Isokinetic testing of muscle strength in older adults with knee osteoarthritis: An integrative review

PubMed was searched for articles potentially related to isokinetic testing and knee arthritis on July 7, 2019. The search string included “isokinet * AND knee osteoarthritis.” Article titles and abstracts were initially reviewed for relevance and if warranted the full text was then examined to determine if the articles addressed isokinetic strength testing for individuals with knee OA. Articles were excluded if published in a language other than English, were review articles, failed to provide procedural specifics or presented no findings relevant to the clinimetric properties of isokinetic testing. Articles retained for this review were mined for information on the sample tested, procedural specifics (e.g., speed and movement tested), and clinically relevant findings.

3. Results and discussion

The PubMed search identified 312 potentially relevant articles. After review, 183 articles were excluded leaving 129 articles for inclusion in this report. Relevant information from these articles are summarized in Table 1 [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136].

The summarized information demonstrates that isokinetic dynamometry is frequently utilized to measure strength of the knee flexor and extensor muscles in the presence of OA. Testing has been reported throughout the world with Asia, Europe, North America, and South America producing the bulk of published work. Various dynamometers have been used although Biodex, Cybex, En-Knee, and Kin-Com are most frequently reported. Most protocols have measured concentric contractions, with a primary outcome of peak torque (PT) or PT adjusted for body weight while fewer studies have reported work, power, fatigue, and agonist/antagonist ratio. Isokinetic test velocities range from 30 to 240 ∘ /s, with most studies utilizing 60 ∘ /s – which appears to produce larger PT values than higher or lower velocities [12, 72, 130].

Isokinetic knee strength measurements have demonstrated convergent validity with several related factors. Knee flexor and extensor strength have been shown to be strongly correlated ( r = 0.79 to 0.97) [61, 67, 74, 123, 124], with knee extensor testing producing larger torque values than flexor testing [72]. Concentric and eccentric measurements have been reported to be highly correlated ( r = 0.61 to 0.88), eccentric efforts demonstrating larger torque values than concentric efforts for both the knee flexor and extensor muscles [53]. A strong correlation exists between the right and left lower extremities for knee flexor and extensor ( r = 0.80 to 0.90) strength [46, 124]. Isokinetic strength has demonstrated strong correlations between testing velocities for the knee extensors [20]. Muscle cross-sectional area and isokinetic strength ( r = 0.49 to 0.78) have been found to be positively correlated with cross-sectional area-partially explaining the variability in isokinetic strength performance [53]. Body mass index (BMI) may or may not be associated with lower knee strength in the presence of knee OA [65, 108].

The clinimetric properties of isokinetic knee strength measurement have been supported via known groups or conditions validity in various ways. Dominant limb strength is larger than non-dominant limb strength [79] and strength measurements in the lower extremity affected with knee OA tend to be lower than in the unaffected limb [64, 67, 122, 130] but no difference has also been reported [87]. Knee extensor strength tends to be greater in individuals without knee OA than with knee OA [8], younger adults have greater strength than older adults [8, 66, 88], and middle-aged adults with knee OA have similar strength to older adults without knee OA [8]. Knee flexor and extensor strength also tends to be greater in males than females [18, 50, 51, 61, 105, 107]. Body weight is inversely correlated with knee extensor strength more frequently in females than males [29, 115], and those who develop incident knee OA [115].

Healthy adults tend to have stronger knee flexors [23, 30, 33, 45, 48, 79, 115, 117] and extensors [19, 21, 23, 30, 33, 38, 45, 48, 66, 79, 82, 91, 110, 114, 115, 117, 132] than individuals with knee OA, however some studies have reported no difference in the flexors [30, 66] and extensors [38, 120] based upon presence or absence of knee OA. Knee flexor [19, 33, 45] and extensor [20, 33, 45] strength tends to be larger in the early stages of OA than later in the natural course, although some studies have reported no strength differences based upon disease severity [29, 51]. Isometric knee strength is reportedly lower in the presence of knee OA [21, 25, 60, 117] and worsens with disease severity [66] with few exceptions [20, 91]. A correlation appears to exist between knee cartilage integrity and isometric knee flexor and extensor strength [66, 121], but not isokinetic strength [121]. The variability in knee extensor isokinetic and isometric strength may be partially explained by patellofemoral cartilage integrity, extent of cartilage lesions, loose bodies, synovitis or effusion, and pain [20]. Significantly higher medial and lateral patellofemoral joint cartilage damage and lateral patellofemoral bone marrow lesions have been found in those in the lowest tertile of knee extensor strength [116].

Pain appears to be negatively correlated with knee strength ( r = - 0.33 to - 0.22) [51] with individuals experiencing knee pain demonstrating lower strength than pain free controls or pain free knee OA participants [30, 38, 63, 82, 114]. An inverse correlation between pain catastrophizing and isometric knee flexor strength has been reported [21]. Muscle strength may not contribute to pain variability [75]. An inverse correlation of pain catastrophizing with isometric knee flexor strength and kinesiophobia with isometric knee extensor strength have been reported [21]. Functional self-efficacy has demonstrated a correlation with knee flexor and extensor strength, with the variability in self-efficacy partial explained by knee stiffness, flexor strength, age, and depression [74].

Functional measurements have been associated with knee strength measurement utilizing isokinetic dynamometers. Knee strength has been identified as inversely associated with functional ambulation measures including the Get Up and Go Test [13, 46, 102, 124, 126], Timed Up and Go Test [26, 49], 6 minute Walk Test [49, 134], 100-meter walk test [123, 124], 20-meter walk test [18], gait speed [92, 93, 122], and walking time [67]. Difficulty rising from a chair has been associated with lower knee extensor strength in females [18]. Functional performance on the 30-second Chair Stand Test has been correlated with knee flexor and extensor strength ( r = 0.51 to 0.56) [49]. Stair ambulation time tends to be inversely correlated to knee strength [26, 46, 49, 53, 67, 102, 122]. Knee extensor strength has been associated with lower levels of impairment on the Knee Injury and Osteoarthritis Outcome Score (KOOS) [94], Western Ontario McMaster Universities Osteoarthritis Index (WOMAC) [13, 14, 51, 102, 109, 123, 124, 126] and Short Form-36 [92] physical function subscales. There may [40] or may not [63, 88] be an association between knee strength and proprioception. An association between knee strength and balance may also be present [61, 63].

Other contributing factors association with knee strength have been reported in patients with knee OA. An inverse correlation may be present between markers of systemic inflammation [40, 103] or vitamin D levels [65] and knee strength in participants with knee OA [40, 103]. Knee strength appears to be negatively influenced by comorbidity [51] and sleep apnea [112]. A history of falling may [41] or may not [10] be associated with lower knee strength values. Levinger et al. [69] have reported fewer steps after an induced forward fall in individuals with higher knee extensor strength performance.

The relative reliability of isokinetic dynamometry in measuring knee strength in the presence of knee OA has been characterized using intraclass correlation coefficients (ICCs). Test-retest reliability for knee isokinetic strength measurement has been noted as strong with ICCs ⩾ 0.72 in all studies [11, 13, 17, 19, 20, 29, 32, 53, 62, 72, 106, 107, 116, 130] except that of Carpenter et al. [32], who reported a range of ICCs from 0.60 to 0.95. Despite strong test-retest reliability, strength values may significantly increase in novice individuals after familiarization with isokinetic measurement suggesting a learning effect [95]. Minimal detectable change (MDC) values have been reported for concentric knee extensor strength PT (33.9 Nm) [62] and body weight adjusted PT for concentric knee extensor (0.03 to 0.05 Nm/kg) and eccentric knee extensor strength = 0.14 to 0.19 Nm/kg [17].

Numerous studies have documented significant increases in knee strength after exercise interven- tions [11, 15, 16, 27, 42, 43, 47, 52, 54, 55, 56, 57, 58, 59, 60, 64, 73, 78, 80, 82, 85, 87, 89, 90, 96, 97, 98, 121, 129, 135] including strength/power training [27, 47, 55, 56, 80, 82, 85, 89, 90, 121, 135], isokinetic exercise [47, 52, 57, 58, 59, 64, 73, 78, 98, 129], aerobic exercise [11, 80, 87, 90], stretching [85, 90], kinesthesia and balance exercises [43], physical therapy/ physiotherapy [24, 60, 97], hydro/aquatic therapy [42, 54], Baduanjin [15, 16], and exercise education [96]. Two studies reported no difference in knee strength subsequent to physiotherapy [28] or aquatic or land-based exercise [70] both however conflicting with the positive effects of exercise reported in the majority of studies.

Thermal modalities including short wave dia- thermy [9, 34], hot packs [34], laser [31], and ultrasound [34, 58, 59] have reportedly been associated with additional improvements in knee strength over exercise interventions alone. Electric modalities including transcutaneous electric stimulation [37], neuromuscular electric stimulation [77, 119], and biofeedback [135] have also been reported to produce improvements in knee strength when combined with exercise interventions. Whole body vibration training in combination with exercise may [27] or may not [85] produce larger increases in knee strength than exercise training alone, or maslinic acid supplementation [136]. Kinesiotape tends to produce no difference in knee strength compared with sham taping [84, 128]; conflicting results have been reported relative to control applications [17]. Although sparsely reported, knee extensor strength may improve with application of a magnetic knee wrap [35] or knee brace [76]. Intraarticular hyaluronic acid injections tend to increase knee strength measures [44, 58, 83, 118] with one study reporting no difference [24]. Wu et al. [133] reported increased knee strength after platelet rich plasma injections although no difference was identified compared to placebo injection.

The available evidence of isokinetic measurement of knee strength the presence of knee OA is extensive, dating to at least the work of Lankhorst et al. [67] in 1985. Numerous studies have evaluated the association of knee strength and the presence and severity of knee OA, age, gender, and body type. Isokinetic dynamometry has been shown to be a valid and reliable measure of knee flexor and extensor muscle strength. In the presence of knee OA, change in knee strength over time and with various interventions have been reported.

This review utilized only one database (PubMed) which may have limited the pool of available articles meeting the search and inclusion criteria. It is unlikely, however, that the inclusion of additional databases would substantially alter the summary provided in this review.

4. Conclusion

Isokinetic measurement of knee strength is well-supported for individuals with knee OA. The evidence supports utilization of isokinetic measurement for the identification of strength impairments and subsequent responsiveness to therapeutic interventions.