Postgrad Med J (1993) 69, 832- 834
© The Fellowship of Postgraduate Medicine, 1993
Letters to the Editor
Recovery of adrenocortical function following treat-
ment oftuberculous Addison's disease
Penrice and Nussey report two cases ofAddison's disease
in which adrenocortical function recovered following
treatment of tuberculosis.' Adrenal glands may be
enlarged or atrophic in patients with Addison's disease
due to t iberculosis. Large glands in tuberculosis mean a
recent and probably active infection requiring treatment,
whereas small calcified glands are in favour ofremote and
probably inactive infection.2 Renner et al. reported two
cases ofadrenal tuberculosis in which adrenal glands had
been enlarged and they concluded that computerized
noninvasive diagnosis of tuberculous adrenalopathy.3 In
another report, the finding ofenlarged adrenal glands on
CT suggested the presence ofearly tuberculosis or in rare
instances other potentially treatable diseases.4
Barnes et al. investigated adrenal function in 90
patients with active tuberculosis (30 pulmonary, 30
miliary, 30 extrapulmonary) before and after starting
anti-tuberculous therapy. They found some degree of
adrenal dysfunction in seven patients. After the treat-
ment, the Synacthen response returned to normal in all
but one patient, and they concluded that adrenal dysfunc-
tion is an uncommon problem in patients with active
favourable effect on adrenal function.'
Recently we have reported a patient with active
pulmonary tuberculosis, acute adrenal failure and an
enlarged adrenal mass demonstrated by CT.6 The mass
was removed surgically and histopathologic examination
disclosed adrenal tuberculoma. We have also shown that
adrenal glands are larger in acute pulmonary tuberculosis
than that in chronic tuberculosis and also healthy sub-
jects.' CT scanning was not carried out in the patients
reported by Penrice and Nussey. Since the symptoms and
signs ofAddison's disease appear after more than 90% of
the glands have been destroyed by tuberculosis,8 we think
that recovery of adrenal insufficiency is not possible in
patients with Addison's disease due to remote tuber-
culosis in which adrenal glands are atrophic and calcified.
In contrast, recovery of adrenocortical function after
anti-tuberculous therapy in reported patients with tuber-
culosis is an expected outcome.2.9
We suggest that CT ofadrenal glands should be carried
out in every patient presenting with Addison's disease
thought to be caused by tuberculosis. If there is adrenal
atrophy anti-tuberculous therapy may not be required. If
adrenal glands are enlarged, anti-tuberculous therapy
may be needed.
( T) is the method of choice for the
1. Penrice, J. & Nussey, S.S. Recovery ofadrenocortical function
following treatment of tuberculous Addison's disease. Post-
grad Med J 1992, 68: 204-205.
2. McMurry, J.F., Long, D., McLure, R. & Kotchen, T.A.
Addison's disease with adrenal enlargement on computed
tomographic scanning. Report oftwo cases oftuberculosis and
review of the literature. Am J Med 1983, 77: 365-368.
3. Renner, F., Karnel, F., Graninger, W. & Imhof, H. Morbus
addison mit sonographisch und computertomographisch
uberlegungen and hand zweiner
adrenalitis. Wien Klin Wochenschr 1986, 98: 244-249.
4. Vita, J.A., Silverberg, S.J., Goland, R.S., Austin, J.H.M. &
Knowlton, A.I. Clinical clues to thecauseofAddison's disease.
Am J Med 1985, 78: 461-466.
5. Barnes, D.J., Naraqi, S., Temu, P. & Turtle, J.R. Adrenal
function in patients with active tuberculosis. Thorax 1989, 44:
6. Keletimur, F., Ozbakir, O., Saglam, A., Ozturk, F. &
Yiucesoy, M. Acute adrenocortical failure due to tuberculosis.
J Endocrinol Invest. 1993, 16: 281-284.
6.Kele§timur,F., Unlui, Y., Ozesmi, M. & Tolu, 1. A hormonal
and radiological evaluation of adrenal gland in patients with
acute and chronic pulmonary tuberculosis. J Endocrinol 1993,
137 (Suppl): P26 (Abstract).
8. Williams, G.H. & Dluhy, R.G. Disease of the adrenal cortex.
In: Braunwald, E. (ed.) Harrison's Principles of Internal
Medicine. McGraw Hill (International edition), Hamburg,
1987, pp. 1753- 1774.
9. Ellis, M.E. &Tayoub, F. Adrenal function in tuberculosis. BrJ
Dis Chest 1986, 80: 7-12.
falle von tuberkuloser
Guillain-Barre syndrome in a patient with chronic
Guillain-Barre syndrome (GBS) has been frequently
especially those of the lymphoid system, for example,
Hodgkin's disease.' However, despite being the com-
monest lymphoid malignancy,
leukaemia has been associated with GBS in only three
reported patients.2'3 I describe a fourth case.
A 77 year old woman with known stage III B-cell
chronic lymphocytic leukaemia (CLL) presented with a
5-day history of progressive symmetrical limb weakness,
distal limb parasthesiae and breathlessness. There was no
history ofrecent viral infection. CLL had been diagnosed
3 months before and treated with chlorambucil as well as
a blood transfusion. Chlorambucil was stopped 8 weeks
later because ofmyelosuppression. Examination revealed
a frail, breathless, afebrile lady with moderate hepato-
splenomegaly. The cranial nerves were normal. She had a
flaccid quadriparesis with generally MRC grade 2 power,
absent tendon reflexes and flexor plantar responses. There
was impaired appreciation of vibration and propriocep-
tion below the ankles.
Investigation revealed a reduced forced vital capacity
of 1.2 litres. Routine biochemistry was normal. Her
haemoglobin was 1.5 g/dl and herlymphocyte count was
99.9 x 109/l. A direct Coombs' test was negative. Paired
group.bmj.com on July 13, 2011 - Published by pmj.bmj.com Downloaded from
LETTERS TO THE EDITOR 833
serology showed no evidence of recent viral or mycoplas-
mal infection. Serum IgA was slightly reduced at 1.17 g/l
(normal 1.25-4.25 g/l) and the IgM was 0.47 g/l (normal
0.5-1.75 g/l), with a normal IgG. A chest X-ray was
normal. Spinal fluid examination showed a protein of
0.7 g/l (normal up to 0.45 g/l), glucose 4.8 mmol/l, with
no white cells, and no growth on culture. Nerve conduc-
tion studies showed motor nerve conduction of the
median nerve was 30 m/second with a distal motor
latency of 4.7 milliseconds (normal, up to 4.2 mill-
seconds). F responses were unobtainable. Distal sensory
nerve conduction velocity of median nerve was 29
m/second, the sensory action potential being reduced at
1.5 jLV. The extensor digitorum muscle was denenervated
and motor conduction along the peroneal nerve could not
be obtained. Motor latency from the capitulum fibulae to
the tibialis anterior was slightly prolonged, but the M
wave was markedly decreased (0.1 mV).
She received a 5-day course of intravenous immuno-
globulin (Venoglobulin, Alpha Pharmaceuticals) at 0.4 g/
kg/day, and steadily improved. After 3 weeks she was able
to walk unaided.
for GBS.' GBS
demyelinating polyradiculoneuropathy in which both
protein and lipid antigens in peripheral nerve are the
target ofimmune attack. Neither the precise antigens nor
the relative roles of humoral or cell-mediated immune
responses are known.' A prominent role of the humoral
immune system in the pathogenesis ofGBS is supported
both by the presence during the acute phase of the illness
of complement-fixing anti-peripheral nerve myelin IgM
antibodies whose kinetics correlate with the clinical
course ofthe disease, and by the efficacy oftreatment with
high-dose intravenous immunoglobulin.4'5
Under certain circumstances
immunosuppression could serve as a contributing factor
in precipitating GBS.6 Impaired cell-mediated immunity
is the proposed mechanism for the increased incidence of
GBS in Hodgkin's disease.6 Indeed, Hughes et al.
reported reduced suppressor T-cell function in about a
quarter of patients during the early phase of GBS.7
In CLL, in contrast with Hodgkin's disease, almost all
patients eventually develop hypogammaglobulinaemia
while the cell-mediated immune system remains largely
intact.89 This may explain why the association between
CLL and GBS is rare. In our patient it is possible that the
transient myelosuppression induced by the chlorambucil
may have predisposed to the development ofGBS, whilst
the almost normal humoral response may have mediated
its subsequent development.
fulfills the well-established diagnostic
is an acute inflammatory
Nottingham NG7 2UH, UK.
Lambert, E.H. & Bunge, R. (eds). Peripheral Neuropathi, 2nd
edn. W.B. Saunders, Philadelphia, 1984, pp. 2050-2100.
2. Lorenz de Haas, M.A.M. Syndrome de Landry dans un cas de
leukemie. Rev Neurol 1951, 85: 306-307.
P.J., Thomas, P.K.,
3. Powles, R.L. & Malpas,
associated with chronic lymphocytic leukaemia. Br Med J
1967, 3: 286-287.
4. Koski, C.L., Gratz, E., Sutherland, J. & Mayer, R.F. Clinical
correlation with antiperipheral nerve myelin antibodies in
Guillain-Barre syndrome. Ann Neurol 1986, 19: 573-577.
5. van der Meche, F.G.A., Schmitz, P.I.M. & The Dutch
Guillain-Barre Syndrome Study Group. A randomised trial
exchange in Guillain-Barre syndrome. N Engl J Med 1992,
6. Lisak, R.P., Mitchell, M., Zweimann, B., Orrechis, E. &
Asbury, A.K. Guillain-Barre syndrome and Hodgkin's
disease: three cases with immunological studies. Ann Neurol
1977, 1: 72-78.
7. Hughes, R.A.C., Aslan, S. & Gray, I.A. Lymphocyte sub-
polyradiculoneuritis (Guillain-Barre syndrome). Clin E.xp
Immunol 1983, 51: 448-454.
8. Gale, R.P. & Foon, K.A. Chronic lymphocytic leukaemia: new
insights into biology and therapy. Ann Intern Med 1990, 113:
9. Slivnick, D.J., Ellis, T.M., Nawrocki, J.F. & Fisher, R.I. The
impact of Hodgkin's disease on the immune system. Semin
Oncol 1990, 17: 673-682.
J.S. Guillain-Barre syndrome
Anti-epileptic therapy and fatal chickenpox
A 15 year old epileptic boy on carbamazepine, primidone,
sodium valproate, acetazolamide and clobazam present-
ed with severe backache following a grand mal seizure.
An X-ray revealed no abnormality and he was sent home
on ibuprofen. Two days later he re-attended with fever,
anorexia, vomiting and a rash which had appeared the
previous night. He had recently been exposed to chicken-
pox. On examination his temperature was 37.8°C and he
had a florid papular, vesicular and purpuric rash involv-
ing his face, trunk and proximal limbs. Ecchymoses were
present at a venepuncture site. His pulse was 120/minute,
blood pressure 165/110 mmHg, the abdomen was dis-
tended and tender, and the urine and vomitus contained
blood. Investigations showed: haemoglobin 16.3 g/dl,
white cell count 17.0 x 109/l, platelets 49 x 109/l, interna-
tional normalized ratio > 8 (normal 0.8-1.2), activated
partial thromboplastin time ratio 3.4 (normal 0.8-1.2),
fibrin degradation products (D-dimer) 32.0 mg/l (normal
< 0.5 mg/l) and glucose 6.7 mmol/l.
Disseminated intravascular coagulation (DIC) pre-
cipitated by a drug, meningococcal septicaemia or
chickenpox was considered. Despite infusions of fresh
frozen plasma, platelets and cryoprecipitate together with
intravenous acyclovir, benzylpenicillin and chloram-
phenicol, he died within 24 hours. Varicella Zoster virus
was isolated from vesicular fluid.
An explanation for the fulminant course was not
apparent. He was, however, on a number ofdrugs and the
possibility that the DIC was due in part to a drug effect
cannot be excluded. In this regard it is relevant that
carbamazepine and acetazolamide can induce autoim-
mune thrombocytopenia,' and sodium valproate can
cause various disorders including abnormal bleeding
times, impaired clotting,3 platelet antibodies,4 and low
fibrinogen levels.5 Fatal DIC has been reported in one
neonatal patient on sodium valproate, but it was not clear
whether causes other than the drug could have been
responsible for the DIC (personal communication from
group.bmj.com on July 13, 2011 - Published by pmj.bmj.comDownloaded from
1993 69: 832-833Postgrad Med J
in a patient with chronic
Updated information and services can be
article. at the top right corner of the online
boxarticles cite this article. Sign up in the
Receive free email alerts when new
To request permissions go to:
To order reprints go to:
To subscribe to BMJ go to:
group.bmj.com on July 13, 2011 - Published by pmj.bmj.comDownloaded from