Preoperative predictors of patient-reported outcomes after joint-preserving forefoot surgery for rheumatoid arthritis

Aims

Forefoot deformities are common in patients with rheumatoid arthritis (RA) and often impair quality of life. Patient-reported outcome measures (PROMs), including the Self-Administered Foot Evaluation Questionnaire (SAFE-Q), are increasingly used to assess surgical outcomes; however, the preoperative conditions needed to achieve favourable postoperative PROMs remain unclear. This study aimed to identify factors associated with postoperative SAFE-Q scores (mean score across five sub-scales, maximum 100) and determined preoperative cutoff values predictive of achieving predefined postoperative SAFE-Q thresholds (≥ 70 or ≥ 90).

Methods

In this multicentre, retrospective case series, 101 feet in 80 patients (including 21 bilateral cases) with RA who underwent forefoot surgery between July 2015 and July 2023 were analyzed (mean age at surgery 65.8 years (SD 10.7); 79 female patients). Inclusion criteria were symptomatic forefoot deformities with available SAFE-Q data pre-surgery and at ≥ one year post-surgery. Procedures included modified scarf osteotomy and offset shortening osteotomies for the metatarsals. SAFE-Q scores, Japanese Society for Surgery of the Foot scores, radiological parameters, and disease activity were evaluated pre- and postoperatively. Multiple regression identified preoperative factors associated with postoperative SAFE-Q scores, and significant variables were further assessed using receiver operating characteristic (ROC) analyses to determine preoperative cutoff values for postoperative SAFE-Q thresholds (≥ 70 and ≥ 90).

Results

Postoperatively, the median SAFE-Q score was 84.8 (IQR 73.0 to 92.3), a significant improvement over the preoperative median score of 48.9 (IQR 34.9 to 67.5; p < 0.001). Multivariable analysis showed that older age and lower preoperative SAFE-Q scores were associated with lower postoperative SAFE-Q scores (β -0.286 (95% CI -0.486 to -0.086) and 0.505 (95% CI 0.319 to 0.691), respectively). ROC analysis identified cutoffs of 54.0 for preoperative SAFE-Q (area under the curve (AUC) 0.644) and 70 years for age (AUC 0.722) when predicting a postoperative score ≥ 90.

Conclusion

Higher preoperative SAFE-Q scores and younger age at surgery were linked to improved postoperative outcomes. Although their predictive accuracy is limited, these thresholds may help estimate postoperative function and support preoperative counselling and shared decision-making.

Cite this article: Bone Joint J  2026;108-B(8):1012–1020.