- Email: [email protected]
Quantification of full-range-of-motion unilateral and bilateral knee flexion and extension torque ratios
971
Quantification of Full-Range-of-Motion Unilateral and Bilateral Knee Flexion and Extension Torque Ratios Michael A. Welsch, PhD, Patty A. Williams, MS, Michael L. Pollock, PhD, James E. Graves, PhD, Daniel N. Foster, MS, Michael N. F&ton, MD ABSTRACT. Welsch MA, Williams PA, Pollock ML, Graves JE, Foster DN, Fulton MN. Quantification of full-range-ofmotion unilateral and bilateral knee flexion and extension torque ratios. Arch Phys Med Rehabil 1998;79:971-8. Objectives: To evaluate the reliability and variability of repeated measurements of isometric knee flexion and extension strength, to quantify the extent of measurement error that may occur due to gravity, and to quantify isometric knee flexion/ extension torque ratios at multiple angles through a full range of motion. Design: Reliability assessment. Setting: A university exercise center. Participants: Seventy-seven healthy men and women recruited from a university and surrounding community. Intervention: Isometric knee flexion and extension strength tests. Main Outcome Measures: Knee flexionlextension strength was measured at 6”, 24”, 42”, 60”, 78”, 96”, and 108” of knee flexion. Before each contraction, subjects were instructed to completely relax the limbs to measure the mass of the lower leg. Torque values obtained during relaxation at each angle were added to or subtracted from “Total Torque” (TTQ) at peak exertion. The adjusted value was recorded as “Net Muscular Torque” (NMT) . Results: Reliability for the unilateral and bilateral tests was high (Y = .88 to Y = .98) and measurement variability low (SEM% = 5.1% to 12.6%). There was a statistically significant difference at each angle of measurement between the TTQ and NMT values for both knee flexion and extension. Knee flexionlextension ratios were highly dependent on the angle tested, ranging from 1.30 (at 6”) to .31 (at 108”). Conclusions: Isometric testing, using standardized angles, can reliably quantify knee flexionlextension strength. Furthermore, these findings emphasize the importance of correcting for the mass of the lower leg when assessing muscle function. Angle-specific knee flexionlextension torque ratios should provide clinicians with a more precise method of evaluating muscular balance (imbalance) throughout the range of motion. o 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Center for Exercise Science, College of Medicine and College of Health and Human Performance, University of Florida, Gainesville (Drs. Welsch, Pollock, Graves. Fulton. Ms. Williams. Mr. Foster): the Deoartment of Kinesioloev. College of Educa&n, Louisiana State Ur&rsity, B&I Rouge (Dr. Welsch); and the&par&&t of Health and Physical Education, Syracuse University, Syracuse, NY (Dr. Graves). Submitted for publication July 3,1997. Accepted in revised form February 20. 1998. No commercial party having a direct or indirect interest in the subject matter of this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Michael A. Welsch, PhD, 112 Huey P. Long Field House, Department of Kinesiology, Louisiana State University, Baton Rouge, LA 70803. @ 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003.9993/98/7908-456653.00/O
A
N IMBALANCE IN force-generating capacity between reciprocal muscle groups can increase the likelihood of injury to a joint. l-3Thus, reciprocal muscle group strength ratios provide clinicians with valuable information regarding muscular balance or imbalance around a joint. The relationship between knee flexor and extensor strength has been of particular interest to many investigators because of the complexity of, and the incidence of injury to, the structures that make up the knee joint. Campbell and Glenn4 have suggested that the flexion/extension strength ratios may be more important in the assessment of muscle function than the measurement of absolute strength. Most studies reporting flexion/extension ratios have used peak torque isokinetic strength testing data and have reported ratios between .55 and .80 (mean ratio, .60).2~3~5-16 It is evident from these studies that knee flexion/extension ratios are highly dependent on the angular velocity of the movement, the subject’s age, the subject’s gender, and status of training.5 Generally, knee flexionlextension ratios are determined from a single peak torque value throughout the range.5 Muscular strength varies greatly with joint position, however, as a result of changes in mechanical advantage and the length of the muscle. In addition, many studies have failed to correct for the effect of gravity on the measurement. Failure to consider this effect may have resulted in erroneous conclusions about muscular balance and function6 Thus, flexion/extension ratios based on a single peak torque value alone provide limited information regarding muscular balance. Recent research from our laboratory has shown that isometric testing using standardized angles and precise positioning of the muscle group of interest is necessary to ensure adequate reproducibility of repeated tests. This provides a safe and accurate means of evaluating isolated muscle strength through a full range of motion in both healthy and clinical populations.i7,ix We hypothesized that isometric knee flexion and extension (both unilateral and bilateral) testing, using standardized test angles and accounting for the mass of the lower limb(s) could provide accurate and reliable information concerning reciprocal llexionlextension muscle strength. Thus, the purposes of this study were (1) to establish the reliability of isometric knee flexion/extension strength measures, (2) to quantify the extent of measurement error that may occur due to gravity, (3) to describe isometric strength curves through a full range of motion for both knee flexion and extension, and (4) to describe the ratio of isometric knee flexionfextension strength through a full range of motion in healthy adults. METHODS Subjects Seventy-seven individuals volunteered for the study. Subject characteristics are presented in table 1. All subjects were healthy, free of overt degenerative disease, and had no history of significant orthopedic problems related to the knee and related musculature. Subjects received information regarding the testing protocol Arch
Phys
Med
Rehabil
Vol 79, August
1998
972
KNEE
Table
1: Characteristics
of the Subjects
Men (n = 39) Variable
Mean 2 SD
Age ha) Height (cm) Weight (Kg)
29.7 i 12.6 179.0 t 8.4* 78.2 + 15.4*
Range
18.0-50.0 167.3-192.4 50.7-108.2
FLEXION/EXTENSION
Women Mean + SD
27.2 + 11.3 165.3 2 8.5 60.4 2 12.8
(n = 38) Range
18.0-47.0 153.6-184.1 46.0-88.8
*ps.O5betweenmenandwomen.
and were required to read and sign an informed consent. They also completed a medical history form and a general activity questionnaire. Procedures The study protocol consisted of five testing sessions that were separated by at least 72 hours (and not more than 1 week), to allow for recovery from residual muscle soreness or loss of strength associated with each test. Before each testing session, subjects were required to complete a short questionnaire that provided information about the person’s perceived health, including information on the use of tobacco and/or other drugs, nutritional and physical activity status, and amount of sleep in the last 24 hours. Individuals who indicated that they were not feeling well or reported other factors that could influence the outcome of the tests (eg, significant alcohol use, lack of sleep, or soreness from recent exercise or earlier strength tests) were rescheduled. Knee flexion and extension strength were measured using an isometric knee flexionlextension strength testing device (fig 1)” The initial session was a practice and orientation session. Subjects were seated in the machine and the joint axis of the knees aligned with that of the machine by adjusting the position of the seat using the seat adjuster (fig 1). The lateral epicondyle of the femur was used as the reference landmark and aligned with the axis of rotation of the machine. Once this was accomplished, the subjects were secured in place with a thigh restraint. A leg restraint attached to the machine’s movement arm was also secured using adjustable straps to the subject’s lower legs (anterior and approximately midway between the knee and ankle) (fig 1). The subject’s range of motion was assessedusing an internal electronic goniometer. Then, subjects performed a dynamic warm-up for both knee flexion and extension. After the warm-up and before each isometric contraction, the subject was instructed to completely relax the limbs.
Fig 1. Knee flexion/extension torque at seven angles.
Arch
Phys
Med
Rehabil
machine
used
Vol 79, August
1998
for
testing
isometric
TORQUE,
Welsch
The amount of torque produced by gravity and other factors were recorded by a strain gauge on the machine’s movement arm and defined as a measure of the mass of the lower leg at that specific angle. The subject’s fully flexed position was chosen as the initial test angle. The subject was then instructed to gradually build up tension against the stationary movement arm for a 3- to 4-second period by slowly extending both legs. Once maximum tension was developed, subjects were encouraged to maintain this force for a brief period (1 to 2 seconds), then gradually release tension by relaxing the quadriceps muscles. Subjects were then tested at three more positions throughout the range of motion, with the last position occurring at 6” of knee flexion. Between isometric contractions, a lo-second recovery period was allowed. Once the isometric strength for the quadriceps muscles was determined, the subject rested for 5 minutes. After this period, hamstring muscle strength was measured at the same positions as the quadriceps muscles, using the procedures outlined above. During test 1 (Tl) and test 2 (T2), maximum isometric knee flexion and extension strength was measured unilaterally. The methodology used during these tests was the same as during the practice session, with the following exceptions: (1) strength was measured at seven positions (6”, 24”, 42”, 60”, 78”, 96”, and 108’ of knee flexion) throughout the subject’s range of motion; (2) during Tl, knee extension strength was assessedfor each leg before assessingknee flexion strength; and (3) during T2, knee flexion strength was assessed on each leg before assessing knee extension strength. Before Tl, the leg to be tested first (right or left) was randomized and balanced among subjects. The order of testing was reversed for T2. During test 3 (T3) and test 4 (T4), maximum isometric knee flexion and extension strength of both legs moving simultaneously was measured. The same seven angles and procedures were followed as described for Tl and T2. The isometric strength values at each angle tested were recorded as “Total Torque” (TTQ). To account for the mass of the lower limbs, torque values obtained during relaxation at each angle were either added to or subtracted from the TTQ. This corrected value was recorded as “Net Muscular Torque” (NMT). Knee flexion/extension ratios were calculated for each angle tested for both TTQ and NMT values. In addition, the knee flexionlextension ratio for peak torque throughout the range was calculated. Data Analysis Maximal voluntary isometric torque was measured in foot times pounds of torque and converted to Newton meters (Nm). Descriptive statistics (means and standard errors) were calculated for each angle of each test. Reliability analysis of the isometric knee flexion and extension strength measures was completed by calculating (1) the mean difference, (2) the Pearson product-moment correlation coefficient, and (3) the standard error of the measurement (SEM = Sydl-r) and percent of SEM (SEM/mean . loo), at each angle of the tests completed (Tl-T2 and T3-T4). The use of the standard error of the measurement was selected to reflect the variability associated with the computed regression line describing the relationship between the tests. The standard error of the measurement is a similar statistic as the standard error of the estimate, but without a clear criterion measure, and provides the added benefit of being expressed in the metric unit of the measurement. This is different from the standard error of the mean, which indicates the precision of the estimated mean among the subjects and would be inappropriate for use as a reproducibility measure. The significance of the calculated mean difference
KNEE
FLEXIONIEXTENSION
was evaluated using paired t tests. Angle-specific and peak flexionlextension ratios were calculated for each test. Tests were compared using analysis of variance (ANOVA) with repeated measures, and an alpha level of p < .05 was required for statistical significance. Strength curves were developed using mean values determined at T2 (unilateral) and T4 (bilateral). All statistical computations were performed using an SAS statistical package.20 RESULTS Isometric torque values at each angle of measurement for both the unilateral and bilateral tests are presented in tables 2 through 5. Because there were no statistically significant differences between left and right leg mean flexion and extension torque values (p > .OS) and reliability data and measurement error for men and women were similar, single leg data are presented on the combined sample for the right leg only. Knee Flexion Isometric Torque Values Unilateral mean isometric flexion torque values (Nm) at each angle of measurement for Tl and T2 are presented in table 2 and figure 2. Correlation coefficients and measurement variability (SEM and SEM%) are also presented in table 2. Mean differences between Tl and T2 were not statistically significant. Correlations for Tl and T2 (for both TTQ and NMT) were high, ranging from r = .86 to Y = .96. Measurement variability ranged from 6 to 9Nm (7% to 19%) for TTQ and 5 to 1ONm (7% to 13%) for NMT. The greatest amount of variability relative to the mean torque value was found at 108” of knee flexion (SEM% = 19.1), although this was greatly reduced after correcting for the mass of the lower leg (SEM% = 12.6). Peak isometric torque values were noted at 24” of knee flexion for both TTQ (114Nm) and NMT (104Nm). Mean isometric torque values from T2 for unilateral knee flexion were used to construct TTQ and NMT strength curves, which are presented in figure 3. The curves for both TTQ and NMT were flat, descending from 6” to 108” of knee flexion. A Table
2: Right Knee Coefficients,
Flexion Isometric and Variability JointAngle
6
SE T2 SE A SE r SEM SEM% NMT Tl SE T2 SE A SE r SEM SEM%
Table
(Degrees
973
Welsch
3: Right Knee Extension Isometric Coefficients, and Variability JointAngle
(Degreesof
60"
78"
96"
109.0 24.5
114.0 t4.4
110.0
99.0 k3.6
80.0 t2.9
57.0
108.0 54.4
114.0 54.5
99.0 1-3.6
82.0 23.0
Flexion)
24"
42"
60"
78"
96"
108"
Tl
63.0
115.0
164.0
216.0
214.0
161.0
150.0
SE T2
12.8 62.0
i4.3 111.0
25.5 166.0
27.3 217.0
1~8.0 217.0
25.6 164.0
25.2 151.0
SE A
22.8 -1.5
24.3 -3.9
k6.0 1.5
27.6 1.1
i-8.5 2.9
k6.1 2.6
15.6 1.3
SE r
21.1
i-1.9 .90
22.5 .91
23.0 .92
22.6 .95
k2.0
il.6
12
15
19
16
12
.92
.95
.96
SEM
7
SEM% NMT Tl
11.0
10.6
9.7
74.0
123.0
168.0
219.0
213.0
156.0
139.0
SE T2
23.0 73.0
24.5 119.0
k5.6 171.0
k7.4 220.0
k8.0 216.0
25.5 159.0
k4.9 141.0
8.7
7.6
10
7.6
6.5
SE
23.0
54.4
26.0
27.6
t&.5
26.0
25.2
A SE
-0.7 21.1
-3.4 21.9
3.1 i-2.5
1.4 23.0
3.0 k2.6
2.9 22.0
1.5 +I.6
12 .91
16 .91
19 .92
16 .95
12 .95
10 .96
9.6
9.2
8.5
7.5
7.7
6.9
:EM
7 .93
SEM%
9.6
ValuesforTlandT2aremean
iSE.Unitsarein
Nm.A=T2-Tl.
statistically significant difference (p < .0.5) between the TTQ and NMT values was present at all angles (except 78”) tested, with the greatest differences noted at the extreme angles in the range of motion (fig 3). Mean bilateral isometric torque values (Nm) for T3 and T4 knee Rexion are presented in table 4 and figure 4. Mean differences between T3 and T4 were not statistically significant. Correlation coefficients ranged from Y = .87 to Y = .98 for both TTQ and NMT. The measurement variability for bilateral TTQ and NMT values was similar to the unilateral tests, ranging from 10 to 15Nm (5% to 19%) and 10 to 16Nm (5% to 12%), Table
4: Knee Flexion Coefficients,
Bilateral Isometric and Variability JointAngle(Degrees
42"
Reliability
6"
of Flexion)
24"
Torque, (n = 77)
TTQ
Torque, Reliability (n = 77)
108"
TTQ Tl
TORQUE,
Torque, (n = 77)
Reliability
of Flexion)
6"
24"
42"
60"
78"
96"
108"
nQ
-1.6 t1.4
0.1 il.4 .95
.95
9
8
8.1
7.4
94.0 i4.1 92.0
104.0 24.2 104.0
24.0 -2.1 ItI.6
24.3 -0.7 21.3
.92
0.3 21.2 .96 7
10.5
8.0
1.9 il.3
.94 7
1.9 kl.l
.90
.89
176.0
124.0
65.0
27.1 211.0
26.0 179.0
24.4 127.0
23.9 65.0
SE
29.1
28.7
28.1
27.1
26.1
k4.8
24.0
A SE :EM
1.5 22.5 9 .95
2.3 t1.9 8 .95
0.5 il.9 7 .96
2.6 21.7 7 .94
2.3 22.0 8 .90
3.2 ?I.8 7 .89
0.4 i2.0 6 .86
8.1
7.4
6.6
7.6
10.2
11.9
19.1
42.0
Tl SE
192.0 28.6
214.0 i-8.4
213.0 27.7
199.0 27.0
176.0 i6.0
135.0 1-4.6
90.0 23.8
T2 SE
196.0 28.5
216.0 28.3
214.0 i7.7
202.0 27.0
179.0 26.1
138.0 25.0
90.0 23.9
63.0 22.3
i3.6 -0.2 t1.2
i3.0 2.1 21.4
64.0 22.6 1.5
k1.7 42.0 k1.7 -0.0
8
21.1 .90 7
kO.9 .88 5
10.3
11.0
12.6
.94 8.0
208.0
28.1 229.0
SEM% NMT
80.0 23.0 83.0
7.3
229.0
28.8 239.0
19.1
96.0 k3.5 95.0
8
237.0
29.4 228.0
6
103.0 23.8
.95
.86
226.0
7 11.9
7
-0.6
Tl SE T2
8 10.2
103.0 24.0 0.1
29.0 Cl.8 to.9
7.6
.95 8
0.2 21.2
22.2 59.0 12.5
6.6
k1.2
10
ValuesforTlandT2aremean
i-4.0 110.0 i-4.1
30.0 21.7
.90
kSE,UnitsareinNm,A=T2-Tl.
A SE
3.2 k2.6
2.3 21.9
0.7 il.9
2.4 11.7
:EM
10 .92
8 .95
7 .95
8 .94
SEM%
10.5
8.0
7.3
8.0
ValuesforTlandT2are
mean
Arch
f SE.Unitsarein
Phys
Med
Rehabil
2.3 il.9 8 .90 10.3 Nm.A
2.8 1~1.8 7 .90 11.0 =T2
Vol 79, August
0.0 21.8 5 .88 12.6 - Tl.
1998
974
KNEE
Table
5: Knee
Extension Coefficients,
Bilateral Isometric Torque, and Variability (n = 77) Joint
6"
24"
Angle (Degreesof 42"
FLEXION/EXTENSION
TORQUE,
between bilateral TTQ and NMT values existed (except at 78”) with the greatest differences noted at the extreme angles in the range of motion (fig 4).
Reliability
Flexion)
60"
78"
96"
1080
Knee Extension Isometric Torque Values Mean isometric torque values (Nm) for unilateral knee extension are shown in table 3 and figure 2. Mean differences between T3 and T4 were not statistically significant. Correlation coefficients for mean unilateral knee extension torque ranged from r = .90 to r = .96 and measurement variability from 7 to 19Nm (6.5% to 11%) for both TTQ and NMT. Peak isometric torque values occurred at 60” of knee flexion for both TTQ (216Nm) and NMT (219Nm). Strength curves for unilateral knee extension were constructed using mean isometric torque values from T2 and are presented in figure 3. The curves for both TTQ and NMT show a characteristic ascending and descending shape with the peak of the curve at 60” of knee flexion and lowest values at 6” and 108” of knee flexion. A statistically significant difference (p < .05) was present between TTQ and NMT values at all angles tested. However, this difference was less marked at the extremes in the range of motion in comparison to the unilateral isometric flexion tests. Bilateral mean isometric torque values (Nm) for knee extension are presented in table 5 and figure 4. Mean differences between Tl and T2 were not statistically significant. Correlation coefficients ranged from r = .93 to r = .98 for both TTQ and NMT. Measurement variability ranged from 11 to 29Nm (5.1% to 10%) and 12 to 29Nm (5.1% to 9%) for TTQ and NMT, respectively. As for the unilateral tests, peak bilateral isometric torque occurred at 60” of knee flexion for TTQ (414Nm) and at 78” NMT (452Nm). Bilateral isometric strength curves for knee extension were constructed using T4 mean values and are presented in figure 4. Although a significant difference (p < .05) at each angle of measurement between the TTQ and NMT values was present, the shape of the curves for both TTQ and NMT were similar to the unilateral strength curves. However, the magnitude of the curve was approximately two times greater for the bilateral test (fig 5).
l-m
Tl
109.0
200.0
300.0
413.0
443.0
360.0
337.0
SE T2
25.0 109.0
27.3 202.0
210.3 304.0
214.9 414.0
217.5 452.0
213.7 365.0
t11.8 333.0
210.8 0.5
214.8 2.6
+19.0 2.3
214.9 3.2
212.1 0.4
SE
A
15.1 1.5
27.8 2.3
SE :EM
21.8 9 .95
k2.9 8 .95
23.4 7 .96
24.7 7 .94
24.0 8 .90
23.0 7 .89
k2.6 6 .86
8.1
7.4
6.6
7.6
10.2
11.9
19.1
SEM% NMT Tl SE
133.0 55.5
220.0 k7.7
314.0 210.6
421.0 k15.0
442.0 t17.4
349.0 k13.4
315.0 ill.2
T2 SE
134.0 -+5.5
222.0 k8.0
319.0 +11.1
423.0 214.9
452.0 k19.0
356.0 214.7
310.0 211.4
A SE
3.2 52.0
r SEM
IO
8
7
8
SEM%
10.5
8.0
7.3
8.0
0.7
2.3 k2.8
.92
2.4 t4.7
k3.4
.95
.95
2.3 k4.0
.94
2.8 23.3
.90
0.0 k2.6
.90
.88
8
7
5
10.3
11.0
12.6
ValuesforTlandT2aremean+-SE.UnitsareinNm.A=T2-T1.
respectively. The greatest amount of variability relative to the mean torque values was found at 108” (SEM% = 19) of knee flexion, although this variability was significantly less when the mass of the leg was accounted for (SEM% = 12). As for the unilateral tests, peak isometric torque occurred at 24” of knee flexion for both TTQ (239Nm) and NMT (216Nm). Bilateral isometric strength curves for knee flexion were constructed using the mean values of T4 and are presented in figure 5. The shape of the curves for bilateral knee flexion (TTQ and NMT) were similar to that of the unilateral curves, although the magnitude of the curves was approximately two times greater. As for the unilateral tests, a significant difference @ < .05)
3‘g‘
300 FI
Welsch
Angle-Specific Ratios
11.30
0.819
0.60
0.44
0.39
0.41
0.31
a4 %oo 8 b u Ol b i -f 100 2 3Q .=
Tl (Flex NMT)
5
e
0
T2 (Flex NMT)
Tl (Ext NMT) -&J-
T2 (Ext NMT)
1
I
I
I
I
I
I
6
24
42
60
78
96
108
Degree of Knee Flexion Arch
Phys
Med
Rehabil
Vol 79, August
1998
Fig 2. Mean (GE) unilateral isometric torque (Nm) (right leg, n = 77) at each angle of measurement in Tl and T2. Knee flexionlextension ratios calculated at each specific angle. Flex, knee flexion; Ext, knee extension.
KNEE
FLEXIONIEXTENSION
TORQUE,
975
Welsch
* B E 200 G .L0 b 8 s: ; 100 BCB Fig 3. Strength curves construtted from mean (*SE) unilateral isometric torque (Nm) (right leg, n = 77) measurements in T2. Flex, knee flexion; Ext, knee extension. *p 5 .05 between TTQ and NMT (knee flexion and extension).
e
-1 9
T2 (Flex TTQ)
+
T2 (Ext TTQ)
0 24
6
60
42
96
78
108
Degree of Knee Flexion
Angle-Specific Knee FlexiodExtension Torque Ratios Unilateral and bilateral knee flexion/extension torque ratios for both TTQ and NMT are presented in table 6. In addition, NMT knee flexionlextension torque ratios are shown in figures 2 and 4. The flexionlextension torque ratios were highly dependent on the angle tested, and ranged from 1.30 (t .04) at 6” to .3 1 (i .Ol) at 108” of the range of motion for the right leg and 1.48 (i.04) at 6” to .30 (t.O1) at 108” of the range of motion for both legs. There was no significant difference in knee flexion/extension torque ratios between men and women. However, there was both a significant angle effect @ < .OS)and difference between NMT and TTQ knee flexionlextension ratios 0, < .05, except at 78”). Peak NMT and TTQ flexion/
extension ratios were also calculated and found to be .49 and 54 for the right leg and .49 and .55 for both legs (table 6). DISCUSSION This study is the first to describe angle-specific knee flexionl extension torque ratios through a full range of motion. Previous studies have addressed knee llexion!extension torque ratios using the peak torque values throughout the range for knee flexion and extension. For example, Gilliam and associates9 tested high school football players using two different angular velocities of movement (30” and 180” per second) and found knee flexionlextension torque ratios of .60 and .77, respectively. Goslin and Charteris,1o examined untrained individuals using
600
Angle-Specific Ratios 0.98
11.48
0,69
0.48
0.41
e
T3 (Flex NMT)
+
T3 (Ext NMT)
-#-=
T4 (Flex NMT)
+
T4 (Ext NMT)
24
42
60
0.40
0.30 I
Fig 4. Mean (&SE) bilateral isometric torque (Nm) (n = 77) at each angle of measurement in T3 and T4. Knee flexion/extension ratios calculated at each specific angle. Flex, knee flexion; Ext, knee extension.
78
Degree of Knee Flexion Arch
Phys
Med
Rehabil
Vol
79, August
1998
976
KNEE
01
FLEXION/EXTENSION
TORQUE,
-8-
T4 (Flex TTQ)
-I+
T4 (Ext TTQ)
e
T4 (Flex NMT)
+
T4 (Ext NMT)
I
24
6
42
60
78
Degree of Knee Flexion an angular velocity of 30” per second and reported a knee flexionlextension torque ratio of .44. Other studies report knee flexion/extension torque ratios ranging from .50 to .80.2-16 Traditionally, the most accepted angle used to represent the knee flexionlextension torque ratio has been 60”. It is clear from these studies that the knee llexionlextension torque ratio increases at higher angular velocities of movement5 Although the exact reason for the changing flexion/extension torque ratio is not fully understood, several possible reasons have been suggested: (1) a decline in quadriceps activity relative to hamstrings at higher speedsof movement; (2) a shift in position Table
6: Knee
Flexion/Extension
Ratios
Joint Angle (Degrees 6”
(n = 77)
of Flexion)
24’
42”
60”
78”
96”
108”
1.05* t.02
.67* 2.01
.46* 2.01
.39* 2.01
.37* k.01
.20* 2.01
1.21” 2.02
.41*
+.02
t.O1
.36* 2.01
.20* 2.01
1.30 Ifi.04 .49 (2.01)
.87 2.02
2.01
k.01
.39 k.01
.41 2.01
.31 2.01
1.48 t.04
.98 2.02
.68 2.01
A8 +.01
.41 2.01
.40 +.01
.30 2.01
Unilateral (right leg) TTQ Mean SE
1.84* 2.06 .54 (2.01)
Peak = Bilateral lTQ Mean SE
2.23” 2.08
Peak = Unilateral
.77*
.52* Ik.01
.55 (t.01)
(right led NMT Mean SE Peak = Bilateral Mean SE Peak =
Phys
44
NMT
.49 (201)
Values are mean * p 5 .05 between
Arch
.90
Med
-+- SE. NMT
Rehabil
and TTQ
knee
Vol 79, August
flexion/extension
1998
ratios.
Welsch
I
I
96
108
Fig 5. Strength curves constructed from mean (%SE) bi,-. . lateral lsometrlc torque (Nm) (n = 77) measurements in T4. Flex, knee flexion; Ext. knee extension. *p 5 .05 between TTQ and NMT (knee flexion and extension).
in the range of motion at which peak torque occurs; and (3) the fact that peak torque for flexion and extension can occur at different positions in the range of motion depending on the mode of testing used (isokinetic, isometric, or isotonic).5 Although the peak torque ratios (table 6) from this study are similar to those found in studies using low angular velocities of movement,5 the use of such ratios is felt to be an inaccurate evaluation of muscle balance.13 A comparison of balance between reciprocal muscle groups should be made at the position of the joint about which they act.15This study supports this concept because the data clearly indicate that knee flexion and extension strength vary significantly throughout the range of motion. Thus, ratios based on peak torque throughout the range provide erroneous information about reciprocal muscle balance at a given angle. Failure to account for the effect of gravity also results in misleading and erroneous conclusions about muscular balance and function.7 Fillyaw and colleagues7 reported that failure to consider “gravity effect torque” seriously underestimates actual quadriceps femoris torque and overestimates hamstring torque at both slow and fast isokinetic speeds and thus leads to incorrect assessment of the muscle’s true ability to develop torque. Data from our study also show the importance of correcting for the mass of the lower leg during isometric test conditions as indicated by the statistically significant differences at each angle of measurement between the TTQ and NMT values. It is important to note that the mass of the lower leg obtained at each angle of measurement is either added to or subtracted from the recorded torque produced by the muscle groups. This is evident from figures 3 and 5, where the isometric strength curve for knee flexion becomes slightly more horizontal after correcting for the mass of the lower leg. The differences between TTQ and NMT were most noticeable at 6”, 24”, and 108” of the range of motion for both unilateral and bilateral tests. Although the magnitude of the differences is small and only clearly evident at the extremes of the range of motion (figs 2 and 4), it is possible that the effects of gravity and other involuntary forces during muscle strength testing could have a greater impact on reciprocal muscle group torque ratios in clinical populations who have limited range of motion,
KNEE
FLEXION/EXTENSION
postsurgical edema, and/or muscle atrophy. In order for therapists to prescribe and subsequently evaluate conditioning or postinjury rehabilitation programs, they must understand those factors that may impact the relationship between the hamstrings and quadriceps femoris muscles. Findings from our study further extend work by Fillyaw, who only considered the effect of gravity at the angle of knee flexion at which peak torque occurred. Our study considered the effect of gravity at specific angles in the range of motion, thereby providing more detailed information about this potential measurement error. Yet our conclusion is similar to Fillyaw’s, demonstrating the need to seriously consider the effects of gravity and other involuntary forces when evaluating the relationship of hamstrings and quadriceps femoris muscle strength. We recognize that our data do not adequately address whether failure to include the mass of the limb is clinically important. Future studies involving serial measurements are needed to further address this question. Reliability coefficients for multiple angle isometric torque measures were high for both knee flexion/extension. Although few data are available on the reliability of repeated measurements at multiple angles for knee extension/flexion, results from this study are in agreement with previous work from our laboratory. 17-19For example, Graves and coworkers’* reported high reliability coefficients for repeated measurements of isometric strength at multiple joint angles (r = .86 -.95) in a study that looked at the specificity of limited range of motion resistance training. The use of an initial practice and orientation session may have helped avoid a significant learning effect, as reported by other investigators. 3,21.22 The high reliability coefficients noted at all angles confirms that when the testing position is carefully standardized, repeated isometric strength measurements at multiple angles of knee flexion are possible. Variability of multiple angle measurements for isometric knee extension strength generally range from 10% to 20% when the reported measurement error is expressed as a percentage of the observed mean torque value. l8 Variability for knee extension in this study is similar to what has been previously reported for multiple joint angle measurements.i* Variability was greatest at the most extreme angles in the range of motion for both flexion and extension. This may be partially explained by the difficulty in generating torque when the leg is fully extended or flexed. However, relative variability was significantly lower after accounting for the mass of the lower leg for both knee flexion and extension (unilateral and bilateral). Thus, correcting for the effect of gravity before multiple joint angle isometric strength testing improves reliability. The ascending-descending shape of the strength curve for knee extension is in agreement with previous studies.18,23,24 The joint angle at which the peak isometric strength measurement occurred is also consistent with other studies using isometric testing.18,23Studies using isokinetic testing modes report that the maximum torque position is affected by the angular velocity of movement.25 Higher velocity movements produce maximum torque at higher angles of knee flexion for both knee flexion and extension.4.24 The implications from this study are that the use of anglespecific knee flexionlextension ratios through an individual’s normal range of joint movement may provide clinicians with a more precise method of evaluating muscle imbalance. Furthermore, the evaluation of angle-specific knee tlexionlextension ratios in the uninjured limb may further characterize the existing deficits in the injured limb. This may provide therapists with an added measure of the patient’s progress after knee lesions and treatments.
TORQUE,
977
Welsch
In summary, isometric testing of knee flexion and extension strength, using standardized test angles and accounting for the mass of the lower leg, provides an accurate and reliable assessmentof leg strength. The results of this study emphasize the importance of correcting for the mass of the lower leg when assessing muscle function. The accurate measure of anglespecific knee flexion/extension strength allowed the calculation of angle-specific knee flexionlextension ratios, which were found to vary greatly through the range of motion, Anglespecific knee flexionlextension ratios should provide clinicians with a more precise method of evaluating muscle imbalance through an individual’s normal range of joint movement. Such information may subsequently result in the development of more appropriate treatment strategies for patients recovering from knee injuries or muscle weakness. References 1. Abdo J. Prevention through weight training-the balanced approach. Nat1 Strength Conditioning Assoc J 1985;6:30-1. 2. Agre J, Baxter T. Musculoskeletal profile of male collegiate soccer players. Arch Phys Med Rehabil 1987;68:147-50. 3. Kelsey J, White A, Pastides H, Bisbee G. The impact of musculoskeletal disorders on the population of the United States. J Bone Joint Surg Am 1979;61A:959-64. 4. Campbell D, Glenn W. Rehabilitation of knee extensor and flexor muscle strength in patients with meniscectomies, ligamentous repairs and chondromalacia. Phys Ther 1982;62: 10-5. 5. Baltzopoulos V, Brodie D. Isokinetic dynamometry: applications and limitations. Sports Med 1989;8:101-16. 6. Davies J, Kirkendall T, Leigh H, Lai L, Reinbold T, Wilson P. Isokinetic characteristics of professional football players: normative data between quadriceps and hamstrings muscle groups and relative to body weight. Med Sci Sports Exert 1981;13:76-7. I. Fillyaw M, Bevins T, Femandez L. Importance of correcting isokinetic peak torque for the effect of gravity when calculating knee flexor to extensor muscle ratios. Phys Ther 1986;66:23-30. 8. Fry A, Powell D. Hamstring/quadricep parity with three different training methods. J Sports Med 1987;27:362-7. 9. Gilliam T, Sady S, Freedson P, Vilanaci J. Isokinetic torque levels for high school football players. Arch Phys Med Rehabil 1979;60: 110-4. 10. Goslin B, Charter-is J. Isokinetic dynamometry: normative data for clinical use in lower extremity (knee cases). Stand J Rehabil Med 1979;11:105-9. 11. Kannus P, Yashuda K. Value of isokinetic angle-specific torque measurements in normal and injured knees. Med Sci Sports Exert 1992;24:292-7. 12. Nunn K, Mayhew J. Comparison of three methods of assessing strength imbalances at the knee. J Orthop Sports Phys Ther 1988;10:134-7. 13. Prietto C, Caiozzo V. The in vivo force-velocity relationship of the knee flexors and extensors. Am J Sports Med 1989;17:607-11. 14. Scudder G. Torque curves produced at the knee during isometric and isokinetic exercise. Arch Phys Med Rehabil 1980;61:68-73. 15. Williams M, Stutzman L. Strength variation through the range of joint motion. Phys Ther Rev 1959;39:145-52. 16. Wyatt M, Edwards A. Comparison of quadriceps and hamstring torque values during isokinetic exercise. J Orthop Sports Ther 1981;3:48-56. 17. Graves J, Pollock M, Carpenter D, Leggett S, Jones A, MacMillan M, et al. Quantitative assessment of full range-of-motion isometric lumbar extension strength. Spine 1990;15:289-94. 18. Graves J, Pollock M, Jones A, Colum A, Leggett S. Specificity of limited range of motion variable resistance training. Med Sci Sports Exert 1989;21:84-9. 19. Risch S, Norvell N, Pollock M, Risch E, Langer H, Fulton M, et al. Lumbar strengthening in chronic back pain-patients: psychological and physiological benefits. Spine 1993; 18: l-7. Arch
Phys
Med
Rehabil
Vol 79, August
1998
978
KNEE
FLEXIONIEXTENSION
20. SAS Institute Inc. SAS user’s guide: Statistics, version 5 edition. Cary (NC): SAS Institute Inc.; 1985. 21. Knapik J, Mawdsley R, Ramos N. Angular specificity and test mode specificity of isometric and isokinetic strength training. J Orthop Sports Phys Ther 1983;5:58-65. 22. Kroll W. Reliability variations of strength in test-retest situations. Res Q Exert Sport 1963;34:50-5. 23. Kulig K, Andrews J, Hay J. Human strength curves. Exert Sport Sci Annu Rev 1984;12:417-66.
Arch
Phys
Med
Rehebil
Vol 79, August
1998
TORQUE,
Welsch
24. Osternig L. Isokinetic dynamometry: implications for muscle testing and rehabilitation. Exert Sport Sci Annu Rev 1986;14:4580. 25. Thorstensson A, Grimby G, Karlsson J. Force-velocity relations and fibre composition in human knee extensor muscles. J Appl Physiol 1976;40: 12-6. Supplier a. MedX Corporation, 1401 NE 77th Street, Ocala, FL 34479.
Quantification of Full-Range-of-Motion Unilateral and Bilateral Knee Flexion and Extension Torque Ratios Michael A. Welsch, PhD, Patty A. Williams, MS, Michael L. Pollock, PhD, James E. Graves, PhD, Daniel N. Foster, MS, Michael N. F&ton, MD ABSTRACT. Welsch MA, Williams PA, Pollock ML, Graves JE, Foster DN, Fulton MN. Quantification of full-range-ofmotion unilateral and bilateral knee flexion and extension torque ratios. Arch Phys Med Rehabil 1998;79:971-8. Objectives: To evaluate the reliability and variability of repeated measurements of isometric knee flexion and extension strength, to quantify the extent of measurement error that may occur due to gravity, and to quantify isometric knee flexion/ extension torque ratios at multiple angles through a full range of motion. Design: Reliability assessment. Setting: A university exercise center. Participants: Seventy-seven healthy men and women recruited from a university and surrounding community. Intervention: Isometric knee flexion and extension strength tests. Main Outcome Measures: Knee flexionlextension strength was measured at 6”, 24”, 42”, 60”, 78”, 96”, and 108” of knee flexion. Before each contraction, subjects were instructed to completely relax the limbs to measure the mass of the lower leg. Torque values obtained during relaxation at each angle were added to or subtracted from “Total Torque” (TTQ) at peak exertion. The adjusted value was recorded as “Net Muscular Torque” (NMT) . Results: Reliability for the unilateral and bilateral tests was high (Y = .88 to Y = .98) and measurement variability low (SEM% = 5.1% to 12.6%). There was a statistically significant difference at each angle of measurement between the TTQ and NMT values for both knee flexion and extension. Knee flexionlextension ratios were highly dependent on the angle tested, ranging from 1.30 (at 6”) to .31 (at 108”). Conclusions: Isometric testing, using standardized angles, can reliably quantify knee flexionlextension strength. Furthermore, these findings emphasize the importance of correcting for the mass of the lower leg when assessing muscle function. Angle-specific knee flexionlextension torque ratios should provide clinicians with a more precise method of evaluating muscular balance (imbalance) throughout the range of motion. o 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Center for Exercise Science, College of Medicine and College of Health and Human Performance, University of Florida, Gainesville (Drs. Welsch, Pollock, Graves. Fulton. Ms. Williams. Mr. Foster): the Deoartment of Kinesioloev. College of Educa&n, Louisiana State Ur&rsity, B&I Rouge (Dr. Welsch); and the&par&&t of Health and Physical Education, Syracuse University, Syracuse, NY (Dr. Graves). Submitted for publication July 3,1997. Accepted in revised form February 20. 1998. No commercial party having a direct or indirect interest in the subject matter of this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Michael A. Welsch, PhD, 112 Huey P. Long Field House, Department of Kinesiology, Louisiana State University, Baton Rouge, LA 70803. @ 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003.9993/98/7908-456653.00/O
A
N IMBALANCE IN force-generating capacity between reciprocal muscle groups can increase the likelihood of injury to a joint. l-3Thus, reciprocal muscle group strength ratios provide clinicians with valuable information regarding muscular balance or imbalance around a joint. The relationship between knee flexor and extensor strength has been of particular interest to many investigators because of the complexity of, and the incidence of injury to, the structures that make up the knee joint. Campbell and Glenn4 have suggested that the flexion/extension strength ratios may be more important in the assessment of muscle function than the measurement of absolute strength. Most studies reporting flexion/extension ratios have used peak torque isokinetic strength testing data and have reported ratios between .55 and .80 (mean ratio, .60).2~3~5-16 It is evident from these studies that knee flexion/extension ratios are highly dependent on the angular velocity of the movement, the subject’s age, the subject’s gender, and status of training.5 Generally, knee flexionlextension ratios are determined from a single peak torque value throughout the range.5 Muscular strength varies greatly with joint position, however, as a result of changes in mechanical advantage and the length of the muscle. In addition, many studies have failed to correct for the effect of gravity on the measurement. Failure to consider this effect may have resulted in erroneous conclusions about muscular balance and function6 Thus, flexion/extension ratios based on a single peak torque value alone provide limited information regarding muscular balance. Recent research from our laboratory has shown that isometric testing using standardized angles and precise positioning of the muscle group of interest is necessary to ensure adequate reproducibility of repeated tests. This provides a safe and accurate means of evaluating isolated muscle strength through a full range of motion in both healthy and clinical populations.i7,ix We hypothesized that isometric knee flexion and extension (both unilateral and bilateral) testing, using standardized test angles and accounting for the mass of the lower limb(s) could provide accurate and reliable information concerning reciprocal llexionlextension muscle strength. Thus, the purposes of this study were (1) to establish the reliability of isometric knee flexion/extension strength measures, (2) to quantify the extent of measurement error that may occur due to gravity, (3) to describe isometric strength curves through a full range of motion for both knee flexion and extension, and (4) to describe the ratio of isometric knee flexionfextension strength through a full range of motion in healthy adults. METHODS Subjects Seventy-seven individuals volunteered for the study. Subject characteristics are presented in table 1. All subjects were healthy, free of overt degenerative disease, and had no history of significant orthopedic problems related to the knee and related musculature. Subjects received information regarding the testing protocol Arch
Phys
Med
Rehabil
Vol 79, August
1998
972
KNEE
Table
1: Characteristics
of the Subjects
Men (n = 39) Variable
Mean 2 SD
Age ha) Height (cm) Weight (Kg)
29.7 i 12.6 179.0 t 8.4* 78.2 + 15.4*
Range
18.0-50.0 167.3-192.4 50.7-108.2
FLEXION/EXTENSION
Women Mean + SD
27.2 + 11.3 165.3 2 8.5 60.4 2 12.8
(n = 38) Range
18.0-47.0 153.6-184.1 46.0-88.8
*ps.O5betweenmenandwomen.
and were required to read and sign an informed consent. They also completed a medical history form and a general activity questionnaire. Procedures The study protocol consisted of five testing sessions that were separated by at least 72 hours (and not more than 1 week), to allow for recovery from residual muscle soreness or loss of strength associated with each test. Before each testing session, subjects were required to complete a short questionnaire that provided information about the person’s perceived health, including information on the use of tobacco and/or other drugs, nutritional and physical activity status, and amount of sleep in the last 24 hours. Individuals who indicated that they were not feeling well or reported other factors that could influence the outcome of the tests (eg, significant alcohol use, lack of sleep, or soreness from recent exercise or earlier strength tests) were rescheduled. Knee flexion and extension strength were measured using an isometric knee flexionlextension strength testing device (fig 1)” The initial session was a practice and orientation session. Subjects were seated in the machine and the joint axis of the knees aligned with that of the machine by adjusting the position of the seat using the seat adjuster (fig 1). The lateral epicondyle of the femur was used as the reference landmark and aligned with the axis of rotation of the machine. Once this was accomplished, the subjects were secured in place with a thigh restraint. A leg restraint attached to the machine’s movement arm was also secured using adjustable straps to the subject’s lower legs (anterior and approximately midway between the knee and ankle) (fig 1). The subject’s range of motion was assessedusing an internal electronic goniometer. Then, subjects performed a dynamic warm-up for both knee flexion and extension. After the warm-up and before each isometric contraction, the subject was instructed to completely relax the limbs.
Fig 1. Knee flexion/extension torque at seven angles.
Arch
Phys
Med
Rehabil
machine
used
Vol 79, August
1998
for
testing
isometric
TORQUE,
Welsch
The amount of torque produced by gravity and other factors were recorded by a strain gauge on the machine’s movement arm and defined as a measure of the mass of the lower leg at that specific angle. The subject’s fully flexed position was chosen as the initial test angle. The subject was then instructed to gradually build up tension against the stationary movement arm for a 3- to 4-second period by slowly extending both legs. Once maximum tension was developed, subjects were encouraged to maintain this force for a brief period (1 to 2 seconds), then gradually release tension by relaxing the quadriceps muscles. Subjects were then tested at three more positions throughout the range of motion, with the last position occurring at 6” of knee flexion. Between isometric contractions, a lo-second recovery period was allowed. Once the isometric strength for the quadriceps muscles was determined, the subject rested for 5 minutes. After this period, hamstring muscle strength was measured at the same positions as the quadriceps muscles, using the procedures outlined above. During test 1 (Tl) and test 2 (T2), maximum isometric knee flexion and extension strength was measured unilaterally. The methodology used during these tests was the same as during the practice session, with the following exceptions: (1) strength was measured at seven positions (6”, 24”, 42”, 60”, 78”, 96”, and 108’ of knee flexion) throughout the subject’s range of motion; (2) during Tl, knee extension strength was assessedfor each leg before assessingknee flexion strength; and (3) during T2, knee flexion strength was assessed on each leg before assessing knee extension strength. Before Tl, the leg to be tested first (right or left) was randomized and balanced among subjects. The order of testing was reversed for T2. During test 3 (T3) and test 4 (T4), maximum isometric knee flexion and extension strength of both legs moving simultaneously was measured. The same seven angles and procedures were followed as described for Tl and T2. The isometric strength values at each angle tested were recorded as “Total Torque” (TTQ). To account for the mass of the lower limbs, torque values obtained during relaxation at each angle were either added to or subtracted from the TTQ. This corrected value was recorded as “Net Muscular Torque” (NMT). Knee flexion/extension ratios were calculated for each angle tested for both TTQ and NMT values. In addition, the knee flexionlextension ratio for peak torque throughout the range was calculated. Data Analysis Maximal voluntary isometric torque was measured in foot times pounds of torque and converted to Newton meters (Nm). Descriptive statistics (means and standard errors) were calculated for each angle of each test. Reliability analysis of the isometric knee flexion and extension strength measures was completed by calculating (1) the mean difference, (2) the Pearson product-moment correlation coefficient, and (3) the standard error of the measurement (SEM = Sydl-r) and percent of SEM (SEM/mean . loo), at each angle of the tests completed (Tl-T2 and T3-T4). The use of the standard error of the measurement was selected to reflect the variability associated with the computed regression line describing the relationship between the tests. The standard error of the measurement is a similar statistic as the standard error of the estimate, but without a clear criterion measure, and provides the added benefit of being expressed in the metric unit of the measurement. This is different from the standard error of the mean, which indicates the precision of the estimated mean among the subjects and would be inappropriate for use as a reproducibility measure. The significance of the calculated mean difference
KNEE
FLEXIONIEXTENSION
was evaluated using paired t tests. Angle-specific and peak flexionlextension ratios were calculated for each test. Tests were compared using analysis of variance (ANOVA) with repeated measures, and an alpha level of p < .05 was required for statistical significance. Strength curves were developed using mean values determined at T2 (unilateral) and T4 (bilateral). All statistical computations were performed using an SAS statistical package.20 RESULTS Isometric torque values at each angle of measurement for both the unilateral and bilateral tests are presented in tables 2 through 5. Because there were no statistically significant differences between left and right leg mean flexion and extension torque values (p > .OS) and reliability data and measurement error for men and women were similar, single leg data are presented on the combined sample for the right leg only. Knee Flexion Isometric Torque Values Unilateral mean isometric flexion torque values (Nm) at each angle of measurement for Tl and T2 are presented in table 2 and figure 2. Correlation coefficients and measurement variability (SEM and SEM%) are also presented in table 2. Mean differences between Tl and T2 were not statistically significant. Correlations for Tl and T2 (for both TTQ and NMT) were high, ranging from r = .86 to Y = .96. Measurement variability ranged from 6 to 9Nm (7% to 19%) for TTQ and 5 to 1ONm (7% to 13%) for NMT. The greatest amount of variability relative to the mean torque value was found at 108” of knee flexion (SEM% = 19.1), although this was greatly reduced after correcting for the mass of the lower leg (SEM% = 12.6). Peak isometric torque values were noted at 24” of knee flexion for both TTQ (114Nm) and NMT (104Nm). Mean isometric torque values from T2 for unilateral knee flexion were used to construct TTQ and NMT strength curves, which are presented in figure 3. The curves for both TTQ and NMT were flat, descending from 6” to 108” of knee flexion. A Table
2: Right Knee Coefficients,
Flexion Isometric and Variability JointAngle
6
SE T2 SE A SE r SEM SEM% NMT Tl SE T2 SE A SE r SEM SEM%
Table
(Degrees
973
Welsch
3: Right Knee Extension Isometric Coefficients, and Variability JointAngle
(Degreesof
60"
78"
96"
109.0 24.5
114.0 t4.4
110.0
99.0 k3.6
80.0 t2.9
57.0
108.0 54.4
114.0 54.5
99.0 1-3.6
82.0 23.0
Flexion)
24"
42"
60"
78"
96"
108"
Tl
63.0
115.0
164.0
216.0
214.0
161.0
150.0
SE T2
12.8 62.0
i4.3 111.0
25.5 166.0
27.3 217.0
1~8.0 217.0
25.6 164.0
25.2 151.0
SE A
22.8 -1.5
24.3 -3.9
k6.0 1.5
27.6 1.1
i-8.5 2.9
k6.1 2.6
15.6 1.3
SE r
21.1
i-1.9 .90
22.5 .91
23.0 .92
22.6 .95
k2.0
il.6
12
15
19
16
12
.92
.95
.96
SEM
7
SEM% NMT Tl
11.0
10.6
9.7
74.0
123.0
168.0
219.0
213.0
156.0
139.0
SE T2
23.0 73.0
24.5 119.0
k5.6 171.0
k7.4 220.0
k8.0 216.0
25.5 159.0
k4.9 141.0
8.7
7.6
10
7.6
6.5
SE
23.0
54.4
26.0
27.6
t&.5
26.0
25.2
A SE
-0.7 21.1
-3.4 21.9
3.1 i-2.5
1.4 23.0
3.0 k2.6
2.9 22.0
1.5 +I.6
12 .91
16 .91
19 .92
16 .95
12 .95
10 .96
9.6
9.2
8.5
7.5
7.7
6.9
:EM
7 .93
SEM%
9.6
ValuesforTlandT2aremean
iSE.Unitsarein
Nm.A=T2-Tl.
statistically significant difference (p < .0.5) between the TTQ and NMT values was present at all angles (except 78”) tested, with the greatest differences noted at the extreme angles in the range of motion (fig 3). Mean bilateral isometric torque values (Nm) for T3 and T4 knee Rexion are presented in table 4 and figure 4. Mean differences between T3 and T4 were not statistically significant. Correlation coefficients ranged from Y = .87 to Y = .98 for both TTQ and NMT. The measurement variability for bilateral TTQ and NMT values was similar to the unilateral tests, ranging from 10 to 15Nm (5% to 19%) and 10 to 16Nm (5% to 12%), Table
4: Knee Flexion Coefficients,
Bilateral Isometric and Variability JointAngle(Degrees
42"
Reliability
6"
of Flexion)
24"
Torque, (n = 77)
TTQ
Torque, Reliability (n = 77)
108"
TTQ Tl
TORQUE,
Torque, (n = 77)
Reliability
of Flexion)
6"
24"
42"
60"
78"
96"
108"
nQ
-1.6 t1.4
0.1 il.4 .95
.95
9
8
8.1
7.4
94.0 i4.1 92.0
104.0 24.2 104.0
24.0 -2.1 ItI.6
24.3 -0.7 21.3
.92
0.3 21.2 .96 7
10.5
8.0
1.9 il.3
.94 7
1.9 kl.l
.90
.89
176.0
124.0
65.0
27.1 211.0
26.0 179.0
24.4 127.0
23.9 65.0
SE
29.1
28.7
28.1
27.1
26.1
k4.8
24.0
A SE :EM
1.5 22.5 9 .95
2.3 t1.9 8 .95
0.5 il.9 7 .96
2.6 21.7 7 .94
2.3 22.0 8 .90
3.2 ?I.8 7 .89
0.4 i2.0 6 .86
8.1
7.4
6.6
7.6
10.2
11.9
19.1
42.0
Tl SE
192.0 28.6
214.0 i-8.4
213.0 27.7
199.0 27.0
176.0 i6.0
135.0 1-4.6
90.0 23.8
T2 SE
196.0 28.5
216.0 28.3
214.0 i7.7
202.0 27.0
179.0 26.1
138.0 25.0
90.0 23.9
63.0 22.3
i3.6 -0.2 t1.2
i3.0 2.1 21.4
64.0 22.6 1.5
k1.7 42.0 k1.7 -0.0
8
21.1 .90 7
kO.9 .88 5
10.3
11.0
12.6
.94 8.0
208.0
28.1 229.0
SEM% NMT
80.0 23.0 83.0
7.3
229.0
28.8 239.0
19.1
96.0 k3.5 95.0
8
237.0
29.4 228.0
6
103.0 23.8
.95
.86
226.0
7 11.9
7
-0.6
Tl SE T2
8 10.2
103.0 24.0 0.1
29.0 Cl.8 to.9
7.6
.95 8
0.2 21.2
22.2 59.0 12.5
6.6
k1.2
10
ValuesforTlandT2aremean
i-4.0 110.0 i-4.1
30.0 21.7
.90
kSE,UnitsareinNm,A=T2-Tl.
A SE
3.2 k2.6
2.3 21.9
0.7 il.9
2.4 11.7
:EM
10 .92
8 .95
7 .95
8 .94
SEM%
10.5
8.0
7.3
8.0
ValuesforTlandT2are
mean
Arch
f SE.Unitsarein
Phys
Med
Rehabil
2.3 il.9 8 .90 10.3 Nm.A
2.8 1~1.8 7 .90 11.0 =T2
Vol 79, August
0.0 21.8 5 .88 12.6 - Tl.
1998
974
KNEE
Table
5: Knee
Extension Coefficients,
Bilateral Isometric Torque, and Variability (n = 77) Joint
6"
24"
Angle (Degreesof 42"
FLEXION/EXTENSION
TORQUE,
between bilateral TTQ and NMT values existed (except at 78”) with the greatest differences noted at the extreme angles in the range of motion (fig 4).
Reliability
Flexion)
60"
78"
96"
1080
Knee Extension Isometric Torque Values Mean isometric torque values (Nm) for unilateral knee extension are shown in table 3 and figure 2. Mean differences between T3 and T4 were not statistically significant. Correlation coefficients for mean unilateral knee extension torque ranged from r = .90 to r = .96 and measurement variability from 7 to 19Nm (6.5% to 11%) for both TTQ and NMT. Peak isometric torque values occurred at 60” of knee flexion for both TTQ (216Nm) and NMT (219Nm). Strength curves for unilateral knee extension were constructed using mean isometric torque values from T2 and are presented in figure 3. The curves for both TTQ and NMT show a characteristic ascending and descending shape with the peak of the curve at 60” of knee flexion and lowest values at 6” and 108” of knee flexion. A statistically significant difference (p < .05) was present between TTQ and NMT values at all angles tested. However, this difference was less marked at the extremes in the range of motion in comparison to the unilateral isometric flexion tests. Bilateral mean isometric torque values (Nm) for knee extension are presented in table 5 and figure 4. Mean differences between Tl and T2 were not statistically significant. Correlation coefficients ranged from r = .93 to r = .98 for both TTQ and NMT. Measurement variability ranged from 11 to 29Nm (5.1% to 10%) and 12 to 29Nm (5.1% to 9%) for TTQ and NMT, respectively. As for the unilateral tests, peak bilateral isometric torque occurred at 60” of knee flexion for TTQ (414Nm) and at 78” NMT (452Nm). Bilateral isometric strength curves for knee extension were constructed using T4 mean values and are presented in figure 4. Although a significant difference (p < .05) at each angle of measurement between the TTQ and NMT values was present, the shape of the curves for both TTQ and NMT were similar to the unilateral strength curves. However, the magnitude of the curve was approximately two times greater for the bilateral test (fig 5).
l-m
Tl
109.0
200.0
300.0
413.0
443.0
360.0
337.0
SE T2
25.0 109.0
27.3 202.0
210.3 304.0
214.9 414.0
217.5 452.0
213.7 365.0
t11.8 333.0
210.8 0.5
214.8 2.6
+19.0 2.3
214.9 3.2
212.1 0.4
SE
A
15.1 1.5
27.8 2.3
SE :EM
21.8 9 .95
k2.9 8 .95
23.4 7 .96
24.7 7 .94
24.0 8 .90
23.0 7 .89
k2.6 6 .86
8.1
7.4
6.6
7.6
10.2
11.9
19.1
SEM% NMT Tl SE
133.0 55.5
220.0 k7.7
314.0 210.6
421.0 k15.0
442.0 t17.4
349.0 k13.4
315.0 ill.2
T2 SE
134.0 -+5.5
222.0 k8.0
319.0 +11.1
423.0 214.9
452.0 k19.0
356.0 214.7
310.0 211.4
A SE
3.2 52.0
r SEM
IO
8
7
8
SEM%
10.5
8.0
7.3
8.0
0.7
2.3 k2.8
.92
2.4 t4.7
k3.4
.95
.95
2.3 k4.0
.94
2.8 23.3
.90
0.0 k2.6
.90
.88
8
7
5
10.3
11.0
12.6
ValuesforTlandT2aremean+-SE.UnitsareinNm.A=T2-T1.
respectively. The greatest amount of variability relative to the mean torque values was found at 108” (SEM% = 19) of knee flexion, although this variability was significantly less when the mass of the leg was accounted for (SEM% = 12). As for the unilateral tests, peak isometric torque occurred at 24” of knee flexion for both TTQ (239Nm) and NMT (216Nm). Bilateral isometric strength curves for knee flexion were constructed using the mean values of T4 and are presented in figure 5. The shape of the curves for bilateral knee flexion (TTQ and NMT) were similar to that of the unilateral curves, although the magnitude of the curves was approximately two times greater. As for the unilateral tests, a significant difference @ < .05)
3‘g‘
300 FI
Welsch
Angle-Specific Ratios
11.30
0.819
0.60
0.44
0.39
0.41
0.31
a4 %oo 8 b u Ol b i -f 100 2 3Q .=
Tl (Flex NMT)
5
e
0
T2 (Flex NMT)
Tl (Ext NMT) -&J-
T2 (Ext NMT)
1
I
I
I
I
I
I
6
24
42
60
78
96
108
Degree of Knee Flexion Arch
Phys
Med
Rehabil
Vol 79, August
1998
Fig 2. Mean (GE) unilateral isometric torque (Nm) (right leg, n = 77) at each angle of measurement in Tl and T2. Knee flexionlextension ratios calculated at each specific angle. Flex, knee flexion; Ext, knee extension.
KNEE
FLEXIONIEXTENSION
TORQUE,
975
Welsch
* B E 200 G .L0 b 8 s: ; 100 BCB Fig 3. Strength curves construtted from mean (*SE) unilateral isometric torque (Nm) (right leg, n = 77) measurements in T2. Flex, knee flexion; Ext, knee extension. *p 5 .05 between TTQ and NMT (knee flexion and extension).
e
-1 9
T2 (Flex TTQ)
+
T2 (Ext TTQ)
0 24
6
60
42
96
78
108
Degree of Knee Flexion
Angle-Specific Knee FlexiodExtension Torque Ratios Unilateral and bilateral knee flexion/extension torque ratios for both TTQ and NMT are presented in table 6. In addition, NMT knee flexionlextension torque ratios are shown in figures 2 and 4. The flexionlextension torque ratios were highly dependent on the angle tested, and ranged from 1.30 (t .04) at 6” to .3 1 (i .Ol) at 108” of the range of motion for the right leg and 1.48 (i.04) at 6” to .30 (t.O1) at 108” of the range of motion for both legs. There was no significant difference in knee flexion/extension torque ratios between men and women. However, there was both a significant angle effect @ < .OS)and difference between NMT and TTQ knee flexionlextension ratios 0, < .05, except at 78”). Peak NMT and TTQ flexion/
extension ratios were also calculated and found to be .49 and 54 for the right leg and .49 and .55 for both legs (table 6). DISCUSSION This study is the first to describe angle-specific knee flexionl extension torque ratios through a full range of motion. Previous studies have addressed knee llexion!extension torque ratios using the peak torque values throughout the range for knee flexion and extension. For example, Gilliam and associates9 tested high school football players using two different angular velocities of movement (30” and 180” per second) and found knee flexionlextension torque ratios of .60 and .77, respectively. Goslin and Charteris,1o examined untrained individuals using
600
Angle-Specific Ratios 0.98
11.48
0,69
0.48
0.41
e
T3 (Flex NMT)
+
T3 (Ext NMT)
-#-=
T4 (Flex NMT)
+
T4 (Ext NMT)
24
42
60
0.40
0.30 I
Fig 4. Mean (&SE) bilateral isometric torque (Nm) (n = 77) at each angle of measurement in T3 and T4. Knee flexion/extension ratios calculated at each specific angle. Flex, knee flexion; Ext, knee extension.
78
Degree of Knee Flexion Arch
Phys
Med
Rehabil
Vol
79, August
1998
976
KNEE
01
FLEXION/EXTENSION
TORQUE,
-8-
T4 (Flex TTQ)
-I+
T4 (Ext TTQ)
e
T4 (Flex NMT)
+
T4 (Ext NMT)
I
24
6
42
60
78
Degree of Knee Flexion an angular velocity of 30” per second and reported a knee flexionlextension torque ratio of .44. Other studies report knee flexion/extension torque ratios ranging from .50 to .80.2-16 Traditionally, the most accepted angle used to represent the knee flexionlextension torque ratio has been 60”. It is clear from these studies that the knee llexionlextension torque ratio increases at higher angular velocities of movement5 Although the exact reason for the changing flexion/extension torque ratio is not fully understood, several possible reasons have been suggested: (1) a decline in quadriceps activity relative to hamstrings at higher speedsof movement; (2) a shift in position Table
6: Knee
Flexion/Extension
Ratios
Joint Angle (Degrees 6”
(n = 77)
of Flexion)
24’
42”
60”
78”
96”
108”
1.05* t.02
.67* 2.01
.46* 2.01
.39* 2.01
.37* k.01
.20* 2.01
1.21” 2.02
.41*
+.02
t.O1
.36* 2.01
.20* 2.01
1.30 Ifi.04 .49 (2.01)
.87 2.02
2.01
k.01
.39 k.01
.41 2.01
.31 2.01
1.48 t.04
.98 2.02
.68 2.01
A8 +.01
.41 2.01
.40 +.01
.30 2.01
Unilateral (right leg) TTQ Mean SE
1.84* 2.06 .54 (2.01)
Peak = Bilateral lTQ Mean SE
2.23” 2.08
Peak = Unilateral
.77*
.52* Ik.01
.55 (t.01)
(right led NMT Mean SE Peak = Bilateral Mean SE Peak =
Phys
44
NMT
.49 (201)
Values are mean * p 5 .05 between
Arch
.90
Med
-+- SE. NMT
Rehabil
and TTQ
knee
Vol 79, August
flexion/extension
1998
ratios.
Welsch
I
I
96
108
Fig 5. Strength curves constructed from mean (%SE) bi,-. . lateral lsometrlc torque (Nm) (n = 77) measurements in T4. Flex, knee flexion; Ext. knee extension. *p 5 .05 between TTQ and NMT (knee flexion and extension).
in the range of motion at which peak torque occurs; and (3) the fact that peak torque for flexion and extension can occur at different positions in the range of motion depending on the mode of testing used (isokinetic, isometric, or isotonic).5 Although the peak torque ratios (table 6) from this study are similar to those found in studies using low angular velocities of movement,5 the use of such ratios is felt to be an inaccurate evaluation of muscle balance.13 A comparison of balance between reciprocal muscle groups should be made at the position of the joint about which they act.15This study supports this concept because the data clearly indicate that knee flexion and extension strength vary significantly throughout the range of motion. Thus, ratios based on peak torque throughout the range provide erroneous information about reciprocal muscle balance at a given angle. Failure to account for the effect of gravity also results in misleading and erroneous conclusions about muscular balance and function.7 Fillyaw and colleagues7 reported that failure to consider “gravity effect torque” seriously underestimates actual quadriceps femoris torque and overestimates hamstring torque at both slow and fast isokinetic speeds and thus leads to incorrect assessment of the muscle’s true ability to develop torque. Data from our study also show the importance of correcting for the mass of the lower leg during isometric test conditions as indicated by the statistically significant differences at each angle of measurement between the TTQ and NMT values. It is important to note that the mass of the lower leg obtained at each angle of measurement is either added to or subtracted from the recorded torque produced by the muscle groups. This is evident from figures 3 and 5, where the isometric strength curve for knee flexion becomes slightly more horizontal after correcting for the mass of the lower leg. The differences between TTQ and NMT were most noticeable at 6”, 24”, and 108” of the range of motion for both unilateral and bilateral tests. Although the magnitude of the differences is small and only clearly evident at the extremes of the range of motion (figs 2 and 4), it is possible that the effects of gravity and other involuntary forces during muscle strength testing could have a greater impact on reciprocal muscle group torque ratios in clinical populations who have limited range of motion,
KNEE
FLEXION/EXTENSION
postsurgical edema, and/or muscle atrophy. In order for therapists to prescribe and subsequently evaluate conditioning or postinjury rehabilitation programs, they must understand those factors that may impact the relationship between the hamstrings and quadriceps femoris muscles. Findings from our study further extend work by Fillyaw, who only considered the effect of gravity at the angle of knee flexion at which peak torque occurred. Our study considered the effect of gravity at specific angles in the range of motion, thereby providing more detailed information about this potential measurement error. Yet our conclusion is similar to Fillyaw’s, demonstrating the need to seriously consider the effects of gravity and other involuntary forces when evaluating the relationship of hamstrings and quadriceps femoris muscle strength. We recognize that our data do not adequately address whether failure to include the mass of the limb is clinically important. Future studies involving serial measurements are needed to further address this question. Reliability coefficients for multiple angle isometric torque measures were high for both knee flexion/extension. Although few data are available on the reliability of repeated measurements at multiple angles for knee extension/flexion, results from this study are in agreement with previous work from our laboratory. 17-19For example, Graves and coworkers’* reported high reliability coefficients for repeated measurements of isometric strength at multiple joint angles (r = .86 -.95) in a study that looked at the specificity of limited range of motion resistance training. The use of an initial practice and orientation session may have helped avoid a significant learning effect, as reported by other investigators. 3,21.22 The high reliability coefficients noted at all angles confirms that when the testing position is carefully standardized, repeated isometric strength measurements at multiple angles of knee flexion are possible. Variability of multiple angle measurements for isometric knee extension strength generally range from 10% to 20% when the reported measurement error is expressed as a percentage of the observed mean torque value. l8 Variability for knee extension in this study is similar to what has been previously reported for multiple joint angle measurements.i* Variability was greatest at the most extreme angles in the range of motion for both flexion and extension. This may be partially explained by the difficulty in generating torque when the leg is fully extended or flexed. However, relative variability was significantly lower after accounting for the mass of the lower leg for both knee flexion and extension (unilateral and bilateral). Thus, correcting for the effect of gravity before multiple joint angle isometric strength testing improves reliability. The ascending-descending shape of the strength curve for knee extension is in agreement with previous studies.18,23,24 The joint angle at which the peak isometric strength measurement occurred is also consistent with other studies using isometric testing.18,23Studies using isokinetic testing modes report that the maximum torque position is affected by the angular velocity of movement.25 Higher velocity movements produce maximum torque at higher angles of knee flexion for both knee flexion and extension.4.24 The implications from this study are that the use of anglespecific knee flexionlextension ratios through an individual’s normal range of joint movement may provide clinicians with a more precise method of evaluating muscle imbalance. Furthermore, the evaluation of angle-specific knee tlexionlextension ratios in the uninjured limb may further characterize the existing deficits in the injured limb. This may provide therapists with an added measure of the patient’s progress after knee lesions and treatments.
TORQUE,
977
Welsch
In summary, isometric testing of knee flexion and extension strength, using standardized test angles and accounting for the mass of the lower leg, provides an accurate and reliable assessmentof leg strength. The results of this study emphasize the importance of correcting for the mass of the lower leg when assessing muscle function. The accurate measure of anglespecific knee flexion/extension strength allowed the calculation of angle-specific knee flexionlextension ratios, which were found to vary greatly through the range of motion, Anglespecific knee flexionlextension ratios should provide clinicians with a more precise method of evaluating muscle imbalance through an individual’s normal range of joint movement. Such information may subsequently result in the development of more appropriate treatment strategies for patients recovering from knee injuries or muscle weakness. References 1. Abdo J. Prevention through weight training-the balanced approach. Nat1 Strength Conditioning Assoc J 1985;6:30-1. 2. Agre J, Baxter T. Musculoskeletal profile of male collegiate soccer players. Arch Phys Med Rehabil 1987;68:147-50. 3. Kelsey J, White A, Pastides H, Bisbee G. The impact of musculoskeletal disorders on the population of the United States. J Bone Joint Surg Am 1979;61A:959-64. 4. Campbell D, Glenn W. Rehabilitation of knee extensor and flexor muscle strength in patients with meniscectomies, ligamentous repairs and chondromalacia. Phys Ther 1982;62: 10-5. 5. Baltzopoulos V, Brodie D. Isokinetic dynamometry: applications and limitations. Sports Med 1989;8:101-16. 6. Davies J, Kirkendall T, Leigh H, Lai L, Reinbold T, Wilson P. Isokinetic characteristics of professional football players: normative data between quadriceps and hamstrings muscle groups and relative to body weight. Med Sci Sports Exert 1981;13:76-7. I. Fillyaw M, Bevins T, Femandez L. Importance of correcting isokinetic peak torque for the effect of gravity when calculating knee flexor to extensor muscle ratios. Phys Ther 1986;66:23-30. 8. Fry A, Powell D. Hamstring/quadricep parity with three different training methods. J Sports Med 1987;27:362-7. 9. Gilliam T, Sady S, Freedson P, Vilanaci J. Isokinetic torque levels for high school football players. Arch Phys Med Rehabil 1979;60: 110-4. 10. Goslin B, Charter-is J. Isokinetic dynamometry: normative data for clinical use in lower extremity (knee cases). Stand J Rehabil Med 1979;11:105-9. 11. Kannus P, Yashuda K. Value of isokinetic angle-specific torque measurements in normal and injured knees. Med Sci Sports Exert 1992;24:292-7. 12. Nunn K, Mayhew J. Comparison of three methods of assessing strength imbalances at the knee. J Orthop Sports Phys Ther 1988;10:134-7. 13. Prietto C, Caiozzo V. The in vivo force-velocity relationship of the knee flexors and extensors. Am J Sports Med 1989;17:607-11. 14. Scudder G. Torque curves produced at the knee during isometric and isokinetic exercise. Arch Phys Med Rehabil 1980;61:68-73. 15. Williams M, Stutzman L. Strength variation through the range of joint motion. Phys Ther Rev 1959;39:145-52. 16. Wyatt M, Edwards A. Comparison of quadriceps and hamstring torque values during isokinetic exercise. J Orthop Sports Ther 1981;3:48-56. 17. Graves J, Pollock M, Carpenter D, Leggett S, Jones A, MacMillan M, et al. Quantitative assessment of full range-of-motion isometric lumbar extension strength. Spine 1990;15:289-94. 18. Graves J, Pollock M, Jones A, Colum A, Leggett S. Specificity of limited range of motion variable resistance training. Med Sci Sports Exert 1989;21:84-9. 19. Risch S, Norvell N, Pollock M, Risch E, Langer H, Fulton M, et al. Lumbar strengthening in chronic back pain-patients: psychological and physiological benefits. Spine 1993; 18: l-7. Arch
Phys
Med
Rehabil
Vol 79, August
1998
978
KNEE
FLEXIONIEXTENSION
20. SAS Institute Inc. SAS user’s guide: Statistics, version 5 edition. Cary (NC): SAS Institute Inc.; 1985. 21. Knapik J, Mawdsley R, Ramos N. Angular specificity and test mode specificity of isometric and isokinetic strength training. J Orthop Sports Phys Ther 1983;5:58-65. 22. Kroll W. Reliability variations of strength in test-retest situations. Res Q Exert Sport 1963;34:50-5. 23. Kulig K, Andrews J, Hay J. Human strength curves. Exert Sport Sci Annu Rev 1984;12:417-66.
Arch
Phys
Med
Rehebil
Vol 79, August
1998
TORQUE,
Welsch
24. Osternig L. Isokinetic dynamometry: implications for muscle testing and rehabilitation. Exert Sport Sci Annu Rev 1986;14:4580. 25. Thorstensson A, Grimby G, Karlsson J. Force-velocity relations and fibre composition in human knee extensor muscles. J Appl Physiol 1976;40: 12-6. Supplier a. MedX Corporation, 1401 NE 77th Street, Ocala, FL 34479.