dependency nursing for certain patients.

We do not yet know

which patients would benefit from such attention, but large
trials would help us to define the true incidence of morbidity
and its relation to the preoperative state and procedure.
Different specialists will need to cooperate to fill in the gaps
in our understanding of optional preoperative, perioperative,
and postoperative management in elderly patients. The
anaesthetist interested in geriatrics is not a subspecialist but
someone concentrating on the biggest challenge presented to
anaesthetists today. It is here that most can be learnt for the
benefit of patients of all ages.
Consultant Anaesthetist,
Northwick Park Hospital,
Harrow,
Middlesex HAL 3UJ

HAROLD T DAVENPORT

1 Lunn JM. Age anaesthesia research. London: Research into Ageing, 1991.
2 Buck N, DevIinHB, Luon JN, Vickers MD. Confidential enquiry into peni-operative deaths. Report of
Nuffield Provincial Hospitals Trust. London: Nuffield Provincial Hospitals Trust, 1987.
3 Seymour DG. Medical assessment of the elderly surgical patient. London: Croom Helm, 1986.
4 Seymour DG, Vaz FG. A prospective study of surgical patients. II Post-operative complications.
Age Ageing 1989;18:316-26.
5 Davis FM, Woolner DF, Frampton C, Wilkinson A, Harrison RT, Roberts MIS, et al. Prospective
multicentre trial of mortality following general or spinal anaesthesia for hip fracture surgery in
the elderly. BrJf Anaesth 1987;59:1080-8.
6 Scott NB, Kehlet H. Regional anaesthesia and surgical morbidity. Br Surg 1988;75:299-304.
7 Jeager MP, Gloss DD, Neff RK, Brink-Johnsen T. Epidural anesthesia and analgesia in high risk
surgical patients. Anesthesiology 1987;66:729-36.
8 Seeling W, Bruckmooser K-P, Hufner C, Kneitinger E, Rigg C, Rockemann M. No reduction in
postoperative complications by the use of catheterized epidural analgesia following major

surgery. Anaesthetist 1990;39:33-40.

9 Dodson ME. Modifications of general anaesthesia for the aged. In: Davenport HT, ed. Anaesthesia
and the aged patient. Oxford: Blackwell Scientific, 1988:204-30.
10 Duncan PG, Cohen MM. Postoperative complications: factors of significance to anaesthetic

practice. CanJ Anaesth 1987;34:2-8.

11 Djikovic JL, Hedley-Whyte J. Prediction of outcome of surgery and anesthesia in patients over 80.
JAMA 1979;212:2301-6.
12 Hosking MB, Warner MA, Lobdell CM, Offord KP, Melton MJ. Outcomes of surgery in patients
90 years of age and older. JAMA 1989;261:1909-15.

Corticosteroids and tuberculosis

For systemic symptoms and local pressure effects
Despite the newer regimens of antituberculous drugs guidelines from both the British and American thoracic societies
suggest that corticosteroids still have a role in the management of tuberculosis.'2 Corticosteroids are most commonly
used for their anti-inflammatory properties to treat either the
disease itself or occasionally the hypersensitivity reactions
that occur with antituberculous drugs.34 As a generalisation
corticosteroids may be tried in any patient in whom systemic
symptoms or local pressure effects are a problem, provided
that the patient is taking antituberculous drugs to which the
organisms are sensitive.
Thoracic disease is the commonest form of tuberculosis
encountered by medical staff. In pleural disease the systemic
and local responses to tuberculoprotein may be reduced by
treatment with corticosteroids, with a prompt improvement
in symptoms.56 Furthermore, steroids limit the organisation
of pleural exudates and subsequent fibrosis, reducing the
chest wall deformity and scoliosis that may develop in
children.57
The use of steroids in patients with tuberculosis of the
pulmonary parenchyma is more controversial. Many trials
have shown that corticosteroids promote weight gain, reduce
fever, and encourage radiographic clearing but have little
effect on sputum conversion, cavity closure, and pulmonary
function.89 Thus corticosteroids cannot be recommended for
routine use in pulmonary disease,8 10 especially as progressive
disease may occur in patients with resistant mycobacteria.9"'
In patients moribund from tuberculosis, many of whom will
have pulmonary disease, nothing is lost by adding corticosteroids to antituberculous treatment.3 810
Tuberculous adenitis rarely causes problems, although
lymphadenopathy may increase after starting antituberculous
treatment.'2 Especially in children enlarged intrathoracic
nodes may cause bronchial obstruction,'3 which treatment
with corticosteroids reduces; occasionally surgery is necessary. Corticosteroids have a role in the management of
pericardial tuberculosis, reducing the risk of adhesions and
subsequent constrictive pericarditis. 14
Several studies using both short and long term end points
have shown that corticosteroids increase survival and reduce
long term sequelae in patients with tuberculous meningitis,
although mortality and morbidity remain high.'" Thus cortiBMJ

VOLUME 303

12 OCTOBER 1991

costeroids should be given to all patients with tuberculous

meningitis.
Tuberculosis of the genitourinary system is uncommon,
but if it is wrongly treated progressive renal damage may
follow. Corticosteroids are helpful in reducing the distressing
symptoms of cystitis,'6 but their main role is in managing the
ureteric obstruction that may be present at diagnosis or
develop during treatment. Although the obstruction usually
responds to treatment with corticosteroids, recourse to
surgical intervention may occasionally be required. 7
Corticosteroids are frequently prescribed in gastrointestinal tuberculosis,8 although there is scant justification for their
use. Intestinal strictures probably rarely complicate gastrointestinal tuberculosis treated with standard antituberculous
drugs; corticosteroids may be helpful in reducing adhesions.'9
In miliary tuberculosis and tuberculosis of soft tissues, bones,
or joints, the role of corticosteroids is poorly defined and there
are no firm indications for their use. In elderly patients the
improvement in constitutional symptoms with corticosteroids
may be helpful, increasing mobility and thus maintaining
independence, although this remains unproved.
Patients with AIDS have an increased risk of developing
tuberculosis and, although the infection usually responds to
standard antituberculous drugs,20 may require a prolonged
course of treatment.2' The role of corticosteroids in these
patients needs further defining.22
Treating other medical conditions with corticosteroids may
reactivate previously untreated infection, and prophylaxis
with isoniazid is recommended for patients receiving two or
more weeks of treatment with corticosteroids,''" although this
is not always successful.23 Despite the widespread use of
corticosteroids tuberculosis developing during treatment
with steroids is uncommon.8 Patients with evidence of
previously untreated tuberculosis (a grade 2 positive result of
a Heaf test if unvaccinated; a grade 3 or 4 positive result after
BCG vaccination) who require regular or protracted immunosuppressive treatment should receive a short course of
standard antituberculous treatment.24
Whatever the indication, the easiest and preferred route of
administration of corticosteroids is by mouth; there is no
advantage in giving them either intravenously or at the site of
infection -for example, intrapleurally.5 In most patients once
871

daily administration is sufficient, though twice daily

corticosteroids may suppress fever more effectively.25 Rifampicin increases the metabolism of steroids thus the dosage of
corticosteroid must be doubled to achieve similar effects.26 As
with antituberculous treatment, the use of corticosteroids
should be supervised by a practitioner with an interest in the
condition.
MARTIN B ALLEN
Senior Registrar
NIGEL J COOKE
Consultant Physician

Department of Respiratory Medicine,

Leeds General Infirmary,
Leeds LS1 3EX.

8 Horne NW. A critical evaluation of corticosteroids in tuberculosis. Advances in Tuberculosis

Researcht 1966;15:1-54.
9 Tuberculosis Research Centre. Study of chemotherapy regimens of 5 and 7 months duration and
the role of corticosteroids in the treatment of sputum positive patients with pulmonary
tuberculosis in South India. l'ubercle 1983;64:73-91.
10 Committee on Tuberculosis Treatment. Adrenal corticosteroids and tuberculosis. Am Rezt Respir
Dis 1968;97:484-5.
11 Bethlem N, Gomes 0, Dias L. The effects of steroid treatment on pulmonary tuberculosis. Dis

Chest 1960;38:131-9.
12 Campbell IA, Dyson AJ. Lymph node tuberculosis: a comparison of various methods of treatment.
Tubercle 1977;58:171-9.
13 Nemir RL, Cardona J, Vaziri F Toledo R. Prednisolone as an adjunct in the chemotherapy of
lymph node-bronchial tuberculosis in childhood: a double blind study. II. Further term studies.
Am Rezv Respir Dis 1967;95:402-10.
14 Strang JIG, Kakaza HHS, Gibson DG, Allen BW, Mitchison DA, Evans DJ, et al. Controlled
clinical trial of complete open surgical drainage and of prednisolone in treatment of tuberculous
pericardial effusion in Transkei. Lancet 1988;ii:759-64.
15 Girling DJ, Darbyshire JH, Humphries MJ, O'Mahonev G. Extrapulmonary tuberculosis.
BrMedBull 1988;44:738-56.
16 Gow JG. The management of genitourinary tuberculosis. 7 Antimicrob Chemother 1981;7:590-1.
17 Horne NW, Tulloch WS. Conservative management of renal tuberculosis. Br J Urol

1975;47:481-7.
I American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and
children. Am Rev Respir Dis 1986;134:355-63.
2 Joint Tuberculosis Committee of the British Thoracic Society. Control and prevention of
tuberculosis in Britain: an updated code of practice. BM_ 1990;300:995-9.
3 Ellis ME, Webb AK. Cause of death in patients admitted to hospital for pulmonary tuberculosis.

Lancet 1983;i:665-7.
4 Thompson JE. The management of hypersensitivity reactions to antituberculous drugs. MedjAust
1969;ii: 1058-63.
5 Menon NK. Steroid therapy in tuberculous pleural effusion. Tubercle 1964;45:17-20.
6 Lee CH, Wang WJ, Lan RS, Tsai YH, Chiang YC. Corticosteroids in the treatment of tuberculous
pleurisy. A double-blind, placebo controlled, randomised study. Chest 1988;94:1256-9.
7 Palev SS, Mihaly JP, Mais EL, Gittens SA, Lupini B. Prednisolone in the treatment of tuberculous
pleural effusions. American Review of Tuberculosis 1959;79:307-14.

18 Bastani B, Shariaztadeh MR, Dehdashti F. Tuberculosis peritonitis-Report of 30 cases and

review of the literature. QJ.MIed 1985;56:549-57.
19 Klimach OE, Ormerod LP. Gastrointestinal tuberculosis: a retrospective review of 109 cases in a
district general hospital. QJ Med 1985;56:569-78.
20 Darbyshire JH. HIV infection and tuberculosis. QJMed 1991;291:563-4.
21 Goldman KP. AIDS and tuberculosis. BM 1987;295:511-2.
22 Masaud T, Kemp E. Corticosteroids in treatment of disseminated tuberculosis in a patient with
HIV infection. BMJ 1988;296:464-5.
23 Marschke G. Tuberculosis, isoniazid and corticosteroids. N Engly Med 1969;280:1245.
24 Milar JW, Horne NW. Tuberculosis in immunosuppressed patients. Lancet 1979;i: 1176-8.
25 Chung KF, Snashall PD. Suppression of prolonged fever during treatment of pulmonary
tuberculosis: importance of using twice versus single daily dose of prednisolone. Postgrad MedJ7
1983;59:373-5.
26 McAllister WA, Thompson PJ, Al-Habet SM, Rogers HJ. Rifampicin reduces effectiveness and
bioavailabilitv of prednisolone. B.J 1983;286:923-5.

Monkey business over AIDS vaccine

Problems with the paradigm
Over the past month the AIDS research community has been
buzzing with consternation over the unexpected and perplexing research at the United Kingdom's National Institute
of Standards and Control. Led by Dr Jim Stott, a team of
investigators at the institute reported its surprising findings at
a Medical Research Council AIDS conference last month and
in a letter in last week's Nature. ' Like several other groups in
the United States, Stott et al had previously reported, with
much fanfare, that macaque monkeys could be protected
against infection with simian immunodeficiency virus (SIV), a
close relative of HIV, by prior immunisation with purified
virus inactivated by formalin or with inactivated cells that
had been infected with SIV.2 Now they have shown that
immunisation with similar preparations of uninfected cells
affords partial protection too. '
This is a totally unexpected result. Although it is well
known that enveloped viruses pick up cellular antigens when
budding from the cell membrane, host cell components have
not previously been shown to afford protection against viral
infection. It means that strategies for producing vaccines
against HIV require re-evaluation, and further basic studies
on the mechanism of protection are needed.
Stott et al find that protection against SIV infection does not
parallel titres of neutralising antibodies to SIV in the immunised monkeys but does correlate with titres of anticellular
antibodies. It should be borne in mind that these results are
preliminary, having been obtained in just four monkeys in
each vaccine group (plus and minus viral antigens in the
immunogen), and that the cellular material was of human, not

simian, origin.
Taken aback by these results, some AIDS researchers have
criticised the group at the institute for not having explored the
effects of control vaccine preparations earlier. On the other
hand, Stott et al are the first group in the world to have used
872

controls in this way and to have realised what they had

stumbled on. Furthermore, the minimum number of monkeys
were used for humanitarian and economic considerations. It
would surely have been unethical to conduct "placebo"
vaccine tests until there was a good indication that the
experimental vaccine containing the virus exhibited some
protective activity. So the research team really deserves credit
for proceeding in an orderly way and in alerting other AIDS
researchers as soon as it obtained its surprising results.
Why human T lymphocytes inactivated by formalin help to
protect macaques from SIV infection remains to be elucidated.
If the end point was the development of clinical AIDS recent
hypotheses on an autoimmune component in the pathogenesis
of HIV and SIV might be invoked.34 But in this case the
efficacy of the experimental vaccines was determined
from early evidence of infection by virus isolation, genome
detection, and seroconversion. Contrary to the view expressed
in an editorial in Nature,5 the results of Stott et al pertain
neither to the aetiology of AIDS nor to its pathogenesis. They
do, however, show that vaccination will be a more complex
process than many were previously led to believe.
ROBIN WEISS

Head, Chester Beatty Laboratories,

Institute Cancer Research,
London SW3 6JB

1 Stott EJ, Kitchin PA, Page M, Flanagan B, Taffs LF, Chan WL, et al. Anti-cell antibodies in
macaques. Nature 1991;353:393.
2 Stott EJ, Chan WL, Mills KHG, Page M, Taffs F, Cranage M, et al. Preliminary report: protection
of cynomolgus macaques gainst simian immunodeficiency virus by fixed infected-cell vaccine.
Lancet 1990;336: 1538-41.
3 Habershaw JA, Dalgleish AG. The relevance of HIV env/CD4 interactions of the pathogenesis of
acquired immunodeficiency syndrome. Acquir Immune Defic Syndr 1989;2:457-68.
4 Kion TA, Hoffmann GW. Anti-HIV and anti-ANTI-MHC antibodies in alloimmune and
autoimmune mice. Science 1991;253:1138-40.
5 Maddox J. AIDS research turned upside down. Nature 1991;353:297.

BMJ

VOLUME

303

12

OCTOBER

1991