- Email: [email protected]
Differences in Range of Motion Between Dominant and Nondominant Sides of Upper and Lower Extremities
DIFFERENCES IN RANGE OF MOTION BETWEEN DOMINANT AND NONDOMINANT SIDES OF UPPER AND LOWER EXTREMITIES Luciana Gazzi Macedo, BPT, MSc, a and David J. Magee, BPT, PhD b,c
ABSTRACT Objective: The objective of this study was to compare ranges of motion (ROM) between dominant and nondominant sides for the joints of the upper and lower extremities. Methods: Ninety healthy white women from 18 to 59 years of age were measured in this study. Active and passive ROM were measured for the ankle, knee, hip, shoulder, elbow, and wrist using a standard goniometer. The order of the joints, motion, sides, and active or passive motion testing was randomly selected. A paired t test was used for the comparison between sides. Results: The results of this study showed a statistically significant difference between dominant and nondominant sides for 34 of the 60 ROM measured. The maximum mean difference between sides for all ROM measured was 7.5°. Conclusion: The results of this show that some ROM are different between body sides and that when these differences exist they are minimal and may not be clinically insignificant. These results support the practice of using the opposite side of the body as an indicator of preinjury or normal extremity ROM. (J Manipulative Physiol Ther 2008;31:577-582) Key Indexing Terms: Range of Motion, Articular; Female
he use of range of motion (ROM) measurements in the health care of musculoskeletal disorders is a common procedure when making diagnoses, setting treatment goals, and measuring treatment progress. Since 1965, when the first standardized manual for recording and measuring joint motion was published, it has been recognized that there is a variation in ROM between individuals.1 This variation creates a problem when trying to establish what preinjury or normal ROM would be for an injured joint. A potentially simple way to estimate normal ROM that accounts for individual variation is to presume that the patient's opposite (uninjured) side has the same ROM as the injured side before the injury. To date, the validity of this assumption has not been adequately tested.
T
a
Doctural Student, The George Institute for International Health, University of Sydney, Sydney, NSW, Australia. b Professor, University of Alberta, Edmonton, Alberta, Canada. c Associate Dean, Faculty of Rehabilitation Medicine, University of Alberta, Edmonton, Alberta, Canada. Submit requests for reprints to: Luciana Gazzi Macedo, BPT, MSc, The George Institute for International Health, University of Sydney, Level 7, 341 George St, Sydney, NSW 2000, Australia (e-mails: [email protected] [email protected]). Paper submitted August 1, 2007; in revised form May 16, 2008; accepted June 15, 2008. 0161-4754/$34.00 Copyright © 2008 by National University of Health Sciences. doi:10.1016/j.jmpt.2008.09.003
In contrast to the proposition that the opposite side is the best estimation of one's ROM of motion, some authors have hypothesized that there is a natural difference between sides.2,3 If this is true, then using the opposite as an estimate of preinjury ROM is inappropriate. The rationale for differences existing from side to side is related to usage. The idea is that the overuse of some joints could lead to overstress of the joint and, consequently, the development of micro injuries. These micro injuries would increase the deposit of scar tissue in the area leading to a decrease in the ROM, most commonly on the dominant side.4 Some authors have tried to determine whether there is a significant difference between body sides, but at present the available literature is contradictory and confusing. Stephanyshyn and Engsberg,5 Roaas and Andersson,6 and Boone and Azen7 found no significant differences between right and left sides when measuring different joints of the lower extremity. Boone and Azen7 studied upper extremity motions and found no significant differences between right and left sides. In contrast, Gunal et al2 found significant differences between right and left sides for 15 of 17 upper extremity motions measured. Gunal et al2 reported that ROM of the elbow and were statistically greater on the nondominant side; however the differences were small, ranging from 1.7° to 2.8°. Murray et al,8 Barnes et al,3 Ellenbecker,9 and Gunal et al2 compared shoulder motion between dominant and nondominant sides with no clear pattern emerging. It is not possible to synthesize these disparate results and determine 577
578
Macedo and Magee ROM Between Sides
whether there are side-to-side differences because the studies have used different measurement procedures and no study has systematically measured ROM for each joint. Although it seems that there is a great amount of information regarding the differences between sides, most studies measured 1 or 2 joints. Only 1 paper compared the difference between sides for the motions of the upper extremity,2 1 for the lower extremity6 and 1 for both upper and lower extremity.7 An important point when analyzing the results of the literature is not only the analysis of the statistically significant differences, but also the clinically significant differences. The majority of the papers that analyzed the differences in ROM between body sides did not report the size of the differences, thus not allowing interpretations about the clinical significance of the results. Therefore, this study helps to clarify the controversy created by previous studies by evaluating whether there is a significant difference between body sides. The objective of this study was to compare the ROM of the ankle, knee, hip, shoulder, elbow, and wrist between dominant and nondominant sides with a clinical significance approach.
Journal of Manipulative and Physiological Therapeutics October 2008
subjects was due to the inclusion criteria of the main project from where these data were extracted. The results of this study are part of a greater study and, therefore, sample size was calculated only for the main project. Participants were excluded if they reported neurologic, systemic, peripheral or rheumatic pathologies, disease or injury that might affect the musculoskeletal system,13 any history of musculoskeletal injury in the past year,14 any type of surgery that may have affected the musculoskeletal system, physical therapy treatment in the last year or any type of therapy that included stretching or manual therapy, been pregnant or having been pregnant in the past year,15 been involved in a high level or professional level of sports activities in the past year,9 or any diseases that might affect the level of the body hormones. Before the beginning of the study, the project was reviewed by the Health Research Ethics Board–Panel B from the University of Alberta and received approval. All participants receive an information sheet about the study and signed a consent form.
Data Collection
METHODS Instruments Two standard transparent plastic goniometers with arm lengths of 20 and 25 cm and a protractor portion divided into 2° segments were used to measure ROM. Previous research has shown that different sizes of goniometer can be used interchangeably.10-12 Before the study, the accuracy of the goniometers was determined by using 10 randomly selected computer-generated angles between 0° and 180°. Six goniometers were tested with 2 discarded because they did not agree (exactly) with the computer-generated angles.
Examiners One examiner who was a physical therapist was responsible for all the ROM measurements. Before the beginning of the study, the examiner was trained in the protocol of the study and the intrarater reliability was tested. To ensure blinding in the reliability testing, a second examiner who was also a physical therapist was responsible for recoding the ROM values. Both examiners had at least one year of experience systematically using the goniometer.
Participants Participants were recruited using advertisement directed at the population attending the University of Alberta (students and staff) and the population around the university area. The participants included in this study were white women aged 18 to 59 years. The selection of white participants avoided unknown biases due to possible differences between races. The inclusion of only female
Intrarater reliability of the goniometric measurements was tested using 12 subjects before beginning the study. The measurement of all 60 different active and passive ROMs was performed twice consecutively. To prevent bias, the goniometers were covered with a white adhesive and the examiner was not able to see the values recorded on the goniometer. A second examiner was responsible for reading the values on the goniometer and recording them. This ensured blinding of the first examiner and avoided bias for the pilot study. Participants were not given an opportunity to warm up because warm-up can change the biomechanical characteristics of the collagen, change the viscoelastic properties of the muscles,16 and could influence ROM available in a joint. The passive ROM measurements were performed according to the methods described by Norkin and White.17 The active ROM measurement used was the same protocol as the one used for the passive ROM, but the subjects actively performed the movement toward the maximum active ROM. During the active data collection, no stabilization was performed by the examiner except for glenohumeral shoulder measurements. The glenohumeral stabilization for the active test was the same as for the passive test. For the other ROMs measured, the examiner watched for compensatory movements such as scapular abduction, anterior tilt, or protraction of the scapula. If the participants used compensatory movements, the motion was corrected. If the participants could not correct the compensatory movement, it meant that the end range had been achieved and compensatory movements were occurring to provide more movement. In this case, the ROM was measured at the point before the compensation occurred.
Journal of Manipulative and Physiological Therapeutics Volume 31, Number 8
Macedo and Magee ROM Between Sides
Table 1. National Occupational Classification for all the participants for each age group Demographics 0 Management occupations 1 Business, finance and administration occupations 2 Natural applied occupations 3 Health occupations 4 Occupations in social science, education, government service, and religion 5 Occupations in art, culture, recreation, and sport 6 Sales and service occupations 7 Trade, transport and equipment operators, and related occupations 8 Occupations unique to primary industry 9 Occupations unique to processing, manufacturing, and utilities Retired Student
18-29 (n = 30)
30-39 (n = 20)
40-49 (n = 20)
50-59 (n = 20)
Overall (N = 90)
2
2 1 3 7 19 5 7
1 1 1
2 1
5
23
1 1 11
2 5 5 4
2
16
5
1 1
1 45
Values presented are the number of people.
Passive and active ROMs were measured for 30 different motions on both sides, consisting of a total of 120 measurements. The order of testing was randomly selected using the “draw out of a hat” method. All measurements were taken once actively and once passively for each joint on each side. The performance of one measurement of each motion avoided any carryover effect due to stretching or viscoelastic changes on the tissues and followed what is normally done on the clinical field.18 All ROM measurements were performed within 2 hours. All subjects completed a demographic and patient characteristics form that included questions such as age, height, and weight.
Statistical Analysis The statistical program SPSS 14.0 for Windows (SPSS, Inc, Chicago, Ill) was used for all data analyses. Descriptive analysis was used to analyze demographic data of the participants to characterize the study population. An intraclass correlation coefficient (ICC, 3.1) was used to analyze the intrarater reliability. The standard error of measurement (SEM) was calculated using the formula SEM = SD * √1 − R. The minimal detectable change (MDC) was calculated using the formula MDC = SEM * 1.96 * √2. A simple paired t test was used to calculate the differences between dominant and nondominant side. A more robust statistical analysis was not performed because it was not appropriate to compare the ROM between active and passive and between each one of the motions measured. Therefore, one paired t test was used for each of the 60 motions measured in this study.
RESULTS Main Study The ICC for intrarater reliability was greater than 0.68 for all ranges of motion measured. Only passive ankle eversion had an ICC of less than 0.70. Standard errors of
Table 2. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for ankle, knee, and hip ROM Motions Ankle ROM Passive Dorsiflexion (talocrural joint) Plantarflexion (talocrural joint) Eversion (tarsal joint) Inversion (tarsal joint) Active Dorsiflexion (talocrural joint) Plantarflexion (talocrural joint) Eversion (tarsal joint) Inversion (tarsal joint) Knee ROM Passive °Flexion °Extension Active °Flexion °Extension Hip ROM Passive Abduction Adduction Flexion Extension Internal rotation External rotation Active Abduction Adduction Flexion Extension Internal rotation External rotation
Mean
Paired t test (95% CI)
P
2.0889 0.507 to 3.671 −2.5333 −4.636 to −0.431 −0.4000 −2.493 to 1.693 0.6444 −1.125 to 2.414
*.010 *.019 .705 .471
−0.968 to 1.679 −5.642 to −1.202 −2.829 to 1.363 −0.174 to 3.219
.595 *.003 .489 *.078
−1.6444 −2.541 to −0.748 −1.0667 −2.380 to 0.246
*.000 .110
−1.7444 −2.695 to −0.794 −0.5000 −1.299 to 0.299
*.000 .217
−1.1667 −0.6556 0.6222 −0.711 1.822 −0.789
−2.497 to 0.163 −1.798 to 0.487 −0.933 to 2.177 −1.930 to 0.508 0.336 to 3.309 −2.214 to 0.636
.085 .257 .429 .250 *.017 .274
−2.0778 −1.9333 −3.333 −1.022 2.678 −3.900
−3.591 to −0.565 −3.083 to −0.784 −4.913 to −1.754 −2.235 to 0.191 1.208 to 4.148 −5.284 to −2.516
*.008 *.001 *.000 .098 *.000 *.000
0.3556 −3.4222 −0.7333 1.5222
* Statistically significant P b .05.
measurement were between 1.0° and 6.3°, and minimal detectable differences were between 2.7° and 17.4° depending on the joint measured.
579
580
Macedo and Magee ROM Between Sides
Journal of Manipulative and Physiological Therapeutics October 2008
Table 3. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for shoulder ROM Motions Shoulder ROM Abduction Passive Glenohumeral abduction Shoulder complex abduction Active Glenohumeral abduction Shoulder complex abduction Flexion Passive Glenohumeral flexion Shoulder complex flexion Active Glenohumeral flexion Shoulder complex flexion Extension Passive Glenohumeral extension Shoulder complex extension Active Glenohumeral extension Shoulder complex extension Internal (medial) rotation Passive Glenohumeral internal rotation Shoulder complex internal rotation Active Glenohumeral internal rotation Shoulder complex internal rotation External (lateral) rotation Passive Glenohumeral external rotation Shoulder complex external rotation Active Glenohumeral external rotation Shoulder complex external rotation
Mean
Pared t test (95% CI)
Table 4. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for elbow and wrist ROM Motions
P
1.189 −1.332 to 3.710 −3.011 −4.624 to −1.398
.351 *.000
1.600 −1.481 to 4.615 −0.644 −3.160 to 1.872
.310 .612
−1.622 −4.691 to 1.447 −0.667 −2.233 to 0.900
.296 .400
−1.344 −4.612 to 1.923 0.556 −0.771 to 1.882
.416 .408
−3.311 −5.170 to −1.453 −3.356 −5.072 to −1.639
*.001 *.000
−2.439 −4.193 to −1.040 −3.567 −5.956 to −1.177
*.001 *.004
−7.544 −9.998 to −5.091 −6.328 −9.013 to −3.642
*.000 *.000
−5.194 −7.793 to −2.596 −4.811 −7.337 to −2.285
*.000 *.000
Elbow ROM Passive Flexion Extension Pronation Supination Active Flexion Extension Pronation Supination Wrist ROM Passive Flexion Extension Ulnar deviation Radial deviation Active Flexion Extension Ulnar deviation Radial deviation
Mean
Paired t test (95% CI)
P
1.389 −0.844 −0.356 −3.644
−0.508 to 3.286 −1.708 to 0.019 −1.894 to 1.182 −5.254 to −2.035
.149 .055 .647 *.000
1.356 −0.556 −1.411 −2.922
0.144 to 2.568 −1.446 to 0.335 −2.975 to 0.153 −4.701 to −1.143
*.029 .218 .076 *.002
−0.267 −3.611 −0.944 −4.678
−1.787 to 1.254 −5.252 to −1.970 −2.370 to 0.481 −5.932 to −3.423
.728 *.000 .191 *.000
2.700 −3.167 −2.200 −5.000
0.989 to 4.411 −4.955 to −1.379 −3.739 to −0.661 −6.706 to −3.294
*.002 *.001 *.006 *.000
* Statistically significant P b .05.
3.422 2.467
1.264 to 5.581 0.253 to 4.680
*.002 *.029
5.089 6.189
2.502 to 7.676 4.057 to 8.320
*.000 *.000
* Statistically significant P b .05.
The 90 participants included in this study had a mean age of 37.2 (±12.4) years (range, 19-59 years). Participants' mean weight was 66.1 kg (±12.9) and mean height was 164.6 cm (±6.2). The demographics for occupation categorized according to the National Occupational Classification developed by Human Resources Development Canada in 1993 is presented in Table 1. All participants had all ranges of motion measured bilaterally. The results of the paired t test showed that 34 of the 60 motions measured (14/30 passive and 20/30 active) were significantly different between the dominant and nondominant sides. There were 10 motions greater on the dominant side and 24 motions greater on the nondominant side (Tables 2-4; positive mean difference means that the dominant side
Fig 1. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of passive ROM of the lower extremity.
had greater ROM than the nondominant side and negative mean difference means that the nondominant side had greater ROM than the dominant side). The greatest difference found between sides was for passive glenohumeral internal rotation that had the nondominant side 7.5° greater than the dominant side. The 95% confidence interval (CI) of the differences between sides and the MDC were plotted in a graph to confirm whether the differences between sides were large
Journal of Manipulative and Physiological Therapeutics Volume 31, Number 8
Fig 2. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of passive ROM of the upper extremity.
Macedo and Magee ROM Between Sides
Fig 4. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of active ROM of the upper extremity.
external rotation. All these 6 motions were also found to be statistically different between sides.
DISCUSSION
Fig 3. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of active ROM of the lower extremity.
enough to be detected for an individual. Because the MDC could cross the CI either on the upper or lower boundary, MDC was plotted as a positive and as a negative value (Figs 1-4). Minimal detectable change values are a reflection of clinically important change and the likelihood that true change has occurred. If the CI contains the values of the MDC, then the differences between sides are likely to be true. The results of the comparison of the MDC with the 95% CI show that, in all but 6 motions, the CI for between side differences contains the value of MDC. These motions were active glenohumeral external rotation, active shoulder complex external rotation, active radial deviation, active hip flexion, active hip internal rotation, and active hip
The results of this study support the practice of using the opposite side of the body in females to estimate what preinjury ROM would be. Thirty-four of the 60 motions measured were significantly different between dominant and nondominant sides, and these differences were small, ranging from 0.26° to 7.54°. Only 5 of the 60 motions had a difference between sides greater than 5°, and none of the motions had a difference greater than 8°. These results lead to the conclusion that, although there was a statistically significant difference between sides for some motions, the differences between sides are small and therefore probably clinically insignificant. The analysis of the CIs with the MDC confirms that the differences found in this study can be considered insignificant. This is because the magnitude of the differences was typically not great enough to exceed the MDC previously established for the goniometric ROM measures. In this study, only 6 motions were found to contain the values of the MDC in the CI. Furthermore, none of these motions had mean difference greater than 5° or upper or lower boundary of the CI greater than 8°. It is important to note that the MDC was calculated using the ICC and SEM of the intrarater reliability of the one rater of this study. For this reason, the results of this study should be taken with caution. A difference of 8° for a motion of 90° represents less than 8.5% of the total range available in the joint. Is a change of this magnitude important enough that it would be a risk factor for injury or decrease functionality of a joint? Or more importantly, is this difference great enough to exclude the possibility of using the opposite
581
582
Macedo and Magee ROM Between Sides
side as references? According to the American Medical Association,19 a change of less than 10° in external rotation is not of sufficient magnitude to be considered an impairment and therefore changes of less than 10° may be considered clinically insignificant.
CONCLUSION The results of the study showed that there was some difference in ROM between dominant and nondominant sides of the body. However, these differences are small and may not affect the estimation of the ROM when using the opposite side as reference. Therefore, these results support the practice of using the opposite side of the body as an indicator of pre-injury or normal ROM.
Practical Applications • In the group of participants in this study, only some ranges of motion are statistically significant between body sides. • Although some ranges of motion have statistically significant differences between sides, these differences may not be clinically significant.
ACKNOWLEDGMENT I would like to thank Professor Chris Maher for his support and helpful guide in the revisions of this manuscript and Jorge Fuentes for his help on the data collection of this study.
REFERENCES 1. American Academy of Orthopaedic Surgeons. Joint motion: method of measuring and recording. Chicago (Ill): American Academy of Orthopaedic Surgeons; 1965. 2. Gunal I, Kose N, Erdogan O, Gokturk E, Seber S. Normal range of motion of the joints of the upper extremity in male subjects, with special reference to side. J Bone Joint Surg Am 1996;78:1401-4.
Journal of Manipulative and Physiological Therapeutics October 2008
3. Barnes CJ, Van Steyn SJ, Fischer RA. The effects of age, sex, and shoulder dominance on range of motion of the shoulder. J Shoulder Elbow Surg 2001;10:242-6. 4. Allander E, Bjornsson OJ, Olafsson O, Sigfusson N, Thorsteinsson J. Normal range of joint movements in shoulder, hip, wrist and thumb with special reference to side: a comparison between two populations. Int J Epidemiol 1974;3:253-9. 5. Stefanyshyn DJ, Engsberg JR. Right to left differences in the ankle joint complex range of motion. Med Sci Sports Exerc 1994;26:551-5. 6. Roaas A, Andersson GB. Normal range of motion of the hip, knee and ankle joints in male subjects, 30-40 years of age. Acta Orthop Scand 1982;53:205-8. 7. Boone DC, Azen SP. Normal range of motion of joints in male subjects. J Bone Joint Surg Am 1979;61:756-9. 8. Murray MP, Gore DR, Gardner GM, Mollinger LA. Shoulder motion and muscle strength of normal men and women in two age groups. Clin Orthop Relat Res 1985;192:268-73. 9. Ellenbecker TS. Shoulder internal and external rotation strength and range of motion of highly skilled junior tennis players. Isokinet Exerc Sci 1992;2:65-72. 10. Elveru RA, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting. Subtalar and ankle joint measurements. Phys Ther 1988;68:672-7. 11. Riddle DL, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting. Shoulder measurements. Phys Ther 1987;67: 668-73. 12. Rothstein JM, Miller PJ, Roettger RF. Goniometric reliability in a clinical setting. Elbow and knee measurements. Phys Ther 1983;63:1611-5. 13. Gajdosik RL, Bohannon RW. Clinical measurement of range of motion. Review of goniometry emphasizing reliability and validity. Phys Ther 1987;67:1867-72. 14. Jonhagen S, Nemeth G, Eriksson E. Hamstring injuries in sprinters: the role of concentric and eccentric hamstring muscle strength and flexibility. Am J Sports Med 1994;22:262-6. 15. Marnach ML, Ramin KD, Ramsey PS, Song SW, Stensland JJ, An KN. Characterization of the relationship between joint laxity and maternal hormones in pregnancy. Obstet Gynecol 2003;101:331-5. 16. Mutingi G, Ranatunga KW. Temperature-dependent changes in the viscoelasticity of intact resting mammalia (rat) fast- and slow-twitch muscle fibers. J Physiol 1998;508:253-65. 17. Norkin CC, White DJ. Measurement of a joint motion: a guide to goniometry. 3rd ed. Philadelphia: FA Davis Co; 2003. 18. Nigg BM, Nigg CR, Reinschmidt C. Reliability and validity of active, passive and dynamic range of motion tests. Sportverletz Sportschaden 1995;9:51-7. 19. Doege TC, Houston TP. Guide to the evaluation of permanent impairment. 4th ed. Chicago: American Medical Association; 1995.
ABSTRACT Objective: The objective of this study was to compare ranges of motion (ROM) between dominant and nondominant sides for the joints of the upper and lower extremities. Methods: Ninety healthy white women from 18 to 59 years of age were measured in this study. Active and passive ROM were measured for the ankle, knee, hip, shoulder, elbow, and wrist using a standard goniometer. The order of the joints, motion, sides, and active or passive motion testing was randomly selected. A paired t test was used for the comparison between sides. Results: The results of this study showed a statistically significant difference between dominant and nondominant sides for 34 of the 60 ROM measured. The maximum mean difference between sides for all ROM measured was 7.5°. Conclusion: The results of this show that some ROM are different between body sides and that when these differences exist they are minimal and may not be clinically insignificant. These results support the practice of using the opposite side of the body as an indicator of preinjury or normal extremity ROM. (J Manipulative Physiol Ther 2008;31:577-582) Key Indexing Terms: Range of Motion, Articular; Female
he use of range of motion (ROM) measurements in the health care of musculoskeletal disorders is a common procedure when making diagnoses, setting treatment goals, and measuring treatment progress. Since 1965, when the first standardized manual for recording and measuring joint motion was published, it has been recognized that there is a variation in ROM between individuals.1 This variation creates a problem when trying to establish what preinjury or normal ROM would be for an injured joint. A potentially simple way to estimate normal ROM that accounts for individual variation is to presume that the patient's opposite (uninjured) side has the same ROM as the injured side before the injury. To date, the validity of this assumption has not been adequately tested.
T
a
Doctural Student, The George Institute for International Health, University of Sydney, Sydney, NSW, Australia. b Professor, University of Alberta, Edmonton, Alberta, Canada. c Associate Dean, Faculty of Rehabilitation Medicine, University of Alberta, Edmonton, Alberta, Canada. Submit requests for reprints to: Luciana Gazzi Macedo, BPT, MSc, The George Institute for International Health, University of Sydney, Level 7, 341 George St, Sydney, NSW 2000, Australia (e-mails: [email protected] [email protected]). Paper submitted August 1, 2007; in revised form May 16, 2008; accepted June 15, 2008. 0161-4754/$34.00 Copyright © 2008 by National University of Health Sciences. doi:10.1016/j.jmpt.2008.09.003
In contrast to the proposition that the opposite side is the best estimation of one's ROM of motion, some authors have hypothesized that there is a natural difference between sides.2,3 If this is true, then using the opposite as an estimate of preinjury ROM is inappropriate. The rationale for differences existing from side to side is related to usage. The idea is that the overuse of some joints could lead to overstress of the joint and, consequently, the development of micro injuries. These micro injuries would increase the deposit of scar tissue in the area leading to a decrease in the ROM, most commonly on the dominant side.4 Some authors have tried to determine whether there is a significant difference between body sides, but at present the available literature is contradictory and confusing. Stephanyshyn and Engsberg,5 Roaas and Andersson,6 and Boone and Azen7 found no significant differences between right and left sides when measuring different joints of the lower extremity. Boone and Azen7 studied upper extremity motions and found no significant differences between right and left sides. In contrast, Gunal et al2 found significant differences between right and left sides for 15 of 17 upper extremity motions measured. Gunal et al2 reported that ROM of the elbow and were statistically greater on the nondominant side; however the differences were small, ranging from 1.7° to 2.8°. Murray et al,8 Barnes et al,3 Ellenbecker,9 and Gunal et al2 compared shoulder motion between dominant and nondominant sides with no clear pattern emerging. It is not possible to synthesize these disparate results and determine 577
578
Macedo and Magee ROM Between Sides
whether there are side-to-side differences because the studies have used different measurement procedures and no study has systematically measured ROM for each joint. Although it seems that there is a great amount of information regarding the differences between sides, most studies measured 1 or 2 joints. Only 1 paper compared the difference between sides for the motions of the upper extremity,2 1 for the lower extremity6 and 1 for both upper and lower extremity.7 An important point when analyzing the results of the literature is not only the analysis of the statistically significant differences, but also the clinically significant differences. The majority of the papers that analyzed the differences in ROM between body sides did not report the size of the differences, thus not allowing interpretations about the clinical significance of the results. Therefore, this study helps to clarify the controversy created by previous studies by evaluating whether there is a significant difference between body sides. The objective of this study was to compare the ROM of the ankle, knee, hip, shoulder, elbow, and wrist between dominant and nondominant sides with a clinical significance approach.
Journal of Manipulative and Physiological Therapeutics October 2008
subjects was due to the inclusion criteria of the main project from where these data were extracted. The results of this study are part of a greater study and, therefore, sample size was calculated only for the main project. Participants were excluded if they reported neurologic, systemic, peripheral or rheumatic pathologies, disease or injury that might affect the musculoskeletal system,13 any history of musculoskeletal injury in the past year,14 any type of surgery that may have affected the musculoskeletal system, physical therapy treatment in the last year or any type of therapy that included stretching or manual therapy, been pregnant or having been pregnant in the past year,15 been involved in a high level or professional level of sports activities in the past year,9 or any diseases that might affect the level of the body hormones. Before the beginning of the study, the project was reviewed by the Health Research Ethics Board–Panel B from the University of Alberta and received approval. All participants receive an information sheet about the study and signed a consent form.
Data Collection
METHODS Instruments Two standard transparent plastic goniometers with arm lengths of 20 and 25 cm and a protractor portion divided into 2° segments were used to measure ROM. Previous research has shown that different sizes of goniometer can be used interchangeably.10-12 Before the study, the accuracy of the goniometers was determined by using 10 randomly selected computer-generated angles between 0° and 180°. Six goniometers were tested with 2 discarded because they did not agree (exactly) with the computer-generated angles.
Examiners One examiner who was a physical therapist was responsible for all the ROM measurements. Before the beginning of the study, the examiner was trained in the protocol of the study and the intrarater reliability was tested. To ensure blinding in the reliability testing, a second examiner who was also a physical therapist was responsible for recoding the ROM values. Both examiners had at least one year of experience systematically using the goniometer.
Participants Participants were recruited using advertisement directed at the population attending the University of Alberta (students and staff) and the population around the university area. The participants included in this study were white women aged 18 to 59 years. The selection of white participants avoided unknown biases due to possible differences between races. The inclusion of only female
Intrarater reliability of the goniometric measurements was tested using 12 subjects before beginning the study. The measurement of all 60 different active and passive ROMs was performed twice consecutively. To prevent bias, the goniometers were covered with a white adhesive and the examiner was not able to see the values recorded on the goniometer. A second examiner was responsible for reading the values on the goniometer and recording them. This ensured blinding of the first examiner and avoided bias for the pilot study. Participants were not given an opportunity to warm up because warm-up can change the biomechanical characteristics of the collagen, change the viscoelastic properties of the muscles,16 and could influence ROM available in a joint. The passive ROM measurements were performed according to the methods described by Norkin and White.17 The active ROM measurement used was the same protocol as the one used for the passive ROM, but the subjects actively performed the movement toward the maximum active ROM. During the active data collection, no stabilization was performed by the examiner except for glenohumeral shoulder measurements. The glenohumeral stabilization for the active test was the same as for the passive test. For the other ROMs measured, the examiner watched for compensatory movements such as scapular abduction, anterior tilt, or protraction of the scapula. If the participants used compensatory movements, the motion was corrected. If the participants could not correct the compensatory movement, it meant that the end range had been achieved and compensatory movements were occurring to provide more movement. In this case, the ROM was measured at the point before the compensation occurred.
Journal of Manipulative and Physiological Therapeutics Volume 31, Number 8
Macedo and Magee ROM Between Sides
Table 1. National Occupational Classification for all the participants for each age group Demographics 0 Management occupations 1 Business, finance and administration occupations 2 Natural applied occupations 3 Health occupations 4 Occupations in social science, education, government service, and religion 5 Occupations in art, culture, recreation, and sport 6 Sales and service occupations 7 Trade, transport and equipment operators, and related occupations 8 Occupations unique to primary industry 9 Occupations unique to processing, manufacturing, and utilities Retired Student
18-29 (n = 30)
30-39 (n = 20)
40-49 (n = 20)
50-59 (n = 20)
Overall (N = 90)
2
2 1 3 7 19 5 7
1 1 1
2 1
5
23
1 1 11
2 5 5 4
2
16
5
1 1
1 45
Values presented are the number of people.
Passive and active ROMs were measured for 30 different motions on both sides, consisting of a total of 120 measurements. The order of testing was randomly selected using the “draw out of a hat” method. All measurements were taken once actively and once passively for each joint on each side. The performance of one measurement of each motion avoided any carryover effect due to stretching or viscoelastic changes on the tissues and followed what is normally done on the clinical field.18 All ROM measurements were performed within 2 hours. All subjects completed a demographic and patient characteristics form that included questions such as age, height, and weight.
Statistical Analysis The statistical program SPSS 14.0 for Windows (SPSS, Inc, Chicago, Ill) was used for all data analyses. Descriptive analysis was used to analyze demographic data of the participants to characterize the study population. An intraclass correlation coefficient (ICC, 3.1) was used to analyze the intrarater reliability. The standard error of measurement (SEM) was calculated using the formula SEM = SD * √1 − R. The minimal detectable change (MDC) was calculated using the formula MDC = SEM * 1.96 * √2. A simple paired t test was used to calculate the differences between dominant and nondominant side. A more robust statistical analysis was not performed because it was not appropriate to compare the ROM between active and passive and between each one of the motions measured. Therefore, one paired t test was used for each of the 60 motions measured in this study.
RESULTS Main Study The ICC for intrarater reliability was greater than 0.68 for all ranges of motion measured. Only passive ankle eversion had an ICC of less than 0.70. Standard errors of
Table 2. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for ankle, knee, and hip ROM Motions Ankle ROM Passive Dorsiflexion (talocrural joint) Plantarflexion (talocrural joint) Eversion (tarsal joint) Inversion (tarsal joint) Active Dorsiflexion (talocrural joint) Plantarflexion (talocrural joint) Eversion (tarsal joint) Inversion (tarsal joint) Knee ROM Passive °Flexion °Extension Active °Flexion °Extension Hip ROM Passive Abduction Adduction Flexion Extension Internal rotation External rotation Active Abduction Adduction Flexion Extension Internal rotation External rotation
Mean
Paired t test (95% CI)
P
2.0889 0.507 to 3.671 −2.5333 −4.636 to −0.431 −0.4000 −2.493 to 1.693 0.6444 −1.125 to 2.414
*.010 *.019 .705 .471
−0.968 to 1.679 −5.642 to −1.202 −2.829 to 1.363 −0.174 to 3.219
.595 *.003 .489 *.078
−1.6444 −2.541 to −0.748 −1.0667 −2.380 to 0.246
*.000 .110
−1.7444 −2.695 to −0.794 −0.5000 −1.299 to 0.299
*.000 .217
−1.1667 −0.6556 0.6222 −0.711 1.822 −0.789
−2.497 to 0.163 −1.798 to 0.487 −0.933 to 2.177 −1.930 to 0.508 0.336 to 3.309 −2.214 to 0.636
.085 .257 .429 .250 *.017 .274
−2.0778 −1.9333 −3.333 −1.022 2.678 −3.900
−3.591 to −0.565 −3.083 to −0.784 −4.913 to −1.754 −2.235 to 0.191 1.208 to 4.148 −5.284 to −2.516
*.008 *.001 *.000 .098 *.000 *.000
0.3556 −3.4222 −0.7333 1.5222
* Statistically significant P b .05.
measurement were between 1.0° and 6.3°, and minimal detectable differences were between 2.7° and 17.4° depending on the joint measured.
579
580
Macedo and Magee ROM Between Sides
Journal of Manipulative and Physiological Therapeutics October 2008
Table 3. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for shoulder ROM Motions Shoulder ROM Abduction Passive Glenohumeral abduction Shoulder complex abduction Active Glenohumeral abduction Shoulder complex abduction Flexion Passive Glenohumeral flexion Shoulder complex flexion Active Glenohumeral flexion Shoulder complex flexion Extension Passive Glenohumeral extension Shoulder complex extension Active Glenohumeral extension Shoulder complex extension Internal (medial) rotation Passive Glenohumeral internal rotation Shoulder complex internal rotation Active Glenohumeral internal rotation Shoulder complex internal rotation External (lateral) rotation Passive Glenohumeral external rotation Shoulder complex external rotation Active Glenohumeral external rotation Shoulder complex external rotation
Mean
Pared t test (95% CI)
Table 4. Paired t test, mean difference, 95% CIs, and P values for the difference between dominant and nondominant sides for elbow and wrist ROM Motions
P
1.189 −1.332 to 3.710 −3.011 −4.624 to −1.398
.351 *.000
1.600 −1.481 to 4.615 −0.644 −3.160 to 1.872
.310 .612
−1.622 −4.691 to 1.447 −0.667 −2.233 to 0.900
.296 .400
−1.344 −4.612 to 1.923 0.556 −0.771 to 1.882
.416 .408
−3.311 −5.170 to −1.453 −3.356 −5.072 to −1.639
*.001 *.000
−2.439 −4.193 to −1.040 −3.567 −5.956 to −1.177
*.001 *.004
−7.544 −9.998 to −5.091 −6.328 −9.013 to −3.642
*.000 *.000
−5.194 −7.793 to −2.596 −4.811 −7.337 to −2.285
*.000 *.000
Elbow ROM Passive Flexion Extension Pronation Supination Active Flexion Extension Pronation Supination Wrist ROM Passive Flexion Extension Ulnar deviation Radial deviation Active Flexion Extension Ulnar deviation Radial deviation
Mean
Paired t test (95% CI)
P
1.389 −0.844 −0.356 −3.644
−0.508 to 3.286 −1.708 to 0.019 −1.894 to 1.182 −5.254 to −2.035
.149 .055 .647 *.000
1.356 −0.556 −1.411 −2.922
0.144 to 2.568 −1.446 to 0.335 −2.975 to 0.153 −4.701 to −1.143
*.029 .218 .076 *.002
−0.267 −3.611 −0.944 −4.678
−1.787 to 1.254 −5.252 to −1.970 −2.370 to 0.481 −5.932 to −3.423
.728 *.000 .191 *.000
2.700 −3.167 −2.200 −5.000
0.989 to 4.411 −4.955 to −1.379 −3.739 to −0.661 −6.706 to −3.294
*.002 *.001 *.006 *.000
* Statistically significant P b .05.
3.422 2.467
1.264 to 5.581 0.253 to 4.680
*.002 *.029
5.089 6.189
2.502 to 7.676 4.057 to 8.320
*.000 *.000
* Statistically significant P b .05.
The 90 participants included in this study had a mean age of 37.2 (±12.4) years (range, 19-59 years). Participants' mean weight was 66.1 kg (±12.9) and mean height was 164.6 cm (±6.2). The demographics for occupation categorized according to the National Occupational Classification developed by Human Resources Development Canada in 1993 is presented in Table 1. All participants had all ranges of motion measured bilaterally. The results of the paired t test showed that 34 of the 60 motions measured (14/30 passive and 20/30 active) were significantly different between the dominant and nondominant sides. There were 10 motions greater on the dominant side and 24 motions greater on the nondominant side (Tables 2-4; positive mean difference means that the dominant side
Fig 1. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of passive ROM of the lower extremity.
had greater ROM than the nondominant side and negative mean difference means that the nondominant side had greater ROM than the dominant side). The greatest difference found between sides was for passive glenohumeral internal rotation that had the nondominant side 7.5° greater than the dominant side. The 95% confidence interval (CI) of the differences between sides and the MDC were plotted in a graph to confirm whether the differences between sides were large
Journal of Manipulative and Physiological Therapeutics Volume 31, Number 8
Fig 2. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of passive ROM of the upper extremity.
Macedo and Magee ROM Between Sides
Fig 4. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of active ROM of the upper extremity.
external rotation. All these 6 motions were also found to be statistically different between sides.
DISCUSSION
Fig 3. Confidence interval of the difference between dominant and nondominant sides and minimal detectable change of active ROM of the lower extremity.
enough to be detected for an individual. Because the MDC could cross the CI either on the upper or lower boundary, MDC was plotted as a positive and as a negative value (Figs 1-4). Minimal detectable change values are a reflection of clinically important change and the likelihood that true change has occurred. If the CI contains the values of the MDC, then the differences between sides are likely to be true. The results of the comparison of the MDC with the 95% CI show that, in all but 6 motions, the CI for between side differences contains the value of MDC. These motions were active glenohumeral external rotation, active shoulder complex external rotation, active radial deviation, active hip flexion, active hip internal rotation, and active hip
The results of this study support the practice of using the opposite side of the body in females to estimate what preinjury ROM would be. Thirty-four of the 60 motions measured were significantly different between dominant and nondominant sides, and these differences were small, ranging from 0.26° to 7.54°. Only 5 of the 60 motions had a difference between sides greater than 5°, and none of the motions had a difference greater than 8°. These results lead to the conclusion that, although there was a statistically significant difference between sides for some motions, the differences between sides are small and therefore probably clinically insignificant. The analysis of the CIs with the MDC confirms that the differences found in this study can be considered insignificant. This is because the magnitude of the differences was typically not great enough to exceed the MDC previously established for the goniometric ROM measures. In this study, only 6 motions were found to contain the values of the MDC in the CI. Furthermore, none of these motions had mean difference greater than 5° or upper or lower boundary of the CI greater than 8°. It is important to note that the MDC was calculated using the ICC and SEM of the intrarater reliability of the one rater of this study. For this reason, the results of this study should be taken with caution. A difference of 8° for a motion of 90° represents less than 8.5% of the total range available in the joint. Is a change of this magnitude important enough that it would be a risk factor for injury or decrease functionality of a joint? Or more importantly, is this difference great enough to exclude the possibility of using the opposite
581
582
Macedo and Magee ROM Between Sides
side as references? According to the American Medical Association,19 a change of less than 10° in external rotation is not of sufficient magnitude to be considered an impairment and therefore changes of less than 10° may be considered clinically insignificant.
CONCLUSION The results of the study showed that there was some difference in ROM between dominant and nondominant sides of the body. However, these differences are small and may not affect the estimation of the ROM when using the opposite side as reference. Therefore, these results support the practice of using the opposite side of the body as an indicator of pre-injury or normal ROM.
Practical Applications • In the group of participants in this study, only some ranges of motion are statistically significant between body sides. • Although some ranges of motion have statistically significant differences between sides, these differences may not be clinically significant.
ACKNOWLEDGMENT I would like to thank Professor Chris Maher for his support and helpful guide in the revisions of this manuscript and Jorge Fuentes for his help on the data collection of this study.
REFERENCES 1. American Academy of Orthopaedic Surgeons. Joint motion: method of measuring and recording. Chicago (Ill): American Academy of Orthopaedic Surgeons; 1965. 2. Gunal I, Kose N, Erdogan O, Gokturk E, Seber S. Normal range of motion of the joints of the upper extremity in male subjects, with special reference to side. J Bone Joint Surg Am 1996;78:1401-4.
Journal of Manipulative and Physiological Therapeutics October 2008
3. Barnes CJ, Van Steyn SJ, Fischer RA. The effects of age, sex, and shoulder dominance on range of motion of the shoulder. J Shoulder Elbow Surg 2001;10:242-6. 4. Allander E, Bjornsson OJ, Olafsson O, Sigfusson N, Thorsteinsson J. Normal range of joint movements in shoulder, hip, wrist and thumb with special reference to side: a comparison between two populations. Int J Epidemiol 1974;3:253-9. 5. Stefanyshyn DJ, Engsberg JR. Right to left differences in the ankle joint complex range of motion. Med Sci Sports Exerc 1994;26:551-5. 6. Roaas A, Andersson GB. Normal range of motion of the hip, knee and ankle joints in male subjects, 30-40 years of age. Acta Orthop Scand 1982;53:205-8. 7. Boone DC, Azen SP. Normal range of motion of joints in male subjects. J Bone Joint Surg Am 1979;61:756-9. 8. Murray MP, Gore DR, Gardner GM, Mollinger LA. Shoulder motion and muscle strength of normal men and women in two age groups. Clin Orthop Relat Res 1985;192:268-73. 9. Ellenbecker TS. Shoulder internal and external rotation strength and range of motion of highly skilled junior tennis players. Isokinet Exerc Sci 1992;2:65-72. 10. Elveru RA, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting. Subtalar and ankle joint measurements. Phys Ther 1988;68:672-7. 11. Riddle DL, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting. Shoulder measurements. Phys Ther 1987;67: 668-73. 12. Rothstein JM, Miller PJ, Roettger RF. Goniometric reliability in a clinical setting. Elbow and knee measurements. Phys Ther 1983;63:1611-5. 13. Gajdosik RL, Bohannon RW. Clinical measurement of range of motion. Review of goniometry emphasizing reliability and validity. Phys Ther 1987;67:1867-72. 14. Jonhagen S, Nemeth G, Eriksson E. Hamstring injuries in sprinters: the role of concentric and eccentric hamstring muscle strength and flexibility. Am J Sports Med 1994;22:262-6. 15. Marnach ML, Ramin KD, Ramsey PS, Song SW, Stensland JJ, An KN. Characterization of the relationship between joint laxity and maternal hormones in pregnancy. Obstet Gynecol 2003;101:331-5. 16. Mutingi G, Ranatunga KW. Temperature-dependent changes in the viscoelasticity of intact resting mammalia (rat) fast- and slow-twitch muscle fibers. J Physiol 1998;508:253-65. 17. Norkin CC, White DJ. Measurement of a joint motion: a guide to goniometry. 3rd ed. Philadelphia: FA Davis Co; 2003. 18. Nigg BM, Nigg CR, Reinschmidt C. Reliability and validity of active, passive and dynamic range of motion tests. Sportverletz Sportschaden 1995;9:51-7. 19. Doege TC, Houston TP. Guide to the evaluation of permanent impairment. 4th ed. Chicago: American Medical Association; 1995.