Background: Knee joint arthritis causes pain, decreased range of motion, and mobility limitation. Knee replacement reduces pain effectively. However, people with knee replacement have decreases in muscle strength (“force-generating capacity”) of the involved leg and difficulties with walking and other physical activities.
Objective and Design: The aim of this cross-sectional study was to determine the extent of deficits in knee extensor and flexor muscle torque and power (ability to perform work over time) and in the extensor muscle cross-sectional area (CSA) after knee joint replacement. In addition, the association of lower-leg muscle deficits with mobility limitations was investigated.
Methods: Participants were 29 women and 19 men who were 55 to 75 years old and had undergone unilateral knee replacement surgery an average of 10 months earlier. The maximal torque and power of the knee extensor and flexor muscles were measured with an isokinetic dynamometer. The knee extensor muscle CSA was measured with computed tomography. The symmetry deficit between the knee that underwent replacement surgery (“operated knee”) and the knee that did not undergo replacement surgery (“nonoperated knee”) was calculated. Maximal walking speed and stair-ascending and stair-descending times were assessed.
Results: The mean deficits in knee extensor and flexor muscle torque and power were between 13% and 27%, and the mean deficit in the extensor muscle CSA was 14%. A larger deficit in knee extension power predicted slower stair-ascending and stair-descending times. This relationship remained unchanged when the power of the nonoperated side and the potential confounding factors were taken into account.
Limitations: The study sample consisted of people who were relatively healthy and mobile. Some participants had osteoarthritis in the nonoperated knee.
Conclusions: Deficits in muscle torque and power and in the extensor muscle CSA were present 10 months after knee replacement, potentially causing limitations in negotiating stairs. To prevent mobility limitations and disability, deficits in lower-limb power should be considered during rehabilitation after knee replacement.
With an aging population, the prevalence of degenerative joint diseases, such as knee joint arthritis, increases and thus adds to the burden of health care systems in Western societies. Knee joint arthritis causes pain, decreased range of motion, and mobility limitations. Knee joint replacement is a common surgical procedure that effectively reduces pain.1–4 However, several studies2,5–12 have shown that people with knee replacement surgery have difficulties with walking and other physical activities. Mobility undergoes an expected decline during the first month after knee replacement.7 Mizner et al7 reported that performance in stair-climbing and “stand-up-and-go” tests returned to the preoperative level at 2 months after surgery. Therefore, although functional ability may improve to the preoperative level, which already is severely impaired because of pain and long-term disuse, it rarely reaches the level in age-matched control subjects.5,12,13 For example, Walsh et al5 and Yoshida et al12 reported that people with knee replacement had a lower maximal walking speed5,12 and negotiated stairs more slowly5 than control subjects even beyond 1 year after surgery.
Mobility limitations are known to be associated with decreases in muscle strength (force-generating capacity) and power (ability to perform work over time). These impairments continue to persist for several months after surgery.5,7,8,10,14–16 Several investigators7,11,14,17–19 have reported declines of 21% to 42% in knee extensor torque and power for the knee that underwent replacement surgery (“operated knee”) compared with the knee that did not undergo replacement surgery (“nonoperated knee”) at 3 to 6 months after surgery. Furthermore, even at 1 to 2 years after knee replacement surgery, a difference of 12% to 29% between the knee extensor muscles has been reported.5,9,17 Similar deficits have been reported for knee flexor muscle strength.5,14 Knee extensor muscle strength has been reported to remain 19% to 35% lower in people with knee replacement than in age-matched people, even at 13 years after surgery.5,11,20–22
Previous studies11,18,23,24 indicated that there is a decline in the knee extensor muscle cross-sectional area (CSA) of the operated leg during the early recovery phase—1 to 3 months after surgery—compared with the preoperative CSA. To our knowledge, no studies comparing muscle CSA between the legs or reporting spontaneous long-term recovery of muscle CSA after knee replacement surgery have been done.
Mobility limitations may be related to lower-limb muscle deficits, that is, side-to-side differences between the operated leg and the nonoperated leg. Previous studies showed that in people who are healthy25,26 and in some clinical populations,27–29 lower-limb power deficits have detrimental effects on mobility. Portegijs et al25 reported that in people who were healthy, knee extensor power asymmetry was associated with a lower walking speed. Additionally, they found that in women recovering from hip fractures, a larger power deficit was associated with limitations in stair-climbing ability.29
To date, little is known about muscle deficits and their persistent effects on mobility limitations in people with knee replacement. Therefore, the purpose of this study was to determine the extent of muscle deficits in knee extensor and flexor muscle torque and power and in the extensor muscle CSA and composition in a group of people who had undergone unilateral knee replacement an average of 10 months earlier. In addition, the association of lower-limb muscle deficits with mobility limitations was investigated.
Setting and Participants
A total of 201 people who, according to the physical therapy records of Kymenlaakso Central Hospital, had undergone unilateral knee replacement 4 to 18 months before the study were informed about the study. Eighty-six people contacted the research personnel. People with bilateral knee arthroplasty, revision arthroplasty, hemiarthroplasty, severe cardiovascular diseases, dementia, rheumatoid arthritis, or major surgery on either of the knees were excluded from the study. Thus, 48 eligible volunteers (29 women and 19 men; age range=55–75 years) participated in the study.
The physical characteristics of the participants are shown in Table 1. All of the participants had undergone knee replacement surgery with cement fixation. Eight of the 48 participants had osteoarthritis diagnosed in the nonoperated knee.
The data used in this cross-sectional study were collected in 2 phases. Because of the small number of eligible subjects in spring 2005, the data collection was repeated in autumn 2005 with the same recruitment protocol, infrastructure, and staff. Before the laboratory examinations, the participants were informed about the study and gave written informed consent.
The clinical history of the participants, including their medications and diseases, were confirmed by a physician before the laboratory examinations. Body height and body weight were measured by use of standard procedures. The day-to-day intrarater reproducibility of the measurements (muscle torque, power, walking speed, stair ascending, and stair descending) was measured in our laboratory with a pilot sample. The measurements were performed twice by use of identical procedures, with an interval of 1 week between the measurement occasions.
In a pilot study (unpublished), 17 volunteers (12 women and 5 men; mean age=77 years, range=55–75) with unilateral knee replacement an average of 8 months (range=4–12) after surgery participated in the measurements. The intraclass correlation coefficient (ICC) was calculated by use of a 1-way random model. The participants in the pilot sample were not included in the sample in the current study. The reliability (ICC) of each measurement is presented in context with the measurement.
Muscle torque and power.
The maximal isokinetic torque (N·m) of the knee extensor and flexor muscles was measured by use of an isokinetic dynamometer* with a sampling frequency of 100 Hz and a measurement error of 1% through the entire range of motion. The dynamometer was calibrated before each measurement session according to the standard procedure recommended by the manufacturer. Before the measurement session, the participants were carefully familiarized with the testing procedure.
For each leg, the axis of rotation of the dynamometer was aligned with the condylus lateralis femoris. The lever arm of the dynamometer was attached around the ankle 2.5 cm above the midpoint of the malleolus lateralis. The hip and thigh were stabilized with straps. The full knee range of motion was measured. The nonoperated leg was measured first. After a few submaximal flexion-extension movements, 3 maximal continuous flexion-extension trials were performed at an angular velocity of 60°/s, and 5 trials were performed at a velocity of 180°/s, with 2 to 3 minutes of rest between trials. The participants were verbally encouraged to make a maximal effort throughout the whole range of motion. The highest peak torque (N·m) at an angular velocity of 60°/s was analyzed. Peak power was analyzed in extension and flexion at an angular velocity of 180°/s. The ICC of the isokinetic parameters for the operated knee in the people with knee replacement varied between .90 and .97.
Muscle CSA and attenuation.
Computed tomography (CT) scans were obtained from both midthighs by use of a Siemens Somatom DR Scanner† with the subject in a supine position. The midthigh was defined as the midpoint between the greater trochanter and the lower edge of the patella. The scans were analyzed by use of software developed for cross-sectional CT image analysis (Geanie 2.1‡), which separates fat and lean tissues on the basis of radiological density (measured as attenuation in Hounsfield units) limits. The quadriceps femoris muscle was determined manually by drawing a line along the fascial plane. A lower mean attenuation value reflects greater fat infiltration within the muscle. The Figure shows an example of the CT analysis. The CT measurements and analyses were conducted in a masked fashion. In our previous study,30 the coefficients of variation between 2 consecutive repeated measurements were calculated and shown to be less than 1% for lean tissue Hounsfield units and 1% to 2% for the CSA.
Maximal walking speed over 10 m was measured in the hospital corridor. Walking time was recorded by use of photocells.§ Participants were instructed to walk as fast as possible without compromising their safety. All participants wore thin aquatic shoes and were allowed 3 m for acceleration. Each participant performed 2 trials, separated by a 1-minute rest period, and the fastest time was accepted as the best result. The ICC for maximal walking speed in people with knee replacement was .86.
Times to ascend and descend a 10-step staircase were measured in the hospital corridor. The stair height was 17 cm, and the depth was 29.5 cm. The participants were instructed to step alternately on each stair and walk as fast as possible without compromising their safety. The use of a handrail or taking a step on each stair with both feet was allowed only when necessary. Three participants stepped on each stair with both feet in the stair-ascending task, and 7 did so in the stair-descending task. Ascending and descending times were recorded by use of photocells.§ Each participant performed 2 ascending trials, followed by a 1-minute rest period, and then performed 2 descending trials. The fastest times were accepted as the best results. The ICCs were .90 for stair ascending and .73 for stair descending in the participants with knee replacement.
The differences in muscle characteristics (torque, power, CSA, and attenuation) between the operated leg and the nonoperated leg were analyzed with a paired 2-tailed Student t test. The muscle symmetry deficit (relative difference) was calculated according to the following equation: symmetry deficit (%) = [(value for nonoperated leg − value for operated leg)/value for nonoperated leg] × 100. Stepwise multiple linear regression models were used to examine the most relevant muscle deficit (muscle torque, power, CSA, and attenuation) and muscle power variable associated with mobility limitations. Variables with nonsignificant independent associations with mobility were removed from the final model. Thus, the final model contained only the explanatory variables that had significant independent associations with mobility limitations and that had the highest possible proportion of the variance explained by coefficients of determination (adjusted R2). The models were further adjusted for age, sex, and time after surgery. For the regression analysis, the results obtained for men and women were pooled because there were no sex differences in age, time after surgery, or any of the muscle deficit variables. Significance was set at P<.05. Statistical analyses were run with SPSS (version 13.0) software.‖
Knee Extensor Muscles
For the entire group, the mean knee extensor torque, power, CSA, and attenuation values for the operated side were significantly (P<.001) lower than those for the nonoperated side. For the knee extensor muscles, more than 97% of the participants had lower or equal values in the operated leg than in the nonoperated leg. The mean knee extension torque deficit was 27% (95% confidence interval [CI]=22%–32%), and the mean knee extension power deficit was 23% (95% CI=17%–29%). The mean knee extensor muscle CSA deficit was 14% (95% CI=11%–18%), and the mean knee extensor attenuation deficit was 9% (95% CI=6%–11%). The results for the knee extensor muscles are shown in Table 2.
Knee Flexor Muscles
For the entire group, the mean knee flexor torque and power values for the operated side were significantly (P<.001) lower than those for the nonoperated side. For the knee flexor muscles, over 87% of the participants had lower or equal values in the operated leg than in the nonoperated leg. The mean knee flexion torque deficit was 13% (95% CI=7%–19%), and the mean knee flexion power deficit was 19% (95% CI=11%–27%). The results for the knee flexor muscles are shown in Table 3.
For the entire group, the mean (SD) maximal 10-m walking speed, stair-ascending time, and stair-descending time were 1.9 (0.5) m/s, 5.1 (2.4) seconds, and 5.6 (3.7) seconds, respectively. The results for mobility are shown separately for women and men in Table 1.
Multivariate regression analysis was performed to examine the association among muscle deficit, muscle power production, and negotiating stairs (Tabs. 4 and 5). A larger knee extension power deficit, together with low knee flexion power on the nonoperated side, predicted slower stair-ascending time (Tab. 4). Adjustments for age, sex, and time after surgery did not materially change the association. In addition, a larger knee extension power deficit, together with low knee flexion power on the nonoperated side, predicted slower stair-descending time (Tab. 5). Adjustments for potential confounding factors did not materially change the association.
The results of this study showed that at an average of 10 months after knee replacement surgery, the operated leg was significantly weaker than the nonoperated leg, and the extensor muscle CSA in the operated leg was smaller than that in the nonoperated leg. A larger knee extension power deficit predicted slower stair-ascending and stair-descending times. This relationship remained unchanged when the power of the nonoperated side and potential confounding factors were taken into account. Lower-limb muscle power, especially the difference between the legs, seemed to be critical for mobility limitations; therefore, it should be considered during evaluations of mobility in both people who are healthy and people who have disabilities.
In the majority of the participants, the operated leg was weaker than the nonoperated leg, and the muscle CSA of the operated leg was smaller than that of the nonoperated leg. The results of the present study are in line with those of previous studies that investigated muscle force 6 to 12 months after unilateral knee replacement.5,7,14,17 A comparison of the results of the present study and those previously reported is difficult because we calculated the muscle strength deficit for each participant individually, whereas in earlier studies, side-to-side differences were estimated from group mean values. Overall, the deficit in the knee extensor muscles after knee replacement surgery is considerable and can also be prolonged; the leg with the knee replacement has been shown to be significantly weaker than the legs of healthy control subjects for as long as 13 years after the surgery.5,20–22 Previous studies7,14,17 showed a difference of 15% to 29% in knee extension torque between the operated leg and the nonoperated leg, which is in line with the 27% difference found in the present study. Rossi and Hasson,16 however, reported a marked, 38% difference in the findings for a single leg press between the operated leg and the nonoperated leg at 16 months after knee replacement. This large side-to-side difference may have been attributable to the multiple muscle groups involved in the leg press.
We also found a marked knee flexor torque deficit (ie, 13%) after an average of 10 months from knee replacement. This finding supports the results of 2 earlier studies reporting a side-to-side difference of 16% to 23% in knee flexor muscles at 6 to 12 months after surgery.5,14 Therefore, this muscle group should receive attention during assessment and rehabilitation of degenerative knee joint problems.
We also found 19% to 23% deficits in knee extensor and flexor muscle power; these values were somewhat higher than the value reported by Lamb and Frost,6 who found a difference of 18% in leg extension power at 6 months after knee replacement. This substantial deficit should be taken into consideration in rehabilitation programs because in daily activities it is important to have the muscle power needed to produce effective force quickly to generate desirable or prevent undesirable movements. In particular, the ability to recover from a stumble is highly dependent on the power and coordination of the leg muscles.31–33 In addition, Portegijs et al26 found that, even in people who were healthy, a knee extension power deficit was associated with falls. Although we did not evaluate falls after knee replacement in the present study, we would argue in accordance with the literature26,28 that a power deficit should be taken seriously as a risk factor for falls and therefore should be considered in knee replacement rehabilitation.
The extensor muscle CSA deficit was marked (14%) in the present study. To our knowledge, a long-term muscle CSA deficit has not been studied. Previous studies11,18,23,24 showed declines of 5% to 20% (relative to preoperative values) in knee extensor muscle CSA in the operated leg at 1 to 3 months after knee replacement. In the present study, the most likely reason for the large side-to-side difference, in addition to long-term pain and disuse because of osteoarthritis, was the surgery itself, which resulted in a long wound, considerable surgical trauma, and a long recovery time. In people with hip osteoarthritis after prolonged unilateral disuse, the preoperative side-to-side difference in quadriceps muscle CSA between the affected leg and the nonaffected leg has been reported to be smaller (8%–10%).34 Loss of muscle CSA (atrophy) is an important mechanism underlying muscle weakness, although the amount of muscle CSA lost is often smaller than the amount of muscle force lost.35 A muscle CSA deficit of 14% may present a challenge for rehabilitation because even in older subjects who were healthy, a progressive strength training regimen lasting 3 to 4 months was shown to have an effect of less than 10% on muscle CSA.30,36
Decreased lower-limb muscle power is one of the factors underlying mobility limitations in older adults.37–39 Mizner et al7 and Mizner and Snyder-Mackler8 reported that weakness of the knee extensor muscles in people with a total knee replacement was closely associated with mobility limitations, especially in stair-climbing tasks and the Timed “Up & Go” Test. According to Lamb and Frost,6 leg extension power is an important determinant of walking speed and stair-ascending time after knee replacement. Portegijs et al25 reported that extension power asymmetry was also associated with a lower walking speed in older women who were healthy. In the present study, large power and torque deficits were associated with slow stair-ascending and stair-descending times but not with maximal walking speed. This result is in line with the results of Portegijs et al,29 who found that in women recovering from hip fracture, a large power deficit was associated with limitations in stair climbing but not with walking speed. It would appear that because walking is a common functional task, the nonoperated leg may be able to compensate for problems with the operated leg. However, to perform more-demanding functional tasks, such as stair ascending and stair descending, a person needs more power and force production in the knee extensor muscles.7,8
The present study had some limitations. The study was a cross-sectional analysis without follow-up; therefore, we cannot speculate on the causal relationships or the associations over time. The study population consisted of people who were relatively healthy and mobile and had undergone successful unilateral knee replacement procedures. It is impossible to know whether people with more-extensive mobility problems might have dropped out; such a situation might have reduced the variance in muscle deficits and in mobility problems. In addition, some of the participants had osteoarthritis in the nonoperated knee, and this condition may have influenced the muscle deficits in the lower legs. The clear strength of the present study is the large number of measurements of deficits in muscle torque, power, and CSA. The results of this cross-sectional study need to be confirmed in future prospective and experimental studies.
Deficits in muscle power or torque are clinically important during evaluations of mobility limitations up to nearly 1 year after surgery. Because the major goals in the rehabilitation of musculoskeletal problems are to restore a person's mobility and functional capacity and to prevent mobility disability, increasing muscle power, especially in the operated leg, may be one of the central issues to address during the rehabilitation process. The findings of this study are potentially useful for planning preventive and rehabilitative strategies; however, further work is needed.
All authors provided concept/idea/research design, writing, and data collection and analysis. Dr Pöyhönen and Dr Heinonen provided project management. Ms Valtonen, Dr Pöyhönen, and Dr Heinonen provided fund procurement. Ms Valtonen and Dr Pöyhönen provided participants. Dr Pöyhönen provided facilities/equipment. Dr Heinonen provided consultation (including review of manuscript before submission).
The study was approved by the ethics committee of Kymenlaakso Central Hospital.
An abstract and oral presentation of this research were given at the 18th Nordic Congress of Gerontology; May 28–31, 2006; Jyväskylä, Finland; and at the 8th Scandinavian Congress of Medicine and Science in Sports; November 9–12, 2006; Vierumäki, Finland.
↵* Biodex Medical Systems Inc, 20 Ramsey Rd, Shirley, NY 11967-4704.
↵† Siemens AG, Erlangen, Germany.
↵‡ Commit Ltd, Espoo, Finland.
↵§ Newtest Oy, Koulukatu 31 B 11, FIN-90100, Oulu, Finland.
↵‖ SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.
- Received October 3, 2007.
- Accepted June 19, 2009.
- American Physical Therapy Association