Functional exercise capacity in rheumatoid arthritis unrelated to lung injury: A comparison of women with and without rheumatoid disease

1. Introduction

Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease characterized by symmetric peripheral polyarthritis that leads to deformity and destruction of the joints [1, 2]. In Brazil, the prevalence of up to 1% was found in the adult population, with the disease occurring more often in women than in men at a ratio of 3:1 [3]. RA causes long-lasting and repeated episodes of inflammation with damage to the articular cartilage and the underlying bone accompanied by symptoms such as pain, stiffness, swelling, and fatigue. The long-term outcome is characterized by permanent changes in the structure and functioning of the joints, which can lead to loss of functional capacity and poor health-related quality of life (HRQoL) [2, 4]. The mortality rate in patients with RA is 1.65 times higher than that in the general population and is associated with the long duration of the disease and the presence of comorbidities [5].

The pathophysiological process of RA results in the progressive involvement of multiple joints [4]. Evaluations of these subjects aim mainly to detect joint inflammation, focusing on the joints of the fingers, wrists, knees, and feet [6]. In more than 90% of individuals with RA, the hand joints are affected, resulting in reduced strength, dexterity, and mobility of the fingers and wrists, which complicates performing activities of daily living (ADLs) [7, 8]. However, pain and deformities of the joints of the lower limbs are also common in RA patients, and approximately 80% of individuals with RA report problems in the feet [9, 10]. In addition to affecting the joints, RA causes decreased peripheral muscle strength and mass, with functional deficits of up to 70% in more severe patients [11, 12]. Chronic inflammation is considered the main mediator of muscle dysfunction; it accelerates cellular catabolism and changes the synthesis-degradation ratio, causing accelerated loss of muscle mass [11]. In addition, RA is accompanied by detrimental changes in body composition that favour an increase in fat mass deposition and a reduction in muscle mass, further deteriorating muscle function and the ability to perform ADLs [1].

Despite advances in the treatment of RA in recent decades, functional disability is still common in these individuals. RA is an aggressive disease with a great impact on HRQoL and well-being; therefore, patient-based measures are especially important because they assess an affected individual’s perspective in relation to the disease burden. Such measures focus on assessing pain, systemic conditions, and physical disability, the latter of which is most frequently assessed with the Health Assessment Questionnaire Disability Index (HAQ-DI) in people with RA [13]. The HAQ-DI is an interesting measure of physical functioning because poor functioning is a predictor of mortality and is associated with lower HRQoL and disability at work [14]. The HAQ-DI is a questionnaire with good reproducibility and sensitivity used to measure changes occurring in an individual [4, 14]. However, its correlation with objective measures of physical limitations (especially those obtained during effort) is not good [4].

Taken together, joint and muscle changes can potentially impact the performance of individuals with RA on tests that assess functional capacity. In this context, early functional assessment of the performance of both the upper and lower limbs is fundamental to obtain data that inform the selection of a therapeutic strategy and to evaluate the efficacy of treatments. Several tests have been used to evaluate the functional capacity for exercise in individuals with RA, including cardiopulmonary exercise tests and the 6-min walk test (6 MWT) [15, 16, 17]; however, no attention has been directed towards upper limb mobility, which is fundamental for ADLs in this population. Thus, evaluating the applicability of submaximal tests that better mimic ADLs and simultaneously incorporate the joints and muscle groups of various body segments in patients with RA has attracted interest.

To this effect, the Glittre-ADL test (G-AT) was developed to address the need for a broader and more representative objective evaluation of functionality using ADL-like activities [18]. The G-AT can not only be used to evaluate upper limb function but also does not neglect the assessment of lower limb function. The G-AT was originally studied in patients with chronic obstructive pulmonary disease (COPD) [18] and was subsequently evaluated for use in different conditions, such as cystic fibrosis, community-acquired pneumonia, cardiovascular diseases, postoperative bariatric surgery, and Parkinson’s disease [19, 20, 21, 22, 23]. More recently, the G-AT has been applied to subjects with RA-related pulmonary disease (RA-PD) to assess the contribution of ventilation distribution heterogeneity [24]. However, since RA is characterized mainly by the involvement of the osteoarticular and muscular systems of both the lower and upper limbs and pulmonary involvement can impair functional exercise capacity, comparing performance on the G-AT in people with and without RA in the absence of RA-PD is essential. Thus, the aim of the present study was to compare functional exercise capacity using the G-AT in women with and without RA in the absence of RA-PD and to correlate these findings with hand functioning, physical functioning, handgrip strength (HGS), and quadriceps strength (QS).

2. Materials and methods

3. Results

Of the 35 women evaluated for inclusion in the study, five were excluded for the following reasons: two had difficulty walking, two had undergone orthopaedic surgery on a lower limb, and one was unable to perform the G-AT due to severe dyspnoea. The median age was 57.5 (44.5–63) years, while the median time since diagnosis was 12 (5–21.3) years. Compared to the women in the control group, the women with RA presented higher scores for the CHFS ( 0.0001) and HAQ-DI ( 0.0001) and lower HGS ( 0.0001) and QS ( p = 0.013) values.

The median time required to perform the G-AT activities was higher in the women with RA than in the healthy women [300 (295–420) vs. 180 (155–203) s], 0.0001). Based on Brazilian values predicted for healthy women with the same anthropometric characteristics [31], the time required for the women with RA to perform the multiple G-AT tasks was approximately 70% higher than the expected time. No significant decrease in SpO 2 was observed during the G-AT in either group, although the degree of dyspnoea was increased in the women with RA compared to the healthy controls. Most of the RA patients reported that the greatest difficulty they experienced at the end of the G-AT was squatting to complete the shelving tasks, while most of the healthy controls reported no difficulty performing the G-AT tasks. A statistically significant difference was found between the two groups in relation to the types of difficulties reported at the end of the test ( 0.0001). The anthropometric data, hand ability, physical functioning, HGS, QS, and G-AT results for the two groups are shown in Table 1. The table also shows the comparisons between the two groups and the effect sizes of the results.

When comparing the medians of the two GA-T trials performed by the patients, no significant difference was noted, although the time to complete the second G-AT was shorter [304 (253–373) seconds vs. 296 (248–356) seconds, p = 0.83]. Highly significant intraobserver agreement was observed between the measurements of the two GA-T trials (ICC = 0.93 and 95% CI = 0.85–0.95; 0.0001). For the healthy controls, no significant difference was found between the time to complete the first G-AT and the time to complete the second G-AT [183 (153–218) seconds vs. 176 (150–209) seconds, p = 0.88], with an ICC of 0.95 (95% CI = 0.92–0.98; 0.0001).

The associations between the time required to perform the G-AT tasks and the measurements of hand ability, physical functioning, HGS, and QS are shown in Table 2 and Fig. 2. The time required to perform the G-AT tasks was positively correlated with HAQ-DI ( r s = 0.668, 0.0001) and CHFS ( r s = 0.586, p = 0.0007) scores and negatively correlated with QS ( r s = - 0.429, p = 0.037). Additionally, we did not observe significant correlations between the time required to perform the G-AT tasks and the time since diagnosis or between the time to perform the tasks of the G-AT and HGS.

OPEN IN VIEWER

Based on an a priori type-I error α = 0.05 (two-tailed), the power analysis detected significant effects in the comparisons between the two groups as follows: CHFS = 99%; HAQ-DI = 99.5%; HGS = 99.5%; QS = 94%; and G-AT time = 99%. For the correlations with G-AT time, significant effects were observed as follows: CHFS = 94%; HAQ-DI = 98%; and QS = 68%. These results show the adequacy of the studied sample size to obtain significant results [32].

Since ADL limitations are best predicted by measures of functional exercise capacity that replicate daily activities, the use of tools involving at least three different tasks has recently been suggested for the evaluation of ADLs [34]. Considering that the risk of disability, which is defined by the presence of at least some self-reported difficulties in most ADLs, is seven times higher in individuals with RA than in the general population [35], we believe that the G-AT may be a useful tool for incorporating the function of both the upper and lower limbs. Thus, when evaluating the applicability of the G-AT for RA patients in the absence of RA-PD, one of the main findings of the present study was that the women with RA required more time to perform the G-AT tasks than the healthy controls, mainly due to the greater difficulty of squatting to perform the shelving tasks. Moreover, the time required to perform the G-AT tasks was correlated with CHFS scores, the HAQ-DI, and QS but not with HGS.

In addition to being an easily administered, valid, and reliable test for measuring functional exercise capacity, the G-AT offers a more complete evaluation of functional capacity, better mimicking the situations experienced during ADLs and more reliably reflecting the burden that patients experience in their day-to-day life [18, 21]. In the present study, we observed that women with RA required more time to perform the G-AT tasks than healthy controls. Curiously, the RA patients’ completion time was 70% higher than that of women of the same age and height according to the Brazilian equations [31]. RA-PD may cause notably worse performance during the G-AT since the median G-AT time of our sample was shorter than that measured in people with interstitial lung disease associated with RA (300 s vs. 360 s) [24]. When we compared the median time required by our patients to perform the G-AT (300 s or 5 min) to the times reported for patients with other clinical conditions in the literature, we found that it was much higher than that observed for patients with cardiovascular diseases, cystic fibrosis, and Parkinson’s disease [19, 20] but lower than that observed for patients with COPD [36].

In the present study, the main difficulty reported by most patients at the end of the test was squatting to perform the shelving tasks. Squatting can alter body alignment and can therefore expose the joints of the lower limbs to excessive torque, requiring adaptive muscle activation strategies to stabilize these joints [37]. In addition to affecting the hip and lower limb joints, RA can cause changes in the spinal sagittal balance that increase the difficulty of squatting during G-AT tasks [38]. A study recently showed that squatting is one of the most difficult exercises for patients with RA to perform, possibly because it requires hip and knee strength and range of motion [39]; this finding reinforces the importance of focusing on these major joints during the clinical treatment and rehabilitation of RA patients. Notably, the second most frequently reported difficulty among the women in our sample was climbing and descending stairs during the G-AT. Interestingly, Mengshoel et al. [40] showed that in individuals with RA, quadriceps muscle strength and age explained 38% of the variation in the time required to climb stairs.

RA has a particular predilection for the hands, resulting in pain, deformity, and functional limitations, which remain a problem even with the use of new immunobiological agents [41, 42, 43]. Thus, using instruments that can assess hand functioning, including manipulation motor skills, manual dexterity, and ADL performance, is important. In the present study, we used the CHFS questionnaire; although this questionnaire is sometimes used for people with RA, its usefulness has been questioned because it has not been validated with another standardized test of functional hand performance [44]. In addition to observing significant differences in CHFS scores between the patients and the healthy controls, we found that the CHFS score was positively correlated with the time required to perform the G-AT tasks. Thus, we believe that the CHFS should be more widely used in the routine assessment of individuals with RA because in addition to evaluating hand performance during ADLs, it is easy to understand, can be administered quickly, and does not require special equipment or training [26, 44].

The marked process of chronic inflammation in RA leads to reduced strength and muscle mass, with a consequent decrease in functional capacity and HRQoL even at early stages of the disease, contributing to increased mortality and a reduction in the life expectancy of affected individuals [11]. Similar to the findings of Cimen et al. [45] and Sferra da Silva et al. [46], we observed reduced HGS in women with RA compared to healthy controls. In line with the studies of Berner et al. [47] and Alfuraih et al. [48], we also noted lower QS in patients than in healthy controls. Muscle weakness as a result of disuse and muscular atrophy is a well-known feature of RA [47, 48]. Almost two-thirds of individuals with RA experience a substantial reduction in skeletal muscle mass, which is characterized by depletion of protein storage and accumulation of fat in the muscle [49]. Interestingly, we observed an association between G-AT time and QS but not between G-AT time and HGS. Since the extensor muscle of the knee plays an important role in the ADL domain in individuals with RA [50], it may have strongly impacted our participants’ performance of the G-AT tasks, causing difficulty with climbing stairs and rising from a chair without a backrest. In line with this hypothesis, a recent study showed that knee extensor force ( β = 0.45, p = 0.001) determined RA subjects’ physical fitness and ability to work much more than HGS ( β = 0.25, p = 0.039) after adjustment for sociodemographic and disease-specific variables [47]. Notably, no significant correlations were found between the time required to perform the G-AT tasks and the time since diagnosis. A possible explanation is that the inflammatory activity intrinsic to the disease, more than the duration of the disease, can negatively impact the functional exercise capacity of people with RA [51].

Since the HAQ-DI is the most widely used disability measure for individuals with RA both clinically and in clinical research, we evaluated physical functioning through questions about ADLs aiming to assess the motor abilities of the upper and lower limbs. Based on previous studies indicating that HAQ-DI scores ⩾ 2 are associated with important functional impairment in RA [4, 52], our sample included only three patients with scores above this cut-off. Despite the HAQ-DI’s limited ability to evaluate more complex ADLs and its lack of sensitivity for detecting changes in patients with low weakness (“floor effect”) [53], we observed a strong correlation between the HAQ-DI and G-AT time. In line with our findings, a recent study conducted by Mellblom Bengtsson et al. [54] in patients with early RA showed significant associations between HAQ-DI scores and different aspects of lower limb functioning, such as overall functional ability, muscle strength, muscle resistance, and balance/coordination, skills that are incorporated into the tasks of the G-AT. Finally, the correlation that we observed between the HAQ-DI and G-AT time was much higher than that recently reported between the distance walked in the 6 MWT and the HAQ-DI ( r = - 0.495, p = 0.0001) in a prospective study with 86 individuals with RA [16]. One possible explanation is that the G-AT incorporates several tasks requiring both upper and lower limb activity unlike the 6 MWT, which evaluates only walking activity.

The strength of this study is that it shows that the G-AT – an easy test to administer that simulates ADLs – can provide a more valid measure of functional exercise capacity in individuals with RA in the absence of RA-PD. In addition to the patients taking longer to perform the G-AT tasks, significant correlations were identified between G-AT time and measurements performed in clinical practice for patients with RA. Despite these interesting results, the present study has limitations. First, the generalization of our results may be limited by convenience sampling and because we evaluated only women, although RA is much more common in women than in men [2, 3]. Second, the cross-sectional design of this study limits our ability to establish causal relationships between variables. Third, we did not objectively measure the participants’ physical activity via physical activity monitoring. Fourth, we used two kinds of questionnaires (the CHFS and HAQ-DI) that are still controversial as tools for measuring diagnostic performance [40, 48]; thus, the use of these questionnaires weakens the level of evidence to some extent.

5. Conclusions

In this pilot study, women with RA without lung injury required more time than healthy controls to perform the multiple tasks of the G-AT. In addition, the total G-AT time exhibited important correlations with measures of hand functioning, physical functioning and QS, but not with HGS.