British Journal of Industrial Medicine 1985;42:838-843

Silica, silicosis, and progressive systemic sclerosisGK SLUIS-CREMER, PA HESSEL, EH NIZDO, AR CHURCHILL, AND EA ZEISS

From the Epidemiology Research Unit, Medical Bureau for Occupational Diseases, Johannesburg 2000,South Africa

ABSTRACT An inquiry into the relation between exposure to silica dust, the presence of silicosis, andprogressive systemic sclerosis was conducted in white South African gold miners by means of a

case-control study. Seventy nine cases of progressive systemic sclerosis were matched by year ofbirth with an equal number of control miners selected randomly but bearing in mind theadministrative channel through which the case had come to be identified. Analysis showed no

association between silicosis and PSS but did show that the cumulative life time silica exposure was

significantly higher in the cases compared with controls. This difference was due to a difference inthe intensity of exposure to silica during mining service rather than a difference in duration ofservice. The results are discussed in the context of current thought on the aetiology of progressivesystemic sclerosis, particularly in relation to autoimmune and genetic factors.

Several case reports documenting the co-existence ofsilicosis and progressive systemic sclerosis (PSS) inthe same patient have appeared, mostly fromEuropean countries but also from Japan' and theUnited States.2 In some of these cases the diagnosis ofPSS was substantiated by a skin biopsy. In 1970Gunther and Schuchard were able to enumerate 49such published cases3 and since then there have beenmany more. Devulder et al discussed the associationand presented a bibliography up to 1977.4 Theysuggested the eponym "Erasmus syndrome" shouldbe used when referring to the co-existence of silicosisand PSS.

Erasmus, however, did not report primarily on theassociation of silicosis and PSS.5 His study wasconcerned with the apparently high prevalence of PSSin Witwatersrand gold miners exposed to dustcontaining a high percentage (± 30%) of free silicacompared with a large male non-mining hospitalpopulation which he used as a referent population. Inonly six of his 17 cases of PSS was definite or probablesilicosis considered to be present on chestradiography.The only other epidemiological study on the

association between silicosis and PSS is one byRodnan et al who noted that of 60 male patientsdiagnosed as suffering from PSS in the period 1955-6,26 had been exposed to "prolonged and heavyexposure to silica dust."2 These cases included 16 coalminers. Coal miners may under certain mining

Received II March 1985Accepted 15 April 1985

conditions be subjected to heavy exposures ofsiliceous dust but this is unusual. In their report onlyeight of 19 radiographs available for review showedsigns of silicosis. Presumably for seven of the 26patients no chest radiographs were available.

Both this report and that of Devulder et al,4 whoalso mention development of PSS in pneumoconioticcoal miners, leave one uncertain as to the intensity ofthe exposure to silica (if any) and the nature of thepneumoconiosis that was present.

Certain hitherto unpublished information on theprevalence of scleroderma in South African goldminers is available.6 A study was carried out tocompare the incidence of PSS among white men overage 19 recognised by the Mines Benefit Society whichprovides health care to miners (most exposed to silicadust) with that reported to the South AfricanRailways and Harbours Sick Fund, which includesfew members exposed to silica dust. During the 10years from 1960 to 1969, 29 members of the MinesBenefit Society (7-73/100000 a year) were diagnosedas having PSS whereas only three of the railway andharbour workers (0-33/100000 a year) developed thedisease. Strict criteria for the diagnosis of PSS couldnot, however, be applied in this study which relied onphysicians' private records and memories.Nevertheless, the large difference in the rates is highlysuggestive. The rate for the Railway and Harbourworkers was similar to that reported for white men inShelby county by Medsger and Masi.7

South African gold miners are exposed to levels ofsilica dust that may be reasonably accuratelyquantitated. The specific occupations and numbers of

838

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from

Silica, silicosis, and progressive systemic sclerosis

shifts worked are recorded and the silica dustintensities for different occupations have beenmeasured.8 The men also have annual chestradiographs and clinical examinations, therebyenhancing the likelihood of discovering silicosis orPSS at onset. Many but not all continue to beexamined regularly after ceasing mining. It wasconsidered that these data could be used withconfidence to carry out a case-control study toexamine the relation between silicosis, exposure tocumulative silica dust, and PSS.

Materials and methods

A register of cases of PSS seen at the Medical Bureaufor Occupational Diseases (MBOD) has beenmaintained since 1955. The MBOD is responsible formonitoring the health of all miners in South Africa. Atotal of 79 definite or probable cases of PSS amongwhite miners according to the criteria of the AmericanRheumatism Association9 had been accumulated byJune 1984. Age at onset ranged from 27 to 71 (mean,45 3). Onset of PSS was deduced from the clinicaldata recorded in the MBOD files, usually from severalsuccessive annual examinations during whichsymptoms were routinely recorded, and by furtherquestioning of the patients at the time of diagnosis.

Cases were identified in several ways. Some werediscovered during routine periodical examinationsthat are annual statutory examinations performed torenew certificates of fitness for work in the mines.Cases discovered in this way usually did not knowthat they had PSS before its detection at theperiodical examination. The others were identifiedduring benefit examinations which are performed toassess whether the worker is suffering from acompensable disease. Compensable diseases includethe pneumoconioses, chronic obstructive airwaysdisease (COAD), PSS, and several other diseases thatare associated with exposure to asbestos and are notrelevant to this study group which is composed ofminers with nearly all of their service in gold mines.Benefit examinations, which are carried out byspecialist physicians, are more comprehensive thanperiodical examinations. They include, for example,an electrocardiograph and lung function tests. Minersapply for benefit examinations if they think that theymay be suffering from a compensable disease. Theymay also request a benefit examination if, havingalready been compensated, they think that theircondition may have worsened to the extent that theyare entitled to increased compensation. Whereas thereis no minimum service requirement to receivecompensation for the pneumoconioses or PSS, longservice in dust is required for compensation forCOAD, which accounts for 55% of all compensation

839

decisions.10As a result of the different ways in which the cases

were identified and the differences in the miners whoundergo periodical compared with benefitexaminations (the latter with a high proportion ofsilicotic patients and those with long service), ascheme was devised for obtaining appropriatelymatched controls. All controls were selected fromminers attending the MBOD for examinationsbetween May 1970 and April 1971 and wererandomly matched to cases by year of birth. Casesidentified during a periodical examination werematched with controls who had undergone periodicalexaminations. Cases identified during a benefitexamination in whom PSS was discovered as anincidental finding-that is to say, the miner was notappearing for a benefit because he knew that he hadPSS-were matched with controls who hadundergone benefit examinations. Cases identifiedduring a benefit examination in which PSS was not anincidental finding, the miner appearing for a benefitexamination because he knew he had PSS, werematched with controls who had undergone periodicalexaminations. The reason for distinguishing betweenthe two types of miners who were identified at benefitexaminations was that miners in whom PSS wasdiscovered as an incidental finding were thought to bemore like others who were attending for benefitexaminations with regard to background levels ofsilicosis and silica dust exposure; the two risk factorsof interest. On the other hand, those cases identifiedduring benefit examinations who were attendingsolely because they knew that they had PSS werethought not to have resembled the remainder of thebenefits with regard to background levels of these riskfactors. It was thought that a periodical examinee wasa more appropriate control for such cases. A furthercomplication resulted from the fact that PSS becamea compensable disease in miners in 1974. For thisreason, cases with prior knowledge that they had PSSwho appeared for benefit examinations before 1974were matched with controls who had undergonebenefit examinations. Through this matchingprocedure, 46 cases were matched with benefits and33 were matched with periodicals. This control groupwill be referred to as the mixed controls.

Despite efforts to determine the circumstancesunder which each case was diagnosed, it was thoughtthat the use of the mixed controls mightunderestimate the true associations since some of thesubjects in whom PSS was diagnosed as an incidentalfinding at a benefit examination might actually haveappeared for a benefit examination to explore thepossibility that they had another compensable diseasebecause they were suffering from disability associatedwith PSS even though they did not know that they

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from

840

had PSS. For this reason, a second control group was

selected. These controls were again individuallymatched by year of birth with the cases and were

selected at random from all the miners appearing atthe MBOD for either a periodical (n = 29245) or abenefit (n = 3876) examination between May 1970and April 1971. It was thought that by increasing theproportion of periodical controls the probabletendency for undiagnosed cases to appear for benefitexaminations would be accounted for. This controlgroup will be referred to as the population controls.Results for both control groups will be presented.For each subject the most recent posteroranterior

chest radiograph was read by a single experiencedreader for silicosis using the major categories forprofusion of rounded opacities of the ILO 1971classification." For those whose most recentradiograph showed evidence of silicosis, the year offirst appearance of each of the major categories was

determined. Also, the year of first appearance ofprogressive massive fibrosis was recorded if present.As radiographs were usually taken annually at theperiodical examinations, the determination of firstappearance was thought to be reasonably accurate.All radiograph readings were made withoutknowledge of the exposure history of the individualor his status as a case or a control. For some

case-control pairs, a large difference was foundbetween the dates of the last available radiographs.Typically, the last radiograph of the control was laterthan that of the case, probably due to the fact thatcases often had to leave employment because ofdisability. For case-control pairs in which only onemember showed signs of silicosis and a differencebetween the dates of the last radiographs was found,films corresponding to the earlier date were read forboth members of the pair. In this way the opportunityfor development of silicosis was thought to be equalunder the null hypothesis of no association betweenPSS and silicosis.

Information on the amount of time spent indifferent occupations (most often the number of shiftsworked) and the dust levels of these occupations wasavailable for the gold miners. Cumulative silica dustexposure (CDE) to onset of PSS for the cases wascalculated by multiplying the number of shiftsworked in a given occupation by a weighting factorproportional to the silica dust concentration of that

Sluis-Cremer, Hessel, Nizdo, Churchill, and Zeiss

occupation.8 The weighting factors were: 12 forunderground high dust; six for undergroundmoderate dust; three for underground low dust andall surface dusty occupations; two for occupations onnon-gold, silica containing mines; and one for serviceon mines containing little silica, dusty factories andquarries, and non-dusty occupations on gold mines.The weighting factors for non-gold mining servicewere based on the available information on the silicacontent of the dusts, mining techniques, and workprocesses. Whereas the weighting factors for theseexposures represent a greater degree ofapproximation than for the gold mines, little of theservice recorded for the study subjects was performedoutside the large gold mines. In addition to CDE thetotal number of shifts and the weighted average dustintensity (weighted by the proportion of shifts at eachlevel of intensity) were also compared for cases andcontrols. CDE, shifts, and average dust intensity forthe controls were calculated up to the date of onset ofPSS of the matched cases.The association between case-control status and

the presence of silicosis was evaluated usingMcNemar's chi-square test and the exact 95%confidence interval was determined from the formulaof Liddell.'2 The odds ratio was adjusted for theeffects of CDE using conditional logistic regressiontechniques.13 Because CDE, shifts, and intensity werenot distributed normally, the Wilcoxon two-sampletest'4 was used to assess differences for thesevariables.

Results

Cases and controls did not differ for the presence ofsilicosis (table 1). The odds ratios did not differsignificantly from one when either control group was

used, either before or after controlling for the effectsof CDE. Both the odds ratios were substantiallyreduced when CDE had been controlled for.The data in table 2 show that the cases had higher

CDE to onset of PSS than the controls, and that thisdifference was due to higher intensity of exposure

rather than to longer duration of service. This was

true when either the mixed control group or thepopulation controls were used for comparison,though the differences were more pronounced whenthe population controls were used. Within the mixed

Table I Matched comparison ofcases and controlsfor silicosisMixed controls Population controlsSil+ Sil- Sil+ Sil-

Cases Sil + 1 9 0 10Sil- 7 62 4 65

Mixed controls: Odds ratio = 1 29 (95% CI: 0 43-409), when controlled for CDE, odds ratio = 0 67 (95% CI: 0-18-248).Population controls: Odds ratio = 2 50 (95% CI: 0-72-10-95), when controlled for CDE, odds ratio = 1 18 (95% CI: 0-26-538).

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from

Silica, silicosis, and progressive systemic sclerosis

Table 2 Comparison ofcases and controlsfor CDE, shifts,and average intensity ofexposure

Cases ControlsCDE:Mixed controls: 29 727 25030

Periodicals 25 188 18 472*Benefits 32983 29 735

Population controls 29 727 19 81 7tShifts:Mixed controls: 5 251 5 204

Periodicals 4831 5 346Benefits 5 552 5 103

Population controls 5251 5211Intensity of exposure:Mixed controls: 5-63 4-73*

Periodicals 5 31 3-43tBenefits 5-85 5-67

Population controls 5 63 3-94tFor cases and controls, n = 79. For mixed controls, n(periodicals)= 33, n(benefits) = 46.*p < 0 05; tp < 0 001.

control group the differences between cases andperiodical controls were more pronounced thandifferences between cases and benefit controls.CDE up to onset of silicosis for the case and

control silicotic subjects was also computed.Exposure was 39909 for the cases, 39077 for themixed controls (p = 0 82), and 45 049 for thepopulation controls (p = 0-36).

Discussion

Although the results obtained using the two controlgroups differed in degree, there was a consistentpattern of higher exposure to dust among the caseswith no association with silicosis. Of the twocomponents of dust exposure-intensity andduration-intensity was more important. Given thatthe results for the two control groups differed bydegree for dust exposure, it is reasonable to questionwhich results more accurately represent the truedifference. The data in table 2 give some indicationthat the distinctions made in classifying the cases forthe mixed control group were at least partiallysuccessful in distinguishing them by mode ofpresentation. Cases matched with benefit controlshad an average CDE of 32983 whereas for thosematched to periodical controls the average was 25 188(p = 0 04). It is possible, however, that since minersappearing for benefit examinations generally do soafter long careers of working in dust or after initialcertification for dust related diseases it may bedifficult to discern an effect of CDE or silicosis usingbenefits as controls. When differences in averageintensity of exposure were examined, it was foundthat for the cases and periodical controls in the mixedcontrol group, the cases had an average intensity 154times that of the controls. The ratio for thepopulation controls was 1-43. The data for these twocomparisons probably reflect the true difference. The

841

comparison using the benefit controls in the mixedcontrol group may not be valid given the highbackground levels of the risk factors among thesecontrols.PSS is generally considered to be an autoimmune

disease, although other possibilities have beensuggested 5; autoimmune phenomena are generallyprominent in PSS, however. Antinuclear antibodiesare commonly present and antibodies to Scl-70antigen, centromere, and nucleolar antigen arespecific for PSS. Rheumatoid factor and antismoothmuscle antibody may be present and immunecomplexes and complement may be present in theglomeruli. Abnormalities in T cell numbers(decreased) and function have also been found. Thedecrease in circulating T cells may be due to theirsegregation in the sclerodermatous lesions. It has alsobeen suggested that there is an inherited or acquireddefect in suppressor T cell function. 6

There appear to be associations between silicaexposure and other autoimmune diseases. Amongthese at least six instances of an association withdiffuse lupus erythematosus have been described,especially in those subjected to high intensity silicaexposure. This disease may have its onset before thedevelopment of radiological silicosis. 17 - 20

Studies of the immune status have been carried outon subjects who have been exposed to silica and whohave developed silicosis. Vigliani and Pernissuggested in 1963 that autoimmune phenomena wereinvolved in the pathogenesis of silicosis.2 ' DepressionofT cell function has also been found in men exposedto silica but without silicosis.20 Webster (personalcommunication) has found a significant occurrence ofhigh titres for rheumatoid factor and antinuclearfactors in South African gold miners with longexposure to silica but who had not developed silicosis.

Experimental studies of the immunological effectsof silica administered intratracheally or intravenouslyhave shown profound changes in both cellular andhumoral immune responses. These include pro-longation of skin allograft and bone marrow graftsurvival and depression of T cell function.2224As in many cases of rheumatoid arthritis25 and

some cases of SLE and PSS, the advent of the auto-immune disease has an unfavourable effect on theprogression and severity of the silicosis if present.

Freneaux et al described four cases of silicosis (allof whom had skin histological findings consistentwith scleroderma) in whom the onset of sclerodermachanged the x ray pattern of silicosis, the nodulesbecoming blurred and asymmetrical.26 In the 10 casesof PSS with silicosis in this study four had atypicalfeatures. Two in particular, although regularly exam-ined, showed a sudden, explosive onset of 3/3 pro-fusion of opacities. One had a normal radiograph in

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from

842

1971 and a reading of r3/3 A in 1974. Another had anormal radiograph in 1970 and was read as q3/3 A in1976. CDE up to onset of silicosis did not, however,differ significantly between cases and controls-thatis, the cases were not more susceptible to silicosis.

Genetic factors have been established in a widevariety of autoimmune diseases; for systemic lupuserythematosus (SLE) concordance for the disease inidentical twins is 70% and another sign of underlyingdisordered immunoregulation is illustrated by the factthat there is impaired function of suppressor T cells insome blood relations of the cases. There is somedegree of linkage of SLE with antigens DR2 and 3.27It could be surmised that the effect of silica on theimmune system acts as a trigger to precipitate theautoimmune disease.PSS appears to be far more common than SLE in

those exposed to silica, although SLE is more com-mon in the general population. At least five of the sixreported cases of SLE in those exposed to silica wereassociated with high intensity exposure. Intensity ofexposure was also important in our cases of PSS, butthe levels were probably far lower than thoseobtained in the environmental conditions describedfor the reported cases of SLE. We have no expla-nation for this.

Instances of familial occurrence of PSS are rare.Soppi et al described an instance of PSS in a sister anda brother and they studied six other siblings in thatfamily.28 Of the other siblings, one had antinuclearand anti-DNA antibodies and a positive Rose-Waalertest. One other sister also had antinuclear and anti-DNA antibodies. One sibling had slight arthralgiaand another, Raynaud's syndrome.

Emerit et al found an increased prevalence of chro-mosomal breakages in 29 patients with PSS and in 54blood relations compared with normal controls.29 Itis of great interest that 16 of the 54 blood relations hestudied reported Raynaud's syndrome and one hadSj6grens syndrome.No relation with HLA antigens of the A and B

locus30 or with any DR antigen3" has been detectedexcept possibly in the case of the CREST syndrome(increased prevalence of DRW3).32 Despite this, onemay speculate that there is an underlying geneticallydetermined constitutional diathesis predisposing aperson to PSS. Silica may, through its known pro-found effects on the immune system, precipitate thedisease in those predisposed.

In this study an association between PSS andcumulative silica dust dose to onset ofPSS was found;the association being primarily due to a difference inthe average intensity of exposure. No association wasseen between PSS and silicosis per se. Both Erasmus5and Rodnan et aP4 found that less than half their casesof PSS who were exposed to silica had in fact con-

Sluis-Cremer, Hessel, Nizdo, Churchill, and Zeiss

tracted silicosis (roentgenographically defined). Silicatherefore seems to exercise its action in those predis-posed to PSS by a mechanism other than is the case inthe pathogenesis of silicosis. This is despite the factthat some of the autoimmune phenomena detected inthe laboratory are similar in both PSS and silicosis.The fact that intensity of exposure is important sug-gests that doses of silica high enough to overwhelmthe lung associated lymphoid system allow appre-ciable amounts of silica access to other tissues,especially immunologically active tissues.

References

Kubo A, Higuchi Y, Miyake Y, Tsunematsu T, Saito T, InamineM. A case of scieroderma associated with silicosis (abstract).Geneva: International Labour Office, 1983:3. (CIS bibliography;pneumoconiosis; No 19.)

2Rodnan GP, Benedek TG, Medsger TA, Cammarata RJ. The asso-ciation of progressive systemic sclerosis (scleroderma) with coalworkers' pneumoconiosis and other forms of silicosis. AnnIntern Med 1967;66:323-34.

3Gunther G, Schuchard E. Silikose und progressive sklerodermie.Dtsch Med Wochenschr 1970;95:467-8.

4Devulder B, Plouvier B, Martin JC, Lenoir L. The associationscleroderma-silicosis or Erasmus' syndrome. Nouv Presse Med1977;6:2877-9.

Erasmus LD. Scleroderma in goldminers on the Witwatersrandwith particular reference to pulmonary manifestations. SouthAfrican Laboratory and Clinical Medicine 1957;3:209-31.

6Sluis-Cremer GK, Erasmus LD, Hins S, Ludwin SK, Webster 1,Sichel HS. Report on incidence of scleroderma (progressive sys-temic sclerosis) in white South African gold miners. Pretoria:Government Printer, 1973. (National Research Institute forOccupational Disease report 9/73.)

Medsger TA, Masi AT. Epidemiology of systemic sclerosis (sclero-derma). Ann Intern Med 1971;74:714-21.

8Beadle DG, Harris E, Sluis-Cremer GK. The relationship betweenthe amount of dust breathed and the incidence of silicosis. In:Shapiro HA, ed. Pneumoconiosis: proceedings of the inter-national conference Johannesburg 1969. London: Oxford Univer-sity Press, 1970:473-7.

9Subcommittee for scleroderma criteria of the American Rheu-matism Association Diagnostic and Therapeutic Criteria Com-mittee. Arthritis Rheum 1980;23:581-90.

10 South African Medical Bureau for Occupational Diseases. Annualreport 1982/3. Pretoria: Government Printer, 1983.

International Labour Organisation. International classification ofradiographs ofpneumoconioses. Geneva: ILO, 1980.

12 Liddell FDK. Simplified exact analysis of case-referent studies:matched pairs, dichotomous exposure. J Epidemiol CommunityHealth 1983;37:82-4.

3 Breslow NE, Day NE. Statistical methods in cancer research. Vol 1.The analysis of case-control studies. Lyon: IARC, 1980.

'4Gibbons JD. Nonparametric statistical inference. Tokyo: McGraw-Hill Kogakusha, 1971.

"Lee EB, Anhalt GJ, Voorhees JJ, Diaz LA. Pathogenesis of sclero-derma. Int J Dermatol 1984;23:85-9.

16 Postlethwaite AE, Kang AH. Pathogenesis of progressive systemicsclerosis. J Lab Clin Med 1984;103:506-10.

" Bernardini P, lannoconi A. Pulmonary silicosis associated withsystemic lupus erythematosus. Lavoro Umano 1982;30:8-16.

'8Suratt PM, Winn WC, Brody AR, Bolton WK, Giles RD. Acutesilicosis in tombstone sandblasters. Am Rev Respir Dis1977;115:521-9.

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from

Silica, silicosis, and progressive systemic sclerosis9 Bailey WC, Brown M, Buechner HA, Weill H, Ichinose H, Ziskind

M. Silico-mycobacterial disease in sandblasters. Am Rev RespirDis 1974;I10:115-25.

20Miossec P, Youinou P, Cledes J, et al. Lowered FcIgG receptorbearing T lymphocytes correlate with non-organ-specific auto-antibodies in silicosis. Int Arch Allergy App! Immunol1 984;73:2 12-5.

21Vigliani EC, Pernis B. Immunological aspects of silicosis. AdvTuberc Res 1963;12:230-72.

22Uber CI, Reynolds RA. Immunotoxicology of silica. CRC CritRev Toxicol 1982;10:303-19.

23 de Shazo RD. Current concepts about the pathogenesis of silicosisand asbestosis. J Allergy Clin Immunol 1982;70:41-9.

24Reichrtova E. Macrophage immune function: a screening for silicadust toxicity. In: Proceedings of the Vlth International Pneu-moconiosis Conference 1984. Vol I. Bochum: Bergbau-Berfsgenossen, 548-59.

2SCaplan A. Certain unusual radiological appearances in the chestsof coal-miners suffering from rheumatoid arthritis. Thorax1953;8:29-37.

843

26Freneaux B, Freneaux B, Buyse N, Le van Quyen H. Associationbetween silicosis and scleroderma. Nouv Presse Med1975;4:2087-90.

2Schoenfield Y, Schwartz RS. Immunologic and genetic factors inautoimmune disease. N Engl J Med 1984;311:1019-29.

28Soppi E, Lehtonen A, Toivanen A. Familial progressive systemicsclerosis (scleroderma): immunological analysis of two patientsand six siblings from a single kindred. Clin Exp Immunol1 982;50:275-82.

29Emerit 1, Housset E, Feingold J. Chromosomal breakage andscleroderma. Studies in family members. J Lab Clin Med1976;88:81-6.

30Mijsky, Kobikova M, Stava Z. HLA and scleroderma. TissueAntigens 1979;14:359-60.

31 Niks M, Rovensky J, Nyulassy S, Buc M, Stefanovic J, Zitnan D.Lack of association of HLA-DR antigens with progressive sys-temic sclerosis (scleroderma). Tissue Antigens 1982;19:238-9.

32Germain BF, Espinoza LR, Bergen LL, Vagesh M, Vasey FB.Increased prevalence of DRW3 in the CREST syndrome.Arthritis Rheum 1981;24:857-9.

Vancouver style

All manuscripts submitted to the Br J Ind Medshould conform to the uniform requirements formanuscripts submitted to biomedical journals(known as the Vancouver style).The Br J Ind Med, together with many other

international biomedical journals, has agreed toaccept articles prepared in accordance with theVancouver style. The style (described in full in BrMed J, 24 February 1979, p 532) is intended tostandardise requirements for authors.

References should be numbered consecutivelyin the order in which they are first mentioned inthe text by Arabic numerals above the line oneach occasion the reference is cited (Manson'confirmed other reports2-5 .. .). In future refer-ences to papers submitted to the Br J Ind Medshould include: the names of all authors if there

are six or less or, if there are more, the first threefollowed by et al; the title of journal articles orbook chapters; the titles of journals abbreviatedaccording to the style of Index Medicus; and thefirst and final page numbers of the article orchapter.Examples of common forms of references are:

'International Steering Committee of Medical Editors.Uniform requirements for manuscripts submitted tobiomedical journals. Br Med J 1979; 1: 532-5.

2 Soter NA, Wasserman SI, Austen KF. Cold urticaria: releaseinto the circulation of histamine and eosinophil chemotac-tic factor of anaphylaxis during cold challenge. N Engl JMed 1976;294:687-90.

Weinstein L, Swartz MN. Pathogenic properties of invadingmicro-organisms. In: Sodeman WA Jr, Sodeman WA, eds.Pathologic physiology: mechanisms of disease. Philadel-phi W B Saunders, 1914:457-72.

copyright. on A

ugust 3, 2020 by guest. Protected by

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.42.12.838 on 1 D

ecember 1985. D

ownloaded from