gms | German Medical Science

Objectives: Mid- and long-term results following angular stable plate fixation of a proximal humeral fracture are limited. Assessment of patient-reported outcome measures (PROM) is difficult in elderly patients because of immobility and dependency of transportation. Therefore, telephone-based assessments of PROM are an alternative in these patients. However, no data is available regarding the test-retest agreement of telephone-based assessment of PROM in humeral head fractures. The aim of the study was to assess mid-term outcomes in patients (>70 years) with proximal humeral fractures treated with angular stable plate fixation and compare the test-retest agreement of the telephone-based score assessment.

Methods: This cohort study analysed PROMS of 90 patients (female: 78; age: 78.1±5.2 range: 70.1-89.8 years) with a proximal humeral fracture (2-part: 34, 3-part: 41, 4-part: 12) treated with angular stable plate fixation at least three years post-operatively (3.7 ± 0.9 years). Two telephone-based interviews assessed the DASH (Disabilities of the Arm, Shoulder and Hand Score), OSS (Oxford Shoulder Score) and CS (Constant Score) and were conducted within three months. Correlations between both interviews were calculated using Spearman's rank correlation coefficient. Cross tabulation was applied and weighted Kappa measure of agreement (weighted k) was calculated to assess the test-retest agreement between the categories of each score. Bland-Altman analyses were calculated to describe absolute individual differences between interviews.

Results: In the first and second interview, CS was median 91 (range: 40 to 100) and 65.5 (23-86), DASH was 12.5 (0-64.2) and 18.3 (0-66.7) and OSS was 58 (33-60) and 55 (25-60) points.

The correlation coefficient between test and retest were r=0.67, r=0.77 and r=0,71 for CS, DASH and OS (all p<0.0001). Bland-Altman analyses showed limited absolute individual score agreement. The mean difference between both interviews was -22.3 (-25 to -20), 4.9 (3 to 7) and -3.0 (-4 to -2) for CS, DASH and OSS. The limits of agreement represented possible differences of 21.6%, 15.5% and 9.0% of the respective scoring scale of CS, DASH and OSS. The agreements of categories between interviews were medium to high: 55.9% (k=0.08), 87.2% (k=0.62) and 99.3% (k=0.74) for CS, DASH and OSS (all p<0.001).

Conclusion: Most patients showed good subjective outcomes at least three years following surgical treatment. However, the test-re-test agreement of the CS was low. Only small differences and high agreement between score categories were found for DASH and OS. Despite higher test-retest agreement, both scores do not analyze the functionality and mobility of the shoulder as detailed as the Constant score. Telephone-based assessment of OSS and DASH present as an alternative to collect and monitor surgical outcomes in elderly patients with low costs, easier assessment and lower lost-to-follow-up rates.