterase type 5 inhibitors (PDE5Is), including sildenafil, tadalafil, vardenafil, and avanafil, for treating ED. However, real-world studies on adverse events (AEs) linked to PDE5Is are limited. This study comprehensively assessed the safety of PDE5Is based on reports from the FDA adverse event reporting system (FAERS) database from January 2004 to June 2024. The disproportionality analysis was used to evaluate the safety profiles of PDE5Is. Based on demographic stratification, correlational analysis and signal differences examination in subgroups were performed in different PDE5Is. Among the 53 517 AEs reports collected from the FAERS database, we identified 135, 73, 72, and 7 preferred terms associated with sildenafil, tadalafil, vardenafil, and avanafil, respectively. The study detected AEs listed on the FDA-approved label of each PDE5I. However, some AEs not listed on the labels were also identified. Some AEs listed by the FDA for PDE5Is were insignificant signals in our analysis. Significant differences were found among PDE5Is across age, weight, and onset time categories. We detected AEs related to the nervous, cardiovascular, and ocular systems that were not listed on the labels of the four PDE5Is, warranting further research on the underlying mechanisms. Additionally, significant differences in PDE5I-associated AEs were observed across age, weight, and onset time, highlighting the need for tailored patient management.
Keywords: adverse event, erectile dysfunction, FAERS, PDE5I, real-world study
INTRODUCTION
Erectile dysfunction (ED) is defined as the consistent or recurrent inability to attain and/or maintain penile erection sufficient for sexual satisfaction.1 Epidemiological data reveal an age-dependent prevalence gradient, affecting 30% of males aged <40 years2 and 52% of males aged 40–70 years.3 While non-life-threatening, ED substantially compromises patients’ quality of life and interpersonal relationships. The current American Urological Association guidelines endorse a therapeutic encompassing, including psychosexual counseling, vacuum erection device therapy, and pharmacological interventions.4 Notably, phosphodiesterase type 5 inhibitors (PDE5Is) are recommended as the first-line regimen. The U.S. Food and Drug Administration (FDA) has approved four PDE5Is agents for ED management: sildenafil, tadalafil, vardenafil, and avanafil, each demonstrating distinct pharmacokinetic profiles.
Phosphodiesterase type 5 (PDE5), an enzyme responsible for degrading cyclic guanosine monophosphate (cGMP), regulates nitric oxide (NO)-cGMP signaling. NO-cGMP is a major signaling pathway for initiating and sustaining penile erection. PDE5Is exert therapeutic effects through selective inhibition of cGMP catabolism, thereby augmenting corpus cavernosum smooth muscle relaxation.5 FDA has approved four PDE5Is with distinct pharmacokinetic profiles for erectogenic therapy: sildenafil, tadalafil, vardenafil, and avanafil.4 Beyond their urological indications, PDE5Is demonstrate pleiotropic effects through cGMP-mediated pathways. Approved therapeutic applications encompass pulmonary arterial hypertension, high-altitude illness, angina, benign prostatic hyperplasia (BPH), and lower urinary tract symptoms. Emerging evidence suggests potential neurovascular protection in ischemic stroke and peripheral neuropathy.6 While previous studies have confirmed the short-term and long-term safety of PDE5Is,7 real-world pharmacovigilance data lack. Despite the wide range of PDE5Is applications in ED and BPH,8 comprehensive data evaluating the safety of PED5Is in the real world remain scarce.
The FDA adverse event reporting system (FAERS), a publicly accessible pharmacovigilance repository, constitutes the largest global database of spontaneously reported adverse events (AEs). This database collects AEs reports related to drug use from both the USA and other countries.9 Given its large size, global data source, and disproportionality analysis capabilities, FAERS database is a good tool for exploring potential associations between AEs and drugs. A comprehensive pharmacovigilance study using the FAERS database demonstrated notable accuracy. Leveraging the FAERS database, we conducted a retrospective disproportionality analysis to systematically evaluate the pharmacovigilance profiles of PDE5Is. It is essential to identify targeted patient demographics, optimal treatment regimens, and appropriate treatment durations. Our findings offer valuable insights that can guide medical stuffs in optimizing the use of PDE5Is.
MATERIALS AND METHODS
Data sources
This study is based on the FAERS database. FAERS database, established and managed by the FDA, contains demographic information, drug details, treatment indications, AEs, patient outcomes, manufacturer information, and report sources. Data are reported by physicians, pharmacists, consumers, and other sources. Since this study utilized publicly available database and did not involve direct interaction with human subjects, patient consent and institutional review board approval were not required.
Reports from the first quarter of 2004 (earliest following FDA market approval of sildenafil, tadalafil, and vardenafil) to the second quarter of 2024 (the most recent FAERS update at the time of the study) were extracted. To retrieve reports on PDE5Is from the FAERS drug information files, both generic names (sildenafil, tadalafil, vardenafil, and avanafil) and brand names (Viagra, Cialis, Levitra, and Stendra) were used. Reports were extracted with a ROLE_COD of primary suspect (PS) to ensure study credibility. To specifically identify ED-related reports, brand names not indicated for ED and reports from females were excluded. Extracted AEs were recorded using Primary Terms (PT) and System Organ Class (SOC) classifications from the Medical Dictionary for Regulatory Activities code.
Additionally, demographic information, including country, year, and weight, was collected from the reports. The onset time of events was calculated as the difference between the end time and start time. Study outcomes were also included. Deduplication based on the primary AE report identification number (PRIMARYID) and report removal were performed as recommended by FAERS.10 When multiple reports shared the same CASEID, reports with the highest PRIMARYID and most recent FDA_DT were retained, while others were excluded as FAERS recommended.
Statistical analyses
This study used descriptive analysis to examine the characteristics of PDE5I-associated AE reports. Descriptive analysis included age, weight, reported country, reporting year, and outcomes. As the denominator cannot be provided by the FAERS, disproportionality analysis, which is commonly used in pharmacovigilance studies, was deemed suitable for this study.11 To explore potential AE signals of PDE5Is, the comparator group was the frequency of the same AE observed with other drugs. Disproportionality analysis was conducted using the reporting odds ratio (ROR) and proportional reporting ratio (PRR) with their corresponding 95% confidence intervals (CIs) through the case/non-case method.12 For the ROR, a significant signal is identified when the number of events exceeds three and the lower limit of the 95% CI is greater than one. For the PRR, a significant signal is detected when the number of reports exceeds three, the Chi-square value is greater than four, and the PRR is greater than two. An AE is considered a significant signal when both the ROR and PRR meet the above thresholds simultaneously. Two-by-two contingency tables were used to calculate the ROR and PRR. The Kruskal-Wallis and Bonferroni-Dunn tests were used to analyze correlations between the four PDE5Is and age, weight, and onset time to event. In the Bonferroni–Dunn test, a significant difference was identified when the P-value was less than 0.008. Additionally, distinctions in AE signals associated with each PDE5I were analyzed based on age, weight, and onset time to event. Fisher’s exact test and the ROR algorithm were used for this analysis. The fourfold table used in this study is provided in Supplementary Table 1. All the data analyses were conducted using the R software (version 4.4.1; The R Foundation, Vienna, Austria).
Supplementary Table 1.
Fourfold table of the disproportionality analysis for phosphodiesterase type 5 inhibitors signal detection
| Reports with target adverse event | Reports with other adverse events | |
|---|---|---|
| Reports with target drug | a | b |
| Reports with other drugs | c | d |
a: number of reports containing both target drug and target adverse event; b: number of reports containing other adverse events of target drug; c: number of reports containing target adverse event of other drugs; d: number of reports containing other drugs and other adverse events
RESULTS
General characteristics
A total of 17 956 653 cases were reported during the study period. After applying the exclusion criteria, 6 279 585 reports were retained in the analysis. A total of 21 779 cases and 53 517 AEs reports were identified (Figure 1). Figure 2 and Supplementary Table 2 present a comprehensive overview of the study population’s general characteristics.
Figure 1.
Flow diagram of data collection and analysis of PDE5Is-associated adverse event reports. PDE5Is: phosphodiesterase type 5 inhibitors; DEMO: demographic information; DRUG: drug information; REAC: adverse events; INDI: indication; PS: primary suspect; ROR: reporting odds ratio; PRR: proportional reporting ratio.
Figure 2.
Clinical characteristics of PDE5Is associated reports from the FAERS database. (a) Reporting countries of PDE5Is. (b) Age of patients in the reports associated with PDE5Is. (c) Weight of patients in PDE5Is-associated reports. (d) Serious outcome of PDE5Is-associated reports. (e) Onset time to event of PDE5Is-associated reports. PDE5Is: phosphodiesterase type 5 inhibitors; FAERS: the FDA adverse event reporting system.
Supplementary Table 2.
General characteristics
| Characteristics | Sildenafil | Tadalafil | Vardenafil | Avanafil | Total |
|---|---|---|---|---|---|
| AE cases | 10 011 | 8685 | 2716 | 379 | 21 791 |
| Age (years), n (%) | |||||
| <18 | 8 (0.1) | 2 (0.0) | 0 (0.0) | 0 (0.0) | 10 (0.0) |
| 18≤age<45 | 834 (11.4) | 608 (10.8) | 219 (10.2) | 16 (5.6) | 1677 (7.7) |
| 45≤age<65 | 4021 (55.0) | 3256 (57.6) | 925 (43.2) | 106 (36.9) | 8308 (38.1) |
| ≥65 | 2451 (33.5) | 1785 (31.6) | 995 (46.5) | 165 (57.5) | 5396 (24.8) |
| Unknown | 2697 | 3034 | 577 | 92 | 6400 (29.4) |
| Weight (kg), n (%) | |||||
| <80 | 1558 (36.3) | 447 (37.6) | 76 (13.2) | 79 (52.7) | 2160 (9.9) |
| 80≤weight<100 | 588 (13.7) | 369 (31.0) | 68 (11.8) | 9 (6.0) | 1034 (4.7) |
| ≥100 | 2148 (50.0) | 374 (31.4) | 431 (75.0) | 62 (41.3) | 3015 (13.8) |
| Unknown | 5717 | 7495 | 2141 | 229 | 15 582 (71.5) |
| Reported countries (top 5), n (%) | |||||
| USA | 8913 (89.0) | 6413 (73.8) | 1616 (59.5) | 341 (90.0) | 17 283 (79.3) |
| France | 51 (0.5) | 247 (2.8) | 16 (0.6) | 15 (4.0) | 329 (1.5) |
| Germany | 46 (0.5) | 187 (2.2) | 88 (3.2) | 2 (0.5) | 323 (1.5) |
| Brazil | 173 (1.7) | 98 (1.1) | 10 (0.4) | 0 (0.0) | 281 (1.3) |
| Canada | 41 (0.4) | 104 (1.2) | 7 (0.3) | 0 (0.0) | 152 (0.7) |
| Outcomes, n (%) | |||||
| Death | 277 (2.8) | 158 (1.8) | 30 (1.1) | 2 (0.5) | 467 (2.1) |
| Disability | 221 (2.2) | 144 (1.7) | 49 (1.8) | 10 (2.6) | 424 (1.9) |
| Hospitalization | 720 (7.2) | 567 (6.5) | 88 (3.2) | 9 (2.4) | 1384 (6.4) |
| Life-threatening | 64 (0.6) | 95 (1.1) | 34 (1.3) | 1 (0.3) | 194 (0.9) |
| Required intervention | 25 (0.2) | 16 (0.2) | 14 (0.5) | 0 (0.0) | 55 (0.3) |
| Onset time to events (days), n (%) | |||||
| <30 | 179 (12.7) | 512 (40.3) | 187 (37.1) | 14 (73.7) | 892 (27.9) |
| 30≤time<60 | 79 (5.6) | 89 (7.0) | 68 (13.5) | 1 (5.3) | 237 (7.4) |
| 60≤time<90 | 45 (3.2) | 48 (3.8) | 28 (5.6) | 0 (0.0) | 121 (3.8) |
| 90≤time<180 | 71 (5.0) | 81 (6.4) | 60 (11.9) | 4 (21.1) | 216 (6.8) |
| 180≤time<365 | 110 (7.8) | 121 (9.5) | 55 (10.9) | 0 (0.0) | 286 (8.9) |
| ≥365 | 924 (65.6) | 418 (32.9) | 106 (21.0) | 0 (0.0) | 1448 (45.3) |
AE: adverse event
As summarized in Supplementary Table 2 and depicted in Figure 2a, the USA (79.3%) submitted the highest number of reports, followed by France (1.5%), Germany (1.5%), Brazil (1.3%), and Canada (0.7%). Among all PDE5Is in this study, the USA accounted for the largest proportion of reports.
Regarding age, as shown in Figure 2b most reports involved individuals aged 45–65 years (38.1%), followed by those aged ≥65 years (24.8%) and 18–45 years (7.7%). Similar to the general analysis, most reports for sildenafil (55.0%) and tadalafil (57.6%) involved individuals aged 45–65 years. However, for vardenafil and avanafil, majority of the reports were from individuals aged ≥65 years (Supplementary Table 2).
Weight data were obtained from 6209 patients (Figure 2c). Among these participants, the majority (3015, 13.8%) weighed more than 100 kg, and 9.91% of them weighed less than 80 kg. When analyzed by medication type, sildenafil (50.0%) showed the highest proportion in individuals with weight ≥100 kg. In contrast, tadalafil (37.6%) and avanafil (52.7%) were predominantly prescribed to patients weighing <80 kg (Supplementary Table 2).
For serious outcomes, as depicted in Figure 2d, hospitalization (6.4%) and death (2.1%) were the most common serious outcomes (Supplementary Table 2). As for AEs onset, 45.3% of AEs (1435 reports) occurred one year after medication initiation (Figure 2e). Most tadalafil and vardenafil AE reports emerged within 30 days, contrasting the overall trend (Supplementary Table 2). As shown in Figure 3, most reports were submitted in 2015.
Figure 3.
Number of reports from 2004 to 2024.
Correlation analysis
Correlation analysis was conducted to examine the relationships between the use of PDE5I and patient age, weight and onset time to events. The Kruskal-Wallis test was first performed on the 21 779 AEs reports to assess differences among each PDE5I. The Dunn-Bonferroni correction was then applied to evaluate the differences among the PDE5Is. For the correlation between PDE5Is and age, the Kruskal-Wallis test yielded P < 0.001.
As depicted in Figure 4a, significant differences in weight were observed between sildenafil and tadalafil, between sildenafil and vardenafil, between tadalafil and vardenafil, and between vardenafil and avanafil (all P < 0.001). In general, patients using tadalafil had the lowest weight, while those using vardenafil had the highest. Stratified by age (Figure 4b), statistically significant differences were observed between sildenafil and vardenafil, between sildenafil and avanafil, between tadalafil and vardenafil, between tadalafil and avanafil, and between vardenafil and avanafil (all P < 0.008). Avanafil users were the oldest, followed by vardenafil, whereas sildenafil and tadalafil users were the youngest. As shown in Figure 4c, significant differences in onset time were observed between sildenafil and tadalafil, between sildenafil and vardenafil, and between tadalafil and avanafil (all P < 0.008). Overall, sildenafil had the longest onset time, while avanafil had the shortest.
Figure 4.
Correlation of demographic and event onset characteristics across PDE5Is. (a) Violin plots showing weight distribution in users of sildenafil, tadalafil, and vardenafil. (b) Violin plots showing age distribution in users of sildenafil, tadalafil, and vardenafil. (c) Violin plots showing onset time to events distribution in users of sildenafil, tadalafil, and vardenafil. **P < 0.008, ***P < 0.001. PDE5Is: phosphodiesterase type 5 inhibitors.
Signal detection
Each PDE5Is has specific AEs listed on the FDA-approved label, which are considered expected when detected in our analysis. Besides, some AEs identified in the present study were not listed on the FDA-approved label and are therefore considered unexpected.
A total of 53 517 AE reports, recorded as PTs, were collected during the study period. Across all four drugs, 109 PTs were identified as significant signals. The ten most frequently reported AEs included headache, hypertension, erection increased, blood cholesterol increased, asthenia, flushing, malignant melanoma, back pain, vision blurred, and blood testosterone decreased.
A total of 135 PTs were identified as significant signals for sildenafil. Among these, AEs listed on the label, including ejaculation disorder (PT: 10014326), penile swelling (PT: 10034319), hepatic function abnormal (PT: 10019670), and cataract (PT: 10007739) were identified in the present study. However, AEs reported on the label, such as headache, flushing, dyspepsia, nasal congestion, dizziness, and rash published during drug labeling, were not identified as significant signals in our study. Additionally, unexpected AEs, including erection increased, malignant melanoma, optic ischemic neuropathy, and blindness, were detected (Supplementary Table 3).
Supplementary Table 3.
Significant signals of sildenafil
| SOC | PT | n | ROR (95% CI) | PRR (χ2) |
|---|---|---|---|---|
| Cardiac disorders | Coronary artery occlusion | 46 | 3.9 (2.4–6.2) | 3.9 (38.1) |
| Cardiac valve disease | 6 | 3.7 (1.0–13.0) | 3.7 (4.7) | |
| Cardiac flutter | 5 | 6.1 (1.2–31.5) | 6.1 (6.1) | |
| Endocrine disorders | Thyroid disorder | 7 | 5.7 (1.5–22.1) | 5.7 (8.2) |
| Eye disorders | Optic ischemic neuropathy | 226 | 4.4 (3.5–5.4) | 4.3 (209.5) |
| Blindness | 184 | 4.7 (3.7–6.0) | 4.7 (183.8) | |
| Cyanopsia | 153 | 5.9 (4.4–7.9) | 5.9 (182.0) | |
| Blindness unilateral | 122 | 3.2 (2.5–4.2) | 3.2 (80.5) | |
| Visual field defect | 71 | 3.1 (2.2–4.4) | 3.1 (44.7) | |
| Cataract | 65 | 3.0 (2.1–4.3) | 3.0 (39.0) | |
| Retinal vein occlusion | 39 | 3.7 (2.2–6.0) | 3.7 (30.3) | |
| Glaucoma | 37 | 2.5 (1.6–3.9) | 2.5 (15.9) | |
| Chromatopsia | 36 | 3.4 (2.1–5.6) | 3.4 (25.4) | |
| Macular degeneration | 31 | 2.3 (1.4–3.8) | 2.3 (11.7) | |
| Papilledema | 27 | 3.5 (1.9–6.3) | 3.5 (19.7) | |
| Vitreous floaters | 24 | 2.3 (1.3–3.9) | 2.3 (8.8) | |
| Retinal hemorrhage | 23 | 2.7 (1.5–4.9) | 2.7 (11.6) | |
| Retinal artery occlusion | 18 | 2.5 (1.3–4.7) | 2.5 (7.7) | |
| Optic atrophy | 18 | 4.4 (2.0–9.6) | 4.4 (16.9) | |
| Optic nerve disorder | 17 | 4.2 (1.9–9.1) | 4.2 (15.1) | |
| Visual brightness | 16 | 3.3 (1.5–6.9) | 3.3 (10.8) | |
| Ocular vascular disorder | 15 | 3.7 (1.7–8.2) | 3.7 (11.7) | |
| Retinal vein thrombosis | 10 | 3.5 (1.3–9.2) | 3.5 (7.3) | |
| Optic disc disorder | 10 | 8.2 (2.3–29.7) | 8.2 (14.5) | |
| Chloropsia | 7 | 17.1 (2.1–139.3) | 17.1 (13.3) | |
| Retinopathy hypertensive | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| General disorders and administration site conditions | Death | 95 | 2.4 (1.8–3.2) | 2.4 (38.4) |
| Therapeutic response unexpected | 83 | 3.6 (2.6–5.0) | 3.6 (62.4) | |
| Pre-existing condition improved | 45 | 15.8 (7.1–35.0) | 15.7 (83.6) | |
| Food interaction | 13 | 5.3 (2.0–14.0) | 5.3 (14.3) | |
| Drug tolerance | 13 | 4.0 (1.7–9.6) | 4.0 (11.0) | |
| Alcohol interaction | 10 | 24.5 (3.1–191.3) | 24.5 (20.5) | |
| Hepatobiliary disorders | Hepatic function abnormal | 6 | 4.9 (1.2–19.6) | 4.9 (6.2) |
| Infections and infestations | Nasopharyngitis | 66 | 2.3 (1.7–3.3) | 2.3 (26.0) |
| Arthritis infective | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| Orchitis | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| Injury, poisoning, and procedural complications | Intentional product use issue | 150 | 7.5 (5.5–10.4) | 7.5 (208.3) |
| Incorrect dose administered | 130 | 2.2 (1.8–2.8) | 2.2 (45.4) | |
| Road traffic accident | 46 | 2.5 (1.7–3.8) | 2.5 (20.5) | |
| Accident | 22 | 4.1 (2.1–8.2) | 4.1 (19.5) | |
| Post procedural complication | 18 | 5.5 (2.4–12.7) | 5.5 (20.4) | |
| Nerve injury | 18 | 2.3 (1.2–4.4) | 2.3 (6.9) | |
| Scar | 17 | 2.6 (1.3–5.2) | 2.6 (8.1) | |
| Back injury | 15 | 2.6 (1.3–5.4) | 2.6 (7.3) | |
| Lower limb fracture | 14 | 3.1 (1.4–6.9) | 3.1 (8.9) | |
| Product dispensing error | 14 | 5.7 (2.2–14.9) | 5.7 (16.3) | |
| Circumstance or information capable of leading to medication error | 13 | 10.6 (3.0–37.2) | 10.6 (21.2) | |
| Joint injury | 12 | 2.9 (1.3–6.8) | 2.9 (7.0) | |
| Accident at work | 12 | 14.7 (3.3–65.6) | 14.7 (21.9) | |
| Arthropod bite | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Upper limb fracture | 7 | 5.7 (1.5–22.1) | 5.7 (8.2) | |
| Meniscus injury | 6 | 7.3 (1.5–36.4) | 7.3 (8.2) | |
| Surgical procedure repeated | 6 | 14.7 (1.8–122.0) | 14.7 (10.9) | |
| Procedural complication | 5 | 6.1 (1.2–31.5) | 6.1 (6.1) | |
| Investigations | Semen volume increased | 35 | 14.3 (6.0–34.0) | 14.3 (63.3) |
| Blood triglycerides increased | 20 | 3.5 (1.8–6.9) | 3.5 (14.7) | |
| Low density lipoprotein increased | 12 | 2.5 (1.1–5.5) | 2.5 (5.1) | |
| Blood cholesterol abnormal | 7 | 8.6 (1.8–41.3) | 8.6 (10.4) | |
| Prostate examination abnormal | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Semen analysis abnormal | 6 | 7.3 (1.5–36.4) | 7.3 (8.2) | |
| Cardiac stress test abnormal | 5 | 6.1 (1.2–31.5) | 6.1 (6.1) | |
| Metabolism and nutrition disorders | Glucose tolerance impaired | 12 | 4.2 (1.7–10.7) | 4.2 (10.8) |
| Weight fluctuation | 8 | 4.9 (1.5–16.3) | 4.9 (8.3) | |
| Metabolic disorder | 5 | 6.1 (1.2–31.5) | 6.1 (6.1) | |
| Musculoskeletal and connective tissue disorders | Arthropathy | 34 | 4.9 (2.7–8.8) | 4.9 (35.1) |
| Back disorder | 29 | 5.5 (2.8–10.5) | 5.5 (32.7) | |
| Musculoskeletal disorder | 15 | 6.1 (2.4–15.8) | 6.1 (18.4) | |
| Exostosis | 6 | 4.9 (1.2–19.6) | 4.9 (6.2) | |
| Neoplasms benign, malignant, and unspecified (incl cysts and polyps) | Malignant melanoma | 586 | 7.2 (6.1–8.4) | 7.0 (788.3) |
| Prostate cancer | 170 | 3.1 (2.5–4.0) | 3.1 (108.5) | |
| Malignant melanoma in situ | 147 | 4.0 (3.1–5.2) | 4.0 (124.4) | |
| Metastatic malignant melanoma | 70 | 4.2 (2.9–6.2) | 4.2 (62.6) | |
| Neoplasm malignant | 41 | 2.4 (1.6–3.7) | 2.4 (16.8) | |
| Malignant melanoma stage iv | 41 | 2.9 (1.8–4.5) | 2.9 (23.0) | |
| Malignant melanoma stage iii | 31 | 2.9 (1.7–4.9) | 2.9 (17.8) | |
| Lentigo maligna | 31 | 5.1 (2.7–9.4) | 5.1 (32.9) | |
| Neoplasm progression | 24 | 14.7 (5.1–42.4) | 14.7 (43.7) | |
| Metastases to lymph nodes | 12 | 7.3 (2.4–22.8) | 7.3 (16.4) | |
| Throat cancer | 7 | 5.7 (1.5–22.1) | 5.7 (8.2) | |
| Metastases to lung | 7 | 5.7 (1.5–22.1) | 5.7 (8.2) | |
| Prostate cancer recurrent | 6 | 14.7 (1.8–122.0) | 14.7 (10.9) | |
| Laryngeal cancer | 6 | 14.7 (1.8–122.0) | 14.7 (10.9) | |
| Leukemia | 5 | 6.1 (1.2–31.5) | 6.1 (6.1) | |
| Hepatic neoplasm | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| Nervous system disorders | Multiple sclerosis | 18 | 2.9 (1.5–5.8) | 2.9 (10.5) |
| Optic neuritis | 17 | 3.0 (1.5–6.0) | 3.0 (10.1) | |
| Coma | 16 | 3.0 (1.5–6.3) | 3.0 (9.6) | |
| Nerve compression | 13 | 3.5 (1.5–8.3) | 3.5 (9.7) | |
| Carotid artery occlusion | 9 | 3.2 (1.2–8.5) | 3.2 (5.8) | |
| Product issues | Suspected counterfeit product | 147 | 2.7 (2.2–3.5) | 2.7 (76.2) |
| Product counterfeit | 24 | 5.3 (2.6–10.9) | 5.3 (26.6) | |
| Product taste abnormal | 21 | 4.3 (2.1–8.7) | 4.3 (19.2) | |
| Product color issue | 8 | 3.9 (1.3–12.0) | 3.9 (6.7) | |
| Psychiatric disorders | Emotional distress | 69 | 2.2 (1.6–3.0) | 2.2 (23.7) |
| Premature ejaculation | 34 | 2.5 (1.6–4.1) | 2.5 (15.4) | |
| Male orgasmic disorder | 24 | 3.7 (2.0–6.9) | 3.7 (18.7) | |
| Reproductive system and breast disorders | Erection increased | 785 | 7.9 (6.8–9.1) | 7.7 (1114.8) |
| Ejaculation disorder | 412 | 14.7 (11.4–18.9) | 14.4 (750.9) | |
| Penis disorder | 161 | 3.7 (2.9–4.8) | 3.7 (127.4) | |
| Painful erection | 53 | 2.0 (1.4–2.9) | 2.0 (14.5) | |
| Penile swelling | 35 | 2.3 (1.4–3.6) | 2.3 (12.7) | |
| Prostatic disorder | 32 | 4.6 (2.6–8.3) | 4.6 (31.4) | |
| Peyronie’s disease | 30 | 2.5 (1.5–4.1) | 2.5 (12.9) | |
| Testicular disorder | 25 | 10.2 (4.2–24.9) | 10.2 (40.2) | |
| Testicular swelling | 23 | 3.8 (2.0–7.2) | 3.8 (18.3) | |
| Penile vascular disorder | 11 | 5.4 (1.9–15.5) | 5.4 (12.3) | |
| Semen discoloration | 10 | 4.1 (1.5–11.2) | 4.1 (8.7) | |
| Pruritus genital | 9 | 11.0 (2.4–51.0) | 11.0 (14.9) | |
| Spontaneous ejaculation | 5 | 12.2 (1.4–104.8) | 12.2 (8.6) | |
| Scrotal disorder | 5 | 12.2 (1.4–104.8) | 12.2 (8.6) | |
| Penile odor | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| Respiratory, thoracic and mediastinal disorders | Sinus disorder | 17 | 2.1 (1.1–4.0) | 2.1 (5.2) |
| Respiratory disorder | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Respiratory distress | 5 | 12.2 (1.4–104.8) | 12.2 (8.6) | |
| Skin and subcutaneous tissue disorders | Skin discoloration | 29 | 2.1 (1.3–3.4) | 2.1 (8.9) |
| Skin odor abnormal | 12 | 5.9 (2.1–16.7) | 5.9 (14.3) | |
| Skin disorder | 9 | 4.4 (1.5–13.2) | 4.4 (8.5) | |
| Social circumstances | Impaired driving ability | 14 | 2.6 (1.2–5.6) | 2.6 (6.8) |
| Alcohol use | 13 | 2.5 (1.1–5.3) | 2.5 (5.6) | |
| Surgical and medical procedures | Prostatic operation | 15 | 6.1 (2.4–15.8) | 6.1 (18.4) |
| Hip arthroplasty | 15 | 6.1 (2.4–15.8) | 6.1 (18.4) | |
| Self-medication | 14 | 2.3 (1.1–4.7) | 2.3 (5.2) | |
| Stent placement | 11 | 3.0 (1.2–7.2) | 3.0 (6.6) | |
| Knee arthroplasty | 11 | 3.9 (1.5–9.9) | 3.9 (9.0) | |
| Cataract operation | 10 | 4.1 (1.5–11.2) | 4.1 (8.7) | |
| Dialysis | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Hernia repair | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Carpal tunnel decompression | 4 | 9.8 (1.1–87.6) | 9.8 (6.3) | |
| Vascular disorders | Arterial occlusive disease | 26 | 5.8 (2.9–11.7) | 5.8 (30.6) |
| Infarction | 14 | 3.4 (1.5–7.7) | 3.4 (10.0) | |
| Arterial disorder | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Vein disorder | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Poor peripheral circulation | 7 | 3.4 (1.1–10.8) | 3.4 (5.0) | |
| Aneurysm | 7 | 5.7 (1.5–22.1) | 5.7 (8.2) | |
| Venous occlusion | 6 | 14.7 (1.8–122.0) | 14.7 (10.9) |
ROR: reporting odd ratio; CI: confidence interval; PRR: proportional reporting ratio; SOC: System Organ Class; PT: primary terms; n: number
AEs listed on the label for tadalafil included asthenia (PT: 10003549), headache (PT: 10019211), dyspepsia (PT: 10013946), back pain (PT: 10003988), gastroesophageal reflux disease (PT: 10066874), pain in extremity (PT: 10033425), myalgia (PT: 10028411), hypertension (PT: 10020772), and vision blurred (PT: 10047513). Notably, this study identified several unexpected AEs associated with tadalafil, including increased blood cholesterol, depression, libido disorder, and decreased body height, among others. However, expected AEs, including dizziness, diarrhea, abdominal pain, pain in limb, and flushing listed, were not detected as significant signals (Supplementary Table 4).
Supplementary Table 4.
Significant signals of tadalafil
| SOC | PT | n | ROR (95% CI) | PRR (χ2) |
|---|---|---|---|---|
| Cardiac disorders | Cardiac disorder | 178 | 4.6 (3.5–5.9) | 4.5 (158.5) |
| Torsade de pointes | 9 | 9.7 (2.1–45.0) | 9.7 (12.8) | |
| Ear and labyrinth disorders | Sudden hearing loss | 24 | 2.5 (1.4–4.4) | 2.5 (9.8) |
| Ear discomfort | 17 | 2.3 (1.2–4.6) | 2.3 (6.0) | |
| Deafness bilateral | 8 | 4.3 (1.3–14.4) | 4.3 (6.8) | |
| Eye disorders | Diplopia | 37 | 2.1 (1.3–3.3) | 2.1 (10.9) |
| Dry eye | 19 | 2.6 (1.3–5.0) | 2.6 (8.3) | |
| Eyelid edema | 16 | 5.8 (2.3–14.7) | 5.8 (17.2) | |
| Eye swelling | 15 | 2.5 (1.2–5.2) | 2.5 (6.2) | |
| Ocular hypertension | 12 | 5.2 (1.8–14.7) | 5.2 (11.9) | |
| Eye inflammation | 9 | 3.2 (1.2–9.1) | 3.2 (5.6) | |
| Central serous chorioretinopathy | 8 | 17.3 (2.2–138.3) | 17.3 (13.6) | |
| Astigmatism | 5 | 10.8 (1.3–92.5) | 10.8 (7.4) | |
| Gastrointestinal disorders | Dyspepsia | 173 | 2.3 (1.9–2.9) | 2.3 (62.3) |
| Gastroesophageal reflux disease | 139 | 3.6 (2.8–4.8) | 3.6 (98.8) | |
| Gastrointestinal disorder | 28 | 2.8 (1.6–4.8) | 2.8 (13.7) | |
| Lip swelling | 18 | 6.5 (2.6–16.3) | 6.5 (20.9) | |
| General disorders and administration site conditions | Asthenia | 726 | 9.4 (8.0–11.1) | 9.2 (1019.8) |
| Therapeutic response decreased | 513 | 43.5 (29.3–64.6) | 42.6 (1013.2) | |
| Decreased activity | 56 | 10.1 (5.4–18.9) | 10.1 (81.0) | |
| Therapeutic response increased | 48 | 20.8 (8.3–52.2) | 20.8 (85.2) | |
| Influenza like illness | 31 | 2.0 (1.2–3.3) | 2.0 (8.4) | |
| Exercise tolerance decreased | 28 | 20.2 (6.1–66.4) | 20.2 (49.4) | |
| Adverse drug reaction | 25 | 3.4 (1.8–6.3) | 3.4 (16.3) | |
| Therapeutic response delayed | 23 | 2.1 (1.2–3.7) | 2.1 (6.5) | |
| Inflammation | 16 | 2.7 (1.3–5.5) | 2.7 (7.4) | |
| Hepatobiliary disorders | Drug-induced liver injury | 19 | 2.7 (1.4–5.4) | 2.7 (9.3) |
| Infections and infestations | Conjunctivitis | 9 | 3.2 (1.2–9.1) | 3.2 (5.6) |
| Ear infection | 7 | 3.8 (1.1–12.9) | 3.8 (5.2) | |
| Injury, poisoning and procedural complications | Intentional product misuse | 277 | 2.4 (2.0–2.8) | 2.4 (105.0) |
| Medication error | 60 | 2.6 (1.8–3.8) | 2.6 (26.7) | |
| Prescribed overdose | 24 | 2.5 (1.4–4.4) | 2.5 (9.8) | |
| Investigations | Blood cholesterol increased | 744 | 10.0 (8.5–11.9) | 9.7 (1072.1) |
| Blood testosterone decreased | 585 | 11.2 (9.2–13.7) | 11.0 (880.9) | |
| Body height decreased | 188 | 14.1 (9.5–20.9) | 14.0 (304.4) | |
| Quality of life decreased | 77 | 15.2 (8.1–28.5) | 15.1 (127.1) | |
| Blood glucose abnormal | 27 | 5.8 (2.8–12.1) | 5.8 (29.3) | |
| Blood glucose fluctuation | 18 | 5.6 (2.3–13.3) | 5.6 (18.8) | |
| Blood alkaline phosphatase increased | 13 | 4.7 (1.8–12.3) | 4.7 (11.9) | |
| Urine output increased | 7 | 5.0 (1.3–19.5) | 5.0 (6.8) | |
| Metabolism and nutrition disorders | Diabetes mellitus | 413 | 5.7 (4.7–6.8) | 5.6 (438.5) |
| Dyslipidemia | 146 | 63.5 (26.0–154.8) | 63.1 (296.1) | |
| Musculoskeletal and connective tissue disorders | Back pain | 565 | 6.3 (5.4–7.4) | 6.2 (643.0) |
| Myalgia | 238 | 5.5 (4.3–7.0) | 5.5 (247.3) | |
| Pain in extremity | 194 | 2.4 (1.9–2.9) | 2.3 (72.0) | |
| Arthritis | 68 | 3.0 (2.1–4.3) | 3.0 (38.0) | |
| Muscle tightness | 17 | 4.6 (2.0–10.7) | 4.6 (15.3) | |
| Neoplasms benign, malignant and unspecified (incl cysts and polyps) | Basal cell carcinoma | 77 | 7.0 (4.4–11.0) | 6.9 (93.0) |
| Squamous cell carcinoma | 65 | 9.4 (5.4–16.5) | 9.4 (91.2) | |
| Malignant melanoma stage I | 46 | 4.7 (2.8–8.0) | 4.7 (42.5) | |
| Malignant melanoma stage II | 40 | 3.6 (2.2–6.0) | 3.6 (28.2) | |
| Nervous system disorders | Somnolence | 161 | 3.5 (2.7–4.4) | 3.4 (108.1) |
| Psychiatric disorders | Depression | 423 | 3.0 (2.6–3.5) | 3.0 (238.1) |
| Depressed mood | 307 | 20.3 (14.2–29.1) | 20.1 (543.3) | |
| Libido disorder | 294 | 160.7 (59.9–431.2) | 158.8 (621.3) | |
| Libido decreased | 288 | 13.4 (9.8–18.2) | 13.2 (459.3) | |
| Sleep disorder | 220 | 10.9 (7.9–15.1) | 10.8 (327.4) | |
| Initial insomnia | 19 | 20.6 (4.8–88.2) | 20.5 (33.6) | |
| Major depression | 16 | 3.8 (1.7–8.7) | 3.8 (12.1) | |
| Renal and urinary disorders | Pollakiuria | 97 | 3.8 (2.7–5.2) | 3.7 (71.5) |
| Urine flow decreased | 71 | 11.8 (6.6–21.4) | 11.8 (108.8) | |
| Urinary hesitation | 20 | 10.8 (3.7–31.6) | 10.8 (29.7) | |
| Nocturia | 19 | 2.4 (1.3–4.7) | 2.4 (7.4) | |
| Micturition urgency | 11 | 2.6 (1.1–6.4) | 2.6 (5.1) | |
| Reproductive system and breast disorders | BPH | 509 | 22.9 (17.1–30.7) | 22.5 (922.5) |
| Spontaneous penile erection | 65 | 2.0 (1.4–2.8) | 2.0 (17.1) | |
| Ejaculation delayed | 32 | 4.1 (2.3–7.3) | 4.1 (25.7) | |
| Prostatitis | 12 | 2.9 (1.2–6.8) | 2.9 (6.3) | |
| Skin and subcutaneous tissue disorders | Actinic keratosis | 61 | 10.2 (5.6–18.5) | 10.1 (88.4) |
| Stevens–Johnson syndrome | 7 | 7.6 (1.6–36.4) | 7.6 (8.9) | |
| Hyperkeratosis | 7 | 7.6 (1.6–36.4) | 7.6 (8.9) | |
| Vascular disorders | Hypertension | 850 | 6.4 (5.6–7.3) | 6.2 (970.8) |
| Pelvic venous thrombosis | 6 | 13.0 (1.6–107.7) | 13.0 (9.5) |
ROR: reporting odd ratio; CI: confidence interval; PRR: proportional reporting ratio; SOC: System Organ Class; PT: primary terms; BPH: benign prostatic hyperplasia; n: number
For vardenafil, expected AEs, including headache, flushing, rhinitis, dyspepsia, dizziness, and nausea, were detected in this study. Significant expected AEs, including nasal congestion, chest pain, increased erection, and elevated heart rate, were identified and warranted clinical attention. Moreover, expected AEs, including sinusitis, flu syndrome, and accidental injury, were not detected (Supplementary Table 5).
Supplementary Table 5.
Significant signals of vardenafil
| SOC | PT | n | ROR (95% CI) | PRR (χ2) |
|---|---|---|---|---|
| Cardiac disorders | Palpitations | 23 | 2.1 (1.4–3.3) | 2.1 (11.5) |
| Ventricular extrasystoles | 4 | 3.4 (1.2–10.3) | 3.4 (5.5) | |
| Congenital, familial, and genetic disorders | Color blindness | 7 | 10.7 (4.0–28.8) | 10.7 (34.7) |
| Ear and labyrinth disorders | Deafness unilateral | 21 | 2.5 (1.6–4.0) | 2.5 (15.9) |
| Vertigo | 13 | 3.1 (1.7–5.8) | 3.1 (15.4) | |
| Eye disorders | Eye pain | 20 | 2.7 (1.7–4.4) | 2.7 (17.9) |
| Photopsia | 17 | 5.1 (2.9–8.9) | 5.1 (40.8) | |
| Blindness transient | 17 | 3.1 (1.8–5.2) | 3.0 (19.1) | |
| Eye disorder | 13 | 2.2 (1.2–3.9) | 2.2 (7.0) | |
| Lacrimation increased | 11 | 3.6 (1.9–7.0) | 3.6 (16.4) | |
| Chromatopsia | 10 | 2.7 (1.4–5.2) | 2.7 (8.6) | |
| Retinal detachment | 10 | 2.2 (1.1–4.3) | 2.2 (5.8) | |
| Abnormal sensation in eye | 8 | 3.2 (1.5–6.8) | 3.2 (9.5) | |
| Visual brightness | 7 | 4.6 (2.0–10.8) | 4.6 (14.7) | |
| Vitreous detachment | 6 | 2.5 (1.1–6.0) | 2.5 (4.6) | |
| Retinopathy | 5 | 5.7 (2.0–16.3) | 5.7 (13.8) | |
| Amaurosis fugax | 4 | 5.5 (1.7–17.6) | 5.5 (10.5) | |
| Gastrointestinal disorders | Nausea | 62 | 2.5 (1.9–3.2) | 2.4 (45.2) |
| Dyspepsia | 47 | 2.3 (1.7–3.1) | 2.3 (28.3) | |
| Diarrhea | 32 | 2.1 (1.5–3.1) | 2.1 (16.2) | |
| Vomiting | 30 | 2.1 (1.4–3.0) | 2.1 (14.2) | |
| Abdominal pain upper | 19 | 2.1 (1.3–3.4) | 2.1 (9.2) | |
| Flatulence | 9 | 2.6 (1.3–5.4) | 2.6 (7.7) | |
| Rectal hemorrhage | 5 | 4.1 (1.5–11.0) | 4.1 (8.9) | |
| Eructation | 5 | 3.3 (1.2–8.7) | 3.3 (6.4) | |
| Toothache | 3 | 5.9 (1.5–22.8) | 5.9 (8.5) | |
| Frequent bowel movements | 3 | 3.8 (1.1–13.5) | 3.8 (4.8) | |
| General disorders and administration site conditions | Chest pain | 41 | 2.3 (1.7–3.2) | 2.3 (26.0) |
| Feeling hot | 34 | 3.0 (2.1–4.4) | 3.0 (37.7) | |
| Chest discomfort | 17 | 2.9 (1.7–4.8) | 2.9 (17.0) | |
| Therapeutic response delayed | 17 | 7.8 (4.3–14.2) | 7.8 (64.4) | |
| Chills | 15 | 3.0 (1.7–5.2) | 3.0 (16.4) | |
| Unevaluable event | 14 | 3.5 (2.0–6.3) | 3.5 (20.0) | |
| Hangover | 9 | 6.2 (2.8–13.6) | 6.2 (27.0) | |
| Influenza like illness | 9 | 2.3 (1.1–4.6) | 2.3 (5.4) | |
| AE | 8 | 2.8 (1.3–5.9) | 2.8 (7.4) | |
| Therapeutic product ineffective | 6 | 41.3 (8.3–204.9) | 41.3 (59.0) | |
| Feeling drunk | 3 | 5.9 (1.5–22.8) | 5.9 (8.5) | |
| Infections and infestations | Rhinitis | 7 | 13.8 (4.8–39.3) | 13.8 (41.4) |
| Investigations | Heart rate increased | 38 | 3.6 (2.5–5.1) | 3.5 (55.1) |
| Blood pressure decreased | 27 | 2.3 (1.6–3.5) | 2.3 (17.7) | |
| Blood urine | 4 | 18.4 (4.1–82.1) | 18.4 (28.1) | |
| Blood pressure systolic increased | 4 | 3.4 (1.2–10.3) | 3.4 (5.5) | |
| Pulse pressure decreased | 3 | 20.7 (3.5–123.7) | 20.6 (22.4) | |
| Nervous system disorders | Headache | 350 | 3.2 (2.8–3.6) | 3.0 (396.3) |
| Dizziness | 103 | 2.2 (1.8–2.8) | 2.2 (59.7) | |
| Migraine with aura | 5 | 17.2 (4.6–64.1) | 17.2 (33.9) | |
| Sinus headache | 5 | 3.4 (1.3–9.2) | 3.4 (6.9) | |
| Epilepsy | 4 | 4.6 (1.5–14.2) | 4.6 (8.4) | |
| Lacunar infarction | 3 | 3.8 (1.1–13.5) | 3.8 (4.8) | |
| Psychiatric disorders | Euphoric mood | 6 | 7.5 (2.8–20.3) | 7.5 (21.9) |
| Tension | 4 | 13.8 (3.4–55.1) | 13.8 (23.7) | |
| Panic reaction | 3 | 4.1 (1.1–15.0) | 4.1 (5.5) | |
| Reproductive system and breast disorders | ED | 1242 | 23.5 (21.5–25.8) | 18.3 (8957.9) |
| Spontaneous penile erection | 26 | 3.3 (2.2–5.1) | 3.3 (33.4) | |
| Ejaculation failure | 23 | 2.4 (1.5–3.7) | 2.3 (15.2) | |
| Hematospermia | 7 | 2.8 (1.2–6.2) | 2.8 (6.5) | |
| Testicular swelling | 6 | 2.6 (1.1–6.2) | 2.6 (4.9) | |
| Edema genital | 3 | 20.7 (3.5–123.7) | 20.6 (22.4) | |
| Genital pain | 3 | 3.8 (1.1–13.5) | 3.8 (4.8) | |
| Genital rash | 3 | 4.6 (1.2–17.0) | 4.6 (6.3) | |
| Respiratory, thoracic, and mediastinal disorders | Nasal congestion | 115 | 3.8 (3.1–4.7) | 3.8 (184.2) |
| Rhinorrhea | 18 | 2.7 (1.6–4.5) | 2.7 (16.4) | |
| Sinus congestion | 10 | 2.1 (1.1–4.0) | 2.1 (4.7) | |
| Epistaxis | 9 | 2.7 (1.3–5.5) | 2.7 (8.0) | |
| Dry throat | 4 | 13.8 (3.4–55.1) | 13.8 (23.7) | |
| Nasal obstruction | 3 | 13.8 (2.8–68.2) | 13.8 (17.8) | |
| Choking sensation | 3 | 41.3 (4.3–397.2) | 41.3 (29.5) | |
| Skin and subcutaneous tissue disorders | Erythema | 32 | 2.5 (1.7–3.6) | 2.5 (24.2) |
| Skin warm | 3 | 5.9 (1.5–22.8) | 5.9 (8.5) | |
| Hair growth abnormal | 3 | 4.6 (1.2–17.0) | 4.6 (6.3) | |
| Vascular disorders | Flushing | 186 | 3.8 (3.2–4.5) | 3.7 (290.1) |
ROR: reporting odd ratio; CI: confidence interval; PRR: proportional reporting ratio; SOC: System Organ Class; PT: primary terms; ED: erectile dysfunction; AE: adverse event; n: number
A total of seven PTs were identified as significant signals for avanafil. Notably, nausea is an expected AE. For avanafil, unexpected AEs such as ineffective, tinnitus, and deafness unilateral also warrant clinicians’ attention. Additionally, expected AEs, including flushing, nasopharyngitis, and nasal congestion, were not detected (Supplementary Table 6).
Supplementary Table 6.
Significant signals of avanafil
| SOC | PT | n | ROR (95% CI) | PRR (χ2) |
|---|---|---|---|---|
| Ear and labyrinth disorders | Tinnitus | 6 | 3.4 (1.5–7.6) | 3.3 (9.5) |
| Deafness unilateral | 5 | 5.0 (2.1–12.3) | 5.0 (15.4) | |
| Gastrointestinal disorders | Nausea | 7 | 2.3 (1.1–4.9) | 2.3 (5.0) |
| General disorders and administration site conditions | Drug ineffective | 247 | 5.4 (4.6–6.4) | 3.6 (504.5) |
| Injury, poisoning, and procedural complications | Wrong technique in product usage process | 20 | 3.0 (1.9–4.8) | 3.0 (25.7) |
| Expired product administered | 9 | 6.5 (3.3–12.8) | 6.5 (39.6) | |
| Nervous system disorders | Disturbance in attention | 3 | 10.5 (3.2–34.0) | 10.4 (23.6) |
ROR: reporting odd ratio; CI: confidence interval; PRR: proportional reporting ratio; SOC: System Organ Class; PT: primary terms; n: number
Signal differences
We illustrate the differences in subgroups of age, weight, and onset time to events (Figure 5). For sildenafil, patients with weight ≥90 kg were more likely to experience AEs such as increased semen volume, blindness unilateral, coronary artery occlusion and blindness. In contrast, individuals weighing <90 kg were prone to suffer penis disorder (Figure 5a). Regarding age, individuals aged over 65 years were more likely to report AEs such as malignant melanoma, prostate cancer and death. Those under 65 years were more likely to experience painful erection, penile swelling, and premature ejaculation (Figure 5b). AEs also vary with onset time. AEs such as malignant melanoma in situ typically emerged after 30 days of sildenafil use, while erection-related issues (increased erection, penile disorders, and painful erection) usually manifested within the first 30 days (Figure 5c).
Figure 5.
Volcano plot for difference detection of sildenafil and tadalafil signals. (a) Signal differences of sildenafil concerning weight (significant threshold: P < 0.01). (b) Signal differences of sildenafil concerning age (significant threshold: P < 0.001). (c) Signal differences of sildenafil concerning onset time to event (significant threshold: P < 0.05). (d) Signal differences of tadalafil concerning weight (significant threshold: P < 0.05). (e) Signal differences of tadalafil concerning age (significant threshold: P < 0.001). (f) Signal differences of tadalafil concerning onset time to event (significant threshold: P < 0.01). ROR: reporting odds ratio; SOC: system organ class.
For tadalafil, patients with weight ≥90 kg were more likely to experience blood alkaline phosphatase increased (Figure 5d). Patients under 65 years were more likely to experience urine flow decreased, whereas those aged over 65 years were prone to arthritis, back pain, dyspepsia, and gastroesophageal reflux disease (Figure 5e). As for the onset time, myalgia and pain in extremity are more likely to occur within the first 30 days of taking tadalafil. In contrast, squamous cell carcinoma, depression, and basal cell carcinoma tend to emerge after 30 days of tadalafil use (Figure 5f).
Vardenafil disclosed significant AEs differences in weight and onset time subgroups. Patients weighing <90 kg were more likely to experience eye pain, retinal detachment, and blindness transient compared to those with weight ≥90 kg (Figure 6a). No age-based correlations were observed (Figure 6b). Regarding onset time to event, diarrhea typically emerged post-30 days use of vardenafil, whereas headache and nasal congestion predominantly manifested within the first 30 days (Figure 6c).
Figure 6.
Volcano plot for difference detection of vardenafil and avanafil signals. (a) Signal differences of vardenafil concerning weight (significant threshold: P < 0.05). (b) Signal differences of vardenafil concerning age (significant threshold: P < 0.05). (c) Signal differences of vardenafil concerning onset time to event (significant threshold: P < 0.05). (d) Signal differences of avanafil concerning weight. (e) Signal differences of avanafil concerning age (significant threshold: P < 0.05). (f) Signal differences of avanafil concerning onset time to event (significant threshold: P < 0.05). ROR: reporting odds ratio; SOC: system organ class.
As for avanafil, no differences were observed in subgroups of weight (Figure 6d). Patients aged over 65 years were more likely to experience drug ineffective compared to those aged under 65 years (Figure 6e). Concerning onset time, no differences were observed (Figure 6f). A summary of the overall findings of this study is presented in Supplementary Table 7.
Supplementary Table 7.
Findings in the study
| Analysis | Findings | |||
|---|---|---|---|---|
|
| ||||
| Sildenafil | Tadalafil | Vardenafil | Avanafil | |
| Signal detection (AEs found in the study not listed in label of FDA) | Erection increased Malignant melanoma Optic ischemic neuropathy Blindness |
Increased blood cholesterol Depression Libido disorder Decreased body height |
Nasal congestion Chest pain Erectile Heart rate increased |
Ineffective Tinnitus Deafness unilateral |
| Signal difference (AEs happening more easily) | ||||
| Age (years) | ||||
| ≥65 | Malignant melanoma Glaucoma Retinal vein thrombosis | Arthritis Back pain Dyspepsia Gastroesophageal reflux disease | - | - |
| <65 | Painful erection Penile swelling Penis disorder | Urine flow decreased | - | Drug ineffective |
| Weight (kg) | ||||
| ≥90 | Semen volume increased Blindness unilateral Cataract Blindness | Increased blood alkaline phosphatase levels | - | - |
| <90 | Painful erection Testicular swelling Prostatic disorder | - | Eye pain Retinal detachment, Blindness | - |
| Onset time to events (days) | ||||
| ≥30 | Malignant melanoma in situ | Decreased therapeutic response, depression, basal cell carcinoma | Diarrhea | - |
| <30 | Penis disorder Painful erection | Myalgia Pain in extremity Ocular hypertension | - | - |
| Correlation analysis | ||||
| Age | Patients receiving avanafil are the oldest, followed by those receiving vardenafil, and patients receiving sildenafil and tadalafil are the youngest | |||
| Weight | The weight of patients using tadalafil is the lowest | |||
| Onset time to events | The onset time of sildenafil is the longest, and the onset time of avanafil is the shortest | |||
-: not mentioned; AEs: adverse events; FDA: Food and Drug Administration
DISCUSSION
Using real-world data from the FERAS database, we assessed the post-marketing safety of PDE5Is (sildenafil, tadalafil, vardenafil, and avanafil) and identified both expected and unexpected AEs for PDE5Is. In different PDE5Is, AEs vary with age, weight, and onset time to events. These findings are beneficial to clinicians optimizing the regimen of ED.
This study included data from 2004 to 2024, but data of avanafil were unavailable until 2013. This delay occurred because avanafil was first approved for sale in 2012.13 Reports peaked in 2015 but declined steadily after 2019. Among the four PDE5Is, sildenafil and tadalafil exhibited the highest AE reporting rates, correlating with their predominant sales volume. Sales volumes may be influenced by their annual costs: USD 459 for sildenafil, USD 723 for tadalafil, USD 2102 for vardenafil, and USD 3455 for avanafil.14 Higher drug costs generally correlate with lower sales volume.
The USA reported most PDE5I-associated AEs, followed by France, Germany, and Brazil. This trend may be due to easy access to the FAERS system, economic development, and national healthcare policies.15 Individuals in developed countries with comprehensive healthcare or insurance are more likely to be diagnosed with ED and prescribed PDE5I, resulting in more AE reports. Brazil is the only developing country among the top five countries, which may be attributed to cultural factors, such as the population’ s permissive attitudes toward sexuality. Hospitalization was the most common serious outcome. Other serious outcomes, including death, disability and life-threatening outcomes, occurred in less than 5% of cases. However, for sildenafil, serious outcomes exceeded 5%, warranting caution in its prescription.
The predominant age group of included patients (≥45 years and <65 years) contrasts with typical ED demographics characterized by older age (≥65 years) and poorer health status. This discrepancy may be attributed to lower sex demand in older ED patients compared to middle-aged patients. Consequently, reduced PDE5I usage in patients ≥65 years compared to patients aged ≥45 years and <65 years explains their elevated AE reporting rates in this analysis. However, unlike the overall PDE5Is trend, vardenafil and avanafil were most frequently reported in patients aged ≥65 years. This may be due to the dominance of sildenafil and tadalafil in sales, particularly among middle-aged ED patients, reducing vardenafil and avanafil use in this group. Thus, patients aged ≥65 years represent the majority of vardenafil and avanafil users, highlighting the potential risk of PDE5Is use in the elderly.
In different PDE5Is, AEs vary with age, weight, and onset time to events, indicating that high-risk groups differ for each drug. Therefore, clinicians should consider age and weight when prescribing PDE5Is to enhance patient safety. Avanafil-treated geriatric patients and high body weight vardenafil users show increased vulnerability. Sildenafil has the longest onset time to event, while avanafil has the shortest. Thus, it is recommended to prescribe sildenafil for the elderly for its safety. In addition, the elderly patients using avanafil should be more vigilant about AEs.
The most reported AE for sildenafil is increased erectile function, mainly due to its therapeutic effect on ED. Malignant melanoma was also detected, consistent with a previous study.16 The underlying mechanism is still unclear, necessitating further research. Vardenafil was associated with common AEs of PDE5Is, including headache, flushing, and dizziness. Avanafil had the fewest detected signals among the four PDE5Is. This may be due to its more recent FDA approval, resulting in limited reports. Recent studies suggest that PDE influences urinary system physiology,17 regulates cardiac cyclic nucleotides,18 and affects neurologic function.19 The reported AEs affected multiple systems where PDE5 is expressed, including the nervous, reproductive, musculoskeletal, and vascular systems. AEs such as optic ischemic neuropathy and blindness not listed on the label but observed in this study, were previously reported.20 The role of PDE5 expression and function underlies the AEs associated with PDE5Is. PDE5 expression across multiple systems contributes to AEs in various organs, complicating safety evaluation and underscoring the need for real-world studies. Understanding this mechanism is essential for clinicians and researchers to identify, manage, and mitigate PDE5I-related side effects.
This study identified significant differences in weight between sildenafil and tadalafil, between sildenafil and vardenafil, between tadalafil and vardenafil, and between vardenafil and avanafil. AEs patterns related to weight varied across different PDE5Is. Interestingly, some AEs were more common in patients weighing ≥90 kg, contradicting prior studies suggesting that lower-weight patients are at higher risk.21 Two possible explanations may account for this finding. First, substantial missing weight data in the FAERS database may affect result accuracy, as an unavoidable limitation of this study. Second, PDE5Is are metabolized in liver. Higher-weight patients may have reduced liver function, leading to prolonged drug retention and an increased risk of adverse event. However, as the FAERS lacks liver function data, this remains a hypothesis. Further pharmacovigilance studies are needed to validate this hypothesis.
This study has several limitations. First, FDA warnings and other factors may influence FAERS reports, leading to underreporting or overreporting of AEs. Second, this study provides only a statistical evaluation of signals, without determining actual AE risk. Missing demographic data is a common challenge in FAERS-based pharmacovigilance studies, potentially affecting the robustness and interpretability of statistical findings. The lack of key demographic variables, such as age or weight, may introduce selection bias and affect the calculation of disproportionality measures such as the ROR and PRR.
A key concern is non-random missingness, where certain patient groups are more likely to have incomplete reports. For instance, older adults may have lower reporting rates for specific demographic attributes, leading to their underrepresentation in signal detection. Underreporting of high-risk groups may underestimate the association between a drug and an adverse event potentially masking true safety signal. Conversely, excessive missing data in lower-risk populations may inflate the calculated signal. Findings showed that while the overall ROR and PRR trends remained stable, subgroup analyses revealed variations, especially when missing data exceeded 30%. This highlights that missing demographic data may affect signal detection, emphasizing the need for cautious results interpretation.
Additionally, FAERS is inherently subject to reporting bias, which can influence the interpretation of pharmacovigilance findings. Two key aspects of reporting bias, underreporting and duplicate reporting, can significantly impact the detection and evaluation of drug safety signals. Underreporting is a well-recognized limitation in FAERS, as not all AEs are documented due to factors such as healthcare provider awareness, patient self-reporting tendencies, and regulatory policies. Serious and unexpected AEs are more likely to be reported than mild or well-known ones, potentially leading to overestimation of certain signals while underestimating others. This selective reporting can create an imbalance in the disproportionality analysis, where rarer but severe AEs appear disproportionately frequent compared to more common but underreported events.
Given these findings, clinicians should adopt enhanced monitoring strategies for high-risk populations. These include regular clinical follow-up to identify early signs of AEs, patient education on recognizing and reporting symptoms, and dose adjustments or alternative treatment for patients with significant risk factors. Additionally, although FAERS-based signal detection is useful for identifying potential safety concerns, it does not establish causality. Consequently, prescribers should integrate these findings with evidence from clinical trials, real-world studies, and professional guidelines to make well-informed patient care decisions. Future studies with controlled designs and comprehensive patient-level data are crucial to validate these observations and refine risk stratification approaches.
In summary, this study expands knowledge on the safety of FDA-approved PDE5Is for ED treatment in real-world settings. Greater attention is needed for AEs with strong real-world signals, such as optic ischemic neuropathy and blindness, which are not listed on the drug label. Additionally, this study identified signal differences based on age, weight, and onset time to event. Future clinical and pharmacovigilance studies are needed to guide clinicians in identifying target patients and optimizing treatment regimens.
AUTHOR CONTRIBUTIONS
ZHZ and YX contribute to conceptualization. ZHZ contributes to the writing of original draft. WW contributes to formal analysis. YX contributes to writing, review and editing. CJW contributes to visualization. JHY contributes to draft reviewing and project administration. All authors read and approved the final manuscript.
COMPETING INTERESTS
All authors declare no competing interests.
ACKNOWLEDGMENTS
We would sincerely like to thank the scientists and researchers who discovered and developed PDE5Is, enabling generations of men to have hope to deal with erectile dysfunction, and FDA for establishing FAERS database which contributed to our understanding of drug safety and adverse events. This work was supported by the Sichuan Science and Technology Program number (No. 2025YFHZ0212).
Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.
REFERENCES
- 1.McCabe MP, Sharlip ID, Atalla E, Balon R, Fisher AD, et al. Definitions of sexual dysfunctions in women and men:a consensus statement from the fourth International Consultation on Sexual Medicine 2015. J Sex Med. 2016;13:135–43. doi: 10.1016/j.jsxm.2015.12.019. [DOI] [PubMed] [Google Scholar]
- 2.Yamazaki H, Tsujimoto N, Koyanagi M, Katoh MC, Tajima K, et al. Real-world safety and effectiveness of tadalafil in patients with lower urinary tract symptoms secondary to benign prostatic hyperplasia:a Japanese post-marketing surveillance study. Pragmat Obs Res. 2020;11:45–54. doi: 10.2147/POR.S237821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nguyen HM, Gabrielson AT, Hellstrom WJ. Erectile dysfunction in young men-a review of the prevalence and risk factors. Sex Med Rev. 2017;5:508–20. doi: 10.1016/j.sxmr.2017.05.004. [DOI] [PubMed] [Google Scholar]
- 4.Burnett AL, Nehra A, Breau RH, Culkin DJ, Faraday MM, et al. Erectile dysfunction:AUA guideline. J Urol. 2018;200:633–41. doi: 10.1016/j.juro.2018.05.004. [DOI] [PubMed] [Google Scholar]
- 5.Barroso F, Crispim Ribeiro J, Miranda EP. Phosphodiesterase type 5 inhibitors and visual side effects:a narrative review. J Ophthalmic Vis Res. 2021;16:248–59. doi: 10.18502/jovr.v16i2.9088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zahir M, Samzadeh M, Poopak A, Khoshdel AR, Armin A. Sildenafil versus tadalafil for the treatment of benign prostatic hyperplasia:a single-arm self-controlled clinical trial. Urol J. 2023;20:7593. doi: 10.22037/uj.v20i.7593. [DOI] [PubMed] [Google Scholar]
- 7.Mohamed Gharib T, Abdel-Al I, Elatreisy A, Kandeel W, El-Shaer W, et al. Short- and long-term follow-up results of daily 5-mg tadalafil as a treatment for erectile dysfunction and premature ejaculation. Arab J Urol. 2022;20:49–53. doi: 10.1080/2090598X.2021.2024695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Liu TT, Pascal LE, Bauer SR, Miles HN, Panksepp JB, et al. Age-dependent effects of voluntary wheel running exercise on voiding behavior and potential age-related molecular mechanisms in mice. J Gerontol A Biol Sci Med Sci. 2024;79:glae007. doi: 10.1093/gerona/glae007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Li H, Wang C, Deng A, Guo C. A real-world disproportionality analysis of mepolizumab based on the FDA adverse event reporting system. Front Pharmacol. 2023;14:1280490. doi: 10.3389/fphar.2023.1280490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hentzen C, Musco S, Amarenco G, Del Popolo G, Panicker JN. Approach and management to patients with neurological disorders reporting sexual dysfunction. Lancet Neurol. 2022;21:551–62. doi: 10.1016/S1474-4422(22)00036-9. [DOI] [PubMed] [Google Scholar]
- 11.Kinoshita S, Hosomi K, Yokoyama S, Takada M. Time-to-onset analysis of amiodarone-associated thyroid dysfunction. J Clin Pharm Ther. 2019;45:65–71. doi: 10.1111/jcpt.13024. [DOI] [PubMed] [Google Scholar]
- 12.Lin X, Yang J, Weng L, Lin W. Differences in hypersensitivity reactions to iodinated contrast media:analysis of the US Food and Drug Administration Adverse Event Reporting System database. J Allergy Clin Immunol Pract. 2023;11:1494–502.e6. doi: 10.1016/j.jaip.2023.01.027. [DOI] [PubMed] [Google Scholar]
- 13.Al-Majed AA, AlKhairallah A, Attwa MW, Alkahtani HM, El-Azab AS, et al. Avanafil:a comprehensive drug profile. Profiles Drug Subst Excip Relat Methodol. 2024;49:115–51. doi: 10.1016/bs.podrm.2023.11.006. [DOI] [PubMed] [Google Scholar]
- 14.Nguyen V, McGovern AM, Rojanasarot S, Patel DP, Bhattacharyya S, et al. Patient out-of-pocket costs for guideline-recommended treatments for erectile dysfunction:a Medicare cost modeling analysis. Int J Impot Res. 2024;37:45–50. doi: 10.1038/s41443-024-00903-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Burnett AL, Rojanasarot S, Amorosi SL. An analysis of a commercial database on the use of erectile dysfunction treatments for men with employer-sponsored health insurance. Urology. 2021;149:140–5. doi: 10.1016/j.urology.2020.11.051. [DOI] [PubMed] [Google Scholar]
- 16.Pottegård A, Schmidt SA, Olesen AB, Achacoso N, Van Den Eeden SK, et al. Use of sildenafil or other phosphodiesterase inhibitors and risk of melanoma. Br J Cancer. 2016;115:895–900. doi: 10.1038/bjc.2016.248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhu X, Zhai K, Mi Y, Ji G. Expression and function of phosphodiesterases (PDEs) in the rat urinary bladder. BMC Urol. 2017;17:54. doi: 10.1186/s12894-017-0244-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kokkonen K, Kass DA. Nanodomain regulation of cardiac cyclic nucleotide signaling by phosphodiesterases. Annu Rev Pharmacol Toxicol. 2017;57:455–79. doi: 10.1146/annurev-pharmtox-010716-104756. [DOI] [PubMed] [Google Scholar]
- 19.Fusco F, Giampa C. Phosphodiesterases as therapeutic targets for Huntington's disease. Curr Pharm Des. 2014;21:365–77. doi: 10.2174/1381612820666140826113957. [DOI] [PubMed] [Google Scholar]
- 20.Capece M, Montorio D, Comune C, Aveta A, Melchionna A, et al. Retinal and optic disc vascular changes in patients using long-term tadalafil:a prospective non-randomized matched-pair study. Diagnostics (Basel) 2021;11:802. doi: 10.3390/diagnostics11050802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Qin HX, Yu P, Yan MA, Hong LY, Ru GU, et al. Prevalence and risk factors associated with adverse drug reactions among previously treated tuberculosis patients in China. Biomed Environ Sci. 2017;30:139–42. doi: 10.3967/bes2017.018. [DOI] [PubMed] [Google Scholar]