Disseminated cryptococcal infection despite treatment for cryptococcal meningitis.

Case Reports (Dr R F Miller) Case 1 A 25 year old Caucasian heterosexual man was found to be HIV I antibody positive in July 1989, when he presented to another hospital with cryptococcal meningitis. At this time his CD4+ count was 190 (normal range = 350-2,200)/mm3. Investigations showed normal urea and electrolyte levels and liver function tests; CT of the head was normal. Blood and urine culture was negative, India ink staining of CSF was positive for Cryptococcus neoformans and this organism was cultured from CSF. The CSF protein was 0.42 g/l, 30 lymphocytes/mm3 were seen and the CSF/plasma glucose ratio was 2.2/5.6 mmol/l. Cryptococcal antigen (CRAG) titre, using latex agglutination, was 1 in 2052 in both CSF and in blood. The patient was treated with intravenous amphotericin B, 1 mg/kg/day for four weeks and responded well clinically; at repeat lumbar puncture performed after four weeks treatment the CSF CRAG had fallen to 1 in 256. Subsequently the patient was maintained on oral fluconazole 400 mg daily. This was discontinued in October 1989 because of abnormal liver function tests. The patient rapidly developed headache and was admitted in November of that year. Investigations then showed normal urea and electrolyte levels and liver function tests and negative blood cultures. Blood CRAG was 1 in 5000, CSF CRAG was 1 in 1026. Cultures of CSF were positive for C neoformans. The patient was given a further course of intravenous amphotericin B for four weeks and subsequently a Hickman line was inserted; he then received 0.5 mg/kg of amphotericin intravenously once weekly, as maintenance therapy and began Fansidar two tablets once weekly for primary prophylaxis against pneumocystis pneumonia. He remained well until April 1990 when he had a further relapse of meningitis whilst receiving maintenance, which was again treated with four weeks of amphotericin B at a dose of 1 mg/kg/day. He then went back onto weekly maintenance amphotericin at the doses given above. He was admitted to the Middlesex Hospital in August 1990 with a four day history of generalised headache and nausea. On examination there was slight ataxia but no other neurological signs were evident. Urea and electrolyte levels and liver function tests were normal, CT of the head showed atrophy only; in particular there was no hydrocephalus and no cryptococcomas were seen. The CD4 count was 40/mm3. Blood cultures were negative for fungi, acid and alcohol fast bacilli (AAFB) and bacteria. At lumbar puncture there were 20 lymphocytes/mm3, CSF/plasma glucose ratio was 17/5.1 mmol/l and CSF protein was 0.69 g/l. Gram, auramine and India ink stains of CSF were negative and culture for bacteria and mycobacteria, viruses and fungi were all negative. Syphilis and toxoplasmosis serology were negative in CSF and blood. The CRAG titre in CSF was 1 in 256 and in blood was 1 in 2052. The Hickman line entry site was painful and red and there was crusting sero-purulent discharge from this site. Culture from the skin site (and from the catheter tip) grew Staphlococcus aureus (sensitive to flucloxacillin). The patient was treated with intravenous flucloxacillin and the Hickman line was removed. After discussions between his wife and his carers the patient declined further treatment with amphotericin and instead chose to receive oral fluconazole 800 mg once daily. He was transferred to a hospice where he developed slurred speech, poor balance and neck stiffness. Pre-terminally he had photophobia and muscle tenderness. Terminally he developed bronchopneumonia and died four weeks after transfer to the hospice. A perimortem serum CRAG was > 1 in 10 000.