1. Introduction
Primary Sjögren’s syndrome (pSS) is an autoimmune disorder of the exocrine glands with unknown etiology, and it mainly affects postmenopausal women [
1]. The prevalence of pSS depends on age and classification criteria [
2], and in Norwegians it is estimated at 0.05% when applying the criteria of the American-European Consensus Group [
3].
Patients with pSS display a wide range of symptoms and low secretion of saliva, which, in particular, contributes to increased risk of dental caries and oral infections. Moreover, digestive manifestations with dysphagia and dysmotility of the pharynx are common, and both the pancreas and the liver may be affected [
4,
5]. There is also a higher risk of gastroesophageal reflux disease among pSS-patients compared with the general population [
6].
These manifestations may affect food intake among pSS patients. Rhodus found significantly lower intakes of twelve analysed nutrients compared both with healthy controls and with the Recommended Daily Average values [
7]. Contrary to this, Hay et al. did not find that pSS-patients had a diet different from the recommended intakes of the New Zealand population, even though the same patient group reported an altered food intake after the onset of xerostomia [
8]. Cermak et al. reported that women with pSS had a significantly higher intake of energy, glutamate, carbohydrates, lactose, phosphorus, caffeine, thiamine, and riboflavin compared with age-matched controls [
9]. Moreover, Tovar et al. indicated that patients with pSS were deficient in vitamin B
6, probably due to a limited intake and not to abnormal uptake mechanisms [
10], and Erten et al. found that pSS-patients were deficient in vitamin D [
11].
There are only a few dietary intervention trials involving pSS-patients, and these have mostly investigated salivary secretion. For example, Singh et al. found no significant effects of supplementation of omega-3 fatty acids and vitamin E on salivary secretion flow compared with placebo oil [
12]. A Norwegian study investigated the effect of a liquid diet in pSS-patients and found that the intervention group had increased salivary flow compared to controls after 4 weeks on the diet [
13].
The reduced ability to taste and smell in pSS contributes to decreased appetite and impaired secretion of saliva, which may cause difficulties when eating and swallowing [
14]. Rusthen et al. found that many pSS-patients have an impaired taste perception and, in addition, often have difficulty in eating dry foods [
15,
16]. Moreover, significant differences in body composition between pSS-patients and controls have been reported [
17,
18,
19].
Currently, there is limited data about what pSS-patients actually eat, but such information is necessary for diet-recommendations. We therefore examined the dietary intake among a well-characterized cohort of Norwegian female pSS-patients and compared it with the dietary intake of a national reference group and with the Nordic dietary recommendations. In addition, we studied associations between dietary intake/body composition and oral health among the patients.
4. Discussion
The pSS-patients had a higher absolute intake of fat and a lower intake of carbohydrates compared to the N3-reference group. In addition, the mean E% from carbohydrates among pSS-patients was below the recommendations. A higher intake of butter, margarine, and oil probably contributed to the higher intake of fat, while a lower intake of bread contributed to the lower intake of carbohydrates. Notably, the daily intake of fish among the pSS-patients was more than twice that of the N3-reference group. None of the pSS-patients was underweight or malnourished, whereas 40% were overweight/obese.
A possible explanation for the higher intake of fat-products among the pSS-patients may be a lubricating effect in the oral cavity, thereby aiding mastication and swallowing [
31]. Impaired taste function can be another explanation, as patients experiencing taste changes have reported change in eating habits to compensate for the missing sensory experience [
32].
The high intake of lozenges among the pSS-patients was expected, as they stimulate salivary secretion, whereas the high intake of chocolate might be due to a higher demand for sweet taste caused by disturbed gustatory function. Rusthen et al. showed that 19% of the pSS patients were aguesic, 32% were hypogeusic, 13% were anosmic, and 29% were hyposmic, which could affect taste [
15]. A Mexican study concluded that pSS-patients were only mildly dysgeusic to the taste of sweet and salt, and more dysgeusic to sour and bitter tastes [
33]. The reduced perception of smell and flavour among the pSS-patients may have led to an increased need to add spices to the food [
34]. Five of our patients avoided spicy food, but there were no significant differences in the intake of spices between ageusic/hyposmic versus normosmic patients.
Although milk may function as a saliva substitute [
35], Cermak et al. found a significantly higher intake of glutamate, lactose, thiamine, and riboflavin among pSS-patients, which was ascribed to a higher intake of milk [
9]. However, this finding was not supported by the current study. The fine, smooth, and viscous texture of fish foods compared with meat may facilitate eating and swallowing among the pSS-patients, thus explaining their higher fish intake compared with the N3-reference group.
Compliance with the nutrient-recommendations, excluding supplements, was good overall, with exceptions for saturated fat, fibre, vitamin D, folate, and iron, which is in agreement with the general Norwegian female population [
22]. The results also indicate that the pSS-patients have a sufficient intake of micronutrients.
The pSS-patients had a lower E% from carbohydrate compared with the N3-reference group and with the recommendations. In contrast, Cermak et al. found higher carbohydrate intake in pSS-patients compared with controls [
9], and Hay et al. found that carbohydrate intake in pSS-patients complied with recommendations [
8]. The low intake of starch could be expected, as many starch foods are considered as dry. The intake of E% from total fat among our patients was within the recommended intake limits, while the absolute fat intake and the intake of E% from monounsaturated fat were higher than in the N3-reference group. The mean E% from saturated fat above the recommended limit of 10 E% among the pSS-patients is similar to that of the Norwegian population [
36]. The median intake (about 3 g/day) of omega-3 fatty acids was higher in the pSS-patients (with and without supplements) compared with the N3-reference group, possibly in part explained by the higher fish intake. Reportedly, an intake above 3–4 g/day reduces the need for anti-inflammatory drugs in rheumatic patients and may be an explanation for the higher intake observed in our pSS-patients [
37].
The intake of vitamin D among the pSS-patients were higher than in the N3-reference group. However, the pSS-patients had a median dietary intake of 5.4 µg/day, and when including supplements it increased to 12.7 µg/day, indicating that there is a need for vitamin D supplementation. Whereas Cermak et al. did not find a different intake of vitamin D between pSS-patients and controls [
9], Erten et al. found lower plasma levels of vitamin D in patients compared with controls [
11]. A recent review [
38] found evidence of lower vitamin D levels in pSS-patients that were associated with extraglandular manifestations, whereas another study did not [
39].
We found that 40% of the pSS-patients were overweight/obese (BMI > 25 kg/m
2). The mean BMI was 25.4 kg/m
2, whereas the mean BMI of all the women in the N3 (not just the N3-reference group) was 24.6 kg/m
2. The limited sample size prevented us from examining statistical associations between the overweight/obese and their intakes, but they appeared to have lower intake of micronutrients and higher energy intake from fat than normal weight patients. The patients studied by Rhodus had a mean BMI of 23.2 kg/m
2, which was significantly lower than the BMI of 25 kg/m
2 among the controls [
7], while a New Zealand pSS-population had a mean BMI of 25.5 kg/m
2 [
8]. The ethnicities of these populations may be different from ours, thus precluding direct comparisons.
Dysgeusia did not influence the intake of the food categories or nutrients, and neither did the salivary flow. There was a higher intake of beverages in the group with more severe xerostomia as indicated by SXI-D, and in the group with lowest OHRQoL. The hyposmic patients had a lower intake of fat, which is in accordance with Kong et al., in which there was a lower intake of fat among women with olfactory dysfunction [
40]. The opposite was found by Duffy et al., in which the women with the best olfactory function had a lower E% intake from total fat [
41].
The main limitations of our study are the small sample size, the lack of biomarkers for nutrient intakes, and no control group regarding the oral parameters. Major strengths include the inclusion of a matched reference group and the detailed information of dietary intakes, body composition, and a wide range of oral health parameters.
The dietary intakes of foods and nutrients among the pSS-patients were not much different from the reference group, and the overall compliance with the Nordic dietary recommendations was fairly good. Salivary secretion was not associated with specific food or nutrient intakes, but consumption of beverages was highest among those with severe oral dryness. Specific dietary guidelines are probably not needed to ensure adequate nutritional health among pSS-patients.