Prognostic Model for Predicting the Probability of Post-Rhinoplastic Complications, Accounting for Patients Potential Confounding Factors

Abstract

Objectives: Post-rhinoplasty complications prognosticating can improve frontal appearance accuracy, but research on their occurrence is limited. Addressing postoperative complications and discussing them before surgery can enhance patient satisfaction and ensure a safe and successful procedure.
Methods: This study examined the outcomes of rhinoplastic surgery in patients with various conditions, including those with previous maxillofacial interventions, cardiovascular or coagulopathy disorders, uncontrolled high blood pressure, and severe nasal deformities. Patients were excluded from the study due to their age, gender, and other factors. The study was approved by the Institutional Review Board/Human Subjects Committee of the Faculty of Medicine at Hashemite University. Patients were given nasal packing, antibiotics, painkillers, and instructions on postoperative care. A cox-regressional proportional hazard model was used to predict the likelihood of complications following rhinoplasty. The study focused on composited outcomes (cOI) and evaluated the significance of ∑points. The study included a primary comparison group, which consisted of better cOI (∑cOI <3) and poorer cOI (∑cOI≥3). The data was collected and analyzed using Microsoft Office LTSC Professional Plus 2021 Excel and IBM SPSS Statistics version 25.
Results: A study of 269 patients underwent rhinoplasty surgery, with 63.57% having better cOI and classified as Group I. However, 36.43% had poorer cOI and were placed in Group II. The study included 143 women (53.2%) and 126 men (468.8%) participants. There were no statistically significant differences between the rates of men and women in Groups I and II. The study also dichotomized patients' ages into those under 35 years and those over 35 years. The distribution of patients' comorbidity burden was significant across Group I-II when the AACCI score was dichotomized into <3 and ≥3 categories. There was a significant difference in the rates of death in the non-administered Dex IV group and the administered Dex IV group. MVs were given less often versus more often, and the rates of distribution were also statistically different between Groups I and II. The study found a significant spread of both conservative strategies across Groups I and II. The cox-regressional proportional hazard modelling revealed a significant adjusted hazard ratio (HR) for all tested potential confounders except for patients' age and gender. Both TXA and TP had a negative impactful coefficient, while both potential confounders of experiencing perioperatively Dex IV and previous regularity of MVs supplementations had significantly reduced directional effects on the probability for cOI..
Conclusion: The study found a significant adjusted hazard ratio (HR) for all potential confounders except age and gender. Intraoperative TXA and post-procedural TP had a negative impactful coefficient. Perioperative Dex IV and previous regularity of MVs supplementations reduced the probability of cOI. However, patients' potential confounders of AACCI and OS had significantly adjusted HR but positively impactful on cOI.