Lancet Article Part 1

[Guest guest]

These tables did not copy correctly! I am sorry they are garbled!

Beverly G.

Volume 351, Number 9109 11 April 1998

Case-control study of risk factors of Creutzfeldt-Jakob disease in Europe

during 1993-95

C M van Duijn, N Delasnerie-Lauprêtre, C Masullo, I Zerr, R de Silva, D P W M

Wientjens, J-P Brandel, T Weber, V Bonavita, M Zeidler, A Alpérovitch, S

Poser, E Granieri, A Hofman, R G Will, for the European Union (EU)

Collaborative Study Group of Creutzfeldt-Jakob disease (CJD)*

------------------------------------------------------------------------------

--

*Members of the EU Collaborative Study Group of CJD are listed at end of paper

------------------------------------------------------------------------------

--

Department of Epidemiology & Biostatistics, Erasmus University Medical School,

PO Box 1738, 3000 DR Rotterdam, The Netherlands (C M van Duijn PhD, D P W M

Wientjens MD, A Hofman MD); U 360 INSERM, Hôpital de la Salpetrière, Paris,

France (N Delasnerie-Lauprêtre MD, J-P Brandel MD, A Alpérovitch MD);

Department of Neurology, Catholic University of Sacred Heart, Rome, Italy (C

Masullo MD); Neurologische Klinik, Georg-August University Göttingen,

Göttingen, Germany (I Zerr MD, S Poser MD); CJD Surveillance Unit, Western

General Hospital, Edinburgh, UK (R de Silva MRCP, M Zeidler MRCP, R G Will

FRCP); Marienkrankenhaus, Alfred Strasse 9, Hamburg, Germany (T Weber MD);

Department of Neurology, Federico II University, Naples, Italy (V Bonavita

MD); Department of Neurology, University of Ferrara, Ferrara, Italy (E

Granieri MD)

------------------------------------------------------------------------------

--

Correspondence to: Dr C M van Duijn

Summary

Introduction

Patients and methods

Results

Discussion

References

Summary

Background Creutzfeldt-Jakob disease (CJD) is a transmissible spongiform

encephalopathy. Genetic and iatrogenic forms have been recognised but most are

sporadic and of unknown cause. We have studied risk factors for CJD as part of

the 1993-95 European Union collaborative studies of CJD in Europe.

Methods The 405 patients with definite or probable CJD who took part in our

study had taken part in population-based studies done between 1993 and 1995 in

Belgium, France, Germany, Italy, the Netherlands, and the UK. Data on putative

risk factors from these patients were compared with data from 405 controls.

Findings We found evidence for familial aggregation of CJD with dementia due

to causes other than CJD (relative risk [RR] 2·26, 95% CI 1·31-3·90). No

significant increased risk of CJD in relation to a history of surgery and

blood tranfusion was shown. There was no evidence for an association between

the risk of CJD and the consumption of beef, veal, lamb, cheese, or milk. No

association was found with occupational exposure to animals or leather. The

few positive findings of the study include increased risk in relation to

consumption of raw meat (RR 1·63 [95% CI 1·18-2·23]) and brain (1·68

[1·18-2·39]), frequent exposure to leather products (1·94 [1·13-3·33]), and

exposure to fertiliser consisting of hoofs and horns (2·32 [1·38-2·91]).

Additional analyses, for example stratification by country and of exposures

pre-1985 and post-1985, suggest that these results should be interpreted with

great caution.

Interpretation Within the limits of the retrospective design of the study, our

findings suggest that genetic factors other than the known CJD mutations may

play an important part in CJD. Iatrogenic transmission of disease seems rare

in this large population-based sample of patients with CJD. There is little

evidence for an association between the risk of CJD and either animal

exposure, or consumption of processed bovine meat or milk products for the

period studied.

Lancet 1998; 351: 1081-85

Introduction

Creutzfeldt-Jakob disease (CJD) is the most clinically significant spongiform

encephalopathy in man.1,2 Although the disease is rare,3 the rapid development

of bovine spongiform encephalopathy (BSE) from a sporadic to an endemic

disease in cattle in the UK4 underscores the potential transmissibility of

these diseases. The discovery of a new variant (nv) of CJD in the UK following

the BSE5 epidemic has re-opened discussion on whether spongiform

encephalopathies may be transmitted from animals to man. Also, the potential

of iatrogenic transmission of CJD remains a matter of concern. CJD has been

transmitted from person to person through human pituitary derived growth and

gonadotropin hormones,6-9 neurosurgery and electroencephalography electrode

implantation,10-13 and corneal transplantation.14 Up until now there has been

no epidemiological evidence of transmission through blood transfusion,15 but

research in animal models indicates that this possibility cannot be

excluded.16,17 Here, we present the findings of a collaborative study of risk

factors for CJD in Europe, in which genetic factors, medical history,

occupational history, animal exposure, and diet were studied.

Patients and methods

The study was embedded within the European Union (EU) collaborative studies

monitoring the incidence of CJD in Belgium (Flanders only), France, Germany,

Italy, the Netherlands, and the UK.3 The registers aimed to ascertain all

patients diagnosed with definite or probable CJD living in a defined

geographical area, mainly at the national level. A standardised diagnostic

protocol was used according to the criteria of Masters and colleagues.18 The

pooled data set of 613 cases of CJD described earlier and 11 additional cases

identified in Belgium.3 Of these patients, 405 (199 definite and 206 probable

CJD) were included in the studies of risk factors, 43 patients from Flanders

and/or the Netherlands, 75 patients from France, 136 patients from Germany, 63

from Italy, and 88 from the UK. Reasons for exclusion in the case-control

study included failure to obtain permission to approach relatives of patients,

refusal to participate in the study, lack of appropriate controls, and late

referral of suspect cases. The patient series consisted of 154 men and 251

women. Nine patients were younger than 40 years, 190 were between 40 years and

64 years, and 206 were age 65 years or older at the time of onset of CJD.

Patients with and without a family history of CJD were included in the study.

In 14 patients with CJD a mutation in the prion protein gene (PRNP) was

found.20-24

The cases were matched to 405 controls according to age (SD 5 years) and sex.

Control participants were recruited from the hospital where the patient who

had CJD had been diagnosed. Patients diagnosed with dementia were not included

in the control group. A standardised core questionnaire for risk-factor

exposure was used. Data were collected by a structured interview done by a

research assistant or physician. For cases, the data were obtained from a

next-of-kin. For controls, a next-of-kin was interviewed whenever possible.

The raw data of the five participating centres were centralised and analysed.

The strength of association between CJD and putative risk factors was assessed

by computing the odds ratio as an estimate of the relative risk (RR).19 RRs

were estimated by maximum likelihood, and the 95% CIs were based on the

asymptotic standard errors. To achieve maximum statistical power, all analyses

were done with an unconditional logistic-regression model, adjusted for the

matching variable age, sex, and study centre. Significant findings were re-

analysed with conditional logistic regression, in accordance with the matched

design of the study. Because exposures may differ considerably between

countries, we tested for heterogeneity by including an interaction term with a

risk factor into the logistic-regression model. Also, a stratified analysis

was done for exposures pre-1985 and post-1985, the year marking the start of

the BSE epidemic in the UK.4

We also analysed the data stratified by sex, age at onset, diagnosis (probable

versus definitive), family history of CJD, PRNP mutations, and the PRNP

polymorphism at codon 129.20-24 Despite the pooling of case-control studies,

the sample size has limited power when testing for differences between risks

in subgroups.25 We therefore report here all significant findings within any

of the subgroups studied, although tests for heterogeneity were not

significant.

Results

54 cases and 22 controls had a positive family history of dementia other than

CJD. In the cases of CJD the frequency of dementia in first degree relatives

proved to be 2·26 (95% CI 1·31-3·90) that of the controls. Table 1 shows the

medical history in patients and controls. Surgery of the vertebral column was

found less often among cases, whereas there was a non-significant increase in

the risk of CJD associated with surgery of the brain. A history of blood

transfusion, electromyography, and lumbar puncture was seen less often in

cases than controls. A history of blood transfusion more than 10 years before

the interview (a period aetiologically relevant because of the long incubation

period of CJD), did not increase the risk of subsequent development of CJD.

------------------------------------------------------------------------------

--

History Exposure frequency Relative risk Relative risk

(at CJD diagnosis) (95% CI) at (95% CI) 10 years

Cases Controls diagnosis before onset

Surgery

Brain 12/400 7/405 1·77 (0·68-4·61) 1·76 (0·69-4·51)

Vertebral column 17/400 31/405 0·53 (0·23-0·98) 0·54 (0·29-0·98)

Other CNS surgery 2/400 2/405 1·03 (0·14-7·40) 1·01 (0·14-7·22)

Eye 33/401 34/406 0·96 (0·57-1·59) 0·98 (0·59-1·62)

Medical treatment/tests

Blood transfusion 38/341 71/378 0·56 (0·37-0·97) 0·74 (0·40-1·37)

Electromyography 9/162 26/176 0·32 (0·14-0·72) 0·25 (0·03-2·22)

Lumbar puncture 38/188 40/144 0·49 (0·28-0·87) 1·00 (0·41-2·43)

Vaccination 109/379 114/385 0·95 (0·68-1·32) 0·93 (0·18-4·90)

Hormone supplements 41/313 30/320 1·50 (0·89-2·54) 0·98 (0·20-4·90)

CNS=central nervous system.

Table 1: Medical history and the risk of CJD

------------------------------------------------------------------------------

--

The medical history of the controls may not be representative of the general

population. If neurological inpatient controls were excluded, there was no

significant difference between cases and controls for surgery of the vertebral

column, electromyography, or lumbar puncture, which suggests the inverse

associations in table 1 may be explained by selection of the controls. There

was no evidence for an association between CJD and any of the disorders

studied in the medical history including neurological and psychiatric disease,

infectious disorders, allergies and atopy, head trauma and for

ophthalmological examination or dental treatment (not in table).

With regard to occupational history, medical or allied professions, and

occupations that involve contact with animals or animal products were assessed

(table 2). None of the professions studied was significantly associated with

increased risk of CJD in the pooled analysis. Occupational exposure to cows

was associated with an increased risk of CJD in the small subset of Italy only

(RR 11·14 [95% CI 1·11-112·02]). However, when compared with the risk observed

at the other sites, no significant difference could be found. With regard to

domestic animals, associations were found at single sites: to cats in the UK

(2·48 [1·35-4·55], significantly [p=0·01] different from the other sites) and

to birds in France (7·57 [1·54-37·24], not significantly different from the

other sites). In the overall analyses, contact with leather products other

than through clothes was associated with a 1·94-fold (1·13-3·33; table 2)

increase in the risk of CJD. This association was absent in the Italian

population. Exposure to fertiliser that contained hoofs and horns was reported

significantly more often for patients than controls (2·32 [1·38-2·91]).

However, this risk was not apparent in analyses after 1985 in the

Netherlands/Belgium, Italy, or the UK.

------------------------------------------------------------------------------

--

Exposure Exposure frequency Relative risk

Cases Controls (95% CI)

Occupational exposures

Health professional* 32/397 35/404 0·92 (0·69-1·32)

Butcher 7/392 5/401 1·43 (0·45-4·60)

Slaughterhouse worker 2/347 2/354 1·05 (0·14-7·29)

Meat/food processor 9/347 9/354 1·01 (0·39-2·61)

Leather worker 2/347 2/354 1·01 (0·14-7·33)

Animals/animal products 73/395 75/402 0·99 (0·68-1·42)

Husbandry 57/393 51/400 1·17 (0·77-1·77)

Farm worker

Cows 43/344 38/354 1·20 (0·75-1·93)

Sheep 23/343 23/354 1·05 (0·57-1·96)

General exposure to animals and animal products

Lived on farm 153/395 142/399 1·14 (0·86-1·54)

Used artificial fertiliser 24/302 32/318 0·76 (0·43-1·36)

Used fertiliser containing 48/366 24/382 2·32 (1·38-2·91)

hoofs and horns

Contact with bone meal 57/358 46/374 1·42 (0·90-2·22)

Contact with fur/leather other 42/391 24/397 1·94 (1·13-3·33)

than through clothes

*Occupations included physician, neuropathologist, nurse, laboratory

technician, dentist, and ambulance worker.