This study shows that ambulatory patients with RA of low disease activity were reported to have a diet intake high in saturated fat. They were in general centrally obese and a fifth had rheumatoid cachexia. A large proportion of the patients had hypertension and MetS, especially those with high FM and rheumatoid cachexia, conditions associated with high levels of oxLDL and low levels of anti-PC. Patients compliant with Mediterranean and vegetarian diets did not differ in body composition from the rest of the patients, but had a higher content of PUFA in adipose tissue and significantly higher serum levels of the atheroprotective anti-PC.
The patients had, according to the FFQ, a lower than recommended caloric intake. It is unclear if this was a real difference or a consequence of underreporting. However, underreporting seems most probable as earlier comparisons between FFQ and weighed diet records concerning fat intake showed lower total intake registered by FFQ than the weighed diet, whereas the proportions of fat components were similar [27
With regard to the caloric proportions of the diet, our patients had a higher intake of saturated fat and lower intake of unsaturated fat compared to the Food Recommendations in Sweden [35
]. A high fat intake has also been noted in RA patients in the USA, of which 40% were obese [21
The reliability of the long-term fatty acid intake according to the FFQ was verified by the significant correlations found between the PUFAs ingested and those in adipose tissue. PUFAs in adipose tissue are largely exogenous and hence valid markers to assess dietary intake [42
]. By contrast, SFAs and monounsaturated fatty acids (MUFAs) are not considered to be good biomarkers of intake because they are also endogenously synthesized [42
], which was evident in the RA patients with no significant correlations concerning 16:0 and 18:1 n-9. Furthermore, the noted significant correlation between 14:0 FA in diet and adipose tissue verifies a report by Wolk et al.
that this FA is also a valid biomarker for FA intake [45
When comparing the proportions of FAs in the adipose tissue in our RA patients with healthy subjects from different countries (Sweden, Denmark, USA, Spain and Costa Rica), our RA patients had higher proportions of SFA and lower proportions of PUFA than healthy subjects from the above-mentioned countries [25
]. However, the SFA and PUFA content of adipose tissue was similar to that described in RA patients from the south of Sweden [48
The patients reported in general a high intake of saturated fat. This high intake was also reflected in the FA composition of the adipose tissue. However, we did not find that the FA intake differed between cachectic and non-cachectic patients. Adherence to a Mediterranean diet did not differ either between cachectic and non-cachectic patients. Thus, diet as analyzed here does not seem to cause the derangement of body composition leading to cachexia. This finding is in accordance with the known difficulties of treating cachexia with diet. However, recently supplements with amino acids have been shown to increase muscle mass [23
]. By contrast, the high intake of dietary fat may have contributed to the dyslipidemia displayed by the patients.
With regard to dietary therapies, 33% to 75% of RA patients believe that food plays an important role in their disease, and 20% to 50% have tried dietary manipulation in an attempt to relieve symptoms [49
]. The present finding that dietary intake of SFAs was significantly associated with disease activity focuses on the immunosuppressive effect of unsaturated FAs. In the scientific community there is, however, no consensus on dietary recommendations to patients with RA. Evidence may emerge that, for patients at an advanced disease stage, a high-protein diet may be recommended and, for patients with early RA, restriction of excessive energy intake may be beneficial to prevent obesity [51
]. A scientific basis for a role of dietary therapy in RA has grown in the last few years, although there is still no agreement on the nature of the optimum diet [52
Several of the RA patients had dyslipidemia, hypertension and MetS, most frequently patients with high FM and rheumatoid cachexia. Of the lipids, oxLDL might be especially important as it has proinflammatory and immune stimulatory properties [53
]. It has previously been detected in synovia from RA patients [55
], and is suggested to be of importance in the pathogenesis of CVD [8
]. Even though there are no existing reference values for oxLDL, the present patients had higher levels than earlier reported for healthy people, with mean (SD) values of 42.2 U/l (± 14.7) and mean (95% confidence interval (CI)) 48 U/l (35 to 68) [7
]. Additionally, when compared with RA patients from Korea who had a mean (SD) value of 53.0 U/l (± 20.9) [7
], our patients had higher levels, a difference that might be secondary to different diet habits.
One novel finding herein is that anti-PC levels decreased in patients with rheumatoid cachexia. Low anti-PC levels could predispose patients to cardiovascular disease [10
] and might thus be one explanation for the increased morbidity in rheumatoid cachexia. The cause of this association is not known.
Also, the very high levels of anti-PC in patients on a Mediterranean diet is of great interest, as patients on this diet are reported to have a lower frequency of CVD [57
]. In RA anti-PC have not been studied in relation to CVD, but we have recently shown that the levels of anti-PC in serum increased in RA patients when changing from a normal Western diet to a gluten-free vegan diet [12
]. The present finding adds strength to our previous report, indicating that diet may influence atheroprotective anti-PC in RA.
A high frequency of MetS has previously been reported in RA patients [59
]. In these patients an association between MetS and disease activity, as well as coronary calcification, has been shown [60
], suggesting that the increased prevalence of coronary heart disease in RA patients may, at least in part, be attributed to inflammation and change of fat metabolism.
In the present RA population every fourth to fifth patient had rheumatoid cachexia, irrespectively of which of the two definitions for cachexia were used. There is no standard definition of this condition as reviewed by Summers et al
], which is why the frequency of rheumatoid cachexia varies. The cachexia and muscle wasting found here could not reflect the changes in body composition that occur with age [63
], as the reference population was matched for age and sex [40
]. Age-related muscle wasting is a separate condition named sarcopenia. Cachexia in chronic inflammatory diseases is suggested to be associated with increased morbidity [19
]. Mostly, the morbidity has been attributing the loss of body lean mass [13
], whereas the coincidental increase of fat mass has not been considered in this context. In the present study it was obvious that FM above the 50th percentile was, together with FFM below the 25th percentile, associated with dyslipidemia as were increased levels of total cholesterol, LDL and oxLDL and also low levels of anti-PC. These patients are therefore suggested to have an increased risk of CVD and premature mortality [10
]. In fact, they also had high frequencies of hypertension and MetS.
Low FFM alone, irrespective of FM, was not associated with these risk factors for CVD. By contrast, only one of the patients was severely emaciated with extremely low BMI known to be associated with cardiovascular mortality in RA [65
A large proportion of the patients were centrally obese, which is in line with earlier reports [16
]. This fact might contribute to the increased presence of risk factors for CVD in the patients with rheumatoid cachexia.
In the present cohort of patients we did not find that DAS28 or CRP differed between those with cachexia and the rest of the patients, probably related to the fact that most patients had low inflammatory activity. However, when separating the patients into those with low DAS28 and those with DAS28 corresponding to moderate activity it was found that the latter had lower FFM and higher FM than the former. Thus, the inflammation per se
might have contributed to the derangement in body composition found here and also described previously [18
]. A further explanation could be that patients with DAS28 of moderate activity reported lower energy intake than those with low activity.