897

2000 parasites/)il of whole blood).4 Both treatment respect to age, body temperature, severity, groups and duration of coma on admission (table). All patients recovered and were discharged 3 days after admission. No obvious harmful effect was observed. In the loading-dose group, there was a significant decrease in the duration of coma, and the parasite clearance times were reduced by an average of 25% (table). Davis et al5 recommended the use of an intravenous quinine loading dose as being a safe and effective treatment. We are unaware of other reports of controlled trials of the clinical efficacy of a quinine loading dose, and our preliminary findings show its effectiveness and good tolerance. Our results need to be confirmed but they indicate the usefulness of this simple regimen as a treatment for severe malaria, even in places where intensive-care facilities are not available.

parasitaemia (>

were similar with

Intensive Care Unit, Hôpital Central de Yaoundé, Yaoundé

Medical Entomology and

Epidemiology Units, OCEAC/ORSTOM, BP288, Yaoundé, Cameroon Hôpital

Central de Yaoundé

J.-J. FARGIER F. J. LOUIS M. COT B. MAUBERT

C. HOUNSINOU

OCEAC/ORSTOM, Yaoundé

J.-P. LOUIS

Centre National de Référence pour la Chimiosensibilité du Paludisme, IMET, Paris, France

J. LE BRAS

Institut de Médecine Tropicale du Service de Santé des Armées, Marseille

J. E. TOUZE

1. Hengy C, Jambou R,

Eberle F, et al. Problèmes thérapeutiques poses par la chimiorésistance de P falciparum aux antimalariques à Yaoundé, Cameroun. Abstracts of the VII International Congress of Parasitology, Paris. Bull Soc Fr Parasitol 1990; 8 (suppl 1): 430. 2. White NJ, Looareesuwan S, Warrell DA, Warrell MJ, Bunnag D, Harinasuta T. Quinine pharmacokinetics and toxicity in cerebral and uncomplicated falciparum malaria. Am J Med 1982; 73: 564-72. 3. Warrell DA, Looareesuwan S, Warrell MJ, et al. Dexamethasone proves deleterious in cerebral malaria: a double blind trial in 100 comatose patients. N Engl J Med 1982; 306: 313-19 4. Warrell DA, Molyneux ME, Beales PF, eds. Severe and complicated malaria. Trans R Soc Trop Med Hyg 1990; 84 (suppl 2): 1-65. 5. Davis TME, Supanaranond W, Pukrittayakamee S, et al. A safe and effective consecutive-infusion regimen for rapid quinine loading in severe falciparum malaria. J Infect Dis 1990; 161: 1305-08.

Malaria, bednets, and mortality SIR,-Dr Velema (Sept 7, p 642) suggests alternative explanations for the large impact on childhood mortality associated with the introduction of insecticide-treated bednets reported by us (June 22, p 1499). He questions the importance of malaria as a cause of death in The Gambia. Velema accepts that insecticide-treated bednets may have prevented deaths caused by malaria. However, there was a further non-specific reduction in mortality, which we attributed to malaria acting as a risk factor for mortality due to other causes. This can be termed indirect mortality. Velema suggests that, rather than malaria being an indirect cause of death, the non-specific reduction in mortality might have resulted from weekly contact of children with village health workers (VHWs). In support of this he cites a study from Benin1 in which children seen by a VHW had a lower risk of dying in the next 6 months than children not seen. This non-specific protective effect of contact with a VHW could not be related to any particular curative or preventive measure. It may be that unprompted contact with health services is a marker of health awareness and of a protective health-conscious attitude, rather than a

childhood mortality is that, in addition to being a major direct cause of death, malaria is also an important factor in mortality due to other causes.

Institute de Parasitologia "Lopez Neyra", 18001 Granada, Spain; London School of Hygiene and Tropical Medicine, London, and Department of Zoology, University of Glasgow, UK

PEDRO L. ALONSO JOANNA R. M. ARMSTRONG STEVE W. LINDSAY

JP, Alihonou EM, Gandaho T, Hounye FH. Childhood mortality among of primary health care in a rural West African community. Int J Epidemiol 1991; 20: 474-79. 2. Greenwood BM, Greenwood AM, Bradley AK, et al. Comparison of two strategies for 1. Velema

users and non-users

control of malaria within a primary health care programme in The Gambia. Lancet 1988; i: 1121-27

Antithrombin III, blood loss, and transfusion requirements during liver transplantation SIR,-Severe intraoperative bleeding is one of the main problems during liver transplantation. Besides specific surgical or technical causes, it may be secondary to acquired haemostatic defects: hypoprothrombinaemia, haemodilution, thrombocytopenia, and primary or secondary hyperfibrinolysis. Between January, 1987, and December, 1990, 101 liver transplants were done at our institution, and the mean blood loss was 14 (SD 16) litres (1987-88), 5 (4) litres (1989), and 4 (3) litres (1990). During the first 22 transplants all patients were infused with antithrombin III (AT III), with the aim of maintaining levels above 80% activity, and were retrospectively divided into two groups according to the AT III value obtained throughout surgery (group A, more than 80%; group B, less than 80%). Group A had a significantly lower blood loss than group B (p :::; 0.05).1 Harper et al2 suggested AT III supplementation during the anhepatic phase and early after reperfusion to maintain values above 70% activity to control clotting activation and secondary fibrinolysis.2 These data were the rationale for a prospective randomised study in cirrhosis. From January, 1989, to April, 1990,29 patients with postnecrotic cirrhosis (Child score B and C) having liver transplantation were randomised to receive (13 patients) AT III supplementation or no supplementation (16 patients) (AT III activity was below 50% in both groups). The groups were similar in age, sex, preoperative Child score, and basal haemostatic indices. AT III (’Thrombhibin’, Immuno, Italy) was given by bolus infusion before the induction of anaesthesia. The dose was calculated to obtain a preoperative level of 100% (AT IIIunit= [100- plasma activity value]/kg bodyweight). Thereafter, 1000 IU/h were given by continuous infusion throughout surgery. Prothrombin; thrombin and reptilase times; activated partial thromboplastin time; platelet count; fibrinogen; AT III; plasminogen; &agr;2-antiplasmin; total (TDP), fibrinogen (FgDP), and fibrin (FbDP) degradation products; thrombin-antithrombin and complexes (TAT); COAGULATION INDICES, BLOOD LOSS, TRANSFUSION NEEDS, AND OPERATION TIME IN PATIENTS RECEIVING AT III AND CONTROLS

protective measure per se. In The Gambia, VHWs have few resources to exercise any direct

curative action: their role in preventive medicine and health education is negligible. As we reported in our paper, the comparison of mortality rates in villages with and without the presence of a simple, village-based primary health care programme suggests that this scheme does not significantly reduce child mortality in The Gambia. A similar conclusion was drawn in a previous study.2 We feel that the most likely explanation for the large reduction in

Mean (SD) values are shown TAT < ttg/mL FgDP80%, CLI=clot lysis mdex, PRC=packed red cells, FFP=fresh frozen plasma,

PLT = platelets *p < 0,001. tp 00001, Mann Whitney test.

898

thromboelastogram (rated by the clot-lysis index) were measured during the preanhepatic, anhepatic, and postanhepatic stages of transfusion. Total blood loss, transfusion requirements, and operation times were also recorded. AT III activity was significantly higher in the treated group throughout surgery (99 [4] vs 52 [2], p < 0 001). TAT, TDP, FgDP, and FbDP progressively increased during the anhepatic phase with maximum values early after reperfusion (table). Blood loss, transfusion requirements, and operation times were also much the same (table). This randomised study does not confirm our previous data: blood loss and transfusion requirements were not affected by AT IIIadministration. Bleeding related to surgical or technical dificulties may be more relevant to the intraoperative blood loss than variations in haemostatic indices. Haematology Department and Liver Transplantation Unit, Ospedale Niguarda Cà Granda, 20100 Milano, Italy

F. BAUDO A. DE GASPERI L. BELLI F. DE CATALDO

Gasperi A, Baudo F, Pannacciulli E, et al. Possible role of antithrombin III replacement therapy in reducing blood loss during orthotopic liver transplantation. Intensive Care Med 1988; 14 (suppl 1): 308. 2. Harper PL, Luddington RJ, Jennings I, et al. Coagulation changes following hepatic revascularization during liver transplantation. Transplantation 1989; 48: 603-07. 1. De

Large granular lymphocytes SIR,-Your Aug 3 editonal and Dr Woodcock and Dr Gordon’s letter

(Sept 7, p 634) on the proliferation of large granulocyte lymphocytes (LGL) prompt us to report findings from the Yorkshire Leukaemia Group. This group has, for the past two years, been studying the frequency of LGL expansion and has been trying to find out if "transient" and "persistent" forms can be differentiated on the basis of cell characteristics and clinical patterns. Since January, 1989, participating haematology departments throughout the Yorkshire region have referred blood samples from patients over 16 years of age who had a lymphocytosis above 4-5 x 109/1 (excluding cases of known lymphoproliferative disease), an increased ( > 25%) proportion of blood lymphocytes with LGL morphology, or neutropenia (< 2-0 x 109/1) of unknown aetiology. These samples were analysed for morphological and immunophenotypic variables, including absolute lymphocyte and LGL numbers and proportions of lymphocytes expressing membrane CD16, CD56, and CD57 NK-associated (NKa) determinants. On the basis of previously defined upper normal limits,1 those patients with absolute LGL, CD 16", CD56, and/or CD57" lymphocyte counts above 10x 109/1 were recorded as potential "persistent LGL expansions". Of 870 patients analysed in this initial phase, 269 (31 %) were recorded as such; the remaining 479 were categorised, after further investigations for the recognition of possible chronic B-cell or non-LGL T-cell lymphoproliferative disorders, as suspected or proven chronic lymphoid malignancy (B-cell, n = 78; non-LGL T-cell, n = 7), lymphocytosis characterised by non-specific increases in B and/or T cell (but not NKa’) subpopulations (n = 298), or as normal in respect of lymphocyte numbers and subpopulation distributions (n = 218). Follow-up investigations at 6 months were requested for all 269

patients registered

as

having potential persistent expansions.

112

(42%) were re-examined. In 92 the abnormal LGL/NKa’ expansion had persisted, being CD3+NKa+ in 47 and CD3-NKa’ in 45. In the remaining 20, LGL/NKa’ counts were normal at follow-up, and they were redefined as having transient LGL expansions. Most patients with persistent LGL/NKaexpansions were more than 60 years old, although 7 cases were aged 40 years or less. Absolute lymphocyte numbers were, in most cases, only slightly increased, and 34/92 patients had normal counts (figure). Clinical evaluation indicated a significant minority with rhematoid arthritis, polymyalgia rheumatica, and various cardiac problems but the clinical spectrum was a very wide one of diagnoses that could not easily be correlated with increased LGL/NKa+ components. Indeed, we suspect that in many of these patients the lymphocyte abnormalities were often unrelated to the primary diagnosis. All

and absolute lymphocyte count distributions in 92 patients with persistent LGL expansion.

Age

Upper age distribution (40, 60, and 80 year cut-offs indicated) Lower. absolute lymphocyte counts (normal limits of 1.5-4.5 x 109/1 indicated). Horizontal axes indicate patient numbers 1 to 92. tested (75/92) were negative for antibodies to HTLV-I/II and only 4 of 61 patients analysed had hypercalcaemia. Whether these LGL/NKa expansions are primary events, with a predisposition to distinct clinical symptoms at a later stage, or secondary is not known. Longer follow-up may clarify matters. This survey does indicate that the population frequency of persistent LGL expansions, primary or secondary, is higher than previously suspected, and that CD3-NKa expansions are more sera

common

than

suggested in published work. Although

some

LGL/NKa’ expansions are transient and related to acute clinical events, a high proportion of such patients may show a prolonged increase in LGL/NKa+ numbers. "Persistence" as defmed here does not necessarily imply permanence or lymphopoietic malignancy but it does suggest, even in symptom-free individuals, a chronic abnormality which clinicians should not dismiss as an incidental finding. Leukaemia Diagnostic and Molecular Haematology Units,

Cookridge Hospital, Leeds LS16 6QB, UK; and Leeds General Infirmary

COLIN S. SCOTT STEPHEN J. RICHARDS M. SIVAKUMARAN, for the Yorkshire Leukaemia

Group

1 Richards SJ, Scott CS. Immunophenotypic dissection of normal peripheral blood NK-associated (NKa) subpopulations by flow cytometry: morphological features and relationships between NKa (CD11b, CD16, CD56, and CD57) and T-cell (CD2, CD3, TCR, CD5, CD7, CD8, and CD38) associated determinant expression. Leuk Lymphoma 1990; 2: 111-26.

CORRECTION Correction by erythropoietin of anaemia in autonomic failure.-In this (Sept 21, p 759), the authors should have been P. Anand (National Hospital for Neurology and Neurosurgery, London WC1), F. E. Chen, C. D. L. Reid, and M. Coates (Northwick Park Hospital), and P. Rudge letter

(National Hospital for Neurology and Neurosurgery).

Antithrombin III, blood loss, and transfusion requirements during liver transplantation.

897 2000 parasites/)il of whole blood).4 Both treatment respect to age, body temperature, severity, groups and duration of coma on admission (table)...
308KB Sizes 0 Downloads 0 Views