ABSTRACT
Although topical corticosteroids are known triggers in the development of perioral dermatitis (PD), investigation into inhalant and nasal corticosteroid preparations, commonly used to manage allergic rhinitis and asthma, remains limited to case series. Using the TriNetX US Collaborative Network, this retrospective cohort study identified pediatric patients (age < 18) with allergic rhinitis or asthma in three groups: corticosteroid nasal/inhalant users, non‐steroid nasal/inhalant users, and controls with no nasal/inhalant product use. Patients were propensity score matched 1:1 from 2000 to 2025 with a new diagnosis of PD as the outcome assessed at 6 and 12 months. Corticosteroid nasal/inhalant usage showed an increased association of newly documented PD with the strongest signal occurring within 6 months (RR 3.25, p < 0.0001) remaining persistently elevated although attenuated by 1 year (RR 2.27, p < 0.0001), whereas nonsteroidal preparations showed no such association. Nasal and inhalant containing corticosteroid preparations were clearly associated with PD in this large retrospective database study. They represent a likely underrepresented trigger necessitating clinician attention when encountering new‐onset pediatric facial dermatoses.
Keywords: allergic rhinitis, asthma, corticosteroid inhalants, corticosteroid intranasal sprays, perioral dermatitis
1. Introduction
Perioral dermatitis (PD) is a chronic inflammatory skin disease that classically affects young to middle‐aged women but has been reported in children as young as 6 months of age. While pathophysiology is multifactorial, corticosteroids represent a well‐established trigger, with most pediatric cases occurring in those with a history of exposure [1, 2]. Although patients with allergic rhinitis (AR) and/or asthma are frequently managed with corticosteroid containing nasal and inhalant products, and case series have suggested possible associations of usage with PD development, no large studies have demonstrated this relationship [3, 4]. Given the widespread availability of over‐the‐counter (OTC) corticosteroid preparations, identifying potential triggers is important to help guide patient counselling. This retrospective cohort study utilizing the TriNetX US Collaborative Network aims to evaluate the association of nasal and inhalant steroid products in pediatric AR and asthma patients with new onset of PD.
2. Methods
Three pediatric cohorts (≤ 18 years of age) were identified from 2000 to 2025 with at least two instances within 3 months of each other of either vasomotor and allergic rhinitis (ICD10: J30), chronic rhinitis or nasopharyngitis (ICD10: J31), chronic sinusitis (ICD10: J32), nasal polyps (ICD10: J33), or eustachian tube disorder (ICD10: H69) for designation as nasal product users; and either asthma (ICD10: J45), chronic obstructive pulmonary disease (ICD10: J44), emphysema (ICD10: J43) or chronic bronchitis (ICD10: J42, J41) designated as inhalant product users all with no prior history of rosacea (ICD 10: L71).
Two test cohorts were defined based on medication exposure: (1) steroidal nasal or inhalant preparation use (ATC: R01AD) and (2) nonsteroidal nasal or inhalant preparation use (ATC: R01AA, R01AB, R01AC), each required to occur within 3 months of qualifying diagnosis. The control cohort consisted of patients with the above diagnoses (ICD10: J30‐33, H69, J41‐45) without any nasal or inhalant product usage. Exclusion criteria included anyone with a history of rosacea or PD prior to the index event.
Propensity score matching performed 1:1 to selected controls based on demographic and clinical factors [5]. (Table 1) The outcome of perioral dermatitis (ICD10: L71.0) was assessed at 6 months and 1 year after the index event of diagnosis (defined as diagnosis plus initiation of designated preparation). Risk ratios (RR) with 95% confidence intervals (CI) and number of instances of outcome were recorded (Table 2). p values of < 0.0001 were considered statistically significant.
TABLE 1.
Baseline characteristics of pediatric patients with corticosteroid and non‐corticosteroid containing nasal and inhalant usage and controls after propensity score matching.
| Characteristics (%) | Control (n = 252,200) | Corticosteroid nasal/inhalant (n = 252,200) | Std diff | Control (n = 50,513) | Non‐corticosteroid nasal/inhalant (n = 50,513) | Std diff |
|---|---|---|---|---|---|---|
| Age at index | 4.86 ± 3.73 | 5.05 ± 3.7 | 0.0510 | 6.39 ± 4.04 | 6.34 ± 3.93 | 0.0131 |
| Gender | ||||||
| Female | 99,270 (40.4%) | 99,281 (40.4%) | < 0.0001 | 19,208 (39.6%) | 18,910 (39.0%) | 0.0126 |
| Male | 146,078 (59.5%) | 146,081 (59.5%) | < 0.0001 | 29,328 (60.4%) | 29,625 (61.0%) | 0.0125 |
| Ethnicity | ||||||
| Hispanic or Latino | 43,192 (17.6%) | 43,422 (17.7%) | 0.0025 | 8957 (18.5%) | 9275 (19.1%) | 0.0168 |
| Not Hispanic or Latino | 160,924 (65.6%) | 156,497 (63.8%) | 0.0377 | 31,567 (65.0%) | 31,307 (64.5%) | 0.0112 |
| Race | ||||||
| White | 111,567 (45.5%) | 111,415 (45.4%) | 0.0012 | 27,102 (55.8%) | 26,633 (54.9%) | 0.0194 |
| Black or African American | 69,921 (28.5%) | 71,957 (29.3%) | 0.0183 | 11,396 (23.5%) | 11,937 (24.6%) | 0.0261 |
| Asian | 9353 (3.8%) | 8927 (3.6%) | 0.0092 | 1621 (3.3%) | 1635 (3.4%) | 0.0016 |
| Prescription | ||||||
| Nasal decongestant | 10,236 (4.2%) | 11,794 (4.8%) | 0.0307 | 5075 (10.5%) | 29,732 (61.2%) | 1.2486 |
| Systemic decongestant | 14,548 (5.9%) | 16,422 (6.7%) | 0.0314 | 4829 (9.9%) | 9160 (18.9%) | 0.2561 |
| Antiasthma/Bronchodilator | 197,662 (80.5%) | 244,009 (99.4%) | 0.6621 | 38,198 (78.7%) | 41,770 (86.0%) | 0.1940 |
| Antitussives/Expectorants | 14,318 (5.8%) | 16,795 (6.8%) | 0.0414 | 4056 (8.4%) | 7524 (15.5%) | 0.2218 |
| Respiratory agents, other | 67,526 (27.5%) | 73,227 (29.8%) | 0.0514 | 30,655 (63.1%) | 33,211 (68.4%) | 0.1111 |
| Cold remedies, combinations | 7222 (2.9%) | 8861 (3.6%) | 0.0375 | 1896 (3.9%) | 4167 (8.6%) | 0.1942 |
| Mucolytics | 299 (0.1%) | 416 (0.2%) | 0.0125 | 137 (0.3%) | 166 (0.3%) | 0.0107 |
| Diagnosis | ||||||
| Acne | 2739 (1.1%) | 2982 (1.2%) | 0.0092 | 523 (1.1%) | 715 (1.5%) | 0.0353 |
| Asthma | 210,812 (85.9%) | 211,204 (86.0%) | 0.0046 | 36,959 (76.1%) | 37,201 (76.6%) | 0.0117 |
| Atopic dermatitis | 34,687 (14.1%) | 36,228 (14.8%) | 0.0179 | 6989 (14.4%) | 7733 (15.9%) | 0.0427 |
| Seborrheic dermatitis | 11,102 (4.5%) | 11,998 (4.9%) | 0.0172 | 1542 (3.2%) | 2004 (4.1%) | 0.0507 |
| Diabetes mellitus | 587 (0.2%) | 613 (0.3%) | 0.0021 | 252 (0.5%) | 199 (0.4%) | 0.0161 |
| Diseases of the circulatory system | 11,251 (4.6%) | 13,636 (5.6%) | 0.0443 | 4420 (9.1%) | 4868 (10.0%) | 0.0314 |
| Disorders of lipoprotein metabolism | 1578 (0.6%) | 1738 (0.7%) | 0.0080 | 578 (1.2%) | 834 (1.7%) | 0.0441 |
| Essential hypertension | 1788 (0.7%) | 2261 (0.9%) | 0.0213 | 809 (1.7%) | 942 (1.9%) | 0.0206 |
| Anxiety disorder | 5451 (2.2%) | 6359 (2.6%) | 0.0241 | 2446 (5.0%) | 3101 (6.4%) | 0.0582 |
| Major depressive disorder | 414 (0.2%) | 466 (0.2%) | 0.0050 | 191 (0.4%) | 98 (0.2%) | 0.0352 |
| Alcohol‐related disorders | 44 (< 0.1%) | 40 (< 0.1%) | 0.0012 | 18 (< 0.1%) | 12 (< 0.1%) | 0.0070 |
| Tobacco use | 75 (< 0.1%) | 72 (< 0.1%) | 0.0007 | 23 (< 0.1%) | 29 (0.01) | 0.0053 |
| Dermatological agents | ||||||
| Topical anti‐inflammatory | 163,512 (66.6%) | 163,123 (66.5%) | 0.0034 | 29,291 (60.3%) | 30,199 (62.2%) | 0.0384 |
| Topical dermatologicals | 87,025 (35.5%) | 86,944 (35.4%) | 0.0007 | 38,196 (78.7%) | 38,027 (78.3%) | 0.0085 |
| Emollients | 39,902 (16.3%) | 38,848 (15.8%) | 0.0117 | 24,138 (49.7%) | 23,740 (48.9%) | 0.0164 |
| Antiacne agents | 22,215 (9.1%) | 22,570 (9.2%) | 0.0050 | 7517 (15.5%) | 7939 (16.4%) | 0.0238 |
| Soaps/shampoos/soap‐free cleaners | 9362 (3.8%) | 10,102 (4.1%) | 0.0155 | 4501 (9.3%) | 4841 (10.0%) | 0.0238 |
Note: Std diff values < 0.1 indicates good balance between cohorts. Bold indicates inclusion in propensity score matching. Dermatologic Agents Definitions: Topical Anti‐inflammatory (VA: DE200): triamcinolone, alclometasone, mometasone, betamethasone, amcinonide, crisaborole, clocortolone, fluocinolone, clobetasol, desonide, desoximetasone, diclofenac, prednicarbate, fluticasone, halobetasol, fluocinonide, flurandrenolide, halcinonide, hydrocortisone, diflorasone. Topical Dermatologicals (VA: DE900): calamine, trypsin, zinc oxide, becaplermin, ruxolitinib, brimonidine, benzyl alcohol, hypochlorite, monobenzone, aluminum acetate, calcium gluconate, camphor, anacaulase, beremagene geperpavec, deoxyribonucleases, pimecrolimus, dimethicone, Peruvian balsam, dihydroxyacetone, diphenhydramine, resorcinol, doxepin, allantoin, histamine, hydroquinone, mineral oil, light, zinc acetate, collagenase, aluminum hydroxide, hyaluronate, methapyrilene, methoxsalen, miconazole, minoxidil, collagen, activated, oxymetazoline, papain, petrolatum, witch hazel, sodium chloride. Emollients (VA: DE350): urea, vitamin A, vitamin E, lactate, colloidal oatmeal, glycerin, allantoin, lanolin, petrolatum. Antiacne agents (VA: DE750): tetracycline, tretinoin, benzoyl peroxide, attapulgite, trifarotene, sulfur, colloidal, tirbanibulin, clascoterone, clindamycin, dapsone, resorcinol, erythromycin, adapalene, ivermectin, azelate, metronidazole, minocycline, tazarotene, salicylic acid, isotretinoin. Soaps/shampoos/soap‐free cleansers (VA: DE400): selenium sulfide, zinc pyrithione, hexachlorophene, povidone‐iodine.
Abbreviation: Std diff, standardized difference.
TABLE 2.
Patient outcomes of new onset perioral dermatitis among pediatric corticosteroid and non‐corticosteroid containing nasal and inhalant product users compared with matched controls.
| Median follow up (Days) | Corticosteroid nasal/inhalant users vs. Controls (n = 252,200) | Non‐corticosteroid nasal/inhalant users vs. Controls (n = 50,513) | |
|---|---|---|---|
| 180 days | Patients with outcomes | 65 vs. 20 | Insufficient data |
| Risk % | 0.026% vs. 0.008% | ||
| ARD | 0.018% | ||
| RR | 3.25 (1.969, 5.365) | ||
| p value | < 0.0001 | ||
| Mean instances of ICD10 code | 1.185 vs. 1.25 (t = −0.527, p = 0.60) | ||
| 365 days | Patients with outcomes | 159 vs. 70 | 27 vs. 16 |
| Risk % | 0.063% vs. 0.028% | 0.053% vs. 0.032% | |
| ARD | 0.035% | 0.022% | |
| RR | 2.272 (1.715, 3.009) | 1.688 (0.91, 3.132) | |
| p value | < 0.0001 | 0.0932 | |
| Mean instances of ICD10 code | 1.195 vs. 1.286 (t = −1.13, p = 0.26) | 1.22 vs. 1.25 (t = −0.165, p = 0.87) |
Note: p values < 0.0001 considered to be statistically significant.
Abbreviations: ARD, absolute risk difference; RR, risk ratio.
3. Results
Among pediatric patients, steroid nasal/inhalant use was associated with an increased likelihood of new documentation of PD diagnosis compared to nonusers. The highest RRs were observed within 6 months of use. At median follow‐up time of 180 days, steroid use was associated with a RR of 3.25 (1.969, 5.365) with absolute risk difference (ARD) 0.018% (p < 0.0001). Decreasing, yet persistent, associations were seen at 365 days [RR 2.272 (1.715, 3.009), ARD 0.035%, p < 0.0001] (Table 2).
Nonsteroidal nasal/inhalant use versus nonuser was largely non‐contributory at early timepoints due to insufficient data, while a non‐significant association was noted at 365 days [RR 1.688 (0.91, 3.132), ARD 0.022%, p = 0.0932]. Mean ICD‐10 code uses for PD did not differ significantly between groups at any interval (Table 2).
4. Discussion
The potential for nasal or inhalant steroid usage to trigger PD has been suggested, but not confirmed outside of isolated case reports [1, 2, 3, 4]. This large‐scale study of matched pediatric patients indicates an increased association of new documentation of PD in nasal/inhalant steroid users, with the strongest signal within 6 months (RR 3.25) remaining persistently elevated, although attenuated, by 1 year (RR 2.27). The lack of significant association noted with other nasal and inhalant preparations supports a corticosteroid class effect, consistent with dermatoses arising from prolonged topical facial steroid exposure, as opposed to a vehicle‐specific effect [6].
Potential diagnosis mimickers, including acne, atopic and seborrheic dermatitis, were included in propensity score matching, while those with a history of rosacea, a common co‐diagnosis, were excluded in original patient counts [5]. Validation of ICD10 coding for PD, important for assessing accurate diagnoses in database studies, has been shown to have high positive predictive values with a single code encounter among all providers [7]. Additional strengths include close follow‐ups with two instances within 3 months for AR and/or asthma, with prescriptions documented within 3 months after a visit further reducing misclassification of indication or exposure.
Key limitations include reliance on database records, which cannot consistently capture OTC product use, or undocumented product usage, and may therefore underestimate true exposure. Although incorporated in propensity score matching, the concurrent usage of dermatologic therapies (Table 1) could not be characterized by anatomical site of application, potency, or duration of use. Nonetheless, these findings highlight nasal/inhalant corticosteroids as a plausible and underrecognized trigger to consider when counselling pediatric patients with new onset facial dermatoses. For parents and caregivers, rinsing or wiping the child's face after nasal or inhaled corticosteroid use may help decrease cutaneous exposure and lower PD risk when treatment cannot be discontinued.
Funding
The authors have nothing to report.
Consent
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Goel N. S., Burkhart C. N., and Morrell D. S., “Pediatric Periorificial Dermatitis: Clinical Course and Treatment Outcomes in 222 Patients,” Pediatric Dermatology 32, no. 3 (2015): 333–336, 10.1111/pde.12534. [DOI] [PubMed] [Google Scholar]
- 2. Nguyen V. and Eichenfield L. F., “Periorificial Dermatitis in Children and Adolescents,” Journal of the American Academy of Dermatology 55, no. 5 (2006): 781–785, 10.1016/j.jaad.2006.05.031. [DOI] [PubMed] [Google Scholar]
- 3. Brunetti T., Chessa M. A., Leuzzi M., Gurioli C., Cedirian S., and Neri I., “Pediatric Perioral Dermatitis Associated With Inhaled Corticosteroids: A Retrospective Case Series,” Dermatology Practical & Conceptual 15, no. 2 (2025): 5239, 10.5826/dpc.1502a5239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Peralta L. and Morais P., “Perioral Dermatitis – The Role of Nasal Steroids,” Cutaneous and Ocular Toxicology 31, no. 2 (2012): 160–163, 10.3109/15569527.2011.621918. [DOI] [PubMed] [Google Scholar]
- 5. Chiriac A., Chiriac A. E., Madke B., Fida M., Shinde A. B., and Wollina U., “Periorificial Dermatitis in Infants and Preschoolers – a Narrative Review,” European Journal of Pediatrics 184, no. 2 (2025): 143, 10.1007/s00431-025-05975-3. [DOI] [PubMed] [Google Scholar]
- 6. Hengge U. R., Ruzicka T., Schwartz R. A., and Cork M. J., “Adverse Effects of Topical Glucocorticosteroids,” Journal of the American Academy of Dermatology 54, no. 1 (2006): 1–18, 10.1016/j.jaad.2005.01.010. [DOI] [PubMed] [Google Scholar]
- 7. Ershadi S., Biba U., Bensellam N., et al., “Validating Optimal International Classification of Diseases‐10 Classification Approaches for Identifying Perioral Dermatitis: A Retrospective Chart Review,” Journal of the American Academy of Dermatology 93, no. 2 (2025): 487–488, 10.1016/j.jaad.2025.04.061. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.