Modified sphygmomanometer versus strain gauge hand-held dynamometer

911

Modified Sphygmomanometer Hand-Held Dynamometer

Versus Strain Gauge

Richard W. Bohannon, EdD, Michelle hf. Lusardi, MS ABSTRACT. Bohannon RW, Lusardi MM: Modified sphygmomanometer versus strain gauge hand-held dynamometer. Arch Phys Med Rehabil 1991;72:911-4. 0 The purpose of this study was to compare measurements of elbow flexor muscle strength obtained with a modified sphygmomanometer (MS) and a strain gauge hand-held dynamometer (HHD). Thirty-two healthy subjects performed two isometric contractions of about five seconds against each device. Repeated measurements with each device did not differ significantly and were correlated (r > -960) and reliable (intraclass correlation coefficients > -960). The means of two measurements obtained with each device were related in a curvilinear manner (R= .882), with MS measurements above 2lOmmHg rising more slowly than corresponding HHD measurements. Although supporting the reliability of each device, the results suggest that the devices do not provide comparable measurements throughout the range of the MS. The MS may not be the best choice for obtaining accurate strength measurements of stronger individuals. KEY WORDS: Arm; Muscle test; Strength

Although manual muscle testing has been the method of choice for assessing muscle strength for much of this century, the measurements that it provides are far less precise and sensitive than those obtained with various instruments.‘-” Among the many types of strength testing instruments that have been advocated, particularly in the last decade, are those that are hand-held by the examiner. Although hand-held devices have limitations, their portability makes them unique and desirable for muscle strength testing in some situations. A number of hand-held dynamometers (HHDs) are commercially available. Availability notwithstanding, an alternative to manufactured HHDs has been described. That alternative is the modified sphygmomanometer (MS).4-h Measurements obtained with HHDs~-~and MSS~J,‘~have each been examined for accuracy and have been shown to be reliable (reliability coefficients typically exceed .90). Measurements obtained with the two devices, however, have not been compared. The comparability of the two devices is important, particularly in light of some of the theoretical sources of error inherent in MSs. Chief among such sources are the MS’s reliance on pressure (which is influenced by cutoff contact area) and its use of air (which is compressible) as a pressure transmission medium.” The purpose of this study was to compare measurements of elbow flexion muscle strength obtained with an MS and a strain gauge HHD. More specificially, three issues were examined: the test-retest reliability of measurements obtained with an MS, the test-retest reliability of measurements obtained with a strain gauge HHD, and the relationship between measurements obtained with the two devices. From the School of Allied Health, U-101, University of CT. Srorrs, CT 06269. 2101. Submitted for publication March 19, 1990. Accepted in revised form August 24, 1990. No commercial party having a direct or indirect interest m the subject matter of this article has conferred or will confer a benefit upon the authors or upon any oreanization with which the authors are associated. xeprint requests to Dr. Bohannon, School of Allied Health, University of Connecticut, Box U-101, 358 Mansfield Road. Storrs, CT 06269-21lll.

METHOD Subjects Thirty-two healthy volunteers part:icipated in this study with written informed consent. Six were men and 26 were women. Their mean age was 30.9 years (SE =. ? 8.9; range = 21 to 53 years). Mean height was 166.6cm (SD = 8.0; range = 149.9 to 180.3cm). Mean weight was 62.lkg (SD = ? 9.4; range = 49.0 to 86.3kg). Based on the toss of a coin, the dominant upper extremity of 14 subjects was tested and the nondominant upper extremity of 18 r;ub.jectswas tested. Instrumentation The HHD available for use in this study” incorporated a strain gauge and provided a digital read-out. It had a range of more than 60kg and was sensitive to .Olkg. It was shown to be accurate before and after the experimental period, when loaded with certified weights. The sphygmomanometer modified for this studyb incorporated a 300-mmHg column and was modified in a manner similar to that described by Helewa and colleagues.’ More specifically, the bladder portion of the cuff was folded into four equal parts., and the rest of the cuff rolled loosely around. The entire rolled cuff was then wrapped with several layers of adhesive tape. When inflated to a baseline 30mmHg, the rolled cuff was 14cm long, 6cm wide, and 4cm thick. The MS gave consistent readings when loaded with certified weights. Procedure All testing was performed by the same examiner. Subjects were tested while supine on a padded mat table. Their designated upper extremity was positioned with the shoulder abducted about 30”, the elbow flexed OO”, and the forearm supinated. During testing, each of the devices was placed just Arch Phys Med Rehabll Vol72, October 1991

COMPARISON OF STRENGTHTESTING DEVICES, Bohannon

the HHD were recorded in kilograms. All measurements with the two devices were made without reference to previous measurements. The measurements from the MS were recorded in millimeters of mercury. The actual measurement recorded was that read off the mercury column minus the 30-mmHg baseline, to which the cuff was inflated before testing. Statistical

Fig l-Measurement of elbow flexion muscle strength with a modified sphygmomanometer.

proximal to the styloids on the flexor surface of the forearm. The arm was manually stabilized (figs 1 and 2). The instructions to the subjects were to “take one to two seconds to come to maximum elbow flexion and then continue bending your elbow as hard as possible until I say stop.” Consistent verbal encouragement was provided throughout each effort. The examiner said “Stop” after appoximately five seconds. Two isometric hold contractions (make tests) were performed against each device in a random order. All measurements were separated by a two-minute rest. Care was taken to hold the HHD perpendicular to the flexor surface of the forearm. Care was also taken to support the rolled sphygmomanometer cuff with a flat hand (so that pressure readings were not augmented by examiner grasp). The measurements from

Analysis

The reliability (consistency) of each device was described using four statistical procedures. Analysis of variance (ANOVA) for repeated measures was used to determine whether a significant difference existed between trial 1 and trial 2. The mean square values of the ANOVA were then used to calculate intraclass correlation coefficients (ICC 3, 1). Pearson product moment correlations and linear regression were then calculated for the measurements of the two trials. Two devices were compared using the means of the two measurements for each device. Initially, Pearson product moment correlations and linear regression were used to compare measurements obtained with the two instruments. After examining the scattergram of measurements obtained with the two instruments, further analysis was performed using multiple regression and a model for curvilinear relationships (y = a+bx+bx2).

RESULTS Results of the test-retest reliability measures for each device are summarized in table 1. Repeated measures ANOVAs failed to identify significant differences between trials for the HHD or for the MS. Consistency of measurements with both devices was further supported by ICCs and Pearson product moment correlations that exceed .960. The scattergrams and linear regression equations for each device (figs 3 and 4) provide further support for the consistency of repeated measures with each device. The relationship between measurement means obtained with the two devices is illustrated in figure 5. Although the scattergram is narrow at low to midrange values on each scale, it widens and demonstrates reduced slope as the magnitude of measurements increases. A drop-off in mean MS measurements with respect to mean HHD measurements is apparent. Given the magnitude of the multiple correlation coefficient (R = -882) for the curvilinear model, the relationship between MS and HHD measurements is best described as curvilinear.

DISCUSSION

Fig 2-Measurement of elbow flexion muscle strength with a digital strain gauge hand-held dynamometer. Arch Phyo Med Aehabll Vol72,

October

1991

Results of this study support previous findings of test-retest reliability for the MS4r5 and for the HHD.7-9 However, the finding of a curvilinear relationship between the mean measurements obtained with the two devices and of a drop-off in MS measurements above 2lOmmHg raises questions concerning clinical application and interpretation of measurements obtained with the MS. Rice and associateslo found a high correlation (r= .994) and a near linear relationship between MS measurements (in mmHg) and the load imparted (in kg) on the modified cuff by standard weights. Helewa and colleagues4 reported similar results for

COMPARISON OF STRENGTHTESTING DEVICES, Bohannon

913

Statistics Relevant to the Rellablllty of Repeated Measurements Obtalned with Two Devices -___ Measurement 1 Device MS (mmHg)'

HHD (kg)**

x 196.0

18.8

SD

jT

30.3

198.9

46

‘Modified spbygmomanometer;

Measurement 2

18.9

ANOVA results Source

Sum of Sguares

29.4 Subjects Measuremenl EWX 4.1 Subjects Measurement EWX

54284.4 129.4 1006.1 2442.7 .30 15.17

SD

Linear regression results .---.--___-

df

ii’

31

1751.1 129.4 32.5 39.4 .30 .49

I 31 31

I 31

lntraclass correlation coetricient

F

P

3.99

.os5

v’)

I Ilh

‘64

.443

I .Oh

1..x

.(.74

.hO

slope intercept __--__--

R

WI4 v71

**

two modifications of cuffs (r= .92/.94). As a result, both authors suggested a comparability between MS strength measurements and other objective measures of strength. Clinical comparability, however, would best be supported by a significant linear relationship between measures obtained from human subjects with an MS and another instrument (eg, HHD) designed for measuring muscle strength. Although there appears, in our comparison, to be a near linear relationship between instrument measurements at lower strengths (fig 5), the drop-off in MS values above 2lOmmHg does not support comparability between instruments throughout the range of the MS. The between-instrument variation may be explained in part by properties of the MS itself. The effect of air compression within the modified cuff” may only become apparent when relatively strong subjects push with narrow limb segments against the cuff of the modified MS. In such cases, the pressure of the limb segment against the cuff may tend to surpass the underlying pressure within the cuff itself. Other sources of variation with the MS include the tester’s failure to (1) maintain a flat palm on the modified cuff; (2) apply appropriate resistance; or (3) accurately read the mercury column or dial.’

The MS has most frequently been used in the clinic to evaluate performance of patients with musculoskeletal dysfunction and associated strength loss. 4~5Even older subjects, whose strength is expected to be lower, have been noted to sometimes exceed the upper limits of the MS. lli Considering these factors and the curvilinear relationship found in this study behveen the measurements obtained with the two devices, we must caution against using the MS when accurate strength measurements are required for stronger subjects. Modified sphygmomanometer values above 2lOmmHg, particularly, may not accurately reflect patient performance as compared to HHD measures. CONCLUSION Both the HHD and MS have been used in clinical and research settings to obtain objective measures of strength. This study confirmed the test-retest reliability of each instrument. The results of this study suggest, however, that MS measurements may not be fully reflective of the actual muscle strength for subjects generating forces above some critical level. Fur-

.

.

t .

/

y R

0

1 c

100

106

+

SPHY6MOMANOMETEN

200 MEASUREMENT

/

99x

964

7.

150

/

250 2 lmm

300 HOI

Fig 3-Scattergram illustrating the relationship between repeated measurements of elbow flexion muscle strength obtained with a modified sphygmomanometer.

10

20

15 UYNAMOMETER

MEASUREMENT

30

25 1 lku

Fig 4-Scattergram illustrating the relationship between repeated measurements of elbow flexion muscle strength obtained with a digital strain gauge hand-held dynamometer. Arch Phys Med Rehabll Vol72,

October

1991

COMPARISON OF STRENGTHTESTING DEVICES, Bohannon

914

6.

7. 8.

4, 10

15 MEAN

20

DYNAMOMETER

MEASUREMENT

30

25

Ikol

Fig S-Scattergram illustrating the relationship between measurements of elbow flexion strength obtained with a modified sphygmomanometer and a hand-held dynamometer.

ther research comparing the MS to other objective strength measures must be performed before the MS can be recommended unequivocally for clinical use.

9.

10.

11.

Bohannon RW. Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil 1986;67:390-2. Aitkens S, Lord J, Bernauer E, Fowler WM, Lieberman JS, Berck P. Relationship of manual muscle testing to objective strength measurements. Muscle Nerve 1989;12:173-7. Helewa A, Goldsmith CH, Smythe HA. The modified sphygmomanometer-an instrument to measure muscle strength: a validation study. J Chron Dis 1981;34:353-61. Helewa A, Goldsmith CH, Smythe HA. Patient, observer and instrument variation in the measurement of strength of shoulder abductor muscles in patients with rheumatoid arthritis using a modified sphygmomanometer. J Rheumatol 1986;13:1044-9. Isherwood L, Lew L, Dean E. Indirect evidence for eccentric muscle contraction during isometric muscle testing performed with a modified sphygmomanometer. Physiother Can 1989;41:138-42. Bohannon RW, Andrews AW. Interrater reliability of hand-held dynamometry. Phys Ther 1987;67:931-3. Bohannon RW, Andrews AW. Accuracy of spring and strain gauge hand-held dynamometers. J Orthop Sports Phys Ther 1989;10:323-5. Bohannon RW. Hand-held versus isokinetic dynamometer for measurement of static knee extension torques. Clin Phys Physiol Meas 1990;11:217-22. Rice CL, Cunningham DA, Paterson DH, Rechnitzer PA. Strength in an elderly population. Arch Phys Med Rehabil 1989;70:3917. Kondraske GV. Measurement of the quality of hand contractions (letter). Med Biol Eng Comput 1985;23:399.

References

1. Beasley WC. Influence of method on estimates of normal knee extensor force among normal and postpolio children. Phys Ther Rev 1956;36:21-41.

Arch Phys Med Rehabil Vol72, October 1991

Suppliers a. Ametek, Mansfield and Green Division, Largo, FL 33543 b. Baumanometer, W.A. Baum Co., Inc., New York, NY