Does hand-behind-back range of motion accurately reflect shoulder internal rotation? Karen A. Ginn, PhD, a Milton L. Cohen, MD, b and Robert D. Herbert, PhD, a Sydney, Australia Shoulder internal rotation (IR) is commonly assessed by an indirect method where the hand is placed behind the back and the distance reached by the tip of the extended thumb is recorded. The aim of this study was to assess the validity of measuring active IR range of motion (ROM) by use of the indirect hand-behind-back (HBB) ROM method in subjects with shoulder pain of mechanical origin. We recruited 137 subjects with unilateral shoulder pain. HBB ROM was determined by measuring the distance between the T1 spinous process and the radial styloid process. Active shoulder IR was measured in the supine position in 45° or 90° abduction. Correlation coefficients adjusted for mea- surement error were calculated. HBB ROM demon- strated only a low to moderate correlation with active shoulder IR. Active HBB ROM is not an accurate method of measuring active shoulder IR in patients with shoulder pain. (J Shoulder Elbow Surg 2006;15: 311-314.) T he measurement of internal rotation (IR) range of motion (ROM) is an essential component of the clini- cal examination of the shoulder joint. 13,17 IR is com- monly assessed by an indirect method called poste- rior reach or hand behind back (HBB), where the hand is actively placed behind the back and the vertebral level reached by the tip of the extended thumb is recorded. 3,8,10,12 This indirect method of measuring shoulder IR is part of the standardized shoulder assessment form adopted by the American Shoulder and Elbow Surgeons. 13 A recent study pub- lished in the Journal of Shoulder and Elbow Surgery evaluated the reliability of shoulder IR measured by the vertebral level reached by the fully extended thumb. 3 In addition to shoulder IR, HBB incorporates shoul- der extension, retraction, and downward rotation of the scapula and flexion of the elbow and will be influenced by mobility at the forearm, wrist, and thumb joints. 10,12 In a study using various imaging procedures on normal subjects, Mallon et al 12 con- cluded that interpreting HBB ROM as a measure of shoulder IR greatly underestimated the contribution from other joints. In this study, 35% of the motion in HBB occurred at the scapulothoracic articulation, with a significant contribution from elbow flexion in inner- range HBB. These results suggest that the interpreta- tion of restricted HBB ROM as an accurate measure of shoulder IR is not valid. However, because this study was conducted on subjects without shoulder dysfunc- tion, the question of whether restriction in HBB ROM reflects decreased shoulder IR in a symptomatic pop- ulation remains unanswered. The aim of this study was to assess the validity of measuring active IR ROM by use of the indirect HBB ROM method in subjects with shoulder pain. MATERIALS AND METHODS This study was approved by the Ethics Committees of the University of Sydney and St Vincent’s Hospital, Sydney, Australia. The sample consisted of 137 subjects who volun- teered to participate in a clinical trial investigating the effectiveness of, and prognostic indicators for, conservative treatment for shoulder pain of mechanical origin. 5,6 All subjects gave written consent before participation. They had pain over the shoulder joint or the proximal arm (or both), which was exacerbated by active shoulder move- ments. Subjects with pain associated with shoulder instabil- ity, inflammatory or neoplastic disorder, referred from ver- tebral column structures or as a result of recent trauma were not included. Of the subjects, 77 had decreased abduction or flexion (or both) accompanying their shoulder pain and 60 had a painful arc of abduction or flexion (or both) but retained full ROM. HBB ROM was determined by asking subjects to reach up the center of their spine with their thumb as far as possible (Figure 1). The distance between the T1 spinous process and the radial styloid process was then measured to the nearest 0.5 cm with a tape measure. The tape measure was used to eliminate the error demonstrated by Edwards et al, 3 which was associated with determining the vertebral level reached. Measuring the distance to the ra- dial styloid process was chosen to eliminate error associ- ated with any coexisting lack of flexibility at the wrist or From the a Faculty of Health Sciences, University of Sydney, and b Faculty of Medicine, University of New South Wales. Reprint requests: Karen A. Ginn, PhD, Faculty of Health Sciences, University of Sydney, PO Box 170, Lidcombe, New South Wales 1825, Australia (E-mail: K.Ginn@fhs.usyd.edu.au). Copyright © 2006 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2006/$32.00 doi:10.1016/j.jse.2005.08.005 311