Research Review By Kevin Neeld©

Date Posted:

December 2009

Study Title:

Influence of Humeral Torsion on Interpretation of Posterior Shoulder Tightness Measures in Overhead Athletes


Myers JB et al.

Author's Affiliations:

Sports Medicine Laboratory and Neuromuscular Research Laboratory, Department of Exercise and Sport Science, University of North Carolina at Chapel Hill, Chapel Hill, NC; Campus Health Services, Division of Sports medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC

Publication Information:

Clinical Journal of Sports Medicine 2009; 19(5): 366-371.

Background Information:

In overhead/throwing athletes, glenohumeral range of motion (ROM) measurements are frequently used to predict or establish injury risk. Limitations in internal rotation ROM of the throwing arm, compared to the non-throwing arm, are thought to be indicative of posterior shoulder tightness. When this side-to-side difference is observed, stretching of the posterior shoulder structures of the throwing arm is often prescribed (in the form of “sleeper” or “cross-body” stretches).

Glenohumeral ROM is determined by both bony geometry (see Related Reviews below) and the surrounding soft-tissue structures. Consequently, there may be a fundamental flaw in the conclusion that internal rotation ROM deficits are always the result of soft-tissue restrictions. One way to account for this is to quantify the amount of torsion of the humerus. Greater torsion could allow for a shift in the rotation ROM of the throwing arm toward external rotation (e.g. more external rotation ROM and less internal rotation ROM), as is commonly measured in overhead throwing athletes.

The relationship between humeral torsion and humeral ROM raises the possibility that the observed ROM differences between the throwing and non-throwing arms of overhead athletes can be fully accounted for my differences in torsion. The purpose of the current study was to assess the relationship between humeral torsion and various measures of glenohumeral range of motion.

Pertinent Results:

  • The throwing arm of baseball players had significantly greater humeral torsion, less internal rotation, and less total rotation ROM compared with their non-throwing arm and control participants.
  • In both the baseball player and control groups, the dominant limb had more external rotation ROM. Baseball players demonstrated more external rotation ROM than controls.
  • Interestingly, once rotational measurements were corrected for humeral torsion*, NO group (baseball players vs. controls) or interlimb differences (difference between throwing and non-throwing arm) in internal rotation were noted (p=0.508). After correction, baseball players demonstrated greater external rotation (p=0.002) in their non-throwing arm compared with their throwing arm and compared to controls.
  • Baseball players demonstrated a significantly greater interlimb difference in internal rotation (p<0.001), total rotation ROM (p=0.002), and horizontal adduction (p=0.002) than controls. However, after correcting for humeral torsion, no differences in interlimb internal rotation ROM were observed (p=0.508), but baseball players had a significantly greater interlimb difference in external rotation (p=0.002) compared to controls.
  • In the combined data of both groups, there was a statistically significant relationship between interlimb difference in humeral torsion and interlimb difference in internal rotation (r=-0.66, p<0.001), horizontal adduction (r=-0.29, p=0.017), and total rotation ROM (r=-0.54, p<0.001).
  • In the combined data of both groups, and after adjustment for torsion, the difference in total rotation ROM between the throwing arm and non-throwing arm was correlated with the difference in internal rotation (r=0.71, p<0.001), meaning less total ROM was associated with a loss of internal rotation ROM.
*The authors described recalculating rotation data after moving the original neutral position (position from which internal and external rotation ROM was initiated) to a corrected neutral, defined as “the anatomic neutral position where the line connecting the apexes of the lesser and greater tubercles is parallel to the horizontal plane.”

Clinical Application & Conclusions:

This study provides quality evidence that differences in glenohumeral internal rotation ROM between throwing and non-throwing arms in overhead athletes may primarily result from differences in humeral torsion (bony geometry), rather than soft-tissue restrictions. Clinically, it would be ideal to be able to assess glenohumeral internal rotation, external rotation, total rotation, horizontal adduction, and humeral torsion. Unfortunately, the widespread use of ultrasonography to assess humeral torsion in healthy populations is far from realistic. However, assessing differences in total rotation ROM between throwing and non-throwing arms may provide similar information.

If differences in internal/external rotation were due strictly to humeral torsion, one could expect the total rotation range of motion to be the same. The arm with more humeral torsion (the throwing arm) would simply demonstrate more external rotation ROM and a decrease in internal rotation ROM of similar magnitude. If soft-tissue restrictions also contributed to ROM deficiencies, it would likely manifest as a decrease in total rotation ROM.

The major take home message from this study is that prescribing stretches for the posterior shoulder based solely on interlimb discrepancies in internal rotation ROM may not always be appropriate. It is necessary to assess total rotation ROM, and if possible humeral torsion, to rule out the possibility that the observed differences are strictly due to differences in humeral torsion.

Study Methods:

Twenty-nine Division I collegiate baseball players (average age: 19.5; height: 183.5 cm; mass: 87.5 kg) and 25 college-aged controls (average age: 20.0; height: 182.3 cm; mass: 81.8 kg) participated in the current investigation. In a single testing session, bilateral humeral internal rotation, external rotation, and horizontal adduction were quantified using a digital inclinometer. All measurements were taken with the shoulder abducted and the elbow flexed 90°. Humeral torsion was quantified using an indirect ultrasonographic technique.

Three trials were taken for each measurement and the average of the three was used for analysis. These values were used to calculate total rotation ROM, inter-limb differences, and torsion-corrected rotation ROM. Relationships between variables were also calculated.

Study Strengths/Weaknesses:

This study was well conducted and without major limitations. It should be remembered when discussing the results of this study that the subjects here were free of injury. Since many of these glenohumeral tests are utilized to minimize injury risk, it may be insightful to see if the same relationships between humeral torsion and glenohumeral ROM are observed in injured overhead throwing populations.

Additional References:

  1. Burkhart S et al. The disabled throwing shoulder: spectrum of pathology. Part I; Pathoanatomy and biomechanics. Arthroscopy 2003; 19: 404-420.
  2. Tyler TF et al. Quantification of posterior capsule tightness and motion loss in patients with shoulder impingement. Sports Med 2000; 28: 668-673.
  3. Myers JB et al. Reliability, precision, accuracy, and validity of posterior shoulder tightness assessment in overhead athletes. Am J Sports Med 2007; 35: 1922-1930.
  4. Kronberg M et al. Retroversion of the humeral head in the normal shoulder and its relationship to the normal range of motion. Clin Orthop Relat Res 1990: 253-113-117.

Related Reviews on RRS:

Please read additional reviews on this topic in Rehabilitaion - Upper Extremities Section to fully understand this topic.