- Email: [email protected]
Impact of micronutrient deficiency & malnutrition in systemic sclerosis: Cohort study and literature review
Accepted Manuscript Impact of Micronutrient Deficiency & Malnutrition in Systemic Sclerosis: Cohort Study and Literature Review
Romain Dupont, Mélanie Longué, Anne Galinier, Christel Cinq Frais, Cécile Ingueneau, Léonardo Astudillo, Philippe Arlet, Daniel Adoue, Laurent Alric, Grégoire Prévot, Bastien Cabarrou, Laurent Sailler, Grégory Pugnet PII: DOI: Reference:
S1568-9972(18)30211-8 doi:10.1016/j.autrev.2018.05.010 AUTREV 2216
To appear in:
Autoimmunity Reviews
Received date: Accepted date:
18 May 2018 25 May 2018
Please cite this article as: Romain Dupont, Mélanie Longué, Anne Galinier, Christel Cinq Frais, Cécile Ingueneau, Léonardo Astudillo, Philippe Arlet, Daniel Adoue, Laurent Alric, Grégoire Prévot, Bastien Cabarrou, Laurent Sailler, Grégory Pugnet , Impact of Micronutrient Deficiency & Malnutrition in Systemic Sclerosis: Cohort Study and Literature Review. Autrev (2018), doi:10.1016/j.autrev.2018.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Impact of Micronutrient Deficiency & Malnutrition in Systemic Sclerosis: Cohort Study and Literature Review Romain Dupont1, Mélanie Longué 2, Anne Galinier3, Christel Cinq Frais3 , Cécile Ingueneau3, Léonardo Astudillo1, Philippe Arlet1, Daniel Adoue 4, Laurent Alric5,6, Grégoire Prévot7 , Bastien Cabarrou2, Laurent Sailler1,8,9 and Grégory Pugnet1,9 1
Service de Médecine interne, URM, CHU de Toulouse, France Service de Biostatistiques, Institut Universitaire du Cancer de Toulouse – Oncopôle 3 Laboratoire de biochimie et nutrition, Institut Fédératif de Biologie, CHU de Toulouse, France 4 Service de Médecine interne et immunopathologie clinique, CHU de Toulouse, France 5 Service de Médecine interne, pavillon Dieulafoy, CHU de Toulouse, France 6 UMR 152 IRD-Université de Toulouse, France 7 Service de Pneumologie, CHU de Toulouse, France 8 CIC 1436, CHU de Toulouse, France 9 UMR 1027 Inserm-Université de Toulouse, France
RI
PT
2
NU
SC
Corresponding author: Romain Dupont, Service de Médecine Interne, salle Le Tallec, URM, Hôpital Purpan, place du Dr Joseph Baylac, TSA 40031, 31059 Toulouse cedex 9, France; Phone: +33 5 61 77 22 78; Fax: +33 5 61 77 71 24; Email: [email protected]
AC C
EP T
ED
MA
ABSTRACT Objectives: The purpose of our study was to determine the prevalence and risk factors associated with malnutrition, and selenium (Se) and vitamin C (vitC) deficiencies in systemic sclerosis (SSc) patients. Methods: We included adult SSc patients fulfilling the 2013 ACR/EULAR criteria from the Toulouse University Hospital cohort who underwent a micronutrient workup (including vitC, Se or thiamine levels) between 2011 and 2016. Results: 82 patients were included, mostly women (76%), with a median age of 60 years. SSc was limited in 76% of the cases, with Scl-70 and centromere antibodies in 32% and 44%, respectively. Median disease duration was 7.4 years. Cardiac involvement was noticed in 19% and gastrointestinal tract in and 95%; 9% had pulmonary artery hypertension (PAH) and 63% had interstitial lung disease. Overt malnutrition was present in 14 (17%) patients. Micronutrient deficiencies included Se (35%), vitC (31%) and/or thiamine (6%). Malnourished patients had significantly a higher summed Medsger disease severity scales (7.5 vs. 5, P = 0.003), lower hemoglobin (10.6 vs. 12.9 g/dL, P < 0.0001) and vitC levels (3.6 vs. 10.6 mg/L, P = 0.003). Cardiac involvement was significantly associated with Se deficiency (OR 6.2, IC 95%[1.48-32.70], P = 0.05). The factors associated with vitC deficiency were malnutrition (OR 8.57, IC 95%[2.16-43.39], P = 0.003), modified Rodnan skin score ≤ 14 (OR 0.33, IC95[0.11-1], P = 0.05), PAH (27% in deficient vs. none in non-deficient patients, P = 0.0006) and esophagitis or Barrett’s mucosa (OR 4.05, IC95[1.27-13.54], P = 0.02). Conclusions: Se testing should be considered as soon as cardiac involvement is suspected. VitC testing should be considered in malnourished SSc patients, especially if skin involvement is extensive. Keywords: systemic sclerosis, malnutrition, selenium, vitamin C, deficiency Highlights: Micronutrient deficiency is frequent in SSc and is not associated with malnutrition SSc primary heart disease is associated with selenium deficiency Malnutrition and SSc severity is associated with vitamin C deficiency Routine micronutrient testing and targeted supplementation might be useful in SSc
Main text 1 INTRODUCTION
ACCEPTED MANUSCRIPT
NU
SC
RI
PT
Systemic sclerosis (SSc) is a rare multisystem disease characterized by three pathogenetic landmarks: microvascular involvement, activation of the immune system and increase of extracellular matrix deposition in the skin and internal organs [1]. This severe chronic disease is of high nutritional risk, estimated to be around 20% [2]. Nutritional status cannot be inferred from gastrointestinal (GI) tract involvement alone [2]. Many factors could also contribute to nutritional impairment: mood disturbance, functional status and disabilities, oropharyngeal manifestations, pancreatic involvement, heart failure, pulmonary interstitial fibrosis, and inflammation [3]. Malnutrition itself involves to varying degrees a deficiency in energy, protein and micronutrients (vitamins and trace elements), depending on respective intakes, absorptions and losses. SSc patients have been shown to have a lower intake in fruits and vegetables [4] and to be deficient in vitamin C (vitC) and selenium (Se) [4–6]. Whereas micronutrient deficiency is associated with a wide range of pathological conditions, the consequences of micronutrient deficiency in SSc patients have been poorly described. Selenium, vitC and thiamine pyrophosphate deficiencies are known causes of dilated, ischemic and high-output heart failure respectively [7]. This may be of utmost importance given that primary heart disease is the main offender in SSc explaining 30% of SSc deaths, while atherosclerosis is responsible for only 5%–8% of deaths [8]. As one of the physiological functions of Se and vitC is to eliminate free radicals, their deficiency could promote oxidative injury and progression of SSc [9]. The prognostic value of malnutrition or micronutrient deficiencies in SSc is still discussed, while survival seems to be associated with prealbumin levels, but neither BMI nor weight loss [2,10]. The aim of this study was to assess the prevalence and the factors associated with malnutrition, Se and vitamin C deficiencies, in a population of SSc patients, when performed during routine patient follow-up evaluation in a tertiary referral center for SSc.
MA
2 METHODS
EP T
ED
2.1 Study population We conducted a retrospective cross-sectional study in a cohort of all unselected consecutive patients diagnosed with a SSc at the Toulouse University Hospital, a tertiary referral center for SSc. After thorough medical chart review, patients were included if they 1) met the ACR/EULAR 2013 classification criteria [11], 2) were more than 18 years of age 3) between 2011 and 2016 had undergone a micronutrient workup including at least Se, Lascorbic acid (vitamin C) or thiamine pyrophosphate levels.
AC C
2.2 Data collection The following clinical and biological data were collected at baseline and annually, according to standard clinical guidelines [12]: a) age, gender, disease duration (since first non-Raynaud manifestation), weight, weight loss, height, body mass index (BMI); b) SSc sub-type according to Leroy and Medsger [13] and SSc treatment at baseline; b) extension of skin fibrosis as measured by the modified Rodnan skin score (mRSS)[14] and inter-incisive distance; c) past or present Raynaud’s phenomenon, and past or present digital ulcers; d ) cardiac and pulmonary evaluation, including results from pulmonary function tests (forced vital capacity (FVC), diffusing capacity of the lung for carbon monoxide (DLCO)), according to the ATS and European Respiratory Society (ATS/ERS) consensu s standards [15], presence of SSc-related interstitial lung disease (ILD) or pleural effusion on chest X -ray or on high resolution CT scans, results from echocardiography including left ventricular ejection fraction (LVEF) and systolic pulmonary arterial pressure (sPAP, mmHg) measurements, and the results of cardiac MRI when available. When resting sPAP on Doppler echocardiography was superior to 35 mmHg, right heart catheterization was performed. Pulmonary arterial hypertension (PAH) was confirmed as a mean PAP ≥25 mmHg and was considered as SSc-associated pre-capillary PAH when combined with a pulmonary artery wedge pressure ≤ 15 mmHg. Cardiac involvement was considered present if stated by the referent physician based on clinical, echocardiography and/or imaging findings; e) GI tract evaluation if needed, including results of gastroscopy and/or colonoscopy, confirmation of small intesti ne bacterial overgrowth by a hydrogen breath test, presence of digestive pseudo-obstruction on abdominal CT scans, presence of any motility disorder on esophageal manometry or gastric scintigraphy, and confirmation of pancreatic insufficiency by fecal elastase. Gastrointestinal involvement was considered present if suspected on clinical grounds; f) other relevant testing if needed for muscle, joint, renal, or hepatobiliary involvement, or neurologic manifestations; g) laboratory parameters, including hemoglobin level, serum creatinine, C-reactive protein, albumin, serum transaminases, alkaline phosphatase, γ-glutamyl transferase, creatinine phosphokinases, ferritin, status of
ACCEPTED MANUSCRIPT antinuclear, anticentromere and anti-topoisomerase-I autoantibodies, and micronutrient testing. Patient survival status was evaluated during and at the end of the follow-up of the entire cohort.
2.3 Nutritional data Micronutrient testing was performed in the Toulouse University Hospital laboratory using fluorescent liquid chromatography for thiamine pyrophosphate (thiamine), pyridoxal phosphate (pyridoxine) and vitC levels; atomic absorption spectrophotometry for serum Se and serum zinc; and chemiluminescence for serum cobalamin (B12) and serum folate. Nutritional workup for indication was classified in four categories: systematic, related to symptoms or signs (in the context of diagnostic investigations), high nutritional risk (overt clinical malnutrition, significant GI tract involvement, parenteral nutrition, or known micronutrient deficiency), before organ or autologous stem cell transplantation.
AC C
EP T
ED
MA
NU
SC
RI
PT
2.4 Outcomes Malnutrition was defined according to French HAS (National Health Authority) statement as follows [16,17]: for patients younger than 70, BMI ≤17, significant weight loss or (albumin ≤30 g/L and CRP ≤10 mg/L); for patients older than 70, BMI ≤21, significant weight loss or (albumin ≤35 g/L and CRP ≤10 mg/L). Weight loss was defined as significant if exceeding 5% body weight in 1 month or 10% body weight in 6 months. The MUST score was calculated as follows: [BMI > 20.0 = 0, 18.5–20.1 = 1, < 18.5 = 2] +[weight loss in the past 3–6 months: < 5% = 0, 5%–10% = 1, > 10% = 2] [18,19]. We did not incorporate a correction based on disease activity score as proposed by some authors [20,21]. MUST scores of 0, 1 and ≥2 correspond to low, medium and high nutritional risk, respectively. The summed Medsger disease severity scales (DSS) was calculated as follows [22]: [weight loss <5% and hemoglobin ≥12.3 g/dL =0, weight loss 5%-9.9% or hemoglobin 11-12.2 g/dL =1, weight loss 10-14.9% or hemoglobin 9.7-10.9g/dL =2, weight loss 15-19.9% or hemoglobin 8.3-9.6g/dL =3, weight loss ≥20% or hemoglobin <8.3 g/dL =4] + [no Raynaud’s or Raynaud’s not requiring vasodilators = 0; Raynaud’s requiring vasodilators = 1; digital pitting scars or ulcers = 2] + [mRSS 0 = 0, mRSS 1-14 = 1, mRSS 15-29 = 2, mRSS 30-39 = 3, mRSS ≥40 = 4] + [no myositis = 0; myositis = 2] + [no motility disorder = 0, distal esophageal hypoperistalsis =1, achalasia or antibiotics required for bacterial overgrowth = 2, chronic pancreatitis or episodes of pseudo-obstruction = 3] + [DLCO ≥80% and FVC ≥80% and no fibrosis on CT and sPAP <35mmHg = 0, DLCO 70-79% or FVC 70-79% or fibrosis on CT or sPAP 35-49 mmHg = 1, DLCO 50-69% or FVC 50-69% or sPAP 50-64 mmHg = 2, DLCO <50% or FVC <50% or sPAP ≥65 mmHg = 3, lung transplant required = 4] + [LVEF ≥50% = 0, LVEF 45-49% =1, arrhythmia not requiring treatment or LVEF 40-44% =2, arrhythmia requiring treatment or LVEF 30-40% = 3, LVEF <30% = 4] + [no history of renal crisis = 0, history of renal crisis and creatinine <132 µmol/L = 1, history of renal crisis and creatinine 132-211 µmol/L = 2, history of renal crisis and creatinine 212-440 µmol/L = 3, history of renal crisis and creatinine ≥440 µmol/L or dialysis required = 4]. The joint component of the Medsger DSS was not included due to the absence of finger-to-palm distance in our data collection. Patients were defined as having severe GI involvement if the corresponding score on the Medsger DSS was 1 or higher. Using our laboratory reference ranges, and according to the literature, micronutrient deficiencies were defined as follows: thiamine pyrophosphate (thiamine) ≤71 nmol/L, pyridoxal phosphate (pyridoxine) ≤30 nmol/L, Lascorbic acid (vitC) ≤4 mg/L [23], Se ≤70 µg/L [24], zinc ≤10.78 µmol/L, B12 ≤197 pg/mL, folate ≤5.38 ng/mL. 2.5 Literature review We reviewed the literature for reports of nutritional status and micronutrient testing in SSc patients. We searched MEDLINE via PubMed for all articles published in English using the following keywords or MeSH terms: “systemic scleroderma”, “malnutrition”, “ascorbic acid”, “selenium”, “thiamine”, “folic acid”, and “vitamin B12”. We reviewed bibliographies of articles retrieved for other relevant articles. Only reports with available clinical and biological data were included. 2.6 Statistical analyses Categorical variables were summarized as count and percentage and numerical variables as me dian ± standard deviation. Comparisons were made using Fisher’s exact test or the χ2–test for proportions, depending on the sample size; and the Wilcoxon signed rank test for continuous variables. Univariate analyses testing the association between Se or vitC deficiency and other variables were realized with logistic regression models. Given the sample size, multivariate analysis and survival modeling could not be performed. Analyses were conducted using the STATA 13 software, with 𝛼 risk set to 5%.
ACCEPTED MANUSCRIPT 2.7 Ethics The data were collected as part of routine clinical care, in agreements with the principles of the declaration of Helsinki, and so in accordance with the standards applied in France, was registered at the French Data Protection Authority (CNIL, No. 2126058v0).
3 RESULTS 3.1 Patient characteristics
NU
SC
RI
PT
We identified 207 relevant medical records for the 2011-2016 period. Eighty-two of these patients met the selection criteria for study inclusion (Fig. 1). Sixty-two patients (76%) were women, with a median age of 60 years (range 21 to 88) (Table 1). SSc was limited, diffuse or sine scleroderma in 76%, 23% and 1% of cases, respectively, with anti-topoisomerase-I (Scl-70) and centromere antibodies in 32% and 44% of cases, respectively. Seventy-seven patients (94%) had symptomatic gastroesophageal reflux. At the time of micronutrient workup, median SSc duration was 7.4 years (0 to 46); 10 (12%) were on steroids, 8 (10%) on another immunosuppressive drug and 12 (15%) on both. Median summed Medsger DSS was 5.5 (1 to 15). Nutritional blood analyses were performed during a diagnostic workup in 10 cases (12%), including 6 cases of heart failure; because of high nutritional risk or known deficiency in 14 cases (17%); and as a routine workup in 57 cases (71%). Sixty patients (73%) were deficient for at least one micronutrient, with on average 1.7 deficiencies per patient. Deficiencies were distributed as follows: pyridoxine, 37 patients (62%); zinc, 15 patients (48%); Se, 19 patients (35%); vitC, 22 patients (31%); folate, 15 patients (23%) and thiamine, 4 patients (6%).
AC C
EP T
ED
MA
3.2 Malnutrition Malnutrition was present in 14/82 patients (17%, see Table 2). Malnourished patients were significantly older (median value 67 vs. 58 years, P = 0.04) than non-malnourished patients. Median BMI was 16.9 in malnourished patients, compared to 23.9 in others (P = 0.01). Significant weight loss was documented in only 3 cases. Gender, disease duration, mRSS, cardiac involvement, ILD, PAH, GI tract involvement, auto-antibodies subtype and baseline SSc treatment were similar between the two groups. Malnutrition was significantly associated with low albumin (35 vs. 39 g/L, P = 0.02), hemoglobin (10.6 vs. 12.9 g/dL, P < 0.0001) and VitC (3.6 vs. 10.6 mg/L, P = 0.003), but not with other micronutrient deficiencies. In the 69 patients that were not malnourished, 29 (42%) had a deficiency in VitC, Se or thiamine, and 47 (68%) had at least one micronutrient deficiency. The summed Medsger DSS was significantly higher in malnourished patients (7.5 vs. 5, P = 0.003). Using MUST scores, we found 12 patients (15%) at high risk for malnutrition, and 5 (6%) at medium risk. Patients at high risk for malnutrition had significantly lower BMI, hemoglobin and vitC levels, and higher summed Medsger DSS than low and medium risk patients. Additionally, ILD was significantly more frequent in high risk patients (10 [91%] vs. 40 [59%], P = 0.05). No association was found with the age or serum albumin level. 3.3 Vitamin C deficiency In the 22/70 deficient patients (31%), median vitC levels were 2.4 mg/dL, compared to 12.9 mg/dL in the others (Table 3). VitC deficient patients were significantly older (median value 70 vs. 58 years, P = 0.03), thinner (BMI 19.2 vs. 24.7, P = 0.02), and more frequently malnourished (8 [67%] vs. 3 [43%], P = 0.003). Sex, SSc duration and auto-antibodies were similar in the two groups. Deficient patients had a significantly higher mRSS (11 vs. 4, P = 0.05) and there was a trend favoring the diffuse subtype (7 [32%] vs. 8 [17%], P = 0.07). Proton pump inhibitor (PPI) use was more frequent in deficient patients (18 [82%] vs. 26 [54%], P = 0.03). PAH was present in 6 deficient (29%) vs. none in non-deficient patients (P = 0.0006). Esophagitis or Barrett’s mucosa were more common in the deficient group (10 [45%] vs. 7 [15%], P = 0.02). VitC deficiency was significantly associated with low albumin (35 vs. 40, P = 0.005), hemoglobin (11.6 vs. 13.4, P < 0.0001) and high CRP levels (5.8 vs. 2.2, P = 0.02). The summed Medsger DSS was significantly higher in vitC deficient patients (8 vs. 5, P < 0.0001). 3.4 Selenium deficiency In the 19/55 deficient patients (35%), median Se levels were 61 µg/L, compared to 82 µg/L in others. Cardiac involvement was significantly more frequent in Se deficient patients (6 [32%] vs. 3 [9%], P = 0.05). Conversely, patients with LVEF ≤50% had significantly lower Se levels than those with LVEF >50% (53.4 ± 9.4 vs. 61.8 ± 9.1 µg/L, p<0.05) (Fig. 2). In the six deficient patients diagnosed with myocardial involvement, dilated cardiomyopathy was observed in three and LEVF≤50% in three. Four patients were symptomatic. Concerning etiology of the
ACCEPTED MANUSCRIPT cardiomyopathy, one patient had coronary heart disease, and one patient had excessive alcohol consumption. SSc and Se deficiency were the only causal factors in the four other patients. Age, sex, nutritional status, SSc duration, SSc cutaneous subtype, auto-antibodies, treatment, laboratory tests, organ involvement and summed Medsger DSS were similar in deficient and non-deficient patients. 3.5 Miscellaneous Thirty-seven patients (over 60) had pyridoxine deficiency (62%). Among them, seven had PAH (19%), whereas there was no PAH in the non-deficient group (P = 0.003). There were no other significant associations between micronutrient status and SSc phenotype (data not shown).
RI
PT
3.6 Follow-up Median follow-up was 5.6 months (0 to 70), 7/82 (9%) patients died during follow-up (supplementary table), among them, two were malnourished. Four patients were deficient in vitC; one patient was deficient for Se and had not been tested for vitC; one patient had been deficient in both micronutrients in the two months before the full micronutrient workup but was no longer deficient; one patient had no deficiency in vitC or Se. The related cause s of death were as follow: pneumonia (n=2), scleroderma renal crisis (n=2), fatal ILD progression (n=1), sepsis (n=1), coma of unknown origin (n=1).
AC C
EP T
ED
MA
NU
SC
4 DISCUSSION This study is to the best of our knowledge the largest series describing micronutrient deficiencies in 82 consecutive and unselected SSc patients. We found that micronutrient deficiency was extremely frequent and occurred independently of clinical and biological markers of malnutrition. Moreover, micronutrient deficiency was associated with organ involvement: vitC deficient patients were more likely to have diffuse cutaneous disease and PAH, while Se deficient patients were more likely to have cardiac involvement. Lastly, SSc severity was positively associated with malnutrition and vitC deficiency. Literature data on nutritional status in SSc is scarce and is reviewed in Table 4. Mean BMI varies between 23 and 27 and is within the normal range. Depending on the definition used, the malnutrition frequency could range from 5.3% to 55.6% [2]. If based on BMI and weight loss, i.e. MUST score, malnutrition affects between 9.4% and 25% of SSc patients, with no obvious correlation with age or cutaneous subset. In our study, 15% of patients were malnourished (as defined with the MUST score), which is consistent with the literature. We nevertheless found similar results when we used a definition according to French HAS statement . The main discrepancies of this definition compared to the MUST score are (i) the use of different BMI and weight loss thresholds (ii) the use of different thresholds in patients below and above 70 years of age and (iii) the incorporation of albumin, provided CRP levels are low. As recommended by a panel of North American experts [25], most studies define malnutrition with the MUST score. In our data, the associations between malnutrition and SSc phenotype were similar when using MUST scores and the French definition of malnutrition, except for ILD. While there was a trend for a higher frequency of ILD in malnourished patients, the association turned to be significant when using MUST scores, but this could be explained by the small sample size. The panel of experts recommended screening all patients with the following laboratory tests: hemoglobin, serum carotene, serum folate, serum albumin; and if malabsorption is suspected: serum methylmalonic acid, zinc, 25-OH vitamin D level, vitamin K level or prothrombin time [25]. In our study, malnutrition was significantly associated with vitC deficiency, but not with deficiencies in thiamine, Se, Zn, folate, B12 or pyridoxine. Most nonmalnourished patients (68%) were deficient in at least one micronutrient. We therefore believe that screening all patients for micronutrient status could uncover unsuspected deficiencies. Earlier studies showed considerably lower vitC levels in SSc patients compared to controls (Table 5). Thirtyone percent of our patients were deficient in vitC. The mechanism underlying vitC deficiency in SSc remains unclear. Mean dietary intake of antioxidants has been described as normal in SSc patients [6], but the absorption of vitC in eight patients with no evidence of bacterial overgrowth has been shown to be similar to that of healthy controls [26]. In addition, vitC blood levels are known to be correlated with malnutrition, PPI use [27], and GI tract diseases that translate into malabsorption, anemia and inflammatory states. Consistently, we found that vitC deficiency was associated with malnutrition, PPI use, hemoglobin and CRP levels. More interestingly, we show that vitC deficiency is associated with extensive skin disease and PAH. In contrast, a previous study had shown lower mean vitC levels in limited SSc patients [5], but the sample size was small (n=18) and these patients were noticeably younger than ours (44 and 47 vs. 66 years). This might explain the discrepancy with our results. Some basic studies have supported the hypothesis that oxidative stress may promote SSc progression [9]. We therefore suggest that vitC deficiency should
ACCEPTED MANUSCRIPT
NU
SC
RI
PT
be sought and treated in SSc patients that are malnourished, have extensive skin involvement or are diagnosed with PAH. Our results are in agreement with several studies that report significantly lower Se levels in patients with SSc than in control subjects [4,5,28]. In our study, selenium deficiency was significantly associated with cardiac involvement. The relationship between selenium and heart diseases was first illustrated by the description of Keshan disease in the 1970s. Keshan disease is a cardiomyopathy that affects people living in rural areas of a Se -deficient zone in China. It is characterized by multifocal necrosis and replacement fibrosis of the myocardium, resulting in acute or chronic heart failure [29]. The role of Se deficiency in the aetiopathogeny of Keshan disease has been suggested in early reports and confirmed in a recent meta-analysis [30]. Se deficiency has also been associated with dilated cardiomyopathy in patients originating from other countries [31,32], and in patients receiving parenteral alimentation [33,34]. Furthermore, in a case-control study nested in a cohort of 11 000 subjects originating from Finland and followed up for seven years from 1972, initial Se levels ≤45 μg/L were associated with a 2 to 3-fold increase in cardiovascular morbidity and mortality [35]. In a Nigerian study, 14 out of 35 women suffering from peripartum cardiomyopathy had plasma Se levels ≤45 µg/L versus none in controls [36]. The efficacy of Se supplementation has only been tested at a population level in Keshan disease endemic areas. The incidence of this disease and of a latent type of this disease was significantly lower in the 1.05 million people supplemented with Se than in the 0.6 million controls [37]. Our findings indicate that six of the nine patients suffering from myocardial involvement tested for Se had low levels. Ultimately, Se deficiency could be suspected to have participated in the cardiomyopathy onset of three patients. Given our patient selection method, based on a historical micronutrient workup, these numbers probably underestimate the real frequency of Se deficiency in SSc and in SSc with myocardial involvement. We therefore suggest testing for Se status in all SSc patients with cardiac involvement. Furthermore, we believe that targeted Se supplementation could be beneficial to patients with cardiac involvement, using preferably sublingual administration to avoid a potential malabsorption.
EP T
ED
MA
In conclusion, we report a high prevalence of micronutrient deficiency in SSc patients independently of clinical and biological markers of malnutrition. Our findings highlight that vitC deficient patients were more likely to have diffuse cutaneous disease and PAH. Interestingly, the present series also shows that Se deficient patients were more likely to have cardiac involvement. Our findings thus underscore that micronutrient deficiency may play a role in the pathophysiology of this chronic progressive disease. We suggest testing vitC, Se and other micronutrient levels routinely and in patients with diffuse skin involvement, cardiac involvement or PAH. Detected deficient patients would be able to benefit from dietary advice (as much as they can follow it) and from supplementation with targeted micronutrients and adequate dosage. The results should be confirmed in larger cohorts and targeted supplementation efficacy and security need to be prospectively evaluated in this population.
AC C
CONTRIBUTORS RD and GP designed the study. Analyses were conducted by BC, ML, CCF and RD. Patients and clinical data for the study were collected by RD, GM, LA, PA, DA, LA, LS, GP. The article has been drafted by RD, ML and GP. All authors critically read, provided input and approved the final version of the article. DECLARATION OF INTEREST The authors declare that there is no conflict of interest regarding the publication of this paper. FUNDING This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. REFERENCES [1] [2] [3]
Varga J, Abraham D. Systemic sclerosis: a prototypic multisystem fibrotic disorder. The Journal of Clinical Investigation 2007;117:557. Codullo V, Cereda E, Crepaldi G, Cappello S, Montecucco C, Caccialanza R, et al. Disease -related malnutrition in systemic sclerosis: evidences and implications. Clin Exp Rheumatol 2015;33:S190-194. Harrison E, Herrick AL, McLaughlin JT, Lal S. Malnutrition in systemic sclerosis. Rheumatology (Oxford, England) 2012;51:1747.
ACCEPTED MANUSCRIPT
[12] [13] [14]
[15] [16] [17] [18]
[19]
[20] [21]
[22] [23] [24] [25]
[26] [27] [28] [29] [30]
PT
[11]
RI
[10]
SC
[9]
NU
[8]
MA
[7]
ED
[6]
EP T
[5]
Lundberg AC, Akesson A, Akesson B. Dietary intake and nutritional status in patients with systemic sclerosis. Ann Rheum Dis 1992;51:1143–8. Herrick AL, Rieley F, Schofield D, Hollis S, Braganza JM, Jayson MI. Micronutrient antioxidant status in patients with primary Raynaud’s phenomenon and systemic sclerosis. J Rheumatol 1994;21:1477–83. Herrick AL, Worthington H, Rieley F, Clarke D, Schofield D, Braganza JM, et al. Dietary intake of micronutrient antioxidants in relation to blood levels in patients with systemic sclerosis. J Rheumatol 1996;23:650–3. Mehta PA, Dubrey SW. High output heart failure. QJM: An International Journal of Medicine 2009;102:235– 241. Elhai M, Meune C, Boubaya M, Avouac J, Hachulla E, Balbir-Gurman A, et al. Mapping and predicting mortality from systemic sclerosis. Annals of the Rheumatic Diseases 2017;76:1897. Wan Y-N, Yan J-W, Peng W-J, Zhang J-Q, Xiao C-C, Wang B-X, et al. Micronutrients, their potential effect on patients with systemic sclerosis. Mod Rheumatol 2014;24:709–14. doi:10.3109/14397595.2013.844383. Codullo V, Cereda E, Klersy C, Cavazzana I, Alpini C, Bonardi C, et al. Serum prealbumin is an independent predictor of mortality in systemic sclerosis outpatients. Rheumatology (Oxford, England) 2016;55:315. Hoogen F, Khanna D, Fransen J, Johnson SR, Baron M, Tyndall A, et al. 2013 classification criteria for systemic sclerosis: an American College of Rheumatology/European League against Rheumatism collaborative initiative. Arthritis & Rheumatology 2013;65:2737–2747. N.d. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA, et al. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202–5. Clements PJ, Lachenbruch PA, Seibold JR, Zee B, Steen VD, Brennan P, et al. Skin thickness score in systemic sclerosis: an assessment of interobserver variability in 3 independent studies. Journal of Rheumatology 1993;20:1892–1896. Miller MR, Crapo R, Hankinson J, Brusasco V, Burgos F, Casaburi R, et al. General considerations for lung function testing. Eur Respir J 2005;26:153–61. doi:10.1183/09031936.05.00034505. N.d. N.d. Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M, et al. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the’malnutrition universal scree ning tool’('MUST’) for adults. The British Journal of Nutrition 2004;92:799. Baron M, Hudson M, Steele R, Canadian Scleroderma Research Group. Malnutrition is common in systemic sclerosis: results from the Canadian scleroderma research group database. J Rheumatol 2009;36:2737–43. doi:10.3899/jrheum.090694. Cereda E, Codullo V, Klersy C, Breda S, Crippa A, Rava ML, et al. Disease -related nutritional risk and mortality in systemic sclerosis. Clinical Nutrition (Edinburgh, Scotland) 2014;33:558. Valentini G, Della R, Bombardieri S, Bencivelli W, Silman A, D’Angelo S, et al. European multicentre study to define disease activity criteria for systemic sclerosis.* II. Identification of disease activity variables and development of preliminary activity indexes. Annals of the Rheumatic Diseases 2001;60:592. Medsger Jr TA, Bombardieri S, Czirjak L, Scorza R, Della Rossa A, Bencivelli W. Assessment of disease severity and prognosis. Clinical and Experimental Rheumatology 2003;21:S42. Jacob RA, Skala JH, Omaye ST. Biochemical indices of human vitamin C status. The American Journal of Clinical Nutrition 1987;46:818. Institute of Medicine (US) Panel on Dietary Antioxidants and Related Compounds. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington (DC): National Academies Press (US); 2000. Baron M, Bernier P, Côté LF, Delegge MH, Falovitch G, Friedman G, et al. Screening and management for malnutrition and related gastro-intestinal disorders in systemic sclerosis: recommendations of a North American expert panel. Clin Exp Rheumatol 2010;28:S42–6. Teh LS, Johns CW, Shaffer JL, Booth EJ, Aarons L, Bennett RJ, et al. Ascorbic acid absorption in patients with systemic sclerosis. J Rheumatol 1997;24:2353–7. Heidelbaugh JJ. Proton pump inhibitors and risk of vitamin and mineral deficiency: evidence and clinical implications. Therapeutic Advances in Drug Safety 2013;4:125–133. Tikly M, Channa K, Theodorou P, Gulumian M. Lipid peroxidation and trace elements in systemic sclerosis. Clinical Rheumatology 2006;25:320. Li GS, Wang F, Kang D, Li C. Keshan disease: an endemic cardiomyopathy in China. Hum Pathol 1985;16:602–9. Zhou H, Wang T, Li Q, Li D. Prevention of Keshan Disease by Selenium Supplementation: a Systematic Review and Meta-analysis. Biol Trace Elem Res 2018. doi:10.1007/s12011-018-1302-5.
AC C
[4]
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
[31] Collipp PJ, Chen SY. Cardiomyopathy and selenium deficiency in a two-year-old girl. N Engl J Med 1981;304:1304–5. doi:10.1056/NEJM198105213042120. [32] Chou HT, Yang HL, Tsou SS, Ho RK, Pai PY, Hsu HB. Status of trace elements in patients with idiopathic dilated cardiomyopathy in central Taiwan. Zhonghua Yi Xue Za Zhi (Taipei) 1998;61:193–8. [33] Fleming CR, Lie JT, McCall JT, O’Brien JF, Baillie EE, Thistle JL. Selenium deficiency and fatal cardiomyopathy in a patient on home parenteral nutrition. Gastroenterology 1982;83:689–93. [34] Johnson RA, Baker SS, Fallon JT, Maynard EP, Ruskin JN, Wen Z, et al. An occidental case of cardiomyopathy and selenium deficiency. N Engl J Med 1981;304:1210–2. doi:10.1056/NEJM198105143042005. [35] Salonen JT, Alfthan G, Huttunen JK, Pikkarainen J, Puska P. Association between cardiovascular death and myocardial infarction and serum selenium in a matched-pair longitudinal study. Lancet 1982;2:175–9. [36] Cénac A, Simonoff M, Moretto P, Djibo A. A low plasma selenium is a risk factor for peripartum cardiomyopathy. A comparative study in Sahelian Africa. Int J Cardiol 1992;36:57–9. [37] Cheng YY, Qian PC. The effect of selenium-fortified table salt in the prevention of Keshan disease on a population of 1.05 million. Biomed Environ Sci 1990;3:422–8. [38] \AAkesson A, \AAkesson B, Gustafson T, Wollheim F. Gastrointestinal function in patients with progressive systemic sclerosis. Clinical Rheumatology 1985;4:441–448. [39] Akesson A, Wollheim FA. Organ manifestations in 100 patients with progressive systemic sclerosis: a comparison between the CREST syndrome and diffuse scleroderma. British Journal of Rheumatology 1989;28:281. [40] Marasini B, Casari S, Bestetti A, Maioli C, Cugno M, Zeni S, et al. Homocysteine concentration in primary and systemic sclerosis associated Raynaud’s phenomenon. The Journal of Rheumatology 2000;27:2621–2623. [41] Caramaschi P, Volpe A, Canestrini S, Bambara LM, Faccini G, Carletto A, et al. Correlation between homocysteine plasma levels and nailfold videocapillaroscopic patterns in systemic sclerosis. Clinical Rheumatology 2007;26:902. [42] Krause L, Becker MO, Brueckner CS, Bellinghausen CJ, Becker C, Schneider U, et al. Nutritional status as marker for disease activity and severity predicting mortality in patients with systemic sclerosis. Annals of the Rheumatic Diseases 2010;69:1951. [43] Murtaugh MA, Frech TM. Nutritional status and gastrointestinal symptoms in systemic sclerosis patients. Clinical Nutrition (Edinburgh, Scotland) 2013;32:130. [44] Caporali R, Caccialanza R, Bonino C, Klersy C, Cereda E, Xoxi B, et al. Disease -related malnutrition in outpatients with systemic sclerosis. Clinical Nutrition (Edinburgh, Scotland) 2012;31:666. [45] Rosato E, Gigante A, Gasperini ML, Molinaro I, Di Lazzaro GG, Afeltra A, et al. Nutritional status measured by BMI is impaired and correlates with left ventricular mass in patients with systemic sclerosi s. Nutrition (Burbank, Los Angeles County, Calif) 2014;30:204. [46] Hughes M, Cooper GJ, Wilkinson J, New P, Guy JM, Herrick AL. Abnormalities of selenium but not of copper homeostasis may drive tissue fibrosis in patients with systemic sclerosis. Rheumatology (Oxford, England) 2015;54:747.
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Figure 1 – Flow chart. Except for the 6 patients excluded from the final analysis, thoroughness of the data was good, with 13% missing values on average
ACCEPTED MANUSCRIPT Table 1 – Study population characteristics. General data (n = 82)
Treatment
Age (y), m (r) Men BMI, m (r) Wei ght l oss Ma l nutrition
60 (21 - 88) 20 (24) 23 (15 - 50) 3 (4) 14 (17)
Scleroderma
PPI Va s odilatator Steroi ds Immunosuppressant
53 (65) 34 (46) 22 (28) 20 (25)
Lab tests 7.4 (0 - 46)
AC C
Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
RI
19 (23) 62 (76) 1 (1)
NU
ED
MA
50 (63) 9 (17) 26 (48)
15 (19) 7 (11) 6 (7) 26 (32) 5 (6) 3 (4) 6 (7)
SC
5 (0 - 40) 40 (16 - 60) 70 (100) 45 (55)
7 (9) 15 (33)
Context Sys tematic Pretra nsplantation Nutri ti onal risk or deficiency Symptoms Usual bloods Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µg/L), m (r) Micronutrients Vi ta min C mg/L, m (r) defi ciency Thi a min nmol /L, m (r) defi ciency Sel enium µg/L, m (r) defi ciency Zi nc µmol /L, m (r) defi ciency Fol ate ng/mL, m (r) defi ciency
PT
26 (32) 36 (44) 82 (100) 10 (12)
EP T
Di s ease duration (y), m (r) Auto - antibody Scl 70 Centromere ANA other Cutaneous form Di ffuse Li mi ted Si ne scleroderma Skin involvement Rodnan, m (r) Interincisive distance (mm), m (r) Ra yna ud Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement
17 (21) 7 (9) 7 (9) 14 (17) 2 (2) 4 (5) 5.5 (1 - 15)
54 (67) 3 (4) 14 (17) 10 (12) 68 (29 - 280) 13 (16) 2.7 (0 - 164) 12 (1 - 64) 38 (16 - 49) 13 (8 - 18) 4.8 (1.7 - 21) 1.4 (0.5 - 2.8) 85.5 (7 - 1661)
8.9 (1 - 26) 22 (31) 109 (53 - 568) 4 (6) 76 (40 - 114) 19 (35) 11 (8.1 - 17) 15 (48) 7.8 (1.6 - 40) 15 (23)
B12 pg/mL, m (r) defi ciency Pyri doxi ne nmol /L, m (r) defi ciency
396 (166 - 1059) 1 (1) 21 (4.5 - 137) 37 (62)
Results are presented as n (%) unless stated otherwise. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, ESR: erythrocyte sedimentation rate, other: other auto-antibodies included anti-RNA polymerase 3, anti-PmScl, anti-Ku, anti-SSA, anti-SSB, LFT: liver function tests, LVEF: left ventricular ejection fraction, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT Table 2 – Malnutrition and SSc phenotype. Malnutrition Patients (n=82) Age (y), m (r) Ma l e sex BMI ≤ 22.9 Wei ght l oss
No malnutrition
OR
95% CI
p
n = 14
n = 68
67 (31 - 88) 2 (14) 9 (90) 3 (33)
58 (21 - 82) 18 (26) 17 (42) 0
1.05 0.46 12.7
[1.00 - 1.10] [0.09 – 2.27] [1.47 – 109.89]
0.04 0.50 0.01 0.21
6.4 (0 - 38)
8.2 (0 - 46)
0.99
[0.93 - 1.05]
0.42
6 (43) 6 (43)
30 (45) 20 (30)
0.93 1.76
[0.29 - 2.96] [0.54 - 5.74]
0.90 0.36
4 (29) 9 (69) 35 (19 - 45) 7 (50)
15 (22) 51 (75) 40 (16 - 60) 38 (56)
0.77 0.75 0.93 0.79
[0.21 - 2.79] [0.20 – 2.75] [0.85 - 1.03] [0.25 - 2.50]
0.74 0.73 0.33 0.69
11 (85) 1 (20) 3 (60)
39 (59) 8 (17) 23 (47)
3.81 1.25 1.70
[0.78 – 18.57] [0.12 – 12.71] [0.26 – 11.06]
0.12 1.00 0.66
3 (21) 5 (56)
4 (6) 10 (27)
4.36 3.37
[0.86 - 22.22] [0.75 – 15.15]
0.09 0.13
12 (18) 5 (9) 6 (9) 23 (34) 3 (4) 2 (3) 6 (9)
1.53 2.45
[0.36 – 6.47] [0.40 – 14.85]
0.53 3.61 2.50
[0.14 – 2.10] [0.54 – 23.96] [0.21 – 29.65]
0.69 0.30 0.58 0.53 0.20 0.44 0.58
5 (36) 2 (14) 2 (14) 4 (29) 0 0 7.5 (3 - 15)
12 (18) 5 (7) 5 (7) 10 (15) 2 (3) 4 (6) 5 (1 - 13)
2.88 2.10 2.10 3.22
[0.81 - 10.25] [0.36 – 12.11] [0.36– 12.11] [0.89 – 11.62]
1.44
[1.14 - 1.83]
0.13 0.34 0.34 0.12 1.00 1.00 0.003
12 (86) 5 (36) 4 (29) 2 (14)
41 (60) 29 (48) 18 (28) 18 (28)
3.95 0.59 1.04 0.44
[0.82 – 19.06] [0.18 - 1.98] [0.29 - 3.76] [0.09 – 2.14]
0.12 0.39 1.00 0.50
83 (29 - 280) 2 (14) 4.6 (0.5 - 28.4) 28 (1 - 64) 35 (16 - 47) 10.6 (9.2 - 14) 4.8 (2.8 - 9.0) 1.0 (0.6 - 2.4) 94 (7 - 389)
68 (31 - 139) 11 (16) 2.5 (0 - 164) 12 (1 - 54) 39 (17 - 49) 12.9 (8.2 - 17.7) 4.6 (1.7 - 21.1) 1.4 (0.5 - 2.8) 79 (9 - 1661)
1.02 0.86 0.99 1.05 0.90 0.40 1.02 0.54 1.00
[1.00 - 1.03] [0.17 – 4.41] [0.96 - 1.03] [0.99 - 1.11] [0.82 - 0.99] [0.24 - 0.68] [0.83 - 1.25] [0.18 - 1.63] [0.99 - 1.00]
0.32 1.00 0.25 0.32 0.02 < 0.0001 0.44 0.20 0.26
8 (73) 1 (10) 3 (33) 2 (29)
14 (24) 3 (6) 16 (35) 13 (54)
8.57 1.85 0.94 0.34
[2.00 – 36.77] [0.17 – 19.85] [0.21 - 4.26] [0.05 – 2.102]
0.003 0.51 1.00 0.39
AC C
Treatment PPI Ca l ci um channel blocker Steroi ds Immunosuppressant drug
Lab tests
Usual bloods Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µ g/L), m (r) Micronutrient deficiency Vi ta min C Thi a min Sel enium Zi nc
RI SC
NU
ED
3 (21) 2 (14) 1 (7) 0
MA
3 (21) 2 (20)
EP T
Di s ease duration (y), m (r) Auto-antibody Anti -centromere Anti -Scl70 Skin involvement Di ffuse form Rodnan score ≤ 14 Interincisive distance, m (r) Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
PT
Scleroderma
ACCEPTED MANUSCRIPT Fol ate B12 Pyri doxi ne
3 (30)
12 (22) 1 (2) 30 (59)
[0.34 - 6.70]
2.45
[0.46 – 12.98]
AC C
EP T
ED
MA
NU
SC
RI
PT
7 (78)
1.50
0.69 1.00 0.46
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Results are presented as n (%) unless stated otherwise. P-values are obtained using the -square or Fisher’s exact test for categorical variables and the Mann–Whitney test for continuous variables. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, DSS : disease severity scales, ESR: erythrocyte sedimentation rate, LFT: liver function tests, LVEF: left ventricular ejection fraction, mRSS: modified Rodnan skin score, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT Table 3 – Vitamin C deficiency and SSc phenotype. Deficiency Patients (n=70) Age (y), m (r) Ma l e sex BMI ≤ 22.9 Wei ght l oss Ma l nutrition
No deficiency
OR
95% CI
p
n = 22
n = 48
70 (21 - 88) 5 (23) 10 (77) 1 (11) 8 (67)
58 (28 - 84) 13 (27) 12 (38) 1 (17) 3 (43)
1.04 0.79 5.56 0.62 8.57
[0.99 - 1.08] [0.24 - 2.58] [1.27 – 24.29] [0.03 - 12.41 [2.00 – 36.77]
0.03 0.70 0.02 1.00 0.003
8.2 (0 - 44)
5.8 (0 - 46)
1.02
[0.98 - 1.07]
0.64
6 (29) 12 (57)
16 (33) 19 (40)
0.8 2.04
[0.26 - 2.45] [0.72 - 5.76]
0.70 0.18
7 (32) 13 (59) 33.5 (16 - 50) 15 (68)
8 (17) 39 (81) 40 (30 - 60) 24 (50)
0.35 0.33 0.91 2.14
[0.11 - 1.11] [0.11 – 1.02] [0.82 - 0.99] [0.74 - 6.19]
0.07 0.05 0.17 0.16
14 (67) 5 (36) 9 (69)
29 (63) 4 (12) 15 (43)
1.17 4.17 3
[0.40 - 3.48] [0.92 – 18.88] [0.77 – 11.63]
0.77 0.10 0.10
6 (27) 7 (50)
0 6 (24)
3.17
[0.79 – 12.75]
0.0006 0.16
8 (17) 2 (5) 4 (8) 19 (40) 4 (8) 0 3 (6)
1.64 2.33 1.10 0.45 0.52
[0.46 - 5.90] [0.30 – 18.15] [0.19 - 6.51] [0.14 - 1.42] [0.06 – 4.98]
1.50
[0.23 - 9.69]
0.50 0.58 1 0.17 1 0.10 0.65
10 (45) 3 (14) 3 (14) 6 (27) 1 (5) 2 (9) 8 (3 - 15)
7 (15) 4 (8) 4 (8) 7 (15) 1 (2) 2 (4) 5 (1 - 10)
4.05 1.74 1.74 1.88 2.24 2.30 1.68
[1.26 - 13.04] [0.35 - 8.52] [0.35 - 8.52] [0.56 – 6.27] [0.13 – 37.52] [0.30 – 17.50] [1.27 - 2.21]
0.01 0.67 0.67 0.34 0.53 0.59 < 0.0001
18 (82) 8 (40) 8 (38) 4 (19)
26 (54) 23 (55) 10 (22) 11 (24)
3.81 0.55 2.22 0.75
[1.12 – 12.94] [0.19 - 1.62] [0.72 - 6.83] [0.21 – 2.70]
0.03 0.28 0.16 0.76
64 (30 - 280) 5 (23) 5.8 (0 - 164) 22 (2 - 64) 35 (25 - 47) 11.6 (8.2 - 13.3) 4.7 (2.3 - 9.1) 1.3 (0.2 - 2.2) 81.5 (7 - 389)
68 (40 - 139) 7 (15) 2.2 (0 - 112) 11 (1 - 54) 40 (16 - 49) 13.4 (10.2 - 17.7) 4.4 (1.7 - 21.1) 1.4 (0.5 - 2.8) 88 (9 - 1661)
1.01 1.72 1.02 1.05 0.91 0.38 1.02 0.41 1.00
[0.99 - 1.02] [0.48 - 6.19] [1.00 - 1.05] [1.00 - 1.10] [0.83 – 1.00] [0.23 - 0.64] [0.86 - 1.21] [0.16 - 1.07] [0.99 - 1.00]
0.98 0.50 0.02 0.08 0.005 < 0.0001 0.33 0.09 0.95
AC C
Treatment
PPI Ca l ci um channel blocker Steroi ds Immunosuppressant drug
Lab tests
Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µ g/L), m (r)
RI
SC
NU MA
ED
6 (27) 2 (12) 2 (9) 5 (23) 1 (5) 2 (9) 2 (9)
EP T
Di s ease duration (y), m (r) Auto-antibody Anti -centromere Anti -Scl70 Skin involvement Di ffuse form Rodnan score ≤ 14 Interincisive distance, m (r) Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
PT
Scleroderma
Results are presented as n (%) unless stated otherwise. P-values are obtained using the Chi-squared or Fisher’s exact test for categorical variables and the Mann–Whitney test for continuous variables. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, DSS : disease severity scales, ESR: erythrocyte sedimentation rate, LFT: liver function tests, LVEF:
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
left ventricular ejection fraction, mRSS: modified Rodnan skin score, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Figure 2 – Plasma Selenium concentration in patients with low and normal left ventricular ejection fraction LVEF: left ventricular ejection fraction; *p<0.05
ACCEPTED MANUSCRIPT Table 4 – Literature review.
n Age (y) Diffuse SSc (%) BMI Malnutrition Thiamine Folate B12 Vitamin C
Selenium
Akesson, 1985 [38]
Akesson, Lundberg, 1989 1992 [39] [4]
Caramaschi, Herrick, Herrick, Herrick, Marasini, 2006 1994 [5] 1994 [5] 1996 [6] 2000 [40] [41]
24 41 38
100 44 37
9 m 44 0
30 48 57 22
9 m 47 100
12 m 47 58
30 58 33
Definition % nmol/L deficiency (%) ng/mL
114 (N >55.2)
β
127
5.7
361 (N >149)
385
deficiency (%) mg/L
5.5
α
m 5.3
70.0 (N >53.7)
deficiency (%)
67.9
T P
α
E C
m 9.7
D E
deficiency (%) μg/L
α
m 82.1
A M
α
m 75.0
α
396
Krause, 2010 [42]
Murtaugh Caporali, 2012 Rosato, , [44] and Cereda, 2014 [45] 2013 [43] 2014 [20]
586 55 39 26
124 54 47 23
24 54 13 27
160 62 19 25
94 53 43 23
MUST ≥2 17
BMI <19 14
MUST ≥2 25
MUST ≥2 9.4
BMI <20 19
T P
I R
C S
U N
β
deficiency (%) pg/mL
60 55 37
Baron, 2009 [19]
4.9 (N >3.0)
α
m 6.0
α
m 74.2
α
Results are presented as mean unless stated otherwise. m: median; MUST: MUST score; α statistically significant difference with control group; β whole blood.
C A
ACCEPTED MANUSCRIPT Supplementary Table – Demographic and clinical characteristics of the seven patients that died during study follow-up. Patient N°
1
2
3
4
5
6
Age (y)
52
77
43
70
21
66
Sex
male
female
female
male
female
male
Malnutrition
no
no
no
no
no
yes
Follow-up duration (y)
0.1
4.2
0.3
0.7
0.2
0.3
Disease duration (y)
1.1
0.6
13
15
1.6
2.9
Auto-antibody
scl70
centromere
scl70
centromere
scl70
ANA
Cutaneous form
diffuse
limited
diffuse
diffuse
diffuse
-
mRSS
27
7
18
25
27
-
ILD
yes
no
yes
yes
no
yes
PAH
no
no
no
no
no
no
Heart involvement
no
no
no
no
yes
no
Renal crisis
yes
no
no
yes
no
no
GI tract involvement
yes
yes
yes
yes
yes
yes
Summed Medsger DSS
4
8
8
11
6
Medication
no
no
Comment
-
-
EP T
Lab tests
RI SC
NU MA
ED
Treatment
PT
Scleroderma
9
steroids & MMF steroids & MTX
steroids
-
screened for HSCT
-
steroids, MMF & tacrolimus lung transplant for ILD
malabsorption and infection
-
transplantation malabsorption planning
HSCT planning
weight loss
CRP (mg/L)
29
29
23
6
71
16
Albumin (g/L)
-
26
28
39
26
-
9
12
12
8
11
yes
yes
yes
yes
-
-
no
-
-
yes
Hemoglobin (g/dL)
AC C
Context
Vitamin C deficiency Selenium deficiency
13
formerly deficient
Death Cause
pneumonia
sepsis
acute respiratory failure
SRC
aspiration pneumonia
coma
SSc-related
yes
-
yes
yes
yes
-
ANA: antinuclear antibodies, DSS : disease severity scales, GI: gastro-intestinal, HSCT: hematopoietic stem cell transplantation, mRSS: modified Rodnan skin score, MMF: mycophenolate mofetil, MTX: methotrexate, ILD: interstitial lung disease, PAH: pulmonary artery hypertension, SRC: scleroderma renal crisis
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Highlights Micronutrient deficiency is frequent in SSc and is not associated with malnutrition SSc primary heart disease is associated with selenium deficiency Malnutrition and SSc severity is associated with vitamin C deficiency Routine micronutrient testing and targeted supplementation might be useful in SSc
Figure 1
Figure 2
Romain Dupont, Mélanie Longué, Anne Galinier, Christel Cinq Frais, Cécile Ingueneau, Léonardo Astudillo, Philippe Arlet, Daniel Adoue, Laurent Alric, Grégoire Prévot, Bastien Cabarrou, Laurent Sailler, Grégory Pugnet PII: DOI: Reference:
S1568-9972(18)30211-8 doi:10.1016/j.autrev.2018.05.010 AUTREV 2216
To appear in:
Autoimmunity Reviews
Received date: Accepted date:
18 May 2018 25 May 2018
Please cite this article as: Romain Dupont, Mélanie Longué, Anne Galinier, Christel Cinq Frais, Cécile Ingueneau, Léonardo Astudillo, Philippe Arlet, Daniel Adoue, Laurent Alric, Grégoire Prévot, Bastien Cabarrou, Laurent Sailler, Grégory Pugnet , Impact of Micronutrient Deficiency & Malnutrition in Systemic Sclerosis: Cohort Study and Literature Review. Autrev (2018), doi:10.1016/j.autrev.2018.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Impact of Micronutrient Deficiency & Malnutrition in Systemic Sclerosis: Cohort Study and Literature Review Romain Dupont1, Mélanie Longué 2, Anne Galinier3, Christel Cinq Frais3 , Cécile Ingueneau3, Léonardo Astudillo1, Philippe Arlet1, Daniel Adoue 4, Laurent Alric5,6, Grégoire Prévot7 , Bastien Cabarrou2, Laurent Sailler1,8,9 and Grégory Pugnet1,9 1
Service de Médecine interne, URM, CHU de Toulouse, France Service de Biostatistiques, Institut Universitaire du Cancer de Toulouse – Oncopôle 3 Laboratoire de biochimie et nutrition, Institut Fédératif de Biologie, CHU de Toulouse, France 4 Service de Médecine interne et immunopathologie clinique, CHU de Toulouse, France 5 Service de Médecine interne, pavillon Dieulafoy, CHU de Toulouse, France 6 UMR 152 IRD-Université de Toulouse, France 7 Service de Pneumologie, CHU de Toulouse, France 8 CIC 1436, CHU de Toulouse, France 9 UMR 1027 Inserm-Université de Toulouse, France
RI
PT
2
NU
SC
Corresponding author: Romain Dupont, Service de Médecine Interne, salle Le Tallec, URM, Hôpital Purpan, place du Dr Joseph Baylac, TSA 40031, 31059 Toulouse cedex 9, France; Phone: +33 5 61 77 22 78; Fax: +33 5 61 77 71 24; Email: [email protected]
AC C
EP T
ED
MA
ABSTRACT Objectives: The purpose of our study was to determine the prevalence and risk factors associated with malnutrition, and selenium (Se) and vitamin C (vitC) deficiencies in systemic sclerosis (SSc) patients. Methods: We included adult SSc patients fulfilling the 2013 ACR/EULAR criteria from the Toulouse University Hospital cohort who underwent a micronutrient workup (including vitC, Se or thiamine levels) between 2011 and 2016. Results: 82 patients were included, mostly women (76%), with a median age of 60 years. SSc was limited in 76% of the cases, with Scl-70 and centromere antibodies in 32% and 44%, respectively. Median disease duration was 7.4 years. Cardiac involvement was noticed in 19% and gastrointestinal tract in and 95%; 9% had pulmonary artery hypertension (PAH) and 63% had interstitial lung disease. Overt malnutrition was present in 14 (17%) patients. Micronutrient deficiencies included Se (35%), vitC (31%) and/or thiamine (6%). Malnourished patients had significantly a higher summed Medsger disease severity scales (7.5 vs. 5, P = 0.003), lower hemoglobin (10.6 vs. 12.9 g/dL, P < 0.0001) and vitC levels (3.6 vs. 10.6 mg/L, P = 0.003). Cardiac involvement was significantly associated with Se deficiency (OR 6.2, IC 95%[1.48-32.70], P = 0.05). The factors associated with vitC deficiency were malnutrition (OR 8.57, IC 95%[2.16-43.39], P = 0.003), modified Rodnan skin score ≤ 14 (OR 0.33, IC95[0.11-1], P = 0.05), PAH (27% in deficient vs. none in non-deficient patients, P = 0.0006) and esophagitis or Barrett’s mucosa (OR 4.05, IC95[1.27-13.54], P = 0.02). Conclusions: Se testing should be considered as soon as cardiac involvement is suspected. VitC testing should be considered in malnourished SSc patients, especially if skin involvement is extensive. Keywords: systemic sclerosis, malnutrition, selenium, vitamin C, deficiency Highlights: Micronutrient deficiency is frequent in SSc and is not associated with malnutrition SSc primary heart disease is associated with selenium deficiency Malnutrition and SSc severity is associated with vitamin C deficiency Routine micronutrient testing and targeted supplementation might be useful in SSc
Main text 1 INTRODUCTION
ACCEPTED MANUSCRIPT
NU
SC
RI
PT
Systemic sclerosis (SSc) is a rare multisystem disease characterized by three pathogenetic landmarks: microvascular involvement, activation of the immune system and increase of extracellular matrix deposition in the skin and internal organs [1]. This severe chronic disease is of high nutritional risk, estimated to be around 20% [2]. Nutritional status cannot be inferred from gastrointestinal (GI) tract involvement alone [2]. Many factors could also contribute to nutritional impairment: mood disturbance, functional status and disabilities, oropharyngeal manifestations, pancreatic involvement, heart failure, pulmonary interstitial fibrosis, and inflammation [3]. Malnutrition itself involves to varying degrees a deficiency in energy, protein and micronutrients (vitamins and trace elements), depending on respective intakes, absorptions and losses. SSc patients have been shown to have a lower intake in fruits and vegetables [4] and to be deficient in vitamin C (vitC) and selenium (Se) [4–6]. Whereas micronutrient deficiency is associated with a wide range of pathological conditions, the consequences of micronutrient deficiency in SSc patients have been poorly described. Selenium, vitC and thiamine pyrophosphate deficiencies are known causes of dilated, ischemic and high-output heart failure respectively [7]. This may be of utmost importance given that primary heart disease is the main offender in SSc explaining 30% of SSc deaths, while atherosclerosis is responsible for only 5%–8% of deaths [8]. As one of the physiological functions of Se and vitC is to eliminate free radicals, their deficiency could promote oxidative injury and progression of SSc [9]. The prognostic value of malnutrition or micronutrient deficiencies in SSc is still discussed, while survival seems to be associated with prealbumin levels, but neither BMI nor weight loss [2,10]. The aim of this study was to assess the prevalence and the factors associated with malnutrition, Se and vitamin C deficiencies, in a population of SSc patients, when performed during routine patient follow-up evaluation in a tertiary referral center for SSc.
MA
2 METHODS
EP T
ED
2.1 Study population We conducted a retrospective cross-sectional study in a cohort of all unselected consecutive patients diagnosed with a SSc at the Toulouse University Hospital, a tertiary referral center for SSc. After thorough medical chart review, patients were included if they 1) met the ACR/EULAR 2013 classification criteria [11], 2) were more than 18 years of age 3) between 2011 and 2016 had undergone a micronutrient workup including at least Se, Lascorbic acid (vitamin C) or thiamine pyrophosphate levels.
AC C
2.2 Data collection The following clinical and biological data were collected at baseline and annually, according to standard clinical guidelines [12]: a) age, gender, disease duration (since first non-Raynaud manifestation), weight, weight loss, height, body mass index (BMI); b) SSc sub-type according to Leroy and Medsger [13] and SSc treatment at baseline; b) extension of skin fibrosis as measured by the modified Rodnan skin score (mRSS)[14] and inter-incisive distance; c) past or present Raynaud’s phenomenon, and past or present digital ulcers; d ) cardiac and pulmonary evaluation, including results from pulmonary function tests (forced vital capacity (FVC), diffusing capacity of the lung for carbon monoxide (DLCO)), according to the ATS and European Respiratory Society (ATS/ERS) consensu s standards [15], presence of SSc-related interstitial lung disease (ILD) or pleural effusion on chest X -ray or on high resolution CT scans, results from echocardiography including left ventricular ejection fraction (LVEF) and systolic pulmonary arterial pressure (sPAP, mmHg) measurements, and the results of cardiac MRI when available. When resting sPAP on Doppler echocardiography was superior to 35 mmHg, right heart catheterization was performed. Pulmonary arterial hypertension (PAH) was confirmed as a mean PAP ≥25 mmHg and was considered as SSc-associated pre-capillary PAH when combined with a pulmonary artery wedge pressure ≤ 15 mmHg. Cardiac involvement was considered present if stated by the referent physician based on clinical, echocardiography and/or imaging findings; e) GI tract evaluation if needed, including results of gastroscopy and/or colonoscopy, confirmation of small intesti ne bacterial overgrowth by a hydrogen breath test, presence of digestive pseudo-obstruction on abdominal CT scans, presence of any motility disorder on esophageal manometry or gastric scintigraphy, and confirmation of pancreatic insufficiency by fecal elastase. Gastrointestinal involvement was considered present if suspected on clinical grounds; f) other relevant testing if needed for muscle, joint, renal, or hepatobiliary involvement, or neurologic manifestations; g) laboratory parameters, including hemoglobin level, serum creatinine, C-reactive protein, albumin, serum transaminases, alkaline phosphatase, γ-glutamyl transferase, creatinine phosphokinases, ferritin, status of
ACCEPTED MANUSCRIPT antinuclear, anticentromere and anti-topoisomerase-I autoantibodies, and micronutrient testing. Patient survival status was evaluated during and at the end of the follow-up of the entire cohort.
2.3 Nutritional data Micronutrient testing was performed in the Toulouse University Hospital laboratory using fluorescent liquid chromatography for thiamine pyrophosphate (thiamine), pyridoxal phosphate (pyridoxine) and vitC levels; atomic absorption spectrophotometry for serum Se and serum zinc; and chemiluminescence for serum cobalamin (B12) and serum folate. Nutritional workup for indication was classified in four categories: systematic, related to symptoms or signs (in the context of diagnostic investigations), high nutritional risk (overt clinical malnutrition, significant GI tract involvement, parenteral nutrition, or known micronutrient deficiency), before organ or autologous stem cell transplantation.
AC C
EP T
ED
MA
NU
SC
RI
PT
2.4 Outcomes Malnutrition was defined according to French HAS (National Health Authority) statement as follows [16,17]: for patients younger than 70, BMI ≤17, significant weight loss or (albumin ≤30 g/L and CRP ≤10 mg/L); for patients older than 70, BMI ≤21, significant weight loss or (albumin ≤35 g/L and CRP ≤10 mg/L). Weight loss was defined as significant if exceeding 5% body weight in 1 month or 10% body weight in 6 months. The MUST score was calculated as follows: [BMI > 20.0 = 0, 18.5–20.1 = 1, < 18.5 = 2] +[weight loss in the past 3–6 months: < 5% = 0, 5%–10% = 1, > 10% = 2] [18,19]. We did not incorporate a correction based on disease activity score as proposed by some authors [20,21]. MUST scores of 0, 1 and ≥2 correspond to low, medium and high nutritional risk, respectively. The summed Medsger disease severity scales (DSS) was calculated as follows [22]: [weight loss <5% and hemoglobin ≥12.3 g/dL =0, weight loss 5%-9.9% or hemoglobin 11-12.2 g/dL =1, weight loss 10-14.9% or hemoglobin 9.7-10.9g/dL =2, weight loss 15-19.9% or hemoglobin 8.3-9.6g/dL =3, weight loss ≥20% or hemoglobin <8.3 g/dL =4] + [no Raynaud’s or Raynaud’s not requiring vasodilators = 0; Raynaud’s requiring vasodilators = 1; digital pitting scars or ulcers = 2] + [mRSS 0 = 0, mRSS 1-14 = 1, mRSS 15-29 = 2, mRSS 30-39 = 3, mRSS ≥40 = 4] + [no myositis = 0; myositis = 2] + [no motility disorder = 0, distal esophageal hypoperistalsis =1, achalasia or antibiotics required for bacterial overgrowth = 2, chronic pancreatitis or episodes of pseudo-obstruction = 3] + [DLCO ≥80% and FVC ≥80% and no fibrosis on CT and sPAP <35mmHg = 0, DLCO 70-79% or FVC 70-79% or fibrosis on CT or sPAP 35-49 mmHg = 1, DLCO 50-69% or FVC 50-69% or sPAP 50-64 mmHg = 2, DLCO <50% or FVC <50% or sPAP ≥65 mmHg = 3, lung transplant required = 4] + [LVEF ≥50% = 0, LVEF 45-49% =1, arrhythmia not requiring treatment or LVEF 40-44% =2, arrhythmia requiring treatment or LVEF 30-40% = 3, LVEF <30% = 4] + [no history of renal crisis = 0, history of renal crisis and creatinine <132 µmol/L = 1, history of renal crisis and creatinine 132-211 µmol/L = 2, history of renal crisis and creatinine 212-440 µmol/L = 3, history of renal crisis and creatinine ≥440 µmol/L or dialysis required = 4]. The joint component of the Medsger DSS was not included due to the absence of finger-to-palm distance in our data collection. Patients were defined as having severe GI involvement if the corresponding score on the Medsger DSS was 1 or higher. Using our laboratory reference ranges, and according to the literature, micronutrient deficiencies were defined as follows: thiamine pyrophosphate (thiamine) ≤71 nmol/L, pyridoxal phosphate (pyridoxine) ≤30 nmol/L, Lascorbic acid (vitC) ≤4 mg/L [23], Se ≤70 µg/L [24], zinc ≤10.78 µmol/L, B12 ≤197 pg/mL, folate ≤5.38 ng/mL. 2.5 Literature review We reviewed the literature for reports of nutritional status and micronutrient testing in SSc patients. We searched MEDLINE via PubMed for all articles published in English using the following keywords or MeSH terms: “systemic scleroderma”, “malnutrition”, “ascorbic acid”, “selenium”, “thiamine”, “folic acid”, and “vitamin B12”. We reviewed bibliographies of articles retrieved for other relevant articles. Only reports with available clinical and biological data were included. 2.6 Statistical analyses Categorical variables were summarized as count and percentage and numerical variables as me dian ± standard deviation. Comparisons were made using Fisher’s exact test or the χ2–test for proportions, depending on the sample size; and the Wilcoxon signed rank test for continuous variables. Univariate analyses testing the association between Se or vitC deficiency and other variables were realized with logistic regression models. Given the sample size, multivariate analysis and survival modeling could not be performed. Analyses were conducted using the STATA 13 software, with 𝛼 risk set to 5%.
ACCEPTED MANUSCRIPT 2.7 Ethics The data were collected as part of routine clinical care, in agreements with the principles of the declaration of Helsinki, and so in accordance with the standards applied in France, was registered at the French Data Protection Authority (CNIL, No. 2126058v0).
3 RESULTS 3.1 Patient characteristics
NU
SC
RI
PT
We identified 207 relevant medical records for the 2011-2016 period. Eighty-two of these patients met the selection criteria for study inclusion (Fig. 1). Sixty-two patients (76%) were women, with a median age of 60 years (range 21 to 88) (Table 1). SSc was limited, diffuse or sine scleroderma in 76%, 23% and 1% of cases, respectively, with anti-topoisomerase-I (Scl-70) and centromere antibodies in 32% and 44% of cases, respectively. Seventy-seven patients (94%) had symptomatic gastroesophageal reflux. At the time of micronutrient workup, median SSc duration was 7.4 years (0 to 46); 10 (12%) were on steroids, 8 (10%) on another immunosuppressive drug and 12 (15%) on both. Median summed Medsger DSS was 5.5 (1 to 15). Nutritional blood analyses were performed during a diagnostic workup in 10 cases (12%), including 6 cases of heart failure; because of high nutritional risk or known deficiency in 14 cases (17%); and as a routine workup in 57 cases (71%). Sixty patients (73%) were deficient for at least one micronutrient, with on average 1.7 deficiencies per patient. Deficiencies were distributed as follows: pyridoxine, 37 patients (62%); zinc, 15 patients (48%); Se, 19 patients (35%); vitC, 22 patients (31%); folate, 15 patients (23%) and thiamine, 4 patients (6%).
AC C
EP T
ED
MA
3.2 Malnutrition Malnutrition was present in 14/82 patients (17%, see Table 2). Malnourished patients were significantly older (median value 67 vs. 58 years, P = 0.04) than non-malnourished patients. Median BMI was 16.9 in malnourished patients, compared to 23.9 in others (P = 0.01). Significant weight loss was documented in only 3 cases. Gender, disease duration, mRSS, cardiac involvement, ILD, PAH, GI tract involvement, auto-antibodies subtype and baseline SSc treatment were similar between the two groups. Malnutrition was significantly associated with low albumin (35 vs. 39 g/L, P = 0.02), hemoglobin (10.6 vs. 12.9 g/dL, P < 0.0001) and VitC (3.6 vs. 10.6 mg/L, P = 0.003), but not with other micronutrient deficiencies. In the 69 patients that were not malnourished, 29 (42%) had a deficiency in VitC, Se or thiamine, and 47 (68%) had at least one micronutrient deficiency. The summed Medsger DSS was significantly higher in malnourished patients (7.5 vs. 5, P = 0.003). Using MUST scores, we found 12 patients (15%) at high risk for malnutrition, and 5 (6%) at medium risk. Patients at high risk for malnutrition had significantly lower BMI, hemoglobin and vitC levels, and higher summed Medsger DSS than low and medium risk patients. Additionally, ILD was significantly more frequent in high risk patients (10 [91%] vs. 40 [59%], P = 0.05). No association was found with the age or serum albumin level. 3.3 Vitamin C deficiency In the 22/70 deficient patients (31%), median vitC levels were 2.4 mg/dL, compared to 12.9 mg/dL in the others (Table 3). VitC deficient patients were significantly older (median value 70 vs. 58 years, P = 0.03), thinner (BMI 19.2 vs. 24.7, P = 0.02), and more frequently malnourished (8 [67%] vs. 3 [43%], P = 0.003). Sex, SSc duration and auto-antibodies were similar in the two groups. Deficient patients had a significantly higher mRSS (11 vs. 4, P = 0.05) and there was a trend favoring the diffuse subtype (7 [32%] vs. 8 [17%], P = 0.07). Proton pump inhibitor (PPI) use was more frequent in deficient patients (18 [82%] vs. 26 [54%], P = 0.03). PAH was present in 6 deficient (29%) vs. none in non-deficient patients (P = 0.0006). Esophagitis or Barrett’s mucosa were more common in the deficient group (10 [45%] vs. 7 [15%], P = 0.02). VitC deficiency was significantly associated with low albumin (35 vs. 40, P = 0.005), hemoglobin (11.6 vs. 13.4, P < 0.0001) and high CRP levels (5.8 vs. 2.2, P = 0.02). The summed Medsger DSS was significantly higher in vitC deficient patients (8 vs. 5, P < 0.0001). 3.4 Selenium deficiency In the 19/55 deficient patients (35%), median Se levels were 61 µg/L, compared to 82 µg/L in others. Cardiac involvement was significantly more frequent in Se deficient patients (6 [32%] vs. 3 [9%], P = 0.05). Conversely, patients with LVEF ≤50% had significantly lower Se levels than those with LVEF >50% (53.4 ± 9.4 vs. 61.8 ± 9.1 µg/L, p<0.05) (Fig. 2). In the six deficient patients diagnosed with myocardial involvement, dilated cardiomyopathy was observed in three and LEVF≤50% in three. Four patients were symptomatic. Concerning etiology of the
ACCEPTED MANUSCRIPT cardiomyopathy, one patient had coronary heart disease, and one patient had excessive alcohol consumption. SSc and Se deficiency were the only causal factors in the four other patients. Age, sex, nutritional status, SSc duration, SSc cutaneous subtype, auto-antibodies, treatment, laboratory tests, organ involvement and summed Medsger DSS were similar in deficient and non-deficient patients. 3.5 Miscellaneous Thirty-seven patients (over 60) had pyridoxine deficiency (62%). Among them, seven had PAH (19%), whereas there was no PAH in the non-deficient group (P = 0.003). There were no other significant associations between micronutrient status and SSc phenotype (data not shown).
RI
PT
3.6 Follow-up Median follow-up was 5.6 months (0 to 70), 7/82 (9%) patients died during follow-up (supplementary table), among them, two were malnourished. Four patients were deficient in vitC; one patient was deficient for Se and had not been tested for vitC; one patient had been deficient in both micronutrients in the two months before the full micronutrient workup but was no longer deficient; one patient had no deficiency in vitC or Se. The related cause s of death were as follow: pneumonia (n=2), scleroderma renal crisis (n=2), fatal ILD progression (n=1), sepsis (n=1), coma of unknown origin (n=1).
AC C
EP T
ED
MA
NU
SC
4 DISCUSSION This study is to the best of our knowledge the largest series describing micronutrient deficiencies in 82 consecutive and unselected SSc patients. We found that micronutrient deficiency was extremely frequent and occurred independently of clinical and biological markers of malnutrition. Moreover, micronutrient deficiency was associated with organ involvement: vitC deficient patients were more likely to have diffuse cutaneous disease and PAH, while Se deficient patients were more likely to have cardiac involvement. Lastly, SSc severity was positively associated with malnutrition and vitC deficiency. Literature data on nutritional status in SSc is scarce and is reviewed in Table 4. Mean BMI varies between 23 and 27 and is within the normal range. Depending on the definition used, the malnutrition frequency could range from 5.3% to 55.6% [2]. If based on BMI and weight loss, i.e. MUST score, malnutrition affects between 9.4% and 25% of SSc patients, with no obvious correlation with age or cutaneous subset. In our study, 15% of patients were malnourished (as defined with the MUST score), which is consistent with the literature. We nevertheless found similar results when we used a definition according to French HAS statement . The main discrepancies of this definition compared to the MUST score are (i) the use of different BMI and weight loss thresholds (ii) the use of different thresholds in patients below and above 70 years of age and (iii) the incorporation of albumin, provided CRP levels are low. As recommended by a panel of North American experts [25], most studies define malnutrition with the MUST score. In our data, the associations between malnutrition and SSc phenotype were similar when using MUST scores and the French definition of malnutrition, except for ILD. While there was a trend for a higher frequency of ILD in malnourished patients, the association turned to be significant when using MUST scores, but this could be explained by the small sample size. The panel of experts recommended screening all patients with the following laboratory tests: hemoglobin, serum carotene, serum folate, serum albumin; and if malabsorption is suspected: serum methylmalonic acid, zinc, 25-OH vitamin D level, vitamin K level or prothrombin time [25]. In our study, malnutrition was significantly associated with vitC deficiency, but not with deficiencies in thiamine, Se, Zn, folate, B12 or pyridoxine. Most nonmalnourished patients (68%) were deficient in at least one micronutrient. We therefore believe that screening all patients for micronutrient status could uncover unsuspected deficiencies. Earlier studies showed considerably lower vitC levels in SSc patients compared to controls (Table 5). Thirtyone percent of our patients were deficient in vitC. The mechanism underlying vitC deficiency in SSc remains unclear. Mean dietary intake of antioxidants has been described as normal in SSc patients [6], but the absorption of vitC in eight patients with no evidence of bacterial overgrowth has been shown to be similar to that of healthy controls [26]. In addition, vitC blood levels are known to be correlated with malnutrition, PPI use [27], and GI tract diseases that translate into malabsorption, anemia and inflammatory states. Consistently, we found that vitC deficiency was associated with malnutrition, PPI use, hemoglobin and CRP levels. More interestingly, we show that vitC deficiency is associated with extensive skin disease and PAH. In contrast, a previous study had shown lower mean vitC levels in limited SSc patients [5], but the sample size was small (n=18) and these patients were noticeably younger than ours (44 and 47 vs. 66 years). This might explain the discrepancy with our results. Some basic studies have supported the hypothesis that oxidative stress may promote SSc progression [9]. We therefore suggest that vitC deficiency should
ACCEPTED MANUSCRIPT
NU
SC
RI
PT
be sought and treated in SSc patients that are malnourished, have extensive skin involvement or are diagnosed with PAH. Our results are in agreement with several studies that report significantly lower Se levels in patients with SSc than in control subjects [4,5,28]. In our study, selenium deficiency was significantly associated with cardiac involvement. The relationship between selenium and heart diseases was first illustrated by the description of Keshan disease in the 1970s. Keshan disease is a cardiomyopathy that affects people living in rural areas of a Se -deficient zone in China. It is characterized by multifocal necrosis and replacement fibrosis of the myocardium, resulting in acute or chronic heart failure [29]. The role of Se deficiency in the aetiopathogeny of Keshan disease has been suggested in early reports and confirmed in a recent meta-analysis [30]. Se deficiency has also been associated with dilated cardiomyopathy in patients originating from other countries [31,32], and in patients receiving parenteral alimentation [33,34]. Furthermore, in a case-control study nested in a cohort of 11 000 subjects originating from Finland and followed up for seven years from 1972, initial Se levels ≤45 μg/L were associated with a 2 to 3-fold increase in cardiovascular morbidity and mortality [35]. In a Nigerian study, 14 out of 35 women suffering from peripartum cardiomyopathy had plasma Se levels ≤45 µg/L versus none in controls [36]. The efficacy of Se supplementation has only been tested at a population level in Keshan disease endemic areas. The incidence of this disease and of a latent type of this disease was significantly lower in the 1.05 million people supplemented with Se than in the 0.6 million controls [37]. Our findings indicate that six of the nine patients suffering from myocardial involvement tested for Se had low levels. Ultimately, Se deficiency could be suspected to have participated in the cardiomyopathy onset of three patients. Given our patient selection method, based on a historical micronutrient workup, these numbers probably underestimate the real frequency of Se deficiency in SSc and in SSc with myocardial involvement. We therefore suggest testing for Se status in all SSc patients with cardiac involvement. Furthermore, we believe that targeted Se supplementation could be beneficial to patients with cardiac involvement, using preferably sublingual administration to avoid a potential malabsorption.
EP T
ED
MA
In conclusion, we report a high prevalence of micronutrient deficiency in SSc patients independently of clinical and biological markers of malnutrition. Our findings highlight that vitC deficient patients were more likely to have diffuse cutaneous disease and PAH. Interestingly, the present series also shows that Se deficient patients were more likely to have cardiac involvement. Our findings thus underscore that micronutrient deficiency may play a role in the pathophysiology of this chronic progressive disease. We suggest testing vitC, Se and other micronutrient levels routinely and in patients with diffuse skin involvement, cardiac involvement or PAH. Detected deficient patients would be able to benefit from dietary advice (as much as they can follow it) and from supplementation with targeted micronutrients and adequate dosage. The results should be confirmed in larger cohorts and targeted supplementation efficacy and security need to be prospectively evaluated in this population.
AC C
CONTRIBUTORS RD and GP designed the study. Analyses were conducted by BC, ML, CCF and RD. Patients and clinical data for the study were collected by RD, GM, LA, PA, DA, LA, LS, GP. The article has been drafted by RD, ML and GP. All authors critically read, provided input and approved the final version of the article. DECLARATION OF INTEREST The authors declare that there is no conflict of interest regarding the publication of this paper. FUNDING This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. REFERENCES [1] [2] [3]
Varga J, Abraham D. Systemic sclerosis: a prototypic multisystem fibrotic disorder. The Journal of Clinical Investigation 2007;117:557. Codullo V, Cereda E, Crepaldi G, Cappello S, Montecucco C, Caccialanza R, et al. Disease -related malnutrition in systemic sclerosis: evidences and implications. Clin Exp Rheumatol 2015;33:S190-194. Harrison E, Herrick AL, McLaughlin JT, Lal S. Malnutrition in systemic sclerosis. Rheumatology (Oxford, England) 2012;51:1747.
ACCEPTED MANUSCRIPT
[12] [13] [14]
[15] [16] [17] [18]
[19]
[20] [21]
[22] [23] [24] [25]
[26] [27] [28] [29] [30]
PT
[11]
RI
[10]
SC
[9]
NU
[8]
MA
[7]
ED
[6]
EP T
[5]
Lundberg AC, Akesson A, Akesson B. Dietary intake and nutritional status in patients with systemic sclerosis. Ann Rheum Dis 1992;51:1143–8. Herrick AL, Rieley F, Schofield D, Hollis S, Braganza JM, Jayson MI. Micronutrient antioxidant status in patients with primary Raynaud’s phenomenon and systemic sclerosis. J Rheumatol 1994;21:1477–83. Herrick AL, Worthington H, Rieley F, Clarke D, Schofield D, Braganza JM, et al. Dietary intake of micronutrient antioxidants in relation to blood levels in patients with systemic sclerosis. J Rheumatol 1996;23:650–3. Mehta PA, Dubrey SW. High output heart failure. QJM: An International Journal of Medicine 2009;102:235– 241. Elhai M, Meune C, Boubaya M, Avouac J, Hachulla E, Balbir-Gurman A, et al. Mapping and predicting mortality from systemic sclerosis. Annals of the Rheumatic Diseases 2017;76:1897. Wan Y-N, Yan J-W, Peng W-J, Zhang J-Q, Xiao C-C, Wang B-X, et al. Micronutrients, their potential effect on patients with systemic sclerosis. Mod Rheumatol 2014;24:709–14. doi:10.3109/14397595.2013.844383. Codullo V, Cereda E, Klersy C, Cavazzana I, Alpini C, Bonardi C, et al. Serum prealbumin is an independent predictor of mortality in systemic sclerosis outpatients. Rheumatology (Oxford, England) 2016;55:315. Hoogen F, Khanna D, Fransen J, Johnson SR, Baron M, Tyndall A, et al. 2013 classification criteria for systemic sclerosis: an American College of Rheumatology/European League against Rheumatism collaborative initiative. Arthritis & Rheumatology 2013;65:2737–2747. N.d. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA, et al. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202–5. Clements PJ, Lachenbruch PA, Seibold JR, Zee B, Steen VD, Brennan P, et al. Skin thickness score in systemic sclerosis: an assessment of interobserver variability in 3 independent studies. Journal of Rheumatology 1993;20:1892–1896. Miller MR, Crapo R, Hankinson J, Brusasco V, Burgos F, Casaburi R, et al. General considerations for lung function testing. Eur Respir J 2005;26:153–61. doi:10.1183/09031936.05.00034505. N.d. N.d. Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M, et al. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the’malnutrition universal scree ning tool’('MUST’) for adults. The British Journal of Nutrition 2004;92:799. Baron M, Hudson M, Steele R, Canadian Scleroderma Research Group. Malnutrition is common in systemic sclerosis: results from the Canadian scleroderma research group database. J Rheumatol 2009;36:2737–43. doi:10.3899/jrheum.090694. Cereda E, Codullo V, Klersy C, Breda S, Crippa A, Rava ML, et al. Disease -related nutritional risk and mortality in systemic sclerosis. Clinical Nutrition (Edinburgh, Scotland) 2014;33:558. Valentini G, Della R, Bombardieri S, Bencivelli W, Silman A, D’Angelo S, et al. European multicentre study to define disease activity criteria for systemic sclerosis.* II. Identification of disease activity variables and development of preliminary activity indexes. Annals of the Rheumatic Diseases 2001;60:592. Medsger Jr TA, Bombardieri S, Czirjak L, Scorza R, Della Rossa A, Bencivelli W. Assessment of disease severity and prognosis. Clinical and Experimental Rheumatology 2003;21:S42. Jacob RA, Skala JH, Omaye ST. Biochemical indices of human vitamin C status. The American Journal of Clinical Nutrition 1987;46:818. Institute of Medicine (US) Panel on Dietary Antioxidants and Related Compounds. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington (DC): National Academies Press (US); 2000. Baron M, Bernier P, Côté LF, Delegge MH, Falovitch G, Friedman G, et al. Screening and management for malnutrition and related gastro-intestinal disorders in systemic sclerosis: recommendations of a North American expert panel. Clin Exp Rheumatol 2010;28:S42–6. Teh LS, Johns CW, Shaffer JL, Booth EJ, Aarons L, Bennett RJ, et al. Ascorbic acid absorption in patients with systemic sclerosis. J Rheumatol 1997;24:2353–7. Heidelbaugh JJ. Proton pump inhibitors and risk of vitamin and mineral deficiency: evidence and clinical implications. Therapeutic Advances in Drug Safety 2013;4:125–133. Tikly M, Channa K, Theodorou P, Gulumian M. Lipid peroxidation and trace elements in systemic sclerosis. Clinical Rheumatology 2006;25:320. Li GS, Wang F, Kang D, Li C. Keshan disease: an endemic cardiomyopathy in China. Hum Pathol 1985;16:602–9. Zhou H, Wang T, Li Q, Li D. Prevention of Keshan Disease by Selenium Supplementation: a Systematic Review and Meta-analysis. Biol Trace Elem Res 2018. doi:10.1007/s12011-018-1302-5.
AC C
[4]
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
[31] Collipp PJ, Chen SY. Cardiomyopathy and selenium deficiency in a two-year-old girl. N Engl J Med 1981;304:1304–5. doi:10.1056/NEJM198105213042120. [32] Chou HT, Yang HL, Tsou SS, Ho RK, Pai PY, Hsu HB. Status of trace elements in patients with idiopathic dilated cardiomyopathy in central Taiwan. Zhonghua Yi Xue Za Zhi (Taipei) 1998;61:193–8. [33] Fleming CR, Lie JT, McCall JT, O’Brien JF, Baillie EE, Thistle JL. Selenium deficiency and fatal cardiomyopathy in a patient on home parenteral nutrition. Gastroenterology 1982;83:689–93. [34] Johnson RA, Baker SS, Fallon JT, Maynard EP, Ruskin JN, Wen Z, et al. An occidental case of cardiomyopathy and selenium deficiency. N Engl J Med 1981;304:1210–2. doi:10.1056/NEJM198105143042005. [35] Salonen JT, Alfthan G, Huttunen JK, Pikkarainen J, Puska P. Association between cardiovascular death and myocardial infarction and serum selenium in a matched-pair longitudinal study. Lancet 1982;2:175–9. [36] Cénac A, Simonoff M, Moretto P, Djibo A. A low plasma selenium is a risk factor for peripartum cardiomyopathy. A comparative study in Sahelian Africa. Int J Cardiol 1992;36:57–9. [37] Cheng YY, Qian PC. The effect of selenium-fortified table salt in the prevention of Keshan disease on a population of 1.05 million. Biomed Environ Sci 1990;3:422–8. [38] \AAkesson A, \AAkesson B, Gustafson T, Wollheim F. Gastrointestinal function in patients with progressive systemic sclerosis. Clinical Rheumatology 1985;4:441–448. [39] Akesson A, Wollheim FA. Organ manifestations in 100 patients with progressive systemic sclerosis: a comparison between the CREST syndrome and diffuse scleroderma. British Journal of Rheumatology 1989;28:281. [40] Marasini B, Casari S, Bestetti A, Maioli C, Cugno M, Zeni S, et al. Homocysteine concentration in primary and systemic sclerosis associated Raynaud’s phenomenon. The Journal of Rheumatology 2000;27:2621–2623. [41] Caramaschi P, Volpe A, Canestrini S, Bambara LM, Faccini G, Carletto A, et al. Correlation between homocysteine plasma levels and nailfold videocapillaroscopic patterns in systemic sclerosis. Clinical Rheumatology 2007;26:902. [42] Krause L, Becker MO, Brueckner CS, Bellinghausen CJ, Becker C, Schneider U, et al. Nutritional status as marker for disease activity and severity predicting mortality in patients with systemic sclerosis. Annals of the Rheumatic Diseases 2010;69:1951. [43] Murtaugh MA, Frech TM. Nutritional status and gastrointestinal symptoms in systemic sclerosis patients. Clinical Nutrition (Edinburgh, Scotland) 2013;32:130. [44] Caporali R, Caccialanza R, Bonino C, Klersy C, Cereda E, Xoxi B, et al. Disease -related malnutrition in outpatients with systemic sclerosis. Clinical Nutrition (Edinburgh, Scotland) 2012;31:666. [45] Rosato E, Gigante A, Gasperini ML, Molinaro I, Di Lazzaro GG, Afeltra A, et al. Nutritional status measured by BMI is impaired and correlates with left ventricular mass in patients with systemic sclerosi s. Nutrition (Burbank, Los Angeles County, Calif) 2014;30:204. [46] Hughes M, Cooper GJ, Wilkinson J, New P, Guy JM, Herrick AL. Abnormalities of selenium but not of copper homeostasis may drive tissue fibrosis in patients with systemic sclerosis. Rheumatology (Oxford, England) 2015;54:747.
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Figure 1 – Flow chart. Except for the 6 patients excluded from the final analysis, thoroughness of the data was good, with 13% missing values on average
ACCEPTED MANUSCRIPT Table 1 – Study population characteristics. General data (n = 82)
Treatment
Age (y), m (r) Men BMI, m (r) Wei ght l oss Ma l nutrition
60 (21 - 88) 20 (24) 23 (15 - 50) 3 (4) 14 (17)
Scleroderma
PPI Va s odilatator Steroi ds Immunosuppressant
53 (65) 34 (46) 22 (28) 20 (25)
Lab tests 7.4 (0 - 46)
AC C
Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
RI
19 (23) 62 (76) 1 (1)
NU
ED
MA
50 (63) 9 (17) 26 (48)
15 (19) 7 (11) 6 (7) 26 (32) 5 (6) 3 (4) 6 (7)
SC
5 (0 - 40) 40 (16 - 60) 70 (100) 45 (55)
7 (9) 15 (33)
Context Sys tematic Pretra nsplantation Nutri ti onal risk or deficiency Symptoms Usual bloods Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µg/L), m (r) Micronutrients Vi ta min C mg/L, m (r) defi ciency Thi a min nmol /L, m (r) defi ciency Sel enium µg/L, m (r) defi ciency Zi nc µmol /L, m (r) defi ciency Fol ate ng/mL, m (r) defi ciency
PT
26 (32) 36 (44) 82 (100) 10 (12)
EP T
Di s ease duration (y), m (r) Auto - antibody Scl 70 Centromere ANA other Cutaneous form Di ffuse Li mi ted Si ne scleroderma Skin involvement Rodnan, m (r) Interincisive distance (mm), m (r) Ra yna ud Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement
17 (21) 7 (9) 7 (9) 14 (17) 2 (2) 4 (5) 5.5 (1 - 15)
54 (67) 3 (4) 14 (17) 10 (12) 68 (29 - 280) 13 (16) 2.7 (0 - 164) 12 (1 - 64) 38 (16 - 49) 13 (8 - 18) 4.8 (1.7 - 21) 1.4 (0.5 - 2.8) 85.5 (7 - 1661)
8.9 (1 - 26) 22 (31) 109 (53 - 568) 4 (6) 76 (40 - 114) 19 (35) 11 (8.1 - 17) 15 (48) 7.8 (1.6 - 40) 15 (23)
B12 pg/mL, m (r) defi ciency Pyri doxi ne nmol /L, m (r) defi ciency
396 (166 - 1059) 1 (1) 21 (4.5 - 137) 37 (62)
Results are presented as n (%) unless stated otherwise. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, ESR: erythrocyte sedimentation rate, other: other auto-antibodies included anti-RNA polymerase 3, anti-PmScl, anti-Ku, anti-SSA, anti-SSB, LFT: liver function tests, LVEF: left ventricular ejection fraction, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT Table 2 – Malnutrition and SSc phenotype. Malnutrition Patients (n=82) Age (y), m (r) Ma l e sex BMI ≤ 22.9 Wei ght l oss
No malnutrition
OR
95% CI
p
n = 14
n = 68
67 (31 - 88) 2 (14) 9 (90) 3 (33)
58 (21 - 82) 18 (26) 17 (42) 0
1.05 0.46 12.7
[1.00 - 1.10] [0.09 – 2.27] [1.47 – 109.89]
0.04 0.50 0.01 0.21
6.4 (0 - 38)
8.2 (0 - 46)
0.99
[0.93 - 1.05]
0.42
6 (43) 6 (43)
30 (45) 20 (30)
0.93 1.76
[0.29 - 2.96] [0.54 - 5.74]
0.90 0.36
4 (29) 9 (69) 35 (19 - 45) 7 (50)
15 (22) 51 (75) 40 (16 - 60) 38 (56)
0.77 0.75 0.93 0.79
[0.21 - 2.79] [0.20 – 2.75] [0.85 - 1.03] [0.25 - 2.50]
0.74 0.73 0.33 0.69
11 (85) 1 (20) 3 (60)
39 (59) 8 (17) 23 (47)
3.81 1.25 1.70
[0.78 – 18.57] [0.12 – 12.71] [0.26 – 11.06]
0.12 1.00 0.66
3 (21) 5 (56)
4 (6) 10 (27)
4.36 3.37
[0.86 - 22.22] [0.75 – 15.15]
0.09 0.13
12 (18) 5 (9) 6 (9) 23 (34) 3 (4) 2 (3) 6 (9)
1.53 2.45
[0.36 – 6.47] [0.40 – 14.85]
0.53 3.61 2.50
[0.14 – 2.10] [0.54 – 23.96] [0.21 – 29.65]
0.69 0.30 0.58 0.53 0.20 0.44 0.58
5 (36) 2 (14) 2 (14) 4 (29) 0 0 7.5 (3 - 15)
12 (18) 5 (7) 5 (7) 10 (15) 2 (3) 4 (6) 5 (1 - 13)
2.88 2.10 2.10 3.22
[0.81 - 10.25] [0.36 – 12.11] [0.36– 12.11] [0.89 – 11.62]
1.44
[1.14 - 1.83]
0.13 0.34 0.34 0.12 1.00 1.00 0.003
12 (86) 5 (36) 4 (29) 2 (14)
41 (60) 29 (48) 18 (28) 18 (28)
3.95 0.59 1.04 0.44
[0.82 – 19.06] [0.18 - 1.98] [0.29 - 3.76] [0.09 – 2.14]
0.12 0.39 1.00 0.50
83 (29 - 280) 2 (14) 4.6 (0.5 - 28.4) 28 (1 - 64) 35 (16 - 47) 10.6 (9.2 - 14) 4.8 (2.8 - 9.0) 1.0 (0.6 - 2.4) 94 (7 - 389)
68 (31 - 139) 11 (16) 2.5 (0 - 164) 12 (1 - 54) 39 (17 - 49) 12.9 (8.2 - 17.7) 4.6 (1.7 - 21.1) 1.4 (0.5 - 2.8) 79 (9 - 1661)
1.02 0.86 0.99 1.05 0.90 0.40 1.02 0.54 1.00
[1.00 - 1.03] [0.17 – 4.41] [0.96 - 1.03] [0.99 - 1.11] [0.82 - 0.99] [0.24 - 0.68] [0.83 - 1.25] [0.18 - 1.63] [0.99 - 1.00]
0.32 1.00 0.25 0.32 0.02 < 0.0001 0.44 0.20 0.26
8 (73) 1 (10) 3 (33) 2 (29)
14 (24) 3 (6) 16 (35) 13 (54)
8.57 1.85 0.94 0.34
[2.00 – 36.77] [0.17 – 19.85] [0.21 - 4.26] [0.05 – 2.102]
0.003 0.51 1.00 0.39
AC C
Treatment PPI Ca l ci um channel blocker Steroi ds Immunosuppressant drug
Lab tests
Usual bloods Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µ g/L), m (r) Micronutrient deficiency Vi ta min C Thi a min Sel enium Zi nc
RI SC
NU
ED
3 (21) 2 (14) 1 (7) 0
MA
3 (21) 2 (20)
EP T
Di s ease duration (y), m (r) Auto-antibody Anti -centromere Anti -Scl70 Skin involvement Di ffuse form Rodnan score ≤ 14 Interincisive distance, m (r) Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
PT
Scleroderma
ACCEPTED MANUSCRIPT Fol ate B12 Pyri doxi ne
3 (30)
12 (22) 1 (2) 30 (59)
[0.34 - 6.70]
2.45
[0.46 – 12.98]
AC C
EP T
ED
MA
NU
SC
RI
PT
7 (78)
1.50
0.69 1.00 0.46
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Results are presented as n (%) unless stated otherwise. P-values are obtained using the -square or Fisher’s exact test for categorical variables and the Mann–Whitney test for continuous variables. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, DSS : disease severity scales, ESR: erythrocyte sedimentation rate, LFT: liver function tests, LVEF: left ventricular ejection fraction, mRSS: modified Rodnan skin score, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT Table 3 – Vitamin C deficiency and SSc phenotype. Deficiency Patients (n=70) Age (y), m (r) Ma l e sex BMI ≤ 22.9 Wei ght l oss Ma l nutrition
No deficiency
OR
95% CI
p
n = 22
n = 48
70 (21 - 88) 5 (23) 10 (77) 1 (11) 8 (67)
58 (28 - 84) 13 (27) 12 (38) 1 (17) 3 (43)
1.04 0.79 5.56 0.62 8.57
[0.99 - 1.08] [0.24 - 2.58] [1.27 – 24.29] [0.03 - 12.41 [2.00 – 36.77]
0.03 0.70 0.02 1.00 0.003
8.2 (0 - 44)
5.8 (0 - 46)
1.02
[0.98 - 1.07]
0.64
6 (29) 12 (57)
16 (33) 19 (40)
0.8 2.04
[0.26 - 2.45] [0.72 - 5.76]
0.70 0.18
7 (32) 13 (59) 33.5 (16 - 50) 15 (68)
8 (17) 39 (81) 40 (30 - 60) 24 (50)
0.35 0.33 0.91 2.14
[0.11 - 1.11] [0.11 – 1.02] [0.82 - 0.99] [0.74 - 6.19]
0.07 0.05 0.17 0.16
14 (67) 5 (36) 9 (69)
29 (63) 4 (12) 15 (43)
1.17 4.17 3
[0.40 - 3.48] [0.92 – 18.88] [0.77 – 11.63]
0.77 0.10 0.10
6 (27) 7 (50)
0 6 (24)
3.17
[0.79 – 12.75]
0.0006 0.16
8 (17) 2 (5) 4 (8) 19 (40) 4 (8) 0 3 (6)
1.64 2.33 1.10 0.45 0.52
[0.46 - 5.90] [0.30 – 18.15] [0.19 - 6.51] [0.14 - 1.42] [0.06 – 4.98]
1.50
[0.23 - 9.69]
0.50 0.58 1 0.17 1 0.10 0.65
10 (45) 3 (14) 3 (14) 6 (27) 1 (5) 2 (9) 8 (3 - 15)
7 (15) 4 (8) 4 (8) 7 (15) 1 (2) 2 (4) 5 (1 - 10)
4.05 1.74 1.74 1.88 2.24 2.30 1.68
[1.26 - 13.04] [0.35 - 8.52] [0.35 - 8.52] [0.56 – 6.27] [0.13 – 37.52] [0.30 – 17.50] [1.27 - 2.21]
0.01 0.67 0.67 0.34 0.53 0.59 < 0.0001
18 (82) 8 (40) 8 (38) 4 (19)
26 (54) 23 (55) 10 (22) 11 (24)
3.81 0.55 2.22 0.75
[1.12 – 12.94] [0.19 - 1.62] [0.72 - 6.83] [0.21 – 2.70]
0.03 0.28 0.16 0.76
64 (30 - 280) 5 (23) 5.8 (0 - 164) 22 (2 - 64) 35 (25 - 47) 11.6 (8.2 - 13.3) 4.7 (2.3 - 9.1) 1.3 (0.2 - 2.2) 81.5 (7 - 389)
68 (40 - 139) 7 (15) 2.2 (0 - 112) 11 (1 - 54) 40 (16 - 49) 13.4 (10.2 - 17.7) 4.4 (1.7 - 21.1) 1.4 (0.5 - 2.8) 88 (9 - 1661)
1.01 1.72 1.02 1.05 0.91 0.38 1.02 0.41 1.00
[0.99 - 1.02] [0.48 - 6.19] [1.00 - 1.05] [1.00 - 1.10] [0.83 – 1.00] [0.23 - 0.64] [0.86 - 1.21] [0.16 - 1.07] [0.99 - 1.00]
0.98 0.50 0.02 0.08 0.005 < 0.0001 0.33 0.09 0.95
AC C
Treatment
PPI Ca l ci um channel blocker Steroi ds Immunosuppressant drug
Lab tests
Crea ti nin (μmol/L), m (r) Abnormal LFT CRP (mg/L), m (r) ESR (mm/h), m (r) Al bumin (g/L), m (r) Hemoglobin (g/dL), m (r) Neutrophils (G/L), m (r) Lymphocytes (G/L), m (r) Ferri tin (µ g/L), m (r)
RI
SC
NU MA
ED
6 (27) 2 (12) 2 (9) 5 (23) 1 (5) 2 (9) 2 (9)
EP T
Di s ease duration (y), m (r) Auto-antibody Anti -centromere Anti -Scl70 Skin involvement Di ffuse form Rodnan score ≤ 14 Interincisive distance, m (r) Di gi tal ulcers Lung involvement Interstitial l ung disease VC ≤ 70% DLCO ≤ 70% Pulmonary artery involvement PAH s PAP ≥ 35 mmHg Heart involvement Ca rdi ac i nvolvement LVEF ≤ 50% Muscle involvement Joint involvement Renal crisis Nerve involvement Liver or biliary tract involvement Gastrointestinal tract involvement Oes ophagitis or Barrett's mucosa Ga s tritis or Ul cer SIBO or CIPO Moti l ity disorders Pa ncreas None Summed Medsger DSS, m (r)
PT
Scleroderma
Results are presented as n (%) unless stated otherwise. P-values are obtained using the Chi-squared or Fisher’s exact test for categorical variables and the Mann–Whitney test for continuous variables. m (r): median (range). ANA: antinuclear antibody, CIPO: chronic intestinal pseudo-obstruction, CRP: C-reactive protein, DLCO: transfer factor for carbon monoxide, DSS : disease severity scales, ESR: erythrocyte sedimentation rate, LFT: liver function tests, LVEF:
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
left ventricular ejection fraction, mRSS: modified Rodnan skin score, PAH: pulmonary artery hypertension, PPI: proton-pump inhibitor, SIBO: small intestine bacterial overgrowth, sPAP: systolic pulmonary artery pressure, VC: vital capacity, y: year.
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Figure 2 – Plasma Selenium concentration in patients with low and normal left ventricular ejection fraction LVEF: left ventricular ejection fraction; *p<0.05
ACCEPTED MANUSCRIPT Table 4 – Literature review.
n Age (y) Diffuse SSc (%) BMI Malnutrition Thiamine Folate B12 Vitamin C
Selenium
Akesson, 1985 [38]
Akesson, Lundberg, 1989 1992 [39] [4]
Caramaschi, Herrick, Herrick, Herrick, Marasini, 2006 1994 [5] 1994 [5] 1996 [6] 2000 [40] [41]
24 41 38
100 44 37
9 m 44 0
30 48 57 22
9 m 47 100
12 m 47 58
30 58 33
Definition % nmol/L deficiency (%) ng/mL
114 (N >55.2)
β
127
5.7
361 (N >149)
385
deficiency (%) mg/L
5.5
α
m 5.3
70.0 (N >53.7)
deficiency (%)
67.9
T P
α
E C
m 9.7
D E
deficiency (%) μg/L
α
m 82.1
A M
α
m 75.0
α
396
Krause, 2010 [42]
Murtaugh Caporali, 2012 Rosato, , [44] and Cereda, 2014 [45] 2013 [43] 2014 [20]
586 55 39 26
124 54 47 23
24 54 13 27
160 62 19 25
94 53 43 23
MUST ≥2 17
BMI <19 14
MUST ≥2 25
MUST ≥2 9.4
BMI <20 19
T P
I R
C S
U N
β
deficiency (%) pg/mL
60 55 37
Baron, 2009 [19]
4.9 (N >3.0)
α
m 6.0
α
m 74.2
α
Results are presented as mean unless stated otherwise. m: median; MUST: MUST score; α statistically significant difference with control group; β whole blood.
C A
ACCEPTED MANUSCRIPT Supplementary Table – Demographic and clinical characteristics of the seven patients that died during study follow-up. Patient N°
1
2
3
4
5
6
Age (y)
52
77
43
70
21
66
Sex
male
female
female
male
female
male
Malnutrition
no
no
no
no
no
yes
Follow-up duration (y)
0.1
4.2
0.3
0.7
0.2
0.3
Disease duration (y)
1.1
0.6
13
15
1.6
2.9
Auto-antibody
scl70
centromere
scl70
centromere
scl70
ANA
Cutaneous form
diffuse
limited
diffuse
diffuse
diffuse
-
mRSS
27
7
18
25
27
-
ILD
yes
no
yes
yes
no
yes
PAH
no
no
no
no
no
no
Heart involvement
no
no
no
no
yes
no
Renal crisis
yes
no
no
yes
no
no
GI tract involvement
yes
yes
yes
yes
yes
yes
Summed Medsger DSS
4
8
8
11
6
Medication
no
no
Comment
-
-
EP T
Lab tests
RI SC
NU MA
ED
Treatment
PT
Scleroderma
9
steroids & MMF steroids & MTX
steroids
-
screened for HSCT
-
steroids, MMF & tacrolimus lung transplant for ILD
malabsorption and infection
-
transplantation malabsorption planning
HSCT planning
weight loss
CRP (mg/L)
29
29
23
6
71
16
Albumin (g/L)
-
26
28
39
26
-
9
12
12
8
11
yes
yes
yes
yes
-
-
no
-
-
yes
Hemoglobin (g/dL)
AC C
Context
Vitamin C deficiency Selenium deficiency
13
formerly deficient
Death Cause
pneumonia
sepsis
acute respiratory failure
SRC
aspiration pneumonia
coma
SSc-related
yes
-
yes
yes
yes
-
ANA: antinuclear antibodies, DSS : disease severity scales, GI: gastro-intestinal, HSCT: hematopoietic stem cell transplantation, mRSS: modified Rodnan skin score, MMF: mycophenolate mofetil, MTX: methotrexate, ILD: interstitial lung disease, PAH: pulmonary artery hypertension, SRC: scleroderma renal crisis
ACCEPTED MANUSCRIPT
AC C
EP T
ED
MA
NU
SC
RI
PT
Highlights Micronutrient deficiency is frequent in SSc and is not associated with malnutrition SSc primary heart disease is associated with selenium deficiency Malnutrition and SSc severity is associated with vitamin C deficiency Routine micronutrient testing and targeted supplementation might be useful in SSc
Figure 1
Figure 2