Reliability and validity of the elevated arm stress test in the diagnosis of neurogenic thoracic outlet syndrome

Objectives: The objective of this retrospective analysis of prospectively collected data was to assess the test-retest reliability and validity of the elevated arm stress test (EAST) as measured by the duration in a cohort of patients with suspected neurogenic thoracic outlet syndrome (NTOS). Methods: Patients evaluated for NTOS between January 2017 and September 2018 were identified. Test-retest reliability by the intraclass correlation coefficient was determined for duration of the EAST. For the validity analysis, patients were classified in a proven NTOS group or a symptomatic control group without NTOS using the Society for Vascular Surgery reporting standards and the outcome of thoracic outlet decompression surgery. A receiver operating characteristic curve was made for the duration of EAST. The area under the curve, and positive and negative predictive values were calculated for the EAST. Results: In total, 428 patients with suspected NTOS were retrospectively analyzed. Of these patients, 61 were excluded because no EAST data was available. Another 101 patients were excluded because of inconclusive reporting standards, arterial or venous TOS, or because thoracic outlet decompression surgery was not performed or had a negative result. The validity analysis in the remaining 266 patients showed an area under the curve for the duration of the EAST of 0.62 (95% confidence interval, 0.55-0.69). The positive predictive value of the duration ranged between 65% and 66%, and the negative predictive value between 53% and 58%. For the test-retest reliability analysis, 118 patients were excluded because they performed only one measurement in a 100-day time period. Analysis in the remaining 148 patients showed an intraclass correlation coefficient value of 0.65 (95% confidence interval, 0.55-0.74) for duration. Conclusions: The EAST measured by the duration showed a moderate test-retest reliability, but the discriminative value was low in the diagnosis of NTOS. The outcome of the EAST measured by the duration should be used with caution.