- Email: [email protected]
A non-sense mutation in Cd36 gene is associated with protection from severe malaria
RESEARCH LETTERS
3
4
5
Gorlin JR, Cohen MM Jr, Levin LS. Ehlers-Danlos syndromes. In: Gorlin JR, Cohen MM Jr, Levin LS eds. Syndromes of the head and neck. 3rd edn. Oxford: Oxford University Press, 1990: 417–41. Rowe PC, Barron DF, Calkins H, Maumenee IH, Tong PY, Geraghty MT. Ehlers-Danlos syndrome. J Pediatr 1999; 135: 513. De Felice C, Di Maggio G, Zagordo L, et al. Hypoplastic or absent mandibular frenulum: a new predictive sign in infantile hypertrophic pyloric stenosis. J Pediatr 2000; 136: 408–10.
Institute of Preventive Paediatrics and Neonatology (C De Felice MD, F Bagnoli MD), Institute of Pathology (P Toti MD), Section of Pediatric Surgery, Institute of General Surgery and Surgical Specialties (G Di Maggio MD), and Institute of Odontostomatologic Sciences (S Parrini MD), Azienda Ospedaliera Senese and University of Siena, I-53100, Italy Correspondence to: Dr C De Felice (e-mail: [email protected])
A non-sense mutation in Cd36 gene is associated with protection from severe malaria Arnab Pain, Britta C Urban, Oscar Kai, Climent Casals-Pascual, Juma Shafi, Kevin Marsh, David J Roberts We sought genetic evidence for the importance of hostparasite interactions involving CD36 in severe malaria. We identified a non-sense mutation in Cd36 gene and looked at the influence of this mutation on the outcome of malaria infection in 693 African children with severe malaria and a similar number of ethnically matched controls. We showed that heterozygosity for this mutation is associated with protection from severe disease (OR 0·74, 95% CI 0·55–0·99; p=0·036). These findings suggest that this Cd36 mutation might have a complex effect on malaria infection by decreasing parasite sequestration, and also by decreasing host immune responses.
Erythrocytes infected with falciparum malaria adhere to a variety of host receptors, including CD36, which are widely expressed on endothelium, platelets, and leucocytes. Adhesion of infected erythrocytes to CD36 might benefit parasite survival and is mediated, at least in part, by the cysteine-rich domain of a highly polymorphic variant antigen expressed at the surface of infected erythrocytes.1 The pathological significance of the adhesion of infected erythrocytes to CD36 is uncertain. CD36 deficiency, though not the pro90-ser mutation, is common in Africans.2 We conjectured that mutations in the Cd36 gene may confer protection from severe malaria. We therefore sequenced DNA from two Afro-Americans who did not express CD36 on their platelets and identified the point mutation from T to G in the Cd36 gene at nucleotide position 188 in exon 10, thus incorporating a stop codon TAG. We looked for this mutation in 693 cases of severe malaria and an equal number of ethnically matched community controls from Kilifi, Kenya. We recruited children with severe malaria who had a parasite count of greater than 2 000 per L and clinical and laboratory findings that excluded illness from other causes. Children had cerebral malaria
1502
PCR- RFLP typing for the T188G mutation W, H, and M denote wild type, heterozygote, and homozygous mutant for the T188G mutation. The genotypes of twenty heterozygotes and mutants selected at random after three rounds of screening by RFLP were confirmed by direct sequencing. Bp=base pair.
with a Blantyre coma score of less than 2 at least 1 h after an epileptic fit, or were hypoglycaemic with a blood glucose lower than 2·2 mmol/L, or had a haemoglobin concentration of less than 50 g/L, or had respiratory distress. All cases were admitted to hospital and treated with parenteral quinine. Ethnically matched controls were taken from consecutive primagravidae attending the antenatal clinic. Severity scale 4 or 3 was assigned to children who had at least three syndromes of severe malaria: cerebral malaria, severe anaemia, respiratory distress, or hypoglycaemia. Severity scale 1 or 2 is represented by children having one or two of these syndromes. Informed consent was obtained from parents of children and from adult controls. The study was approved by ethics committees in Oxford and in Kenya. Genotypes were identified from NdeI digested fragments of PCR amplified products of the exon 10 of Cd36 gene (Gen Bank Accession number Z32760) with the primer pairs 5' CTATGCTGTATTTGAATCCGACGTT 3' and 5' ATGGACTGTGCTACTGAGGTTATTTACTC 3', respectively. The mutant allele remains uncut with NdeI and runs as a single 212 base pair band, whereas digestion of the PCR product in those carrying the wild type allele produces two fragments of 148 base pairs and 64 base pairs (figure). Heterozygotes for the T188G mutation were present in 121/693 (17%) of the control population but in only 95/693 (14%) of cases of severe malaria (odds ratio 0·74, 95% CI 0·55–0·99). Furthermore, the association was stronger in those children suffering from multiple syndromes of malaria (table). The wide clinical spectrum of malaria indicates different patho-physiological processes. We therefore analysed the association of the T188G mutation with well-defined syndromes of severe malaria. Intriguingly, this mutation was associated with protection from either respiratory distress, severe anaemia, or hypoglycaemia but not with protection from cerebral malaria (table). Our results show that this T188G Cd36 polymorphism influences the outcome of malaria infection. The mutation may directly influence the expression of CD36 on endothelium and reduce the sequestration of malariainfected erythrocytes. The association of the T188G Cd36 polymorphism with protection from some syndromes of severe malaria, such as severe anaemia and hypoglycaemia, may be due to reduced sequestration and parasite burden. The absence of protection from cerebral malaria in patients heterozygous for T188G Cd36 is at first sight puzzling. However, infected erythrocytes may interact with CD36 expressed not only on endothelial cells but also on monocytes, macrophages, and dendritic cells. The failure of the T188G Cd36 mutation to confer protection from
THE LANCET • Vol 357 • May 12, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
RESEARCH LETTERS
Disease state
T/G
Controls (n⫽693) 121 (17·5%) Cases (n⫽693) 95 (13·7%) 4 or 3 syndromes (n⫽104) 10 (9·6%) 1 or 2 syndromes (n⫽589) 85 (14·4%) Cerebral malaria (n⫽413) 62 (15·0%) Respiratory distress (n⫽250) 32 (12·8%) Severe anaemia (n⫽304) 37 (12·2%) Hypoglycaemia (n⫽146) 16 (11·0%)
G/G
OR (95% CI)
p*
8 (1·1%) 5 (0·7%) 0 (0·0%) 5 (0·9%) 4 (1·0%) 1 (0·4%) 1 (0·3%) 0 (0·0%)
0·74 (0·55–0·99) 0·47 (0·22–0·95) 0·79 (0·58–1·07) 0·84 (0·60–1·18) 0·67 (0·43–1·04) 0·62 (0·41–0·94) 0·54 (0·30–0·96)
0·036 0·024 0·11 0·27 0·051 0·017 0·026
*Mantel-Haenszel corrected p value for 2 test homozygotes (G/G) and heterozygotes (T/G) versus controls (2⫻2); OR⫽odds ratio.
Association of the Cd36 T188G polymorphism with severe malaria
cerebral malaria suggests the beneficial effect of reduced sequestration in those carrying this mutation is counterbalanced by an altered immune response. For example, this mutation may abrogate the modulation of dendritic cells by adhesion of infected erythrocytes.3 In a combined data-set from children in east and west Africa, Aitman and colleagues4 showed there was a significant association of Cd36 mutations with susceptibility to severe malaria in general, and to cerebral malaria in particular. Despite the heterogeneity of parasite and human populations, the study by Aitman and coworkers, and our own results, suggest that a non-sense T188G Cd36 mutation has a complex effect in malaria infection—by decreasing parasite sequestration on the one hand, and by depressing host immune response on the other. Identifying the mechanism of protection for Cd36 mutations will require integrated genetic, functional, and field studies with careful case definition of the symptoms of severe disease. Given the mortality from malaria, protective polymorphisms for severe malaria would become fixed unless the mutation is associated with susceptibility to other diseases.5 CD36 is widely expressed and has functions including immune responses. Thus, altered function of CD36 could affect diverse pathological processes and might exist as a balanced polymorphism in malaria endemic areas. The work was supported by the Wellcome Trust, the Kenyan Medical Research Institute (KEMRI), and the National Blood Service, UK. We thank Brian Curtis (Wisconsin) for Naka (-) platelets. Sir EPA Abraham Trust supports BCU. KM is a Wellcome Trust Senior Fellow in Clinical Science. 1
2
3
4 5
Baruch DI, Ma XC, Pasloske B, Howard RJ, Miller LH. CD36 peptides that block cytoadherence define the CD36 binding region for Plasmodium falciparum-infected erythrocytes. Blood 1999; 94: 2121–27. Kashiwagi H, Honda S, Tomiyama Y, et al. A novel polymorphism in glycoprotein IV (replacement of proline-90 by serine) predominates in subjects with platelet GPIV deficiency. Thromb Haemost 1993; 69: 481–84 Urban BU, Ferguson DJP, Pain A, Austyn J, Roberts DJ. Malaria infected erythrocytes modulate the maturation of dendritic cells. Nature 1999; 400: 73–77 Aitman TJ, Cooper LD, Norsworthy PJ, et al. Malaria susceptibility and CD36 mutation. Nature 2000; 405: 1015–16. Miller LH. Impact of malaria on genetic polymorphism and genetic diseases in Africans and African Americans. Proc Natl Acad Sci USA 1994; 91: 2415–19.
Institute of Molecular Medicine (A Pain PhD, B C Urban, PhD, C Casals-Pascual MD) and National Blood Service (D J Roberts MRCPath); John Radcliffe Hospital, Oxford OX3 9DS, UK; and KEMRI Centre for Geographic Medicine Research Coast, PO Box 230, Kilifi, Kenya (O Kai DAB, J Shafi BSc, K Marsh FRCP) Correspondence to: Dr David Roberts (email: [email protected])
THE LANCET • Vol 357 • May 12, 2001
Pulmonary hypersensitivity reaction induced by efavirenz Georg M N Behrens, Matthias Stoll, Reinhold E Schmidt HIV-infected patients are at increased risk of developing adverse drug reactions such as a maculopapular rash due to antiretroviral drugs. For example, exanthema induced by nonnucleoside reverse transcriptase inhibitors leads to drug discontinuation in many patients. We present a patient with a severe pulmonary hypersensitivity reaction due to efavirenz.
The appearance of a maculopapular rash is one of the most frequent and limiting adverse drug reactions within the first few weeks of treatment with non-nucleoside analogue reverse transcriptase inhibitors (NNRTI). The frequency of this rash varies from 9–32% with a higher rate for nevirapine than for efavirenz.1-3 Both drugs induce similar cutaneous signs and symptoms—fever, rigor, myalgia, and arthralgias. Only the nucleoside analogue reverse transcriptase inhibitor abacavir has been linked to respiratory symptoms.1 A 39-year-old HIV-infected man presented with diarrhoea, fatigue, weight loss of 8% bodyweight within 3 months, night sweats, and fever of more than 39°C. Laboratory tests on admission showed increased concentrations of aspartate aminotransferase 73 U/L (reference range <18 U/L), alanine aminotransferase 100 U/L (<22 U/L), γglutamyltransferase 126 U/L (<28 U/L), alkaline phosphatase 366 U/L (<190 U/L), and C-reactive protein 53 mg/L (<5 mg/L). The viral load for HIV was above 750 000 copies HIV-RNA/mL plasma (HIV-Monitor, Roche, Mannheim, Germany) and CD4 cell counts of 11 cells/µL indicated an advanced cellular immunodeficiency. Repeated examination of sputum, stools, blood, and bone marrow aspiration was negative for mycobacteria. Diarrhoea was attributed to cryptosporidiosis which was diagnosed after colonoscopy. Based on bilateral mild interstitial and reticular opacities on chest radiography, fever up to 40°C, mild dyspnoea, and positive immunofluorescence, we diagnosed Pneumocysitis carinii pneumonia (PCP). Initiation of high-dose co-trimoxazole led to rapid and complete remission of radiographic abnormalities and symptoms. 10 days after the initiation of highly active combination therapy consisting of stavudine (40 mg twice daily), didanosine (200 mg twice daily), and efavirenz (600 mg, four times daily) the patient developed a confluent maculopapular pruritic rash of the trunk and upper extremities accompanied by myalgia. In addition, the man had a fever up to 40°C, and interstitial infiltration without lymphadenopthy in both lungs detected by chest radiography (figure). C-reactive protein concentration increased up to 239 mg/L as did aminotransferase concentrations, and there was mild eosinophilia (1000 cells/µL [reference range <700/µL]). Direct cytology and microbial culture obtained after bronchoalveolar lavage provided no evidence for a relapse of PCP or other bacterial, viral, or fungal pathogens. Efavirenz treatment was stopped, while stavudine and didanosine were continued, and the patient’s symptoms including the pulmonary infiltration resolved immediately under treatment with oral corticosteroids (initially 75 mg per day). 2 weeks after complete remission of the symptoms,
1503
For personal use. Only reproduce with permission from The Lancet Publishing Group.
3
4
5
Gorlin JR, Cohen MM Jr, Levin LS. Ehlers-Danlos syndromes. In: Gorlin JR, Cohen MM Jr, Levin LS eds. Syndromes of the head and neck. 3rd edn. Oxford: Oxford University Press, 1990: 417–41. Rowe PC, Barron DF, Calkins H, Maumenee IH, Tong PY, Geraghty MT. Ehlers-Danlos syndrome. J Pediatr 1999; 135: 513. De Felice C, Di Maggio G, Zagordo L, et al. Hypoplastic or absent mandibular frenulum: a new predictive sign in infantile hypertrophic pyloric stenosis. J Pediatr 2000; 136: 408–10.
Institute of Preventive Paediatrics and Neonatology (C De Felice MD, F Bagnoli MD), Institute of Pathology (P Toti MD), Section of Pediatric Surgery, Institute of General Surgery and Surgical Specialties (G Di Maggio MD), and Institute of Odontostomatologic Sciences (S Parrini MD), Azienda Ospedaliera Senese and University of Siena, I-53100, Italy Correspondence to: Dr C De Felice (e-mail: [email protected])
A non-sense mutation in Cd36 gene is associated with protection from severe malaria Arnab Pain, Britta C Urban, Oscar Kai, Climent Casals-Pascual, Juma Shafi, Kevin Marsh, David J Roberts We sought genetic evidence for the importance of hostparasite interactions involving CD36 in severe malaria. We identified a non-sense mutation in Cd36 gene and looked at the influence of this mutation on the outcome of malaria infection in 693 African children with severe malaria and a similar number of ethnically matched controls. We showed that heterozygosity for this mutation is associated with protection from severe disease (OR 0·74, 95% CI 0·55–0·99; p=0·036). These findings suggest that this Cd36 mutation might have a complex effect on malaria infection by decreasing parasite sequestration, and also by decreasing host immune responses.
Erythrocytes infected with falciparum malaria adhere to a variety of host receptors, including CD36, which are widely expressed on endothelium, platelets, and leucocytes. Adhesion of infected erythrocytes to CD36 might benefit parasite survival and is mediated, at least in part, by the cysteine-rich domain of a highly polymorphic variant antigen expressed at the surface of infected erythrocytes.1 The pathological significance of the adhesion of infected erythrocytes to CD36 is uncertain. CD36 deficiency, though not the pro90-ser mutation, is common in Africans.2 We conjectured that mutations in the Cd36 gene may confer protection from severe malaria. We therefore sequenced DNA from two Afro-Americans who did not express CD36 on their platelets and identified the point mutation from T to G in the Cd36 gene at nucleotide position 188 in exon 10, thus incorporating a stop codon TAG. We looked for this mutation in 693 cases of severe malaria and an equal number of ethnically matched community controls from Kilifi, Kenya. We recruited children with severe malaria who had a parasite count of greater than 2 000 per L and clinical and laboratory findings that excluded illness from other causes. Children had cerebral malaria
1502
PCR- RFLP typing for the T188G mutation W, H, and M denote wild type, heterozygote, and homozygous mutant for the T188G mutation. The genotypes of twenty heterozygotes and mutants selected at random after three rounds of screening by RFLP were confirmed by direct sequencing. Bp=base pair.
with a Blantyre coma score of less than 2 at least 1 h after an epileptic fit, or were hypoglycaemic with a blood glucose lower than 2·2 mmol/L, or had a haemoglobin concentration of less than 50 g/L, or had respiratory distress. All cases were admitted to hospital and treated with parenteral quinine. Ethnically matched controls were taken from consecutive primagravidae attending the antenatal clinic. Severity scale 4 or 3 was assigned to children who had at least three syndromes of severe malaria: cerebral malaria, severe anaemia, respiratory distress, or hypoglycaemia. Severity scale 1 or 2 is represented by children having one or two of these syndromes. Informed consent was obtained from parents of children and from adult controls. The study was approved by ethics committees in Oxford and in Kenya. Genotypes were identified from NdeI digested fragments of PCR amplified products of the exon 10 of Cd36 gene (Gen Bank Accession number Z32760) with the primer pairs 5' CTATGCTGTATTTGAATCCGACGTT 3' and 5' ATGGACTGTGCTACTGAGGTTATTTACTC 3', respectively. The mutant allele remains uncut with NdeI and runs as a single 212 base pair band, whereas digestion of the PCR product in those carrying the wild type allele produces two fragments of 148 base pairs and 64 base pairs (figure). Heterozygotes for the T188G mutation were present in 121/693 (17%) of the control population but in only 95/693 (14%) of cases of severe malaria (odds ratio 0·74, 95% CI 0·55–0·99). Furthermore, the association was stronger in those children suffering from multiple syndromes of malaria (table). The wide clinical spectrum of malaria indicates different patho-physiological processes. We therefore analysed the association of the T188G mutation with well-defined syndromes of severe malaria. Intriguingly, this mutation was associated with protection from either respiratory distress, severe anaemia, or hypoglycaemia but not with protection from cerebral malaria (table). Our results show that this T188G Cd36 polymorphism influences the outcome of malaria infection. The mutation may directly influence the expression of CD36 on endothelium and reduce the sequestration of malariainfected erythrocytes. The association of the T188G Cd36 polymorphism with protection from some syndromes of severe malaria, such as severe anaemia and hypoglycaemia, may be due to reduced sequestration and parasite burden. The absence of protection from cerebral malaria in patients heterozygous for T188G Cd36 is at first sight puzzling. However, infected erythrocytes may interact with CD36 expressed not only on endothelial cells but also on monocytes, macrophages, and dendritic cells. The failure of the T188G Cd36 mutation to confer protection from
THE LANCET • Vol 357 • May 12, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
RESEARCH LETTERS
Disease state
T/G
Controls (n⫽693) 121 (17·5%) Cases (n⫽693) 95 (13·7%) 4 or 3 syndromes (n⫽104) 10 (9·6%) 1 or 2 syndromes (n⫽589) 85 (14·4%) Cerebral malaria (n⫽413) 62 (15·0%) Respiratory distress (n⫽250) 32 (12·8%) Severe anaemia (n⫽304) 37 (12·2%) Hypoglycaemia (n⫽146) 16 (11·0%)
G/G
OR (95% CI)
p*
8 (1·1%) 5 (0·7%) 0 (0·0%) 5 (0·9%) 4 (1·0%) 1 (0·4%) 1 (0·3%) 0 (0·0%)
0·74 (0·55–0·99) 0·47 (0·22–0·95) 0·79 (0·58–1·07) 0·84 (0·60–1·18) 0·67 (0·43–1·04) 0·62 (0·41–0·94) 0·54 (0·30–0·96)
0·036 0·024 0·11 0·27 0·051 0·017 0·026
*Mantel-Haenszel corrected p value for 2 test homozygotes (G/G) and heterozygotes (T/G) versus controls (2⫻2); OR⫽odds ratio.
Association of the Cd36 T188G polymorphism with severe malaria
cerebral malaria suggests the beneficial effect of reduced sequestration in those carrying this mutation is counterbalanced by an altered immune response. For example, this mutation may abrogate the modulation of dendritic cells by adhesion of infected erythrocytes.3 In a combined data-set from children in east and west Africa, Aitman and colleagues4 showed there was a significant association of Cd36 mutations with susceptibility to severe malaria in general, and to cerebral malaria in particular. Despite the heterogeneity of parasite and human populations, the study by Aitman and coworkers, and our own results, suggest that a non-sense T188G Cd36 mutation has a complex effect in malaria infection—by decreasing parasite sequestration on the one hand, and by depressing host immune response on the other. Identifying the mechanism of protection for Cd36 mutations will require integrated genetic, functional, and field studies with careful case definition of the symptoms of severe disease. Given the mortality from malaria, protective polymorphisms for severe malaria would become fixed unless the mutation is associated with susceptibility to other diseases.5 CD36 is widely expressed and has functions including immune responses. Thus, altered function of CD36 could affect diverse pathological processes and might exist as a balanced polymorphism in malaria endemic areas. The work was supported by the Wellcome Trust, the Kenyan Medical Research Institute (KEMRI), and the National Blood Service, UK. We thank Brian Curtis (Wisconsin) for Naka (-) platelets. Sir EPA Abraham Trust supports BCU. KM is a Wellcome Trust Senior Fellow in Clinical Science. 1
2
3
4 5
Baruch DI, Ma XC, Pasloske B, Howard RJ, Miller LH. CD36 peptides that block cytoadherence define the CD36 binding region for Plasmodium falciparum-infected erythrocytes. Blood 1999; 94: 2121–27. Kashiwagi H, Honda S, Tomiyama Y, et al. A novel polymorphism in glycoprotein IV (replacement of proline-90 by serine) predominates in subjects with platelet GPIV deficiency. Thromb Haemost 1993; 69: 481–84 Urban BU, Ferguson DJP, Pain A, Austyn J, Roberts DJ. Malaria infected erythrocytes modulate the maturation of dendritic cells. Nature 1999; 400: 73–77 Aitman TJ, Cooper LD, Norsworthy PJ, et al. Malaria susceptibility and CD36 mutation. Nature 2000; 405: 1015–16. Miller LH. Impact of malaria on genetic polymorphism and genetic diseases in Africans and African Americans. Proc Natl Acad Sci USA 1994; 91: 2415–19.
Institute of Molecular Medicine (A Pain PhD, B C Urban, PhD, C Casals-Pascual MD) and National Blood Service (D J Roberts MRCPath); John Radcliffe Hospital, Oxford OX3 9DS, UK; and KEMRI Centre for Geographic Medicine Research Coast, PO Box 230, Kilifi, Kenya (O Kai DAB, J Shafi BSc, K Marsh FRCP) Correspondence to: Dr David Roberts (email: [email protected])
THE LANCET • Vol 357 • May 12, 2001
Pulmonary hypersensitivity reaction induced by efavirenz Georg M N Behrens, Matthias Stoll, Reinhold E Schmidt HIV-infected patients are at increased risk of developing adverse drug reactions such as a maculopapular rash due to antiretroviral drugs. For example, exanthema induced by nonnucleoside reverse transcriptase inhibitors leads to drug discontinuation in many patients. We present a patient with a severe pulmonary hypersensitivity reaction due to efavirenz.
The appearance of a maculopapular rash is one of the most frequent and limiting adverse drug reactions within the first few weeks of treatment with non-nucleoside analogue reverse transcriptase inhibitors (NNRTI). The frequency of this rash varies from 9–32% with a higher rate for nevirapine than for efavirenz.1-3 Both drugs induce similar cutaneous signs and symptoms—fever, rigor, myalgia, and arthralgias. Only the nucleoside analogue reverse transcriptase inhibitor abacavir has been linked to respiratory symptoms.1 A 39-year-old HIV-infected man presented with diarrhoea, fatigue, weight loss of 8% bodyweight within 3 months, night sweats, and fever of more than 39°C. Laboratory tests on admission showed increased concentrations of aspartate aminotransferase 73 U/L (reference range <18 U/L), alanine aminotransferase 100 U/L (<22 U/L), γglutamyltransferase 126 U/L (<28 U/L), alkaline phosphatase 366 U/L (<190 U/L), and C-reactive protein 53 mg/L (<5 mg/L). The viral load for HIV was above 750 000 copies HIV-RNA/mL plasma (HIV-Monitor, Roche, Mannheim, Germany) and CD4 cell counts of 11 cells/µL indicated an advanced cellular immunodeficiency. Repeated examination of sputum, stools, blood, and bone marrow aspiration was negative for mycobacteria. Diarrhoea was attributed to cryptosporidiosis which was diagnosed after colonoscopy. Based on bilateral mild interstitial and reticular opacities on chest radiography, fever up to 40°C, mild dyspnoea, and positive immunofluorescence, we diagnosed Pneumocysitis carinii pneumonia (PCP). Initiation of high-dose co-trimoxazole led to rapid and complete remission of radiographic abnormalities and symptoms. 10 days after the initiation of highly active combination therapy consisting of stavudine (40 mg twice daily), didanosine (200 mg twice daily), and efavirenz (600 mg, four times daily) the patient developed a confluent maculopapular pruritic rash of the trunk and upper extremities accompanied by myalgia. In addition, the man had a fever up to 40°C, and interstitial infiltration without lymphadenopthy in both lungs detected by chest radiography (figure). C-reactive protein concentration increased up to 239 mg/L as did aminotransferase concentrations, and there was mild eosinophilia (1000 cells/µL [reference range <700/µL]). Direct cytology and microbial culture obtained after bronchoalveolar lavage provided no evidence for a relapse of PCP or other bacterial, viral, or fungal pathogens. Efavirenz treatment was stopped, while stavudine and didanosine were continued, and the patient’s symptoms including the pulmonary infiltration resolved immediately under treatment with oral corticosteroids (initially 75 mg per day). 2 weeks after complete remission of the symptoms,
1503
For personal use. Only reproduce with permission from The Lancet Publishing Group.