Dr. Cook has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Amgen and CaridianBCT.
Seasonal variation in vitamin D levels in psoriatic arthritis patients from different latitudes and its association with clinical outcomes
Article first published online: 27 SEP 2011
Copyright © 2011 by the American College of Rheumatology
Arthritis Care & Research
Volume 63, Issue 10, pages 1440–1447, October 2011
How to Cite
Touma, Z., Eder, L., Zisman, D., Feld, J., Chandran, V., Rosen, C. F., Shen, H., Cook, R. J. and Gladman, D. D. (2011), Seasonal variation in vitamin D levels in psoriatic arthritis patients from different latitudes and its association with clinical outcomes. Arthritis Care Res, 63: 1440–1447. doi: 10.1002/acr.20530
- Issue published online: 27 SEP 2011
- Article first published online: 27 SEP 2011
- Accepted manuscript online: 11 JUL 2011 10:27AM EST
- Manuscript Accepted: 7 JUN 2011
- Manuscript Received: 11 MAR 2011
- Krembil Foundation
- The Arthritis Society through the Spondyloarthritis Research Consortium of Canada National Research Initiative
- Geoff Carr Lupus Fellowship
- Arthritis Centre of Excellence Fellowship
- Canadian Arthritis Network Fellowship
- Abbott Psoriatic Arthritis Fellowship
- Canadian Institutes of Health Research
Vitamin D insufficiency appears to be a pandemic problem and is more common in inhabitants of high latitude compared to low latitude areas. We aimed to determine the prevalence of vitamin D deficiency/insufficiency in patients with psoriatic arthritis (PsA), its seasonal and geographic variation, and the possible association with demographics and disease activity.
This study was conducted in a northern geographic area and in a subtropical region from March 2009 to August 2009. Most subjects were assessed in both winter and summer. Demographics, clinical data, skin phototype, and serum 25-hydroxyvitamin D (25[OH]D) levels were determined. Multivariate linear and logistic mixed models were used to assess the relationship with serum 25(OH)D levels.
In total, 302 PsA patients were enrolled. Two hundred fifty-eight patients were evaluated during the winter, while 214 patients were evaluated during the summer. 25(OH)D levels in winter and summer were adequate (north: 41.3% winter and 41.4% summer, south: 42.1% winter and 35.1% summer), insufficient (north: 55.7% winter and 58.6% summer, south: 50.9% winter and 62.2% summer), and deficient (north: 3% winter and 0% summer, south: 7% winter and 2.7% summer) among patients. There was no association between 25(OH)D levels, geographic and seasonal interaction, race, employment status, and skin phototype or disease activity in both seasons. No association between disease activity in summer and vitamin D levels in winter could be found.
A high prevalence of vitamin D insufficiency among PsA patients was found. There was no seasonal variation in 25(OH)D levels among PsA patients in the southern and northern sites. No association could be established between disease activity and vitamin D level.
Psoriasis is a common chronic skin disorder typically characterized by erythematous scaly papules and plaques. Psoriatic arthritis (PsA) is an inflammatory arthritis associated with psoriasis, usually seronegative for rheumatoid factor. Psoriasis occurs in 1–3% of the population, and approximately one-third of these patients have PsA, with an estimated prevalence of 6–42% (1, 2). Vitamin D (cholecalciferol) is synthesized in the skin under the influence of ultraviolet B (UVB) radiation in sunlight and ingested from fish or plant sources (ergocalciferol). Vitamin D is then hydroxylated in the liver to 25-hydroxyvitamin D (25[OH]D). Therefore, plasma 25(OH)D concentration is the best clinical indicator of vitamin D status because it represents the combined contributions of cutaneous synthesis and dietary intake of vitamin D (3).
Hypovitaminosis D appears to be a pandemic problem. Latitude and season affect both the quantity and quality of solar UVB radiation reaching the earth's surface, which influences the cutaneous synthesis of vitamin D (4). Studies have shown that in areas with latitudes above 37° north or below 35° south, there is a marked decrease in UVB irradiance during the winter months, increasing the risk of vitamin D deficiency (5). Studies have shown that vitamin D insufficiency/deficiency is more common in people living at higher latitudes during the winter. These people were thought to have vitamin D deficiency as a result of reduced sun exposure (6–8).
In temperate areas such as Boston and Edmonton, for example, cutaneous production of vitamin D virtually ceases in winter, when dietary supplementation of the vitamin is advisable (4). Epidemiology has linked vitamin D status with autoimmune disease susceptibility and severity (9, 10). This hypothesis emerged from the observation that people living near the equator were at a decreased risk of developing autoimmune diseases (11). Several studies have found reduced levels of vitamin D in patients with scleroderma, polymyositis/dermatomyositis, antiphospholipid syndrome, rheumatoid arthritis, and systemic lupus erythematosus compared with controls (12–17). It was hypothesized that the overall effect of vitamin D is immunosuppressive through its effect on both T and B lymphocytes (18). However, it is very important to note that the recent review by the Institute of Medicine states that there is not enough evidence at this time to support this hypothesis (19).
Although studies have pointed to a low 25(OH)D level among healthy Canadians, the prevalence of 25(OH)D among PsA patients has not been determined, nor has its correlation with disease activity been assessed. In this study, we aimed to 1) determine the prevalence of vitamin D deficiency/insufficiency in patients with PsA, 2) determine its seasonal and geographic variation, and 3) determine the association of vitamin D deficiency/insufficiency with disease activity.
Significance & Innovations
Vitamin D insufficiency/deficiency is common among patients with psoriatic arthritis (PsA).
There was no seasonal variation in vitamin D level among PsA patients in the southern and northern sites.
No association could be established between disease activity and vitamin D level.
PATIENTS AND METHODS
This is a cross-sectional study conducted from March 2009 to August 2009 in 2 geographic areas. The first one, at 43° 40′ N (north), is the University of Toronto Psoriatic Arthritis Clinic, and the second, at 32° 46′ N (south), included Lin Medical Centre and Carmel Medical Centre, Haifa, Israel. All consecutive patients seen between March and August 2009 were enrolled. Most subjects were assessed in both winter and summer. The study was approved by the ethics boards of the respective centers, and all patients gave written informed consent that was obtained according to the Declaration of Helsinki. All patients fulfilled the Classification Criteria for Psoriatic Arthritis criteria for PsA (20). Patients were assessed initially after the winter season in March 2009 and at the end of the summer in August 2009.
Patients were evaluated by rheumatologists according to a standard protocol (20). Demographics on all patients were collected. Employment status was categorized as employed, retired, homemaker, student, disabled, sick leave, looking for work, and other. Functional class was defined as the following: grade 1 = all activities without pain or handicap, grade 2 = adequate for most activities of daily living (ADL) but some discomfort or limitation, grade 3 = ADL limited to self-care and/or a few daily activities, and grade 4 = little or no self-care or confined to a bed or wheelchair. Data regarding dietary intake, especially dairy products and vitamin supplements, were recorded as well.
At each visit, symptoms (joint pain was reported as present or absent) and physical examination (including complete musculoskeletal examination) with joint counts (68 joints for tenderness and 66 for swelling) were performed. The number of actively inflamed joints was recorded. An “actively inflamed” joint was defined as the presence of joint tenderness and/or effusion. Inflammatory spinal pain attributed to PsA was determined based on physician clinical judgment as concluded from subjective and objective collected data and was reported as present or absent.
Psoriasis activity and skin phototypes.
Psoriasis severity was assessed by the Psoriasis Area and Severity Index (PASI) (21). Fitzpatrick skin phototypes were determined based on the reaction to sun exposure as: type I = always burns, never tans; type II = may burn, tans minimally; type III = burns minimally, tans gradually; type IV = burns minimally, tans well; type V = brown skin, rarely burns, tans profusely; and type VI = black skin, never burns, tans deeply (22).
Laboratory and radiologic assessment.
We collected information on the current use of medications (nonsteroidal antiinflammatory drugs [NSAIDs], disease-modifying antirheumatic drugs [DMARDs], and biologic drugs), and laboratory findings were recorded. During the visit, plasma 25(OH)D, calcium, phosphorus, creatinine and estimated glomerular filtration rate, and liver enzymes were measured. Patients with known liver or kidney disease were excluded from this study, since this might alter the synthesis of 25(OH)D. Serum collected at the time of the visit was assayed for concentrations of 25(OH)D by radioimmunoassay (DiaSorin) according to the manufacturer's instructions.
Recent reports have recommended that the concentration of serum 25(OH)D in adults should be greater than 75 nmoles/liter, while serum 25(OH)D levels of <75 nmoles/liter reflect deficiency or insufficiency (23). In our study, 25(OH)D levels were categorized as deficient at <30 nmoles/liter, insufficient at 30–74 nmoles/liter, and adequate at ≥75 nmoles/liter (24).
Clinically damaged joints (68 assessed joints) were defined as the presence of limitation of range of movement of >20% of the range not related to the presence of joint effusion, presence of joint deformity, subluxation, loosening, or ankylosis (25).
Descriptive statistics were computed with continuous variables summarized by their means and SDs and categorical variables summarized by numbers and proportions. Regression analyses were used to assess the relationship between vitamin D levels and disease activity. A multivariate linear mixed model was used for continuous outcomes with random effects introduced to accommodate dependence within subjects in related measurements. Reduced models were obtained by backward elimination (in the first model the covariates were centers and seasons; in the second model the covariates were centers, seasons, and centers and seasons interaction). Logistic regression was used for analyzing the categorical outcomes. The following measures of disease activity/damage were assessed as outcomes: PASI, total damaged joint counts, active joint counts, and inflammatory back pain. We included in the models the following covariates: age, sex, centers (north versus south), seasons (summer versus winter), and duration since onset of PsA (1-year increase), use of DMARDs, use of NSAIDs, use of biologic agents, and 25(OH)D levels. The second model included the same covariates but added the 25(OH)D categorical variable: normal versus deficient/insufficient. For this analysis, patients with insufficient and deficient levels of 25(OH)D were collapsed into one group: “abnormal = deficient/insufficient.”
In addition, to assess whether 25(OH)D levels in the winter predict disease activity in the summer, we used the same statistical approach as above, using 25(OH)D as both a continuous and a categorical covariate (normal versus abnormal, where abnormal = insufficient and deficient).
The statistical software R, version 2.11.0, was used for all statistical analyses, and the significance level was set at 5%.
Three hundred two PsA patients were enrolled: 258 were enrolled in the winter (201 north/57 south) and 214 patients were enrolled in the summer (140 north/74 south). The majority of the patients in both centers were men, with 63.7% in the northern center and 58.4% in the southern center. Overall, 61.9% of the patients were men and 38.1% were women. The majority of the patients (96.7%) were white, with a mean ± SD age at the initial visit of 53.4 ± 12.9 years (Table 1). The patients had normal kidney function and no evidence of liver disease. All patients had normal calcium and phosphorus levels.
|North (n = 201)||South (n = 101)||North and south|
|Age at visit, years||51.8 ± 12.5||56.4 ± 13.4||53.4 ± 12.9|
|Age at onset of psoriasis, years||28.2 ± 14.6||37.1 ± 16.0||31.2 ± 15.5|
|Age at onset of PsA, years||36.5 ± 12.9||46.4 ± 14.4||39.8 ± 14.2|
|Duration since onset of psoriasis, years||23.9 ± 13.9||19.3 ± 15.2||22.4 ± 14.5|
|Duration since onset of PsA, years||15.2 ± 11.3||10.1 ± 11.9||13.5 ± 11.7|
|Sex, no. (%)|
|Male||128 (63.7)||59 (58.4)||187 (61.9)|
|Female||73 (36.3)||42 (41.6)||115 (38.1)|
|Race, no. (%)|
|White||189 (94.0)||101 (100)||290 (96.7)|
|South Asian||1 (0.5)||1 (0.3)|
|Chinese||4 (2.0)||4 (1.3)|
|Filipino||2 (1.0)||2 (0.7)|
|Others||3 (1.5)||3 (1.0)|
|Fitzpatrick skin classification, no. (%)|
|I||12 (6.2)||11 (13.8)||23 (8.4)|
|II||58 (30.1)||40 (50.0)||98 (35.9)|
|III||67 (34.7)||15 (18.8)||82 (30.0)|
|IV||39 (20.2)||13 (16.2)||52 (19.0)|
|V||17 (8.8)||1 (1.2)||18 (6.6)|
|VI||0 (0)||0 (0)||0 (0)|
The mean ± SD overall 25(OH)D level in the winter was 69.3 ± 28.5 nmoles/liter, and in the summer was 71.6 ± 25.3 nmoles/liter. The mean ± SD 25(OH)D levels in the northern center were 70.3 ± 28.7 and 73.9 ± 25.8 nmoles/liter in the winter and summer, respectively, and in the southern center they were 65.7 ± 27.6 and 67.3 ± 23.8 nmoles/liter, respectively. The 25(OH)D level in the northern study population was adequate in 41.3% of PsA patients in the winter and in 41.4% in the summer. In the southern population, 42.1% of PsA patients had adequate levels in the winter and 35.1% had adequate levels in the summer. At the northern site, the 25(OH)D level was insufficient in 55.7% of PsA patients in the winter and in 58.6% in the summer. At the southern site, 50.9% of patients had an insufficient level of 25(OH)D in the winter, with 62.2% having insufficient levels in the summer. At the northern site, 25(OH)D deficiency was detected in 3% of patients in the winter and in 0% in the summer. At the southern site, 7% of patients were deficient in 25(OH)D in the winter, with 2.7% found to be deficient in the summer (Table 2). No statistically significant difference in 25(OH)D level was found between the winter and summer in the northern (P = 0.91) and southern centers (P = 0.36). No statistically significant difference in 25(OH)D level was found in all patients (north and south) between winter and summer (P = 0.62).
|North||South||North and south|
|Adequate||83 (41.3)||24 (42.1)||107 (41.5)|
|Insufficient||112 (55.7)||29 (50.9)||141 (54.7)|
|Deficient||6 (3)||4 (7)||10 (3.8)|
|Adequate||58 (41.4)||26 (35.1)||84 (39.3)|
|Insufficient||82 (58.6)||46 (62.2)||128 (59.8)|
|Deficient||0 (0)||2 (2.7)||2 (0.9)|
PsA disease activity.
Table 3 shows disease activity in patients from both centers categorized by vitamin D levels as deficient, insufficient, normal, and overall in winter and summer. Overall, 161 patients (67.9%) reported joint pain in the winter and 132 (65%) reported joint pain in the summer. The mean ± SD number of actively inflamed joints was 5.01 ± 8.83 in the winter and 4.26 ± 7.11 in the summer. Fifty (21.3%) and 26 (12.9%) patients had inflammatory back pain in the winter and summer, respectively. Mean ± SD PASI scores ranged from 3.59 ± 5.09 in the winter to 3.44 ± 5.59 in the summer (Table 3).
|Variables/sites||North (n = 201)||South (n = 101)||North and south (n = 302)|
|Categorical variables, no. (%) when variable is present|
|Winter||4 (80.0)||67 (63.2)||56 (69.1)||127 (66.1)||2 (50.0)||20 (90.9)||12 (63.2)||34 (75.6)||6 (66.7)||87 (68.0)||68 (68.0)||161 (67.9)|
|Summer||N/A||50 (65.8)||40 (72.7)||90 (68.7)||1 (50.0)||25 (56.8)||16 (61.5)||42 (58.3)||1 (50.0)||75 (62.5)||56 (69.1)||132 (65.0)|
|Inflammatory back pain|
|Winter||2 (40.0)||25 (23.8)||16 (20.0)||43 (22.6)||1 (25.0)||4 (18.2)||2 (10.5)||7 (15.6)||3 (33.3)||29 (22.8)||18 (18.2)||50 (21.3)|
|Summer||N/A||11 (14.7)||8 (14.5)||19 (14.6)||0 (0.0)||3 (7.0)||4 (15.4)||7 (9.9)||0 (0.0)||14 (11.9)||12 (14.8)||26 (12.9)|
|Continuous variables, mean ± SD|
|Winter||9.83 ± 13.40||4.75 ± 9.03||5.01 ± 8.35||5.01 ± 8.89||2.00 ± 2.30||5.90 ± 10.50||4.63 ± 7.07||5.02 ± 8.64||6.70 ± 10.86||4.94 ± 9.25||4.94 ± 8.09||5.01 ± 8.83|
|Summer||N/A||4.37 ± 7.47||3.65 ± 6.31||4.06 ± 6.99||0 ± 0.00||4.24 ± 5.92||5.57 ± 9.58||4.60 ± 7.37||0.0 ± 0.00||4.32 ± 6.91||4.27 ± 7.51||4.26 ± 7.11|
|Winter||9.16 ± 16.42||8.37 ± 13.88||8.24 ± 13.69||8.34 ± 13.80||0 ± 0.00||9.09 ± 11.70||8.26 ± 8.88||7.90 ± 10.16||5.50 ± 13.12||8.49 ± 13.52||8.24 ± 12.90||8.26 ± 13.20|
|Summer||N/A||11.86 ± 16.59||6.80 ± 11.70||9.72 ± 14.88||0 ± 0.00||5.17 ± 8.62||6.76 ± 7.33||5.60 ± 8.10||0 ± 0.00||9.35 ± 14.46||6.79 ± 10.45||8.24 ± 12.99|
|Winter||2.70 ± 1.98||3.86 ± 5.23||3.03 ± 3.78||3.48 ± 4.61||3.60 ± 1.47||4.06 ± 6.40||4.16 ± 8.17||4.06 ± 6.86||3.06 ± 1.77||3.90 ± 5.41||3.24 ± 4.87||3.59 ± 5.09|
|Summer||N/A||3.05 ± 94.62||2.929 ± 3.50||3.00 ± 4.17||10.50 ± 8.91||3.21 ± 7.20||5.59 ± 8.30||4.35 ± 7.71||10.50 ± 8.91||3.10 ± 5.57||3.73 ± 5.51||3.44 ± 5.59|
Association between vitamin D level and geographic location and other covariates.
Demographic and geographic determinants and seasonal interaction.
There was no association between vitamin D levels and race, employment status, skin type, and other demographic covariates. There was no association between vitamin D levels, geographic location, or season. No significant interaction was found between the northern and southern centers and seasons (Table 4).
|Univariate model||Multivariate full model||Multivariate reduced model|
|Covariate first model|
|Centers: south vs. north||−5.095||0.105||−5.64||0.074||−5.64||0.074|
|Seasons: summer vs. winter||2.264||0.183||2.609||0.128||2.609||0.128|
|Covariate second model|
|Centers: south vs. north||−5.095||0.105||−7.134||0.063||−7.134||0.063|
|Seasons: summer vs. winter||2.264||0.183||1.976||0.308||1.976||0.308|
|Centers and seasons interaction||2.81||0.492|
Vitamin D level and association with disease activity.
Multivariate regression analysis, including linear mixed model for active joint counts, total damaged joints, and PASI, and logistic regression for inflammatory spinal pain showed no association between vitamin D levels (normal versus abnormal) and measures of disease activity/damage (results not shown). No association between disease activity in summer and vitamin D level (normal versus abnormal) in winter was found (results not shown).
In this cross-sectional study of patients with PsA, we found a high prevalence of vitamin D insufficiency. These results were surprising. Sun exposure and ultraviolet irradiation have long been recognized as beneficial for the control of psoriatic skin lesions, and patients with psoriasis and PsA usually seek sun exposure or are treated with ultraviolet irradiation. On the other hand, patients with psoriasis often wear clothing that covers their skin to hide their psoriasis, which would decrease their sun exposure. If they were not receiving phototherapy, they might be exposed to less UVB. Nevertheless, we hypothesized that patients with PsA would have a lower prevalence of vitamin D insufficiency/deficiency as compared to the general population and patients with other diseases. For instance, patients with lupus are advised to avoid sun exposure and recommended to use sunscreen, and thus might be expected to have vitamin D insufficiency. The peak of vitamin D levels usually occurs in late summer, and the nadir occurs in the beginning of spring (5). Vitamin D levels were insufficient in 55.7% of PsA patients from the northern center in the winter and 58.6% in the summer. Three percent of patients from the northern center were deficient in vitamin D in the winter and none were deficient in the summer. Despite being at lower latitude, vitamin D levels were low in Israeli patients from the southern center as well. In patients from the southern center, 50.9% and 62.2% were insufficient in vitamin D in winter and summer, respectively. Similar to patients from the northern center, deficient levels of vitamin D were prevalent in Israeli patients during the winter (7%) and to a lesser extent in the summer (2.7%). In areas with latitudes above 37° north or below 35° south, there is a marked decrease in UVB irradiance during the winter months, increasing the risk of vitamin D deficiency (5). Toronto, Canada (43° 40′ N), falls in the area that has been linked with a marked decrease in UVB irradiance during the winter, potentially explaining the high prevalence in vitamin D insufficiency/deficiency. Haifa, Israel (32° 46′ N), does not fall in this range; however, the high prevalence of vitamin D insufficiency could be related to other factors. For instance, the disability due to PsA might lead to a more sedentary lifestyle and less exposure to the sun. In addition, the increased awareness of sun-related damage could cause avoidance to exposure to the sun and wide use of sunscreen in Israel (26). On the other hand, studies have shown that the use of sunscreen does not decrease serum vitamin D concentration sufficiently to induce changes in metabolic markers, and the normal usage of sunscreen does not generally result in vitamin D insufficiency (27–29). Cultural differences in certain communities may also lead to decreased sun exposure.
Recent reports described a high prevalence of low 25(OH)D among healthy Canadians (8, 24, 30). The prevalence of 25(OH)D among patients with PsA has not been determined, nor has its correlation with disease activity. No consensus on vitamin D replacement regimens among PsA patients has been determined. We aimed to raise the awareness of vitamin D insufficiency/deficiency in this group of patients and understand whether there is a seasonal variation. In this study, we used linear mixed models and logistic regression models to adjust for demographic and clinical covariates, and we demonstrated that there was no statistically significant seasonal variation in 25(OH)D levels among PsA patients in the northern and southern centers. However, more patients were deficient in vitamin D in the northern center and in the southern center during winter as compared to summer. More patients had vitamin D insufficiency in the southern center during the summer as compared to winter. Moreover, no association could be established between disease activity and vitamin D levels.
Previous reports have estimated the prevalence of vitamin D deficiency to be up to 35% of elderly Israelis. However, this could be attributed to the decreased ability of the skin to synthesize vitamin D in the elderly (31). The insufficiency of vitamin D could have been linked with current advice to avoid sunlight exposure for fear of skin cancer and because of lifestyle modifications. However, we have witnessed a trend in the population to look for other sources of vitamin D and/or vitamin D–fortified milk and other products that could have decreased the implications of seasonal variations in vitamin D levels (31). On the other hand, individuals with adequate UVB exposure have been also reported to have low 25(OH)D levels (32). A study conducted on a sample of adults in Honolulu, Hawaii, suggests that variable responsiveness to UVB radiation is evident among individuals, causing some to have low vitamin D status despite abundant sun exposure (32). Vitamin D deficiency was also common in Chilean healthy postmenopausal women with adequate sun exposure but without vitamin D supplements (33).
Seasonal variation in vitamin D levels has been shown in previous studies in Canada. A high prevalence of 25(OH)D insufficiency and seasonal variations was shown in a community-dwelling population of healthy Canadians living in Calgary (8). Rucker et al also observed seasonal fluctuations in serum 25(OH)D in western Canadians (8). A recent study from Toronto found a decrease in serum 25(OH)D levels of 16 nmoles/liter (range 54.4–39.4) from fall to winter (24). Vieth et al observed that in Toronto, mean levels of vitamin D in young white women decreased in summer from 76 nmoles/liter to 58 nmoles/liter (30). The Canadian Health Measures Survey showed that for those ages 20–39 years, there was a significant seasonal variation in 25(OH)D levels of 9 nmoles/liter between winter and spring (34). In our study, the mean ± SD difference of serum 25(OH)D levels from winter to fall in Toronto was 3.6 ± 2.9 nmoles/liter, in Haifa was 1.6 ± 3.8 nmoles/liter, and from both centers (Toronto and Haifa) was 2.3 ± 3.2 nmoles/liter, and this was not clinically or statistically different.
Several factors affect the interpretation of vitamin D levels. In people with greater melanin in the skin, the vitamin D synthesis rate is decreased. African Americans are at higher risk for vitamin D deficiency (5, 16). In our study, we determined the skin phototypes using the Fitzpatrick classification. We found no association between vitamin D levels and skin phototype. One of the limitations of our study is that the majority of patients had skin phototypes II, III, and IV, with the minority having types I and V. Type VI represents the darker skin phototype, which was not represented in our study.
Obesity is a risk factor for lower levels of 25(OH)D, as this vitamin is stored in fat. A previous study of 644 PsA patients from the International Psoriasis and Arthritis Research Team database showed that the mean body mass index (BMI) was 29.6 kg/m2, which was higher than patients with psoriasis (BMI 27.9 kg/m2), patients with rheumatoid arthritis (BMI 27.3 kg/m2), and the normal population (BMI 26.1 kg/m2) (35).
Evidence suggests that aging decreases the capacity of human skin to produce vitamin D as the content of 7-dehydrocholesterol decreases, and that the elderly experience intestinal malabsorption of fat-soluble vitamin D (31). These studies evaluated a population age 65 years and older (36, 37). In our study, the mean ± SD age of the patients was 53.4 ± 12.9 years, and no association was found between vitamin D levels and age.
Although studies have shown that several drugs alter vitamin D levels, including anticonvulsants, corticosteroids, cimetidine, and others, we did not find an association between vitamin D levels and NSAIDs, DMARDs, and biologic treatments (5). In our patients, the mean ± SD PASI scores were 3.59 ± 5.09 and 3.44 ± 5.59 in winter and summer, respectively. Topical steroid use in PsA patients is very common as a treatment modality for psoriasis. However, in this study we did not evaluate if there is a link between topical steroid use and 25(OH)D levels. Factors such as vitamin D intake and sun exposure, in addition to other lifestyle determinants that affect vitamin D levels, need to be evaluated in future studies.
The prevalence of vitamin D insufficiency has been studied in few rheumatologic diseases. In a predominantly elderly male rheumatoid arthritis population, results showed that 84% of patients were insufficient and 42% were deficient (17). The prevalence of vitamin D insufficiency/deficiency is even higher than the results shown in our study; however, the variations in the laboratory techniques employed, the 25(OH)D cutoff values used to define insufficiency/deficiency, and patient characteristics (including ethnicity) varied between studies, which made direct comparisons difficult. In this study, they found that latitude variation between northern and southern sites did not appear to affect vitamin D levels or disease activity (17). In patients with lupus, suboptimal 25(OH)D serum levels were found in 66.7% of the patients and deficient levels were found in 17.9% of the patients, with lower levels in winter as compared to summer. Similar to our findings, in lupus patients no association with lupus disease activity could be found (14).
The specification of cut points for serum 25(OH)D levels has serious ramifications on the conclusions about vitamin D nutritive and clinical practice. The Dietary Reference Intake Committee's review of data suggests that persons are at risk of deficiency at 25(OH)D levels below 30 nmoles/liter (12 ng/ml). Moreover, they stated that some but not all persons are potentially at risk for inadequacy at serum 25(OH)D levels between 30 and 50 nmoles/liter (between 12 and 20 ng/ml). More importantly, they highlighted that serum 25(OH)D levels above 75 nmoles/liter (30 ng/ml) are not consistently associated with an increased benefit (19). The Dietary Reference Intake reported that the estimated average requirement for vitamin D in an individual age >9 to 70 years is 400 IU/day, is 800 IU/day in an individual age >70 years, and the recommended dietary allowance is 600 IU/day for individuals ages >9 to 70 years and 800 IU/day in those ages >70 years (19). These guidelines could be adopted by our patients as long as they do not have other medical issues or take medications that can interfere with vitamin D absorption and metabolism; for instance, steroid use, patients who have limited exposure to sun because of inability to ambulate, or because of joint pain or damage and/or handicap.
In summary, to our knowledge, this is the first study to provide data on the prevalence of vitamin D insufficiency/deficiency in PsA patients and to evaluate vitamin D levels in these patients with respect to seasonal and latitude variation. We report that the prevalence of 25(OH)D insufficiency/deficiency is high in PsA patients, as has been demonstrated in other rheumatologic diseases. Seasonal and latitude variations in our study were not statistically significant. Additional research is required to elucidate if the vitamin D intake needed to achieve and maintain optimal vitamin D levels in PsA patients is greater than that recommended for the general population.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Gladman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Touma, Eder, Zisman, Chandran, Rosen, Cook, Gladman.
Acquisition of data. Touma, Eder, Zisman, Feld, Chandran, Rosen, Gladman.
Analysis and interpretation of data. Touma, Eder, Zisman, Chandran, Rosen, Shen, Cook, Gladman.
- 2Psoriatic arthritis. In: Gordon K, Ruderman E, editors. Psoriasis and psoriatic arthritis. Berlin: Springer-Verlag; 2005. p. 57–65..
- 19Dietary reference intakes for calcium and vitamin D. Washington, DC: Institute of Medicine; 2010., , , .
- 27Clinically prescribed sunscreen (sun protection factor 15) does not decrease serum vitamin D concentration sufficiently either to induce changes in parathyroid function or in metabolic markers. Br J Dermatol 1998; 139: 422–7., , , , , .
- 34Vitamin D status of Canadians as measured in the 2007 to 2009 Canadian Health Measures Survey. Health Rep 2010; 21: 47–55., , , , .
- 35Differences in body mass index among individuals with psoriatic arthritis, psoriasis, rheumatoid arthritis, and the general population [abstract]. Arthritis Rheum 2010; 62 Suppl: S218., , , , , , et al.