EJINME-03132; No of Pages 4 European Journal of Internal Medicine xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case–control study M.E. Sarre a, C. Annweiler b,c, E. Legrand d, L. Martin a, O. Beauchet e,f,g,⁎ a
Department of Dermatology, UNAM, Angers University Hospital, Angers, France Department of Neuroscience, Division of Geriatric Medicine, UPRES EA 4638, UNAM, Angers University Hospital, Angers, France Robarts Research Institute, Department of Medical Biophysics, Schulich School of Medicine and Dentistry, The University of Western Ontario, London, Ontario, Canada d Division of Rheumatology, UNAM, Angers University Hospital, Angers, France e Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis — Jewish General Hospital and Lady Davis Institute for Medical Research, McGill University, Montreal, Quebec, Canada f Dr. Joseph Kaufmann Chair in Geriatric Medicine, Faculty of Medicine, McGill University, Montreal, Quebec, Canada g Centre of Excellence on Aging and Chronic Diseases of McGill integrated University Health Network, Quebec, Canada b c
a r t i c l e
i n f o
Article history: Received 6 November 2015 Received in revised form 23 January 2016 Accepted 3 February 2016 Available online xxxx Keywords: Bullous pemphigoid Hypovitaminosis Vitamin D Aged, 80 and over
a b s t r a c t Objectives: To compare serum vitamin D status in older inpatients with bullous pemphigoid (BP) and matched inpatients without BP, and to examine whether hypovitaminosis D, a high comorbidity burden or their combination were associated with BP. Methods: This prospective case–control study was performed from November 2012 to February 2014. A total of 90 consecutive older inpatients (31 consecutive inpatients with a de novo diagnosis of active BP, and 59 matched controls without BP) were recruited in the Department of Dermatology of Angers University Hospital, France. Hypovitaminosis D was defined as serum 25-hydroxyvitamin D (25OHD) concentration b 50 nmol/L. Age, gender, functional level, sun exposure, season, comorbidity burden and cognitive performance were used as covariates. Results: There was no significant difference between older inpatients with and without BP. Fully adjusted logistic regression showed a significant association between BP and hypovitaminosis D (odds ratio [OR] = 3.7, P = 0.046). The analysis of interaction between hypovitaminosis D and comorbidity burden showed that only the association of both was significantly associated with PB (OR = 3.1, P = 0.042). Conclusions: BP was significantly associated with hypovitaminosis D solely in patients with a high comorbidity burden among the older in-patients studied. This result suggests a complex interplay between hypovitaminosis D and BP, explaining the mixed results reported previously in the literature. © 2016 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Bullous pemphigoid (BP) is an autoimmune subepidermal blistering disease mainly affecting older adults (i.e., ≥70 years) with an incidence of around 20 new cases per 1,000,000 persons per year [1,2]. BP is a severe disease because of the high mortality rate in older patients with a high comorbidity burden (i.e., a high number of chronic or acute diseases other than BP index-disease) [2,3]. It justifies developing prevention strategies of BP. However, such an objective cannot be achieved without first understanding the immunopathogenic
Abbreviations: BMI, body mass index; BP, bullous pemphigoid; IADL, instrumental activity daily living; OR, odds ratio; S-MMSE, short mini mental status examination; 25OHD, 25-hydroxy vitamin D. ⁎ Corresponding author at: Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis Jewish General Hospital, Room E-0078.13755 chemin de la Côte-Sainte-Catherine, Montréal, QC H3T 1E2, Canada. Tel:. + 1 514 340 8222x4765; Fax: + 1 514 340 7547. E-mail address: [email protected] (O. Beauchet).
mechanisms of BP, which remain to date not fully elucidated. Among the biological determinants likely to influence autoimmunity in older adults, serum 25-hydroxyvitamin D (25OHD) concentration, which is considered the best indicator of vitamin D supply from cutaneous synthesis and diet, could be a candidate. Hypovitaminosis D (i.e., serum 25OHD concentration b 50 nmol/L) is highly frequent in older adults with a prevalence estimated around 70% over 70 years of age [4,5]. The clinical relevance is that hypovitaminosis D effects are not restricted to bone, but target a large number of non-bone processes including the immune system and the maintenance of self-tolerance [6–8]. As an illustration, higher vitamin D status has been associated with risk reduction and clinical improvement of autoimmune disorders such as lupus, multiple sclerosis [9,10]. Thereby, hypovitaminosis D may trigger or exacerbate autoimmunity, and hypovitaminosis D is associated with the incidence and/or severity of various autoimmune disorders [6–8]. We hypothesized that these mixed results could be explained by an interaction between hypovitaminosis D and a high comorbidity burden, hypovitaminosis D being associated with BP only in older patients with a high comorbidity burden.
http://dx.doi.org/10.1016/j.ejim.2016.02.004 0953-6205/© 2016 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
2
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
This study aims to 1) compare serum vitamin D status in older inpatients with BP and matched-controls without BP, and 2) examine whether hypovitaminosis D, a high comorbidity burden or their combination were associated with BP. 2. Materials and methods 2.1. Participants Using a case–control design, 90 consecutive older inpatients (31 consecutive inpatients with a de novo diagnosis of active BP and 59 matchedcontrol inpatients without BP) hospitalized in the Department of Dermatology of Angers University Hospital (France) between November 2012 and February 2014, were recruited for this study. Controls were consecutively recruited in the same ward in the Department of Dermatology of Angers University Hospital, France, and during the same period of inclusion as cases (i.e., between November 2012 and February 2014). For each case included in the study with the diagnosis of bullous pemphigoid, we included consecutively and successively 2 controls who were the two first patients hospitalized after the inclusion of the case. They were matched on age (±3 years), gender and skin complexion (i.e., black skin or not). Inpatients with non-BP bullous diseases and those who had received vitamin D supplements within the previous months were excluded. The diagnosis of active BP was based on clinical criteria [11], skin pathology results and a positive direct immunofluorescence test [12]. 2.2. Assessment Early morning venous blood was collected on the day of inclusion (i.e., before BP treatment) from fasted resting in-patients. Serum 25OHD level was measured by radioimmunoassay (DiaSorin Inc., Stillwater, MN). Hypovitaminosis D was defined as a serum 25OHD concentration b 50 nmol/L [13]. With this method, there is no lipid interference, which is often observed in other non-chromatographic assays of serum 25OHD concentrations. All measurements were performed at the University Hospital of Angers, France. During a full examination by a physician the following information was recorded: demographic (i.e., age and gender), functional and environmental characteristics (i.e., instrumental activities of daily living score (IADL), developed to assess more complex activities necessary for functioning in community settings like shopping, cooking, and managing finances) [14], sun exposure at midday assessed using the following standardized question “When the weather is nice, do you stay outside more than 15 min exposed to the sun (face and hands uncovered) between 11 am and 3 pm?” [15] and season of evaluation, as well as morbidity burden defined by a cognitive decline (i.e., score on the short version of the Mini-Mental State Examination (S-MMSE) b 6) [16], undernutrition (i.e., body mass index [BMI] b 20 kg/m2) and number of chronic diseases (i.e., diseases of indefinite duration or running a course with minimal change). A high morbidity burden was defined as a combination of cognitive decline and/or undernutrition and/or a number of chronic diseases N3 [17]. 2.3. Standard protocol approvals, registrations, and participant consents All eligible participants were included in the study after giving their consent. The study was conducted in accordance with ethical standards set forth in the Helsinki Declaration (1983). The project was approved by the local Ethics Committee (No. 1635125v0). 2.4. Statistical analyses The participants' characteristics were summarized using means and standard deviations (SD) or frequencies and percentages, as appropriate. Comparisons between groups were performed using
the Mann–Whitney test, unpaired t-test, Chi-square test or Fisher exact test, as appropriate. Multiple logistic regressions were used to examine the associations between BP (dependent variable) and hypovitaminosis D, morbidity burden or the three possible combinations (i.e., hypovitaminosis D, high morbidity burden, combination of hypovitaminosis D plus high morbidity burden) as independent variables, while adjusting for functional and environmental factors. P-values b0.05 were considered significant. All statistical analyses were performed using SPSS (v19.0, IBM Corporation, Chicago, IL). 3. Results There was no significant difference between older inpatients with and without BP (Table 1). The diagnoses in the control group were skin diseases (n = 41, 69.5%) including the management of chronic wounds (n = 10); skin infections (n = 9) including erysipelas (n = 7), varicella (n = 1) and ecthyma (n = 1); tumors (n = 9) including squamous cell carcinoma (n = 1), extramammary Paget disease (n = 1), Kaposi disease (n = 1), Merkel cell carcinoma (n = 1), mycosis fungoides (n = 5); pruritus (n = 8) including eczema (n = 2), psoriasis (n = 3), pruritus sine materia (n = 3); and connective tissue diseases (n = 5) including lupus (n = 1), Sjögren syndrome (n = 1), panarteritis nodosa (n = 1). A total of 18 (30.5%) controls were admitted for nondermatologic diseases with a diagnosis of cardiovascular diseases (n = 7) (heart failure (n = 5), cardiac arrhythmia (n = 1), hypertension (n = 1)), fall (n = 2), digestive occlusion (n = 1), Basedow disease (n = 1) and inability to stay at home (n = 7). All participants had a serum 25OHD concentration below 75 nmol/L with a concentration ranged from 10 to 73 nmol/L. The association between BP and hypovitaminosis D was significant in the fully adjusted logistic regression (odds ratio [OR] = 3.7 with P = 0.046; Table 2). Fig. 1 shows that only the combination of hypovitaminosis D with high cormorbidity burden was significantly associated with PB (OR = 3.1 with P = 0.042). 4. Discussion The findings showed that BP was significantly associated with hypovitaminosis D in older inpatients only in the case of high comorbidity burden, suggesting a more complex interplay between hypovitaminosis D and BP than suggested to date. There is growing epidemiological evidence of a beneficial effect of higher vitamin D status in the onset and progression of autoimmune disorders [18]. Only three studies [19–21] focused specifically on the association between low serum 25OHD concentrations and BP, and they reported mixed results. While Marzano et al. [19,21] showed a significant association, Turkaj et al. [20] reported no association. Various limitations precluded any definite conclusion regarding the involvement of hypovitaminosis D in the aetiopathogenesis of BP. Apart from rather small sample sizes, important confounders such as comorbidities, cognition or functional status were not taken into account in previous analyses. Our results suggest that these confounders, and in particular the comorbidity burden, may greatly influence expression of an association between hypovitaminosis D and PB. Our results are therefore consistent with evidence that vitamin D may influence autoimmune disorders [8,18]. Measuring serum 25OHD concentrations with comprehensive clinical examinations in consecutive BP cases and matched controls in a single research center and a robust a priori hypothesis represent the main strengths of the present study. However, the case–control design provides no information on causal relationships and, like in previous studies, the sample size was relatively small. Furthermore, we included in our study only two controls per case. In addition, no control on vitamin D dietary intake was done in the studied participants, and information on sun exposure was limited to face and hands and has no reproducible period of time.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
3
Table 1 Comparison of characteristics of cases with bullous pemphigoid (n = 31), and matched controls without bullous pemphigoid (n = 59). Characteristics
Total sample (n = 90)
Serum 25-hydroxyvitamin D concentration Mean ± SD (nmol/L) b50 nmol/L, (n%) Demographic Age, mean ± SD (years) Female gender, n (%) Functional and environmental IADL score, /4 Mean ± SD b4, n (%) Sun exposure at midday, n (%) Season of evaluation Spring, n (%) Summer, n (%) Fall, n (%) Winter, n (%) Disease burden Body mass index, kg/m2 Mean ± SD b20, n (%) Number of chronic diseases Mean ± SD N3 chronic diseases, n (%) S-MMSE score (/6) Mean ± SD b6, n (%)
P-value⁎
Bullous pemphigoid Yes (n = 31)
No (n = 59)
33.0 ± 18.8 69 (76.7)
29.3 ± 17.3 26 (86.7)
34.9 ± 19.4 43 (71.7)
0.246 0.113
83.0 ± 6.6 60 (66.7)
82.9 ± 6.9 20 (66.7)
83.0 ± 6.5 40 (66.7)
0.925 1.000
1.9 ± 1.5 71 (78.9) 36 (40.0)
1.6 ± 1.6 25 (83.3) 11 (36.7)
2.1 ± 1.5 46 (76.7) 25 (41.7)
12 (13.3) 24 (26.7) 24 (26.7) 30 (33.3)
4 (13.3) 8 (26.7) 8 (26.7) 10 (33.3)
8 (13.3) 16 (26.7) 16 (26.7) 20 (33.3)
0.465 0.767 0.648 1.000 – – – –
26.6 ± 6.4 10 (11.1)
25.5 ± 5.0 3 (10.0)
27.2 ± 6.9 7 (11.7)
0.204 0.813
5.1 ± 2.5 64 (71.1)
5.3 ± 2.3 22 (73.3)
5.1 ± 2.5 42 (70.0)
0.489 0.742
3.8 ± 1.9 70 (77.8)
3.4 ± 2.1 26 (86.7)
4.0 ± 1.8 44 (73.3)
0.232 0.151
Data presented as mean ± standard deviation where applicable. IADL, Instrumental Activities of Daily Living; S-MMSE, Short version of Mini-Mental State Examination; *Comparisons between participants with and without bullous pemphigoid based on the Mann–Whitney test, t-test or Chi-square, as appropriate.
In conclusion, BP was significantly associated with hypovitaminosis D only when there was a high comorbidity burden in the older inpatients studied. This result suggests a complex interplay between hypovitaminosis D and BP, which may explain the mixed results reported to date.
more complex interplay between hypovitaminosis D and BP than suggested to date.
Author's contributions
5. Learning points • Bullous pemphigoid is an autoimmune subepidermal blistering disease mainly affecting adults aged over 70 years. • Hypovitaminosis D is associated with a greater incidence and/or severity of different autoimmune disorders. • The association between hypovitaminosis D and BP is still open to debate, some studies showing an association, while others have failed to find any association. • BP was significantly associated with hypovitaminosis D in older inpatients solely in the case of a high comorbidity burden, suggesting a
− MES has full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analyses. − Study concept and design: MES, LM, OB and CA. − Acquisition of data: MES and LM. − Analysis and interpretation of data: MES and OB. − Drafting of the manuscript: MES and OB. − Critical revision of the manuscript for significant intellectual content: LM, EL and CA. − Funding obtained: MES. − Statistical expertise: OB.
Table 2 Multiple logistic regression models examining the cross-sectional association between hypovitaminosis D and bullous pemphigoid (n = 90). Bullous pemphigoid Model 1
Hypovitaminosis D⁎ Age Female gender IADL b 4 Sun exposure at midday Season of evaluation Body mass index b20 kg/m2 Number chronic diseases N3 S-MMSE score b 6
Model 2
Model 3
Adjusted OR [95% CI]
P-value
Adjusted OR [95% CI]
P-value
Fully adjusted OR [95% CI]
P-value
2.7 [0.8; 9.1] 1.0 [0.9; 1.1] 1.0 [0.4; 2.7] – – – – – –
0.110 0.680 0.950 – – – – – –
2.8 [0.8; 9.6] 1.0 [0.9; 1.1] 0.9 [0.3; 2.5] 1.8 [0.5; 6.4] 0.8 [0.3; 2.3] 1.1 [0.6; 1.7] – – –
0.097 0.482 0.813 0.390 0.735 0.830 – – –
3.7 [1.0; 13.5] 1.0 [0.9; 1.0] 1.0 [0.3; 2.8] 1.3 [0.3; 5.1] 0.9 [0.3; 2.6] 1.1 [0.7; 1.8] 0.7 [0.2; 3.6] 1.1 [0.4; 3.5] 3.3 [0.8; 13.5]
0.046 0.244 0.945 0.727 0.861 0.655 0.697 0.823 0.093
Model 1) adjusted for demographic characteristics (i.e., age and gender); Model 2) Model 1 + adjustment for functional and environmental characteristics (Instrumental Activities of Daily Living score b 4, sun exposure and season of evaluation); Model 3) Model 2 + adjustment for disease burden (Body mass index b20 kg/m2, Number of chronic diseases N3, Short version of Mini-Mental State Examination score b 6); CI, confidence interval; IADL, Instrumental Activities of Daily Living; S-MMSE, Short Mini-Mental State Examination; OR, odds ratio; ⁎serum 25hydroxyvitamin D b 50 nmol/L; significant OR (P b 0.05) indicated in bold.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
4
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
Conflict of interest disclosures The authors have no conflicts of interest to report. References
for bullous pemphigoid serum 25-hydroxyvitamin D < 50 nmol/L S-MMSE score 3 Significant P-values (P
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case–control study M.E. Sarre a, C. Annweiler b,c, E. Legrand d, L. Martin a, O. Beauchet e,f,g,⁎ a
Department of Dermatology, UNAM, Angers University Hospital, Angers, France Department of Neuroscience, Division of Geriatric Medicine, UPRES EA 4638, UNAM, Angers University Hospital, Angers, France Robarts Research Institute, Department of Medical Biophysics, Schulich School of Medicine and Dentistry, The University of Western Ontario, London, Ontario, Canada d Division of Rheumatology, UNAM, Angers University Hospital, Angers, France e Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis — Jewish General Hospital and Lady Davis Institute for Medical Research, McGill University, Montreal, Quebec, Canada f Dr. Joseph Kaufmann Chair in Geriatric Medicine, Faculty of Medicine, McGill University, Montreal, Quebec, Canada g Centre of Excellence on Aging and Chronic Diseases of McGill integrated University Health Network, Quebec, Canada b c
a r t i c l e
i n f o
Article history: Received 6 November 2015 Received in revised form 23 January 2016 Accepted 3 February 2016 Available online xxxx Keywords: Bullous pemphigoid Hypovitaminosis Vitamin D Aged, 80 and over
a b s t r a c t Objectives: To compare serum vitamin D status in older inpatients with bullous pemphigoid (BP) and matched inpatients without BP, and to examine whether hypovitaminosis D, a high comorbidity burden or their combination were associated with BP. Methods: This prospective case–control study was performed from November 2012 to February 2014. A total of 90 consecutive older inpatients (31 consecutive inpatients with a de novo diagnosis of active BP, and 59 matched controls without BP) were recruited in the Department of Dermatology of Angers University Hospital, France. Hypovitaminosis D was defined as serum 25-hydroxyvitamin D (25OHD) concentration b 50 nmol/L. Age, gender, functional level, sun exposure, season, comorbidity burden and cognitive performance were used as covariates. Results: There was no significant difference between older inpatients with and without BP. Fully adjusted logistic regression showed a significant association between BP and hypovitaminosis D (odds ratio [OR] = 3.7, P = 0.046). The analysis of interaction between hypovitaminosis D and comorbidity burden showed that only the association of both was significantly associated with PB (OR = 3.1, P = 0.042). Conclusions: BP was significantly associated with hypovitaminosis D solely in patients with a high comorbidity burden among the older in-patients studied. This result suggests a complex interplay between hypovitaminosis D and BP, explaining the mixed results reported previously in the literature. © 2016 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Bullous pemphigoid (BP) is an autoimmune subepidermal blistering disease mainly affecting older adults (i.e., ≥70 years) with an incidence of around 20 new cases per 1,000,000 persons per year [1,2]. BP is a severe disease because of the high mortality rate in older patients with a high comorbidity burden (i.e., a high number of chronic or acute diseases other than BP index-disease) [2,3]. It justifies developing prevention strategies of BP. However, such an objective cannot be achieved without first understanding the immunopathogenic
Abbreviations: BMI, body mass index; BP, bullous pemphigoid; IADL, instrumental activity daily living; OR, odds ratio; S-MMSE, short mini mental status examination; 25OHD, 25-hydroxy vitamin D. ⁎ Corresponding author at: Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis Jewish General Hospital, Room E-0078.13755 chemin de la Côte-Sainte-Catherine, Montréal, QC H3T 1E2, Canada. Tel:. + 1 514 340 8222x4765; Fax: + 1 514 340 7547. E-mail address: [email protected] (O. Beauchet).
mechanisms of BP, which remain to date not fully elucidated. Among the biological determinants likely to influence autoimmunity in older adults, serum 25-hydroxyvitamin D (25OHD) concentration, which is considered the best indicator of vitamin D supply from cutaneous synthesis and diet, could be a candidate. Hypovitaminosis D (i.e., serum 25OHD concentration b 50 nmol/L) is highly frequent in older adults with a prevalence estimated around 70% over 70 years of age [4,5]. The clinical relevance is that hypovitaminosis D effects are not restricted to bone, but target a large number of non-bone processes including the immune system and the maintenance of self-tolerance [6–8]. As an illustration, higher vitamin D status has been associated with risk reduction and clinical improvement of autoimmune disorders such as lupus, multiple sclerosis [9,10]. Thereby, hypovitaminosis D may trigger or exacerbate autoimmunity, and hypovitaminosis D is associated with the incidence and/or severity of various autoimmune disorders [6–8]. We hypothesized that these mixed results could be explained by an interaction between hypovitaminosis D and a high comorbidity burden, hypovitaminosis D being associated with BP only in older patients with a high comorbidity burden.
http://dx.doi.org/10.1016/j.ejim.2016.02.004 0953-6205/© 2016 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
2
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
This study aims to 1) compare serum vitamin D status in older inpatients with BP and matched-controls without BP, and 2) examine whether hypovitaminosis D, a high comorbidity burden or their combination were associated with BP. 2. Materials and methods 2.1. Participants Using a case–control design, 90 consecutive older inpatients (31 consecutive inpatients with a de novo diagnosis of active BP and 59 matchedcontrol inpatients without BP) hospitalized in the Department of Dermatology of Angers University Hospital (France) between November 2012 and February 2014, were recruited for this study. Controls were consecutively recruited in the same ward in the Department of Dermatology of Angers University Hospital, France, and during the same period of inclusion as cases (i.e., between November 2012 and February 2014). For each case included in the study with the diagnosis of bullous pemphigoid, we included consecutively and successively 2 controls who were the two first patients hospitalized after the inclusion of the case. They were matched on age (±3 years), gender and skin complexion (i.e., black skin or not). Inpatients with non-BP bullous diseases and those who had received vitamin D supplements within the previous months were excluded. The diagnosis of active BP was based on clinical criteria [11], skin pathology results and a positive direct immunofluorescence test [12]. 2.2. Assessment Early morning venous blood was collected on the day of inclusion (i.e., before BP treatment) from fasted resting in-patients. Serum 25OHD level was measured by radioimmunoassay (DiaSorin Inc., Stillwater, MN). Hypovitaminosis D was defined as a serum 25OHD concentration b 50 nmol/L [13]. With this method, there is no lipid interference, which is often observed in other non-chromatographic assays of serum 25OHD concentrations. All measurements were performed at the University Hospital of Angers, France. During a full examination by a physician the following information was recorded: demographic (i.e., age and gender), functional and environmental characteristics (i.e., instrumental activities of daily living score (IADL), developed to assess more complex activities necessary for functioning in community settings like shopping, cooking, and managing finances) [14], sun exposure at midday assessed using the following standardized question “When the weather is nice, do you stay outside more than 15 min exposed to the sun (face and hands uncovered) between 11 am and 3 pm?” [15] and season of evaluation, as well as morbidity burden defined by a cognitive decline (i.e., score on the short version of the Mini-Mental State Examination (S-MMSE) b 6) [16], undernutrition (i.e., body mass index [BMI] b 20 kg/m2) and number of chronic diseases (i.e., diseases of indefinite duration or running a course with minimal change). A high morbidity burden was defined as a combination of cognitive decline and/or undernutrition and/or a number of chronic diseases N3 [17]. 2.3. Standard protocol approvals, registrations, and participant consents All eligible participants were included in the study after giving their consent. The study was conducted in accordance with ethical standards set forth in the Helsinki Declaration (1983). The project was approved by the local Ethics Committee (No. 1635125v0). 2.4. Statistical analyses The participants' characteristics were summarized using means and standard deviations (SD) or frequencies and percentages, as appropriate. Comparisons between groups were performed using
the Mann–Whitney test, unpaired t-test, Chi-square test or Fisher exact test, as appropriate. Multiple logistic regressions were used to examine the associations between BP (dependent variable) and hypovitaminosis D, morbidity burden or the three possible combinations (i.e., hypovitaminosis D, high morbidity burden, combination of hypovitaminosis D plus high morbidity burden) as independent variables, while adjusting for functional and environmental factors. P-values b0.05 were considered significant. All statistical analyses were performed using SPSS (v19.0, IBM Corporation, Chicago, IL). 3. Results There was no significant difference between older inpatients with and without BP (Table 1). The diagnoses in the control group were skin diseases (n = 41, 69.5%) including the management of chronic wounds (n = 10); skin infections (n = 9) including erysipelas (n = 7), varicella (n = 1) and ecthyma (n = 1); tumors (n = 9) including squamous cell carcinoma (n = 1), extramammary Paget disease (n = 1), Kaposi disease (n = 1), Merkel cell carcinoma (n = 1), mycosis fungoides (n = 5); pruritus (n = 8) including eczema (n = 2), psoriasis (n = 3), pruritus sine materia (n = 3); and connective tissue diseases (n = 5) including lupus (n = 1), Sjögren syndrome (n = 1), panarteritis nodosa (n = 1). A total of 18 (30.5%) controls were admitted for nondermatologic diseases with a diagnosis of cardiovascular diseases (n = 7) (heart failure (n = 5), cardiac arrhythmia (n = 1), hypertension (n = 1)), fall (n = 2), digestive occlusion (n = 1), Basedow disease (n = 1) and inability to stay at home (n = 7). All participants had a serum 25OHD concentration below 75 nmol/L with a concentration ranged from 10 to 73 nmol/L. The association between BP and hypovitaminosis D was significant in the fully adjusted logistic regression (odds ratio [OR] = 3.7 with P = 0.046; Table 2). Fig. 1 shows that only the combination of hypovitaminosis D with high cormorbidity burden was significantly associated with PB (OR = 3.1 with P = 0.042). 4. Discussion The findings showed that BP was significantly associated with hypovitaminosis D in older inpatients only in the case of high comorbidity burden, suggesting a more complex interplay between hypovitaminosis D and BP than suggested to date. There is growing epidemiological evidence of a beneficial effect of higher vitamin D status in the onset and progression of autoimmune disorders [18]. Only three studies [19–21] focused specifically on the association between low serum 25OHD concentrations and BP, and they reported mixed results. While Marzano et al. [19,21] showed a significant association, Turkaj et al. [20] reported no association. Various limitations precluded any definite conclusion regarding the involvement of hypovitaminosis D in the aetiopathogenesis of BP. Apart from rather small sample sizes, important confounders such as comorbidities, cognition or functional status were not taken into account in previous analyses. Our results suggest that these confounders, and in particular the comorbidity burden, may greatly influence expression of an association between hypovitaminosis D and PB. Our results are therefore consistent with evidence that vitamin D may influence autoimmune disorders [8,18]. Measuring serum 25OHD concentrations with comprehensive clinical examinations in consecutive BP cases and matched controls in a single research center and a robust a priori hypothesis represent the main strengths of the present study. However, the case–control design provides no information on causal relationships and, like in previous studies, the sample size was relatively small. Furthermore, we included in our study only two controls per case. In addition, no control on vitamin D dietary intake was done in the studied participants, and information on sun exposure was limited to face and hands and has no reproducible period of time.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
3
Table 1 Comparison of characteristics of cases with bullous pemphigoid (n = 31), and matched controls without bullous pemphigoid (n = 59). Characteristics
Total sample (n = 90)
Serum 25-hydroxyvitamin D concentration Mean ± SD (nmol/L) b50 nmol/L, (n%) Demographic Age, mean ± SD (years) Female gender, n (%) Functional and environmental IADL score, /4 Mean ± SD b4, n (%) Sun exposure at midday, n (%) Season of evaluation Spring, n (%) Summer, n (%) Fall, n (%) Winter, n (%) Disease burden Body mass index, kg/m2 Mean ± SD b20, n (%) Number of chronic diseases Mean ± SD N3 chronic diseases, n (%) S-MMSE score (/6) Mean ± SD b6, n (%)
P-value⁎
Bullous pemphigoid Yes (n = 31)
No (n = 59)
33.0 ± 18.8 69 (76.7)
29.3 ± 17.3 26 (86.7)
34.9 ± 19.4 43 (71.7)
0.246 0.113
83.0 ± 6.6 60 (66.7)
82.9 ± 6.9 20 (66.7)
83.0 ± 6.5 40 (66.7)
0.925 1.000
1.9 ± 1.5 71 (78.9) 36 (40.0)
1.6 ± 1.6 25 (83.3) 11 (36.7)
2.1 ± 1.5 46 (76.7) 25 (41.7)
12 (13.3) 24 (26.7) 24 (26.7) 30 (33.3)
4 (13.3) 8 (26.7) 8 (26.7) 10 (33.3)
8 (13.3) 16 (26.7) 16 (26.7) 20 (33.3)
0.465 0.767 0.648 1.000 – – – –
26.6 ± 6.4 10 (11.1)
25.5 ± 5.0 3 (10.0)
27.2 ± 6.9 7 (11.7)
0.204 0.813
5.1 ± 2.5 64 (71.1)
5.3 ± 2.3 22 (73.3)
5.1 ± 2.5 42 (70.0)
0.489 0.742
3.8 ± 1.9 70 (77.8)
3.4 ± 2.1 26 (86.7)
4.0 ± 1.8 44 (73.3)
0.232 0.151
Data presented as mean ± standard deviation where applicable. IADL, Instrumental Activities of Daily Living; S-MMSE, Short version of Mini-Mental State Examination; *Comparisons between participants with and without bullous pemphigoid based on the Mann–Whitney test, t-test or Chi-square, as appropriate.
In conclusion, BP was significantly associated with hypovitaminosis D only when there was a high comorbidity burden in the older inpatients studied. This result suggests a complex interplay between hypovitaminosis D and BP, which may explain the mixed results reported to date.
more complex interplay between hypovitaminosis D and BP than suggested to date.
Author's contributions
5. Learning points • Bullous pemphigoid is an autoimmune subepidermal blistering disease mainly affecting adults aged over 70 years. • Hypovitaminosis D is associated with a greater incidence and/or severity of different autoimmune disorders. • The association between hypovitaminosis D and BP is still open to debate, some studies showing an association, while others have failed to find any association. • BP was significantly associated with hypovitaminosis D in older inpatients solely in the case of a high comorbidity burden, suggesting a
− MES has full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analyses. − Study concept and design: MES, LM, OB and CA. − Acquisition of data: MES and LM. − Analysis and interpretation of data: MES and OB. − Drafting of the manuscript: MES and OB. − Critical revision of the manuscript for significant intellectual content: LM, EL and CA. − Funding obtained: MES. − Statistical expertise: OB.
Table 2 Multiple logistic regression models examining the cross-sectional association between hypovitaminosis D and bullous pemphigoid (n = 90). Bullous pemphigoid Model 1
Hypovitaminosis D⁎ Age Female gender IADL b 4 Sun exposure at midday Season of evaluation Body mass index b20 kg/m2 Number chronic diseases N3 S-MMSE score b 6
Model 2
Model 3
Adjusted OR [95% CI]
P-value
Adjusted OR [95% CI]
P-value
Fully adjusted OR [95% CI]
P-value
2.7 [0.8; 9.1] 1.0 [0.9; 1.1] 1.0 [0.4; 2.7] – – – – – –
0.110 0.680 0.950 – – – – – –
2.8 [0.8; 9.6] 1.0 [0.9; 1.1] 0.9 [0.3; 2.5] 1.8 [0.5; 6.4] 0.8 [0.3; 2.3] 1.1 [0.6; 1.7] – – –
0.097 0.482 0.813 0.390 0.735 0.830 – – –
3.7 [1.0; 13.5] 1.0 [0.9; 1.0] 1.0 [0.3; 2.8] 1.3 [0.3; 5.1] 0.9 [0.3; 2.6] 1.1 [0.7; 1.8] 0.7 [0.2; 3.6] 1.1 [0.4; 3.5] 3.3 [0.8; 13.5]
0.046 0.244 0.945 0.727 0.861 0.655 0.697 0.823 0.093
Model 1) adjusted for demographic characteristics (i.e., age and gender); Model 2) Model 1 + adjustment for functional and environmental characteristics (Instrumental Activities of Daily Living score b 4, sun exposure and season of evaluation); Model 3) Model 2 + adjustment for disease burden (Body mass index b20 kg/m2, Number of chronic diseases N3, Short version of Mini-Mental State Examination score b 6); CI, confidence interval; IADL, Instrumental Activities of Daily Living; S-MMSE, Short Mini-Mental State Examination; OR, odds ratio; ⁎serum 25hydroxyvitamin D b 50 nmol/L; significant OR (P b 0.05) indicated in bold.
Please cite this article as: Sarre ME, et al, Association between bullous pemphigoid and hypovitaminosis D in older inpatients: Results from a case– control study, Eur J Intern Med (2016), http://dx.doi.org/10.1016/j.ejim.2016.02.004
4
M.E. Sarre et al. / European Journal of Internal Medicine xxx (2016) xxx–xxx
Conflict of interest disclosures The authors have no conflicts of interest to report. References
for bullous pemphigoid serum 25-hydroxyvitamin D < 50 nmol/L S-MMSE score 3 Significant P-values (P
