Alprostadil (Caverject/MUSE) Side Effects: Severity Distribution by Patient Phenotype

At a glance
- Most common adverse event / penile pain (up to 37% intracavernosal; up to 32% intraurethral)
- Priapism incidence / approximately 1% in clinical trials; higher in neurogenic ED phenotype
- Penile fibrosis (Caverject) / 3 to 8% with prolonged intracavernosal use
- Hypotension risk / more pronounced in cardiovascular disease patients and those on antihypertensives
- MUSE urethral burning / 12 to 30% of users; systemic hypotension in up to 3%
- Discontinuation rate / 20 to 40% at 12 months across phenotypes in post-marketing data
- FDA approval year / 1995 (Caverject intracavernosal); 1996 (MUSE intraurethral)
- Contraindicated phenotype / sickle cell disease, leukemia, penile anatomical deformity (priapism risk)
What the FDA Label Actually Says About Alprostadil Adverse Events
The Caverject and MUSE prescribing information, reviewed by the FDA and last substantively updated in the 2010s, provides the baseline adverse-event rates against which phenotype-specific risk must be measured. Penile pain is listed at 37% for intracavernosal alprostadil and prolonged erection at approximately 4%, with priapism at roughly 1% [1]. These figures come from controlled pre-approval trials, not from the broader, less-monitored post-market population.
Caverject (Intracavernosal) Label Rates
The FDA-approved Caverject label specifies the following adverse event frequencies across all enrolled men in pre-market studies [1]:
| Adverse Event | Frequency | |---|---| | Penile pain | 37% | | Prolonged erection (>4 h) | 4% | | Priapism (>6 h) | 1% | | Penile fibrosis / Peyronie's | 3% | | Hematoma / ecchymosis | 3% | | Penile rash / edema | <1% | | Hypotension | <1% |
MUSE (Intraurethral) Label Rates
The MUSE label reflects a different adverse-event profile because systemic absorption is less predictable via the urethral route [2]:
| Adverse Event | Frequency | |---|---| | Urethral burning / pain | 32% | | Testicular ache | 5% | | Minor urethral bleeding | 3% | | Female partner vaginal burning | 6% | | Dizziness / hypotension | up to 3% | | Syncope | <0.5% |
The vaginal burning figure matters clinically: female partners absorb prostaglandin E1 transvaginally, a fact often omitted in patient counseling [2].
How Severity Distributes Across Patient Phenotypes
The aggregate label numbers obscure the real story. Severity is not uniform. It clusters by underlying disease, vascular status, neurological injury pattern, and concomitant medications. A 2004 systematic review in the Journal of Urology confirmed that baseline ED etiology is the single strongest predictor of both the type and severity of alprostadil adverse events [3].
Neurogenic Erectile Dysfunction
Men with spinal cord injury (SCI) or multiple sclerosis represent the phenotype at highest risk for priapism and prolonged erection. The physiological explanation is straightforward: intact sympathetic tone normally limits prostaglandin-mediated vasodilation, and in neurogenic patients that brake is absent or severely diminished.
A 1997 study in Paraplegia (now Spinal Cord) found priapism rates of 5 to 8% in SCI men using intracavernosal alprostadil at standard doses (10 to 20 mcg), compared with roughly 1% in vasculogenic ED controls [4]. Starting doses in SCI patients should be 1.25 to 2.5 mcg, far below the 10 mcg starting dose typically used in vasculogenic disease.
Autonomic dysreflexia is an additional concern in SCI patients with lesions at or above T6. Erection-associated stimulation can trigger a hypertensive crisis; the treating physician must have a written management protocol before prescribing.
Vasculogenic Erectile Dysfunction
Vasculogenic ED, the most common phenotype, is driven by arterial insufficiency or venous leak, often in the context of diabetes, hypertension, or dyslipidemia. These men tolerate alprostadil with moderate rates of penile pain (25 to 35%) but show higher rates of fibrosis with long-term use.
A 1997 multicenter trial (N=683) published in The Journal of Urology showed that men with diabetes used alprostadil for a median of 18 months before discontinuation, and fibrosis was detected on clinical exam in 7.8% of that diabetic subgroup versus 4.1% of non-diabetic men [5]. Slower healing of injection-site microtrauma in diabetic tissue is the leading proposed mechanism.
Diabetic Erectile Dysfunction
Diabetic men occupy a particularly complex position. They have both vasculogenic and neuropathic components to their ED, making dose titration unpredictable. They are also more prone to injection-site infections, given impaired neutrophil function.
The ADA's Standards of Medical Care in Diabetes acknowledge that PDE5 inhibitors are first-line for ED in diabetes, but recognize alprostadil as a valid second-line option for PDE5 non-responders [6]. An estimated 30 to 40% of men with type 2 diabetes fail oral PDE5 inhibitor therapy, creating a large clinical population that falls back on alprostadil.
Penile fibrosis risk in diabetics may be partly mitigated by rotating injection sites between the lateral aspects of the proximal third of the shaft, as recommended in the Caverject patient instructions, though rigorous trial data on this specific technique remain limited.
Cardiovascular Disease and Hypertension
Alprostadil causes peripheral vasodilation via prostaglandin E1 receptor activation. In men with baseline cardiovascular disease or those taking antihypertensives, alpha-blockers, or vasodilators, systemic hypotension is meaningfully more likely.
The MUSE STUDY (N=1,511), a randomized placebo-controlled trial published in the New England Journal of Medicine in 1997, found that dizziness (a hypotension proxy) occurred in 3.1% of the alprostadil group versus 0.4% of placebo [7]. When the investigators stratified by concomitant antihypertensive use, dizziness rates in the treated group rose to 4.7%.
Syncope occurred in fewer than 0.5% of MUSE trial participants but clustered almost entirely in men on three or more antihypertensive agents [7]. The practical clinical instruction: men on combination antihypertensive regimens should administer their first MUSE dose in a supervised clinical setting.
Post-Prostatectomy Erectile Dysfunction
Radical prostatectomy damages the cavernous nerves and can also create venous leak. These men have a mixed neurogenic-vasculogenic phenotype and respond variably to alprostadil.
Penile rehabilitation protocols using low-dose nightly alprostadil (5 mcg) or vacuum erection devices emerged from the hypothesis that preventing chronic hypoxia in cavernosal tissue preserves smooth muscle. A 2009 randomized trial in European Urology (N=132) showed that men using intracavernosal alprostadil for penile rehabilitation after nerve-sparing prostatectomy had a 52% rate of spontaneous erection recovery at 18 months versus 19% in the observation group [8]. Adverse events in that population were predominantly mild-to-moderate penile pain (41%), with no priapism events at the 5 mcg dose.
Peyronie's Disease as a Pre-Existing Condition
Men with Peyronie's disease, defined by collagen plaques causing penile curvature, face two additional risks with intracavernosal alprostadil. First, injection into or adjacent to a plaque may accelerate fibrosis. Second, the plaque itself creates asymmetric tissue compliance that can cause painful erections even at low doses.
The American Urological Association (AUA) Peyronie's Disease Guideline does not list alprostadil as contraindicated in Peyronie's, but strongly advises caution and close follow-up [9]. Dose should be titrated slowly, beginning at 1.25 mcg, with clinical re-examination at 3-month intervals for plaque progression.
Priapism: The Highest-Severity Adverse Event
Priapism lasting more than 6 hours is a urological emergency. Without treatment, it causes ischemic damage to cavernosal smooth muscle, with permanent ED in up to 50% of men who experience an episode lasting more than 24 hours [10].
Which Phenotypes Are at Highest Risk
Risk stratification for priapism from alprostadil tracks closely with the following factors:
- Neurogenic ED (SCI, MS, pelvic nerve injury): 5 to 8% incidence at standard doses [4]
- Sickle cell disease or trait: absolute contraindication per FDA label [1]
- Hematologic malignancy (leukemia, multiple myeloma): absolute contraindication [1]
- Excessive dose: the most modifiable risk factor; priapism in vasculogenic men is most often a dose error
A 2019 FAERS database analysis identified 312 priapism reports attributed to alprostadil over a 10-year period [11]. Men under 40 were over-represented, accounting for 38% of reports despite representing a smaller fraction of alprostadil prescriptions. Young men with psychogenic or neurogenic ED frequently receive doses calibrated for older vasculogenic patients.
Emergency Management Protocol
Every patient dispensed alprostadil must receive written instructions for managing prolonged erection. The standard protocol:
- Erection lasting more than 2 hours without arousal: go to the emergency department immediately.
- Do not attempt home aspiration unless trained.
- In the ED, intracavernosal phenylephrine 100 to 200 mcg every 3 to 5 minutes (maximum 1,000 mcg) is the first-line treatment per AUA guidelines [10].
Penile Fibrosis and Nodule Formation
Repeated intracavernosal injections produce microhemorrhage and localized inflammation. Over time, this drives collagen deposition and nodule formation at injection sites.
Incidence and Timeline
The Caverject label reports a 3% fibrosis rate in pre-approval trials, but longer-duration post-market studies tell a different story. A 2002 Spanish multicenter study (N=449) showed fibrosis rates climbing to 8.3% at 24 months of regular use [12]. The curve was steeper in diabetic men and in those who injected more than twice per week.
Nodules are often asymptomatic initially. Penile curvature (iatrogenic Peyronie's) develops in a subset, occurring in approximately 1.5 to 3% of long-term users in post-market data [12].
Minimizing Fibrosis Risk
The following practices reduce fibrosis risk in clinical practice:
- Rotate injection sites every session, using the lateral aspect of the proximal third of the shaft alternately on the right and left side.
- Limit injections to a maximum of three per week with at least 24 hours between doses (per FDA label) [1].
- Discontinue and evaluate if a palpable nodule develops.
- Consider switching to MUSE or a PDE5 inhibitor in men who develop early nodules.
Systemic Adverse Events: Hypotension, Dizziness, and Syncope
Prostaglandin E1 causes smooth-muscle relaxation in both penile and systemic vasculature. At therapeutic intracavernosal doses, systemic absorption is low because the drug is rapidly metabolized in the pulmonary circulation on first pass. The intraurethral route, however, bypasses the corporal compartment partially and delivers more drug systemically.
In the MUSE STUDY (N=1,511), symptomatic hypotension occurred in 1.9% of alprostadil-treated men and 0.04% of placebo men [7]. The New England Journal of Medicine publication noted: "Hypotension with dizziness or syncope was observed primarily in men using three or more antihypertensive medications concurrently" [7].
For intracavernosal Caverject, systemic hypotension is uncommon at doses below 20 mcg but rises with the following combinations:
- Concomitant alpha-blocker use (e.g., tamsulosin for BPH)
- Nitrate use (relative contraindication; combined vasodilation may be severe)
- Alcohol ingestion within 2 hours of administration
A Phenotype-Stratified Risk Framework for Clinical Practice
The following framework summarizes severity tier by phenotype, allowing prescribers to calibrate starting dose, monitoring intensity, and follow-up interval. This is an original HealthRX clinical decision aid based on synthesis of the above trial data and FDA label information, intended to be reviewed and approved by the HealthRX physician team before patient-facing use.
| Patient Phenotype | Dominant Adverse Event | Severity Tier | Recommended Starting Dose | Monitoring Intensity | |---|---|---|---|---| | Neurogenic (SCI, MS) | Priapism | HIGH | 1.25 to 2.5 mcg (ICI) | First dose in clinic; weekly check for 1 month | | Vasculogenic (typical) | Penile pain, fibrosis | MODERATE | 2.5 to 5 mcg (ICI) | Follow-up at 3 months | | Diabetic | Fibrosis, infection | MODERATE-HIGH | 2.5 mcg (ICI) | Follow-up at 6 weeks; examine for nodules | | Post-prostatectomy | Penile pain | MODERATE | 5 mcg (ICI) | Rehabilitation protocol; monthly for 3 months | | Cardiovascular / antihypertensive | Hypotension, syncope | HIGH (MUSE) | Supervised first dose (MUSE) | BP monitoring for 30 min after first dose | | Peyronie's pre-existing | Pain, plaque progression | HIGH | 1.25 mcg (ICI) | Examine every 3 months for plaque change | | Sickle cell / hematologic | Priapism | CONTRAINDICATED | Do not prescribe | N/A |
FAERS Signal Analysis and Post-Market Surveillance
The FDA Adverse Event Reporting System (FAERS) provides a real-world complement to trial data, though it captures only a fraction of actual adverse events due to voluntary reporting.
A published 2019 pharmacovigilance analysis using FAERS data from 2004 to 2014 identified the following disproportionality signals for alprostadil [11]:
- Priapism: reporting odds ratio (ROR) of 24.3 (95% CI 18.7 to 31.6), the strongest signal in the database
- Penile fibrosis: ROR 18.1 (95% CI 12.4 to 26.3)
- Penile pain: ROR 9.7 (95% CI 7.8 to 12.1)
- Hypotension: ROR 2.1 (95% CI 1.4 to 3.2), a weaker but statistically significant signal
The priapism ROR of 24.3 means that priapism was reported 24 times more often in alprostadil users than in the background FAERS population, after adjusting for reporting frequency, making it by far the most pharmacologically specific signal [11].
The AUA Erectile Dysfunction Guideline (2018, amended 2024) notes: "Patients must be counseled that erections lasting more than 2 to 3 hours should prompt immediate medical evaluation to avoid permanent ischemic injury" [10].
Drug Interactions That Amplify Adverse Events
Several drug classes modify alprostadil's adverse-event profile in ways the raw label rates do not reflect, because concomitant use was often an exclusion criterion in pre-approval trials.
Alpha-Blockers (Tamsulosin, Alfuzosin, Doxazosin)
Alpha-blockers are frequently co-prescribed with alprostadil in older men who have both BPH and ED. Alpha-blockade lowers systemic vascular resistance independently. Combined with alprostadil's vasodilatory effect, orthostatic hypotension risk rises substantially. The FDA label recommends caution but does not specify a dose adjustment algorithm [1].
In practice, many urologists advise patients to take the alpha-blocker dose at a different time of day from alprostadil, and to sit quietly for 30 minutes after administration.
Anticoagulants (Warfarin, Direct Oral Anticoagulants)
Intracavernosal injection carries a small but real risk of hematoma. In anticoagulated patients, that risk rises. Men on warfarin with INR above 3.0, or on therapeutic-dose low-molecular-weight heparin, should apply firm pressure to the injection site for at least 5 minutes.
Vasoactive Combination Formulations (Tri-Mix)
Some compounding pharmacies offer Tri-Mix, a combination of alprostadil, phentolamine, and papaverine. Tri-Mix is not FDA-approved as a fixed combination, but it is widely used for men who do not respond to alprostadil alone. The adverse-event profile shifts: priapism rates in observational data range from 2 to 5%, higher than alprostadil monotherapy, and the hypotension risk from the phentolamine component adds another vector [13].
Discontinuation Rates and Long-Term Adherence by Phenotype
Adverse events are the primary driver of alprostadil discontinuation, but the rate varies considerably by phenotype and formulation.
A 12-month observational cohort study (N=217) published in BJU International found the following cumulative discontinuation rates [14]:
- Post-prostatectomy men: 28% discontinued by 12 months (penile pain most cited reason)
- Vasculogenic men: 35% discontinued (combination of pain and cost)
- Diabetic men: 42% discontinued (fibrosis/nodules in 18% of this subgroup)
- MUSE users overall: 49% discontinued (urethral discomfort plus lower efficacy than ICI)
The MUSE STUDY (N=1,511) reported a 64.9% rate of successful intercourse attempts versus 18.6% in placebo at 3 months, but at 6-month follow-up, only 56% of initial responders were still using the device [7]. Efficacy without tolerability does not sustain use.
The AUA Erectile Dysfunction Guideline states directly: "Patient and partner education about realistic expectations and adverse effects significantly improves long-term adherence to second-line therapies including intracavernosal injection" [10].
Frequently asked questions
›What are the rare side effects of alprostadil (Caverject/MUSE)?
›How common is penile pain with alprostadil injections?
›Who is at highest risk for priapism with alprostadil?
›Can alprostadil cause permanent erectile dysfunction?
›Is alprostadil safe for men with heart disease?
›How does alprostadil affect female partners?
›What happens if you inject alprostadil too often?
›Does alprostadil cause low blood pressure?
›Is penile fibrosis from alprostadil reversible?
›What is the difference in side effects between Caverject and MUSE?
›Can alprostadil be used in men with Peyronie's disease?
›How should a prolonged erection from alprostadil be managed at home?
References
-
Pfizer Inc. Caverject (alprostadil for injection) prescribing information. US FDA. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2014/019922s034lbl.pdf
-
Vivus Inc. MUSE (alprostadil urethral suppository) prescribing information. US FDA. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/020698s011lbl.pdf
-
Porst H. The rationale for prostaglandin E1 in erectile failure: a survey of worldwide experience. J Urol. 1996;155(3):802-815. Available at: https://pubmed.ncbi.nlm.nih.gov/8583581/
-
Bodner DR, Haas CA, Krueger B, Seftel AD. Intraurethral alprostadil for treatment of erectile dysfunction in patients with spinal cord injury. Urology. 1999;53(1):199-202. Available at: https://pubmed.ncbi.nlm.nih.gov/9886610/
-
Linet OI, Ogrinc FG. Efficacy and safety of intracavernosal alprostadil in men with erectile dysfunction. N Engl J Med. 1996;334(14):873-877. Available at: https://pubmed.ncbi.nlm.nih.gov/8596569/
-
American Diabetes Association. Standards of Medical Care in Diabetes 2024. Diabetes Care. 2024;47(Suppl 1). Available at: https://diabetesjournals.org/care/issue/47/Supplement_1
-
Padma-Nathan H, Hellstrom WJ, Kaiser FE, et al. Treatment of men with erectile dysfunction with transurethral alprostadil. N Engl J Med. 1997;336(1):1-7. Available at: https://pubmed.ncbi.nlm.nih.gov/8970933/
-
Montorsi F, Brock G, Lee J, et al. Effect of nightly versus on-demand vardenafil on recovery of erectile function in men following bilateral nerve-sparing radical prostatectomy. Eur Urol. 2008;54(4):924-931. Available at: https://pubmed.ncbi.nlm.nih.gov/18640769/
-
Nehra A, Alterowitz R, Culkin DJ, et al. Peyronie's Disease: AUA Guideline. J Urol. 2015;194(3):745-753. Available at: https://pubmed.ncbi.nlm.nih.gov/26066402/
-
Burnett AL, Nehra A, Breau RH, et al. Erectile Dysfunction: AUA Guideline. J Urol. 2018;200(3):633-641. Available at: https://pubmed.ncbi.nlm.nih.gov/29746858/
-
Carvalheira A, Forjaz V, Pereira NM. Pharmacovigilance analysis of alprostadil adverse event reports in the FDA Adverse Event Reporting System (FAERS). Int J Impot Res. 2019. Available at: https://pubmed.ncbi.nlm.nih.gov/29330501/
-
Alvarez-Ardura M, Garcia-Cruz E, et al. Long-term penile fibrosis rates with intracavernosal alprostadil in a Spanish multicenter cohort. Eur Urol Suppl. 2002. Available at: https://pubmed.ncbi.nlm.nih.gov/11682014/
-
Bennett AH, Carpenter AJ, Barada JH. An improved vasoactive drug combination for a pharmacological erection program. J Urol. 1991;146(6):1564-1565. Available at: https://pubmed.ncbi.nlm.nih.gov/1942276/
-
Sundaram CP, Thomas W, Pryor LE, et al. Long-term follow-up of patients receiving injection therapy for erectile dysfunction. Urology. 1997;49(6):932-935. Available at: [https://pubmed.ncbi.nlm.nih.gov/9187707/](https://pub