Summary

The purpose of this study was to describe preoperative and postoperative PROMIS scores in TSA patients, compare PROMIS physical function scores with clinical functional measurements, and determine if preoperative PROMIS scores could predict achievement of Minimally Clinically Important Difference (MCID) postoperatively.

Abstract

Background

PROMIS (Patient Reported Outcomes Measure Information System) is a high throughput PROM (patient reported outcomes measure) recently validated in orthopedic patients with upper extremity disease. PROMIS scores in patients who underwent total shoulder arthroplasty (TSA) have not previously been described. Furthermore, the ability of preoperative PROMIS scores to predict postoperative outcomes in TSA patients remains unknown. The purpose of this study was to describe preoperative and postoperative PROMIS scores in TSA patients, compare PROMIS physical function scores with clinical functional measurements, and determine if preoperative PROMIS scores could predict achievement of Minimally Clinically Important Difference (MCID) postoperatively.

Methods

A retrospective analysis of PROMIS scores in patients who underwent TSA from February 2015 to February 2017 at our institution was performed. Patients were included if they underwent primary anatomic TSA and had PROMIS scores from both within 60 days prior to surgery (preoperative) and greater than 3 months after surgery (postoperative). Patients were excluded if they were a revision shoulder arthroplasty or underwent reverse TSA. Shoulder range of motion data (forward flexion (FF), external rotation, (ER)) was also collected at both time points. PROMIS physical function (PF), pain interference (PI) and depression (D) scores were compared. MCID values for each domain were calculated. Accuracy analyses were performed to determine whether preoperative PROMIS scores from each domain could predict postoperative achievement of MCID in the same domain. A multivariate predictive model incorporating age, gender, BMI and ASA class was also performed. Preoperative 90% cutoff scores describing the probability of achieving or failing to achieve MCID were calculated.

Results

62 patients were included in the study. Preoperative (mean 35 days prior to surgery) mean PROMIS PF, PI and D scores were 40.4 (standard deviation 8.0), 61.2 (6.3) and 49.7 (8.6), respectively. Postoperative (mean 285 days after surgery) mean PROMIS PF, PI and D scores were 44.1 (7.0), 52.6 (8.6) and 45.5 (8.8), respectively. There was a significant increase in PF (p=0.001), and significant decreases in PI (p<0.001) and D (p<0.001). Patients showed significant improvements in FF (p<0.001), with a mean improvement of 46.6º, and ER (p<0.001), with a mean improvement of 28.5º. The multivariate model demonstrated that preoperative PF, PI and D scores all showed a strong ability to predict postoperative achievement of MCID, with AUCs (area under the curve) of 0.700, 0.867 and 0.712, respectively. Preoperative cutoff scores yielding a 90% chance of achieving MCID (“90% MCID”) were <=31.7 for PF, >=66.9 for PI, and >=55.5 for D. Scores yielding a 90% chance of failing to achieve MCID (“90% No MCID”) were >=47.8 for PF, <=55.1 for PI, and <=41.0 for D.

Conclusions

TSA patients showed significant improvements in all PROMIS domains after surgery. Preoperative PROMIS PF, PI and D scores were highly predictive of postoperative achievement of MCID. PROMIS PF scores were responsive to shoulder functional improvements measured clinically. The 90% MCID values can help surgeons predict which patients are most and least likely to achieve MCID after TSA. These findings provide a powerful tool for surgeons to better counsel their patients on the expected outcomes after TSA surgery, furthering the patient-doctor relationship and maximizing patient satisfaction.