**3.2 Effects on rheumatoid cachexia (body composition)**

The effects of resistance training on body composition in RA are less well reported (Table 1). In 1976, Nordemar et al. observed increased cross-sectional area of type I and especially type II fibres in 10 RA patients following 6 weeks of cycling, walking and quadriceps strength training. Similarly, Hakkinen et al. (1994) observed increases in quadriceps muscle crosssectional area in RA patients following 6 months PRT. However, when Rall et al. (1996b) reported no changes in whole-body composition (DXA assessed) in 8 RA subjects following 12 weeks PRT (despite significant improvements in strength), the conclusion was that RA patients are resistant to the anabolic effects of exercise. This concern has subsequently been refuted by methodologically more robust trials. Initially, we (Marcora et al., 2005a) reported significant increases in (DXA assessed) LBM, ALM and estimated total body protein (TBP), and reductions in %BF, with a trend toward reduced trunk fat (-0.75kg) following 12 weeks of high-intensity PRT. Subsequently, these effects were confirmed by our RCT (Lemmey et al., 2009); LBM, ALM (≈1.2kg), and TBP were all significantly increased (p's=0.002-0.006) and total and especially trunk FM (-2.5kg, i.e. 18%) were substantially reduced following 24 weeks of PRT. Additionally, Hakkinen et al. (2005) have reported quadriceps femoris hypertrophy (p<0.001) and reduced quadriceps subcutaneous fat thickness (p<0.001) in female RA patients following 21 weeks of combined PRT and aerobic training.

Whilst aerobic exercise training, by increasing daily energy expenditure, has been shown to be an effective adjunct to restricted energy intake for weight loss in young adults, its efficacy in middle aged and elderly individuals is questionable. This is because sedentary individuals of this aged are usually so deconditioned that they are unable to perform exercise of sufficient intensity and duration to significantly elevate daily energy expenditure (Evans, 1999). In contrast, in elderly men and women an elevation of approximately 15% in resting metabolic rate (RMR) has been observed following 12 weeks PRT as a consequence of increased LBM (Campbell et al., 1994). An increase in RMR of this magnitude is very relevant as RMR typically accounts for 60-75% of 24 hr energy expenditure.

In our PRT studies (Lemmey et al., 2009; Marcora et al., 2005a), the elicited increases in muscle mass were significantly associated with improvements in objectively assessed physical function (i.e. 30 sec arm curl, 30 sec sit-to-stand, 50' walk, hand-grip strength, and knee extensor strength; tests taken from the Senior Fitness Test (Rikli & Jones, 2001), and designed to reflect the ability to perform ADL's). Interestingly, the increased muscle mass and reduced fat mass in the PRT subjects in our RCT (Lemmey et al., 2009) caused a reclassification of the body types of many of these patients. Wherein, whereas at baseline, 9 (out of 13) were classified as cachectic, 10 as obese, and 5 as both (i.e. "cachectic-obese"),

Resistance Training for Patients with Rheumatoid Arthritis: Effects on Disability,

prolonged and its impact on BMD given longer to accrue.

1997; Nelson et al., 1994; Rhodes et al., 2000).

**3.5 Effects on bone** 

Rheumatoid Cachexia, and Osteoporosis; and Recommendations for Prescription 295

The benefit of weight-bearing and strengthening exercise in maximising and maintaining BMD, and reducing the risk of falling by improving strength and balance is well accepted in the general population (ACSM, 2010a). With specific reference to RA, a sedentary lifestyle confers a relative risk of 1.6 for low BMD in RA patients, and even moderate physical activity has been shown to reduce this risk by 50% (Tourinho et al., 2008). Additionally, de Jong et al. (2004) showed that bone loss at the hip was reduced in RA patients participating in the 2 year, high-intensity RAPIT exercise program (median -1.1% vs -1.9% for nonexercising controls, p<0.05). Further analysis revealed that these changes in BMD were significantly and independently associated with changes in strength and aerobic power, and that the high-intensity training had a benefit comparable to that of biphosphonate treatment. This finding led the investigators to conclude that intense weight-bearing exercise, including PRT, is essential for improving BMD in RA patients. Similar conclusions were made by Hakkinen et al. following their RCT (1999, 2001, 2004a). In this trial, twelve months PRT by RA patients resulted in mean BMD gains of +1.10% at the femoral head and +0.19% at the lumbar spine in contrast to losses of -0.03% and -1.14%, respectively, in the ROM controls (Hakkinen et al., 1999). Following a further 12 months PRT, the mean differences between the groups increased with the changes in BMD at the femoral head and the lumbar spine now +0.51% and +1.17% for the training group and -0.70% and -0.91%, respectively, for the controls (Hakkinen et al., 2001). These observed trends in BMD were noted again at a 3 year follow-up (Hakkinen et al., 2004a). Whilst the differences between the groups were not statistically significant, except for the femoral head at 24 months, it was suggested by the authors that such an effect would be substantial and of clinical significance if PRT was

Treatments for osteopenia or osteoporosis are judged on their ability to increase BMD, or failing that, to minimise bone loss. Thus, although the evidence from RA patients is limited, PRT appears to be as efficacious in this population as it is generally (e.g. Dornemann et al.,

In RCT's conducted to evaluate the effect of PRT on BMD in the general population, the evidence is compelling that intensity (i.e. loading) is the key variable (see Layne and Nelson, 2001 for a review). This is consistent with Wolff's law which states that the magnitude of the stress or mechanical load applied to bone via muscles and tendons directly determines the osteogenic response (Chamay & Tschantz, 1972). The results of Kerr et al. (1996) serve to illustrate this. In this study, post-menopausal women (aged 51-62 years) were randomised to either high-intensity (HI) "strength" PRT (high load, low repetitions i.e. 3 sets of 8 repetitions) or low-intensity (LI) "endurance" PRT (low load, high repetitions i.e. 3 sets of 20 repetitions). After training 3x's/week for 12 months, the HI group had increased femoral head and distal radial BMD significantly more than the LI group; with the site-specific gains in BMD significantly correlated to the site-specific strength increases. In patients recovering from surgery, strength training has also been shown to be effective in countering glucocorticoid-induced bone loss (Braith et al., 1996). However, as for the general population, the greatest benefit of PRT in reducing osteoporotic fractures in RA patents is likely to be a consequence of lowering the incidence of falling due to improved strength and balance (Layne and Nelson, 2001; Nelson et al., 1994; Vanderhoek et al., 2000). With regards to the suitability of high-intensity PRT for individuals with low BMD; Vanderhoek et al. (2000) specifically chose osteopenic or osteoporotic elderly women (mean±sd; age = 69.0±1.3

after 24 weeks of PRT the number of patients in these high disability risk categories (Morley et al., 2001) were reduced to 4, 7 and 2, respectively. Given the reported links between adverse body composition and physical disability in RA patients (Giles et al., 2008) and the general elderly population (Morley et al., 2001), the positive effects of PRT on function in RA patients are anticipated. To emphasise the crucial role played by training intensity, in our RCT study (Lemmey et al., 2009) range-of-movement (ROM) exercises (i.e. the form of exercise most commonly prescribed for RA patients) were performed by the control group. Despite good compliance to the intervention, this low intensity exercise failed to have any effect on the various measures of body composition or objective physical function.

#### **3.3 Impact on mechanisms of rheumatoid cachexia**

As mentioned earlier, the precise mechanisms underlying rheumatoid cachexia have not been clarified. However, an additional insight was provided by Lemmey et al's RCT (2009). In this study diminished muscle levels of insulin-like growth factor-I (mIGF-I) were identified in our RA patients. This finding is consistent with reports of reduced mIGF-I levels in other conditions characterised by muscle wasting: chronic heart failure (CHF) (Hambrecht et al., 2005), chronic obstructive pulmonary disease (COPD) (Vogiatzis et al., 2007), chronic renal failure (Macdonald et al., 2004, 2005), and advanced aging (Fiatarone Singh et al., 1999); and with the proposed role of mIGF-I in regulating the maintenance of adult skeletal muscle (Adams, 2002). Following 24 weeks PRT, along with muscle hypertrophy, mIGF-I levels were observed to increase 50% in our RA patients. Again, this finding of coincident increases in mIGF-I levels and muscle mass in cachectic individuals following exercise training is consistent with responses in COPD (Vogiatzis et al., 2007) and dialysis (Macdonald et al., 2005) patients, and the frail elderly (Fiatarone Singh et al., 1999); and the pivotal role put forward for mIGF-I in muscle's hypertrophic response to loading (Adams, 2002).

#### **3.4 Responsiveness of RA patients to PRT**

The magnitude of effects of PRT on strength and body composition observed in RA patients are similar to those reported for healthy middle-aged or older individuals (e.g. Frontera et al., 1991; Morse et al., 2007; Nichols et al., 1993; Pedersen & Saltin, 2006). The study by Hakkinen et al. (2005) described previously provides a direct comparison of training responses. This investigation featured female RA patients and age-matched healthy women who completed the same 21 week combined resistance and aerobic exercise training program, and noted remarkably similar improvements in strength and body composition (with regard to both absolute and relative increases in quadriceps femoris cross-section and reductions in quadriceps femoris subcutaneous fat thickness) following training. This similarity in training response is consistent with recent reports that muscle quality (muscle force per size) is not compromised in RA patients (Matschke et al., 2010a, 2010b). In these studies, a range of skeletal muscle parameters (e.g. specific force, muscle architecture, coactivation of antagonist muscles, voluntary activation capacity) were observed to be the same for well controlled RA patients, including those classified as cachectic, as for matched healthy subjects. This finding that rheumatoid muscle is normal both qualitatively and in its response to resistance training is important for health professionals involved in prescribing exercise for people with RA.
