Introduction

Traditional risk scores, like the Charlson Comorbidity Index and the Frailty Index, have had limited accuracy and applicability in predicting outcomes in hand surgery [1-3] These risk assessment tools are necessary to address modifiable risk factors or to have more informed discussion of realistic outcomes.

Objectives

To develop an objective and granular risk score for readmissions in hand surgery To validate this score nationally, and demonstrate accuracy and usability

Patient 2: office consult for elective case. On exam: male, elderly, history of COPD, end stage renal disease (on dialysis), congestive heart failure,

A Nationally Validated Risk Assessment Score For Predicting Unplanned Readmissions In Hand Surgery

Olachi Oleru, MD 1; Nargiz Seyidova, MD, MHQS 1; Peter J. Taub, MD, MS, FACS 1 Icahn School of Medicine at Mount SInai, Division of Plastic and Reconstructive Surgery

Acknowledgements

Dr. Oleru was supported the by the National Center for Advancing Translational Sciences (NCATS) TL1TR004420 NRSA TL1 Training Core in Transdisciplinary Clinical and Translational Science (CTSA) grant.

Many thanks to the Icahn School of Medicine at Mount Sinai Department of Surgery and Division of Plastic and Reconstructive Surgery

Results

Related literature Lipira AB, Sood RF, Tatman PD, Davis JI, Morrison SD, Ko JH. Complications Within 30 Days of Hand Surgery: An Analysis of 10,646 Patients. J Hand Surg. 2015;40(9):1852-1859.e3. doi:10.1016/j.jhsa.2015.06.103 Usami S, Inami K, Kawahara S, Hirase Y, Shimizu H, Mori H. Investigation of Predictors of Successful Replantation of Distal Digits at the Nail Bed Level: The Contribution of Digital Nerve Repair to Survival Rate. Plast Reconstr Surg. 2022;149(4):889-896 Hall DE, Arya S, Schmid KK, et al. Association of a Frailty Screening Initiative With Postoperative Survival at 30, 180, and 365 Days. JAMA Surg. 2017;152(3):233-240. doi:10.1001/jamasurg.2016.4219

Conclusions

We present a nationally validated risk stratification tool for unplanned readmissions following hand surgery. This risk score outperforms the commonly used modified frailty and Charlson comorbidity indices.

Database

Modeling

ACS National Surgical Quality Improvement Program

Hand/UE specific CPT codes

2012 - 2019 Training

2020 Testing

> 1 million

>100,000

>87,000

15,000

Multivariate regression with backward elimination

Unexponentiated regression coefficient (x10, to nearest integer)

Calculator

Total score for each patient

outcome of interest: 30 day readmissions

applied to demographics, comorbidities, perioperative factors

Future studies should clinically validate this model and determine if implementation of this risk score optimizes safety and reduces readmission rates in hand surgery patients.

The area under the curve (AUC) represents the discriminatory ability of the model. Generally, >0.7 represents an accurate test. Our model demonstrated higher accuracy than the mFI-5 and mCCI, and rendered an AUC >0.7 in both the training and testing datasets.

The integer score ranged from 0-90 when tabulated.

Examples

Patient 1: ED consult for trauma patient. On exam: female, history of hypertension, diabetes,, and has hgb < 7 (requiring preoperative transfusion)

Patient 2: Readmission score: 23 Highest risk category No easily modifiable risk factors discussion with patient about reasonable expectations

Patient 1: Readmission score: 17 Highest risk category Wait until anemia is treated and stabilized New readmission score: 7

Methodology

Patients were stratified by total score to fit a normal distribution. The strata correlate to risk categories: 1 - lowest, 2 - low, 3 - intermediate, 4 - high, 5 - highest risk

sum of all weights

risk percent correlating to total score