Can Acromio-Humeral Distance difference predicts rotator cuff lesion?--A study among Unilateral Shoulder Pain in Makassar.
Self-reported prevalence of shoulder problem is estimated up to 66% lifetime; it is the third most common cause of musculoskeletal consultation in primary care.  Approximately 1% of adults consult a general practitioner with new shoulder pain annually.  Stiff and pain shoulder complaints are closely related to frozen shoulder or rotator cuff lesion. 
Rotator cuff injury covers up to 70% of shoulder problems in orthopedic practices. Diagnosis of rotator cuff injury is established from history taking, physical examination and radiologic workup. Magnetic resonance imaging (MRI) is the gold standard for confirming the presence of rotator cuff injury. However, MRI machine availability is very limited in the east part of Indonesia due to its expensive cost.
Rotator cuff lesion can be predicted by X-ray measurements such as proximal humeral migration, acromion shape, and acromiohumeral distance (AHD). [4, 5] Altough MRI and Ultrasonogram is the best way to detect it.  Simple X-ray measurement could be helpful enough in places where MRI is not available, in determining the possibility of rotator cuff problems.
Saupe et al. and Mayerhoever et al. found that narrowing of AHD correlates significantly with rotator cuff tear and concomitant muscle degeneration. [5,7] Other study stated that AHD reflects clinical condition in patients with supraspinatus muscle disorder. However, Gumina et al. found that AHD is more influenced by genetic factor rather than adaptive/ environmental factor.  Therefore, in patients with narrow AHD, does not mean suffer from rotator cuff tear, meanwhile, patients with wide AHD does not necessarily free from rotator cuff injury. We have a same consideration as the Gumina et al. stated before. We assumed that the physiologic AHD is individual fashioned.
The mechanism of AHD narrowing due to the ascension of humeral head is still not precisely known. Proposed probable mechanism is overpull of deltoid muscle due to loss of infraspinatus injury or loss of subacromial volume due to retracted supraspinatus muscle/tendon. 
Based on several considerations mentioned above, authors decided to study the difference of AHD between symptomatic and asymptomatic shoulders, whether it is correlated significantly with rotator cuff tear or not. We evaluated the AHD difference between the symptomatic and the asymptomatic shoulders, rather than solely evaluate single symptomatic shoulder without contralateral comparison. To the extent of our literature review, no previous research ever evaluating AHD difference between symptomatic and asymptomatic shoulder. Moreover, no study ever measuring the AHD of Indonesian population. We hope that this study could be a proper additional reference about AHD research in Indonesia or world-widely.
The objectives of this study are to prove the correlation of AHD difference between symptomatic and asymptomatic shoulders with rotator cuff tear presence seen in MRI among unilateral shoulder pain patients in Makassar; to measure the predictive performance of AHD difference in detecting rotator cuff tear in MRI, and to measure AHD value of Indonesian population.
MATERIALS AND METHODS
Study Design and Site
This is a cross-sectional study. The study conducted in Makassar, from September 2016 to July 2017. We collected data from patient age 45 to 65 years old, with a complaint of unilateral shoulder pain and stiffness.
Inclusion and Exclusion Criteria
The following criteria were included in this study:
1. Male or female age 45-65 years old,
2. Unilateral shoulder pain and deficit in range of motion,
3. Complaint duration at least 3 months.
The following criteria were excluded from the study: Exclusion criteria such as concomitant neurologic disorder on affected shoulder, bilateral shoulder pain, evidence of fracture around shoulder seen from X-ray, recent shoulder overactivity.
Thorax anteroposterior (AP) upright X-ray (symmetric neutral or slight abduction of both arms) was taken from patients with matched inclusion criteria. AHD was measured from both shoulders, and the difference is calculated using radiological software assisted with manual mouse-pointer clicking at the radiology monitor. Next, MRI 1.5T was performed on symptomatic shoulders to confirm the rotator cuff tear. Thorax AP upright X-ray also taken from healthy subjects, and AHD was measured as a significance comparison (control group).
This research got ethical approval from Medical Research Ethical Committee of Hasanuddin University Hospital/ Wahidin Sudirohusodo General Hospital, Makassar.
SPSS 22 was employed for this study. We performed the descriptive test and comparative test (independent sample t-test) between AHD of sample group and control group. Next, we assess the correlation between AHD difference and rotator cuff tear severity seen in MRI using Spearman rank test. The predictive performance of AHD difference was assessed using the area under curve of receiver operating characteristic (ROC) curves of sensitivity over 1-specificity.
This study collected sample from two hospitals in Makassar, September 2016 through July 2017. This study was included total 30 patients in sample group and 30 patients in control group.
Sample and Control Characteristics
Characteristic of sample and control was arranged according to sex, age, symptomatic shoulder, and MRI finding. 14 (46.7%) of samples are male, meanwhile, 16 (53.3%) of samples are females. Mean age of this study is 54.63 ([+ or -] 6) years old for the sample group, with range 45-65 years old; and 55.23 ([+ or -] 6) years old for the control group, with range 46-65 years old. There are more samples with right shoulder pain than left shoulder pain (18 samples (60%) vs. 12 samples (40%)). Nine samples had no tear (30%), 11 samples had partial supraspinatus tear (36.7%), 3 samples had total supraspinatus tear (10%), and 7 samples had multiple rotator cuff tear (23.3%).
AHD was measured and collected from control group representing normal value of AHD. 15 (50%) of control are male, 15 (50%) of control are female. Mean AHD of male is 10.17 [+ or -] 1.32 mm on the right shoulder, 9.98 [+ or -] 1.55 mm on the left shoulder. Mean AHD of female is 9.12 [+ or -] 0.80 mm on the right shoulder, 8.94 [+ or -] 0.62 mm on the left shoulder. Mean AHD of male is significantly wider than female (P = 0.016 for right shoulder and P = 0.027 for left shoulder).
Correlation between AHD Differences with Shoulder MRI Finding on Sample Group
Based on Spearman rank test, there is a positive correlation between AHD difference with shoulder MRI findings [Table 1]. It means that the wider the AHD difference, the worst the rotator cuff tear present in symptomatic shoulder MRI. Correlation coefficient is 0.749, signifies that this two variable are strongly correlated each other (strong correlation coefficient = 0.50-0.89) (P < 0.05 [0.000]).
Comparative Test between AHD Difference Symptomatic-asymptomatic Shoulders in Sample and Control Group
The mean of AHD difference in sample group is 1.42 mm ([+ or -] 0.85 mm), meanwhile, the mean of AHD difference in the control group is 0.25 mm ([+ or -] 0.18 mm). AHD difference between sample and control group is significantly different, P < 0.05 (0.001) [Table 2].
Sensitivity and Specificity (Predictive Performance) of AHD Difference to Determine Rotator Cuff Tear in Shoulder MRI
From the analysis of ROC curve, to predict at least partial tear of supraspinatus on MRI cut off point 1.05 mm of AHD was noted with sensitivity 85.7% and specificity 89.9%. To predict at least total tear supraspinatus (with/without other rotator cuff muscle) on MRI, cut off point 1.75 mm of AHD was noted with sensitivity 80.0% and specificity 90.0%. To predict multiple rotator cuff tear on MRI, cut off point 1.75 mm of AHD was noted with sensitivity 71.4% and specificity 78.3% [Table 3].
Our samples have the same mean age with the control (sample 54 [+ or -] 6 years old vs. control 55 [+ or -] 6 years old). This finding is similar with the descriptive study conducted by Petersson et al., in which the mean age of a patient with supraspinatus tear is 55 years old. 
There is more right shoulder pain than the left shoulder pain, with ratio 3:2. This could be caused by the hand dominance (right hand dominant is more prevalent than the left hand dominant). Based on the shoulder MRI findings, partial supraspinatus tear is the most prevalent (11 subjects), followed by no rotator cuff tear (9 subjects), multiple rotator cuff tear (7 subjects), and total supraspinatus tear (3 subjects). If categorized by "tear" versus "no tear," then the samples divided into 9 subjects without rotator cuff tear versus 21 subjects with rotator cuff tear.
In Makassar population, mean AHD on normal male is 10.17 [+ or -] 1.32 mm on the right shoulder and 9.98 [+ or -] 1.55 mm on the left shoulder. Meanwhile, on a normal female is 9.1 [+ or -] 0.8 mm on the right shoulder and 8.9 [+ or -] 0.62 mm on the left shoulder. Statistically, the difference is significant (P = 0.016 for the right shoulder, and P = 0.027 for the left shoulder). Our findings are closely similar with the previous researcher. Petersson et al., found that the mean AHD for a male is 10.2 mm (range 6.6-13.8 mm) and 9.5 mm (range 7.1-11.9 mm) for female. Possible explanation for this difference is because the size of tendon and muscle in male is relatively larger than female, and the workload of the right shoulder may be higher than the left shoulder. Hence, the tendon and rotator cuff muscle may be more hypertrophic, in turns, will produce wider AHD.
Our hypothesis is the AHD difference on X-ray examination between the symptomatic and asymptomatic shoulder correlates positively with the degree of rotator cuff tear detected on the MRI. To the extent of our literature review, no previous research ever evaluating AHD difference between symptomatic and asymptomatic shoulder. Spearman rank test proves there is a positive correlation between the AHD difference and the MRI findings, means that the wider the AHD difference, the worst the degree of rotator cuff tear found at symptomatic shoulder MRI. The correlation coefficient is 0.79 signifies that these two variables are correlated each other strongly (range for the strong correlation is 0.70-0.89) and statistically significant, P < 0.05 (0.000). Our hypothesis is proven in this study.
Keener et al. studied that the proximal migration of humeral head correlated significantly with the number of rotator cuff tear (1.01 [+ or -] 1.5 mm on supraspinatus and infraspinatus tear v. -0.09 [+ or -] 1.5 mm on isolated supraspinatus tear) among the shoulder pain with visual analog scale more than 5/10.  In our opinion, proximal migration of humeral head is analog with the AHD difference in patients with unilateral shoulder pain.
Saupe et al. describe that narrowing of AHD is reliable for predicting rotator cuff tear.  Research on 1970 by Weiner and Macnab found that mean for normal AHD is 10.5 mm, meanwhile, at the rotator cuff tear condition is only 8.2 mm.  Other study said AHD <7 mm as a sign of total rotator cuff tear.  However, Petersson et al. using a cut off <6 mm to determine a supraspinatus tear.  Gouttalier et al. stated that AHD <6 mm is definitive for infraspinatus tear. 
Based on ROC curve analysis (ROC), we found that AHD difference 1.05 mm, this variable has sensitivity 85.7% and specificity 89.9% to predict any type of rotator cuff tear on symptomatic shoulder MRI. Statistically, this finding is useful, but, in our opinion, this finding is less applicable since comparing a 1 mm distance accurately using bare eyes and simple ruler is very difficult.
AHD difference has sensitivity 80.0% and specificity 90.0% to predict a total tear of supraspinatus or multiple rotator cuff tear on symptomatic shoulder MRI at cut off point 1.75 mm. AHD difference also has sensitivity 71.4% and specificity 78.3% to predict multiple rotator cuff tear on symptomatic shoulder MRI at cut off point 1.75 mm.
Limitation and Auto critics
We noticed several flaws in our research. First, the measurement of AHD is using the software at the radiology monitor which is operator dependent, in which the point of measurement is manually clicked using mouse pointer. Second, we have only 30 samples; it is needed a larger study with more samples to get a more accurate result and conclusion. Third, we did not have any samples with an isolated tear on teres minor, infraspinatus, or subscapularis. Fourth, we did not exclude the osteoarthritic shoulder joint which may cause bias in evaluating rotator cuff tear.
AHD difference has a positive correlation with the severity of rotator cuff tear found at shoulder MRI. AHD difference can be considered as a diagnostic aid for rotator cuff tear in 45-65 years old patient with non-traumatic unilateral shoulder pain and stiffness, in places which MRI is not available, especially if the AHD difference is more than 1.75 mm. AHD in males is relatively wider than the females on Makassar population. Larger samples are needed to evaluate the usefulness of AHD difference further. Another study can assess and compare AHD difference with other proven parameters of shoulder such as single symptomatic AHD, or proximal humeral head migration.
[1.] Luime JJ, Koes BW, Hendriksen IJ, Burdorf A, Verhagen AP, Miedema HS, et al. Prevalence and incidence of shoulder pain in the general population; a systematic review. Scand J Rheumatol 2004;33:73-81.
[2.] Linsell L, Dawson J, Zondervan K, Rose P, Randall T, Fitzpatrick R, et al. Prevalence and incidence of adults consulting for shoulder conditions in UK primary care; patterns of diagnosis and referral. Rheumatology (Oxford) 2006;45:215-21.
[3.] Murphy R, Carr A. Shoulder Pain. Am Fam Physician 2011;83:137-8.
[4.] Keener JD, Wei AS, Kim M, Steger-May K, Yamaguchi K. Proximal humeral migration in shoulders with symptomatic and asymptomatic rotator cuff tears. J Bone Joint Surg Am 2009;91:1405-13.
[5.] Saupe N, Pfirrmann CW, Schmid MR, Jost B, Werner CM, Zanetti M, et al. Association between rotator cuff abnormalities and reduced Acromiohumeral distance. AJR Am J Roentgenol 2006;187:376-82.
[6.] Franca FO, Godinho AC, Ribiero EJ, Falster L, Burigo LE, Nunes RB. Evaluation of acromiohumeral distance by means of MRI humerus. Rev Bras Ortop 2016;51:169-74.
[7.] Mayerhoefer ME, Breitenseher MJ, Wurnig C, Roposch A. Shoulder impingement: Relationship of clinical symptoms and imaging criteria. Clin J Sport Med 2009;19:83-9.
[8.] Gumina S, Arceri V, Fagnani C, Venditto T, Catalano C, Candela V, et al. Subacromial space width: Does overuse or genetics play a greater role in determining it? An MRI study on elderly twins. J Bone Joint Surg Am 2015;97:1647-52.
[9.] Petersson CJ, Redlund-Johnell I. The subacromial space in normal shoulder radiographs. Acta Orthop Scand 1984;55:57-8.
[10.] Weiner DS, Macnab I. Superior migration of humeral head. A radiological aid in the diagnosis of tears of rotator cuff. J Bone Joint Surg Br 1970;52:524-7.
[11.] Gouttalier D, Le Guilloux P, Postel JM, Radier C, Bernageu J, Zilber S. Acromio-humeral distance less than six millimeters; Its meaning in full-thickness rotator cuff tear. Orthop Traumatol Surg Res 2011;97:246-51.
How to cite this article: Then Z, Sakti M, Saleh MR. Can Acromio-Humeral Distance difference predicts rotator cuff lesion?--A study among Unilateral Shoulder Pain in Makassar. Int J Med Sci Public Health 2018;7(4):264-268.
Source of Support: Nil, Conflict of Interest: None declared.
Zuwanda Then, Muhammad Sakti, M. Ruksal Saleh
Department of Orthopaedic and Traumatology, Faculty of Medicine, Hasanuddin University, Makassar, Indonesia
Correspondence to: Zuwanda Then, E-mail: firstname.lastname@example.org
Received: November 30, 2017; Accepted: January 26, 2018
Access this article online
Table 1: Spearman's correlation test between AHD difference and shoulder MRI findings Variable AHD difference Shoulder MRI findings Spearman's rank test AHD difference Coefficient 1.000 0.749 ** P (1-tailed) 0.0 0.000 n 30 30 Shoulder MRI findings Coefficient 0.749 ** 1.000 P (1-tailed) 0.000 0.0 n 30 30 ** Correlation is significant on P<0.01 (1-tailed). AHD: Acromiohumeral distance, MRI: Magnetic resonance imaging Table 2: Comparison between sample and control group Variable Group n Mean SD P (mm) (mm) AHD Sample 30 1.42 0.85 0.001 difference Control 30 0.82 0.38 SD: Standard deviation Table 3: Sensitivity and specificity of AHD difference to predict several types of rotator cuff tear Prediction AHD cut Sensitivity Specificity off (mm) (%) (%) To predict at least partial 1.05 85.7 89.9 tear supraspinatus To predict at least total 1.75 80.0 90.0 tear supraspinatus To predict multiple 1.75 71.4 78.3 rotator cuff tear AHD: Acromiohumeral distance
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Article|
|Author:||Then, Zuwanda; Sakti, Muhammad; Saleh, M. Ruksal|
|Publication:||International Journal of Medical Science and Public Health|
|Date:||Apr 1, 2018|
|Previous Article:||Effect of six-minute walk test in obesity.|
|Next Article:||Effectiveness of breathing exercises as play method on cardiopulmonary parameters among children with acute respiratory tract infections in a...|