- Email: [email protected]
Malignancies and mortality in patients with coeliac disease and dermatitis herpetiformis: 30-year population-based study
Digestive and Liver Disease 38 (2006) 374–380
Alimentary Tract
Malignancies and mortality in patients with coeliac disease and dermatitis herpetiformis: 30-year population-based study M. Viljamaa a,b , K. Kaukinen a,b , E. Pukkala d , K. Hervonen a,c , T. Reunala a,c , P. Collin a,b,∗ a
Departments of Gastroenterology and Alimentary Tract Surgery, Tampere University Hospital, P.O. Box 2000, FIN-33521 Tampere, Finland b Medical School, University of Tampere, FIN-33014 Tampere, Finland c Department of Dermatology, Tampere University Hospital, P.O. Box 2000, FIN-33521 Tampere, Finland d Finnish Cancer Registry, Institute for Statistical and Epidemiological Cancer Research, Liisankatu 21 B, FIN-00170 Helsinki, Finland Received 5 December 2005; accepted 6 March 2006 Available online 18 April 2006
Abstract Background and aim. To assess the long-term risks of malignant diseases and mortality in patients with coeliac disease and dermatitis herpetiformis in a centre, where the prevalence of these diseases is high. The risks have probably been overestimated, as patients with subtle forms have earlier remained undetected. Patients. The study comprised 17,245 person-years of follow-up in 1147 patients. Methods. The observed numbers of malignancies and causes of deaths were assessed, and compared to those expected, and standardised incidence ratio and standardised mortality ratio given. Results. The occurrence of all malignant conditions was equal to that in the population both in coeliac disease and dermatitis herpetiformis: standardised incidence ratios of 1.2 (95% confidence intervals 0.9–1.5) and 1.0 (0.6–1.5), respectively. Five patients with coeliac disease and seven with dermatitis herpetiformis had developed non-Hodgkin lymphoma; standardised incidence ratios of 3.2 (1.0–7.5) and 6.0 (2.4–12.4), respectively. Four patients with coeliac disease and one with dermatitis herpetiformis had enteropathy-associated T-cell lymphoma, associated with inadequate dietary compliance. Mortality was increased (standardised mortality ratio 1.26; 1.00–1.55) in coeliac disease, but decreased in dermatitis herpetiformis (standardised mortality ratio 0.52; 0.36–0.72). Conclusion. The overall prognosis in our patients was good. Non-Hodgkin lymphoma emerged in patients with undiagnosed or poorly treated coeliac disease. The mortality rate in dermatitis herpetiformis was even lower than in the population. Our data support the early diagnosis and dietary treatment of these conditions. © 2006 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: Cancer; Coeliac disease; Dermatitis herpetiformis; Mortality
1. Introduction Coeliac disease (CD) is classified as a gluten-induced disease causing small bowel mucosal damage with villous shortening, crypt hyperplasia and inflammation, which recovers on a gluten-free diet [1]. Dermatitis herpetiformis (DH) is a blistering, itching skin disease, where the diagnosis is based on the demonstration of granular IgA deposits in the dermal ∗
Corresponding author. Tel.: +358 3 3116 7869; fax: +358 3 3551 8402. E-mail address: [email protected] (P. Collin).
papillae of uninvolved perilesional skin as observed by direct immunofluorescence [2,3]. About 75% of our patients with DH have mucosal damage identical to that in CD, and the rest have mucosal inflammation, again consistent with CD [4]. We have, within the last 20 years, taken action to augment the diagnostics of CD and DH [5]. The disorder has been considered even in the presence of mild or atypical symptoms, and serologic tests have been widely employed in patients with other autoimmune diseases and in first-degree relatives of patients with CD and DH. As a result, we have recorded a clinical prevalence of CD and DH in our area as high as 0.3%
1590-8658/$30 © 2006 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2006.03.002
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
already in 1997 [6], two to three times than usually reported [7–9]. Irrespective of the plausible symptoms at diagnosis, our policy has been to treat all the patients similarly by glutenfree diet. From our previous studies, we know that over 80% of our patients follow strict gluten-free diet, and only 2% use normal gluten-containing diet, the remainder having some dietary lapses [10,11]. CD is associated with an increased risk of enteropathyassociated non-Hodgkin’s lymphoma (NHL) [12–16]. Gluten-free diet treatment seems to protect patients from NHL [13]. Oesophageal, colorectal and small bowel cancers have been reported to occur in CD [14,15,17], whereas the risk of breast and lung cancer may be reduced [17]. Due to prolonged diagnostic delay and poor dietary compliance, NHL has thought to be the main cause of the two-fold increased mortality in CD, but when the diagnosis of CD has been made early enough, the additional risk seems to be lower or even negligible [18]. In a study by Swerdlow et al. [19] the mortality rate in DH was not increased; to our knowledge, no other studies have been carried out on mortality in patients with DH. Our large prospective series of patients with CD and DH sampled during 30 years in a high-prevalence area enables us to determine whether the malignancy and mortality rates are really increased in these patients, when compliance with gluten-free dietary treatment has been shown to be good.
2. Materials and methods This population-based study comprised all the 1147 patients; 781 with CD (women 68%) and 366 with DH (48%), diagnosed at the Tampere University Hospital area between 1960 and 2000. The diagnosis of CD was based on villous atrophy and crypt hyperplasia shown on a small bowel biopsy [1]; 79% had been detected because of gastrointestinal symptoms or malabsorption, and 21% by serologic screening in risk groups. The diagnosis of DH was based on the demonstration of granular IgA deposit finding in skin biopsy [4].
375
CD was diagnosed at the age of 39 years (median, range 1–84 years) and DH at 38 years (5–84 years); the respective ages at the time of follow-up were 53 (9–89) and 57 (12–93). The number of person-years at risk was 17,245; 10,956 for CD and 6289 for DH. Malignant diseases were identified by linking the personal identification codes with records from the nationwide database of the Finnish Cancer Registry, which includes more than 99% of incident cases diagnosed in Finland since 1953 [20]. Causes of death were obtained from the files of Statistics Finland. Follow-up commenced on January 1st, 1971 for patients whose diagnosis had been established before 1970 and in the rest from the year of diagnosis. The end-points were 31st December 2002, or the date of emigration (n = 2) or death. The expected figures for malignant diseases and deaths were calculated by multiplying the observed person-years of follow-up by the incidence rate of each malignant disease and the mortality rate in the respective sex, age and calendar period in the Finnish population. Standardised incidence ratio (SIR) was used to measure the relative risk of cancer; that is, the ratio of observed-to-expected cancers. For subgroup analysis, we chose the commonest cancers in CD and DH in Finland, and cancers for which an increased or decreased risk has been reported in CD or DH. The analyses were stratified according to the time since diagnosis. Similarly, the standardised mortality ratio (SMR) was used to assess the relative risk of death. Exact 95% confidence intervals (CI) were computed by assuming that the observed figures followed Poisson distribution. Institution’s ethics committee approved the study.
3. Results 3.1. Malignant diseases The total risk of malignant diseases did not differ from that in the population in general (Table 1). There was a significantly increased risk of NHL in both CD and DH; SIR
Table 1 Observed (O) and expected (E) cancers and SIR for malignant diseases with 95% CI in patients with CD and DH Site of malignant disease
CD O
E
SIR (95% CI)
O
E
SIR (95% CI)
All cancers
49
41.9
1.2 (0.9–1.5)
20
20.7
1.0 (0.6–1.5)
NHL Hodgkin’s lymphoma
5 0
1.6 0.3
3.2 (1.0–7.5)a 0.0 (0.0–14.4)
7 0
1.2 0.2
6.0 (2.4–12.4)a 0.0 (0.0–21.6)
Digestive system Stomach Colon and rectum Oesophagus Small bowel cancer
10 2 4 0 0
8.0 1.2 3.7 0.3 0.1
1.3 (0.6–2.3) 1.2 (0.2–4.5) 1.1 (0.3–2.8) 0.0 (0.0–10.7) 0.0 (0.0–80.1)
6 3 2 0 0
6.6 2.1 2.9 0.3 0.04
0.9 (0.3–2.0) 2.1 (0.4–6.3) 0.7 (0.1–2.5) 0.0 (0.0–11.7) 0.0 (0.0–104.0)
2 9
3.4 10.2
0.6 (0.1–2.1) 0.9 (0.4–1.7)
1 5
3.5 4.71
0.3 (0.0–1.6) 1.06 (0.34–2.47)
Lung and trachea Breast a
Significantly increased risk of NHL.
DH
376
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
Table 2 Characteristics of patients with CD and DH with NHL Number
Gender
Patients with CD 1 F
Dg of CD/DH; year (age)
Dg of NHL; year (age)
Organ involved
Histology of NHL
Adherence to gluten-free diet
Villous atrophy at dg
Time and cause of death
1987 (58)
1995 (66)
Ileum
EATL
1987–1990 no diet, 1991–1995 strict 1989–1990 no diet, 1991–1992 strict 1990–1995 no diet, 1995–1996 strict 1994–1995 strict 1997 strict
Yes
1995; NHL
Yes Yes
1997; pulmonary embolism 1996; NHL
Yes Yes
1999; pneumonia 2003; NHL
Yes
Alive
Yes
1995; obscure
2
M
1989 (65)
1992 (68)
Jejunum
EATL
3
M
1990 (51)
1996 (57)
Ileum
EATL
4 5
F M
1994 (53) 1997 (73)
1995 (54) 1997 (73)
Jejunum Jejunum
DLBCL EATL
1967 (16)
1999 (48)
Stomach
DLBCL
Patients with DH 1 M
Yes
Alive
EATL DLBCL
1964–1988 no diet, 1988–1999 strict 1971–1975 no diet, 1976–1979 partial, 1980–1992 strict 1972–1975 no diet, 1975–1979 strict 1973–1976 partial 1978–1983 strict
Yes No
1976; NHL 1983; NHL
FL
1978–1980 strict
Yes
Alive
DLBCL
1982–1996 partial
No
2003; NHL
2
M
1971 (58)
1992 (79)
Penis
DLBCL
3
M
1972 (30)
1979 (37)
Diffuse
4 5
M F
1973 (47) 1978 (64)
1976 (50) 1983 (69)
6
F
1978 (40)
1980 (42)
7
F
1981 (31)
1996 (46)
Mesenterium, colon Ileum Mesenterium, cervical lymph node Inguinal lymph node Mesenterium, ovario
EATL, enteropathy-associated T-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma. Partial diet: the diet was mostly gluten-free, but dietary transgressions occurred at least every month.
3.2 and 6.0, respectively. Of the five cases with CD and NHL, four were enteropathy-associated T-cell lymphomas (EATL). NHL was diagnosed virtually simultaneously with the diagnosis of CD in two; the other three had had dietary transgressions (Table 2). In DH, only one of the seven with NHL had EATL. Two (29%) of these seven had no small intestinal villous atrophy at the time of diagnosis. Five had at least partial dietary transgression after the introduction of gluten-free diet. The diagnosis of DH had been made in all seven patients with subsequent NHL in 1981 or earlier (Table 2).
3.2. Mortality rate The mortality rate during the follow-up period was significantly increased in CD: SMR 1.26; 95% CI 1.00–1.55. Within the first 5 years of diagnosis, SMR was 1.42 and 1.18 thereafter; this difference was not statistically significant (Table 3). In DH, the mortality was significantly lower than expected (0.52; 0.36–0.72) and the risk reduction was similar before and after 5 years of the diagnosis (Table 3). At the end of follow-up, all five patients with CD and lymphoma and four out of seven with DH and lymphoma had died (Table 2).
Table 3 Mortality during the follow-up time in patients with CD and DH after the diagnosis Follow-up time after diagnosis of CD and DH
Person-years of follow-up
Deaths Observed
Expected
Total follow-up time CD DH
10955.6 6288.6
85 34
0–5 years CD DH
3728.0 1774.0
≥5 years CD DH
7227.6 4514.3
SMR, standardised mortality ratio; CI, confidence interval. a Statistically significant difference compared to general population.
SMR
95% CI
67.6 65.2
1.26 0.52
1.00–1.55a 0.36–0.72a
31 7
21.9 14.0
1.42 0.50
0.96–2.01 0.20–1.02
54 27
45.7 51.2
1.18 0.53
0.89–1.54 0.35–0.76a
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
377
Table 4 Observed (O) and expected (E) deaths and SMR with 95% CI for patients with CD and DH Cause of death
Patients with CD (n = 781)
Patients with DH (n = 366)
O
E
SMR (95% CI)
O
E
SMR (95% CI)
Diseases, all
75
60.9
1.23 (0.97–1.54)
33
59.1
0.56 (0.38–0.78)a
Diseases of circulatory system Ischaemic heart diseases Cerebrovascular diseases Other heart diseases
34 27 1 5
28.2 16.9 6.7 2.3
1.20 (0.83–1.68) 1.59 (1.05–2.31)a 0.15 (0.00–0.83)a 2.14 (0.70–5.00)
21 13 3 3
29.3 18.1 6.5 2.5
0.72 (0.44–1.09) 0.72 (0.38–1.23) 0.46 (0.10–1.35) 1.22 (0.25–3.57)
Malignant neoplasms Lymphoproliferative/haematopoietic diseases Malignant neoplasms of stomach Malignant neoplasms of respiratory tract
17 7 2 1
17.8 1.7 1.2 3.0
0.96 (0.56–1.53) 4.12 (1.66–8.51)a 1.71 (0.21–6.18) 0.33 (0.01–1.85)
9 3 2 1
14.6 1.4 1.1 3.2
0.62 (0.28–1.17) 2.18 (0.45–6.36) 1.88 (0.23–6.79) 0.31 (0.01–1.74)
Non-malignant diseases of digestive system Endocrine diseases Dementia Other diseases of nervous system Infectious diseases Respiratory diseases Urinary diseases Other diseases (non-alcohol) Diseases associated with alcohol misuse
6 1 3 2 0 7 1 3 1
1.7 1.0 2.2 1.2 0.6 4.1 0.6 0.9 2.3
3.59 (1.32–7.82)a 1.02 (0.03–5.70) 1.34 (0.28–3.91) 1.65 (0.20–5.94) 0.00 (0.00–6.75) 1.73 (0.69–3.56) 1.69 (0.04–9.43) 3.44 (0.71–10.05) 0.44 (0.01–2.45)
0 0 2 0 0 1 0 0 0
1.6 0.8 2.7 0.9 0.5 4.5 0.6 0.6 2.2
0.00 (0.00–2.27) 0.00 (0.00–4.45) 0.74 (0.09–2.68) 0.00 (0.00–3.79) 0.00 (0.00–7.46) 0.22 (0.01–1.23) 0.00 (0.00–6.06) 0.00 (0.00–6.28) 0.00 (0.00–1.64)
External causes
8
6.7
1.20 (0.52–2.36)
1
6.1
0.16 (0.00–0.91)a
a
Statistically significant difference compared to general population.
A significant difference in mortality rates between CD and DH was observed in patients with diseases of the circulatory system. Patients with CD had an increased mortality due to ischaemic heart diseases (SMR 1.59), whereas the risk was decreased (0.72), though not significantly, in patients with DH (Table 4). A similar trend was observed in malignant neoplasms of lymphoproliferative and haematopoietic organs, and in non-malignant diseases of the digestive system (not possible to specify in detail). In DH, the mortality due to external causes was significantly lower than expected (Table 4).
4. Discussion The overall cancer risk in our patients remained at population level. The risk of NHL was increased both in patients with CD and DH; by and large in line with data published elsewhere (Table 5). In DH, SIR for NHL was even higher than in CD, but in only one of the seven cases was the organ involved with the small bowel, compared to four of the five in CD. The clinical course of patients with DH and lymphoma has been presented earlier when the risk of relatives to develop lymphoma was assessed: the study also [21] included patients outside our region; that is, in Helsinki area. Here we wanted to survey the risk of lymphoma and mortality in a single centre, and make comparisons between CD, DH and general population. We cannot explain why the patients with DH developed lymphomas other than EATL, including B-cell lymphomas. The less severe small bowel mucosal damage in DH may be the reason for the
difference. Nevertheless, NHL no longer appeared in DH patients whose diagnosis was made recently. Twenty years ago, a gluten-free diet was not considered as essential as today in DH, and the gluten intake may have contributed to the development of these lymphomas. It was impossible to obtain reliable data on each patient’s dietary history, but again, our results show unanimously that the compliance to gluten-free diet was lower in our CD and DH patients with concomitant lymphoma than in our patients in general. We did not calculate the risk of lymphoma between DH patients with normal mucosa and mucosal villous atrophy, as the number of patients in both groups would have been too small, and mucosal biopsy results were not available in each DH case. However, two (29%) out of seven patients with lymphoma and DH had normal villous architecture while on the gluten-containing diet. The percentage was comparable to what has been reported earlier in our patients with DH [22,23]. This suggests that the risk of lymphoma, other than EATL, was not dependent on the degree of villous damage in DH. In contrast to some recent reports [14,15,17], we did not observe any increased occurrence of gastrointestinal cancers other than EATL. Further studies are needed to confirm that the risk of lung cancer is reduced in CD and DH, as seen here (SIR 0.6 in CD and 0.3 in DH). Even though this did not reach statistical significance, the observation is still of note, as the same finding (SIR 0.4; 95% CI 0.1–1.2) was made in the UK [17]. It has been reported that patients with CD and DH smoke less often than people in general [24–26]. The mortality rate in the present study was somewhat lower than shown in other studies [17–19,27–29] (Table 6).
378
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
Table 5 Latest studies of the risk of lymphoma in patients with CD and DH Author (year) [ref.]
Country
Patients (n)
Patient-years of follow-up
SIR (95% CI)
Patient series
Studies on patients with CD Collin et al. (1996) [12]
Finland
383
3107
2.7 (0.1–14.8)
Askling et al. (2002) [14]
Sweden
11019
97236
6.3 (4.2–125)
Green et al. (2003) [15]
US
381
1977
5.3 (2.3–13.0)
Card et al. (2004) [16]
UK
637
5684
5.8 (1.6–14.9)
Current study
Finland
781
10956
3.2 (1.0–7.5)
Studies on patients with DH Collin et al. (1996) [12]
Finland
305
3029
10.3 (2.8–26.3)
Askling et al. (2002) [14]
Sweden
1354
14451
1.9 (0.7–4.0)
Current study
Finland
366
6289
6.0 (2.4–12.4)
Consecutive patients from one unit Hospitalised patients from countrywide in-patient register. Lymphomas within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit Consecutive patients from two units. Lymphomas within 2 years after the diagnosis of CD were excluded Consecutive patients from one unit Consecutive patients from one unit Hospitalised patients from countrywide in-patient register. Lymphomas within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit
SIR, standardised incidence ratio.
Good dietary compliance in our patients may have influenced in the better prognosis: virtually all of our patients adhere to strict gluten-free diet or use gluten-containing products only occasionally [10,21]. Surprisingly, in contrast to CD, the mortality rate in DH was significantly reduced (SMR 0.52). This low rate was seen in almost all categories, though not statistically significantly (Table 4). We well recognise that the reliability of death certificates is far from complete, but this source of
bias is probably equal in the present study series and in the population in general. Nevertheless, there was a conspicuous difference between CD and DH in mortality due to ischaemic heart diseases. In UK, the risk of myocardial infarction was observed to be 15% decreased in 3590 CD patients [30]. Similarly, decreased mortality due to ischaemic heart disease in DH (SMR 0.37; 95% CI 0.12–0.86) has been reported in UK [19]. The fat intake has been reported to be higher in Finnish CD patients than in healthy controls [31],
Table 6 The latest mortality studies in patients with CD and DH Author (year) [ref.]
Country
Patients (n)
Person-years of follow-up
SMR (95% CI)
Patient series
Studies on patients with CD Logan et al. (1989) [29] Cottone et al. (1999) [28] Corrao et al. (2001) [18] Peters et al. (2003) [27]
UK Italy Italy Sweden
653 228 1072 10032
8823 1349 6444 81182
1.9 (1.5–2.2) 3.8 (1.9–6.7) 2.0 (1.5–2.7) 2.0 (1.8–2.1)
4372
18923
1.3 (1.1–1.5)
West et al. (2004) [17]
UK
Current study
Finland
781
10956
1.12 (1.0–1.55)
Patients attending hospitals in two units Consecutive adult patients from one unit Consecutive patients from 11 units Hospitalised patients from countrywide in-patient register. Deaths within 1 year after the diagnosis of CD were excluded Patients from nationwide general practice research database. Deaths within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit
UK Finland
152 366
2288 6289
0.9 (0.6–1.2) 0.52 (0.36–0.72)
Consecutive patients from one unit Consecutive patients from one unit
Studies on patients with DH Swerdlow et al. (1993) [19] Current study
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
but we have no data on DH. On the other hand, serum cholesterol levels have been relatively low both in patients with CD and DH, even when they have adhered to a gluten-free diet [24,26]. In CD, folic acid malabsorption entails high levels of plasma homocysteine, which again may increase the risk of cardiovascular diseases [32]. This mucosal damage and consequently the folic acid malabsorption is milder in DH [33]. The mortality due to external causes in DH was significantly lower than expected, but we consider that our series were unlikely to be biased; the patients in the area had been diagnosed and remained under surveillance in a single centre and there were no drop-outs, which suggest that the DH population was representative of the population in general. The issues are altogether highly speculative, and the reasons for the decreased mortality in DH remain obscure. When comparing the studies in Tables 5 and 6, it must be noted that the study designs are different: some authors had decided to exclude the lymphomas detected within 1 or 2 years after the diagnosis of CD. We consider it essential to include these cases also, when the overall risk of lymphoma is estimated. Most lymphomas in patients with CD will be detected shortly after the diagnosis of CD or initiation of strict gluten-free diet [13,16], as also seen in Table 2. The exclusion might be justified when the protective effect of gluten-free diet is assessed. In conclusion, our long-term follow-up study showed that the risk of NHL was comparable to what has been reported elsewhere. EATL was not observed in patients with longterm adherence to a gluten-free diet. The mortality rate in CD was only marginally increased compared to the population in general, and even decreased in DH. The overall prognosis in CD and DH was thus good in our high-prevalence area with well-treated patients. Conflict of interest statement None declared.
Acknowledgements This study was supported by the Research Fund of the Finnish Coeliac Society, the Mary and George C. Ehrnrooth Foundation, the Finnish Medical Foundation, the Finnish Foundation for Paediatric Research and the Medical Research Fund of Tampere University Hospital.
References [1] Walker-Smith JA, Guandalini S, Schmitz J, Shmerling DH, Visakorpi JK. Revised criteria for diagnosis of coeliac disease. Arch Dis Child 1990;65:909–11. [2] van der Meer JB. Granular deposits of immunoglobulins in the skin of patients with dermatitis herpetiformis. An immunofluorescent study. Br J Dermatol 1969;81:493–503.
379
[3] Zone JJ, Meyer LJ, Petersen MJ. Deposition of granular IgA relative to clinical lesions in dermatitis herpetiformis. Arch Dermatol 1996;132:912–8. [4] Reunala T. Dermatitis herpetiformis: coeliac disease of the skin. Ann Med 1998;30:416–8. [5] Collin P, H¨allstr¨om O, M¨aki M, Viander M, Keyril¨ainen O. Atypical coeliac disease found with serologic screening. Scand J Gastroenterol 1990;25:245–50. [6] Collin P, Reunala T, Rasmussen M, Kyr¨onpalo S, Pehkonen E, Laippala P, et al. High incidence and prevalence of adult coeliac disease. Augmented diagnostic approach. Scand J Gastroenterol 1997;32:1129–33. [7] Midhagen G, J¨arnerot G, Kraaz W. Adult coeliac disease within a defined geographic area in Sweden. A study of prevalence and associated diseases. Scand J Gastroenterol 1988;23:1000–4. [8] Schweizer JJ, von Blomberg BME, Bueno-de Mesquita HB, Mearin ML. Coeliac disease in the Netherlands. Scand J Gastroenterol 2004;39:359–64. [9] Talley NJ, Valdovinos M, Petterson TM, Carpenter HA, Melton Jr L. Epidemiology of celiac sprue: a community-based study. Am J Gastroenterol 1994;89:843–6. [10] Viljamaa M, Collin P, Huhtala H, Sievanen H, Maki M, Kaukinen K. Is coeliac disease screening in risk groups justified? Fourteenyear follow-up with special focus on compliance and quality of life. Aliment Pharmacol Ther 2005;22:317–24. [11] Hervonen K, Hakanen M, Kaukinen K, Collin P, Reunala T. Firstdegree relatives are frequently affected in coeliac disease and dermatitis herpetiformis. Scand J Gastroenterol 2002;37:51–5. [12] Collin P, Pukkala E, Reunala T. Malignancy and survival in dermatitis herpetiformis: a comparison with coeliac disease. Gut 1996;38:528–30. [13] Holmes GKT, Prior P, Lane MR, Pope D, Allan RN. Malignancy in coeliac disease—effect of a gluten free diet. Gut 1989;30:333–8. [14] Askling J, Linet M, Gridley G, Halstensen TS, Ekstrom K, Ekbom A. Cancer incidence in a population-based cohort of individuals hospitalized with celiac disease or dermatitis herpetiformis. Gastroenterology 2002;123:1428–35. [15] Green PH, Fleischauer AT, Bhagat G, Goyal R, Jabri B, Neugut AI. Risk of malignancy in patients with celiac disease. Am J Med 2003;115:191–5. [16] Card TR, West J, Holmes GK. Risk of malignancy in diagnosed coeliac disease: a 24-year prospective, population-based, cohort study. Aliment Pharmacol Ther 2004;20:769–75. [17] West J, Logan RF, Smith CJ, Hubbard RB, Card TR. Malignancy and mortality in people with coeliac diease: population based cohort study. Br Med J 2004;329:716–9. [18] Corrao G, Corazza GR, Bagnardi V, Brusco G, Ciacci C, Cottone M, et al. Mortality in patients with coeliac disease and their relatives: a cohort study. Lancet 2001;358:356–61. [19] Swerdlow AJ, Whittaker S, Carpenter LM, English JSC. Mortality and cancer incidence in patients with dermatitis herpetiformis: a cohort study. Br J Dermatol 1993;129:140–4. [20] Teppo L, Pukkala E, Lehtonen M. Data quality and quality control of a population-based cancer registry. Experience in Finland. Acta Oncol 1994;33:365–9. [21] Hervonen K, Vornanen M, Kautiainen H, Collin P, Reunala T. Lymphoma in patients with dermatitis herpetiformis and their first-degree relatives. Br J Dermatol 2005;152:82–6. [22] Savilahti E, Reunala T, M¨aki M. Increase of lymphocytes bearing the gamma/delta T cell receptor in the jejunum of patients with dermatitis herpetiformis. Gut 1992;33:206–11. [23] Reunala T, Kosnai I, Karpati S, Kuitunen P, T¨or¨ok E, Savilahti E. Dermatitis herpetiformis: jejunal findings and skin response to gluten-free diet. Arch Dis Child 1984;59:517–22. [24] West J, Logan RFA, Hill PG, Lloyd A, Lewis S, Hubbard R, et al. Seroprevalence, correlates, and characteristics of undetected coeliac disease in England. Gut 2003;52:960–5.
380
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
[25] Snook JA, Dwyer L, Lee-Elliot C, Khan S, Wheeler DW, Nicholas DD. Adult coeliac disease and cigarette smoking. Gut 1996;39:60–2. [26] Lear JT, English JSC, Jones PW. Adult coeliac disease, dermatitis herpetiformis and smoking. Gut 1997;40:289. [27] Peters U, Askling J, Gridley G, Ekbom A, Linet M. Causes of death in patients with celiac disease in a population-based Swedish cohort. Arch Intern Med 2003;163:1566–72. [28] Cottone M, Termini A, Oliva L, Magliocco A, Marrone C, Orlando A, et al. Mortality and causes of death in celiac disease in a Mediterranean area. Dig Dis Sci 1999;44:2538–47. [29] Logan RFA, Rifkind EA, Turner ID, Ferguson A. Mortality in celiac disease. Gastroenterology 1989;97:265–71.
[30] West J, Logan RFA, Card TR, Smith C, Hubbard R. Risk of vascular disease in adults with diagnosed coeliac disease: a population-based study. Aliment Pharmacol Ther 2004;20:73–9. [31] Kemppainen T, Uusitupa M, Janatuinen E, J¨arvinen R, Julkunen R, Pikkarainen P. Intakes of nutrients and nutritional status in coeliac disease. Scand J Gastroenterol 1995;30:575–9. [32] Hallert C, Grant C, Grehn S, Granno C, Hulten S, Midhagen G, et al. Evidence of poor vitamin status in coeliac patients on a gluten-free diet for 10 years. Aliment Pharmacol Ther 2002;16: 1333–9. [33] Marks J, Shuster S, Watson AJ. Small bowel changes in dermatitis herpetiformis. Lancet 1966;II:1280–2.
Alimentary Tract
Malignancies and mortality in patients with coeliac disease and dermatitis herpetiformis: 30-year population-based study M. Viljamaa a,b , K. Kaukinen a,b , E. Pukkala d , K. Hervonen a,c , T. Reunala a,c , P. Collin a,b,∗ a
Departments of Gastroenterology and Alimentary Tract Surgery, Tampere University Hospital, P.O. Box 2000, FIN-33521 Tampere, Finland b Medical School, University of Tampere, FIN-33014 Tampere, Finland c Department of Dermatology, Tampere University Hospital, P.O. Box 2000, FIN-33521 Tampere, Finland d Finnish Cancer Registry, Institute for Statistical and Epidemiological Cancer Research, Liisankatu 21 B, FIN-00170 Helsinki, Finland Received 5 December 2005; accepted 6 March 2006 Available online 18 April 2006
Abstract Background and aim. To assess the long-term risks of malignant diseases and mortality in patients with coeliac disease and dermatitis herpetiformis in a centre, where the prevalence of these diseases is high. The risks have probably been overestimated, as patients with subtle forms have earlier remained undetected. Patients. The study comprised 17,245 person-years of follow-up in 1147 patients. Methods. The observed numbers of malignancies and causes of deaths were assessed, and compared to those expected, and standardised incidence ratio and standardised mortality ratio given. Results. The occurrence of all malignant conditions was equal to that in the population both in coeliac disease and dermatitis herpetiformis: standardised incidence ratios of 1.2 (95% confidence intervals 0.9–1.5) and 1.0 (0.6–1.5), respectively. Five patients with coeliac disease and seven with dermatitis herpetiformis had developed non-Hodgkin lymphoma; standardised incidence ratios of 3.2 (1.0–7.5) and 6.0 (2.4–12.4), respectively. Four patients with coeliac disease and one with dermatitis herpetiformis had enteropathy-associated T-cell lymphoma, associated with inadequate dietary compliance. Mortality was increased (standardised mortality ratio 1.26; 1.00–1.55) in coeliac disease, but decreased in dermatitis herpetiformis (standardised mortality ratio 0.52; 0.36–0.72). Conclusion. The overall prognosis in our patients was good. Non-Hodgkin lymphoma emerged in patients with undiagnosed or poorly treated coeliac disease. The mortality rate in dermatitis herpetiformis was even lower than in the population. Our data support the early diagnosis and dietary treatment of these conditions. © 2006 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: Cancer; Coeliac disease; Dermatitis herpetiformis; Mortality
1. Introduction Coeliac disease (CD) is classified as a gluten-induced disease causing small bowel mucosal damage with villous shortening, crypt hyperplasia and inflammation, which recovers on a gluten-free diet [1]. Dermatitis herpetiformis (DH) is a blistering, itching skin disease, where the diagnosis is based on the demonstration of granular IgA deposits in the dermal ∗
Corresponding author. Tel.: +358 3 3116 7869; fax: +358 3 3551 8402. E-mail address: [email protected] (P. Collin).
papillae of uninvolved perilesional skin as observed by direct immunofluorescence [2,3]. About 75% of our patients with DH have mucosal damage identical to that in CD, and the rest have mucosal inflammation, again consistent with CD [4]. We have, within the last 20 years, taken action to augment the diagnostics of CD and DH [5]. The disorder has been considered even in the presence of mild or atypical symptoms, and serologic tests have been widely employed in patients with other autoimmune diseases and in first-degree relatives of patients with CD and DH. As a result, we have recorded a clinical prevalence of CD and DH in our area as high as 0.3%
1590-8658/$30 © 2006 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2006.03.002
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
already in 1997 [6], two to three times than usually reported [7–9]. Irrespective of the plausible symptoms at diagnosis, our policy has been to treat all the patients similarly by glutenfree diet. From our previous studies, we know that over 80% of our patients follow strict gluten-free diet, and only 2% use normal gluten-containing diet, the remainder having some dietary lapses [10,11]. CD is associated with an increased risk of enteropathyassociated non-Hodgkin’s lymphoma (NHL) [12–16]. Gluten-free diet treatment seems to protect patients from NHL [13]. Oesophageal, colorectal and small bowel cancers have been reported to occur in CD [14,15,17], whereas the risk of breast and lung cancer may be reduced [17]. Due to prolonged diagnostic delay and poor dietary compliance, NHL has thought to be the main cause of the two-fold increased mortality in CD, but when the diagnosis of CD has been made early enough, the additional risk seems to be lower or even negligible [18]. In a study by Swerdlow et al. [19] the mortality rate in DH was not increased; to our knowledge, no other studies have been carried out on mortality in patients with DH. Our large prospective series of patients with CD and DH sampled during 30 years in a high-prevalence area enables us to determine whether the malignancy and mortality rates are really increased in these patients, when compliance with gluten-free dietary treatment has been shown to be good.
2. Materials and methods This population-based study comprised all the 1147 patients; 781 with CD (women 68%) and 366 with DH (48%), diagnosed at the Tampere University Hospital area between 1960 and 2000. The diagnosis of CD was based on villous atrophy and crypt hyperplasia shown on a small bowel biopsy [1]; 79% had been detected because of gastrointestinal symptoms or malabsorption, and 21% by serologic screening in risk groups. The diagnosis of DH was based on the demonstration of granular IgA deposit finding in skin biopsy [4].
375
CD was diagnosed at the age of 39 years (median, range 1–84 years) and DH at 38 years (5–84 years); the respective ages at the time of follow-up were 53 (9–89) and 57 (12–93). The number of person-years at risk was 17,245; 10,956 for CD and 6289 for DH. Malignant diseases were identified by linking the personal identification codes with records from the nationwide database of the Finnish Cancer Registry, which includes more than 99% of incident cases diagnosed in Finland since 1953 [20]. Causes of death were obtained from the files of Statistics Finland. Follow-up commenced on January 1st, 1971 for patients whose diagnosis had been established before 1970 and in the rest from the year of diagnosis. The end-points were 31st December 2002, or the date of emigration (n = 2) or death. The expected figures for malignant diseases and deaths were calculated by multiplying the observed person-years of follow-up by the incidence rate of each malignant disease and the mortality rate in the respective sex, age and calendar period in the Finnish population. Standardised incidence ratio (SIR) was used to measure the relative risk of cancer; that is, the ratio of observed-to-expected cancers. For subgroup analysis, we chose the commonest cancers in CD and DH in Finland, and cancers for which an increased or decreased risk has been reported in CD or DH. The analyses were stratified according to the time since diagnosis. Similarly, the standardised mortality ratio (SMR) was used to assess the relative risk of death. Exact 95% confidence intervals (CI) were computed by assuming that the observed figures followed Poisson distribution. Institution’s ethics committee approved the study.
3. Results 3.1. Malignant diseases The total risk of malignant diseases did not differ from that in the population in general (Table 1). There was a significantly increased risk of NHL in both CD and DH; SIR
Table 1 Observed (O) and expected (E) cancers and SIR for malignant diseases with 95% CI in patients with CD and DH Site of malignant disease
CD O
E
SIR (95% CI)
O
E
SIR (95% CI)
All cancers
49
41.9
1.2 (0.9–1.5)
20
20.7
1.0 (0.6–1.5)
NHL Hodgkin’s lymphoma
5 0
1.6 0.3
3.2 (1.0–7.5)a 0.0 (0.0–14.4)
7 0
1.2 0.2
6.0 (2.4–12.4)a 0.0 (0.0–21.6)
Digestive system Stomach Colon and rectum Oesophagus Small bowel cancer
10 2 4 0 0
8.0 1.2 3.7 0.3 0.1
1.3 (0.6–2.3) 1.2 (0.2–4.5) 1.1 (0.3–2.8) 0.0 (0.0–10.7) 0.0 (0.0–80.1)
6 3 2 0 0
6.6 2.1 2.9 0.3 0.04
0.9 (0.3–2.0) 2.1 (0.4–6.3) 0.7 (0.1–2.5) 0.0 (0.0–11.7) 0.0 (0.0–104.0)
2 9
3.4 10.2
0.6 (0.1–2.1) 0.9 (0.4–1.7)
1 5
3.5 4.71
0.3 (0.0–1.6) 1.06 (0.34–2.47)
Lung and trachea Breast a
Significantly increased risk of NHL.
DH
376
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
Table 2 Characteristics of patients with CD and DH with NHL Number
Gender
Patients with CD 1 F
Dg of CD/DH; year (age)
Dg of NHL; year (age)
Organ involved
Histology of NHL
Adherence to gluten-free diet
Villous atrophy at dg
Time and cause of death
1987 (58)
1995 (66)
Ileum
EATL
1987–1990 no diet, 1991–1995 strict 1989–1990 no diet, 1991–1992 strict 1990–1995 no diet, 1995–1996 strict 1994–1995 strict 1997 strict
Yes
1995; NHL
Yes Yes
1997; pulmonary embolism 1996; NHL
Yes Yes
1999; pneumonia 2003; NHL
Yes
Alive
Yes
1995; obscure
2
M
1989 (65)
1992 (68)
Jejunum
EATL
3
M
1990 (51)
1996 (57)
Ileum
EATL
4 5
F M
1994 (53) 1997 (73)
1995 (54) 1997 (73)
Jejunum Jejunum
DLBCL EATL
1967 (16)
1999 (48)
Stomach
DLBCL
Patients with DH 1 M
Yes
Alive
EATL DLBCL
1964–1988 no diet, 1988–1999 strict 1971–1975 no diet, 1976–1979 partial, 1980–1992 strict 1972–1975 no diet, 1975–1979 strict 1973–1976 partial 1978–1983 strict
Yes No
1976; NHL 1983; NHL
FL
1978–1980 strict
Yes
Alive
DLBCL
1982–1996 partial
No
2003; NHL
2
M
1971 (58)
1992 (79)
Penis
DLBCL
3
M
1972 (30)
1979 (37)
Diffuse
4 5
M F
1973 (47) 1978 (64)
1976 (50) 1983 (69)
6
F
1978 (40)
1980 (42)
7
F
1981 (31)
1996 (46)
Mesenterium, colon Ileum Mesenterium, cervical lymph node Inguinal lymph node Mesenterium, ovario
EATL, enteropathy-associated T-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma. Partial diet: the diet was mostly gluten-free, but dietary transgressions occurred at least every month.
3.2 and 6.0, respectively. Of the five cases with CD and NHL, four were enteropathy-associated T-cell lymphomas (EATL). NHL was diagnosed virtually simultaneously with the diagnosis of CD in two; the other three had had dietary transgressions (Table 2). In DH, only one of the seven with NHL had EATL. Two (29%) of these seven had no small intestinal villous atrophy at the time of diagnosis. Five had at least partial dietary transgression after the introduction of gluten-free diet. The diagnosis of DH had been made in all seven patients with subsequent NHL in 1981 or earlier (Table 2).
3.2. Mortality rate The mortality rate during the follow-up period was significantly increased in CD: SMR 1.26; 95% CI 1.00–1.55. Within the first 5 years of diagnosis, SMR was 1.42 and 1.18 thereafter; this difference was not statistically significant (Table 3). In DH, the mortality was significantly lower than expected (0.52; 0.36–0.72) and the risk reduction was similar before and after 5 years of the diagnosis (Table 3). At the end of follow-up, all five patients with CD and lymphoma and four out of seven with DH and lymphoma had died (Table 2).
Table 3 Mortality during the follow-up time in patients with CD and DH after the diagnosis Follow-up time after diagnosis of CD and DH
Person-years of follow-up
Deaths Observed
Expected
Total follow-up time CD DH
10955.6 6288.6
85 34
0–5 years CD DH
3728.0 1774.0
≥5 years CD DH
7227.6 4514.3
SMR, standardised mortality ratio; CI, confidence interval. a Statistically significant difference compared to general population.
SMR
95% CI
67.6 65.2
1.26 0.52
1.00–1.55a 0.36–0.72a
31 7
21.9 14.0
1.42 0.50
0.96–2.01 0.20–1.02
54 27
45.7 51.2
1.18 0.53
0.89–1.54 0.35–0.76a
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
377
Table 4 Observed (O) and expected (E) deaths and SMR with 95% CI for patients with CD and DH Cause of death
Patients with CD (n = 781)
Patients with DH (n = 366)
O
E
SMR (95% CI)
O
E
SMR (95% CI)
Diseases, all
75
60.9
1.23 (0.97–1.54)
33
59.1
0.56 (0.38–0.78)a
Diseases of circulatory system Ischaemic heart diseases Cerebrovascular diseases Other heart diseases
34 27 1 5
28.2 16.9 6.7 2.3
1.20 (0.83–1.68) 1.59 (1.05–2.31)a 0.15 (0.00–0.83)a 2.14 (0.70–5.00)
21 13 3 3
29.3 18.1 6.5 2.5
0.72 (0.44–1.09) 0.72 (0.38–1.23) 0.46 (0.10–1.35) 1.22 (0.25–3.57)
Malignant neoplasms Lymphoproliferative/haematopoietic diseases Malignant neoplasms of stomach Malignant neoplasms of respiratory tract
17 7 2 1
17.8 1.7 1.2 3.0
0.96 (0.56–1.53) 4.12 (1.66–8.51)a 1.71 (0.21–6.18) 0.33 (0.01–1.85)
9 3 2 1
14.6 1.4 1.1 3.2
0.62 (0.28–1.17) 2.18 (0.45–6.36) 1.88 (0.23–6.79) 0.31 (0.01–1.74)
Non-malignant diseases of digestive system Endocrine diseases Dementia Other diseases of nervous system Infectious diseases Respiratory diseases Urinary diseases Other diseases (non-alcohol) Diseases associated with alcohol misuse
6 1 3 2 0 7 1 3 1
1.7 1.0 2.2 1.2 0.6 4.1 0.6 0.9 2.3
3.59 (1.32–7.82)a 1.02 (0.03–5.70) 1.34 (0.28–3.91) 1.65 (0.20–5.94) 0.00 (0.00–6.75) 1.73 (0.69–3.56) 1.69 (0.04–9.43) 3.44 (0.71–10.05) 0.44 (0.01–2.45)
0 0 2 0 0 1 0 0 0
1.6 0.8 2.7 0.9 0.5 4.5 0.6 0.6 2.2
0.00 (0.00–2.27) 0.00 (0.00–4.45) 0.74 (0.09–2.68) 0.00 (0.00–3.79) 0.00 (0.00–7.46) 0.22 (0.01–1.23) 0.00 (0.00–6.06) 0.00 (0.00–6.28) 0.00 (0.00–1.64)
External causes
8
6.7
1.20 (0.52–2.36)
1
6.1
0.16 (0.00–0.91)a
a
Statistically significant difference compared to general population.
A significant difference in mortality rates between CD and DH was observed in patients with diseases of the circulatory system. Patients with CD had an increased mortality due to ischaemic heart diseases (SMR 1.59), whereas the risk was decreased (0.72), though not significantly, in patients with DH (Table 4). A similar trend was observed in malignant neoplasms of lymphoproliferative and haematopoietic organs, and in non-malignant diseases of the digestive system (not possible to specify in detail). In DH, the mortality due to external causes was significantly lower than expected (Table 4).
4. Discussion The overall cancer risk in our patients remained at population level. The risk of NHL was increased both in patients with CD and DH; by and large in line with data published elsewhere (Table 5). In DH, SIR for NHL was even higher than in CD, but in only one of the seven cases was the organ involved with the small bowel, compared to four of the five in CD. The clinical course of patients with DH and lymphoma has been presented earlier when the risk of relatives to develop lymphoma was assessed: the study also [21] included patients outside our region; that is, in Helsinki area. Here we wanted to survey the risk of lymphoma and mortality in a single centre, and make comparisons between CD, DH and general population. We cannot explain why the patients with DH developed lymphomas other than EATL, including B-cell lymphomas. The less severe small bowel mucosal damage in DH may be the reason for the
difference. Nevertheless, NHL no longer appeared in DH patients whose diagnosis was made recently. Twenty years ago, a gluten-free diet was not considered as essential as today in DH, and the gluten intake may have contributed to the development of these lymphomas. It was impossible to obtain reliable data on each patient’s dietary history, but again, our results show unanimously that the compliance to gluten-free diet was lower in our CD and DH patients with concomitant lymphoma than in our patients in general. We did not calculate the risk of lymphoma between DH patients with normal mucosa and mucosal villous atrophy, as the number of patients in both groups would have been too small, and mucosal biopsy results were not available in each DH case. However, two (29%) out of seven patients with lymphoma and DH had normal villous architecture while on the gluten-containing diet. The percentage was comparable to what has been reported earlier in our patients with DH [22,23]. This suggests that the risk of lymphoma, other than EATL, was not dependent on the degree of villous damage in DH. In contrast to some recent reports [14,15,17], we did not observe any increased occurrence of gastrointestinal cancers other than EATL. Further studies are needed to confirm that the risk of lung cancer is reduced in CD and DH, as seen here (SIR 0.6 in CD and 0.3 in DH). Even though this did not reach statistical significance, the observation is still of note, as the same finding (SIR 0.4; 95% CI 0.1–1.2) was made in the UK [17]. It has been reported that patients with CD and DH smoke less often than people in general [24–26]. The mortality rate in the present study was somewhat lower than shown in other studies [17–19,27–29] (Table 6).
378
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
Table 5 Latest studies of the risk of lymphoma in patients with CD and DH Author (year) [ref.]
Country
Patients (n)
Patient-years of follow-up
SIR (95% CI)
Patient series
Studies on patients with CD Collin et al. (1996) [12]
Finland
383
3107
2.7 (0.1–14.8)
Askling et al. (2002) [14]
Sweden
11019
97236
6.3 (4.2–125)
Green et al. (2003) [15]
US
381
1977
5.3 (2.3–13.0)
Card et al. (2004) [16]
UK
637
5684
5.8 (1.6–14.9)
Current study
Finland
781
10956
3.2 (1.0–7.5)
Studies on patients with DH Collin et al. (1996) [12]
Finland
305
3029
10.3 (2.8–26.3)
Askling et al. (2002) [14]
Sweden
1354
14451
1.9 (0.7–4.0)
Current study
Finland
366
6289
6.0 (2.4–12.4)
Consecutive patients from one unit Hospitalised patients from countrywide in-patient register. Lymphomas within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit Consecutive patients from two units. Lymphomas within 2 years after the diagnosis of CD were excluded Consecutive patients from one unit Consecutive patients from one unit Hospitalised patients from countrywide in-patient register. Lymphomas within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit
SIR, standardised incidence ratio.
Good dietary compliance in our patients may have influenced in the better prognosis: virtually all of our patients adhere to strict gluten-free diet or use gluten-containing products only occasionally [10,21]. Surprisingly, in contrast to CD, the mortality rate in DH was significantly reduced (SMR 0.52). This low rate was seen in almost all categories, though not statistically significantly (Table 4). We well recognise that the reliability of death certificates is far from complete, but this source of
bias is probably equal in the present study series and in the population in general. Nevertheless, there was a conspicuous difference between CD and DH in mortality due to ischaemic heart diseases. In UK, the risk of myocardial infarction was observed to be 15% decreased in 3590 CD patients [30]. Similarly, decreased mortality due to ischaemic heart disease in DH (SMR 0.37; 95% CI 0.12–0.86) has been reported in UK [19]. The fat intake has been reported to be higher in Finnish CD patients than in healthy controls [31],
Table 6 The latest mortality studies in patients with CD and DH Author (year) [ref.]
Country
Patients (n)
Person-years of follow-up
SMR (95% CI)
Patient series
Studies on patients with CD Logan et al. (1989) [29] Cottone et al. (1999) [28] Corrao et al. (2001) [18] Peters et al. (2003) [27]
UK Italy Italy Sweden
653 228 1072 10032
8823 1349 6444 81182
1.9 (1.5–2.2) 3.8 (1.9–6.7) 2.0 (1.5–2.7) 2.0 (1.8–2.1)
4372
18923
1.3 (1.1–1.5)
West et al. (2004) [17]
UK
Current study
Finland
781
10956
1.12 (1.0–1.55)
Patients attending hospitals in two units Consecutive adult patients from one unit Consecutive patients from 11 units Hospitalised patients from countrywide in-patient register. Deaths within 1 year after the diagnosis of CD were excluded Patients from nationwide general practice research database. Deaths within 1 year after the diagnosis of CD were excluded Consecutive patients from one unit
UK Finland
152 366
2288 6289
0.9 (0.6–1.2) 0.52 (0.36–0.72)
Consecutive patients from one unit Consecutive patients from one unit
Studies on patients with DH Swerdlow et al. (1993) [19] Current study
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
but we have no data on DH. On the other hand, serum cholesterol levels have been relatively low both in patients with CD and DH, even when they have adhered to a gluten-free diet [24,26]. In CD, folic acid malabsorption entails high levels of plasma homocysteine, which again may increase the risk of cardiovascular diseases [32]. This mucosal damage and consequently the folic acid malabsorption is milder in DH [33]. The mortality due to external causes in DH was significantly lower than expected, but we consider that our series were unlikely to be biased; the patients in the area had been diagnosed and remained under surveillance in a single centre and there were no drop-outs, which suggest that the DH population was representative of the population in general. The issues are altogether highly speculative, and the reasons for the decreased mortality in DH remain obscure. When comparing the studies in Tables 5 and 6, it must be noted that the study designs are different: some authors had decided to exclude the lymphomas detected within 1 or 2 years after the diagnosis of CD. We consider it essential to include these cases also, when the overall risk of lymphoma is estimated. Most lymphomas in patients with CD will be detected shortly after the diagnosis of CD or initiation of strict gluten-free diet [13,16], as also seen in Table 2. The exclusion might be justified when the protective effect of gluten-free diet is assessed. In conclusion, our long-term follow-up study showed that the risk of NHL was comparable to what has been reported elsewhere. EATL was not observed in patients with longterm adherence to a gluten-free diet. The mortality rate in CD was only marginally increased compared to the population in general, and even decreased in DH. The overall prognosis in CD and DH was thus good in our high-prevalence area with well-treated patients. Conflict of interest statement None declared.
Acknowledgements This study was supported by the Research Fund of the Finnish Coeliac Society, the Mary and George C. Ehrnrooth Foundation, the Finnish Medical Foundation, the Finnish Foundation for Paediatric Research and the Medical Research Fund of Tampere University Hospital.
References [1] Walker-Smith JA, Guandalini S, Schmitz J, Shmerling DH, Visakorpi JK. Revised criteria for diagnosis of coeliac disease. Arch Dis Child 1990;65:909–11. [2] van der Meer JB. Granular deposits of immunoglobulins in the skin of patients with dermatitis herpetiformis. An immunofluorescent study. Br J Dermatol 1969;81:493–503.
379
[3] Zone JJ, Meyer LJ, Petersen MJ. Deposition of granular IgA relative to clinical lesions in dermatitis herpetiformis. Arch Dermatol 1996;132:912–8. [4] Reunala T. Dermatitis herpetiformis: coeliac disease of the skin. Ann Med 1998;30:416–8. [5] Collin P, H¨allstr¨om O, M¨aki M, Viander M, Keyril¨ainen O. Atypical coeliac disease found with serologic screening. Scand J Gastroenterol 1990;25:245–50. [6] Collin P, Reunala T, Rasmussen M, Kyr¨onpalo S, Pehkonen E, Laippala P, et al. High incidence and prevalence of adult coeliac disease. Augmented diagnostic approach. Scand J Gastroenterol 1997;32:1129–33. [7] Midhagen G, J¨arnerot G, Kraaz W. Adult coeliac disease within a defined geographic area in Sweden. A study of prevalence and associated diseases. Scand J Gastroenterol 1988;23:1000–4. [8] Schweizer JJ, von Blomberg BME, Bueno-de Mesquita HB, Mearin ML. Coeliac disease in the Netherlands. Scand J Gastroenterol 2004;39:359–64. [9] Talley NJ, Valdovinos M, Petterson TM, Carpenter HA, Melton Jr L. Epidemiology of celiac sprue: a community-based study. Am J Gastroenterol 1994;89:843–6. [10] Viljamaa M, Collin P, Huhtala H, Sievanen H, Maki M, Kaukinen K. Is coeliac disease screening in risk groups justified? Fourteenyear follow-up with special focus on compliance and quality of life. Aliment Pharmacol Ther 2005;22:317–24. [11] Hervonen K, Hakanen M, Kaukinen K, Collin P, Reunala T. Firstdegree relatives are frequently affected in coeliac disease and dermatitis herpetiformis. Scand J Gastroenterol 2002;37:51–5. [12] Collin P, Pukkala E, Reunala T. Malignancy and survival in dermatitis herpetiformis: a comparison with coeliac disease. Gut 1996;38:528–30. [13] Holmes GKT, Prior P, Lane MR, Pope D, Allan RN. Malignancy in coeliac disease—effect of a gluten free diet. Gut 1989;30:333–8. [14] Askling J, Linet M, Gridley G, Halstensen TS, Ekstrom K, Ekbom A. Cancer incidence in a population-based cohort of individuals hospitalized with celiac disease or dermatitis herpetiformis. Gastroenterology 2002;123:1428–35. [15] Green PH, Fleischauer AT, Bhagat G, Goyal R, Jabri B, Neugut AI. Risk of malignancy in patients with celiac disease. Am J Med 2003;115:191–5. [16] Card TR, West J, Holmes GK. Risk of malignancy in diagnosed coeliac disease: a 24-year prospective, population-based, cohort study. Aliment Pharmacol Ther 2004;20:769–75. [17] West J, Logan RF, Smith CJ, Hubbard RB, Card TR. Malignancy and mortality in people with coeliac diease: population based cohort study. Br Med J 2004;329:716–9. [18] Corrao G, Corazza GR, Bagnardi V, Brusco G, Ciacci C, Cottone M, et al. Mortality in patients with coeliac disease and their relatives: a cohort study. Lancet 2001;358:356–61. [19] Swerdlow AJ, Whittaker S, Carpenter LM, English JSC. Mortality and cancer incidence in patients with dermatitis herpetiformis: a cohort study. Br J Dermatol 1993;129:140–4. [20] Teppo L, Pukkala E, Lehtonen M. Data quality and quality control of a population-based cancer registry. Experience in Finland. Acta Oncol 1994;33:365–9. [21] Hervonen K, Vornanen M, Kautiainen H, Collin P, Reunala T. Lymphoma in patients with dermatitis herpetiformis and their first-degree relatives. Br J Dermatol 2005;152:82–6. [22] Savilahti E, Reunala T, M¨aki M. Increase of lymphocytes bearing the gamma/delta T cell receptor in the jejunum of patients with dermatitis herpetiformis. Gut 1992;33:206–11. [23] Reunala T, Kosnai I, Karpati S, Kuitunen P, T¨or¨ok E, Savilahti E. Dermatitis herpetiformis: jejunal findings and skin response to gluten-free diet. Arch Dis Child 1984;59:517–22. [24] West J, Logan RFA, Hill PG, Lloyd A, Lewis S, Hubbard R, et al. Seroprevalence, correlates, and characteristics of undetected coeliac disease in England. Gut 2003;52:960–5.
380
M. Viljamaa et al. / Digestive and Liver Disease 38 (2006) 374–380
[25] Snook JA, Dwyer L, Lee-Elliot C, Khan S, Wheeler DW, Nicholas DD. Adult coeliac disease and cigarette smoking. Gut 1996;39:60–2. [26] Lear JT, English JSC, Jones PW. Adult coeliac disease, dermatitis herpetiformis and smoking. Gut 1997;40:289. [27] Peters U, Askling J, Gridley G, Ekbom A, Linet M. Causes of death in patients with celiac disease in a population-based Swedish cohort. Arch Intern Med 2003;163:1566–72. [28] Cottone M, Termini A, Oliva L, Magliocco A, Marrone C, Orlando A, et al. Mortality and causes of death in celiac disease in a Mediterranean area. Dig Dis Sci 1999;44:2538–47. [29] Logan RFA, Rifkind EA, Turner ID, Ferguson A. Mortality in celiac disease. Gastroenterology 1989;97:265–71.
[30] West J, Logan RFA, Card TR, Smith C, Hubbard R. Risk of vascular disease in adults with diagnosed coeliac disease: a population-based study. Aliment Pharmacol Ther 2004;20:73–9. [31] Kemppainen T, Uusitupa M, Janatuinen E, J¨arvinen R, Julkunen R, Pikkarainen P. Intakes of nutrients and nutritional status in coeliac disease. Scand J Gastroenterol 1995;30:575–9. [32] Hallert C, Grant C, Grehn S, Granno C, Hulten S, Midhagen G, et al. Evidence of poor vitamin status in coeliac patients on a gluten-free diet for 10 years. Aliment Pharmacol Ther 2002;16: 1333–9. [33] Marks J, Shuster S, Watson AJ. Small bowel changes in dermatitis herpetiformis. Lancet 1966;II:1280–2.