Assessment of strength of individual digits in persons with osteoarthritis of the hand

Introduction

Osteoarthritis (OA) of the hand is common and known to cause a decrease in hand strength that exceeds the age-related strength deficits. ¹ Lower grip strength generally results in decreased hand function and many studies indicate that hand function is limited in persons with OA of the hand. ^2,3 During grip, forces are primarily exerted at the proximal interphalangeal (PIP), metacarpophalangeal (MCP) and carpometacarpal (CMC) joints. ^3–5 Clinical and radiological evaluation of the small joints of the dominant hand reveal that OA severity is greatest in the interphalangeal (IP) joints in the radial 3 digits. ⁶ In individuals without hand problems, strength of digits 2 and 3 makes a greater contribution to overall grip strength than digits 4 and 5. ^7–9 It follows that OA involvement of the radial digits may have a greater impact on deficits in grip and pinch strength. Thus, hand function, and limitations thereof, may have a stronger association with strength of the index and middle fingers than other measures of hand strength.

Hand therapists traditionally measure hand grip and pinch strength using digital dynamometry. ¹⁰ The NK Digits-Grip device permits measurement of the flexor strength of individual digits in addition to overall grip strength. ^7–9 Thus, individual finger strength is recorded during a multifinger task rather than during maximum voluntary contraction of individual finger flexor muscles. Although the maximum strength of each digit may be underestimated by this method (especially when the resistance is applied distal to the PIP joints), ¹¹ the NK Digits-Grip device could provide a measure of functional strength of each digit.

This method of digital dynamometry has been used previously to show that the ratio of radial and ulnar digit strength to overall grip strength is approximately 60:40 in persons without hand pathology. ^7–9 Only one publication reported the actual force values for the individual digits and subjects were 46 healthy men between 20 and 39 years of age. ⁹ The percentage contribution of digital strength to overall grip strength in persons with hand OA and whether OA involvement in the joints of a finger is related to digit strength is not known since this has not been investigated to date.

Purpose of the study

The purpose of this study was to determine the extent to which individual digit strength measures correlate with overall hand strength, function and pain in persons with OA of the hand, and thereby judge whether individual digit strength measures are relevant to the clinical assessment and rehabilitation of OA-related hand disability. Specifically, we address the following research questions:

What is the relative contribution of strength of digits 2–5 to total grip strength?

Methods

Measures

Strength of grip, individual digit flexion, tripod pinch, key (narrow and wide) pinch and dexterity were measured using components of the NK Hand Assessment System (NK Biotechnical, Minneapolis, MN, USA) and the corresponding standardized instructions. Strength of individual finger flexion and total grip were measured using the NK Digits-Grip device. The size of the grip on this device is similar to position 3 on the regular grip strength instrument of the NK apparatus (Figure 1A). Tripod pinch involves three fingers (thumb and digits 2 and 3) (Figure 1B). Key pinch involves positioning of the thumb on top and the radial side of digit 2 below. Narrow key was performed with the thumb and digit 2 spaced 12 mm apart (Figure 1C) and wide key with the thumb and digit 2 spaced 32 mm apart (Figure 1D). Dexterity of the dominant hand was assessed using the NK dexterity board and recorded as the time taken to manipulate small, and then large, objects. The average of three trials was the outcome measure for the strength and dexterity tests. Reliability of the standardized protocols has been established. ^13–15

OPEN IN VIEWER

Figure 1

Grip and pinch strength instruments of the NK apparatus. (A) Digits-Grip device for measuring force produced by individual digits and total grip, (B) tripod pinch device, (C) key pinch device positioned to measure narrow key pinch force and (D) key pinch device positioned to measure wide key pinch force

The Patient-Rated Wrist and Hand Evaluation (PRWHE) was used to measure self-reported pain and disability. The PRWHE contains three subscales: a five-item pain scale, a six-item specific activities subscale (patient describes difficulty with specific tasks) and a four-item usual activities subscale (patient describes difficulty in four domains of daily life). All questions were rated on a 0–10 scale, where 10 represents higher pain or disability. High reliability has been confirmed for subscale scores, ¹⁶ and the subscales for pain and specific activities were used in this study. The items in each scale were added to produce a pain subscale score (maximum 50) and a disability subscale score (maximum 60).

Involvement of the small joints of the hand was determined by a physiotherapist (JW) based on the following clinical examination. The presence of pain, swelling, nodules and/or deformities was recorded for the MCP, PIP and distal interphalangeal (DIP) joints of digits 2–5 and IP, MCP and CMC joints of digit 1.

Results

Table 1 summarizes the characteristics of the 104 participants (84 women). Women tended to be younger and shorter and have less hand strength. On average, women manipulated small objects more quickly than men, but the time required to manipulate large objects was similar for both genders. Self-reported pain and disability were similar. More women had involvement of digit 5, but most participants had involvement of digits 1–3 regardless of gender.

Strength was greatest for digit 3 compared with the other digits and greater for the radial digits combined (digits 2 and 3) than for the ulnar digits combined (digits 4 and 5) (Table 1, P < 0.001). The greater contribution of radial digit strength to total grip strength is illustrated in Figure 2. With the exception of digits 2 and 4, the percentage contributions of the four digits were significantly different from each other (P < 0.001).

OPEN IN VIEWER

Figure 2

Ratio of grip strength of individual digits, digits 2 and 3 combined (radial digits) and digits 4 and 5 combined (ulnar digits) to total grip strength (%) in the dominant hand of 104 persons with hand osteoarthritis

As expected, moderate to excellent positive relationships were observed between measures of hand muscle strength (Table 2). Digital involvement did not have a clear impact on the distribution of digit strength since there were no significant associations between involvement of individual digits (2–5) and strength of individual digits or total grip (eta = −0.05 to 0.18, P > 0.05). In contrast, involvement of digit 1 was significantly correlated with all measures of strength (digit strength: eta = 0.20–0.26, P ≤ 0.05; grip: eta = 0.26, P < 0.01; pinch: eta = 0.24–0.28, P ≤ 0.01).

The correlations between hand strength, age, pain, self-report disability and dexterity are presented in Table 3. Digit 2 and tripod pinch strength had a similar association with age (r = −0.22 and −0.24, respectively). Strength of digit 2, total grip and key pinch had similar associations with pain (r = −0.28 to −0.31). All measures of hand strength were related to dexterity with large objects (r = −0.25 to −0.37). With the exception of strength of digits 4 and 5, all strength measures were related to dexterity with small objects (r = −0.25 to −0.41). The same pattern was observed between strength and self-reported hand function, although the associations tended to be weak or non-significant (r ≤ −0.27).

Self-reported function (PRWHE-specific activities) explained less than 8% of the variance in performance-based function (dexterity with small and large objects, r = 0.28 and 0.24, respectively).

Discussion

This study is the first to demonstrate that, while grip strength in persons with hand OA is lower than age- and gender-matched normative values, the percentage contribution of individual digit strength to total grip strength is preserved. Despite the fact that our subjects had more OA involvement of their radial digits than their ulnar digits (Table 1), the percentage contribution to total grip strength of the radial and ulnar digits (59% and 41%, respectively) was remarkably similar to that previously reported for younger adults without hand pathology (61% and 39%, respectively). ^7–9 The paper which reported mean values for individual digits tested healthy young men who had values 2–2.5 times higher than the 20 men in the present study. ⁹ Similarly, the magnitude of total grip strength in our volunteers was lower as compared with age- and gender-matched normative values. ¹⁸ However, values for total grip were similar to previous reports for men and women with OA of the hand. ² The pattern of digit strength suggests global deficits in the force generated by the extrinsic and intrinsic finger flexor muscles ¹¹ and do not appear to reflect inhibitory mechanisms arising from OA involvement in individual digits.

Measures of strength in individual digits (2–5) have moderate to excellent positive associations with total grip (0.74–0.93) and tripod pinch strength (0.56–0.74). However, the strength of digits 2 and 5 explain only 55% and 56% of the variance in total grip strength, respectively. Individuals with greater strength in total grip, tripod pinch and wide key pinch require less time to manipulate large objects with the strength of the association being comparable for all three strength measures (see confidence intervals in Table 3). Similarly, individuals with greater digit 2 and tripod pinch strength require less time to manipulate small objects. Pinch strength has the strongest association with self-reported pain and function. The very high correlation observed between measures of narrow and wide key pinch (r = 0.91) indicates that it is unnecessary to measure strength using both types of key pinch in individuals with hand OA. Taken together, our findings suggest that measurement of pinch strength (tripod and narrow or wide key) and individual digit strength provides unique information regarding hand function in persons with uncomplicated hand OA.

The presence of pain, swelling, nodes and/or deformity in specific fingers did not result in preferential loss of grip strength in those digits. Similarly, total grip strength and pinch (tripod and narrow and wide key) strength were not related to the presence of clinical features of OA in individual digits. In contrast, OA involvement of the thumb had a weak (but statistically significant) association with all measures of hand strength. As the most force during tripod pinch is on the CMC-I ⁵ and the thumb is required to stabilize the hand dynamometer during total grip strength testing, signs and symptoms in the thumb require particular attention in order to maximize strength in the OA hand. These results indicate that thumb involvement may be a limiting factor in production of force for patients with hand OA.

Limitations of our study must be considered when interpreting these results. One limitation is that we investigated the clinical utility of individual digit strength in persons with hand OA using a cross-sectional design, which permits only correlational analyses. Sensitivity to change and predictive validity of these strength measures were not investigated. A second limitation associated with testing muscle strength is the inability to confirm that the participants produce maximum force during their voluntary effort. We addressed this challenge by averaging three consistent trials for each measure of hand strength, which resulted in maximum grip forces that compare favourably with values previously reported for similar study participants. ² A third limitation is that our study confirms the previously reported percentage contribution of digit strength to overall grip strength using the same method of assessment (Digits-Grip device). Previous work comparing digit force generated during isolated testing of that digit with the force generated in that digit during a four-finger task, suggests that using the Digit-Grip device to record individual digit strength in all four fingers simultaneously may underestimate individual finger force. ¹¹ Finally, the construct of hand disability was evaluated using a single self-report measure and a single performance-based measure. The relations between hand strength and function reported in this study may not apply to different measure(s) of hand disability. Further research is needed to confirm our findings in longitudinal studies using a variety of methodologies available for measuring impairment, physical function and disability in persons with varying severity of hand OA.

The purpose of this study was to determine the extent to which individual digit strength measures correlate with overall hand strength in persons with OA of the hand, and thereby judge whether individual digit strength measures are relevant to the clinical assessment and rehabilitation of OA-related hand disability. Although persons with hand OA demonstrate generalized loss of hand strength, our findings suggest that assessment of individual digit strength (particularly digit 2) may provide useful information regarding functional status and effectiveness of strengthening exercises. The weak associations between physical impairment measures, self-reported function and performance-based measures of hand dexterity indicate that these tests provide different information and all should be included in a comprehensive assessment of persons with hand OA.