The relationships among knee extensor torques produced during maximal voluntary contractions under various test conditions.

Isokinetic dynamometers are commonly used to measure the forces generated during voluntary muscle contractions. Measurements from more than one contraction type or during more than one velocity of movement are often obtained.[1-6] Presumably, physical therapists measure the forces or torques generated during different muscle contractions and during different velocities of limb movement because these measurements provide useful information. Reports from some investigators,[7-9] however, suggest that these measurements may actually be providing similar information about muscle function.

Lankhorst et al[7] reported strong linear correlations between isometric torques and concentric peak torques obtained during maximal voluntary contractions of the quadriceps femoris muscle group. Otis and Godbold[8] reported a significant linear correlation between isometric torques and concentric angle-specific torques obtained during maximal voluntary contractions of the elbow flexor and extensor muscles. Singh and Karpovich[9] reported significant linear correlations between angle-specific forces obtained during maximal voluntary concentric and eccentric contractions of the elbow flexor and extensor muscles. The data of these investigators suggest that different tests used to assess performance during maximal voluntary contractions of a muscle may actually be assessing common components of muscle function. The different tests may, therefore, be parallel forms of the same test.

Unfortunately, in the three studies cited,[7-9] the investigators did not report the reliability of their measurements. Some of the investigators who used isokinetic dynamometers also failed to follow acceptable protocols (eg, they did not check calibration of the dynamometer, control for impact artifact, and correct for the effects of gravity), which must be followed if the measurements obtained with these devices are to be of the greatest value.[10,11] The results of these studies may, therefore, be questioned.

Knowledge of the correlations (relationships) between measurements of force or torque obtained during different maximal voluntary contractions is limited. Determining whether relationships exist in nondisabled subjects would add to physical therapists' understanding of normal muscle physiology. Knowledge of the presence of relationships in nondisabled subjects would provide therapists with a point of departure and a standard of comparison when determining whether the same relationships exist in various patient populations. The presence of relationships in nondisabled subjects would also provide some evidence to suggest that isokinetic testing protocols may be redundant. Further research would be necessary to determine whether strong relationships are present in various patient populations.

The purpose of our study was to describe the relationships among measurements of knee extensor torque produced by nondisabled subjects during maximal voluntary concentric, eccentric, and isometric contractions.

Method

Subjects

The subjects of this study were 20 nondisabled, female volunteers who had no limitations in ranges of motion of their right hip or knee joints, had no pain on resisted motion of their right hips and knees, had never had surgery on their right knees, and had no known pathology of their right quadriceps femoris muscles or associated right knee structures. The subjects' ages ranged from 21 to 33 years, with a mean age of 25 years (SD=3.5). Characteristics of the subject sample are presented in Table 1. Subjects were asked to read and sign a consent form prior to testing.

Table 1. Characteristics of Subject
Sample (N = 20)
Variable
Age (y)
  [X bar]        25
  SD               3.5
  Range          21-33
Height (cm)
  [X bar]        163.8
  SD               5.1
  Range        154.9-175.3
Weight (kg)
  [X bar]         57
  SD               9.5
  Range         41.9-81.0

Instrumentation

A Kin-Como[R](*) dynamometer (model 500-11, software version 3.01) was used to measure the torques produced during maximal voluntary concentric, eccentric, and isometric contractions of the quadriceps femoris muscles. In this study, concentric and eccentric contractions were tested with the Kin-Comp[R] in the evaluation mode. The control constant was set on speed. The acceleration (turn point, acceleration) and deceleration (turn point, deceleration) of the lever arm of the Kin-Com[R] were set on high. The high settings were used so that the subject's limb accelerated to and decelerated from constant velocity in the shortest possible period of time. This procedure maximized the amount of time the subject's limb moved at constant velocity. The force required to initiate motion of the lever arm (initial force) was 150% of the weight of the subject's limb. During preliminary testing on the Kin-Com[R], we found that using 150% of the subject's limb weight prevented sudden movements of the lever arm during testing and enabled the subject to generate some muscle tension before the lever arm began to move. Isometric contractions were tested with the Kin-Com[R] in the training mode. The time (pause) for each isometric contraction was set for 3 seconds.

The Kin-Com[R](*) dynamometer's recording system samples the analog signals from the strain gauge (force signal), potentiometer (angle signal), and tachometer (velocity signal) at a rate of 100 Hz. Preliminary testing conducted in our laboratory indicated that during high-velocity testing, a sampling rate of 100 Hz did not adequately represent the signals. The analog signals were, therefore, digitized by an external recording system (AMM1 Analog-to-digital Board in a Keithley DAS Measurement and Control Systems[R],([dagger]) series 500). Data acquisition was controlled using DADISP I[R]([double dagger] software (version 1). Signal acquisition was at a rate of 500 Hz for each channel.

The calibration of the Kin-Com[R] dynamometer's strain gauge and potentiometer was checked prior to and after testing all of the subjects. The results of the prestudy and poststudy calibration checks indicated that the strain gauge and potentiometer of the Kin-Com[R] remained calibrated during this study. Calibration of the tachometer was tested several times prior to the study and was found to be accurate.

Procedure

All subjects participated in two practice sessions and one test session. The first 15 subjects also participated in a retest session to determine the reliability of the torque measurements. One investigator (JMK) conducted all of the practice, test, and retest sessions. The practice, test, and retest sessions occurred on separate days, with no less than 48 hours and no greater than 96 hours between sessions. The practice and test protocols used in this study were based on those of a previous study that we conducted.[12] The current study was undertaken to develop a protocol that could be used to obtain reliable measurements of knee extensor torque.

Practice Sessions

On the first day of practice, each subject's limb was weighed by use of the Kin-Com[R] gravity-correction mode. Each subject positioned herself in the supine position on the Kin-Com[R] table with her right leg next to the lever arm of the dynamometer. The axis of rotation of the lever arm was aligned with the subject's right lateral femoral epicondyle, which was used as the reference for the knee joint axis. A strap was then placed across the midportion of the subject's right thigh, and the pad of the lever arm was placed just proximal to the medial malleolus. The distance from the pad to the axis of rotation of the lever arm was recorded so that the pad of the lever arm was placed in the same position during subsequent practice, test, and retest sessions. This value was also used to convert a subject's force measurements to torque measurements during data reduction.

With the subject still in a supine position, her knee was placed in 90 degrees of flexion. A universal goniometer was used to determine the position of the knee joint. The 90-degree value was entered into the Kin-Com[R] computer so that the recording of the lever arm position corresponded to the knee joint position, The subject's knee was then placed in 0 degrees of flexion and, with the subject relaxed, her leg was weighed by use of the Kin-Com[R] dynamometer's strain gauge and gravity correction mode. The weight of the subject's leg was used to determine the force required to initiate motion of the lever arm when performing concentric and eccentric contractions. The weight of the subject's leg was also used during data reduction to correct torque measurements for the effect of gravity.[11]

After the subject's limb was weighed, she sat up on the Kin-Com[R] table with her hips in approximately 80 degrees of flexion. The position of the subject's hips was determined by visual estimation of the angle formed by her trunk and thighs. The seat back of the Kin-Com[R] was placed behind the subject to maintain this position. The location of the seat back was recorded so that the seat back was placed in the same position during subsequent sessions. A strap was then placed securely across the subject's pelvis, and the axis of rotation of the lever arm was aligned with the axis of rotation of her right knee joint. This was done as described previously. The subject's knee was then placed in 90 degrees of flexion. A universal goniometer was used to determine this position. The 90-degree value was entered into the Kin-Com[R] computer so the recording of the lever arm position corresponded to the knee joint position.

A similar procedure was followed on the second day of practice. The only difference was that the subject's limb was not weighed again. For gravity correction, the weight obtained during the initial session was used. The pad of the lever arm and the seat back were placed in the same position as on the first practice day.

On each practice day, subjects performed six consecutive maximal voluntary contractions of their right quadriceps femoris muscles for each of the following 10 conditions: isometric contractions at 40 and 60 degrees of knee flexion (conditions 1 and 2); concentric contractions at velocities of 30[degrees], 90[degrees], 120[degrees], and 180[degrees]/s (conditions 3-6); and eccentric contractions at velocities of 30[degrees], 90[degrees], 120[degrees], and 180[degrees]/s (conditions 7-10).

The conditions we tested were chosen because these conditions are often used when testing patients with knee problems or when conducting research with isokinetic devices. The order in which the conditions were tested was random for each subject and remained the same between sessions. There was a 45-second rest period between consecutive contractions and a 2-minute rest period between conditions.

Prior to obtaining measurements during a condition, subjects were informed of which type of contraction they had to perform. Subjects were instructed to hold onto the sides of the seat and to "kick as hard as possible" during all contractions. The subjects began a contraction following the examiner's verbal cue of "One, two, three, GO!"

Isometric contractions were performed for 3 seconds. Concentric contractions were performed through a 75-degree arc of motion, starting at 90 degrees and ending at 15 degrees of knee flexion. Eccentric contractions were performed through the same arc of motion, starting at 15 degrees and ending at 90 degrees of knee flexion. No data were collected during the practice sessions.

The subjects did not view the Kin-Com[R] cathode ray tube (CRT) screen when performing contractions. We felt that viewing the CRT screen would be distracting and would interfere with the subject's ability to make a maximal effort. Schenck and Forward[13] found that subjects who had knowledge of the force values they produced during maximal voluntary contractions did not perform any differently than subjects who did not know their force values.

Test Session and Retest Session

Prior to the start of each test session, each subject was positioned on the Kin-Com[R] table and prepared for testing using the procedure followed during the practice sessions. The subject then performed four consecutive maximal voluntary contractions of her right quadriceps femoris muscle for the 10 conditions that she practiced. The order of testing and amount of rest between contractions was the same as during the practice sessions. The subject also received the same verbal instructions that she received during the practice sessions. Unlike the practice sessions, the subject was given strong verbal encouragement while performing each contraction.

The first 15 subjects were tested a second time, within 96 hours of the test session. This retesting was done in order to collect data to determine the reliability of the torque measurements. Intraclass correlation coefficients ICC[1,1])[14] were calculated to determine the degree of agreement reliability) between the repeated measurements for the test and retest sessions. The ICC values ranged from .87 to .98 (Tab. 2) and indicate that the torque measurements obtained in this study were highly reliable.

Table 2. Reliability Estimates for
Knee Extensor Torque Measurements
                          ICC(b)
Condition(a)             (N=15)
IKET([degrees]
   40                      .90
   60                      .94
EKEPT ([degrees]/s)
   30                      .94
   90                      .94
  120                      .98
  180                      .98
EKEPT ([degrees]/s)
   30                      .87
   90                      .95
  120                      .96
  180                      .96
(a)IKET=isometric knee extensor torque,
CKEPT=concentric knee extensor peak torque,
EKEPT=eccentric knee extensor peak torque.
(a)ICC=intraclass correlation coefficient (1,1).

Data Reduction

The force, velocity, and angle data were analyzed using DADISP II[R](sections] version 1.01) software. For each maximal voluntary contraction, the voltage signal from the strain gauge was converted to newtons, the voltage signal from the tachometer was converted to degrees per second, and the voltage signal from the potentiometer was converted to degrees. These conversions were done by multiplying the signals by calibration factors that were determined during preliminary testing of the Kin-Com[R] used in this study (Thomas P Mayhew, PhD, PT; personal communication; January 1989).

The calibrated force data for each subject were converted to newton-meters (torque) by multiplying the force values by the length of the subject's lever arm. The calibrated torque data were then corrected for gravity."[11]

The peak torque for each isometric torque set was determined. The peak torque for each concentric and eccentric torque set was also determined. Prior to determining the peak torques, the velocity data for each concentric and eccentric condition were sampled to determine the portion of the range during which limb velocity was constant. During isokinetic testing, the subject accelerated her limb and the lever arm of the machine at the beginning of the movement and decelerated her limb and the lever arm at the end of the movement. Because it is difficult to interpret forces produced during acceleration and deceleration of the limb, the concentric and eccentric torque sets were only analyzed for the portion during which limb velocity was constant. The method used to analyze the velocity data is described in detail elsewhere.[12]

For each subject, for each condition, the greatest of the four peak torque values was determined and represented the subject's concentric knee extensor peak torque (CKEPT), eccentric knee extensor peak torque (EKEPT), or isometric knee extensor peak torque (IKEPT). Because we were interested in the relationships among knee extensor torques produced during a maximal effort, we chose the greatest of the four peak torque values for our analysis.

Data Analysis

Each subject had a total of 10 peak torque measurements, 1 for each condition, that were used in the data analysis. Linear regression equations (LREs) and coefficients of determination (CD) ([r.sup.2] were calculated, using Statistical Analysis Software[R],([parallel]) to determine the nature and strength of the linear relationships among the knee extensor torque measurements. From the data we obtained, a total of 45 different relationships among the torque measurements could have been examined. We chose to examine 25 relationships that we believed were the most relevant.

Results

The Relationship Among Isometric Peak Torques

The CD describing the strength of the relationship between IKEPT at 40 degrees of knee flexion and IKEPT at 60,degrees of knee flexion was .84. The LRE describing the nature of the relationship was Y = .71X + 11.

The Relationships Among Peak Torques Produced During Concentric or Eccentric Contractions

The CDs and LREs describing the relationships among CKEPTs are presented in Table 3. The CDs ranged from .73 to .94. The CDs and LREs describing the relationships among EKEPTs are presented in Table 4. The CDs ranged from .63 to .88. The CDs and LREs describing the relationships between CKEPTs and EKEPTs are presented in Table 5. The CDs ranged from .60 to .86.

[TABULAR DATA 3 & 4 OMITTED]

Table 5. Matrix of Coefficients of
Determination and Linear Regression
Equations Describing the Relationships
Between Concentric Knee Extensor Peak
Torques and Eccentric Knee Extensor
Peak Torques(a)
Velocity ([degrees]/s)   CKEPT(b) Versus EKEPT(c)
30                      .86
                         Y = 1.25X - 20
90                      .60
                         Y = 1.33X   20
120                     .63
                        Y = 1.52X   6
180                     .79
                        Y = 1.80X - 15
(a)N=20, P<.0001.
(b)Concentric knee extensor peak torque
(X=independent variable).
(c)Eccentric knee extensor peak torque
(Y=dependent variable).

The Relationships Between Isometric Peak Torques and Concentric or Eccentric Peak Torques

The CDs and LREs describing the relationships between CKEPTs and IKEPTs at 60 degrees of knee flexion are presented in Table 6. The CDs ranged from .73 to .90. The CDs and LREs describing the relationships between EKEPTs and IKEPTs at 60 degrees of knee flexion are also presented in Table 6. The CDs ranged from .72 to .82.

[TABULAR DATA 6 OMITTED]

Discussion

Our study was conducted to determine the relationships among measurements of torque produced during maximal voluntary contractions of the quadriceps femoris muscles. We hypothesized, based on muscle physiology, that the torque measurements would be highly correlated. Our methods, however, did not assess only a muscle's ability to produce forces. Force production of the quadriceps femoris muscle was dependent on subject performance. Although subjects practiced before being tested, several factors (eg, fatigue, emotional state, motivation) in addition to a muscle's ability to produce forces could have influenced their performance. We also tested only 20 subjects in our study. With this sample size, a subject who deviates from the rest of the group can significantly influence the results of a statistical analysis. Because of these two factors (ie, variability in subject performance and small sample size), we believed that a CD of .70 or greater indicated a strong correlation, a coefficient between .50 and .69 indicated a moderate correlation, and a coefficient of less than .50 indicated a weak correlation. Out of the 25 relationships examined in our study, 20 of the CDs were .70 or greater. The 5 coefficients of less than .70 ranged from .60 to .69. Based on our classification, the torque measurements were moderately to highly correlated.

The relationship between concentric and eccentric peak torques at 90[degrees]/s was the weakest, with a CD of .60 and an LRE of Y = 1.33X + 20 (Tab. 5). We used the LRE and the subjects' CKEPTs at 90[degrees]/s to predict the subjects' EKEPTs at 90[degrees]/s. We then compared the predicted value for EKEPT with the actual EKEPT value. The predicted value was within 10% or less of the actual value for 10 of the subjects and was within 20% or less of the actual value for 7 of the subjects. The predicted values for the other 3 subjects were within 22%, 46%, and 47%, respectively, of the actual values.

The relationship between CKEPTs at 120[degrees]and 180[degrees]/s was the strongest relationship we found, with a CD of .94 and an LRE of Y = .87X - 4. Using the LRE and the subjects' CKEPTs at 120[degrees]/s to predict the subjects' CKEPTs at 180[degrees]/s, we found that the predicted value for CKEPT at 180[degrees]/s was within 10% or less of the actual CKEPT value for 18 of the 20 subjects. The predicted value for the other 2 subjects was within 13% of the actual value.

Based on the information we obtained by examining our data, we believe the classification system we chose for the CDs was appropriate. We also believe that the results of our study suggest that measurements of torque obtained during different maximal voluntary contractions of the quadriceps femoris muscles are assessing some common components of muscle function. Because the different measurements were not perfectly correlated (ie, CD of 1.00), there are still some "noncommon" components that are unique to each type of contraction.

In the following sections, the strength and nature of the linear relationships examined in this study are discussed in more detail. The discussion focuses on the Cds and slopes of the LREs describing these relationships. The values of the y-intercepts of the LREs were quite variable and did not follow any discernible trends. We felt that attempting to discuss these values would be difficult and would not be meaningful to the reader. The reader should be cautioned that we did not statistically test the slopes of the LREs. Because we were only looking for general trends in our data, we did not believe that an analysis of the slopes of the LREs was necessary. In addition, because we had such high CDs, we did not test for higher order models. Our results, therefore, describe linear relationships among the torque measurements.

The Relationships Among Isometric Peak Torques

No studies were found that examined the relationships among the forces or torques produced during maximal voluntary isometric contractions of the quadriceps femoris muscles. Several investigators,[15-19] however, have described force-joint angle or torque-joint angle curves produced during maximal voluntary isometric contractions of the quadriceps femoris muscles. These studies have shown that during maximal effort, force or torque production of the quadriceps femoris muscle varies in a predictable way with respect to joint position. The results of these studies suggest that the force produced by the quadriceps femoris muscle at one position of knee joint flexion should be highly correlated with the force produced at another position of knee joint flexion. The results of our study support this hypothesis.

The LRE for predicting IKEPT at 40 degrees of knee flexion from IKEPT at 60 degrees of knee flexion was Y = .71X + 11. The subjects' IKEPTs at 60 degrees of knee flexion ranged from 83 to 234 N[multiplied by]m. When actual torque measurements in this range are used to predict IKEPT at 40 degrees of knee flexion (using the LRE determined in this study), the predicted torque is less than IKEPT at 60 degrees of knee flexion. This finding is in agreement with those of studies that have examined force-joint angle or torque-joint angle curves produced during maximal voluntary isometric contractions of the quadriceps femoris muscles.[15-19] These studies have shown that the maximum force or torque produced during isometric knee extension occurred when the knee was in approximately 60 degrees of flexion and that the force or torque decreased with extension, reaching a minimum value with the knee fully extended.

The Relationships Among Peak Torques Produced During Concentric or Eccentric Contractions

Concentric knee extensor peak torques were highly correlated. These findings are in agreement with those of Lankhorst et al,[7] who measured CKEPTs during isokinetic movements at velocities of 30[degrees], 60[degrees], 120[degrees], and 180[degrees]/s. Unfortunately, Lankhorst et al did not determine the reliability of their torque measurements. They also failed to provide details regarding the damp setting of the Cybex[R] II dynamometer,(#) calibration of the Cybex[R] II, subject positioning, and control of impact artifact. Therefore, any comparisons between the two studies should be done with caution.

One trend apparent in the data of Lankhorst et al[7] is that the correlation coefficients decreased as the difference between the velocities of the isokinetic movements increased. A similar trend was seen in our data (Tab. 3). For example, the CD for CKEPT at 30[degrees] and 90[degrees]/s was .83 and the CD for CKEPT at 30[degrees] and 120[degrees]/s was .73. These results suggest that high-velocity and low-velocity isokinetic testing may be assessing some different components of performance. These data contrast with data presented by Rothstein et al.[20] Rothstein et al reported a strong relationship among CKEPTs obtained at velocities of 30[degrees], 60[degrees], 90[degrees], and 120[degrees]/s. They used a Cybex[R] Il dynamometer to measure CKEPT in 19 patients with rheumatic diseases and 11 age-matched control subjects. Linear regression equations calculated to determine the concentric peak torque-velocity relationship for each subject indicated strong linear relationships (average r =.87 for patients and .96 for control subjects). Rothstein and colleagues' results suggest that CKEPT produced by a subject at one velocity of isokinetic testing can be used to predict CKEPTs that subject would produce at other velocities of isokinetic testing. These results also suggest that low-velocity and high-velocity isokinetic testing could be assessing common components of performance during concentric contractions.

We plotted concentric peak torque-velocity curves for each subject to examine our results in more detail and to make more direct comparisons with the findings of Rothstein et al.[20] The CDs and LREs are presented in Table 7. For each subject, there was a strong linear relationship between CKEPT and velocity, with CDs ranging from .80 to .99. These results indicate that, for a given subject, CKEPT at one velocity of isokinetic testing was highly correlated with CKEPT at any other velocity of testing. The results also suggest that, for a given subject, CKEPTs obtained during high-velocity and low-velocity isokinetic movements are assessing a common component of performance. Unfortunately, the LREs describing the concentric peak torque-velocity relationships were subject specific. Therefore, the equations are only useful in predicting a subject's CKEPTs after the subject has been tested at several velocities and the concentric peak torque-velocity relationship has been determined.

When determining the nature of the relationship between two CKEPTs in our study, the CKEPT at the lower velocity of testing was always the independent variable (X) and the CKEPT at the higher velocity was always the dependent variable (Y). The slopes of the LREs describing the relationships among CKEPTs ranged from .64 to .88 and decreased as the difference between the velocities of the isokinetic movements increased (Tab. 3).

The results of the linear regression analyses are in agreement with the subjects' concentric peak torque-velocity relationships (Tab. 7). The slopes of the LREs for the subjects' concentric peak torque-velocity relationships were always negative. As the velocity of the isokinetic movements increased, the amount of peak torque produced by the subjects decreased.

Table 7. Coefficients of
Determination and Linear Regression
Equations Describing Concentric Peak
Torque-Velocity Relationships
               Coefficient of   Regression
Subject No.    Determination    Equation
 1           .96              -.23X    102
 2            99              -.41X    221
 3           .93              -.33X    206
 4           .95              -.40X    180
 5            89              -.38X    213
 6           .90              -.34X    224
 7           .97              -.21X    185
 8           .96              -.34X    136
 9           .89              -.47X    243
10           .99              -.48X    207
11           .96              -.12X    102
12           .91              -.35X    204
13           .98              -.56X    203
14           .89              -.67X    235
15           .96              -.56X    201
16           .80              -.24X    143
17           .99              -.40X    187
18           .97              -.44X    221
19           .96              -.58X    195
20           .95              -.32X    177

Because CKEPTs at the lower velocities of testing were the X values and CKEPTs at the higher velocities of testing were the Y values, the slopes of the lines describing the relationships among CKEPTs would decrease as the difference between the velocities of the concentric contractions increased. Other investigators, who have examined in vivo peak-torque velocity relationships[21-24] and in vitro force-velocity relationships[25] for concentric contractions, have reported similar results. These studies have shown that when the velocity of a concentric contraction increases, the amount of torque or force produced decreases.

Eccentric knee extensor peak torques were moderately to highly correlated. The CDs tended to decrease as the difference between the velocities of the isokinetic movements increased. This trend is similar to that seen for the CKEPTs (Tab. 3). The CDs for the EKEPTs, however, were generally lower than the CDs for the CKEPTs.

The relationships among EKEPTs may have been weaker because EKEPTs were less reliable than CKEPTs (Tab. 2). Because the error associated with measurements of CKEPT was less than with measurements of EKEPT, the relationships among CKEPTs may have been stronger. The lower ICC values for EKEPTs suggest that subjects may not have "learned" to perform eccentric contractions as well as concentric contractions. Most of the subjects in our study stated that eccentric contractions were more difficult to perform than concentric contractions. Performing eccentric contractions may have been a more novel task for the subjects. Unfortunately, we did not determine each subject's level of fitness or her experience with isokinetic devices. Knowledge of these factors may explain some of the inconsistency in performance, especially during the eccentric contractions.

The CDs for the relationships among EKEPTs suggest that high-velocity and low-velocity isokinetic testing during eccentric contractions could be assessing different components of performance, We plotted eccentric peak torque-velocity curves for each subject to examine our results in more detail and to determine whether individual eccentric peak torque-velocity relationships were any stronger. For a majority of the subjects, there was little, if any, relationship between EKEPT and velocity. In most instances, the EKEPTs produced by a subject were similar at all velocities tested.

The eccentric peak torque-velocity relationships for our subjects differed from in vitro force-velocity[25,26] and in vivo torque-velocity[2,27,28] relationships that have been described by previous investigators. These investigators found that force or torque increased with increases in the velocity of the eccentric contraction. Westing et al,[24] however, have described in vivo eccentric peak torque-velocity relationships that are similar to those found in our study. They measured peak torque during eccentric contractions of the quadriceps femoris muscles and found that, for each subject, the peak torques produced were similar at all velocities tested.

The study by Westing et al24 Was the only study found that examined the torque-velocity relationship for eccentric contractions of the quadriceps femoris muscles. Other investigators,[2,27,28] who described in vivo eccentric torque-velocity relationships for the elbow flexor or extensor muscles, found that the torque produced during an eccentric contraction increased with increased velocity of movement.

The slopes of the LREs describing the relationships among EKEPTs were generally similar in value and did not follow a discernible trend (Tab. 4). The EKEPTs produced by each subject tended to be similar at each velocity tested. Because there was little difference in EKEPTs across velocities, the slopes of the LREs describing the relationships among EKEPTs would be similar to each other.

Concentric knee extensor peak torques and EKEPTs were moderately to highly correlated (Tab. 5). The CDs did not follow a discernible trend. The relationship between CKEPTs and EKEPTs, however, was strongest at the lowest velocity of testing (ie, 30[degrees]/s).

The slopes of the LREs describing the relationships between EKEPTs and CKEPTs decreased as the velocities of the isokinetic movements increased. This finding is consistent with the subjects' concentric peak torque-velocity and eccentric peak torque-velocity relationships. The CKEPTs decreased with increases in the velocity of the isokinetic movement. The EKEPTs produced by most of the subjects were similar (for a given subject) at all velocities tested. Because CKEPT was the independent variable (X) and EKEPT was the dependent variable (Y), the slopes of the lines would increase as the velocities of the isokinetic movements increased.

The Relationships Between Isometric Peak Torques and Concentric or Eccentric Peak Torques

The CDs for the relationships between IKEPTs and CKEPTs decreased as the velocity of the isokinetic movement increased. These results are similar to those reported by Lankhorst et al[7] and Knapik and Ramos.[29] Although the results reported by these authors are in agreement with the results of our study, methodological limitations (eg, failure to report the reliability of the torque measurements and lack of details regarding control of impact artifact, damp, and calibration of the dynamometer) bring the results of these two studies into question.

The slopes of the LREs describing the relationships between IKEPTs and CKEPTs decreased as the velocity of the isokinetic movement increased (Tab. 6). The decrease in the values of the slopes is consistent with the concentric peak torque-velocity relationships for the subjects of this study (Tab. 7). As the velocity of the isokinetic movement increased, the CKEPT produced by the subjects decreased.

Because IKEPT was the independent variable (X) and CKEPT was the dependent variable (Y), the slopes of the lines would decrease as the velocity of the isokinetic movement increased.

The IKEPTs were generally less predictive of EKEPTs than of CKEPTs (Tab. 6). The ICC values estimating the reliability of EKEPTs and CKEPTs indicate that there was more error associated with EKEPTs (Tab. 2). The IKEPTs may have been less correlated with EKEPTs because measurements of EKEPT had more error.

Summary of the Relationships

The relationships among IKEPTs, CKEPTs, and EKEPTs were moderately to highly correlated. These findings suggest that measurements of torque produced during maximal voluntary contractions of the quadriceps femoris muscles are generally assessing common components of performance. The trends of the CDs, however, indicate that contraction type and the velocity of the isokinetic movement influence the strength of the correlation between one knee extensor torque measurement and other knee extensor torque measurements. The relationships among EKEPTs were generally weaker than the relationships among CKEPTs. The IKEPTs were also less correlated with EKEPTs than with CKEPTs. The relationships among CKEPTs decreased as the difference between the velocities of the isokinetic movements increased. A similar trend was seen with the relationships among EKEPTs. The relationship between CKEPTs and EKEPTs was greatest at the lowest velocity of testing.

Clinical Implications

Our study was conducted to determine the relationships among measurements of torque produced during maximal voluntary contractions of the quadriceps femoris muscles. We hypothesized, based on normal muscle physiology, that the torque measurements would be highly correlated. We therefore tested nondisabled subjects.

Because we found moderate to strong correlations among the knee extensor torque measurements of nondisabled subjects, a correlation may exist among the knee extensor torque measurements of patient populations. Additional factors, such as pain and limited range of motion, may influence how strong the correlations are.

The method of our study did not simulate a clinical situation. The subjects in our study were 20 nondisabled women. The equipment we used to process the signals from the Kin-Comp[R] sampled at a rate that was higher than the sampling rate of the Kin-Com[R]. The subjects in our study also practiced for 2 days before being tested and performed more contractions than a patient would perform during a physical therapy evaluation. The generalizability of our results is limited.

Despite these limitations, the results of our study suggest that physical therapists may want to reevaluate their protocols for testing subjects on isokinetic dynamometers. The results of our study provide evidence to suggest that some of the same components of muscle function are measured by different tests, These different tests could, therefore, be used interchangeably to obtain similar information about muscle function. Until further research is conducted to determine whether these relationships exist in patient populations, we suggest that physical therapists give some thought to the number and types of tests they perform with isokinetic dynamometers. Those tests that provide the most reliable measurements, that take the least time to conduct, or that are the most comfortable for the patient may provide sufficient information.

Decisions regarding which tests to use may also depend on the reasons for performing the isokinetic evaluation. For example, if an isokinetic evaluation is being conducted to determine whether an individual is ready to return to work, the therapist may only be willing to eliminate tests that are very highly correlated (eg, CDs of .90 or greater). if isokinetic evaluations are being conducted to screen athletes, however, one or two tests may provide sufficient information.

Conclusion

Twenty-five different relationships among measurements of knee extensor torque were examined in this study. The CDs for these relationships were moderately to highly correlated. These results suggest that measurements of torque produced by nondisabled subjects during different maximal voluntary contractions of the quadriceps femoris muscle may be assessing common components of performance, Further research is needed to determine whether knee extensor torque measurements in various patient populations are strongly correlated. (*)Chattecx Corp, 101 Memorial Dr, PO Box 4287, Chattanooga, TN 37045. [dagger])Keithley Data Acquisition & Control Inc, 28775 Aurora Rd, Cleveland, OH 44139 ([double dagger])DSP Development Corp, 55 Cambridge Pkwy, Cambridge, MA 02142. ([sections])IDSP Development Corp, One Kendall Sq, Cambridge, MA 02139. ([parallel])SAS Institute Inc, PO Box 8000, Cary, NC 27511-8000. (#) Cybex Div of Lumex Inc, 2100 Smithtown Ave, Ronkonkoma, NY 11779.

References

[1] Elsner RC, Redegan LR, Lang J, Protocol for strength testing of the upper extremity. J Orthop Sports Phys Ther. 1983;4:229-235. [2] Griffin JW. Differences in elbow flexion torque measured concentrically, eccentrically, and isometrically. Phys Ther. 1987;67:1205-1208. [3] Murray M, Gardner G, Mollinger L, Sepic S. Strength of isometric and isokinetic contractions. Phys Ther. 1980;60:412-419. [4] Timm KE, Patch DG. Case study: use of a Cybex[R] II velocity spectrum in the rehabilitation of post-surgical knees. J Orthop Sports Phys Ther. 1985;6:347-349. [5] Watkins MP, Harris BA, Kozlowski BA. Isokinetic testing in patients with hemiparesis: a pilot study. Phys Ther. 1984;64:184-189. [6] Wong DL, Glasheen-Wray M, Andrews L. Isokinetic evaluation of the ankle invertors and evertors. J Orthop Sports Phys Ther. 1984;6: 246-252. [7] Lankhorst GJ, Van de Stadt RJ, Van der Korst JK. The relationships of functional capacity, pain, and isometric and isokinetic torque in osteoarthritis of the knee. Scand J Rehabil Med. 1985;17:167-172. [8] Otis JC, Godbold JH. Relationship of isokinetic torque to isometric torque. J Orthop Res. 1983;1:165-171. [9] Singh M, Karpovich PV. Isotonic and isometric forces of forearm flexors and extensors. J Appl Physiol 1966;21:1453-1457. [10] Rothstein JM, Lamb RL, Mayhew TP. Clinical uses of isokinetic measurements: critical issues. Phys Ther. 1987;67:1840-1844. [11] Winter DA, Wells RP, Orr GW. Errors in the use of isokinetic dynamometers. Eur J Appl Physiol. 1981;46:397. [12] Kues JK, Rothstein JM, Lamb RL. Obtaining reliable measurements of knee extensor torque produced during maximum voluntary contractions. Phys Ther, 1992;72:492-504. [13] Schenck JM, Forward EM. Quantitative strength changes with test repetitions, Phys Ther. 1965;45:562-569. [14] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979,86:420-428. [15] Haffajee D, Moritz U, Svantesson G. Isometric knee strength as a function of joint angle, muscle length, and motor unit activity. Acta Orthop Scand 1969;40:79-85. [16] Knapik JJ, Wright JE, Mawsdley RH, Braun J. Isometric, isotonic, and isokinetic torque variations in four muscle groups through a range of joint motion. Phys Ther. 1983;63:938-947. [17] Lindahl D, Movin A, Ringqvist I. Knee extension: measurement of the isometric force in different positions of the knee joint. Acta Orthop Scand 1969;40:79-85. [18] Scudder GN. Torque curves produced at the knee during isometric and isokinetic exercise. Arch Phys Med Rehabil. 1980;61:68-73. [19] Smidt GL. Biomechanical analysis of knee flexion and extension. J Biomech. 1973;6:79-92. [20] Rothstein JM, Delitto A, Sinacore DR, Rose SJ. Muscle function in rheumatic disease patients treated with corticosteroids. Muscle Nerve. 1983;6:128-135. [21] Froese EA, Houston ME. Torque-velocity characteristics and muscle fiber type in human vastus lateralis. J Appl Physiol. 1985;59:309-314. [22] Perrine JJ, Edgerton VR. Muscle force-velocity and power-velocity relationships under isokinetic loading. Med Sci Sports. 1978;10: 159-166. [23] Thorstensson A, Grimby G, Karisson J. Force-velocity relationships and fiber composition in human knee extensor muscles. J Appl Physiol 1976;40:12-16. [24] Westing SH, Seger JY, Karlson E, Ekbolm B. Eccentric and concentric torque-velocity characteristics of the quadriceps femoris in man. Eur J Appl Physiol 1988;58:100-104. [25] Hill AV. The heat of shonening and the dynamic constants of muscle. Proc R Soc Lond 1938;126:136-195. [26] Joyce GC, Rack PM, Westbury DR. The mechanical properties of cat soleus muscle during controlled shortening and lengthening movements. J Physiol (Lond). 1969;204:468-474. [27] Komi PV. Relationship between tension, EMG, and velocity of contraction under concentric and eccentric work. In: Desmedt JE, ed. New Developments in Electromyography and Clinical Neurophysiology. Basel, Switzerland: S Karger AG, Medical and Scientific Publishers; 1973:540-546. [28] Rodgers KL, Berger RA. Motor-unit involvement and tension during maximum voluntary concentric, eccentric, and isometric contractions of the elbow flexors. Med Sci Sports. 1974;6:253-259. [29] Knapik JJ, Ramos MU. Isokinetic and isometric torque relationships in the human body. Arch Phys Med Rehabil, 1980;61:64-67.

JM Kues, PT, is Assistant Professor, Department of Physical Therapy, Medical College of Virginia, Virginia Commonwealth University, Box 224, MCV Station, Richmond, VA 23298 (USA). She was a student in the Department of Physical Therapy, Medical College of Virginia, Virginia Commonwealth University, when this study was completed in partial fulfillment of the requirements for her Master of Science degree. Address all correspondence to Ms Kues.

JM Rothstein, PhD, PT, is Professor and Head, Department of Physical Therapy, College of Associate Health Professions, University of Illinois at Chicago, 1919 W Taylor St, Chicago, IL 60612, and Chief, Physical Therapy Services, Physical Therapy Department, University of Illinois Hospital, 1740 W Taylor St, Chicago, IL 60612.

RL Lamb, PhD, PT, is Chairman and Professor, Department of Physical Therapy, Medical College of Virginia, Virginia Commonwealth University.

This article was submitted may 22, 1992, and was accepted December 7, 1993.