Predicting Elective Surgical Patient Outcome Destination Based on the Preoperative Modified Frailty Index and Laboratory Values

Incidence and Risk Factors of Postoperative Complications After Rhinoplasty: A Multi-Institutional ACS-NSQIP Analysis

OBJECTIVE

Despite the popularity of rhinoplasty, outcome research has been mainly limited to single-surgeon or single-institution reports. Therefore, we performed a multi-institutional analysis to present a broader portrait of the postoperative outcomes and risk factors for adverse events after rhinoplasty surgery.

METHODS

We reviewed the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database (2009-2019) to identify patients who underwent rhinoplasty. The postoperative outcomes of interest included 30-day mortality, reoperation, readmission, and surgical and medical complications. We also analyzed risk factors for complication occurrence, including patient comorbidities and preoperative laboratory values.

RESULTS

We identified 835 patients, 72% (n=602) of whom underwent a primary, 21% (n=175) a secondary, and 6.7% (n=58) a cleft nasal deformity procedure. The average patient age was 41±17 years, with most patients being female (n=472; 57%) and white (n=643; 77%). Complications rates were generally low, with reoperation (n=19; 2.3%) and superficial incisional infection (n=9; 1.1%) account for the most common general and surgical adverse event, respectively. Multivariable analysis revealed male sex ( P =0.04) and higher ASA scores ( P <0.0001) as risk factors for complications. Low serum albumin ( P =0.04) and hematocrit ( P =0.003) levels were associated with the occurrence of any complication, whereas low serum albumin ( P =0.02) also correlated with the incidence of surgical adverse events.

CONCLUSION

Complication rates after rhinoplasty were overall low and seemed to correlate with male sex and ASA scores. We identified preoperative albumin and hematocrit as predictive biomarkers of adverse events. Preoperative nutritional optimization and management of low hematocrit may improve postoperative outcomes.

A Surgeon's Guide to Artificial Intelligence-Driven Predictive Models

Artificial intelligence (AI) focuses on processing and interpreting complex information as well as identifying relationships and patterns among complex data. Artificial intelligence- and machine learning (ML)-driven predictions have shown promising potential in influencing real-time decisions and improving surgical outcomes by facilitating screening, diagnosis, risk assessment, preoperative planning, and shared decision-making. Fundamental understanding of the algorithms, as well as their development and interpretation, is essential for the evolution of AI in surgery. In this article, we provide surgeons with a fundamental understanding of AI-driven predictive models through an overview of common ML and deep learning algorithms, model development, performance metrics and interpretation. This would serve as a basis for understanding ML-based research, while fostering new ideas and innovations for furthering the reach of this emerging discipline.