Abstract
Background: Tendinopathies and ligament disorders are significant musculoskeletal adverse events associated with various drugs, leading to restricted mobility and reduced quality of life. Although certain drug classes such as fluoroquinolones and corticosteroids have established links to these conditions, there is limited research on other potential drug associations. This study aimed to comprehensively evaluate drugs associated with tendinopathies and ligament disorders using data from the USFDA Adverse Event Reporting System (AERS). Methods: A retrospective pharmacovigilance study utilizing spontaneous reports from the USFDA AERS database between March 2004 and June 2024 was conducted. The Standardized Medical Dictionary for Regulatory Activities (MedDRA) query “Tendinopathies and ligament disorders” and relevant Preferred Terms were used to identify cases. Disproportionality analysis was performed using both frequentist and Bayesian methods. Subgroup analyses were conducted by age, gender, and clinical outcomes. Results: Out of 29 153 222 reports, 40 485 unique reports were included, with 6641 related to tendon rupture and 2121 to ligament rupture. Fluoroquinolones, corticosteroids, lipid-modifying agents, immunosuppressants, and bisphosphonates were confirmed to have strong associations with tendon and ligament disorders. Emergent signals were identified for anti-inflammatory drugs, and various other drugs, including vaccines. Hospitalization rates were significantly higher in cases of tendon rupture compared to ligament rupture (P < .0001). Conclusion: This study confirms established drug associations and identifies new signals for tendinopathies and ligament disorders. Continued pharmacovigilance is necessary to validate these findings and enhance our understanding of drug-induced musculoskeletal disorders.
Keywords: tendon rupture, ligament rupture, adverse events, tendinopathies
Introduction
Lower extremity tendinopathy is increasingly recognized as a common musculoskeletal condition. A recent Dutch study reported the prevalence and incidence rates at 11.83 and 10.52 per 1000 person-years, respectively, underscoring its widespread occurrence. 1 Tendinopathy is not only associated with restricted mobility but also significantly impair the quality of life, contributing to chronic pain and functional limitations.2,3 Notably, four drug classes have been strongly implicated in the development of tendon disorders: corticosteroids, quinolone antimicrobials, aromatase inhibitors, and statins. 4 These drug-associated tendon disorders predominantly affect vulnerable populations, with risk factors including older age (60 years and above), a prior history of tendon disorders, renal impairment, diabetes mellitus, and the concurrent use of other medications linked to tendinopathy.5,6
Among these, fluoroquinolone-associated tendinopathies are particularly noteworthy for their potential severity. The onset of these conditions has been shown to be largely dose-independent, 7 and although any tendon may be affected, the large weight-bearing tendons in the lower extremities, such as the Achilles tendon, are most involved. Alarmingly, Achilles tendon rupture occurs in 30% to 40% of affected individuals, further highlighting the serious nature of this adverse drug reaction. 8
Pharmacovigilance systems like the United States Food and Drug Administration Adverse Event Reporting System (USFDA AERS) play a crucial role in identifying such adverse events. This system relies on the spontaneous reporting of adverse drug reactions by healthcare professionals. 9 Retrospective pharmacovigilance studies, leveraging statistical methods like disproportionality analysis, offer significant potential for detecting early warning signals of drug-associated adverse events, including those related to tendinopathies. 10
Although fluoroquinolones have been the primary focus of studies on drug-associated tendon disorders, research remains limited. A recent analysis of USFDA AERS data from 2016 to 2021 identified significant signals associated with three fluoroquinolones: ciprofloxacin (strongly linked to tendonitis), levofloxacin (most strongly associated with tendon rupture), and moxifloxacin, which showed a weaker association. 11 In parallel, data from VigiBase, the World Health Organization’s global database of adverse drug reactions, revealed musculoskeletal and connective tissue disorders such as arthralgia (16.34%), tendonitis (11.04%), pain in the extremities (9.98%), tendon pain (7.63%), and myalgia (7.17%) as commonly reported adverse events. 12
Despite these findings, there remains a significant gap in the literature concerning the comprehensive evaluation of drug-associated tendinopathies, particularly regarding ligament disorders and tendon ruptures. To address this gap, the present study undertakes a thorough analysis of spontaneously filed reports, aiming to identify potential signals related to drugs implicated in tendinopathies and ligament disorders.
Methods
Data Source
We searched the USFDA AERS database using a set of three queries: a Standardized Medical Dictionary for Regulatory Activities (MedDRA) queries (SMQ) “Tendinopathies and ligament disorders” (MedDRA code: 20000223), Preferred Term (PT) “Tendon rupture” (MedDRA code: 10043248), and “Ligament rupture” (MedDRA code: 10065433). The SMQ “Tendinopathies and ligament disorders” is described by MedDRA as follows: “Chronic tendinopathies and ligament disorders are pathologic conditions of tendons or ligaments, resulting mainly from repeated minor strain. Formerly called ‘tendonitis’ or ‘ligamentitis’. Etiologic factors: degenerative processes are thought to be principal underlying pathology; repeated minor strain is considered to be main precipitating factor; drug-associated forms have been described; an understanding of the pathophysiology continues to evolve. Most commonly injured tendons and ligaments: supraspinatus tendon and long head of the biceps muscle; Medial and lateral extensors of the elbow; patellar tendon; Achilles tendon; posterior tibialis tendon. Intrinsic and extrinsic risk factors: overuse (sports activities, training errors, fatigue); cold environment during outdoor training; faulty footwear/equipment; drugs (e.g., fluoroquinolone antibiotics, oral contraceptives, injected corticosteroids and statins). Commonly described presenting symptoms: pain at the site of the affected tendon or ligament; morning stiffness, local tenderness, swelling and reduced articular range of motion. Therapeutic approaches included: exercise, shock wave therapy, growth factors, nitric oxide, sclerosant therapy, gene therapy, and tissue engineering”. 13 The list of PTs included in the SMQ are outlined in the Supplemental Table 1. The data obtained pertained to the period of spontaneously filed reports between March 2004 and June 2024, encompassing a total of 82 quarterly reports.
Data Processing
The USFDA’s guidelines was followed to exclude duplicate reports. The spontaneously filed reports were initially sorted by Case ID to identify potential duplicates. For cases with identical IDs, only the record with the highest FDA ID was retained, and earlier versions were excluded. From each unique report, key demographic information including age, gender, reporting year, suspected drugs (categorized as primary suspect, secondary suspect, interacting, or concomitant), and reported outcomes were extracted. Only non-proprietary drug names were considered, and only drugs listed as the primary suspect were included in the analysis. The primary outcomes of interest were disability and hospitalization (both initial and prolonged).
Data Mining Algorithms
To detect potential signals of drug-associated tendinopathies and ligament disorders, we applied a case-non-case analysis, where cases referred to reports involving these conditions and non-cases represented reports of other adverse events. 14 Disproportionality analysis was performed using both frequentist and Bayesian methods via OpenVigil 2.1 software.
For the frequentist approach, we employed reporting odds ratio (ROR) and proportional reporting ratio (PRR). 15 The ROR was estimated as the ratio of odds of tendon/ligament disorders being reported for a given drug to the odds of the same event being reported for all other drugs. The PRR is estimated by the ratio of proportion of reports for a specific drug with the tendon/ligament disorders over the proportion of reports with the specific tendon/ligament disorders for all other drugs. Signal detection criteria required at least three unique reports, a PRR of 2 or higher, and a Chi-square (χ2) value of 4 or greater. 15 For Bayesian analysis, we used the confidence propagating neural network and multi-item gamma Poisson Shrinker methods to estimate the information component (IC). Signals were considered present if the lower limit of the 95% CI of the IC (IC025) exceeded zero. Similarly, signals in the empirical Bayes geometric mean (EBGM) method were detected when the lower 95% CI limit (EBGM05) exceeded 2. 15 All drugs were coded using the first level of the Anatomical Therapeutic Chemical (ATC) classification system of the World Health Organization. 16 According to the Council for International Organizations of Medical Sciences (CIOMS) Working Group VIII report, there is no “gold standard” measure for signal detection. The Working Group recommends using a combination of statistical approaches from both frequentist and Bayesian paradigms for more robust signal identification. 17 Additionally, the report indicates that Bayesian algorithms are more effective in mitigating false-positive associations that are inevitable with any signal detection measure. 17 To enhance the readability and understanding on the most common drugs implicated with tendon (and/or ligament) rupture, we focused on the drugs with at least 100 spontaneously filed adverse event reports. This manuscript adheres to the Reporting of a Disproportionality Analysis for Drug Safety Signal Detection Using Spontaneously Reported Adverse Events in Pharmacovigilance (READUS-PV) guidelines. 18
Statistical Analysis
Descriptive statistics were employed to summarize the demographic characteristics of the reports. Separate analyses were conducted for SMQ and the two PTs. Chi-square tests were used to compare the proportion of outcomes across different drug classes. Sub-group analyses were also performed based on age categories (0-17, 18-64, and ≥65 years), gender (male and female), and outcomes (hospitalization or disability). A P-value of ≤.05 was considered statistically significant.
Results
Search Results
A total of 29 153 222 reports were initially identified, of which 40 485 unique reports were included after deduplication and sorting for primary suspect drugs (Figure 1). Of these, 6641 reports pertained to tendon rupture, and 2121 reports involved ligament rupture. The demographic characteristics for patients with tendon and ligament disorders, as well as those specifically reporting tendon and ligament ruptures, are presented in the Supplemental Table 2. Most patients were middle-aged (18-64 years) and female. Most reports were filed between 2009 and 2012, predominantly from the United States. Reports also included complementary and alternative medicines such as fish oil, turmeric, cranberry, cinnamon, milk thistle, peppermint, ginseng, flax seed, and others; these were excluded from further analysis.
Figure 1.
Study flow diagram.
Note. The figure highlights the number of reports at every stage of this study. A total of 40 485 reports were included for the final analysis.
Signal Detection for Tendon and Ligament Disorders
The Supplemental Table 3 lists drugs with signals detected by frequentist measures. The most common drugs identified with signals for tendon and ligament disorders from both frequentist and Bayesian measures (Table 1) include the following: systemic antibacterials [levofloxacin (n = 6961), ciprofloxacin (n = 4820), moxifloxacin (n = 968), ofloxacin (n = 198), and metronidazole (n = 239)]; immunosuppressants [adalimumab (n = 3403), methotrexate (n = 2710), tofacitinib (n = 1105), leflunomide (n = 960), secukinumab (n = 901), abatacept (n = 637), tocilizumab (n = 529), certolizumab pegol (n = 500), upadacitinib (n = 452), golimumab (n = 395), and risankizumab (n = 296)]; corticosteroids [prednisone (n = 1964), hydrocortisone (n = 834), prednisolone (n = 814), methylprednisolone (n = 660), cortisone (n = 144), and desoximetasone (n = 103)], statins [atorvastatin (n = 1409), simvastatin (n = 1001), rosuvastatin (n = 669), and ezetimibe (n = 345)]; analgesics, anti-inflammatory, antirheumatic, and antigout preparations [naproxen (n = 852), rofecoxib (n = 748), diclofenac (n = 705), meloxicam (n = 534), indomethacin (n = 135), colchicine (n = 165), and febuxostat (n = 127)]; anti-osteoporotic drugs [alendronate (n = 855) and risedronate (n = 187)]; antiprotozoals, antidiarrheals and intestinal anti-inflammatory drugs [hydroxychloroquine (n = 884) and sulfasalazine (n = 755)]; endocrine drugs [amitriptyline (n = 377), anastrozole (n = 311), and letrozole (n = 300)]; psycholeptics, psychoanaleptics, anesthetics and muscle relaxants [sodium oxybate (n = 316), amitryptiline (n = 377), cyclobenzaprine (n = 263), nitrazepam (n = 114)]; diuretics and drugs acting on renin-angiotensin system [triamterene (n = 112) and telmisartan (n = 145)]; and miscellaneous drugs [tretinoin (n = 156), celecoxib (n = 801), folic acid (n = 1066), cholecalciferol (n = 1003), ergocalciferol (n = 276), antihemophilic factor (n = 109), isopropyl alcohol (n = 3175), and benzyl alcohol (n = 2207)].
Table 1.
Most Common Drugs (Number of Reports >100) with Signals from Both Frequentist and Bayesian Measures for Tendon and Ligament Disorders.
| ATC system | Drugs | PRR | χ2 | RRR | ROR | Lower 95% CI of ROR | Upper 95% CI of ROR | IC025 | EBGM05 |
|---|---|---|---|---|---|---|---|---|---|
| Agents acting on the renin-angiotensin system | Telmisartan | 2.5 | 123.4 | 2.4 | 2.5 | 2.1 | 2.90 | 1.1 | 2.1 |
| Analgesics | Diclofenac | 2.3 | 524.3 | 2.3 | 2.3 | 2.2 | 2.52 | 1.1 | 2.1 |
| Indomethacin | 5.4 | 477.9 | 5.4 | 5.5 | 4.6 | 6.48 | 2 | 4.5 | |
| Anesthetics | Sodium oxybate | 3.3 | 514.9 | 3.3 | 3.4 | 3.0 | 3.77 | 1.6 | 3 |
| Anti-acne preparations | Tretinoin | 2.8 | 182.4 | 2.8 | 2.8 | 2.4 | 3.33 | 1.3 | 2.4 |
| Antibacterials for systemic use | Levofloxacin | 51.2 | 284 460.2 | 42.6 | 59.1 | 57.5 | 60.75 | 5.3 | 41.4 |
| Ciprofloxacin | 32.2 | 128 557.2 | 28.4 | 35.3 | 34.2 | 36.40 | 4.7 | 27.6 | |
| Moxifloxacin | 14.1 | 11 513.8 | 13.8 | 14.7 | 13.8 | 15.68 | 3.5 | 12.1 | |
| Metronidazole | 2.4 | 188.5 | 2.4 | 2.4 | 2.1 | 2.71 | 1.1 | 2.1 | |
| Ofloxacin | 15.7 | 2701.6 | 15.6 | 16.5 | 14.3 | 18.98 | 3.4 | 13.5 | |
| Antidiarrheals, intestinal antiinflammatory/antiinfective agents | Sulfasalazine | 6.2 | 3266.7 | 6.1 | 6.3 | 5.9 | 6.83 | 2.4 | 5.7 |
| Antigout preparations | Colchicine | 3.8 | 338.3 | 3.8 | 3.8 | 3.3 | 4.48 | 1.6 | 3.3 |
| Febuxostat | 4.3 | 320.1 | 4.3 | 4.4 | 3.7 | 5.20 | 1.8 | 3.6 | |
| Antihemorrhagics | Antihemophilic factor | 6.2 | 470.7 | 6.2 | 6.3 | 5.2 | 7.63 | 2.2 | 5.1 |
| Anti-inflammatory and antirheumatic products | Naproxen | 3.1 | 1167.7 | 3.0 | 3.1 | 2.9 | 3.31 | 1.5 | 2.8 |
| Rofecoxib | 6.0 | 3075.8 | 5.9 | 6.1 | 5.7 | 6.58 | 2.4 | 5.5 | |
| Meloxicam | 5.5 | 1949.4 | 5.5 | 5.6 | 5.1 | 6.10 | 2.2 | 5 | |
| Antiprotozoals | Hydroxychloroquine | 4.0 | 1938.0 | 3.9 | 4.0 | 3.8 | 4.30 | 1.8 | 3.7 |
| Antiseptics and disinfectants | Isopropyl alcohol | 2.7 | 3175.9 | 2.6 | 2.7 | 2.6 | 2.83 | 1.3 | 2.5 |
| Corticosteroids for systemic use | Cortisone | 6.9 | 725.9 | 6.9 | 7.1 | 6.0 | 8.35 | 2.4 | 5.9 |
| Hydrocortisone | 7.2 | 4337.8 | 7.0 | 7.3 | 6.8 | 7.83 | 2.6 | 6.6 | |
| Prednisolone | 2.5 | 725.6 | 2.5 | 2.5 | 2.4 | 2.70 | 1.2 | 2.3 | |
| Methylprednisolone | 3.3 | 1033.6 | 3.2 | 3.3 | 3.1 | 3.58 | 1.6 | 3 | |
| Coticosteroids, dermatological preparations | Desoximetasone | 6.4 | 459.2 | 6.3 | 6.5 | 5.3 | 7.86 | 2.2 | 5.2 |
| Diuretics | Triamterene | 2.7 | 122.2 | 2.7 | 2.8 | 2.3 | 3.32 | 1.2 | 2.3 |
| Drugs for treatment of bone diseases | Alendronate | 5.6 | 3183.0 | 5.5 | 5.7 | 5.3 | 6.11 | 2.3 | 5.2 |
| Risedronate | 4.1 | 429.3 | 4.1 | 4.1 | 3.6 | 4.75 | 1.7 | 3.5 | |
| Ectoparasiticides | Benzyl alcohol | 2.9 | 2532.7 | 2.8 | 2.9 | 2.8 | 3.01 | 1.4 | 2.6 |
| Endocrine therapy | Anastrozole | 5.1 | 1002.0 | 5.0 | 5.1 | 4.6 | 5.73 | 2.1 | 4.5 |
| Letrozole | 2.6 | 287.0 | 2.6 | 2.6 | 2.3 | 2.90 | 1.2 | 2.3 | |
| Immunosuppressants | Adalimumab | 3.2 | 4660.4 | 3.0 | 3.2 | 3.1 | 3.31 | 1.5 | 2.9 |
| Methotrexate | 3.3 | 3997.9 | 3.1 | 3.3 | 3.2 | 3.43 | 1.6 | 3 | |
| Tofacitinib | 2.8 | 1205.6 | 2.7 | 2.8 | 2.6 | 2.94 | 1.4 | 2.6 | |
| Leflunomide | 6.4 | 4244.0 | 6.2 | 6.5 | 6.1 | 6.90 | 2.5 | 5.9 | |
| Secukinumab | 2.4 | 708.6 | 2.4 | 2.4 | 2.2 | 2.56 | 1.2 | 2.2 | |
| Abatacept | 2.8 | 727.2 | 2.8 | 2.8 | 2.6 | 3.05 | 1.4 | 2.6 | |
| Tocilizumab | 3.1 | 740.3 | 3.1 | 3.1 | 2.9 | 3.40 | 1.5 | 2.8 | |
| Certolizumab pegol | 2.3 | 380.3 | 2.3 | 2.4 | 2.2 | 2.57 | 1.1 | 2.1 | |
| Upadacitinib | 3.9 | 978.9 | 3.9 | 4.0 | 3.6 | 4.36 | 1.8 | 3.6 | |
| Golimumab | 3.4 | 647.7 | 3.3 | 3.4 | 3.1 | 3.74 | 1.6 | 3 | |
| Risankizumab | 2.6 | 281.3 | 2.6 | 2.6 | 2.3 | 2.90 | 1.2 | 2.3 | |
| Lipid modifying agents | Atorvastatin | 2.3 | 993.8 | 2.2 | 2.3 | 2.2 | 2.43 | 1.1 | 2.1 |
| Simvastatin | 2.4 | 798.0 | 2.4 | 2.4 | 2.3 | 2.57 | 1.2 | 2.2 | |
| Rosuvastatin | 2.5 | 617.9 | 2.5 | 2.6 | 2.4 | 2.76 | 1.2 | 2.3 | |
| Ezetimibe | 2.4 | 265.4 | 2.3 | 2.4 | 2.1 | 2.63 | 1.1 | 2.1 | |
| Rosuvastatin calcium | 3.3 | 352.8 | 3.3 | 3.4 | 2.9 | 3.85 | 1.5 | 2.9 | |
| Muscle relaxants | Cyclobenzaprine | 2.7 | 284.1 | 2.7 | 2.7 | 2.4 | 3.09 | 1.3 | 2.4 |
| Other antineoplastic agents | Celecoxib | 3.5 | 1439.2 | 3.5 | 3.6 | 3.3 | 3.84 | 1.7 | 3.3 |
| Psychoanaleptics | Amitriptyline | 2.5 | 325.8 | 2.5 | 2.5 | 2.2 | 2.75 | 1.2 | 2.2 |
| Psycholeptics | Nitrazepam | 15.4 | 1519.1 | 15.3 | 16.1 | 13.4 | 19.47 | 3.3 | 12.7 |
| Systemic hormonal preparations | Prednisone | 2.7 | 2082.9 | 2.7 | 2.8 | 2.6 | 2.89 | 1.4 | 2.5 |
| Vitamins | Folic acid | 3.2 | 1555.7 | 3.1 | 3.2 | 3.0 | 3.41 | 1.5 | 2.9 |
| Cholecalciferol | 2.4 | 796.5 | 2.4 | 2.4 | 2.3 | 2.56 | 1.2 | 2.2 | |
| Ergocalciferol | 2.5 | 240.6 | 2.5 | 2.5 | 2.2 | 2.80 | 1.2 | 2.2 |
Note. ATC = anatomical therapeutic chemical; PRR = proportional reporting ratio; χ2 = chi square; RRR = relative reporting ratio; ROR = reporting odds ratio; IC = information component; EBGM = empirical Bayes geometric mean.
Signal Detection for Tendon Ruptures
The Supplemental Table 4 lists the drugs identified with signals for tendon ruptures by frequentist measures. The most common drugs identified with signals for tendon ruptures from both frequentist and Bayesian measures (Table 2) include the following: systemic antibacterials [levofloxacin (n = 2019), ciprofloxacin (n = 1011), and moxifloxacin (n = 248)]; immunosuppressants [adalimumab (n = 500) and methotrexate (n = 333)]; corticosteroids [prednisone (n = 411), hydrocortisone (n = 229), prednisolone (n = 229), and methylprednisolone (n = 159)]; statins [simvastatin (n = 331), atorvastatin (n = 304), and rosuvastatin (n = 124)]; and other drugs [isopropyl alcohol (n = 437) and benzyl alcohol (n = 378)].
Table 2.
Most Common Drugs (Number of Reports >100) with Signals from Both Frequentist and Bayesian Measures for Tendon Rupture.
| ATC system | Drugs | PRR | χ2 | RRR | ROR | Lower 95% CI of ROR | Upper 95% CI of ROR | IC025 | EBGM05 |
|---|---|---|---|---|---|---|---|---|---|
| Antibacterials for systemic use | Levofloxacin | 107.7 | 148 504.6 | 75.2 | 112.1 | 106.3 | 118.2 | 5.9 | 71.4 |
| Ciprofloxacin | 42.7 | 34 908.9 | 36.4 | 43.5 | 40.7 | 46.6 | 4.8 | 34 | |
| Moxifloxacin | 22.3 | 4842.2 | 21.5 | 22.6 | 19.9 | 25.6 | 3.9 | 18.9 | |
| Antiseptics and disinfectants | Isopropyl alcohol | 2.2 | 281.3 | 2.2 | 2.2 | 2.0 | 2.5 | 1 | 2 |
| Corticosteroids for systemic use | Prednisone | 3.6 | 706.3 | 3.4 | 3.6 | 3.2 | 3.9 | 1.6 | 3.1 |
| Hydrocortisone | 12.2 | 2255.0 | 11.8 | 12.2 | 10.7 | 14.0 | 3.1 | 10.3 | |
| Prednisolone | 4.4 | 571.5 | 4.3 | 4.4 | 3.8 | 5.0 | 1.8 | 3.7 | |
| Methylprednisolone | 4.9 | 472.8 | 4.8 | 4.9 | 4.2 | 5.7 | 1.9 | 4.1 | |
| Ectoparasiticides | Benzyl alcohol | 3.0 | 472.1 | 2.9 | 3.0 | 2.7 | 3.3 | 1.4 | 2.6 |
| Immunosuppressants | Adalimumab | 2.8 | 540.9 | 2.7 | 2.8 | 2.6 | 3.1 | 1.3 | 2.4 |
| Methotrexate | 2.4 | 259.0 | 2.3 | 2.4 | 2.2 | 2.7 | 1.1 | 2.1 | |
| Lipid modifying agents | Simvastatin | 5.0 | 992.5 | 4.8 | 5.0 | 4.5 | 5.6 | 2 | 4.1 |
| Atorvastatin | 3.1 | 398.2 | 3.0 | 3.1 | 2.7 | 3.4 | 1.4 | 2.6 | |
| Rosuvastatin | 2.9 | 147.9 | 2.8 | 2.9 | 2.4 | 3.4 | 1.3 | 2.4 |
Note. ATC = anatomical therapeutic chemical; PRR = proportional reporting ratio; χ2 = chi square; RRR = relative reporting ratio; ROR = reporting odds ratio; IC = information component; EBGM = empirical Bayes geometric mean.
Signal Detection for Ligament Ruptures
The Supplemental Table 5 lists drugs identified with signals for ligament ruptures by frequentist measures. The most common drugs identified with signals for ligament ruptures from both frequentist and Bayesian measures (Table 3) include the following: immunomodulators [adalimumab (n = 267), interferon beta-1a (n = 119), and methotrexate (n = 113)]; systemic antibacterials [levofloxacin (n = 143)]; corticosteroids [prednisone (n = 110)]; and miscellaneous drugs [isopropyl alcohol (n = 292) and benzyl alcohol (n = 118)].
Table 3.
Most Common Drugs (Number of Reports >100) with Signals from Both Frequentist and Bayesian Measures for Ligament Rupture.
| ATC system | Drugs | PRR | χ2 | RRR | ROR | Lower 95% CI of ROR | Upper 95% CI of ROR | IC025 | EBGM05 |
|---|---|---|---|---|---|---|---|---|---|
| Antibacterials for systemic use | Levofloxacin | 17.8 | 2098.9 | 16.7 | 17.8 | 15.1 | 21.1 | 3.4 | 14.1 |
| Antiseptics and disinfectants | Isopropyl alcohol | 5.1 | 822.9 | 4.5 | 5.1 | 4.5 | 5.8 | 1.9 | 4 |
| Corticosteroids for systemic use | Prednisone | 2.9 | 132.1 | 2.8 | 2.9 | 2.4 | 3.6 | 1.2 | 2.3 |
| Ectoparasiticides | Benzyl alcohol | 2.9 | 138.8 | 2.8 | 2.9 | 2.4 | 3.5 | 1.2 | 2.3 |
| Immunostimulants | Interferon beta-1a | 5.2 | 373.8 | 4.9 | 5.2 | 4.3 | 6.2 | 1.9 | 4.1 |
| Immunosuppressants | Adalimumab | 5.0 | 738.5 | 4.5 | 5.0 | 4.4 | 5.7 | 1.9 | 4 |
| Methotrexate | 2.6 | 100.5 | 2.5 | 2.6 | 2.1 | 3.1 | 1.1 | 2.1 |
Note. ATC = anatomical therapeutic chemical; PRR = proportional reporting ratio; χ2 = chi square; RRR = relative reporting ratio; ROR = reporting odds ratio; IC = information component; EBGM = empirical Bayes geometric mean.
Signal Detection in Various Sub-Group Analyses
Detailed signal detection measures for various age groups, genders, and key outcomes (hospitalization and disability) are provided in the Supplemental Tables 6 to 12. Table 4 summarizes additional drugs that were identified with positive signals in these subgroups. Compared to the overall analysis, new signals were detected in specific subgroups, including age and gender-related differences for some drugs (Table 4).
Table 4.
List of Drugs with Positive Signals by Frequentist and Bayesian Measures in Various Sub-Groups.
| Age categories | Gender | Key reported outcomes | ||||
|---|---|---|---|---|---|---|
| ≤17 years | 18 to <65 years | ≥65 years | Males | Females | Hospitalization | Disability |
| Denosumab | Acrivastine | Cannabidiol | Colchicine | Abatacept | Bazedoxifene | Apremilast |
| Emicizumab | Bilastine | Cyclobenzprine | Testosterone | Efinaconazole | Dexlansoprazole | |
| Esomeprazole | Dextropropxyphene | Dexlansoprazole | Elbasvir | Entrectinib | ||
| Idursulfase | Glucosamine | Doxycycline | Estropipate | Febuxostat | ||
| Ipratropium | Hydroxychloroquine | Rofecoxib | Fexofenadine | |||
| Minocycline | Pegvisomant | Tiagabine | Meprobomate | |||
| Vitamin A | Tiaprofenic acid | Testosterone | ||||
| Zaleplon | Tildakizumab | |||||
Comparison of Outcomes Between the Tendon and Ligament Ruptures
The distribution of key outcomes, such as disability and hospitalization, between the tendon and ligament rupture groups is shown in Figure 2. There was a statistically significant higher occurrence of hospitalization among patients with drug-associated tendon ruptures compared to those with ligament ruptures (χ2: 114.5; P < .0001).
Figure 2.
Comparison of key outcomes between the groups.
Note. The horizontal bars depict the distributions of the number of patients with specific outcomes in each group. LR = ligament rupture; TR = tendon rupture; TLD = tendon and ligament disorders.
Discussion
Key Findings
This large-scale pharmacovigilance study identified significant safety signals for tendon and ligament disorders across multiple drug classes. Fluoroquinolone antibiotics, particularly levofloxacin and ciprofloxacin, demonstrated the strongest signals with substantial reporting frequencies. Immunosuppressive agents, particularly adalimumab and methotrexate, emerged as another major drug class associated with these disorders. Systemic corticosteroids, bisphosphonates, and statins also showed notable signals. When specifically analyzing tendon ruptures, fluoroquinolones maintained the strongest association, followed by immunosuppressants and corticosteroids. For ligament ruptures, immunomodulators, particularly adalimumab and interferon beta-1a, showed the most prominent signals. These findings not only confirm previously known associations but also highlight potential risks with newer therapeutic agents, particularly in the immunosuppressant class, warranting careful clinical consideration and ongoing surveillance. Emergent signals were identified for anti-inflammatory drugs, and various other drugs, including vaccines.
Comparison with Existing Literature
A recent study analyzing 35 667 reports on three fluoroquinolones found that ciprofloxacin had the strongest association with tendonitis (ROR 98.50, PRR 93.25, IC 6.15, and EBGM 76.80), while levofloxacin had the strongest association with tendon rupture (ROR 76.38, PRR 73.75, IC 5.84, and EBGM 63.89). 11 Our findings align with this study, as we also observed that levofloxacin showed the strongest association with tendon rupture (ROR 112.1, PRR 107.7, IC 5.9, and EBGM 71.4), followed by ciprofloxacin and moxifloxacin. Additionally, we identified significant associations of tendon rupture with other fluoroquinolones, such as ofloxacin, gatifloxacin, norfloxacin, gemifloxacin, and lomefloxacin. While we observed tendinopathy as a class effect for fluoroquinolones, a previous study refuted this. 19 Ciprofloxacin-related tendon rupture also showed the strongest signal for disability, followed by gatifloxacin, gemifloxacin, levofloxacin, and others, with prulifloxacin associated with the lowest risk. This aligns with findings by Shu et al, 11 who reported the highest risk of disability with ciprofloxacin. Although previous research has found that most tendonitis or tendon rupture cases occur within a month of fluoroquinolone initiation, this data was not available in our study, precluding similar analysis. Moreover, even in pediatric populations, levofloxacin and ciprofloxacin have shown signals of tendon rupture. 20 In contrast to the USFDA AERS findings, Baik et al 19 conducted a 10-year retrospective study of one million patients and observed a significant risk of tendon rupture only with levofloxacin (hazard ratio: 1.16; 95% CI 1.06-1.28) for rotator cuff and Achilles tendon rupture, but not with ciprofloxacin (hazard ratio: 0.96; 95% CI 0.89-1.03) or moxifloxacin (hazard ratio: 0.59; 95% CI 0.37-0.93). However, it is important to note that Baik et al compared the risk of tendon rupture with control antimicrobials, including amoxicillin, amoxicillin-clavulanate, cephalexin, and azithromycin. In our study, we also observed tendinopathy, particularly tendon rupture, with cephalexin. Therefore, it is possible that the risk of tendon rupture in other fluoroquinolones, such as ciprofloxacin and moxifloxacin, was not detected due to the comparison with control drugs also linked to tendon rupture. Additionally, Baik et al’s study focused only on elderly patients, which may not have been sufficiently powered to detect tendon rupture risk with all fluoroquinolones, highlighting only levofloxacin with the strongest association. We also observed an association between corticosteroid use and tendinopathy, including tendon rupture. van der Linden et al 21 reported an increased risk of tendon rupture in elderly patients receiving both corticosteroids and quinolones. Our findings indicated that tendon rupture was significantly associated with disability. Furthermore, even topical fluoroquinolone ear drops have been linked to an increased risk of tendon rupture. 22 Given the severity of these adverse events, fluoroquinolones should be reserved for serious infections like acute bacterial sinusitis, acute exacerbations of chronic bronchitis, and uncomplicated urinary tract infections, where alternative antimicrobials are unsuitable. 23
Lipid-modifying drugs, particularly statins, have also been associated with tendinopathy and tendon rupture, often within the first year of use. 24 Our study observed signals for all statins, which is consistent with a previous 15-year report from the French pharmacovigilance center. 25 A recent Swedish cohort study similarly found an increased risk of tendinopathy among statin users, with adjusted hazard ratios of 1.5 for men and 1.21 for women. 26 Multiple tendon ruptures without recurrence after statin withdrawal were also noted in patients with familial hypercholesterolemia. 27 Another well-known class associated with tendinopathies in the present study includes corticosteroids, that interfere with the immune system’s role in tendon healing. 28 Among bisphosphonates, our findings align with a recent study that found alendronate had the highest risk of tendinopathy (ROR 16.30, PRR 15.47, and IC 3.88), while zoledronate had the lowest (ROR 2.13, PRR 2.12, and IC 1.08). 29 In our study, we also found the highest risk associated with alendronate (PRR 5.6, ROR 5.7, IC 2.3).
Emergent signals were also observed for anti-inflammatory drugs and analgesics, including both non-selective and selective cyclooxygenase-2 inhibitors, which were linked to tendinopathy and ligament disorders. Interestingly, these drugs have been used to treat tendinopathies, but recent studies have questioned their therapeutic effects, citing no impact on the genetic expression of collagen or related growth factors. 30 Furthermore, non-steroidal anti-inflammatory drugs have been shown to impair tendon healing after injury.31,32 Hence, while our findings corroborate previous studies on the association between certain drugs and tendinopathies, there remain important nuances regarding drug class effects, co-administration, and patient populations that warrant further investigation. Given the potential severity of these adverse events, careful consideration of drug choice and monitoring is crucial in clinical practice.
The signals detected in this pharmacovigilance study reflect established clinical associations, particularly for fluoroquinolones, where the risk of tendinopathy is well-documented and has led to boxed warnings. For immunosuppressants, especially tumor necrosis factor-α (TNF- α) inhibitors and Janus kinase inhibitors, the signals align with their mechanism of action affecting collagen metabolism and tissue repair.33,34 While corticosteroids’ association with tendon weakening is known, the signals for statins suggest a need for vigilance, especially in elderly patients or those with concurrent risk factors. The frequency of reports should be interpreted in the context of each drug’s prescription volume and reporting biases. For newer agents like upadacitinib and risankizumab, these signals warrant prospective monitoring. Clinicians should consider these associations when prescribing, particularly in patients with pre-existing tendon disorders or multiple risk factors and may need to adjust monitoring strategies or consider alternative therapies in high-risk cases.
Strengths and Limitations
This study has several strengths, including the use of the USFDA AERS, a robust pharmacovigilance database that covers a wide array of spontaneously reported adverse events over an extended period. The comprehensive scope of the analysis, which examines a diverse range of drug classes, enables the identification of both established and emergent drug signals associated with tendinopathies and ligament disorders. The use of advanced signal detection methods, including both frequentist and Bayesian approaches, further strengthens the reliability of the findings. Subgroup analyses based on age, gender, and clinical outcomes add depth to the understanding of demographic and clinical factors that may influence drug-associated tendon and ligament disorders. Moreover, the rigorous data processing, adherence to the READUS-PV guidelines, and exclusion of duplicate reports enhance the credibility and reproducibility of the results.
However, there are limitations inherent to the study design. The spontaneous nature of the reporting system is prone to underreporting and reporting biases, and the lack of detailed clinical information in the reports restricts the ability to fully explore potential risk factors such as dosing regimens or concomitant conditions. The study’s reliance on disproportionality analysis limits its ability to establish causality, and the clustering of reports from specific time periods may introduce temporal bias. The interpretation of these pharmacovigilance signals requires careful consideration of several potential confounding factors. Many patients receiving these medications, particularly immunosuppressants and corticosteroids, often have underlying inflammatory conditions that independently increase their risk of tendon and ligament disorders. Additionally, the frequent use of combination therapy, especially in rheumatologic conditions, makes it challenging to attribute adverse events to a single agent. For instance, patients on TNF- α inhibitors often receive concurrent methotrexate, while those on antibiotics may be simultaneously prescribed corticosteroids. Pre-existing conditions such as diabetes, advanced age, or previous tendon injuries could also predispose patients to these complications, potentially leading to confounding by indication. The sporadic nature of spontaneous reporting and the lack of denominator data further complicate the assessment of true risk attribution. These limitations highlight the need for controlled observational studies to better quantify the independent effects of these medications on tendon and ligament disorders. Additionally, the exclusion of complementary and alternative medicine reports, as well as proprietary drug names, may overlook other contributing factors or formulations relevant to tendon and ligament disorders. Finally, given that most reports originated from the United States, the findings may not be fully generalizable to other populations or regions considering the potential geographic variations in drug use patterns, genetic predispositions, and healthcare reporting biases. Despite these limitations, the study provides valuable insights into drug safety related to tendinopathies and ligament disorders.
Conclusion
In conclusion, this study provides a comprehensive analysis of drug-associated tendinopathies and ligament disorders, leveraging data from the USFDA AERS. Our findings corroborate known associations between tendon and ligament injuries and specific drug classes, such as fluoroquinolones, corticosteroids, lipid-modifying agents, immunosuppressants, and bisphosphonates while also identifying emergent signals for other drugs, including anti-inflammatory agents and vaccines. These findings have important clinical implications: fluoroquinolones should be reserved for serious infections where alternative antibiotics are unsuitable, with careful monitoring in high-risk patients, particularly the elderly and those on concurrent corticosteroids. For chronic medications like statins and immunosuppressants, clinicians should implement regular tendon assessments during follow-up visits, especially in the first months of therapy. Risk mitigation strategies should include patient education about early warning symptoms, dose optimization when possible, and consideration of protective rehabilitation exercises for high-risk individuals. While the study’s strengths lie in its extensive data source, robust signal detection methodologies, and detailed subgroup analyses, limitations related to the spontaneous nature of adverse event reporting and potential underreporting should be acknowledged. These findings underscore the importance of ongoing pharmacovigilance and highlight the need for further research to confirm these signals and explore potential mechanisms underlying drug-induced tendinopathies and ligament disorders, ultimately informing evidence-based guidelines for risk management.
Supplemental Material
Supplemental material, sj-docx-1-hpx-10.1177_00185787251337621 for Drug-Associated Tendinopathies and Ligament Disorders: Results from a Retrospective Pharmacovigilance Study Using Disproportionality Analysis by Kannan Sridharan in Hospital Pharmacy
Acknowledgments
The authors acknowledge the use of Claude-Instant for improving the grammar and language clarity of this manuscript.
Footnotes
Authors’ Contributions: KS conceived the idea, did data analysis, interpreted the data and wrote the manuscript.
Availability of Data and Materials: The study was carried out from the publicly available data from the USFDA that can be accessed at: https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/fda-adverse-event-reporting-system-faers-public-dashboard.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Kannan Sridharan 
https://orcid.org/0000-0003-3811-6503
Supplemental Material: Supplemental material for this article is available online.
References
- 1. Albers IS, Zwerver J, Diercks RL, Dekker JH, Van den Akker-Scheek I. Incidence and prevalence of lower extremity tendinopathy in a Dutch general practice population: a cross sectional study. BMC Musculoskelet Disord. 2016;17:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Fearon AM, Cook JL, Scarvell JM, Neeman T, Cormick W, Smith PN. Greater trochanteric pain syndrome negatively affects work, physical activity and quality of life: a case control study. J Arthroplasty. 2014;29:383-386. [DOI] [PubMed] [Google Scholar]
- 3. van der Worp H, Zwerver J, Kuijer PP, Frings-Dresen MH, van den Akker-Scheek I. The impact of physically demanding work of basketball and volleyball players on the risk for patellar tendinopathy and on work limitations. J Back Musculoskelet Rehabil. 2011;24:49-55. [DOI] [PubMed] [Google Scholar]
- 4. Knobloch K. Drug-induced tendon disorders. In: Ackermann P, Hart D, eds. Metabolic Influences on Risk for Tendon Disorders. Advances in Experimental Medicine and Biology. Vol. 920, Springer; 2016. doi: 10.1007/978-3-319-33943-6_22 [DOI] [PubMed] [Google Scholar]
- 5. Kirchgesner T, Larbi A, Omoumi P, et al. Drug-induced tendinopathy: from physiology to clinical applications. Joint Bone Spine. 2014;81(6):485-492. [DOI] [PubMed] [Google Scholar]
- 6. Yu C, Guiffre BM. Achilles tendin-opathy after treatment with fluoro-quinolone. Australas Radiol. 2005;49:407-410. [DOI] [PubMed] [Google Scholar]
- 7. Williams RJ, Attia E, Wickiewicz TL. The effect of ciprofloxacin on tendon, paratenon, and capsular fibroblast metabolism. Am J Sports Med. 2000;28(3):364-369. [DOI] [PubMed] [Google Scholar]
- 8. Bolon B. Mini-review: toxic tendinopathy. Toxicol Pathol. 2017;45(7):834-837. [DOI] [PubMed] [Google Scholar]
- 9. Potential Signals of Serious Risks. New Safety Information Identified from the FDA Adverse Event Reporting System (FAERS). 2024. Accessed September 27, 2024. https://www.fda.gov/drugs/questions-and-answers-fdas-adverse-event-reporting-system-faers/potential-signals-serious-risksnew-safety-information-identified-fda-adverse-event-reporting-system
- 10. Caster O, Aoki Y, Gattepaille LM, Grundmark B. Disproportionality analysis for pharmacovigilance signal detection in small databases or subsets: recommendations for limiting false-positive associations. Drug Saf. 2020;43(5):479-487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Shu Y, Zhang Q, He X, Liu Y, Wu P, Chen L. Fluoroquinolone-associated suspected tendonitis and tendon rupture: a pharmacovigilance analysis from 2016 to 2021 based on the FAERS database. Front Pharmacol. 2022;13:990241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Huruba M, Farcas A, Leucuta DC, Bucsa C, Sipos M, Mogosan C. A VigiBase descriptive study of fluoroquinolone induced disabling and potentially permanent musculoskeletal and connective tissue disorders. Sci Rep. 2021;11(1):14375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Abate M, Silbernagel, et al. Pathogenesis of tendinopathies: inflammation or degeneration? Arthritis Res Ther. 2009;11(3):235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Faillie JL. Case-non-case studies: principle, methods, bias and interpretation. Therapie. 2019;74(2):225-232. [DOI] [PubMed] [Google Scholar]
- 15. Meyboom RH, Egberts AC, Edwards IR, Hekster YA, de Koning FH, Gribnau FW. Principles of signal detection in pharmacovigilance. Drug Saf. 1997;16(6):355-365. [DOI] [PubMed] [Google Scholar]
- 16. Hollingworth S, Kairuz T. Measuring medicine use: applying ATC/DDD methodology to real-world data. Pharmacy (Basel). 2021;9(1):60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. WHO. Practical aspects of signal detection in pharmacovigilance. Report of CIOMS working group VIII. World Health Organization; 2010. [Google Scholar]
- 18. Fusaroli M, Salvo F, Begaud B, et al. The REporting of A Disproportionality Analysis for DrUg Safety Signal Detection Using Individual Case Safety Reports in PharmacoVigilance (READUS-PV): explanation and elaboration. Drug Saf. 2024; 47(6): 585-599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Baik S, Lau J, Huser V, McDonald CJ. Association between tendon ruptures and use of fluoroquinolone, and other oral antibiotics: a 10-year retrospective study of 1 million US senior medicare beneficiaries. BMJ Open. 2020;10(12):e034844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Li Y, Wang J, Wang C, Chen L. Safety analysis of quinolones use in minors-based on the FAERS database. Front Med (Lausanne). 2024;11:1437376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. van der Linden PD, Sturkenboom MC, Herings RM, Leufkens HM, Rowlands S, Stricker BH. Increased risk of achilles tendon rupture with quinolone antibacterial use, especially in elderly patients taking oral corticosteroids. Arch Intern Med. 2003;163(15):1801-1807. [DOI] [PubMed] [Google Scholar]
- 22. Tran PT, Antonelli PJ, Winterstein AG. Quinolone ear drops and achilles tendon rupture. Clin Infect Dis. 2023;76(3):e1360-e1368. [DOI] [PubMed] [Google Scholar]
- 23. FDA Drug Safety Communication. FDA updates warnings for oral and injectable fluoroquinolone antibiotics due to disabling side effects. 2024. Accessed October 5, 2024. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-updates-warnings-oral-and-injectable-fluoroquinolone-antibiotics
- 24. Deren ME, Klinge SA, Mukand NH, Mukand JA. Tendinopathy and tendon rupture associated with statins. JBJS Rev. 2016;4(5):e4. [DOI] [PubMed] [Google Scholar]
- 25. Marie I, Delafenêtre H, Massy N, Thuillez C, Noblet C; Network of the French pharmacovigilance centers. Tendinous disorders attributed to statins: a study on ninety-six spontaneous reports in the period 1990-2005 and review of the literature. Arthritis Rheum. 2008;59(3):367-372. [DOI] [PubMed] [Google Scholar]
- 26. Eliasson P, Dietrich-Zagonel F, Lundin AC, Aspenberg P, Wolk A, Michaëlsson K. Statin treatment increases the clinical risk of tendinopathy through matrix metalloproteinase release—a cohort study design combined with an experimental study. Sci Rep. 2019;9(1):17958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Gowdar SD, Thompson PD. Multiple tendon ruptures associated with statin therapy. J Clin Lipidol. 2020;14(2):189-191. [DOI] [PubMed] [Google Scholar]
- 28. Chisari E, Rehak L, Khan WS, Maffulli N. The role of the immune system in tendon healing: a systematic review. Br Med Bull. 2020;133(1):49-64. [DOI] [PubMed] [Google Scholar]
- 29. Ji LH, Zhao CL, Wang YQ, Fu ZH. Bisphosphonates-related tendinopathies and ligament disorders: Cases analysis from the U.S. Food and Drug Administration adverse event reporting system. Bone. 2023;177:116919. [DOI] [PubMed] [Google Scholar]
- 30. Heinemeier KM, Øhlenschlæger TF, Mikkelsen UR, et al. Effects of anti-inflammatory (NSAID) treatment on human tendinopathic tissue. J Appl Physiol (1985). 2017;123(5):1397-1405. [DOI] [PubMed] [Google Scholar]
- 31. Brown KM, Saunders MM, Kirsch T, Donahue HJ, Reid JS. Effect of COX-2-specific inhibition on fracture-healing in the rat femur. J Bone Joint Surg Am. 2004;86(1):116-123. [DOI] [PubMed] [Google Scholar]
- 32. Connizzo BK, Yannascoli SM, Tucker JJ, et al. The detrimental effects of systemic Ibuprofen delivery on tendon healing are time-dependent. Clin Orthop Relat Res. 2014;472(8):2433-2439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Lee C, Oh JI, Park J, et al. TNF α mediated IL-6 secretion is regulated by JAK/STAT pathway but not by MEK phosphorylation and AKT phosphorylation in U266 multiple myeloma cells. Biomed Res Int. 2013;1:580135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Gillesberg FS, Pehrsson M, Bay-Jensen AC, et al. Regulation of fibronectin and collagens type I, III and VI by TNF-α, TGF-β, IL-13, and tofacitinib. Sci Rep. 2025;15:1087. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-hpx-10.1177_00185787251337621 for Drug-Associated Tendinopathies and Ligament Disorders: Results from a Retrospective Pharmacovigilance Study Using Disproportionality Analysis by Kannan Sridharan in Hospital Pharmacy