Abstract
BACKGROUND: Rhinophyma, a progressive nasal deformity resulting from advanced rosacea, presents significant cosmetic and functional challenges. Fully ablative CO(2) laser therapy is a recognized treatment modality, but data on its long-term efficacy and safety, and recurrence rates remain limited. OBJECTIVES: To assess the long-term outcomes, safety and patient satisfaction associated with fully ablative CO(2) laser therapy for rhinophyma and to identify predictors of treatment success using an artificial intelligence (AI)-assisted model. METHODS: A retrospective study was conducted on 152 patients with rhinophyma (grades I-III) treated with CO(2) laser therapy at an outpatient clinic affiliated with McGill University. Patients were evaluated for aesthetic improvement using the Global Aesthetic Improvement Scale (GAIS), satisfaction surveys and follow-up assessments at 1, 3, 6 and 12 months post-treatment. Demographic and clinical data were analysed using descriptive statistics, logistic regression and a deep learning model to determine predictors of hypopigmentation, recurrence and patient outcomes. RESULTS: Significant aesthetic improvement (GAIS ≥ 3) was observed in 84% of p atients, with an average satisfaction score of 2.51/3. Recurrence was rare (4%), occurring primarily in older men with grade III rhinophyma. Side effects included mild hypopigmentation (9%) and textural changes (2%). A deep learning model identified rhinophyma grade, age and Fitzpatrick phototype as key predictors of treatment outcomes. Logistic regression confirmed that advanced rhinophyma grades significantly reduced hypopigmentation risk [adjusted odds ratio (aOR) 0.43, 95% confidence interval (CI) 0.18-0.97; P = 0.041], while age significantly increased it (aOR 1.08, 95% CI 1.00-1.17, P = 0.026). Higher rhinophyma grades were also strongly associated with increased patient satisfaction (aOR 4.97, 95% CI 2.79-9.48; P < 0.001) and showed a trend toward higher recurrence risk (aOR = 6.11, 95% CI: 0.92-690.69, P = 0.064). Fitzpatrick phototype was significantly associated with patient satisfaction; decreasing Fitzpatrick phototype was associated with reduced odds of patient-reported satisfaction (aOR 0.48, 95% CI 0.25-0.87; P = 0.015). CONCLUSIONS: Fully ablative CO(2) laser therapy is highly effective and safe for rhinophyma, with most patients achieving significant improvements after a single session. AI-assisted analysis provides valuable insights into predictors of success, enabling personalized treatment plans. Older patients had an 8% increased risk of hypopigmentation per year of age, while patients with severe rhinophyma were at lower risk of hypopigmentation but at higher risk of recurrence. These findings reinforce the role of CO(2) laser therapy as a cornerstone treatment for rhinophyma while highlighting the utility of predictive analytics in guiding treatment strategies.