Some foods make my arthritis worse... is it my imagination?

For centuries, the effects of food on arthritis symptoms have been discussed both in the scientific literature as well as in lay writings. Food allergy has been reported to play a role particularly in RA.

In a study published in 1980, 22 patients with RA consumed a diet that excluded common allergens. Twenty patients (91%) experienced an improvement in symptoms, and 19 found that specific foods repeatedly exacerbated their symptoms. The mean time on the elimination diet before improvement occurred was 10 days, and the longest time was 18 days. The mean number of food sensitivities per patient was 2.5; the most common symptom-provoking foods were grains, milk, nuts, beef, and egg. (Hicklin JA, McEwen LM, Morgan JE. The effect of diet in rheumatoid arthritis. Clin Allergy 1980;10:463.)

In a later study, 53 patients with RA were randomly assigned to consume a diet that excluded common allergens, or their usual diet (control group) for six weeks. After one week, the patients on the exclusion diet began reintroducing one food at a time; any foods producing symptoms were removed from the diet. The hypoallergenic diet group fared significantly better than the control group for each of 13 different parameters of disease activity. The patients in the control group then underwent the same elimination-and-challenge procedure that the diet group had, and experienced similar, though somewhat less impressive, improvements (Darlington LG, Ramsey NW, Mansfield JR. Placebo-controlled, blind study of dietary manipulation therapy in rheumatoid arthritis. Lancet 1986;1:236-238).

Long-term follow-up of 100 patients who underwent dietary therapy in this study revealed that one-third remained well on diet alone, without any medication, for up to 7.5 years after starting treatment (Darlington LG, Ramsey NW. Diets for rheumatoid arthritis. Lancet 1991;338:1209).

While the possibility of placebo effect needs to be considered, the long-term benefit experienced by these patients is noteworthy. Also, while there was some weight loss noted in the treated patients, there was no significant correlation between weight loss and clinical improvement in RA symptoms.

In another double-blind controlled study, 94 patients with RA were randomized to one of two diets for four weeks, followed by a return to their usual diets for another four weeks. One diet ("allergen free") was free of common allergens, additives and preservatives. The other diet ("low allergen") was similar to the allergen-free diet, but contained milk allergens and azo dyes. Seventy-eight patients completed the study. The effects of food elimination and re-challenge varied considerably among patients. Nine patients (11.5% of the total; 6 in the allergen-free group, 3 in the low-allergen group) had a favorable response to the elimination diet, followed by marked disease flare during re-challenge. In these patients, subjective improvements were confirmed by improvements in objective parameters of disease activity (Van de Laar MA, van der Korst JK. Food intolerance in rheumatoid arthritis. I. A double blind, controlled trial of the clinical effects of elimination of milk allergens and azo dyes. Ann Rheum Dis 1992;51:298-302). The small number of patients exhibiting changes is an argument against significant benefits associated with a diet manipulation.  Nonetheless it appears that there is a subset of patients for whom diet is an important component of their symptom complex.

A smaller study looked at eleven RA patients. Two of the 11 RA patients showed a favorable response to an elimination diet and experienced worsening after eating offending foods. In that study, the elimination diet did not exclude certain common allergens (wheat, corn, egg whites, sugar, and coffee). It could be argued that the response rate would have been higher if the elimination diet had been more restrictive (Panush RS, Carter RL, Katz P, et al. Diet therapy for rheumatoid arthritis. Arthritis Rheum 1983;26:462-471). Small numbers in this study make comprehensive validation impossible.

These studies seem to imply that avoidance of allergenic foods might benefit a subset of patients with RA, although the proportion of patients responding to dietary change varies a lot from one study to the next. The difference in response rates may be related in part to the patient populations studied. Some authorities feel that younger patients (under the age of 40) with less aggressive RA respond best to avoidance of allergens. Older patients and those with relatively severe RA may not respond to dietary manipulation as well. The divergent results in published studies may also be explained in part by strictness of dietary restriction and/or compliance. Finally, RA is a spectrum of diseases; not all therapies work for all patients.  This divergence of effect has been noted even with biologic therapies.

The possibility of food allergies accounting for RA symptoms makes the use of food allergy testing a viable weapon in the arsenal of the clinical rheumatologist. In our clinic we use food allergy testing in patients where symptoms don’t seem to respond to conventional therapies.