41

have antibody to this agent and such antibodies are clearly notprotective. In Denmark, cryoprecipitate concentrate may be

purchased from the United States or European commercial sources.However, since the cryoprecipitate concentrate commerciallymanufactured in Europe uses large quantities of sera purchasedfrom the United States, the distinction about place of manufactureis likely to be important only if it is also certain that sera used in themanufacture are obtained solely from populations at low risk ofAIDS.Clinicians caring for haemophiliacs should consider alternative

forms of therapy for the care of new patients not yet exposed tocryoprecipitate concentrate. We anticipate that, with developmentsin retrovirus research, exposure to LAV through cryoprecipitatemay be only a temporary situation, but until screening tests ortechniques to neutralise the agent are in widespread use

commercially available cryoprecipitate products should beconsidered as probably contaminated.

Institute of Cancer Research,Radiumstationen,8000 Aarhus C, Denmark

Environmental Epidemiology Branch,National Cancer Institute,Bethesda, Maryland, USA

M. MELBYE

Institut Pasteur,Pans, France

Aarhus Municipal Hospital

Institute of Cancer Research, Aarhus

R. J. BIGGAR

J. C. CHERMANNL. MONTAGNIER

S. STENBJERG

P. EBBESEN

1. Menitove JE, Aster RH, Casper JT, et al. T-lymphocyte subpopulations in patientswith classic hemophilia treated with cryoprecipitate and lyophilized concentrate. NEngl J Med 1983; 308: 83-86.

2 Rickard KA, Joshua DE, Campbell J, Wearne A, Hodgson J, Kronenberg H. Absenceof AIDS in haemophiliacs in Australia treated from an entirely voluntary blooddonor system. Lancet 1983; i: 956-58.

3. Barré-Sinoussi F, Chermann JC, Rey F, et al. Isolation of T-lymphotropic retrovirusesfrom a patient at risk for acquired immune deficiency syndrome. Science 1983; 220:868-70.

4. Brun-Vezinet F, Rouzioux C, Barre-Sinoussi F, et al. Detection of IgG antibodies tolymphadenopathy-associated virus in patients with AIDS or lymphadenopathysyndrome. Lancet 1984; i: 1253-56.

5. Gallo RC, Salahuddin SZ, Popovic M, et al. Frequent detection and isolation ofcytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS.Science 1984; 224: 500-03.

IDIOTYPES AND AIDS

SIR,-Dr Hsia and colleagues’ hypothesis (June 2, p 1212) is basedon a concept which is highly improbable, on current knowledge ofidiotypes. They propose expression of an identical major cross-reacting idiotype (IdX) on antibodies directed against viruses orsperm (VST antibodies) in different individuals.

It is presumed that 1% of the total immunoglobulin repertoiremay bear this idiotype. Suppose VST antibodies make up 2% of thetotal antibody repertoire. The alleged IdX would then be expressedon 50% of VST antibodies. This, however, is unlikely. Availableevidence indicates extreme idiotypic heterogeneity of immuneresponses,1-3 especially for immune responses against compleximmunogenic particles such as viruses or sperm. There are majorcross-reacting idiotypes-the IdX on 20-70% of anti-arsenateantibodies in A/J mice4 is but one example. That, however, is arestricted immune response against a hapten and is very differentfrom the immune response against a human pathogen whereidiotypic heterogeneity, not idiotypic restriction, is the rule.3 3

Expression of IdX in the mouse is genetically controlled.5 In theunlikely event of a dominant Id on VST antibodies in a givenperson, it is even less probable that a genetically distinct person (inHsia’s hypothesis a contact with acquired immunodeficiencysyndrome) would be capable of expressing an identical IdX. This isrequired, though, if the IdX is significantly to upsetimmunoregulation, since this implies the existence of IdX and anti-IdX bearing regulatory T lymphocytes. BMoreover, there is little to suggest that manipulation of idiotypes

significantly alters overall immunocompetence. Even the specificimmune response is usually maintained after, for example,suppression of an IdX. In the anti-arsenate response of A/J mice,

suppression of the major IdX can be achieved by anti-IdXantibodies. This, however, is associated with an increase of anti-arsenate antibodies bearing other idiotypes. Thus the anti-arsenateantibody titre is maintained. Idiotype expression can readily bemanipulated with anti-idiotype, but the overall level of the immuneresponse cannot.

Idiotype-specific immunoregulation is a fact. It provides a

mechanism for maintaining a highly dynamic antigen-specificimmune response. Idiotypic heterogeneity would seem to bebeneficial to that effect. 7 Existence of a major IdX in the immuneresponse to a pathogen could probably interfere with

pathogenesis-but the chances that such an IdX exists in anantiviral or anti-sperm immune response in genetically distinctindividuals are very small indeed.

Laboratory of Neurochemistry,Born Bunge Foundation,University of Antwerp,B-2610 Wilrijk, Belgium JAN GHEUENS

1. Oudin J, Michel M. Une nouvelle forme d’allotypie de globulines gamma du serum dulapin apparement ee à la fonction et à la specificité anti-corps. C R Acad Sci (Paris)1963, 257: 805.

2. Kunkel HG, Mannik M, Williams RC Individual antigenic specificity of isolatedantibodies Science 1963; 140: 1218-19.

3 Urbain J, Wuilmart C, Cazenave PA Idiotypic regulation in immune networks.Contemp Topics Mol Immunol 1981; 8: 113-48.

4 Kuettner MG, Wang AL, Nisonoff S. Quantitative investigations of idiotypicantibodies VI Idiotypic specificity as a potential genetic marker for the variableregions of mouse immunoglobulin polypeptide chains. J Exp Med 1972; 135:579-95.

5. Eichmann K. Expression and function of idiotypes on lymphocytes. Adv Immunol1978; 26: 195-254

6. Nisonoff A, Shyr-Te JU. Relation of a cross reactive idiotype to genetic control of theimmune response. Fed Proc 1977; 36: 207-13.

7 Gheuens J, McFarlin DE, Rammohan KW, Bellini WJ. Idiotypes and biologicalactivity of murine monoclonal antibodies against the hemagglutinin of measlesvirus. Infect Immun 1981; 34: 200-07.

AIDS AND AUTHORS

SlR,—I am sure that many scientists believe that the elucidation ofthe epidemiology, cause, and pathogenesis of the acquiredimmunodeficiency syndrome (AIDS) will advance our knowledge ofimmunology considerably. AIDS appeared out of the blue a fewyears ago and, apart from causing immunodeficiency, it has beenresponsible for two other syndromes-the "minimum publishableunit syndrome" (MPUS) and the "how many authors can I cramonto one paper syndrome" (HMACICOOPS). These syndromesmay well be responsible for as many deaths as AIDS itself. Manyimportant medical papers must have been squeezed out by theinterminable reporting of AIDS, and, more importantly, a great dealof useful and potentially more beneficial research has not beenfunded or carried out because so many scientists have jumped on theAIDS bandwagon knowing that most of their work, whatever theresults, will be published in reputable journals, which seem to beAIDS struck.

In your June 23 issue (p 1415) 18 authors publish a letter statingthat they are doing an experiment with 23 monkeys. I am pleased tohear that all the monkeys are well and send them my regards, but Icannot believe that it takes 18 people to do these experiments on 23monkeys or that 18 people in six centres can write a letter.

It is this sort of publication that has encouraged MPUS andHMACICOOPS to such an extent that they threaten to strangle ourjournals and stop good work being done or published. It is timejournals of international repute took a stand and stamped thesemalignant syndromes out.

Department of Medical Microbiology,University of Sheffield Medical School,Sheffield S10 2RX A. R. MELLERSH

DIAGNOSING FAMILIAL MEDITERRANEAN FEVER

SIR,-We agree with Dr Cattan and colleagues (May 19, p 1130)that in most cases of familial Mediterranean fever (FMF) thediagnosis is not difficult for clinicians familiar with the disease. Thisdoes not mean that an objective test would not be useful, and FMF,which is most often encountered among ethnic groups from theMiddle East and from Mediterranean countries, will often be a

42

diagnostic problem for clinicians unfamiliar with the disease. Evenwhen FMF is strongly suspected the patient is usually subjected toexhaustive and costly investigations to exclude diseases with similarpresentation: 25-55% of patients with FMF undergo unnecessarylaparotomies for "acute abdomen" and more than 10% have

repeated surgery.I-5 Large cities today are so cosmopolitan thatFMF may be encountered far away from traditional areas. The

availability of a single test, such as the metaraminol provocation test(March 24, p 656), could, in our opinion, facilitate early diagnosis.

Cattan et al note an apparent discrepancy between fever in naturalattacks (100%) and in metaraminol-induced attacks (29%). Webelieve that fever in FMF is a secondary symptom, reflecting theintensity of the inflammatory reaction in the affected serousmembrane, perhaps related to lysosomal degranulation of thepolymorphonuclear leucocytes; fever as an isolated symptom inFMF is very rare.4,6 The absence of fever in more than two-thirds ofmetaraminol-induced attacks indicates the very mild nature of theseattacks. We observed that induced attacks which were accompaniedby fever were more severe than those when fever was not present.Clinical findings, such as fever, friction rub, rebound tenderness,and arthritis, in induced attacks reflect the severity of the attack, just

, as they do in spontaneous FMF. In our series of 21 patientsobjective findings were seen in only 7. If abdominal tenderness isthought of as an objective sign, all of 21 patients challenged withmetaraminol had one or more objective responses to the test. 9 hadwhat Cattan calls paucisymptomatic manifestation of the disease:on metaraminol provocation, 2 had fever and abdominal tenderness,1 fever and friction rub, and the other 6 showed abdominaltenderness only.We agree that our test should be further assessed, but not just in

paucisymptomatic cases. We have now tested 80 patients, withoutserious side-effects. The metaraminol provocation test, in our

experience, is far less risky than intravenous pyelography orendoscopy, not to mention laparotomy.

Faculty of Medicine,Kuwait University,Kuwait

M. H. BARAKATN. I. EL-SOBKIA. O. EL-KHAWADK. A. GUMMAF. F. FENECH

1. Seigal S. Benign paroxysmal peritonitis. Ann Intern Med 1945; 23: 1-21.2. Heller H, Sohar E, Sherf I. Familial Mediterranean fever Arch Intern Med 1958; 102:

50-58.3. Armenian HK, Khashadurian AK. Familial paroxysmal polyserositis. Leb Med J 1973;

26: 605-14.4. Schwabe AD, Peters RS. Familial Mediterranean fever in Armenians. Medicine 1974;

53: 453-62.5. Bakir F. Periodic peritonitis: Present management and future prospects. Arch Intern

Med 1979; 139: 781-83.6. Sohar E, Gafni J, Pras M, Heller H. Familial Mediterranean fever: a survey of 470 cases

and review of the literature. Am J Med 1967; 43: 227-53.

EPIDEMIC YELLOW FEVER IN UPPER VOLTA

SIR,-An epidemic of yellow fever raged from mid-September tothe beginning of December, 1983, in south-east Upper Volta. Itspread over some 10 000 km2 of bushy savannah, affecting onlypeople living by forest galleries, notably those in the Peul ethnicgroup. The Centre Muraz in Upper Volta and the Pasteur Institutesin Ivory Coast and Senegal, in coordination with the Ministry ofHealth of Upper Volta and the World Health Organisation, havedone an epidemiological study of this epidemic.Transmission of the virus was by sylvatic vectors, essentially

Aedesfurcifer. 25 strains of yellow fever virus were isolated from 32batches of Aedes furcifer, and 1 strain from Aedes metallicus.There were 800-1700 deaths in a population estimated at

25 000-35 000, for a mortality rate of about 4%. 14 out of the 16post mortem hepatic samples examined showed histological lesionssuggestive of yellow fever and the yellow fever virus was isolatedfrom 11 samples. 5100 sera were studied and data suggested thatabout 50% of the population had anti-yellow fever specific IgM,pointing to recent infection. The case fatality rate was estimated at6-10%. 43 strains of yellow fever virus were isolated from bloodsamples.The epidemic quickly declined in the sylvatic area as climatic

conditions operated against the survival of vectors. A mass

vaccination campaign, rapidly instituted, prevented spread to

nearby towns. Vaccine (17D, Pasteur Institute, Dakar) provedeffective in the field.We have called this epidemic "intermediate sylvatic"

- intermediate because the virus circulated both in man andin monkey, sylvatic because it arose in a sylvatic environmentonly, involving only wild vectors reaching receptive humanpopulations that had only recently (within 10 years) becomeestablished in the area.

Centre Muraz is part of Organisation de Coordination et de Cooperationpour la lutte contre les grandes endemies (OCCGE).Centre Muraz,BP 153,

Bobo-Dioulasso,Upper Volta;and ORSTOMBobo-Dioulasso

Institut Pasteur,Abidjan, Ivory Coast

Institut Pasteur,Dakar, Sénégal

Ministry of Health,Upper Volta

Centre Muraz

D. BAUDON

J. ROUXP.GAZIN

M. LHUILLIER

J. F. SALUZZO

L. BIBANE

T. R. GUIGUEMDE

V. ROBERTA. STANGHELLINIJ. F. MOLEZ

J. L. SARTHOLI

M. CORNET

L. SOME

B. R. SOUDRE

CHLOROQUINE RESISTANT PLASMODIUMFALCIPARUM MALARIA IN NAMIBIA

SIR,-To two recent Lancet reports of chloroquine-resistantfalciparum malaria in visitors to Angola (June 2, p 1244; June 30, p1462) we would add two South African patients who acquiredclinical malaria while on chloroquine prophylaxis and did notrespond to therapeutic courses of chloroquine, both oral andparenteral. The patients had been stationed in northern Namibia forless than a year and both had also been in southern Angola for a shortperiod. Following this report we did an independent survey in lateMay, 1984, of local people living in central Ovambo, northernNamibia. Twenty people attending the outpatient department atthe Finnish Mission Hospital, Onandjokwe, with a pyrexial illnesswere confirmed to have P falciparum malaria. Six had moderateparasitaemias (1000-100 000 asexual parasites/1 blood); had noevidence of having taken 4-aminoquinolines, as shown by the Dill-Glazko test on their urine; and showed >10% of schizont

development in five control cultures and >5% in one. Schizont

development was completely inhibited at levels of 5 or 6 pmolchloroquine per well in four patients but not in the other two whentested by the Rieckman micromethod with minor modifications.Complete inhibition in these two blood samples was achieved onlyat 12 pmol chloroquine/well. We report this finding as in vitroconfirmation of the suspected presence of moderately chloroquineresistant P falciparum malaria in the indigenous population ofsouthern Africa.

Chloroquine continues to be the drug of choice for both

prophylaxis and therapy in this region. Medical practitionersadvising would-be visitors or treating returning visitors should,however, be alerted to the possibility of malaria in patients who havereliably practised chloroquine prophylaxis.

Full details of these studies will be published in the South African Medical_7ournaL

School of Pathology,South African Institute

for Medical Researchand University of the Witwatersrand,

Johannesburg 2000, South Africa

University of Pretoriaand South African Medical Service

MARGARETHA ISAÄCSONG. A. COX

W. L. SIELING

CRYPTOSPORIDIUM RELATED TO MEASLESDIARRHOEA IN RWANDA

SIR,-Cryptosporidium spp, the enteropathogenic protozoanparasite associated with diarrhoea in immunodepressed patients, isa frequent cause of diarrhoea in the tropics. The depression ofOKT3 and OKT4 cell population2 or the suppression of