Abstract

Purpose: An increasing number of anecdotal reports on social media platforms and medical blogs describe dysesthesia, particularly burning skin sensations, in association with glucagon-like peptide-1 receptor (GLP-1R) agonists. The authors performed a pharmacovigilance data-mining analysis to characterise cases of dysesthesia related to GLP-1R agonists.

Methods: A disproportionality analysis was conducted using VigiBase data on GLP-1R agonists (ATC code A10BJ) and tirzepatide, focusing on the High Level Term “Paraesthesia and dysesthesia,” using the Information Component (IC). All narratives of dysesthesia cases in the French Pharmacovigilance Database were reviewed to extract clinical and pharmacological characteristics. A literature review complemented the analysis.

Results: Exenatide was significantly associated with hypoesthesia or oral paraesthesia; semaglutide and tirzepatide with hyperesthesia; and semaglutide with dysesthesia and burning sensation. Dysesthesia appears dose-dependent, occurring more frequently at higher doses and with more potent GLP-1R agonists, whether used for weight management or type 2 diabetes. Discontinuation was often performed, followed by spontaneous favorable outcomes; cases of rechallenge were observed. Skin burning sensations represent a distinctive form of dysesthesia.

Conclusion: Pharmacovigilance quantitative and qualitative data strengthen evidence for dysesthesias associated with GLP-1R agonists already observed in clinical trials of semaglutide, tirzepatide, and retatrutide.

Study at a Glance

Study Snapshot

IC₀₂₅ > 0 = statistically significant disproportionate reporting. PT = MedDRA Preferred Term. Data from VigiBase through October 25, 2025.

Study Facts Table

What the Researchers Actually Did

Laroche and colleagues conducted a two-pronged pharmacovigilance investigation. First, they queried VigiBase, the WHO’s global spontaneous adverse event reporting database containing over 33 million anonymized individual case safety reports (ICSRs), for all reports involving GLP-1R agonists (ATC code A10BJ) and tirzepatide filed through October 25, 2025. They restricted their analysis to the MedDRA High Level Term “Paraesthesia and dysesthesia,” excluding administration-site reactions. For each drug-adverse event pair, they calculated the Information Component (IC), a Bayesian disproportionality measure developed by the Uppsala Monitoring Centre. An IC₀₂₅ value exceeding zero, the lower bound of the 95% credibility interval, was used as the threshold for a statistically significant reporting signal. The IC was selected because of its conservative properties, which minimize false-positive signal generation.

Second, they reviewed all 28 narratives of dysesthesia cases in the French Pharmacovigilance Database (FPVD) involving GLP-1R agonists, retaining 15 sufficiently informative cases for qualitative analysis. The FPVD cases were evaluated by physicians or pharmacists, providing structured narratives unavailable in most VigiBase entries. The authors extracted time to onset, dose at onset, evidence of dose-dependence, dechallenge and rechallenge outcomes, and differential diagnoses. A parallel literature review identified relevant case reports and clinical trial data to contextualize the pharmacovigilance findings.

Key Findings: Primary Outcomes

• Semaglutide — hyperesthesia: 157 reports; IC 2.6, IC₀₂₅ 2.4 (statistically significant, highest IC value among all drug-reaction pairs in the analysis)
• Semaglutide — dysesthesia: 62 reports; IC 2.1, IC₀₂₅ 1.8 (statistically significant)
• Semaglutide — skin burning sensation/burning sensation (combined): 470 reports; IC 0.1, IC₀₂₅ 0.0 (borderline statistically significant at the IC₀₂₅ threshold)
• Tirzepatide — hyperesthesia: 52 reports; IC 0.7, IC₀₂₅ 0.3 (statistically significant)
• Exenatide — hypoesthesia oral: 89 reports; IC 0.8, IC₀₂₅ 0.5 (statistically significant)
• Exenatide — paraesthesia oral: 79 reports; IC 0.5, IC₀₂₅ 0.1 (statistically significant)
• In the full VigiBase cohort (n = 4,281), the most common preferred terms were paraesthesia (35.1%), hypoesthesia (29.5%), burning sensation (17.8%), skin burning sensation (8.9%), and hyperesthesia (5.4%)
• Most reports originated from the USA (64.5%) and were submitted by consumers rather than health professionals (70.6%)
• The most affected age group was 45 to 64 years (36.1% of cases with known age); women comprised 69.2% of cases
• Of 845 serious cases, 69.7% were classified as medically important conditions and 36.9% resulted in hospitalization or prolonged hospitalization

Key Findings: Secondary Outcomes and Qualitative Analysis

French Pharmacovigilance Database (n = 15)

• Semaglutide was the most frequently implicated drug (10 cases: 7 type 2 diabetes, 3 weight management); dulaglutide accounted for 4 cases; liraglutide for 1
• Mean age 58.7 ± 9.5 years; BMI range 25.3 to 45.3 kg/m²; 9 women (60%), 6 men (40%)
• A history of neuropathy was present in 4 patients, including 3 with diabetic neuropathy; no co-administered drugs known to cause dysesthesia were identified in any case
• Predominant manifestation was burning sensation (8 cases with semaglutide), localized primarily to the back, abdomen, thighs, and arms; paraesthesia (5 cases) predominantly affected the extremities
• Symptom onset occurred during dose escalation across all cases; time from first injection to onset ranged from 3 to 93 days
• Dose-dependence was documented in the majority of cases; onset was recorded at doses ranging from 0.75 mg to 2.4 mg for semaglutide and 0.75 mg to 3.0 mg for dulaglutide
• In all 3 cases with formal rechallenge, symptoms recurred at the same dose and with a comparable time-to-onset pattern as the initial episode
• Clinical workup, when performed, was unremarkable
• Dysesthesia resolved after drug discontinuation in cases where dechallenge outcome was known

Supporting Clinical Trial Data (Literature Review)

• OASIS 1 (oral semaglutide 50 mg daily): Altered skin sensation in 13% of semaglutide recipients vs. 1% placebo; generally mild to moderate, occurring during dose escalation, resolving without discontinuation
• Scoping review (semaglutide dermatologic AEs): Oral semaglutide 50 mg associated with dysesthesia (1.8%), hyperesthesia (1.2%), neuralgia (0.9%), skin pain (2.4%), paraesthesia (2.7%), sensitive skin (2.7%); skin burning sensation led to discontinuation in 1.8%; only 0.2% reported skin burning sensation with subcutaneous semaglutide 2.4 mg weekly
• High-dose semaglutide phase 2 (up to 16 mg SC weekly in type 2 diabetes/overweight): Dysesthesia rates of 0% (2 mg), 8% (8 mg), 18% (16 mg), and 2% (placebo); 2 cases required withdrawal, 4 did not recover within the follow-up period
• STEP UP (semaglutide 7.2 mg vs. 2.4 mg vs. placebo, obesity): Dysesthesia in 22.9% (7.2 mg), 6.0% (2.4 mg), and 0.5% (placebo); one in five patients in the 7.2 mg group required dose reduction; 4 discontinued due to dysesthesia
• STEP UP T2D (semaglutide 7.2 mg vs. 2.4 mg vs. placebo, obesity + T2D): Dysesthesia in 18.9% (7.2 mg), 4.9% (2.4 mg), and 0% (placebo)
• Retatrutide phase 2 (triple GLP-1/GIP/GCG agonist): Cutaneous hyperesthesia and skin sensitivity in 7% retatrutide vs. 1% placebo, dose-dependent, mild to moderate, without treatment discontinuation
• ATTAIN-1 (orforglipron, oral small-molecule GLP-1R agonist, obesity): Dysesthesia cluster (allodynia, dysesthesia, burning sensation, hyperesthesia) reported in 0.1% (6 mg), 0.1% (12 mg), 1.2% (36 mg), and 0.6% (placebo), demonstrating dose-dependent effect
• Regulatory actions: FDA added dysesthesia to post-marketing adverse reaction labeling for injectable semaglutide in January 2025; EMA included dysesthesia in the SmPC for semaglutide and tirzepatide; SmPCs for exenatide, dulaglutide, and liraglutide do not currently mention dysesthesia

Adverse Events and Safety Profile

The paper is itself a characterization of an adverse event class. Of the 4,281 dysesthesia-related ICSRs, 79.5% were non-serious. Among the 845 serious cases, 69.7% met criteria for medically important condition and 36.9% resulted in hospitalization or prolonged hospitalization. Clinical workup in the French narrative cases, when performed, yielded unremarkable findings, suggesting that the mechanism is pharmacodynamic rather than structural. No co-administered drugs known to independently cause peripheral neuropathy were identified in the French cohort. The temporal pattern, onset within 3 to 93 days of initiation, resolution after discontinuation, and recurrence at rechallenge, is consistent with a drug-attributable effect rather than a coincidental neuropathic process. Dulaglutide and liraglutide, which carry no dysesthesia labeling as of this publication, each generated cases in the French database, though their IC₀₂₅ values for most dysesthesia preferred terms did not reach statistical significance in VigiBase.

Statistical Approach and Rigor

The IC is a Bayesian disproportionality measure that compares the observed frequency of a specific drug-adverse event pair against what would be expected given overall reporting rates in VigiBase. The authors correctly selected IC₀₂₅ > 0 as their signal threshold, a conservative criterion that reduces false-positive rate. The READUS

Physician-Led, Whole-Person Care

A doctor who takes the time to truly understand you.

Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness

One-on-One Cannabis Guidance

Metabolic Balance

Leave a Message
Metabolic Care
Medical Consulting
Cannabis Care