- Email: [email protected]
Potential coeliac disease in Type 1 diabetes mellitus: Does a positive antibody lead to increased complications?
Nutrition, Metabolism & Cardiovascular Diseases (2014) 24, 378e383
Available online at www.sciencedirect.com
Nutrition, Metabolism & Cardiovascular Diseases journal homepage: www.elsevier.com/locate/nmcd
Potential coeliac disease in Type 1 diabetes mellitus: Does a positive antibody lead to increased complications? J.S. Leeds a,*, A.D. Hopper a, M. Hadjivassiliou b, S. Tesfaye c, D.S. Sanders a a
Gastroenterology & Liver Unit, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK Department of Neurology, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK c Department of Diabetes, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK b
Received 17 November 2012; received in revised form 12 August 2013; accepted 3 September 2013 Available online 23 October 2013
KEYWORDS Type 1 diabetes; Glycaemic control; Coeliac disease; Complications
Abstract Background and aims: Coeliac disease (CD) is more common in people with Type 1 diabetes and is associated with poorer glycaemic control, lipid profiles, nephropathy and retinopathy. Potential CD (positive serology but normal duodenal biopsy) is associated with neuropathy but patients with coexisting Type 1 diabetes were excluded. The aim was to determine whether potential CD is associated with increased microvascular complications in patients with Type 1 diabetes. Methods and results: Four groups were recruited; 1) patients with Type 1 diabetes and potential CD, 2) patients with Type 1 diabetes and newly identified CD, 3) patients with Type 1 diabetes alone and 4) patients with CD alone. Glycaemic control, quality of life, lipid profile and microvascular complication rates were examined. As many as 76 individuals were included in the study: 22 in group 1, 14 in group 2, 24 in group 3 and 16 in group 4. There were no differences in age, gender, BMI and diabetes duration between the groups. Patients in group 1 had significantly lower total cholesterol compared to group 3 (p Z 0.003) but higher than group 2 (p Z 0.027). There were no significant differences in HbA1c, HDL cholesterol, cholesterol:HDL ratio, creatinine, quality of life scores or prevalence of neuropathy between individuals in group 1 and the other groups. Conclusions: This is the first study to assess the effects of potential CD in patients with Type 1 diabetes. It appears that an enteropathy is required as well as antibody positivity in order to increase the risk of diabetes related complications. This pilot data requires further longitudinal validation. ª 2013 Elsevier B.V. All rights reserved.
Introduction The association between coeliac disease (CD) and Type 1 diabetes mellitus has been well described with the prevalence of CD ranging between 3 and 5% in the largest Abbreviations: CD, coeliac disease; GFD, gluten free diet; DAFNE, Dose Adjustment for Normal Eating; HbA1c, glycated haemoglobin; IgA, immunoglobulin A; HDL, high density lipoprotein; CASE, computed aided sensory evaluator; BMI, body mass index. * Corresponding author. Room 2.39, Ashgrove House, Department of Gastroenterology, Aberdeen Royal Infirmary, Aberdeen AB25 2ZN, UK. Tel.: þ44 114 2261179; fax: þ44 114 2712692. E-mail addresses: [email protected], [email protected] (J.S. Leeds). 0939-4753/$ - see front matter ª 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.numecd.2013.09.005
studies [1e3]. Recently our group has shown that patients with Type 1 diabetes mellitus and newly identified CD have worse glycaemic control and lipid profiles as well as higher prevalence of microvascular complications. The underlying mechanism was not investigated by this study but the effect upon glycaemic control is likely to be responsible [4]. CD, in the absence of diabetes, is also associated with neurological manifestations with one study showing changes in electroneuromyography even in well controlled CD [5]. These neurological complications can also occur in patients with positive coeliac serology in the absence of enteropathy. Hadjivassiliou et al. looked at gluten ataxia
Type 1 diabetes mellitus and potential coeliac disease
patients and showed that only 13/28 patients had gross histological changes in the duodenum [6]. This offers insight into the role of auto-antibodies as being part of the pathophysiological mechanism. This is separate from gluten sensitivity which is a gluten-mediated immune reaction and gluten allergic reactions which are IgE mediated. Individuals with gluten sensitivity are believed to have a different type of immune mediated reaction and do not have enteropathy but can test positive to antibodies to gliadin [7]. The implications of potential CD to the individual are still unknown as it may be that in time they will develop the histological changes of CD [8,9]. Support for this hypothesis comes from an Italian study of children with Type 1 diabetes, where histological progression was demonstrable following the rectal instillation of gliadin [10]. In gluten ataxia it has been demonstrated that these neurological complications can be arrested or reversed by the institution of a gluten free diet (GFD) [11]. The pathophysiological changes appear to lie in the humoural immune response and a GFD appears to be beneficial [11,12]. Finally recent Swedish work has suggested that individuals with a positive antibody and no enteropathy have an increased mortality when compared to controls [13]. Theoretically there may be benefits to having CD with studies in non-diabetic CD cohorts showing a reduced risk of ischaemic heart disease and hypertension as well as lower cholesterol levels [14,15]. Our previous study showed that patients with Type 1 diabetes and CD had significantly lower cholesterol levels prior to commencement of a GFD [4]. A recent study showed that following a GFD, there was no change in total cholesterol but HDL cholesterol increased by an average of 12% [16]. However, this issue is of more importance to those patients who are already at increased cardiac risk such as those with preexisting diabetes. In support of this, one study has shown that patients with CD and Type 1 diabetes have a higher rate of subclinical atherosclerosis compared to either condition alone [17]. There is no data in patients with Type 1 diabetes and potential CD particularly with respect to glycaemic control, lipid profiles and microvascular complications. The aim of this study was to examine the effect of potential CD on important outcomes in patients with Type 1 diabetes. Methods Study groups 4 specific groups were identified and studied. Group 1: patients with Type 1 diabetes and potential CD (positive antibodies but normal duodenal biopsy), group 2: patients with Type 1 diabetes and newly identified CD, group 3: patients with Type 1 diabetes alone and group 4: patients with newly diagnosed CD but without diabetes. Participants with type 1 diabetes The Sheffield Diabetes Centre operates at both the Royal Hallamshire Hospital and the Northern General Hospital.
379
This covers a population of approximately 500 000 and provides tertiary referral services for the South Yorkshire region. There are approximately 2000 patients with Type 1 diabetes defined using the American Diabetes Association position statement [18]. Around 95% of patients with Type 1 diabetes are managed by the diabetes centre in secondary care. Inclusion criteria were patients with Type 1 diabetes >16 years. Exclusion criteria were age <16 years, inability to consent or diabetes other than Type 1. Where diabetes type was uncertain, the notes were reviewed with the treating consultant and a decision made concerning diabetes type. If diabetes type was still uncertain then the individual was excluded. Patients were prospectively recruited when attending for annual review, foot clinic or Dose Adjustment for Normal Eating (DAFNE) clinic. 1000 individuals with Type 1 diabetes were previously recruited [4] and from this cohort the study groups were identified. Participants with CD Patients with newly identified CD were recruited consecutively from the gastrointestinal outpatient’s clinic. All patients had biopsy confirmed CD with at least partial villous atrophy (using the modified Marsh criteria) [19] and underwent the neurological investigations prior to commencement on a GFD. Study group investigations Study participants completed a health questionnaire and the Short Form 36 (version 2) quality of life assessment questionnaire (QualityMetric Incorporated TM). Data were prospectively collected including age, gender, ethnicity, drug history and other medical co-morbidities. Blood was taken for HbA1c (in the diabetes cohorts), renal profile, ile (including total and HDL cholesterol), full blood count, IgA endomysial antibody, IgA anti-tissue transglutaminase antibody and total IgA level. All participants with either a positive antibody or IgA deficiency underwent duodenal biopsy to confirm the presence or absence of enteropathy. Retinopathy assessed by notes review and graded as no retinopathy, background changes, pre-proliferative changes or proliferative changes as described by the National Screening Committee for Diabetic Retinopathy [20]. Patients undergo annual retinal photography and were reviewed within 3 months of diagnosis. All groups were assessed for peripheral neuropathy by a combination of quantitative sensory threshold testing, cardiac autonomic function tests and electrophysiological testing of four nerves. Quantitative sensory threshold testing was performed using the Computer Aided Sensory Evaluator (CASE IV version 4.27) which provides cold, vibration and heat pain detection thresholds on the dorsum of the right foot. Cardiovascular autonomic function tests are measured using the O’Brien protocol. A 3-lead ECG is attached whilst
380
J.S. Leeds et al.
Table 1 Demographics and characteristics of the individuals with potential CD. Patient
Age
Sex
Symptoms
Associations
EMA þve
Anaemic
1 2 3 4 5 6
40 48 53 69 26 43
M M M M M F
Alternating bowel habit None IBS None None None
Yes No No No No Yes
No No No No No Yes
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
26 53 47 24 49 35 46 49 63 45 45 62 57 57 42 52
F M M M M M F F M M M M M M M F
None None None None None Constipation Alternating bowel habit IBS None None Alternating bowel habit None None None Diarrhoea None
Epilepsy Nil IgA deficiency Nil Nil Pernicious anaemia, hypothyroidism Nil Hypothyroidism IgA deficiency Nil Nil Pancreatitis Nil Hypothyroidism Nil Hypothyroidism IgA deficiency Nil None Osteoporosis Nil Nil
No No Yes No No No Yes No No No No No No No No Yes
No No No No No Yes No Yes No No No No No No No No
IBS Z irritable bowel syndrome, IgA Z immunoglobulin A.
lying supine. Following a rest period, measurements of the time between ‘R’ waves on the ECG are made and changes in heart rate based on RR variation over time are recorded. Degree of variation is calculated using the RR ratio and is corrected for age. RR ratios were measured at rest, during deep breathing, during the Valsalva manoeuvre and following rising from supine. Change in blood pressure from lying to standing was also measured. Autonomic dysfunction was diagnosed if two or more of these tests were abnormal. Electrophysiological testing was performed on the radial nerve, sural nerve, common peroneal nerve and the tibial nerve using surface electrodes (SYNERGY version 10.0, Oxford instruments). The nerves were subjected to supramaximal stimulation to allow measurement of the amplitude and latency.
Results
Statistical analyses were performed using SPSS v17.0. To determine whether there were any differences between the averages in each group either a KruskaleWallis test (nonparametric) or a one way ANOVA (parametric) was performed. Comparison between groups was performed using the median values and the Mann Whitney U test or the means and the Students t test. A p value less than 0.05 was considered significant.
76 individuals were included in the study. There were 22 individuals with Type 1 diabetes and potential CD (group 1: positive antibody but normal biopsy), 14 individuals with newly identified CD and Type 1 diabetes (group 2), 24 individuals with Type 1 diabetes alone (group 3) and 16 with CD alone (group 4). The demographics and characteristics of the individuals with potential CD are shown in Table 1. There were no significant differences between the groups with respect to age, diabetes duration, body mass index (BMI), cholesterol:HDL ratio, triglyceride levels or creatinine. There were no significant differences in HbA1c, HDL cholesterol, cholesterol:HDL ratio or creatinine between individuals in group 1 and the other groups. Total cholesterol was significantly lower in group 1 compared to group 3. As previously shown [4], group 2 had significantly higher HbA1c compared to group 3, and significantly lower total cholesterol compared to both groups 1 and 3 (summarised in Table 2). There were no significant differences in quality of life scores between any of the groups. With respect to other autoimmune conditions, 4/22 in group 1, 4/14 in group 2, 4/24 in group 3 and 1/16 in group 4 had an autoimmune condition previously recognised (p Z 0.56). In group 2 there were 2 individuals with multiple autoimmune conditions; one with Type 1 diabetes, CD, Addisons and hyperthyroidism, one with Type 1 diabetes, CD, dermatitis herpetiformis and pernicious anaemia.
Ethical approval
Results of peripheral nerve assessment
Ethical approval was obtained from the South Sheffield Research Ethics Committee and written informed consent obtained from all individuals.
There were no differences in vibration, cold or heat pain detection levels between all of the groups studied. None of those in group 4 had any abnormalities in the autonomic
Statistical analysis
Type 1 diabetes mellitus and potential coeliac disease
381
Table 2 Summary of values for all variables in all groups with Type 1 diabetes. Variable
Group 1 CD potential
Group 2 CDDM
Group 3 type 1 DM only
p Value
Age (years) Duration (years) BMI (Kg/m2) HbA1c (%) (mmol/mol) Cholesterol HDL Chol:HDL ratio Triglycerides Creatinine
46.9 24 28.4 8.0 64 4.8 1.34 3.1 0.9 79.5
44.2 22.5 25.4 8.2 66 4.05 1.1 3.4 0.85 78.5
46.6 28.5 24.2 7.5 58 5.1 1.56 2.8 0.75 83.0
(21e71) (4e71) (17.3e32.6) (1.93)
NS NS NS NSa
(1.4) (0.8) (1.8) (0.6) (10.3)
0.011b 0.04c NS NS NS
(24e69) (2e40) (19.7e34.4) (1.38) (1.38) (0.7) (1.3) (1.0) (27.0)
(21e82) (5e44) (20.1e36.9) (2.5) (1.43) (0.6) (1.7) (0.8) (35.0)
HbA1c Z glycosylated haemoglobin, HDL Z high density lipoprotein. Values in brackets are interquartile ranges. a KruskaleWallis test. CDDM HbA1c vs. Type 1 DM only Z 0.05. b KruskaleWallis test. CDDM potential cholesterol vs. CDDM Z 0.027, vs. Type 1 DM only Z 0.003. c KruskaleWallis test. CDDM HDL vs. Type 1 DM only Z 0.017.
function tests and there were no differences in the frequency of abnormalities between groups 1 and 3. With respect to peripheral nerve conduction studies there were no significant differences in amplitude or conduction velocity when comparing between groups 1 and 3 but there was a significant difference seen in conduction velocity when comparing any of the groups with diabetes to group 4. This is summarised in Table 3. Retinopathy Overall, there was no difference in the prevalence of retinopathy between the cases and controls. When looking at higher grades of retinal disease (pre-proliferative or higher) there was a significantly higher prevalence of more advanced retinopathy in group 2 compared to group 3 (p Z 0.02). However, there was no significant difference between group 1 and either group 2 or 3. Discussion This is the first study to examine the effect of potential CD in people with known Type 1 diabetes with respect to
important diabetes related outcomes. Although we have previously reported that newly identified CD and Type 1 diabetes patients had significantly higher HbA1c compared to controls with Type 1 diabetes alone [4] these difference were not observed when assessing those patients with potential CD and Type 1 diabetes. Furthermore when we looked at lipid profiles, renal function, peripheral nerve function and retinal changes no significant differences were observed. As in the previous study, there was a higher prevalence of more advanced retinal lesions in those individuals with both Type 1 diabetes and CD but this was not observed in those with potential CD and Type 1 diabetes. We have previously shown that patients with both CD and Type 1 diabetes have a higher prevalence of microvascular complications [4]. The mechanism for this was not specifically assessed but was hypothesised to be due to the adverse effect on glycaemic control. The present study suggests that positive antibodies are insufficient to cause a noticeable effect on important outcomes and that enteropathy is also required. We also found no evidence of neuropathy in patients with newly identified CD alone. However, as the prevalence of neurological abnormalities in cohorts with CD ranges between 2 and 8% [6,21], the size of the CD group in the present study is probably too
Table 3 Results of mean nerve conduction studies in each group. Peripheral nerve Radial Amplitude (uV) Velocity (m s1) Sural Amplitude (uV) Velocity (m s1) Common peroneal Amplitude (uV) Velocity (m s1) Tibial Amplitude (uV) Velocity (m s1)
CDDM
CD potential
Type 1 DM alone
CD alone
p
28.2 (6.4) 49.2 (3.8)
22.2 (3.5) 51.5 (2.1)
32.0 (4.6) 51.2 (2.2)
36.2 (6.7) 57.9 (1.2)
NS 0.052a
11.1 (2.4) 38.0 (3.1)
11.2 (1.3) 39.6 (1.4)
12.1 (3.9) 37.4 (0.7)
15.8 (2.5) 44.0 (1.4)
NS 0.05b
4.3 (0.7) 41.7 (2.0)
6.8 (2.5) 39.5 (1.4)
3.6 (0.8) 36.7 (2.5)
5.3 (0.6) 49.6 (0.9)
NS 0.001c
11.1 (2.0) 41.9 (2.0)
11.8 (3.9) 37.4 (1.5)
8.7 (1.8) 37.8 (1.9)
11.2 (1.9) 45.8 (0.8)
NS 0.001d
Results in brackets are standard errors. a Differences only seen between CD alone b Differences only seen between CD alone c Differences only seen between CD alone d Differences only seen between CD alone
group group group group
and and and and
those those those those
with with with with
diabetes. diabetes. diabetes. diabetes.
382
small to allow accurate assessment. Furthermore, the most common gluten related neurological finding is ataxia secondary to cerebellar damage which was not specifically measured in this study [6,11,12]. Changes in lipid levels have been described previously in patients with CD with and without Type 1 diabetes [4,14e16]. In the present study, patients with Type 1 diabetes and potential CD had significantly higher total cholesterol compared to patients with Type 1 diabetes and CD but not HDL cholesterol. This may reflect subtle changes in the small intestine which were not detected on routine duodenal biopsy and standard histopathology techniques. There are limitations to this study. The main issue is the study size is small and therefore subtle changes may therefore have been missed. The prevalence of potential CD also seems high suggesting a potential for bias. The cohort that these individuals were selected from had an overall prevalence of CD of 3.3% compared to 2.2% with potential CD but selected only those with newly diagnosed CD. The mechanism by which CD increases the risk of microvascular disease is unclear and is likely to be multifactorial. Unrecognised CD is associated with raised homocysteine levels probably due to deficiency in folic acid and other B vitamins, which is a risk factor for endothelial dysfunction [22]. A recent study showed that supplementation of vitamin B in CD was associated with significantly lower homocysteine levels potentially reducing the risk of vascular disease [23]. Further studies are required to determine whether early detection of CD in Type 1 diabetes and treatment with a GFD affects development of microvascular complications. Longitudinal follow up of the present study group over several years may also provide further insight into the interplay between Type 1 diabetes and CD. Although our cohort is small we would consider this to be novel pilot data, the natural progression of this work would be a larger multicentre study population and corresponding control group. In summary, this is the first study to assess the effects of potential CD in patients with Type 1 diabetes. It appears that in Type 1 diabetes, enteropathy as well as antibody positivity is required to increase diabetes related complications. Authors contributions DSS conceived the study. JSL, ST, MH, ADH and DSS wrote the protocol and were clinical investigators. JSL and DSS wrote the manuscript. JSL, ST, MH, ADH and DSS edited the final manuscript. Declaration of competing interests Nothing to declare. Acknowledgements This study was funded by the Bardhan Research and Education Trust of Rotherham and Solvay. This data has been
J.S. Leeds et al.
presented in abstract form at the British Society of Gastroenterology annual scientific meeting as a poster presentation. References [1] Page SR, Lloyd CA, Hill PG, Peacock I, Holmes GKT. The prevalence of coeliac disease in adult diabetes mellitus. Q J Med 1994;87: 631e7. [2] De Vitis I, Ghirlanda G, Gasbarrini G. Prevalence of coeliac disease in type I diabetes: a multicentre study. Acta Paediatr Suppl 1996; 412:56e7. [3] Sjöberg K, Eriksson KF, Brendberg A, Wassmuth R, Eriksson S. Screening for coeliac disease in adult insulin-dependent diabetes mellitus. J Intern Med 1998;243:133e40. [4] Leeds JS, Hopper AD, Hadjivassiliou M, Tesfaye S, Sanders DS. High prevalence of microvascular complications in adults with type 1 diabetes and newly diagnosed celiac disease. Diabetes Care 2011; 34(10):2158e63. [5] Luostarinen L, Himanen SL, Luostarinen M, Collin P, Pirttilä T. Neuromuscular and sensory disturbances in patients with well treated coeliac disease. J Neurol Neurosurg Psychiatry 2003;74: 490e4. [6] Hadjivassiliou M, Grunewald RA, Chattopadhyay AK, DaviesJones GAB, Gibson A, Jarratt JA, et al. Clinical, radiological, neurophysiological and neuropathological characteristics of gluten ataxia. Lancet 1998;347:369e71. [7] Jackson JR, Eaton WW, Cascella NG, Fasano A, Kelly DL. Neurologic and psychiatric manifestations of celiac disease and gluten sensitivity. Psychiatr Q 2012;83(1):91e102. [8] Ferguson A, Arranz E, O’Mahony S. Clinical and pathological spectrum of coeliac disease e active, silent, latent, potential. Gut 1993;34:150e1. [9] Troncone R, Greco L, Mayer M, Paparo F, Caputo N, Micillo M, et al. Latent and potential coeliac disease. Acta Paediatr Suppl 1996;412: 10e4. [10] Troncone R, Franzese A, Mazzarella G, Paparo F, Auricchio R, Coto I, et al. Gluten sensitivity in a subset of children with insulin dependent diabetes mellitus. Am J Gastroenterol 2003;98(3): 590e5. [11] Hadjivassiliou M, Davies-Jones GA, Sanders DS, Grunewald RA. The dietary treatment of gluten ataxia. J Neurol Neurosurg Psychiatry 2003;74:1221e4. [12] Hadjivassiliou M, Boscolo S, Davies-Jones GA, Grunewald RA, Not T, Sanders DS, et al. The humoral response in the pathogenesis of gluten ataxia. Neurology 2002;58:1221e6. [13] Ludvigsson JF, Montgomery SM, Ekbom A, Brandt L, Granath F. Small-intestinal histopathology and mortality risk in celiac disease. J Am Med Assoc 2009;302(11):1171e8. [14] West J, Logan RF, Card TR, Smith C, Hubbard R. Risk of vascular disease in adults with diagnosed coeliac disease: a populationbased study. Aliment Pharmacol Ther 2004;20(1):73e9. [15] West J, Logan RF, Hill PG, Khaw KT. The iceberg of celiac disease: what is below the waterline? Clin Gastroenterol Hepatol 2007; 5(1):59e62. [16] Lewis NR, Sanders DS, Logan RF, Fleming KM, Hubbard RB, West J. Cholesterol profile in people with newly diagnosed coeliac disease: a comparison with the general population and changes following treatment. Br J Nutr 2009;102(4):509e13. [17] Pitocco D, Giubilato S, Martini F, Zaccardi F, Pazzano V, Manto A, et al. Combined atherogenic effects of celiac disease and type 1 diabetes mellitus. Atherosclerosis 2011;217(2):531e5. [18] American Diabetes Association Position Statement. Diagnosis and classification of diabetes mellitus. Diabetes Care 2007;30(Suppl. 1):S42e7. [19] Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol 1999;11:1185e94. [20] Clements C, Prince CB, Prasad S, Aralikatti S, Swindlehurst H, Clearkin LG. National screening programme for diabetic retinopathy. BMJ 2002;324:849. [21] Chin RL, Sander HW, Brannagan TH, Green PH, Hays AP, Alaedini A, et al. Celiac neuropathy. Neurology 2003;60(10):1581e5.
Type 1 diabetes mellitus and potential coeliac disease [22] Saibeni S, Lecchi A, Meucci G, Cattaneo M, Tagliabue L, Rondonotti E, et al. Prevalence of hyperhomocysteinemia in adult gluten-sensitive enteropathy at diagnosis: role of B12, folate, and genetics. Clin Gastroenterol Hepatol 2005;3(6):574e80.
383 [23] Hadithi M, Mulder CJ, Stam F, Azizi J, Crusius JB, Peña AS, et al. Effect of B vitamin supplementation on plasma homocysteine levels in celiac disease. World J Gastroenterol 2009;15(8): 955e60.
Available online at www.sciencedirect.com
Nutrition, Metabolism & Cardiovascular Diseases journal homepage: www.elsevier.com/locate/nmcd
Potential coeliac disease in Type 1 diabetes mellitus: Does a positive antibody lead to increased complications? J.S. Leeds a,*, A.D. Hopper a, M. Hadjivassiliou b, S. Tesfaye c, D.S. Sanders a a
Gastroenterology & Liver Unit, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK Department of Neurology, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK c Department of Diabetes, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK b
Received 17 November 2012; received in revised form 12 August 2013; accepted 3 September 2013 Available online 23 October 2013
KEYWORDS Type 1 diabetes; Glycaemic control; Coeliac disease; Complications
Abstract Background and aims: Coeliac disease (CD) is more common in people with Type 1 diabetes and is associated with poorer glycaemic control, lipid profiles, nephropathy and retinopathy. Potential CD (positive serology but normal duodenal biopsy) is associated with neuropathy but patients with coexisting Type 1 diabetes were excluded. The aim was to determine whether potential CD is associated with increased microvascular complications in patients with Type 1 diabetes. Methods and results: Four groups were recruited; 1) patients with Type 1 diabetes and potential CD, 2) patients with Type 1 diabetes and newly identified CD, 3) patients with Type 1 diabetes alone and 4) patients with CD alone. Glycaemic control, quality of life, lipid profile and microvascular complication rates were examined. As many as 76 individuals were included in the study: 22 in group 1, 14 in group 2, 24 in group 3 and 16 in group 4. There were no differences in age, gender, BMI and diabetes duration between the groups. Patients in group 1 had significantly lower total cholesterol compared to group 3 (p Z 0.003) but higher than group 2 (p Z 0.027). There were no significant differences in HbA1c, HDL cholesterol, cholesterol:HDL ratio, creatinine, quality of life scores or prevalence of neuropathy between individuals in group 1 and the other groups. Conclusions: This is the first study to assess the effects of potential CD in patients with Type 1 diabetes. It appears that an enteropathy is required as well as antibody positivity in order to increase the risk of diabetes related complications. This pilot data requires further longitudinal validation. ª 2013 Elsevier B.V. All rights reserved.
Introduction The association between coeliac disease (CD) and Type 1 diabetes mellitus has been well described with the prevalence of CD ranging between 3 and 5% in the largest Abbreviations: CD, coeliac disease; GFD, gluten free diet; DAFNE, Dose Adjustment for Normal Eating; HbA1c, glycated haemoglobin; IgA, immunoglobulin A; HDL, high density lipoprotein; CASE, computed aided sensory evaluator; BMI, body mass index. * Corresponding author. Room 2.39, Ashgrove House, Department of Gastroenterology, Aberdeen Royal Infirmary, Aberdeen AB25 2ZN, UK. Tel.: þ44 114 2261179; fax: þ44 114 2712692. E-mail addresses: [email protected], [email protected] (J.S. Leeds). 0939-4753/$ - see front matter ª 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.numecd.2013.09.005
studies [1e3]. Recently our group has shown that patients with Type 1 diabetes mellitus and newly identified CD have worse glycaemic control and lipid profiles as well as higher prevalence of microvascular complications. The underlying mechanism was not investigated by this study but the effect upon glycaemic control is likely to be responsible [4]. CD, in the absence of diabetes, is also associated with neurological manifestations with one study showing changes in electroneuromyography even in well controlled CD [5]. These neurological complications can also occur in patients with positive coeliac serology in the absence of enteropathy. Hadjivassiliou et al. looked at gluten ataxia
Type 1 diabetes mellitus and potential coeliac disease
patients and showed that only 13/28 patients had gross histological changes in the duodenum [6]. This offers insight into the role of auto-antibodies as being part of the pathophysiological mechanism. This is separate from gluten sensitivity which is a gluten-mediated immune reaction and gluten allergic reactions which are IgE mediated. Individuals with gluten sensitivity are believed to have a different type of immune mediated reaction and do not have enteropathy but can test positive to antibodies to gliadin [7]. The implications of potential CD to the individual are still unknown as it may be that in time they will develop the histological changes of CD [8,9]. Support for this hypothesis comes from an Italian study of children with Type 1 diabetes, where histological progression was demonstrable following the rectal instillation of gliadin [10]. In gluten ataxia it has been demonstrated that these neurological complications can be arrested or reversed by the institution of a gluten free diet (GFD) [11]. The pathophysiological changes appear to lie in the humoural immune response and a GFD appears to be beneficial [11,12]. Finally recent Swedish work has suggested that individuals with a positive antibody and no enteropathy have an increased mortality when compared to controls [13]. Theoretically there may be benefits to having CD with studies in non-diabetic CD cohorts showing a reduced risk of ischaemic heart disease and hypertension as well as lower cholesterol levels [14,15]. Our previous study showed that patients with Type 1 diabetes and CD had significantly lower cholesterol levels prior to commencement of a GFD [4]. A recent study showed that following a GFD, there was no change in total cholesterol but HDL cholesterol increased by an average of 12% [16]. However, this issue is of more importance to those patients who are already at increased cardiac risk such as those with preexisting diabetes. In support of this, one study has shown that patients with CD and Type 1 diabetes have a higher rate of subclinical atherosclerosis compared to either condition alone [17]. There is no data in patients with Type 1 diabetes and potential CD particularly with respect to glycaemic control, lipid profiles and microvascular complications. The aim of this study was to examine the effect of potential CD on important outcomes in patients with Type 1 diabetes. Methods Study groups 4 specific groups were identified and studied. Group 1: patients with Type 1 diabetes and potential CD (positive antibodies but normal duodenal biopsy), group 2: patients with Type 1 diabetes and newly identified CD, group 3: patients with Type 1 diabetes alone and group 4: patients with newly diagnosed CD but without diabetes. Participants with type 1 diabetes The Sheffield Diabetes Centre operates at both the Royal Hallamshire Hospital and the Northern General Hospital.
379
This covers a population of approximately 500 000 and provides tertiary referral services for the South Yorkshire region. There are approximately 2000 patients with Type 1 diabetes defined using the American Diabetes Association position statement [18]. Around 95% of patients with Type 1 diabetes are managed by the diabetes centre in secondary care. Inclusion criteria were patients with Type 1 diabetes >16 years. Exclusion criteria were age <16 years, inability to consent or diabetes other than Type 1. Where diabetes type was uncertain, the notes were reviewed with the treating consultant and a decision made concerning diabetes type. If diabetes type was still uncertain then the individual was excluded. Patients were prospectively recruited when attending for annual review, foot clinic or Dose Adjustment for Normal Eating (DAFNE) clinic. 1000 individuals with Type 1 diabetes were previously recruited [4] and from this cohort the study groups were identified. Participants with CD Patients with newly identified CD were recruited consecutively from the gastrointestinal outpatient’s clinic. All patients had biopsy confirmed CD with at least partial villous atrophy (using the modified Marsh criteria) [19] and underwent the neurological investigations prior to commencement on a GFD. Study group investigations Study participants completed a health questionnaire and the Short Form 36 (version 2) quality of life assessment questionnaire (QualityMetric Incorporated TM). Data were prospectively collected including age, gender, ethnicity, drug history and other medical co-morbidities. Blood was taken for HbA1c (in the diabetes cohorts), renal profile, ile (including total and HDL cholesterol), full blood count, IgA endomysial antibody, IgA anti-tissue transglutaminase antibody and total IgA level. All participants with either a positive antibody or IgA deficiency underwent duodenal biopsy to confirm the presence or absence of enteropathy. Retinopathy assessed by notes review and graded as no retinopathy, background changes, pre-proliferative changes or proliferative changes as described by the National Screening Committee for Diabetic Retinopathy [20]. Patients undergo annual retinal photography and were reviewed within 3 months of diagnosis. All groups were assessed for peripheral neuropathy by a combination of quantitative sensory threshold testing, cardiac autonomic function tests and electrophysiological testing of four nerves. Quantitative sensory threshold testing was performed using the Computer Aided Sensory Evaluator (CASE IV version 4.27) which provides cold, vibration and heat pain detection thresholds on the dorsum of the right foot. Cardiovascular autonomic function tests are measured using the O’Brien protocol. A 3-lead ECG is attached whilst
380
J.S. Leeds et al.
Table 1 Demographics and characteristics of the individuals with potential CD. Patient
Age
Sex
Symptoms
Associations
EMA þve
Anaemic
1 2 3 4 5 6
40 48 53 69 26 43
M M M M M F
Alternating bowel habit None IBS None None None
Yes No No No No Yes
No No No No No Yes
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
26 53 47 24 49 35 46 49 63 45 45 62 57 57 42 52
F M M M M M F F M M M M M M M F
None None None None None Constipation Alternating bowel habit IBS None None Alternating bowel habit None None None Diarrhoea None
Epilepsy Nil IgA deficiency Nil Nil Pernicious anaemia, hypothyroidism Nil Hypothyroidism IgA deficiency Nil Nil Pancreatitis Nil Hypothyroidism Nil Hypothyroidism IgA deficiency Nil None Osteoporosis Nil Nil
No No Yes No No No Yes No No No No No No No No Yes
No No No No No Yes No Yes No No No No No No No No
IBS Z irritable bowel syndrome, IgA Z immunoglobulin A.
lying supine. Following a rest period, measurements of the time between ‘R’ waves on the ECG are made and changes in heart rate based on RR variation over time are recorded. Degree of variation is calculated using the RR ratio and is corrected for age. RR ratios were measured at rest, during deep breathing, during the Valsalva manoeuvre and following rising from supine. Change in blood pressure from lying to standing was also measured. Autonomic dysfunction was diagnosed if two or more of these tests were abnormal. Electrophysiological testing was performed on the radial nerve, sural nerve, common peroneal nerve and the tibial nerve using surface electrodes (SYNERGY version 10.0, Oxford instruments). The nerves were subjected to supramaximal stimulation to allow measurement of the amplitude and latency.
Results
Statistical analyses were performed using SPSS v17.0. To determine whether there were any differences between the averages in each group either a KruskaleWallis test (nonparametric) or a one way ANOVA (parametric) was performed. Comparison between groups was performed using the median values and the Mann Whitney U test or the means and the Students t test. A p value less than 0.05 was considered significant.
76 individuals were included in the study. There were 22 individuals with Type 1 diabetes and potential CD (group 1: positive antibody but normal biopsy), 14 individuals with newly identified CD and Type 1 diabetes (group 2), 24 individuals with Type 1 diabetes alone (group 3) and 16 with CD alone (group 4). The demographics and characteristics of the individuals with potential CD are shown in Table 1. There were no significant differences between the groups with respect to age, diabetes duration, body mass index (BMI), cholesterol:HDL ratio, triglyceride levels or creatinine. There were no significant differences in HbA1c, HDL cholesterol, cholesterol:HDL ratio or creatinine between individuals in group 1 and the other groups. Total cholesterol was significantly lower in group 1 compared to group 3. As previously shown [4], group 2 had significantly higher HbA1c compared to group 3, and significantly lower total cholesterol compared to both groups 1 and 3 (summarised in Table 2). There were no significant differences in quality of life scores between any of the groups. With respect to other autoimmune conditions, 4/22 in group 1, 4/14 in group 2, 4/24 in group 3 and 1/16 in group 4 had an autoimmune condition previously recognised (p Z 0.56). In group 2 there were 2 individuals with multiple autoimmune conditions; one with Type 1 diabetes, CD, Addisons and hyperthyroidism, one with Type 1 diabetes, CD, dermatitis herpetiformis and pernicious anaemia.
Ethical approval
Results of peripheral nerve assessment
Ethical approval was obtained from the South Sheffield Research Ethics Committee and written informed consent obtained from all individuals.
There were no differences in vibration, cold or heat pain detection levels between all of the groups studied. None of those in group 4 had any abnormalities in the autonomic
Statistical analysis
Type 1 diabetes mellitus and potential coeliac disease
381
Table 2 Summary of values for all variables in all groups with Type 1 diabetes. Variable
Group 1 CD potential
Group 2 CDDM
Group 3 type 1 DM only
p Value
Age (years) Duration (years) BMI (Kg/m2) HbA1c (%) (mmol/mol) Cholesterol HDL Chol:HDL ratio Triglycerides Creatinine
46.9 24 28.4 8.0 64 4.8 1.34 3.1 0.9 79.5
44.2 22.5 25.4 8.2 66 4.05 1.1 3.4 0.85 78.5
46.6 28.5 24.2 7.5 58 5.1 1.56 2.8 0.75 83.0
(21e71) (4e71) (17.3e32.6) (1.93)
NS NS NS NSa
(1.4) (0.8) (1.8) (0.6) (10.3)
0.011b 0.04c NS NS NS
(24e69) (2e40) (19.7e34.4) (1.38) (1.38) (0.7) (1.3) (1.0) (27.0)
(21e82) (5e44) (20.1e36.9) (2.5) (1.43) (0.6) (1.7) (0.8) (35.0)
HbA1c Z glycosylated haemoglobin, HDL Z high density lipoprotein. Values in brackets are interquartile ranges. a KruskaleWallis test. CDDM HbA1c vs. Type 1 DM only Z 0.05. b KruskaleWallis test. CDDM potential cholesterol vs. CDDM Z 0.027, vs. Type 1 DM only Z 0.003. c KruskaleWallis test. CDDM HDL vs. Type 1 DM only Z 0.017.
function tests and there were no differences in the frequency of abnormalities between groups 1 and 3. With respect to peripheral nerve conduction studies there were no significant differences in amplitude or conduction velocity when comparing between groups 1 and 3 but there was a significant difference seen in conduction velocity when comparing any of the groups with diabetes to group 4. This is summarised in Table 3. Retinopathy Overall, there was no difference in the prevalence of retinopathy between the cases and controls. When looking at higher grades of retinal disease (pre-proliferative or higher) there was a significantly higher prevalence of more advanced retinopathy in group 2 compared to group 3 (p Z 0.02). However, there was no significant difference between group 1 and either group 2 or 3. Discussion This is the first study to examine the effect of potential CD in people with known Type 1 diabetes with respect to
important diabetes related outcomes. Although we have previously reported that newly identified CD and Type 1 diabetes patients had significantly higher HbA1c compared to controls with Type 1 diabetes alone [4] these difference were not observed when assessing those patients with potential CD and Type 1 diabetes. Furthermore when we looked at lipid profiles, renal function, peripheral nerve function and retinal changes no significant differences were observed. As in the previous study, there was a higher prevalence of more advanced retinal lesions in those individuals with both Type 1 diabetes and CD but this was not observed in those with potential CD and Type 1 diabetes. We have previously shown that patients with both CD and Type 1 diabetes have a higher prevalence of microvascular complications [4]. The mechanism for this was not specifically assessed but was hypothesised to be due to the adverse effect on glycaemic control. The present study suggests that positive antibodies are insufficient to cause a noticeable effect on important outcomes and that enteropathy is also required. We also found no evidence of neuropathy in patients with newly identified CD alone. However, as the prevalence of neurological abnormalities in cohorts with CD ranges between 2 and 8% [6,21], the size of the CD group in the present study is probably too
Table 3 Results of mean nerve conduction studies in each group. Peripheral nerve Radial Amplitude (uV) Velocity (m s1) Sural Amplitude (uV) Velocity (m s1) Common peroneal Amplitude (uV) Velocity (m s1) Tibial Amplitude (uV) Velocity (m s1)
CDDM
CD potential
Type 1 DM alone
CD alone
p
28.2 (6.4) 49.2 (3.8)
22.2 (3.5) 51.5 (2.1)
32.0 (4.6) 51.2 (2.2)
36.2 (6.7) 57.9 (1.2)
NS 0.052a
11.1 (2.4) 38.0 (3.1)
11.2 (1.3) 39.6 (1.4)
12.1 (3.9) 37.4 (0.7)
15.8 (2.5) 44.0 (1.4)
NS 0.05b
4.3 (0.7) 41.7 (2.0)
6.8 (2.5) 39.5 (1.4)
3.6 (0.8) 36.7 (2.5)
5.3 (0.6) 49.6 (0.9)
NS 0.001c
11.1 (2.0) 41.9 (2.0)
11.8 (3.9) 37.4 (1.5)
8.7 (1.8) 37.8 (1.9)
11.2 (1.9) 45.8 (0.8)
NS 0.001d
Results in brackets are standard errors. a Differences only seen between CD alone b Differences only seen between CD alone c Differences only seen between CD alone d Differences only seen between CD alone
group group group group
and and and and
those those those those
with with with with
diabetes. diabetes. diabetes. diabetes.
382
small to allow accurate assessment. Furthermore, the most common gluten related neurological finding is ataxia secondary to cerebellar damage which was not specifically measured in this study [6,11,12]. Changes in lipid levels have been described previously in patients with CD with and without Type 1 diabetes [4,14e16]. In the present study, patients with Type 1 diabetes and potential CD had significantly higher total cholesterol compared to patients with Type 1 diabetes and CD but not HDL cholesterol. This may reflect subtle changes in the small intestine which were not detected on routine duodenal biopsy and standard histopathology techniques. There are limitations to this study. The main issue is the study size is small and therefore subtle changes may therefore have been missed. The prevalence of potential CD also seems high suggesting a potential for bias. The cohort that these individuals were selected from had an overall prevalence of CD of 3.3% compared to 2.2% with potential CD but selected only those with newly diagnosed CD. The mechanism by which CD increases the risk of microvascular disease is unclear and is likely to be multifactorial. Unrecognised CD is associated with raised homocysteine levels probably due to deficiency in folic acid and other B vitamins, which is a risk factor for endothelial dysfunction [22]. A recent study showed that supplementation of vitamin B in CD was associated with significantly lower homocysteine levels potentially reducing the risk of vascular disease [23]. Further studies are required to determine whether early detection of CD in Type 1 diabetes and treatment with a GFD affects development of microvascular complications. Longitudinal follow up of the present study group over several years may also provide further insight into the interplay between Type 1 diabetes and CD. Although our cohort is small we would consider this to be novel pilot data, the natural progression of this work would be a larger multicentre study population and corresponding control group. In summary, this is the first study to assess the effects of potential CD in patients with Type 1 diabetes. It appears that in Type 1 diabetes, enteropathy as well as antibody positivity is required to increase diabetes related complications. Authors contributions DSS conceived the study. JSL, ST, MH, ADH and DSS wrote the protocol and were clinical investigators. JSL and DSS wrote the manuscript. JSL, ST, MH, ADH and DSS edited the final manuscript. Declaration of competing interests Nothing to declare. Acknowledgements This study was funded by the Bardhan Research and Education Trust of Rotherham and Solvay. This data has been
J.S. Leeds et al.
presented in abstract form at the British Society of Gastroenterology annual scientific meeting as a poster presentation. References [1] Page SR, Lloyd CA, Hill PG, Peacock I, Holmes GKT. The prevalence of coeliac disease in adult diabetes mellitus. Q J Med 1994;87: 631e7. [2] De Vitis I, Ghirlanda G, Gasbarrini G. Prevalence of coeliac disease in type I diabetes: a multicentre study. Acta Paediatr Suppl 1996; 412:56e7. [3] Sjöberg K, Eriksson KF, Brendberg A, Wassmuth R, Eriksson S. Screening for coeliac disease in adult insulin-dependent diabetes mellitus. J Intern Med 1998;243:133e40. [4] Leeds JS, Hopper AD, Hadjivassiliou M, Tesfaye S, Sanders DS. High prevalence of microvascular complications in adults with type 1 diabetes and newly diagnosed celiac disease. Diabetes Care 2011; 34(10):2158e63. [5] Luostarinen L, Himanen SL, Luostarinen M, Collin P, Pirttilä T. Neuromuscular and sensory disturbances in patients with well treated coeliac disease. J Neurol Neurosurg Psychiatry 2003;74: 490e4. [6] Hadjivassiliou M, Grunewald RA, Chattopadhyay AK, DaviesJones GAB, Gibson A, Jarratt JA, et al. Clinical, radiological, neurophysiological and neuropathological characteristics of gluten ataxia. Lancet 1998;347:369e71. [7] Jackson JR, Eaton WW, Cascella NG, Fasano A, Kelly DL. Neurologic and psychiatric manifestations of celiac disease and gluten sensitivity. Psychiatr Q 2012;83(1):91e102. [8] Ferguson A, Arranz E, O’Mahony S. Clinical and pathological spectrum of coeliac disease e active, silent, latent, potential. Gut 1993;34:150e1. [9] Troncone R, Greco L, Mayer M, Paparo F, Caputo N, Micillo M, et al. Latent and potential coeliac disease. Acta Paediatr Suppl 1996;412: 10e4. [10] Troncone R, Franzese A, Mazzarella G, Paparo F, Auricchio R, Coto I, et al. Gluten sensitivity in a subset of children with insulin dependent diabetes mellitus. Am J Gastroenterol 2003;98(3): 590e5. [11] Hadjivassiliou M, Davies-Jones GA, Sanders DS, Grunewald RA. The dietary treatment of gluten ataxia. J Neurol Neurosurg Psychiatry 2003;74:1221e4. [12] Hadjivassiliou M, Boscolo S, Davies-Jones GA, Grunewald RA, Not T, Sanders DS, et al. The humoral response in the pathogenesis of gluten ataxia. Neurology 2002;58:1221e6. [13] Ludvigsson JF, Montgomery SM, Ekbom A, Brandt L, Granath F. Small-intestinal histopathology and mortality risk in celiac disease. J Am Med Assoc 2009;302(11):1171e8. [14] West J, Logan RF, Card TR, Smith C, Hubbard R. Risk of vascular disease in adults with diagnosed coeliac disease: a populationbased study. Aliment Pharmacol Ther 2004;20(1):73e9. [15] West J, Logan RF, Hill PG, Khaw KT. The iceberg of celiac disease: what is below the waterline? Clin Gastroenterol Hepatol 2007; 5(1):59e62. [16] Lewis NR, Sanders DS, Logan RF, Fleming KM, Hubbard RB, West J. Cholesterol profile in people with newly diagnosed coeliac disease: a comparison with the general population and changes following treatment. Br J Nutr 2009;102(4):509e13. [17] Pitocco D, Giubilato S, Martini F, Zaccardi F, Pazzano V, Manto A, et al. Combined atherogenic effects of celiac disease and type 1 diabetes mellitus. Atherosclerosis 2011;217(2):531e5. [18] American Diabetes Association Position Statement. Diagnosis and classification of diabetes mellitus. Diabetes Care 2007;30(Suppl. 1):S42e7. [19] Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol 1999;11:1185e94. [20] Clements C, Prince CB, Prasad S, Aralikatti S, Swindlehurst H, Clearkin LG. National screening programme for diabetic retinopathy. BMJ 2002;324:849. [21] Chin RL, Sander HW, Brannagan TH, Green PH, Hays AP, Alaedini A, et al. Celiac neuropathy. Neurology 2003;60(10):1581e5.
Type 1 diabetes mellitus and potential coeliac disease [22] Saibeni S, Lecchi A, Meucci G, Cattaneo M, Tagliabue L, Rondonotti E, et al. Prevalence of hyperhomocysteinemia in adult gluten-sensitive enteropathy at diagnosis: role of B12, folate, and genetics. Clin Gastroenterol Hepatol 2005;3(6):574e80.
383 [23] Hadithi M, Mulder CJ, Stam F, Azizi J, Crusius JB, Peña AS, et al. Effect of B vitamin supplementation on plasma homocysteine levels in celiac disease. World J Gastroenterol 2009;15(8): 955e60.