54 TRANSACTIONS OF THE ROYAL Somrv OF TROPICAL MEDICINE AND HYGIENE (1993) 87, 5655 1 Short Report 1 Interleukin-8 and Plasmodium falciparum malaria in Thailand Jon S. Friedland’, May H02,3, Daniel G. Remick”, llyl;;i Bunnagz, Nicholas J. White*,5 and George E. * 1 ‘Division of Communicable Diseases, St George’s Hospital Medical School, London, UK; *Hospital for Tropical Diseases and Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand; 3Department of Microbiology and Infectious Diseases, University of Calgay, Calgay, Canada; 4Department of Pathology, University of Michigan Medical School, Ann Arbor, Michigan, USA; 5Nuffield Department of Clinical Medicine,John Radcliffe Hospital, Oxford, UK The role of interleukin 8 (IL-8) in Plasmodium falcipa- rum malaria is unknown, although activation of the cyto- kine network and secretion of pro-inflammatory cyto- kines such as tumour necroiis factor (TNF) &d interleukin 6 (IL-6) doesoccur. IL-8 is secretedbv manv cells including monocytes, macrophagesand end&heli& cells. It is a chemoattractant and activator of neutrophils and is also chemotactic for T cells (BAGGIOLINI et al., 1989). However, there are few in v&o data on IL-8 in clinical infections. We have investigated plasma IL-8 concentrations in a pilot study of 6 patients with rela- tively severe? but non-fatal, non-cerebral, I’. falciparum malaria admltted consecutively to the Hospital for Tropi- cal Diseases, Bangkok, Thailand. The study was ap- proved by the ethics committee of the Faculty of Tropi- cal Medicine, Mahidol University, Thailand. Following diagnosis of malaria by microscopical exam- ination of thick and thin blood films. 5 ml blood samules were taken immediatelv before initiation of treatGent (artemether or artesunate and appropriate supportive therapy), then daily for 7 d and weekly thereafter until discharge at 1 month. The prolonged admission policy at the Hosuital for Tronical Diseases ensures that there is no secoid episode of-infection in patients and allows in- sight into the natural history of a single treated episode of malaria. Samples were collected into endotoxin-free tubes containing the anticoagulant potassium ethylene diaminetetraacekte and 100 cl1 trasylol, a protease in- hibitor. Thev were then centrifuged (1000 a for 5 min at 4°C) and plasma was stored at -70°6for a maximum of 2 months. IL-8 was measured with a previously described enzyme-linked immunosorbent assay (ELISA) (DEFORGE & REMICK, 1991). Plasma TNF and IL-6 concentrations were measured usinn established bioassavs (ESPEVIK & NISSEN-MEYER, 198%; AARDEN et al., 1987): The lower limit of sensitivitv of the TNF assav was 23 De/ml. of the IL-6 assay 1 pgjml, and of the iL-8 EL&A-95 ‘pg/ml. Standard curves were determined for all assays, which were performed with appropriate control samples and carried out in a ‘blinded’ fashion by a laboratory unaware of the clinical details. The 6 patients were all male, aged between 16 and 45 years; their initial peripheral parasitaemias were 2.1%, 5.6%, 7.0%, 7.6%, 7.8% and 13%. None had evidence of cerebral malaria and all recovered following treatment. Five patients had abnormal liver function tests with raised serum bilirubin and transaminases. One patient required haemodialysis and cytokine analyses were dis- continued as renal replacement therapy may activate the cybokine network. A neutrophil le&ocyiosis (11.9~ 10 /L, 84% neutronhils) was documented in only one pa- tient. ‘Malarial parasites’werecleared from the ci&latIon Address for correspondence: Dr Jon Friedland, Division of Communicable Diseases, St George’s Hospital Medical School, Cranmer Terrace, London, SW17 ORE, UK. within 3 d of starting specific treatment and all patients were afebrile within one week. The initial mean nlasma IL-8 concentration was raised at 216k9.5 pg/ml, bhich is similar to levels we have pre- viously recorded in patients with fatal Gram-negative sepsis(97-362 pgiml.). In normal healthy subjects with- out malaria no plasma IL-8 is detectable using this ELISA (DEFORGE & REMICK, 1991), and in control experiments we have found that the freezing and storage protocol used in this study does not affect plasma measurements. The prognostic significance of raised plasma IL-8 con- centration cannot be deduced from this small pilot study. IL-8 may be the stimulus for the leucocytosis found in somepatients with severe malaria, although a raised neu- trophil count was found in only one patient in our study. The current data suggest that IL-8 has as yet uncharac- terized functions in host defence to this protozoa1 pa- thogen, unrelated to activation of neutrophils. Mononu- clear cells express the IL-8 gene folldwing adhesion (KASAHARA et al., 1991), suggesting a relationship be- tween IL-8 and the regulation of adhesion molecules, such as ICAM- 1. Such an interaction may be important if endothelial expression of intercellular adhesion-molecule 1 (ICAM-1). an adhesion recentor for P. falcibarum in vitro (BERE~DT et al., 1989), is involved in capillary se- questration of the parasite with consequent organ dys- function in vivo. Admission plasma IL-6 concentrations detected by bioassay in these patients were slightly elevated at 44k21.6 pgiml. This is similar to levels we have de- tected in relatively trivial bacterial infections but lower than the concentrations measured in African children with cerebral malaria (MOLYNEUX et al., 1991). This dif- ference may be important, as high plasma IL-6 concen- trations have been associatedwith poor prognosis in the African study and with organ damage in severe malaria in non-immune European adults (KERN et al., 1989). However, this discrepancy might reflect the fact that we have used a biological rather than an immunoradiometric assay. Bioactive TNF was detectable in the plasma of 4 oatients on admission to the studv (12Ok65 &ml), at concentrations similar to those - previously- ;ep&ted (GRAU et al., 1989). An unexpected and interesting finding of the current study was that plasma IL-8 and IL-6 levels remained elevated throughout the period of investigation (Figure). Patients were studied for at least 3 weeks after they had become afebrile and recovered clinically from their ill- ness. In only one patient was plasma TNF bioactivity de- tected beyond day 7, the period during which fever set- Fever Peripheral Parasitaemia Fieure.Mean duration of fever.oeriuheral oarasitaemia and olasma IL-8 (0) and IL-6 (0) concentrations in patients during treatment and after recovery from I’. falcipamm malaria; verticalbars indicate standard error of the means.