Elsevier

The Lancet

Volume 363, Issue 9419, 1 May 2004, Pages 1422-1426
The Lancet

Articles
Asbestos exposure as a risk factor for retroperitoneal fibrosis

https://doi.org/10.1016/S0140-6736(04)16100-XGet rights and content

Summary

Background

Retroperitoneal fibrosis (RPF) is an uncommon disease with unknown causation in most cases. The pathognomonic finding is a fibrous mass covering the abdominal aorta and the ureters. Our aim was to clarify the possible role of asbestos exposure in the development of RPF. The hypothesis was based on the ability of asbestos to cause fibrosis in pulmonary and pleural tissue.

Methods

We undertook a case-control study of 43 patients with the disease (86% of eligible cases) treated in three university hospital districts of Finland in 1990–2001. For every patient, five population-based controls were selected, matched by age, sex, and central hospital district. We assessed asbestos exposure and medical history using a postal questionnaire and a personal interview. Of the 215 eligible controls, 179 (83%) participated in the study.

Findings

The age-standardised incidence of RPF was 0·10 (95% CI 0·07–0·14) per 100 000 person-years. The disease was strongly associated with asbestos exposure. The odds ratio (OR) was 5·54 (1·64–18·65) for less than 10 fibre-years of asbestos exposure and 8·84 (2·03–38·50) for 10 or more fibre-years, the attributable fraction being 82% and 89%, respectively. Other risk factors were previous use of ergot derivates (OR 9·92 [1·63–60·26]), abdominal aortic aneurysm (OR 6·73 [0·81–56·08]), and smoking for more than 20 pack-years (OR 4·73 [1·28–17·41]).

Interpretation

Our results show that occupational asbestos exposure is an important causal factor for RPF. For patients with work-related asbestos exposure, RPF should be considered an occupational disease.

Introduction

Idiopathic retroperitoneal fibrosis, Ormond's disease, is a rare condition that was first described by Albarran in 1905,1 with comprehensive characterisation by Ormond2 in 1948. The pathognomonic feature of retroperitoneal fibrosis (RPF) is a thick retroperitoneal fibrotic mass covering the abdominal aorta and compressing the ureters (figure). The process of fibrosis can result in obstruction of the ureters and renal failure. The main treatment options include surgical liberation of the ureters and systemic administration of corticosteroids. Investigators have reported remission rates exceeding 90%.3

Previous work on the cause of this disease have been mainly case reports and series, which indicate that RPF can be induced by different factors. About a third of cases develop as a result of malignant disease, radiation therapy, abdominal surgery, pancreatitis, haematomas, and infections.4, 5 RPF has also been associated with the use of several drugs, especially methysergide and other ergotamine derivates.6

Little is known about the role of occupational factors such as asbestos exposure in RPF. Asbestos fibres cause interstitial lung fibrosis (asbestosis), pleural fibrosis, pleural plaques, lung cancer, and pleural and peritoneal mesothelioma.7 Asbestos exposure has been proposed as a causal factor for pleural and retroperitoneal fibrosis in three previous case reports.8, 9, 10 RPF has been suggested as the most severe form of chronic periaortitis, caused by the autoimmune response to the components of atherosclerotic plaques.11 The histological appearances of RPF and atherosclerotic chronic periaortitis are often identical.12 An association between abdominal aortic aneurysms and RPF has also been shown.5

The inflammatory nature of idiopathic RPF is supported by an elevated erythrocyte sedimentation rate at presentation and positive antinuclear antibodies,13 good response to anti-inflammatory drugs, and an occasional association with different autoimmune diseases.14 However, in most cases, the causative factors remain elusive. Our aim was to assess the aetiological significance of occupational asbestos exposure in the development of RPF in a case-control study.

Section snippets

Methods

Patients

We selected all adult patients alive with RPF diagnosed and treated in 12 secondary and tertiary hospitals within three university hospital districts in Finland in 1990–2001. The population of the catchment area was 3·62 million in 2001. Demographic data were obtained from Statistics Finland. The ethics committee of Tampere University Hospital approved the study plan.

Potential cases were retrieved from the hospital discharge data register. Case identification was based on diagnostic

Controls

For every case, five living controls matched for year of birth, sex, and central hospital district were randomly assigned from the Finnish population register centre. Matching for central hospital district was used to avoid bias from different industrial structures in the regions. At the time of enrolment, 1 597 483 people born in 1918–63 lived in the study area.

Data collection

A five-page questionnaire was sent to all patients and controls to obtain information about sociodemographic factors, smoking, and medical history (including chronic diseases, abdominal surgery, aortic aneurysm, migraine and migraine medication used, and other permanent medication and malignant diseases). A structured questionnaire with 21 questions compiled and validated by the Finnish Institute of Occupational Health15 was used for assessment of asbestos exposure.

Postal questionnaires are

Evaluation of asbestos exposure

Assessment of cumulative asbestos exposure was based on fibre-years as determined by an expert according to industrial hygiene knowledge of certain occupations. A fibre-year was defined as working in a full shift (40 hours per week) for 1 year at an average dust level of 1 fibre/mL of air. The cumulative dose of 10 fibre-years has been estimated to cause a 1% risk of developing clinically recognisable asbestosis.7 We defined three grades of cumulative exposure before obtaining data: 0=no

Analysis

We calculated age-standardised incidence using the European 5-year group standard population.17 Multivariate analysis was done with conditional logistic regression in STATA (version 7.0). The attributable fraction (proportion of cases caused by exposure) was calculated with the formula (OR-1)/OR and the population attributable fraction (PAF) was determined as q(OR-1)/OR, in which OR was the odds ratio and q was the proportion of cases exposed to the factor.18 In calculation of the CIs, q was

Role of the funding source

The funding sources of the study had no role in the study design, data collection, data analysis, data interpretation, or writing of the report.

Results

The age-standardised incidence of RPF was 0·10 (95% CI 0·07–0·14) per 100 000 person-years overall, 0·14 (0·08–0·21) for men and 0·07 (0·04–0·10) for women, in 1990–2001.

The prevalence of the disease was 1·38 per 100 000 inhabitants in the study area. 22 (51%) cases and 48 (27%) controls had a positive asbestos exposure history (table 2). Exposures exceeding 10 fibre-years were exclusively occupational (table 3). Asbestos exposure was strongly associated with RPF, with crude OR 4·2 (95% CI

Discussion

Occupational asbestos exposure seems to be an important risk factor for RPF. To our knowledge, in addition to our first report,8 only two case reports have been published that suggest asbestos exposure as a common cause for diffuse pleural thickening, pleural calcification, and RPF.9, 10

We recorded a nine-fold risk for RPF in association with ≥10 fibre-years of asbestos exposure. The fraction of exposed cases attributable to asbestos exposure was more than 80%. This fraction is equivalent to

References (29)

  • R Bégin et al.

    Asbestos

  • R Sauni et al.

    Asbestos exposure: a potential cause of retroperitoneal fibrosis

    Am J Ind Med

    (1998)
  • GP Maguire et al.

    Association of asbestos exposure, retroperitoneal fibrosis, and acute renal failure

    NY State J Med

    (1991)
  • DV Parums et al.

    Serum antibodies to oxidized low-density lipoprotein and ceroid in chronic periaortitis

    Arch Pathol Lab Med

    (1990)
  • Cited by (0)

    View full text