Isokinetic Performance After Total Hip Replacement

Abstract

Bertocci GE, Munin MC, Frost KL, Burdett R, Wassinger CA, Fitzgerald SG: Isokinetic performance after total hip replacement. Am J Phys Med Rehabil 2004;83:1–9.

Objective 

To evaluate differences in isokinetic hip flexion, extension, and abduction muscle performance of operated vs. nonoperated hips in older adults who have undergone elective, unilateral, total hip replacement (THR) surgery and completed rehabilitation.

Design 

Quasi-experimental study using a nonequivalent posttest-only control group design, comprising 20 unilateral THR patients and a convenience sample of 22 healthy older adults. THR patients participated between 4 and 5 mos after surgery. THR subjects received an average of 13 outpatient or home-based physical therapy sessions. Isokinetic muscle strength and fatigue was assessed through measurement of hip peak torque per body weight, total work, and average power using a robotic dynamometer.

Results 

Comparisons of THR subjects’ operated vs. nonoperated hips showed no significant differences in isokinetic performance for any of the examined variables. THR subjects’ operated hips generated significantly less peak torque per body weight, total work, and average power across all exercises as compared with a population of healthy subjects.

Conclusions 

THR subjects’ operated and nonoperated hips showed similar biomechanical performance. THR patients are not being restored to the same level of strength and muscular endurance as compared with a population of healthy adults. These findings may be useful in providing a preliminary rationale for revising current approaches in THR rehabilitation protocols.

More than 280,000 total hip replacement (THR) procedures are performed annually in the United States in an effort to increase mobility and reduce pain in those having arthritic, deteriorated, or injured hips. ¹ Osteoarthritis is the most common preoperative diagnosis for THR, ² accounting for approximately 70% of elective THR surgeries. ³ Severe or moderate osteoarthritis of the hip affects an estimated 3.1% of all individuals between the ages of 65 and 74 yrs. ⁴ As the number of elderly adults in the United States continues to increase, we can reasonably anticipate that the number of THR operations will also increase.

Published data assessing biomechanical performance after THR are sparse. The majority of studies assessing post-THR hip function using robotic dynamometry have focused on evaluating surgical outcomes and prosthetic selection. Borja et al. ⁵ compared postoperative operated vs. nonoperated hip abductor strength across two surgical approaches. The trochanteric surgical approach led to decreased isokinetic and isometric performance of the operated hip, whereas the modified posterior lateral approach did not result in a difference between operated and nonoperated hip abductor performance. In another study, Mostardi et al. ⁶ used isokinetic strength to evaluate functional outcomes after various trochanteric surgical management techniques. In this study, patients receiving the wafer surgical technique had superior performance in operated hip flexion, extension, and abduction at 2–11 yrs postoperatively as compared with patients undergoing other surgical techniques. More recently, Reardon et al. ⁷ found improved preoperative to postoperative (5 mos) quadriceps strength in THR subjects’ nonoperated hips, yet no difference in the operated hips. Compared with controls, hip patients exhibited significantly reduced quadriceps concentric peak torque but no difference in eccentric peak torque. Trudelle-Jackson et al. ⁸ used measures of postural stability, range of motion, and isometric strength to postoperatively compare operated vs. nonoperated hips. Postural stability was significantly less for the operated hip, though no differences were seen in isometric strength measured using hand-held dynamometry.

These studies provide insight regarding strength deficits of older adults after THR. However, the data are not consistent, making it difficult to assess the nature of the deficit. The purpose of this study is to describe isokinetic strength postoperatively between 4 and 5 mos after THR surgery and rehabilitation in a population of older adults and to evaluate these results in comparison with a population of healthy older adults. Additional information that aids in the understanding of the progression of biomechanical performance after THR may prove useful for clinicians to identify areas of deficit during rehabilitation and develop additional therapeutic strategies to improve rehabilitation outcomes.

METHODS

Subjects.

A total of 42 individuals gave written informed consent to participate in this study, in accordance with University of Pittsburgh Institutional Review Board procedures. THR subjects were identified and recruited through physician offices or physical therapy departments within the University of Pittsburgh Medical Center. Healthy subjects were recruited from the community through placement of paid advertisements in a community newspaper. All subjects participated in one test session. THR subjects participated between 4 and 5 mos postsurgery. This time period was chosen for several reasons: (1) postoperative healing is significantly complete, (2) hip precautions are relaxed, (3) risk of dislocation is significantly reduced, (4) patients have resumed activities of daily living, and (5) patients are able to execute the exercises included in the study protocol. A total of 20 THR subjects and 22 healthy subjects between the ages of 55 and 75 yrs were recruited. Subject characteristics are presented in Table 1. Groups were similar in sex distribution and number of comorbidities; however, the convenience sample of healthy adults yielded a difference in mean age. Mean age for THR subjects was 68 yrs (±5 yrs), and mean age for healthy subjects was 64 yrs (± 6 yrs). Women represented 55% of THR subjects and 64% of healthy subjects. All subjects were white. Average number of comorbidities for each group was two, excluding the presence of osteoarthritis. Eligibility criteria, shown in Table 2, ruled out persons with major cardiovascular disease history, those actively taking corticosteroids, and persons requiring ambulatory assist devices at the time of participation. All enrolled THR subjects had a diagnosis of osteoarthritis as the primary cause for elective THR. Healthy subjects were excluded for discrete hip pathology or a diagnosis of hip arthritis. Although surgical approach was not specified as an inclusion criterion, all THR subjects in this study had either a posterior lateral or straight lateral approach. THR subjects self-reported no discrete pathology of their contralateral hips.

TABLE 1: Select subjects’ demographics

TABLE 2: Inclusion criteria for total hip replacement (THR) and healthy adult subjects

Medical records revealed an average length of stay for surgery and acute care was 4 days for THR subjects. All but two THR subjects were discharged directly to a subacute rehabilitation facility, where the average length of stay was 8 days. After subacute rehabilitation, THR subjects received an average of 13 sessions of outpatient or home-based physical therapy.

Isokinetic Testing.

Isokinetic measurements were assessed using the Biodex System III Isokinetic Dynamometer v3.03 (Biodex Medical, Shirley, NY). Subjects were positioned on the Biodex Dynamometer in a supine position for hip flexion and extension measurements and in a side-lying position for hip abduction measurement. Subjects were properly positioned by a physical therapist before initiation of each test. To stabilize the body and minimize muscle compensation, especially during abduction-adduction, subjects were secured by placing a 6-inch wide strap above the iliac crests. Subjects were observed by testers across all repetitions and given verbal feedback as needed to maintain proper positioning. Measurements were assessed for both the left and right hip (healthy subjects) and for operated and nonoperated hip (THR subjects). Hip precautionary measures were observed at all times during exercise (i.e., no hip flexion >90 degrees, no hip adduction past neutral, and no internal rotation past neutral). The greater trochanter was used as the bony landmark for matching the axis of rotation of the hip joint with the axis of the dynamometer lever arm. Gravity correction was obtained before each test by measuring the torque exerted on the dynamometer lever arm by the weight of the limb. Once measured, the effect of gravity is automatically incorporated in subsequent calculations and the output provided by the Biodex Advantage Software program. Calibration of the Biodex dynamometer was performed according to the specifications outlined by the manufacturer’s service manual. ¹⁰

Reciprocal concentric isokinetic hip flexion and extension were assessed at a preset velocity of 90 degrees/sec over a range of motion of 0–45 degrees. A fixed number of 15 flexion-extension repetitions were completed by each subject. Reciprocal concentric isokinetic hip adduction and abduction were assessed at a preset velocity of 60 degrees/sec over a range of motion of 0–45 degrees, with subjects completing 25 repetitions for adduction/abduction. Each subject’s actual range of motion was based on his or her physiologic capability within the maximal set range. The selected number of repetitions and preset velocities were based on the clinical experience of the research physiatrist and physical therapist and on quality control testing performed using selected research team members. Subjects were provided with instructions before each test. Once instructed, subjects were allowed practice trials of up to three repetitions. A waiting period of 1 min was observed between practice trials and actual testing. No verbal encouragement was provided by the testers. Reliability of the Biodex for hip flexion, extension, and abduction measures was conducted using five members of the research team.

Data Collection.

The following three measurements were assessed during each test: (1) peak torque per body weight (highest value of torque, regardless of repetition), (2) total work (torque multiplied by angular displacement), and (3) average power (total work divided by time). Total work and average power calculations accounted for varying ranges of motion achieved by each subject. Data were obtained from the comprehensive evaluation reports generated by the Biodex at the completion of each test.

Statistical Analysis.

The data were analyzed for homogeneity and normal distribution. Because the comparison groups were not equal in sample size and the data were not normally distributed, nonparametric statistics were used. Wilcoxon’s signed-ranks test was used to test for significant differences between THR subjects’ operated and nonoperated hips. Healthy subjects’ data were first examined for bilateral differences between the right and left leg using Wilcoxon’s signed-ranks. No statistically significant difference was found; therefore, the average value of healthy subjects’ right and left legs was calculated for subsequent use in comparing data between groups. Comparisons between groups were made using Mann-Whitney U to test for significant differences between THR subjects’ operated leg and averaged healthy subject data. An alpha level of 0.05 was set for all statistical tests of significance. Test-retest reliability of the Biodex was estimated by calculating the intraclass correlation coefficient (ICC_3,1) for all biomechanical measures using five members of our research team. Statistical analysis was performed using SPSS Version 11. ¹¹

RESULTS

Reliability.

The intraclass correlation coefficients (ICC_3,1) for Biodex biomechanical measurements (flexion, extension, abduction) ranged from 0.91 to 0.95 for peak torque, 0.89 to 0.95 for total work, and 0.87 to 0.95 for average power. In general, reliability coefficients above 0.75 are indicative of good reliability. ¹² Therefore, use of Biodex for evaluating hip musculature performance in our study protocol is justified as an acceptable measurement tool.

Demographics.

Table 2 shows the comparison of selected demographic factors between both groups. As shown, both groups were similar in sex and mean number of comorbidities. No subjects reported pain on the day of testing.

It should be noted that there are variations in the number of subjects whose data were available for test-specific analysis (flexion-extension, abduction-adduction). The total number of subjects ranged from a low of 35 to a high of 41 for a given test-specific analysis. There are two reasons for this loss of data; subject inability or equipment malfunction. Five THR subjects were unable to complete the required number of isokinetic repetitions during one or more tests identified in the protocol. There was no difference between individuals unable to complete one or more tests and THR group subject characteristics. Because the Biodex system does not generate report data unless the full number of programmed repetitions is executed, no data were obtained in these instances. On six occasions, Biodex equipment/computer software problems were experienced. As a result, one or more isokinetic tests could not be completed (two of these occasions occurred during THR subject testing, and four occasions occurred during healthy subject testing), and again, data were not available for analysis. Resulting variations in sample sizes are indicated in the figures.

THR Operated vs. Nonoperated Hips.

As shown in Table 3, analysis of THR subject data showed that there were no significant differences in the performance of isokinetic exercises between THR subjects’ operated and nonoperated hips. THR subjects generated the highest values for peak torque per body weight during extension, followed by abduction, then flexion. This pattern was repeated for average power. Total work values were greatest for extension, followed by flexion and abduction. A high value for total work indicates a greater level of endurance. Thus, THR subjects showed greater endurance in extension, followed by flexion and abduction.

TABLE 3: Mean values: total hip replacement subjects’ operated and nonoperated hips

THR Operated vs. Healthy Subjects’ Hips.

As stated previously, analysis of healthy subject data showed that there were no statistically significant differences between healthy subjects’ right and left hips. Therefore, healthy subjects’ right and left hip data were combined to generate average values for subsequent comparison of healthy subjects to THR operated hips.

THR subjects’ operated hips generated decreased peak torque per body weight, decreased total work, and lower average power as compared with healthy subjects. As shown in Table 4, these differences were significant for each variable.

TABLE 4: Mean values: total hip replacement subjects’ operated and healthy subjects’ hips

For isokinetic flexion, mean peak torque per body weight (expressed as a percentage) was 1.6 times greater (20.00%vs. 12.54%) in healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.02). Total work was 2.1 times greater (728.9 ft-lb vs. 352.8 ft-lb) in healthy subjects’ hips as compared with THR subjects’ operated hips (P < 0.001). The difference in average power for isokinetic flexion was 4.4 times greater (12.72 W vs. 2.87 W) in healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.007). These differences are summarized in Figure 1.

Figure 1: Isokinetic flexion: healthy subject hips vs. total hip replacement operated and nonoperated hips. Total hip replacement subjects’ operated and nonoperated hips performed at significantly lower levels of isokinetic flexion as compared with healthy subjects. There is no significant difference between total hip replacement subjects’ operated and nonoperated hips. Mean values are shown with error bars depicting standard deviation.

For isokinetic extension, significant differences were also evident between healthy subjects’ hips and THR subjects’ operated hips. Peak torque per body weight was 1.7 times greater (53.77%vs. 31.63%) for healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.004). Total work was 1.6 times greater (597.3 ft-lb vs. 369.8 ft-lb) for healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.004). Average power was 1.6 times greater (77.71 W vs. 47.67 W) in healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.015). These results are summarized in Figure 2.

Figure 2: Isokinetic extension: healthy subject hips vs. operated and nonoperated hips. Total hip replacement subjects’ operated and nonoperated hips performed at significantly lower levels of isokinetic extension as compared with healthy subjects. There is no significant difference between total hip replacement subjects’ operated and nonoperated hips. Mean values are shown with error bars depicting standard deviation.

Isokinetic abduction comparisons of THR operated hips and healthy subjects’ hips yielded significant differences across all three variables. For isokinetic abduction, mean peak torque per body weight was 1.8 times greater (45.23%vs. 25.02%) in healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.001). Total work was 5.0 times greater (96.2 ft-lb vs. 18.5 ft-lb), and average power was 2.1 times greater (51.00 W vs. 24.37 W) in healthy subjects’ hips as compared with THR subjects’ operated hips (P = 0.006 and P < 0.001 respectively). These results are shown in Figure 3.

Figure 3: Isokinetic abduction: controls vs. operated and nonoperated hips. Total hip replacement subjects’ operated and nonoperated hips performed at significantly lower levels of isokinetic abduction as compared with healthy subjects. There is no significant difference between THR subject’s operated and nonoperated hips. Mean values are shown with error bars depicting standard deviation.

THR Nonoperated Hips vs. Healthy Subject Hips.

As stated above, THR operated and nonoperated hips showed no statistically significant difference for any of the three variables. However, THR subjects’ nonoperated hips generated decreased peak torque per body weight, decreased total work, and lower average power as compared with healthy subjects.

For isokinetic flexion, mean peak torque per body weight was 1.6 times greater (20.00%vs. 12.13%) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P = 0.012). Total work was 2 times greater (728.9 ft-lb vs. 360.8 ft-lb) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P = 0.000). The difference in average power for isokinetic flexion was 4.3 times greater (12.72 W vs. 2.98 W) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P = 0.002). These differences are summarized in Figure 1.

For isokinetic extension, significant differences were also evident between healthy subjects’ hips and THR subjects’ nonoperated hips. Peak torque per body weight was 1.7 times greater (53.77%vs. 31.38%) for healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P < 0.001). Total work was 1.6 times greater (597.3 ft-lb vs. 370.7 ft-lb) for healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P < 0.001). Average power was 1.7 times greater (77.71 W vs. 46.6 W) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P = 0.001). These results are summarized in Figure 2.

Comparisons of isokinetic abduction data also yielded significant differences across all three variables. For isokinetic abduction, mean peak torque per body weight was 1.8 times greater (24.65%vs. 45.23%) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P < 0.001). Total work was 4.1 times greater (23.2 ft-lb vs. 96.2 ft-lb), and average power was 2.1 times greater (24.63 W vs. 51.00 W) in healthy subjects’ hips as compared with THR subjects’ nonoperated hips (P = 0.005 and P < 0.001 respectively). These results are shown in Figure 3.

Coefficient of Variation.

Coefficients of variation, defined as mean torque divided by peak torque across all repetitions, are provided in Table 5 for THR and healthy subjects. Variation in performance across repetitions increases as coefficients of variation values decrease. THR subjects tended to have increased variation in extension as compared with flexion and abduction. Healthy adults had similar variation in extension to THR subjects but had even greater variation in abduction exercises.

TABLE 5: Coefficient of variation for isokinetic tests

DISCUSSION

We found no significant difference in isokinetic performance between THR subjects’ operated and nonoperated hips. This is likely due to mobility and functional limitations imposed by the operated hip that result in reduced levels of patient activity over a prolonged period of time, causing disuse muscle atrophy in the unaffected hip. These findings provide quantitative performance data during a time frame that has received limited study and that coincides more closely with completion of clinical and therapeutic services after THR. A number of previously conducted studies have sought to compare operated with nonoperated hip performance. ^5,7,8,13,14

Reardon et al. ⁷ studied eccentric and concentric peak torque in a group of seven THR patients 5 mos after surgery and found no difference in concentric peak torque between operated and nonoperated hips. Borja et al. ⁵ found similar results to our findings at 9 mos for THR patients undergoing a trochanteric approach, yet THR patients who received a posterior lateral surgical approach demonstrated significant differences in isokinetic strength between operated and nonoperated hips 6 mos postoperatively. It is possible that differences in the timing of postoperative tests (4–5 mos in our study vs. 6 mos in the study by Borja et al. ⁵) and isokinetic test protocols might contribute to varying results. Another major difference in the study by Borja et al. ⁵ is that none of the subjects received physical therapy, whereas in our study, all subjects had received physical therapy. Trudelle-Jackson et al. ⁸ used hand-held dynamometry to investigate hip strength and found no significant difference between operated and nonoperated hips 1 yr postoperatively. Subjects participating in the Trudelle-Jackson et al. ⁸ study had a mean age of 62 yrs, and all subjects had undergone home-health physical therapy. These studies have either in part or in total corroborated our findings of no significant differences between operated and nonoperated hip performance. ^5,7,8

However, some studies seem to be in disagreement with our findings of no difference between operated and nonoperated hips, and these require closer examination of study details. Shih et al. ¹⁴ studied 20 women and 20 men 1 yr after THR. As compared with patients’ nonoperated hips, the investigators found deficits in flexion, extension, and abduction ranging from 11–16% in the men, and 19–21% in the women, though statistical significance was not established. Differences in findings between our study and that by Shih et al. ¹⁴ may be due to (1) the subjects in the Shih et al. ¹⁴ study did not receive physical therapy, (2) male subjects in the study by Shih et al. ¹⁴ received THR due to osteonecrosis associated with alcoholism, (3) the subjects in the Shih et al. ¹⁴ study ranged from 24 to 71 yrs of age, and (5) the subjects in the Shih et al. ¹⁴ study were evaluated 1 yr postoperatively. Long et al. ¹³ used force-plate data to compare indirect measures of muscle strength between operated and nonoperated hips up to 2 yrs after THR. As compared with patients’ nonoperated hips, Long et al. ¹³ found strength deficits in the operated hips ranging from 8% to 14% at 1 yr postoperatively and deficits ranging from 5% to 24% at 2 yrs postoperatively, although again, statistical significance was not established. Subjects in the Long et al. ¹³ study were significantly younger than our subjects, with a mean age of 55 yrs and an age range of 35–85 yrs. No details were provided regarding physical therapy services received by patients in the Long et al. ¹³ study. It is important to note that although these studies show deficits on the operated side, ^13,14 these deficits were not statistically significant from the performance on the nonoperated side. Furthermore, a number of differences in subject characteristics and standard of care, as described above, existed between our study and these studies.

Rehabilitation after THR seeks to return patients to levels of normal physical function and activity. Thus, if rehabilitation is successful, muscle performance would be expected to improve to near normal levels of healthy adults. We did not find this to be the case when comparing THR patients with our convenience sample of healthy subjects. Our findings indicate that THR subject isokinetic hip performance at 4–5 mos postsurgery is significantly lower than that of a healthy older adult population. Previous research has documented differences between THR subjects and healthy subjects using gait analysis and isometric testing methods 1.5–3.8 yrs after surgery. ^5,15,16 Reardon et al. ⁷ found similar differences between THR subjects and controls when comparing concentric peak torque at 5 mos postsurgery. We found significant differences in muscle performance across flexion, extension, and abduction when comparing THR subjects’ operated hips with healthy subjects’ hips. This was the case for all isokinetic biomechanical measures (peak torque per body weight, total work, and average power), suggesting that THR subjects experience not only a deficit in strength when compared with healthy adults, but in endurance as well. Although the mean age for our convenience sample of healthy older adults was 4 yrs less than that of our hip patient population (mean age, = 68 yrs for THR patients and 64 yrs for healthy adults), we do not believe that this difference in age is clinically meaningful, especially given that the composition of the groups is similar with respect to sex and number of comorbidities. Post hoc analysis confirmed that within our sample of THR subjects, age was not correlated with muscle performance. Accordingly, our analyses did not attempt to control for age. Our findings therefore suggest that therapy should focus on both endurance and strength exercises after THR for both operated and nonoperated hips.

In clinical practice, rehabilitation to improve strength occurs for only a brief time after surgery. Therefore, early postoperative evaluation of muscle performance, such as that conducted in our study, may be a more appropriate marker of rehabilitation effectiveness. Our results and those of Reardon et al. ⁷ suggest that THR patients exhibit strength deficits in the short-term after rehabilitation. One implication of these findings is that current, brief, postoperative rehabilitation protocols may not be adequate to restore THR patients’ hip strength and endurance. In addition, our findings also suggest that therapeutic exercises after post-THR should be prescribed for both the operated and nonoperated hip to ensure maximum functional gains.

Our protocol took into consideration that hip musculature is typically healed by 4–5 mos postsurgery, minimizing the risk for spontaneous dislocation during testing protocols and further providing a basis for testing rehabilitation gains at this point in time. Indeed, many surgeons discontinue aggressive dislocation precautions by 4–6 mos postsurgery. During the course of this study we anecdotally observed that rehabilitation for THR subjects had been completed by 3 mos after surgery once functional goals such as stair climbing, ambulation in the community, and lower limb self-care were met and that employed patients had returned to full-time work. Moreover, at the 4–5 mos assessment, no THR subject reported pain on the day of testing, nor did THR subjects report pain during or after testing.

In our study, we conducted isokinetic performance testing to evaluate strength and dynamic neuromuscular performance. ^17–19 Compared with isometric measures, isokinetic measures require torque to be applied across a range of motion. For this reason, it has been suggested that isokinetic measures may be more closely associated with function than isometric measures. Indeed, studies have found correlations between isokinetic leg performance and balance in the elderly, suggesting a link to risk of falls. ^20,21 Some of the most reliable biomechanical muscle performance and endurance measures assessed using a robotic dynamometer are isokinetic peak torque, total work, and power. ^22,23 Although many studies have used work fatigue index (decrease of work production across repetitions) as a measure of endurance, Porter et al. ²³ recently affirmed that total work and isokinetic peak torque are more reliable indicators than work fatigue.

Although our study takes a first step toward attempting to quantify performance through isokinetic measures, it falls short in that there is not a direct link to functional tasks and performance of activities of daily living. Studies that correlate muscle performance to functional tasks and activities of daily living are also greatly needed to assess the effectiveness of current rehabilitation strategies. Future studies should also attempt to assess rehabilitation status of THR subjects using isokinetic performance on a longer-term basis, such as 1 yr postoperatively.

A limitation of our study is that exercises were performed in the same order of progression (flexion, extension, abduction) for all subjects, and thus, fatigue could potentially have had an effect on the relative ranking of strength and endurance across the various muscle groups. Furthermore, testing of THR subjects consistently started with the operated hip. This arrangement could present an advantage to the nonoperated hip due to a learning effect. However, our findings did not seem to result in an advantage given that nonoperated hip performance was statistically equivalent to operated hip performance. Future studies should, however, randomize the order of testing for all subjects. The use of age- and sex-matched healthy subjects could also have further enhanced our study; however, recruitment of subjects meeting inclusion criteria posed considerable challenges. Preoperative assessment of muscle performance would also be of value in describing the full continuum of progression in THR patients. Similarly, the effect of surgical approach on muscle performance was not evaluated in our study, but should be considered in future studies of THR recovery and rehabilitation outcomes.

Previous studies support our findings of reduced performance as demonstrated through diminished strength, ^5,7 quadriceps muscle atrophy, ⁷ and abnormal gait patterns. ^15,16 Although not a focus of this study, deficits in isokinetic strength and endurance may contribute to an underlying mechanism that affects gait patterns even when pain has been successfully treated with surgery. Our data suggest that patients may not be receiving adequate endurance and strength training postoperatively, perhaps due to ineffective rehabilitation protocols or shortened therapy programs. It is also possible that third-party restrictions regarding therapy services contribute to both of these potential factors. Brief postoperative rehabilitation protocols may not be adequate to restore THR patient hip function. Rehabilitation protocols that include endurance training and are longer in duration, as recommended by others, ^8,13,14 may be needed.

CONCLUSIONS

To date, few studies have utilized biomechanical measurements as a means to evaluate rehabilitation status after THR surgery. Our findings suggest that THR patients are not being restored to the same level as healthy adults from both a strength and muscular endurance perspective. Significant differences were found when comparing biomechanical hip performance across 4–5 mos postoperative THR subjects and healthy adults. Although several studies have evaluated various aspects of outcome after THR, the authors are aware of only one study in which THR patient performance was also compared with that of healthy older adults. Our findings also found no differences in biomechanical performance when comparing the operated with nonoperated hips in THR subjects. These findings provide a preliminary rationale for revising current approaches in THR rehabilitation protocols.

ACKNOWLEDGMENTS

We thank the following individuals for their assistance in recruitment of subjects: Raj K. Sinha, MD, Lawrence Crossett, MD, and the physicians and staff at the Division of Adult Reconstructive Surgery, University of Pittsburgh; Terry Breisinger, PT, Brian Urso, MS, OTR/L, and therapists at UPMC Rehabilitation Hospital, UPMC Shadyside Hospital, UPMC Southside Hospital, and UPMC St. Margaret’s Hospital; Edward McClain III, MD, and physicians and staff at Three Rivers Orthopedic Associates. We thank the Pittsburgh VA Healthcare System-HERL for the use of their laboratory and Biodex equipment and Fernando Aguel and Dongran Ha for their testing assistance.