Sage Journals: Discover world-class research

Abstract

Background:

Blepharoptosis is one of the most common eyelid disorders in clinical ophthalmology. Previous evidence on drug-related blepharoptosis limited to case reports.

Objectives:

This study aims to systematically evaluate the disproportionality signals of drugs associated with blepharoptosis using large-scale real-world data from the US FDA Adverse Event Reporting System (FAERS).

Design:

A retrospective disproportionality analysis was conducted based on the FAERS database.

Methods:

A total of 21,838,627 reports from the FAERS database, spanning 2004 to 2024, were included, with 19,541,994 reports retained after deduplication. Disproportionality analysis methods including reporting odds ratio, proportional reporting ratio, Bayesian confidence propagation neural network, and multi-item gamma Poisson shrinker were employed to detect positive signals. The therapeutic purposes of the drugs, drug-related reporting patterns, drug signals strength, and onset times of adverse reactions were comprehensively assessed.

Results:

A total of 9324 cases of blepharoptosis were confirmed, involving 20 identified with significant signals. They primarily include antineoplastic and immunomodulating agents, sensory organ drugs, and neuropsychiatric drugs. Eleven drugs had previously been reported to be associated with blepharoptosis, while 9 were newly identified. Botulinum toxin A, ravulizumab, and latanoprost were found to exhibit the strongest signals. Neuropsychiatric drugs (e.g., lidocaine) had a median onset time of 1–9 days, while antineoplastic and immunomodulating agents (e.g., ravulizumab) had a median onset time of 164–912 days. Avelumab and rosuvastatin showed stronger signals for elderly males, while botulinum toxin A and bupivacaine were more significant for younger females.

Conclusion:

This study validates known associations such as immune checkpoint inhibitors and neuroregulatory agents, and identifies drug-related signals. There are significant differences in the onset times of adverse reactions across drug categories, suggesting the need for targeted monitoring.

Plain language summary

Drug-related blepharoptosis

Blepharoptosis refers to excessive drooping of the upper eyelid, which may obstruct vision or affect appearance. In the past, physicians have observed that a small number of individuals developed blepharoptosis after using certain medications, but most reports were limited to small-scale studies or case reports. This study systematically evaluates which drugs are more likely to induce blepharoptosis, the time frame in which this adverse event occurs, and whether certain demographic groups (such as different genders and age groups) are more susceptible. The analysis was conducted using data from the US Food and Drug Administration Adverse Event Reporting System (FAERS) from 2004 to 2024.

The results indicate that among more than 20 million adverse event reports, 20 drugs were significantly associated with blepharoptosis. These drugs fall into four main categories: (1) antineoplastic and immunomodulating agents, (2) sensory system medications, (3) nervous system drugs, and (4) other categories. Notably, 11 of these drugs had been previously mentioned in the literature, while the remaining 9 were newly identified in this study as potential contributors to blepharoptosis. The study also revealed that the onset time of drug-related blepharoptosis varied: some medications (such as certain anesthetics) could cause blepharoptosis shortly after administration, whereas others (such as specific antineoplastic and immunomodulating agents) might lead to gradual onset over several months or even close to a year. Furthermore, gender and age appeared to influence the signal strength of blepharoptosis, with some drugs showing more positive strong signals in young females and others requiring greater caution in elderly males.

These findings are valuable for both physicians and patients. For physicians, early monitoring and prevention of ptosis can be implemented when prescribing strong signals medications. For patients, understanding potential symptoms following drug use enables timely medical consultation if ptosis occurs. Overall, this study provides a “warning list” that aids clinical decision-making, urging both healthcare providers and patients to enhance risk management and achieve a balance between effective treatment and medication safety.

Keywords

blepharoptosis drug-related adverse events pharmacovigilance real-world study

Introduction

Blepharoptosis is one of the most common eyelid disorders in clinical ophthalmology, with reported prevalence rates in adults ranging from 4.7% to 13.5% according to studies.^1–3 It can affect appearance and impair visual function, both of which negatively impact quality of life.⁴ Patients with blepharoptosis may experience social withdrawal and fear of negative judgment from others, often reporting elevated levels of anxiety and depression.⁵ Typically, blepharoptosis is classified as either congenital or acquired,⁶ with acquired blepharoptosis being the predominant form among adults,⁷ categorized by etiology into aponeurotic, myogenic, neurogenic, mechanical, or traumatic origins.⁸ Drug-related blepharoptosis, as a relatively rare clinical manifestation, has received limited attention. However, with the increasing variety and widespread use of medications in recent years, it has emerged as a hidden risk in clinical practice.⁹ Drug-related blepharoptosis may occur through various mechanisms, such as effects on the neuromuscular junction or inflammation of the levator palpebrae superioris muscle.^10,11

Although some studies have explored the potential risk of drug-related blepharoptosis, existing literature mainly focuses on small sample case analyses and lacks extensive, systematic research to comprehensively assess the risk of drug-related blepharoptosis. With the diversification of drug therapies and the increasing prevalence of long-term medication use, drug-related blepharoptosis may become an increasingly important clinical issue, but our understanding of it remains limited. Real-world studies provide us with extensive and rich data sources that can more accurately reflect adverse drug events (ADEs) across diverse clinical settings and patient groups.^12,13 FDA Adverse Event Reporting System (FAERS) aggregates a large amount of real-world drug adverse reaction data, making it an important tool for evaluating drug safety.^14,15 By conducting an in-depth analysis of the FAERS database, we can more comprehensively identify the potential signals of drug-related blepharoptosis and provide more valuable drug safety assessment results for clinical practice.

This study aims to utilize the FAERS database, based on large-scale real-world data, to systematically assess the correlation between different drugs and blepharoptosis. Through this research, we hope to provide clinicians with more precise drug usage guidelines and signal stratification lists, helping to improve drug safety, reduce the occurrence of drug-related adverse reactions, and provide a solid theoretical foundation for future research and treatment strategies.

Methods

Study data source and cleaning process

In this pharmacovigilance study, we conducted an analysis of adverse events related to blepharoptosis based on the FAERS database. This study was reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. The data were collected from the FAERS database covering the period from 2004 to 2024 available at: (https://www.fda.gov/drugs/drug-approvals-and-databases/fda-adverse-event-reporting-system-faers-database). The database primarily includes spontaneous reports of adverse events from global healthcare professionals, drug users, and pharmaceutical manufacturer. Each report in FAERS contains structured information, including patient demographics, drug exposure, reported adverse events, and relevant medical history. This study did not require ethical approval or informed consent because the FAERS data are openly accessible and do not contain any personally identifiable information.

This study began with 21,838,627 original individual case safety reports (ICSRs) submitted to the FAERS database from 2004 to 2024. Following FDA guidelines for deduplication, reports were organized by PRIMARYID, CASEID, and FDA_DT. For multiple entries with identical CASEID and FDA_DT, only the report with the highest PRIMARYID and the most recent FDA_DT was retained to ensure inclusion of the most updated version of each case.¹⁶ After removing 2,296,636 duplicate reports based on this approach, a total of 19,541,994 unique ICSRs remained. Among these, 9324 ICSRs involved cases of blepharoptosis, corresponding to reports that may or may not reflect unique patients, as multiple reports can be submitted for the same individual. To enhance specificity in causal attribution, only drug-event pairs where the drug was designated as the primary suspect were included in the analysis. In total, 1484 drug products were identified in association with these blepharoptosis-related ICSRs. After applying four disproportionality analysis methods, 1441 drugs were excluded. Ultimately, 43 drugs met the criteria across all four methods and were considered to have potential signals associated with blepharoptosis. After further standardizing drug brand and generic names and excluding ambiguous drugs, the list was reduced to 20 drugs ultimately confirmed to be associated with blepharoptosis ADEs after rigorous review. The specific data cleaning process is illustrated in Figure 1.

Figure 1.

Data cleaning flowchart for drug-related blepharoptosis based on FAERS database analysis.

ADEs identification

The definition of ADEs analyzed in this study was based on the MedDRA^® (Medical Dictionary for Regulatory Activities) 20.0 version (http://www.meddra.org/).¹⁷ Adverse events were coded using MedDRA preferred terms (PTs), and standardized MedDRA queries were employed to identify PTs related to blepharoptosis. In this study, we utilized a “narrow” range of PTs.¹⁸

Disproportionality analysis

According to previous research,^12,19–21 disproportionality methods including reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS) were used for ADEs signal detection in pharmacovigilance studies. These four methods are based on comparing target events and target drugs with all other events and drugs using contingency table-based calculations to detect potential positive signals (Supplemental Tables 1 and 2). The criteria for identifying positive signals are as follows: (1) For ROR, the criterion is ROR ⩾ 3 with a 95% confidence interval (CI; lower bound of ROR 95% CI) >1; (2) For PRR, the criterion is PRR ⩾ 3 with a PRR 95% CI lower bound >1; (3) For BCPNN, the criterion is information component (IC025) >0; (4) For MGPS, the criterion is Empirical Bayes Geometric Mean (EBGM) ⩾ 2 and posterior distribution (EBGM05) >0. Once drugs meet the standards established by all four methods, it indicates a potential association between the drug and the event. Descriptive analyses were performed on patients reporting blepharoptosis. The onset times for drug-related blepharoptosis were compared, and the signals strength of drugs associated with blepharoptosis were classified based on BCPNN values. Subgroup analyses based on age and gender were also performed. Statistical analysis was conducted using R (version 4.4.1) (R Foundation for Statistical Computing, Vienna, Austria), with p < 0.05 considered statistically significant.

Results

Patient baseline data

Among the 9324 patients, the mean age was 54.0 ± 18.8 years, and the mean weight was 69.6 ± 25.0 kg. The majority of patients were female (5932; 63.6%), while 2443 (26.2%) were male. The number of female patients reported each year was consistently higher than that of male patients. Regarding clinical outcomes reported in ICSRs involving blepharoptosis, 1720 patients (18.4%) were hospitalized, 370 (4.0%) reported disability, 270 (2.9%) died, 194 (2.1%) faced life-threatening conditions, and 6770 (72.6%) experienced other serious conditions. It is important to note that these serious outcomes may be related to other adverse events listed in the same reports, rather than blepharoptosis itself. The routes of drug administration varied widely: intramuscular administration was reported in 2088 cases (22.4%), oral in 1530 (16.4%), intravenous in 914 (9.8%), subcutaneous in 588 (6.3%), ophthalmic in 276 (3.0%), and other routes in 3928 (42.1%). Reports were primarily submitted by consumers (3684; 39.5%), followed by physicians (2562; 27.5%), other healthcare professionals (770; 8.3%), pharmacists (282; 3.0%), and other occupations (2026; 21.7%). Geographically, the majority of reports originated from the United States (5616; 58.4%), with additional contributions from Canada (450; 4.7%), Japan (353; 3.7%), the United Kingdom (299; 3.1%), France (234; 2.4%), Germany (155; 1.6%), Italy (106; 1.1%), Brazil (79; 0.8%), and other countries (2032; 21.1%). The detailed information can be found in Table 1 and Figure 2.

Table 1.

Baseline data of patients with blepharoptosis reported in the FAERS database.

Variables	Data presentation	Value
Age (years)	Mean ± SD	54.0 ± 18.8
	Median (Q1, Q3)	56 (44, 68)
Weight (kg)	Mean ± SD	69.6 ± 25.0
	Median (Q1, Q3)	68 (56, 81)
Gender
Female	n (%)	5932 (63.6%)
Male	n (%)	2443 (26.2%)
Unknown	n (%)	949 (10.2%)
Outcome
Hospitalization—initial or prolonged	n (%)	1720 (18.4%)
Disability	n (%)	370 (4.0%)
Death	n (%)	270 (2.9%)
Life-threatening condition	n (%)	194 (2.1%)
Other serious condition	n (%)	6770 (72.6%)
Route
Intramuscular	n (%)	2088 (22.4%)
Oral	n (%)	1530 (16.4%))
Intravenous	n (%)	914 (9.8%)
Subcutaneous	n (%)	588 (6.3%)
Ophthalmic	n (%)	276 (3.0%)
Other administration	n (%)	3928 (42.1%)
Occupation reported
Consumer	n (%)	3684 (39.5%)
Physician	n (%)	2562 (27.5%)
Health-professional	n (%)	770 (8.3%)
Pharmacist	n (%)	282 (3.0%)
Others	n (%)	2026 (21.7%)
Country
United States	n (%)	5616 (58.4%)
Canada	n (%)	450 (4.7%)
Japan	n (%)	353 (3.7%)
United Kingdom	n (%)	299 (3.1%)
France	n (%)	234 (2.4%)
Germany	n (%)	155 (1.6%)
Italy	n (%)	106 (1.1%)
Brazil	n (%)	79 (0.8%)
Others	n (%)	2032 (21.1%)

FAERS, FDA Adverse Event Reporting System.

Figure 2.

Distribution of baseline data of patients with blepharoptosis adverse events reported in the FAERS database. (a) Age distribution of patients by gender; (b) patient distribution over time of adverse event reporting; (c) distribution of reporters’ occupations; (d) distribution of drug administration routes; (e) distribution of patient outcomes; (f) distribution of patients by country; and (g) geographic distribution of patients.

Positive signal values of drugs

After conducting disproportionality analysis, a total of 20 drugs were identified as having positive signals (Figure 3 and Supplemental Table 3). Botulinum toxin type A emerged as the drug with the highest number of blepharoptosis-related adverse events, with a total of 970 positive signals. Pembrolizumab followed with 121 positive signals, also showing a strong association with blepharoptosis. Eculizumab (26 positive signals) and oxaliplatin (28 positive signals) were among the drugs with moderate numbers of blepharoptosis-related adverse events. Other drugs included latanoprost (24 positive signals), rosuvastatin (18 positive signals), avelumab (11 positive signals), lidocaine (9 positive signals), bupivacaine (7 positive signals), and bimatoprost (7 positive signals). Drugs with a lower association with blepharoptosis included teprotumumab (5 positive signals), trastuzumab deruxtecan (3 positive signals), cemiplimab (3 positive signals), zidovudine (3 positive signals), prednisolone acetate (4 positive signals), and interferon alfa-2b (4 positive signals).

Figure 3.

Bubble chart of adverse signal counts by group. Different colors represent different groups, and the size of the circles corresponds to the magnitude of the adverse signal count.

List of potential positive drugs categorized by treatment purpose

The 20 drugs identified in the disproportionality analysis were classified into four categories based on the Anatomical Therapeutic Chemical (ATC) classification system (Figures 3 and 4; Supplemental Table 3): antineoplastic and immunomodulating agents, sensory organ drugs, nervous system drugs, and other drugs. Among the antineoplastic and immunomodulating agents, ravulizumab had the highest signal strength (ROR = 24.39, 95% CI: 14.41–41.28), followed by avelumab (ROR = 11.95, 95% CI: 6.61–21.62), cemiplimab (ROR = 9.07, 95% CI: 2.92–28.15), trastuzumab deruxtecan (ROR = 8.03, 95% CI: 2.59–24.95), nivolumab (ROR = 7.53, 95% CI: 5.74–9.89), melphalan (ROR = 6.8, 95% CI: 5.49–8.42), pembrolizumab (ROR = 5.9, 95% CI: 4.92–7.08), teprotumumab (ROR = 5.71, 95% CI: 2.37–13.74), ipilimumab (ROR = 5.53, 95% CI: 3.96–7.73), eculizumab (ROR = 4.69, 95% CI: 3.18–6.90), interferon alfa-2b(ROR = 4.64, 95% CI: 1.74–12.37), and oxaliplatin (ROR=3.57, 95% CI: 2.46–5.18). In the sensory organ drugs category, latanoprost (ROR = 16.88, 95% CI: 11.29–25.25) and bimatoprost (ROR = 5.33, 95% CI: 2.54–11.20) were identified. Nervous system drugs included lidocaine (ROR = 5.92, 95% CI: 3.07–11.39) and bupivacaine (ROR = 5.78, 95% CI: 2.75–12.15). Under other drugs, botulinum toxin type A emerged as the drug with the strongest signal of blepharoptosis (ROR = 181.96, 95% CI: 168.02–197.05), followed by prednisolone acetate (ROR = 16.16, 95% CI: 6.05–43.15), zidovudine (ROR = 8.06, 95% CI: 2.60–25.03), and rosuvastatin (ROR = 7.45, 95% CI: 4.68–11.84).

Figure 4.

Signal values of drugs in different ATC categories.

List of drugs reported and unreported in the literature

Several drugs have been previously reported to be associated with blepharoptosis (Table 2; Figure 5), including avelumab, botulinum toxin type A, bupivacaine, interferon alfa-2b, ipilimumab, latanoprost, nivolumab, oxaliplatin, pembrolizumab, rosuvastatin, and zidovudine. These drugs have shown a potential link to the development of blepharoptosis, which is consistent with the findings of this study.

Table 2.

Statistical values and distribution of drug-related blepharoptosis.

Drug	ATC classification	Number	ROR (95% CI)	PRR (95% CI)	MGPS (95% CI)	BCPNN (95% CI)	PRR (X²)	p Value
Ravulizumab	Antineoplastic and immunomodulating agents	14	24.39 (14.41–41.28)	24.29 (23.76–24.81)	24.16 (15.56–37.53)	4.59 (2.93–6.26)	24.29 (310.99)	<0.001
Avelumab	Antineoplastic and immunomodulating agents	11	11.95 (6.61–21.62)	11.93 (11.34–12.52)	11.88 (7.24–19.51)	3.57 (1.9–5.24)	11.93 (109.69)	<0.001
Cemiplimab	Antineoplastic and immunomodulating agents	3	9.07 (2.92–28.15)	9.05 (7.92–10.18)	9.04 (3.5–23.34)	3.18 (1.51–4.85)	9.05 (21.47)	<0.001
Trastuzumab deruxtecan	Antineoplastic and immunomodulating agents	3	8.03 (2.59–24.95)	8.02 (6.89–9.15)	8.02 (3.11–20.69)	3 (1.33–4.67)	8.02 (18.43)	<0.005
Nivolumab	Antineoplastic and immunomodulating agents	53	7.53 (5.74–9.89)	7.52 (7.25–7.8)	7.39 (5.89–9.28)	2.89 (1.22–4.55)	7.52 (293.88)	<0.001
Melphalan	Antineoplastic and immunomodulating agents	87	6.8 (5.49–8.42)	6.79 (6.58–7.01)	6.6 (5.52–7.89)	2.72 (1.06–4.39)	6.79 (415.55)	<0.001
Pembrolizumab	Antineoplastic and immunomodulating agents	121	5.9 (4.92–7.08)	5.9 (5.71–6.08)	5.67 (4.87–6.61)	2.5 (0.84–4.17)	5.9 (469.48)	<0.001
Teprotumumab	Antineoplastic and immunomodulating agents	5	5.71 (2.37–13.74)	5.71 (4.83–6.58)	5.7 (2.73–11.87)	2.51 (0.84–4.18)	5.71 (19.37)	<0.001
Ipilimumab	Antineoplastic and immunomodulating agents	35	5.53 (3.96–7.73)	5.53 (5.2–5.86)	5.47 (4.14–7.23)	2.45 (0.78–4.12)	5.53 (128.14)	<0.001
Eculizumab	Antineoplastic and immunomodulating agents	26	4.69 (3.18–6.9)	4.68 (4.3–5.07)	4.65 (3.36–6.42)	2.22 (0.55–3.88)	4.68 (74.56)	<0.001
Interferon alfa-2b	Antineoplastic and immunomodulating agents	4	4.64 (1.74–12.37)	4.63 (3.65–5.62)	4.63 (2.04–10.52)	2.21 (0.54–3.88)	4.63 (11.39)	<0.005
Oxaliplatin	Antineoplastic and immunomodulating agents	28	3.57 (2.46–5.18)	3.57 (3.19–3.94)	3.54 (2.59–4.83)	1.82 (0.16–3.49)	3.57 (51.17)	<0.001
Lidocaine	Nervous system	9	5.92 (3.07–11.39)	5.91 (5.26–6.57)	5.9 (3.41–10.2)	2.56 (0.89–4.23)	5.91 (36.62)	<0.001
Bupivacaine	Nervous system	7	5.78 (2.75–12.15)	5.78 (5.04–6.52)	5.77 (3.1–10.73)	2.53 (0.86–4.19)	5.78 (27.59)	<0.001
Latanoprost	Sensory organs	24	16.88 (11.29–25.25)	16.84 (16.44–17.24)	16.69 (11.92–23.38)	4.06 (2.39–5.73)	16.84 (354.32)	<0.001
Bimatoprost	Sensory organs	7	5.33 (2.54–11.2)	5.33 (4.59–6.07)	5.32 (2.86–9.9)	2.41 (0.74–4.08)	5.33 (24.56)	<0.001
Botulinum toxin type A	Other drugs	970	181.96 (168.02–197.05)	178.3 (178.22–178.38)	112.98 (105.69–120.77)	6.82 (5.15–8.49)	178.3 (108033.51)	<0.001
Prednisolone acetate	Other drugs	4	16.16 (6.05–43.15)	16.12 (15.14–17.09)	16.09 (7.07–36.6)	4.01 (2.34–5.68)	16.12 (56.63)	<0.001
Zidovudine	Other drugs	3	8.06 (2.6–25.03)	8.05 (6.92–9.18)	8.04 (3.12–20.76)	3.01 (1.34–4.68)	8.05 (18.51)	<0.001
Rosuvastatin	Other drugs	18	7.45 (4.68–11.84)	7.44 (6.97–7.9)	7.39 (5.01–10.9)	2.89 (1.22–4.55)	7.44 (99.62)	<0.001

ATC, Anatomical Therapeutic Chemical; BCPNN, Bayesian confidence propagation neural network; MGPS, multiitem gamma Poisson shrinker; PRR, proportional reporting ratio; ROR, reporting odds ratio.

Figure 5.

List of drugs classified based on literature reports.

However, other drugs such as bimatoprost, cemiplimab, eculizumab, lidocaine, melphalan, prednisolone acetate, ravulizumab, teprotumumab, and trastuzumab deruxtecan have not been previously associated with blepharoptosis. In this study, significant association was found between these drugs and blepharoptosis, highlighting the variability in adverse event profiles across different medications.

Differences in the onset time of ADEs

The cumulative reporting rate curve shows that adverse reactions to neuropsychiatric drugs occur earlier, with the curve rising quickly and then leveling off. In contrast, adverse reactions to antineoplastic and immunomodulatory drugs occur later, with a slower increase in the curve (Figure 6(a)). The reported times for the other two groups (sensory organ drugs and other drugs) fall between these two, with statistically significant differences (p < 0.0001). Additionally, the median onset time for neuropsychiatric drugs was 34.22 days, while the median onset time for antineoplastic and immunomodulatory drugs was 164.79 days. The onset times for sensory organ drugs and other drugs were between these two, at 140.92 and 153.09 days, respectively (Figure 6(b)).

Figure 6.

Onset time of adverse reactions in blepharoptosis. (a) Cumulative reporting timeline of drug-related blepharoptosis based on different ATC classifications; (b) Differences in median onset times by classification. The horizontal line indicates a statistically significant difference between the two drug categories.

For individual drugs (Figure 7(b)), the median onset times for lidocaine, bimatoprost, botulinum toxin type A, and prednisolone acetate were relatively short, at 1, 2, 6, and 9 days, respectively. In contrast, the median onset times for ravulizumab and zidovudine were the longest, at 912 and 676 days, respectively.

Figure 7.

Drug-related blepharoptosis signals and onset time ranking. (a) Signals ranking of drugs associated with blepharoptosis; (b) Median onset time ranking of drug-related blepharoptosis.

Signal intensity ranking of drugs

The BCPNN algorithm was used to rank the drug-related signals of blepharoptosis. Values ranging from 0 to 1.5 were considered to indicate a weak association, those between 1.5 and 3 suggested a moderate association, and values greater than 3 were interpreted as reflecting a strong association.^12,19 Botulinum toxin type A, ravulizumab, and latanoprost were ranked as the top three drugs with the highest signals. Eculizumab, interferon alfa-2b, and oxaliplatin exhibited relatively lower drug-related signals. Detailed information can be found in Figure 7(a).

Age and gender differences in ADEs-related blepharoptosis

We conducted a subgroup analysis based on age and gender for patients. The highest ROR value among all subgroups for each drug was set as 1. The ROR values of the other subgroups were then expressed as relative proportions compared to the highest ROR value subgroup. A ROR value of 0 indicates that the drug had no statistically significant association within the corresponding subgroup. It was found that avelumab, eculizumab, ipilimumab, nivolumab, oxaliplatin, pembrolizumab, and rosuvastatin showed a stronger signal in the population aged over 65 years. Bimatoprost, botulinum toxin type A, bupivacaine, interferon alfa-2b, latanoprost, lidocaine, melphalan, ravulizumab, trastuzumab deruxtecan, and zidovudine showed a stronger signal in the population aged under 65 years. Additionally, avelumab, eculizumab, latanoprost, melphalan, nivolumab, pembrolizumab, ravulizumab, rosuvastatin, and zidovudine exhibited a stronger signal in the male population. Bimatoprost, botulinum toxin type A, bupivacaine, ipilimumab, lidocaine, oxaliplatin, prednisolone acetate, and trastuzumab deruxtecan showed a stronger signal in the female population. Detailed information can be found in Figure 8.

Figure 8.

Normalized radar plot of subgroup analysis for drug-related blepharoptosis: (a) Subgroup analysis based on age and (b) subgroup analysis based on gender.

Sensitivity analysis

To assess the robustness of our findings, we conducted a sensitivity analysis including only adverse event reports submitted by physicians. All 20 previously identified drugs remained positively associated with blepharoptosis in this restricted dataset. Given that almost 40% of ADE reports were submitted by consumers, which may introduce reporting bias, this physician-only analysis helps to mitigate such bias and reinforces the reliability of our results. Detailed findings are provided in Supplemental Table 4.

Discussion

Previous studies on drug-related blepharoptosis have been mainly limited to case reports and small sample analyses. In this study, we conducted a disproportionality analysis of over 21 million adverse event reports from the FAERS database spanning from 2004 to 2024, identifying 20 drugs with significant associations to blepharoptosis. These drugs were primarily categorized into four groups: antineoplastic and immunomodulating agents, sensory organ drugs, nervous system drugs, and other drugs. We ranked these drugs by signals strength, assessed the onset time of blepharoptosis, and comprehensively reviewed relevant literature to provide supporting evidence for each identified drug. Furthermore, we performed subgroup analyses based on age and sex. Our findings aim to provide a reliable list of drugs associated with blepharoptosis as a clinical reference. To the best of our knowledge, this study represents the first large-scale, real-world investigation of drug-related blepharoptosis using the FAERS database.

From a therapeutic perspective, our drug disproportionality analysis, based on the ATC classification, highlighted antineoplastic and immunomodulatory agents, sensory organ drugs, and nervous system drugs as categories with significant associations. These results are consistent with previous research. Antineoplastic and immunomodulatory agents may induce muscle inflammation and neuromuscular junction dysfunction, leading to painless blepharoptosis accompanied by muscle inflammation and immune system abnormalities.²² Additionally, ocular drugs, such as prostaglandin analogs and local corticosteroids, may cause blepharoptosis through alterations in periorbital fat metabolism and nerve conduction abnormalities.^23,24 Among drugs affecting the nervous system, anesthetics may cause degeneration and regeneration of the levator palpebrae superioris or extraocular muscle fibers, leading to transient or permanent muscle weakness and, consequently, blepharoptosis.²⁵ Therefore, clinicians should remain cautious of the potential risks of these drugs, especially in oncology, anesthesia, and ophthalmology treatments. From previous clinical evidence, among the 20 identified drugs, 11 exhibited literature-supported associations with blepharoptosis. Nine drugs were newly identified as potential contributors to blepharoptosis. Notably, immune checkpoint inhibitors (such as avelumab and nivolumab) and local neuro-modulating agents (such as botulinum toxin type A and bupivacaine) accounted for a significant proportion of the known drugs, which may be related to their mechanisms of action. The former may induce immune-mediated myasthenia syndrome,²⁶ while the latter may cause neuromuscular conduction inhibition,^27,28 leading to levator palpebrae superioris dysfunction. Although the mechanisms for the newly identified drugs remain to be fully elucidated, this study expands the boundaries of our understanding of drug-related blepharoptosis and highlights the importance of recognizing these associations. Further research integrating electrophysiological and perceptual approaches may help to elucidate the broader visual system implications of drug-induced blepharoptosis.^29,30

Regarding drug signal strength, botulinum toxin type A was identified as the strongest signal drug. Commonly used for treating facial wrinkles, migraines, and blepharospasm, botulinum toxin type A works by blocking nerve signal transmission to reduce muscle activity.³¹ However, in certain cases, it may lead to blepharoptosis.³² Therefore, clinicians should exercise caution when injecting this drug, ensuring correct dosing and injection sites. Ravulizumab, an immunosuppressant used for complement-mediated diseases,^33,34 was found to be another drug with strong signals. Although ravulizumab is used to treat myasthenia gravis by inhibiting complement C5 to prevent neuromuscular junction destruction,^35,36 it may inadvertently disrupt neuromuscular signaling, leading to levator palpebrae dysfunction and blepharoptosis. Latanoprost, a drug used to lower intraocular pressure by increasing aqueous outflow,³⁷ was also identified as a strong signal drug, with prolonged use potentially causing changes in eyelid muscle tone or positioning, contributing to blepharoptosis.³⁸

Our analysis of onset times revealed that drugs from the neuropsychiatric category, such as lidocaine and bupivacaine, were commonly associated with blepharoptosis in temporal proximity to administration, with onset times averaging only a few days. This is a common occurrence, usually following the administration of anesthetics during surgery.²⁶ In contrast, drugs in the antineoplastic/immunomodulatory category, such as ravulizumab and zidovudine, demonstrated longer onset times, indicating a delayed development of adverse reactions. This delayed onset is often associated with neurotoxicity, which is typically related to drug dosage and cumulative exposure.³⁹ These differences in onset times are clinically significant, as they assist clinicians in understanding the appropriate monitoring timelines for patients receiving these strong positive signal medications.

Furthermore, subgroup analyses revealed age- and gender-specific signal patterns. For example, avelumab and rosuvastatin exhibited stronger signals in males and older adults, while botulinum toxin type A and bupivacaine were more strongly associated with younger populations and females. These differences may reflect variations in drug usage patterns, hormonal influences, or age-related susceptibility to neuromuscular dysfunction. A previous study found that statins have a higher sensitivity in males and older adults, which aligns with our findings.¹⁰ Despite this, there is still limited research on the age- and sex-specific signals of other drugs.

These findings have significant clinical implications. Clinicians should be aware of the potential risks of drug-related blepharoptosis, particularly when prescribing medications known to affect neuromuscular function. The signal strength ranking and onset time data presented in this study provide a practical reference for identifying high-risk drugs and developing individualized monitoring strategies. For instance, patients receiving botulinum toxin type A or ravulizumab should be closely monitored for early signs of blepharoptosis, especially if they are more vulnerable due to age or gender-related factors.

While this study provides valuable insights, several limitations should be acknowledged. First, FAERS data rely on spontaneous reports, which may underrepresent mild or unrecognized cases and overemphasize severe outcomes. It is worth noting that approximately 40% of the ADE reports were submitted by consumers, which may introduce reporting bias due to differences in clinical expertise, symptom interpretation, and reporting behavior. This potential limitation should be considered when interpreting the findings. Second, the lack of granular clinical data (e.g., comorbidities, concomitant medications) limits our ability to draw causal inferences. It is important to acknowledge that some of the observed associations between drugs and blepharoptosis may be partially attributable to the underlying diseases for which these drugs are prescribed. For example, corticosteroids are commonly used in patients with autoimmune or inflammatory disorders, some of which (e.g., myasthenia gravis or systemic lupus erythematosus) may independently contribute to the development of blepharoptosis. Similarly, antineoplastic agents may be administered in cancer patients who experience paraneoplastic syndromes or treatment-related neuropathies affecting the oculomotor system. Therefore, these associations should be interpreted with caution, and further studies adjusting for indication bias are warranted to distinguish drug effects from disease-related risks. Third, the generalizability of the results may be limited, as the majority of reports originated from the United States (58.4%), potentially skewing the findings toward regional prescribing practices. Controlled cohort studies and prospective trials are needed to validate the associations identified here, particularly for newly implicated drugs. Fourth, it should also be noted that a high disproportionality signal does not necessarily reflect a higher actual risk. Signal strength indicates the extent of disproportionate reporting within the FAERS database but may be influenced by external factors such as publicity, reporting habits, or usage frequency. Therefore, these signals should be interpreted as hypotheses-generating findings rather than definitive evidence of causality or risk magnitude.

Conclusion

Through disproportionality analysis of the FAERS database from 2004 to 2024, this study identified 20 drugs significantly associated with blepharoptosis, including 9 novel associations, with neuropsychiatric drugs showing rapid onset and antineoplastic/immunomodulatory drugs exhibiting delayed effects; subgroup analyses revealed age- and gender-specific signals, supporting tailored monitoring strategies for vulnerable populations. Prospective validation is needed to address potential underreporting and regional biases, and future longitudinal studies should confirm causality.

Supplemental Material

sj-docx-1-taw-10.1177_20420986251371983 – Supplemental material for Real-world large sample evaluation of drug-related blepharoptosis: a pharmacovigilance analysis of the FDA Adverse Event Reporting System database

Supplemental material, sj-docx-1-taw-10.1177_20420986251371983 for Real-world large sample evaluation of drug-related blepharoptosis: a pharmacovigilance analysis of the FDA Adverse Event Reporting System database by Kunhong Xiao, Xiaodong Chen, Shinan Wu, Yiyan Zhang, Ruiye Chen, Haixing Wu, Ziyi Qi, Jiahao Liu, Yuru Wang, Yanliang Miao, Yan Huang and Li Li in Therapeutic Advances in Drug Safety

Supplemental Material

sj-docx-2-taw-10.1177_20420986251371983 – Supplemental material for Real-world large sample evaluation of drug-related blepharoptosis: a pharmacovigilance analysis of the FDA Adverse Event Reporting System database

Supplemental material, sj-docx-2-taw-10.1177_20420986251371983 for Real-world large sample evaluation of drug-related blepharoptosis: a pharmacovigilance analysis of the FDA Adverse Event Reporting System database by Kunhong Xiao, Xiaodong Chen, Shinan Wu, Yiyan Zhang, Ruiye Chen, Haixing Wu, Ziyi Qi, Jiahao Liu, Yuru Wang, Yanliang Miao, Yan Huang and Li Li in Therapeutic Advances in Drug Safety

Footnotes

Acknowledgements

We thank the US Food and Drug Administration for providing open access to the FAERS database, which made this research possible.

Declarations

ORCID iDs

Kunhong Xiao

Shinan Wu

Supplemental material

Supplemental material for this article is available online.

References

Sridharan

Tallis

Leatherbarrow

, et al. A community survey of ptosis of the eyelid and pupil size of elderly people. Age Ageing 1995; 24(1): 21–24.

Hashemi

Khabazkhoob

Emamian

, et al. The prevalence of ptosis in an Iranian adult population. J Curr Ophthalmol 2016; 28(3): 142–145.

Kim

Cho

Zhao

, et al. Prevalence and associated factors of blepharoptosis in Korean adult population: The Korea National Health and Nutrition Examination Survey 2008–2011. Eye (Lond) 2017; 31(6): 940–946.

Bacharach

Lee

Harrison

, et al. A review of acquired blepharoptosis: prevalence, diagnosis, and current treatment options. Eye (Lond) 2021; 35(9): 2468–2481.

Richards

Jenkinson

Rumsey

, et al. The psychological well-being and appearance concerns of patients presenting with ptosis. Eye (Lond) 2014; 28(3): 296–302.

Richards

Jenkinson

White

, et al. Patient reported psychosocial functioning following successful ptosis surgery. Eye (Lond) 2022; 36(8): 1651–1655.

Lim

Hou

Singa

, et al. Relative incidence of blepharoptosis subtypes in an oculoplastics practice at a tertiary care center. Orbit 2013; 32(4): 231–234.

Sudhakar

Kosoko-Lasaki

, et al. Upper eyelid ptosis revisited. Am J Clin Med 2009; 6(3): 5–14.

Nestor

Han

Gade

, et al. Botulinum toxin-induced blepharoptosis: anatomy, etiology, prevention, and therapeutic options. J Cosmet Dermatol 2021; 20(10): 3133–3146.

10.

Fraunfelder

Richards

AB.

Diplopia, blepharoptosis, and ophthalmoplegia and 3-hydroxy-3-methyl-glutaryl-CoA reductase inhibitor use. Ophthalmology 2008; 115(12): 2282–2285.

11.

Collier

Ptosis explored: a known complication following facial botulinum toxin injections. J Aesthet Nurs 2013 ;2(1): 8–17.

12.

Chen

Yan

, et al. Drug-associated glaucoma: a real-world study based on the Food and Drug Administration adverse event reporting system database. Clin Exp Ophthal 2025; 53: 140–160.

13.

Ohashi

Munesue

Shimazaki

, et al. Long-term safety and effectiveness of diquafosol for the treatment of dry eye in a real-world setting: a prospective observational study. Adv Ther 2020; 37: 707–717.

14.

Zhou

Peng

Lin

, et al. Psychiatric disorders associated with immune checkpoint inhibitors: a pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database. EClinicalMedicine 2023; 59: 101967.

15.

Sun

, et al. Thromboembolic events associated with immune checkpoint inhibitors: a real-world study of data from the Food and Drug Administration Adverse Event Reporting System (FAERS) database. Int Immunopharmacol 2021; 98: 107818.

16.

Yin

Shu

Zhu

, et al. A real-world pharmacovigilance study of FDA Adverse Event Reporting System (FAERS) events for osimertinib. Sci Rep 2022; 12(1): 19555.

17.

Brown

Wood

The medical dictionary for regulatory activities (MedDRA). Drug Saf 1999; 20(2): 109–117.

18.

Kinoshita

Hosomi

Yokoyama

, et al. Time-to-onset analysis of amiodarone-associated thyroid dysfunction. J Clin Pharm Therap 2020; 45(1): 65–71.

19.

S-N

Huang

Wang

Y-Q

, et al. Real-world large sample assessment of drug-related dry eye risk: based on the FDA Adverse Event Reporting System database. Asia Pac J Ophthalmol 2024; 13(5): 100104.

20.

Chen

X-D

Xiao

K-H

Zhou

C-B.

Drug-induced retinal vein occlusion: a disproportionality analysis from the FDA adverse event reporting system (2004–2023). Front Pharmacol 2024; 15: 1480269.

21.

Bate

Lindquist

Edwards

, et al. A Bayesian neural network method for adverse drug reaction signal generation. Eur J Clin Pharmacol 1998; 54: 315–321.

22.

Williams

Allen

RC.

Immune checkpoint inhibitor-induced ptosis in a patient with prostate cancer. J Neuroophthalmol 2021; 41(1): e71–e72.

23.

Zheng

Roufas

Healey

, et al. Contact lens induced Baerveldt tube extrusion. Clin Exp Ophthalmol 2015; 43(7): 690–692.

24.

Habchane

Khatem

Alioua

, et al. Ptosis following the use of dexamethasone-based eye drops. Sch J Med Case Rep 2024; 2: 215–217.

25.

Rainin

Carlson

BM.

Postoperative diplopia and ptosis: a clinical hypothesis based on the myotoxicity of local anesthetics. Arch Ophthalmol 1985; 103(9): 1337–1339.

26.

Hamada

Maeda

Takase-Minegishi

, et al. Incidence and distinct features of immune checkpoint inhibitor-related myositis from idiopathic inflammatory myositis: a single-center experience with systematic literature review and meta-analysis. Front Immunol 2021; 12: 803410.

27.

Martyn

Fagerlund

Eriksson

Basic principles of neuromuscular transmission. Anaesthesia 2009; 64: 1–9.

28.

Aghai

Vazirnia

Poormatin

, et al. Effects of local anesthesia with bupivacaine plus epinephrine on blepharoptosis and levator palpebrae muscle function. Ophthalmic Plast Reconstr Surg 2013; 29(3): 198–200.

29.

Kremers

Huchzermeyer

Electroretinographic responses to periodic stimuli in primates and the relevance for visual perception and for clinical studies. Vis Neurosci 2024; 41: E004.

30.

Dering

Wright

Gheorghiu

The interaction between luminance polarity grouping and symmetry axes on the ERP responses to symmetry. Vis Neurosci 2024; 41: E005.

31.

Bhidayasiri

Truong

DD.

Expanding use of botulinum toxin. J Neurol Sci 2005; 235(1–2): 1–9.

32.

Camargo

Xia

Costa

, et al. Botulinum toxin type A for facial wrinkles. Cochrane Database Syst Rev 2021; 7(7): CD011301.

33.

Lee

Kulasekararaj

AG.

Ravulizumab for the treatment of paroxysmal nocturnal hemoglobinuria. Expert Opin Biol Ther 2020; 20(3): 227–237.

34.

Stern

Connell

NT.

Ravulizumab: a novel C5 inhibitor for the treatment of paroxysmal nocturnal hemoglobinuria. Ther Adv Hematol 2019; 10: 2040620719874728.

35.

Wiendl

Katsuno

, et al. Ravulizumab in myasthenia gravis: a review of the current evidence. Neuropsychiatr Dis Treat 2023; 19: 2639–2655.

36.

Tokuyasu

Suzuki

Uzawa

, et al. Real-world experience with eculizumab and switching to ravulizumab for generalized myasthenia gravis. Ann Clin Transl Neurol 2024; 11(5): 1338–1346.

37.

Stalmans

Lim

Oddone

, et al. MERCURY-3: a randomized comparison of netarsudil/latanoprost and bimatoprost/timolol in open-angle glaucoma and ocular hypertension. Graefes Arch Clin Exp Ophthalmol 2024; 262(1): 179–190.

38.

Nakakura

Yamamoto

Terao

, et al. Prostaglandin-associated periorbitopathy in latanoprost users. Clin Ophthalmol 2015; 9: 51–56.

39.

Smitka

von der Hagen

Schallner

, et al. Treatment of vincristin-induced bilateral ptosis. Neuropediatrics 2008; 39(1): P74.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB