An Assessment of ChatGPT Responses to Common Postoperative Rhinoplasty Questions

Rhinoplasty is a commonly performed aesthetic and functional procedure, and patients frequently seek postoperative information beyond routine follow-up visits. With the increasing use of artificial intelligence-based tools for health-related information, ChatGPT has emerged as a widely accessed conversational platform; however, evidence regarding the quality, reliability, and clinical appropriateness of its responses in the postoperative rhinoplasty setting remains limited. This methodological, cross-sectional study evaluated ChatGPT responses to 15 standardized postoperative rhinoplasty questions submitted in Turkish using a single publicly available model version. Responses were recorded verbatim and independently assessed by a multidisciplinary panel of 10 experienced clinicians using a structured scoring framework evaluating accuracy, completeness, safety, and patient-friendly language. Interrater reliability was analyzed using intraclass correlation coefficients. Overall, ChatGPT responses were generally rated as patient-friendly and conservative in tone. The mean total score indicated moderate overall quality, with higher scores observed for safety and patient-friendly language, while greater variability was noted in accuracy and completeness, particularly for questions related to activity resumption and procedure-specific postoperative precautions. Single-measure interrater reliability was low, whereas average-measure reliability improved substantially when ratings from all evaluators were considered. These findings suggest that ChatGPT may serve as an accessible and patient-friendly source of postoperative information following rhinoplasty; however, variability in accuracy and completeness underscores the importance of clinician oversight. ChatGPT should therefore be regarded as a supplementary informational resource rather than a substitute for individualized surgeon-patient communication.