Journal of Infection (I994) 28, 59-64

Cryptococcal meningit is in Lilongwe and Blantyre, Malawi

Dermot Maher* and Henry Mwandumbat

* Department of Medicine, Queen Elizabeth Central Hospital, P.O. Box 95, Blantyre, t Department of Medicine, Kamuzu Central Hospital,

P.O. Box 149, Lilongwe, Malawi

Accepted for publication 19 July 1993

Summary

Infection with Human Immunodeficiency Virus is widespread in Malawi and cryptococcal meningitis is a common problem in those with AIDS. A review of microbiology laboratory records in Lilongwe and Blantyre between July 1991 and January 1993 identified 31 patients with cryptococcal meningitis. Diagnosis was based on a positive India ink stain of CSF and/or culture of Cryptococcus neoformans.

There were 16 men (median age 38 years) and 15 women (median age 28 years) in the investigation. The median duration of symptoms was 2 weeks. The clinical presentation was varied, the most frequent features being headache (97 %), neck stiffness (74%), fever (61%) and altered consciousness (58%). CSF WBC count, glucose and protein concentrations were non-specific.

Most patients could not afford anti-cryptococcal chemotherapy and their median survival time after diagnosis was 4 days. Patients who could afford such treatment survived for up to several months. Diagnosis is useful for prognostic reasons and may save patients unnecessary treatment if tuberculous meningitis is the alternative diagnosis. Cryptococcal antigen detection tests may improve diagnostic accuracy.

The problem of cryptococcal meningitis is likely to become increasingly common as HIV infection becomes more widespread.

Introduction

Cryptococcus neoformans, an opportunis t ic pathogen, is an encapsulated yeast which is common in the general and domest ic environment . Crypto- coccosis is the commones t invasive fungal infection in those who are H I V - infected 1 and is an AIDS-de f in ing i l lness) Infect ion is by inhalation of a fungus-contamina ted aerosol, resulting in pu lmonary infection, which is self- limiting in the immunocompe ten t host. In the immunocompromised host, however , the initial pu lmonary infection may be fol lowed by haematogenous spread to other organs, including the brain and meninges. Cryptococcal meningit is (or, more strictly, meningo-encephali t is , since lesions are spread diffusely th roughou t the brain) is the most common presentat ion of cryptococcosis in patients with A I D S . a

Cryptococcal meningit is is common in those with A I D S in sub-Saharan Africa, occurr ing in be tween 6 and 12 % of patients. 4 Dur ing the past 5 years as H I V has become widespread, clinicians in Malawi have observed an increase in the n u m b e r of patients admit ted to hospital with cryptococcal meningitis. T h e main hospitals in Li longwe and Blantyre, Malawi ' s two

* Address correspondence to: Dr Dermot Maher.

oi63-4453/94/OlOO59+o6 $08.00/0 © I994 The British Society for the Study of Infection

60 D. M A H E R A N D H. M W A N D U M B A

Table I Frequency of clinical features in 3I patients with cryptococcal meningitis

Clinical features Number of patients (%)

Symptoms Headache 3o (97) Fever 19 (6I) Weight loss z5 (48) Mental disturbance I4 (45) Vomiting 9 (29) Convulsions 4 (x3)

Signs Neck stiffness 23 (74) Altered consciousness I8 (58) Wasting 15 (48) Fever 13 (42) Oral candidiasis 12 (39) Cranial nerve palsies 2 (6)

(2 bilateral sixth I unilateral sixth i unilateral seventh)

Kaposi's sarcoma 2 (6) Generalised lymphadenopathy 2 (6) Perianal warts I (3)

largest cities, are K a m u z u Central Hospi ta l ( K C H ) and Queen Elizabeth Central Hospi ta l ( Q E C H ) respectively. T h e aim of this s tudy is to review the clinical features, C S F findings, t reatment and outcome of patients with cryptococcal meningitis admit ted to K C H and Q E C H during I8 months be tween July I99I and January I993.

M e t h o d s

Cases of cryptococcal meningitis were identified retrospectively f rom the microbiology laboratory records in K C H and Q E C H from July I99I to January I993. A case was defined as one in which C. neoformans was identified in C S F either by a posit ive India ink stain or by culture. T h e microbiology laboratories per form these investigations only when reques ted by a clinician who suspects cryptococcal meningitis. T h e case records of patients with cryptococcal meningit is were reviewed to obtain the following information: demographic details, clinical findings, results of biochemical and micro- biological investigations of CSF , t reatment and outcome.

R e s u l t s

There were 3I patients with cryptococcal meningitis be tween July I99I and January I993. In 3o, C. neoformans was identified in C S F by a posit ive India ink stain, and in one the result of India ink stain was not recorded bu t C. neoformans was cul tured f rom the CSF . There were 16 men (median age 38

Cryptococcal meningitis in Malawi 6I

Tab le II Results of CSF biochemistry and WBC counts in 25 patients with cryptococcal meningitis

N u m b e r wi th abnormal finding

C S F Median Range N u m b e r tested (%)

Prote in concentrat ion (mg/d l ) 67 I5-428 > 45 20/25 (80)

Glucose concentrat ion (mg/d l ) 32 o-56 < 4 ° I9 /25 (76)

W B C count ( x Io6/1) 2o o--I2OO > 5 I9 /25 (76)

25 24 23 22 21 20 19 18 17

?. 16 • ~ 15

14 13 12

Z 8 7 6

4 3 [ 2 I 1

I i i 1 I 1 I ~ i t i i i I

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Survival after diagnosis (days)

Fig. i. Survival of patients without anti-fungal therapy following diagnosis of cryptococcal meningitis.

years, range 23-54 years) and z5 women (median age z8 years, range I9 -49 years). The median durat ion of symptoms was 2 weeks (range I day to 5 months) . Nine patients (29%) had a previous history of pu lmonary tuberculosis and three (IO %) gave a history of herpes zoster. Tab le I shows the f requency of clinical features in the 31 patients.

W e do not have complete results of C S F examinations. O f Io cases in which the naked eye appearance of the C S F was recorded, it was clear in 9 cases and c loudy in one. Biochemistry and W B C findings were recorded in z5 of 3I cases. T h e results are shown in Tab le II .

T h e median differential W B C count (L = lymphocytes , N = neutrophils) , g rouped in deciles, was 9 o % L / I o % N , range I o o % L / o % / N - I o % L / 90 ~o/1'4.

62 D. M A H E R A N D H. M W A N D U M B A

Table I I I Survival of patients without anti-fungal therapy following diagnosis of cryptococcal meningitis

Survival after diagnosis (days) Number of survivors

o 20

I Z 5

2 14

3 I3 4 zo 6 8 7 7 8 6

z i 5 z2 3

17 2 20 i

30 o

India ink stain was positive in 3 ° cases and not recorded in I case. The result of CSF culture for C. neoformans was recorded in 22 cases, of which I9 (86 %) were positive and 3 (z4 %) were negative.

Six (z9%) of the 3z patients received specific antifungal chemotherapy: four were treated with fluconazole, one with amphotericin and one with amphotericin followed by fluconazole. Of the four patients treated with fluconazole, two died (one after 4 months and one after 6 months) and two were alive in March I993 (after 4 months and 9 months). The patient treated with amphotericin died after 3 months and the one treated with amphotericin followed by fluconazole died after z month. A total of 25 (8z %) of the 3 I patients received no specific antifungal chemotherapy. Information concerning their survival was available in 2o cases (Fig. i ; Table III). Median survival time was 4 days (range I-3o days).

D i s c u s s i o n

Between July I99z and January I993, 3 z patients with cryptococcal meningitis were identified. During this z8 month period approximately 3o,ooo patients were admitted to the medical departments of Kamuzu and Queen Elizabeth Central Hospitals giving an incidence of cases of cryptococcal meningitis among medical admissions to the two hospitals of o-x % per year. This is an underestimate of the true incidence because of under-recognition of cases by clinicians and low sensitivity of laboratory diagnostic methods.

The median age of patients with cryptococcal meningitis was significantly greater for men (38 years) than for women (28 years). This reflects the age distribution of people with H I V / A I D S in Malawi. The age distribution is shifted towards an older group in men than in women, probably because men tend to have younger sexual partners than women. Many patients in Malawi tend to underestimate the duration of their symptoms, with the result that the median duration of 2 weeks is probably not accurate.

Cryptococcal meningitis in Malawi 63

The clinical presentation was varied. The most frequent features were headache (97 %), neck stiffness (74 %), fever (61%) and altered consciousness (58 %). Clinical differentiation of cryptococcal meningitis from meningitis due to other infections, malignant infiltration or sarcoid is unreliable. The abnormalities in CSF white blood cell count, and protein and glucose concentrations are non-specific, since similar abnormalities may occur in tuberculous meningitis, other bacterial meningitides, meningovascular syph- ilis and trypanosomal meningo-encephalitis. The definitive diagnosis of cryptococcal meningitis rests on culture of C. neoformans from the CSF. In this series of 31 cases, the CSF culture results were recorded in 22, of which three (14 %) were negative. Technical problems in the laboratory which may have led to false negative CSF culture results for C. neoformans include delay in setting up the culture and use of culture medium 'past its sell-by date'.

There are several ways of improving the detection of cases of cryptococcal meningitis. In view of the varied clinical presentation clinicians need to have a high index of suspicion. Given the unreliability of clinical diagnosis, the microbiology laboratory should perform India ink staining and culture for C. neoformans on every CSF sample submitted and not just on those from patients with a clinically suspected diagnosis of cryptococcal meningitis. Laboratory procedures can be improved, e.g. by reducing delay in setting up cultures, by avoiding the use of non-expired culture media and by better recording and reporting of results. Finally the laboratory could introduce more sensitive diagnostic techniques, e.g. the use of cryptococcal antigen detection tests on CSF and blood. 5 Symptomatic treatment of cryptococcal meningitis is with analgesics, antipyretics and anticonvulsants. Effective anti-cryptococcal drugs, e.g. amphotericin and fluconazole, are not available in government hospitals in Malawi except on private prescription. These drugs are expensive and of the 31 patients in this series, only six (19 %) could afford them. The survival time after diagnosis of patients treated with anticryptococcal drugs was usually of the order of several months, whereas that of patients who could not afford them was a median of 4 days. Cost-benefit considerations mitigate against the provision of expensive anti-cryptococcal drugs as a priority of health care expenditure in developing countries.

Despite the lack of affordable specific drug treatment for most patients with cryptococcal meningitis, it is important to make the diagnosis because knowing the prognosis may be important for patients and relatives and a definite diagnosis saves the patient from unnecessary treatment of other conditions, most commonly tuberculous meningitis.

The problem of cryptococcal meningitis is likely to become more common in sub-Saharan Africa and elsewhere as HIV infection becomes more widespread.

(We thank the staff of the Kamuzu Central Hospital and Queen Elizabeth Central Hospital microbiology laboratories for their assistance, and Professor A. D. Harries, Professor of Medicine, University of Malawi College of Medicine, for his helpful comments.)

64 D. MAHER AND H. MWANDUMBA

R e f e r e n c e s

i. Panther LA, Sande MA. Cryptococcal meningitis in the acquired immunodeficiency syndrome. Semin Respir Infect 199o; 5(2): 138-145.

2. World Health Organisafion: Acquired Immunodeficiency Syndrome (AIDS), 1987 revisions of the CDC/WHO case definition for AIDS. W/My Epidemiol Rec 1988; 63: 1-7.

3. Chuck SL, Sande MA. Infections with Cryptococcus neoformam in the acquired immunodeficiency syndrome. N EnglJ Med 1989; 321: 794-799.

4. Desmet P, Kayembe KD, De Vroey C. The value of cryptococcal antigen screening among HIV-positive/AIDS patients in Kinshasa, Zaire. AIDS 1989; 3: 77-78.

5. Bloomfield N, Gordon MA, Elmendorf DF Jr. Detection of Cryptococcus neoformans antigen in body fluids by latex particle agglutination. Proc Soc Exp Biol Med 1963; 114: 64-67.