Correlation between mobility restrictions and radiologic changes in ankylosing spondylitis

Abstract

Study design: In 151 adult patients with ankylosing spondylitis who participated in an inpatient rehabilitation program of 3-4 weeks, 10 different range of motion (ROM) values were measured, and their lumbar spine and sacroiliac joints were radiographed.

Objectives: To determine whether a correlation obtained between restriction of ROM and progression of radiologic findings in ankylosing spondylitis, together with a high enough reliability level in measurements.

Summary of background data: A significant correlation was observed between the restriction of eight ROMs: the Schober test, thoracolumbar rotation, thoracolumbar flexion, cervical rotation, occiput-wall distance, chin-chest distance, chest expansion, finger-floor distance, and overall radiologic changes in both lumbar spine and sacroiliac joints; straight leg raise did not correlate, and vital capacity only with sacroiliac joint changes. Spearman's correlation coefficients for ROMs were slightly higher to lumbar spine changes than to sacroiliac joint changes. A corresponding correlation was also observed between spinal mobility restrictions and six detailed changes in lumbar roentgenogram: syndesmophytes, apophyseal arthritis, sclerotic anterior borders of vertebrae, straightened anterior surface of vertebrae, and ossification of interspinous and anterior longitudinal ligaments. Other detailed lumbar spine findings did not correlate. As assessed by erythrocyte sedimentation rate values the disease activity increased in the course of radiologic progression, decreasing again, however, to the end stage.

Methods: Conventional methods with a tape and (Myrin) inclinometer were used to measure thoracolumbar flexion, cervical rotation, occiput-wall distance, chin-chest distance, finger-floor distance, chest expansion, vital capacity, and straight leg raise. In addition, a new method of thoracolumbar rotation and a new modification of the Schober test were introduced. Thirty-nine patients were randomized for a reliability assessment using repeated measurements of ROMs. Radiologic changes were evaluated (in a masked fashion) using the method of Dale and Vinje.

Results: The reliability of all ROMs was good (except for interrater intraclass correlation coefficients of chest expansion: 0.53).

Conclusions: The clear correlation between radiologic sacroiliac joint and lumbar spine progression and eight ROMs showed that these are useful noninvasive measurements of disease progression and severity in ankylosing spondylitis that can be used in daily practice.