Eur J Clin Pharmacol (1992) 42:107-110
@ Springer-Verlag 1992
Short communications
Absorption of intramuscular phenobarbitone in children with severe falciparum malaria E ter Kuile 1' 2, 4, F. Nosten 1, 4, T. Chongsuphajaisiddhi 1, P. Holloway s, L. Maelankirri 4, and N. J. White 1' 3 Faculty of Tropical Medicine, Mahidol University, Bangkok 10400, Thailand 2 Unit of Infectious Diseases and Tropical Medicine, Academic Medical Center, University of Amsterdam, The Netherlands Nuffield Department of Clinical Medicine, University of Oxford, U. K. 4 Shoklo Malaria Research Unit, Mae Sod, Thailand Received: May 7, 1991/Accepted in revised form: May 22, 1991
Summary. T h e absorption of intramuscular p h e n o b a r b i tone 7 mg. k g - 1was studied in 11 K a r e n children aged bet w e e n 1.7 and 11 y with severe falciparum malaria. Eight of the children were comatose. Clinical findings were c o m p a r e d with those in 9 further children with severe malaria of similar age range (four of w h o m were unconscious), w h o received an identical placebo. O n e child, w h o had received placebo, had r e p e a t e d convulsions and died 1 h after admission to hospital. T h e r e m a i n d e r m a d e an u n c o m p l i c a t e d recovery. T h e r e were no convulsions subsequent to treatment, although the study was too small to assess anticonvulsant efficacy. T h e r e was no observable toxicity, but p h e n o b a r b i t o n e recipients had a significant t e n d e n c y to d e e p e n in their level of c o m a or to b e c o m e sleepy within the 4 h after drug administration. P h e n o b a r bitone was rapidly absorbed, reaching a m e a n (range) p e a k c o n c e n t r a t i o n of 34.2 [29.342.6] g m o l - 1 - ~ in a median (range) of 4 (2.5-12) h. These values are c o m p a r a b l e to those previously r e p o r t e d in healthy children and in children with febrile convulsions. I n t r a m u s c u l a r p h e n o barbitone is well a b s o r b e d in children with severe malaria; the o p t i m u m prophylactic anticonvulsant dose remains to be determined. Key words: P h e n o b a r b i t o n e , Cerebral malaria; E falciparum, kinetics, drug absorption, children
Convulsions are an i m p o r t a n t complication of severe malaria, particularly in childhood. T h e y m a y cause death f r o m aspiration p n e u m o n i a or neurological sequelae in survivors (Brewster et al. 1990). In a recent, double blind, placebo-controlled trial, a single small dose (3.5 mg- kg -1) of intramuscular p h e n o b a r b i t o n e was shown to reduce the incidence of convulsions in adults with cerebral malaria (White et al. 1988). H o w e v e r most cases of cerebral m a laria occur in children, and seizures are a particular problem in this age group. Severe malaria is associated with dist u r b a n c e of microvascular b l o o d flow, skeletal muscle damage, and alteration in the disposition of the antimalarial drugs (White 1986). T h e objective of the study was
to determine the rate o f p h e n o b a r b i t o n e absorptioni.s, following intramuscular injection and the plasma concentration profile during c o m a in children with severe malaria.
Patients and methods
Patients Children of the Karen ethnic minority aged 18 months to 12 y and presenting with severe malaria (WHO 1990) were studied. The investigation was part of a placebo controlled double blind study of phenobarbitone anticonvulsant prophylaxis in children with severe malaria and forms part of a prospective study of severe malaria in childhood. The children were admitted to a specially constructed hospital located in the largest of a series of camps for displaced persons, situated in a malarious area of the Thai-border 110 km north of the town of Mae Sod. The epidemiology and clinical pattern of malaria in this area has been described previously (Nosten et al. 1987). Children were included only if their parents gave fully informed consent. The studies were approved by the Ethical Committee of the Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand.
Clinical procedures On admission a blood smear was performed to confirm the diagnosis, baseline blood samples ( < 5 ml) were taken and an infusion of 0.9% saline was started. Patients were randomised in pairs to receive either quinine dihydrochloride (20 mg salt, kg- ~) given by intravenous infusion over 4 h, or by split intramuscular injection to the anterior thighs (Waller et al. 1990). Patients were also independently randomised to receive either phenobarbitone sodium (Gardenal, May & Baker, 200mg-m1-1) or an identical placebo. The drugs were drawn up in a 1 ml syringe, and a volume corresponding to 7 mg. kg- 1was given by single intramuscular injection to the anterior thigh when quinine treatment was started. The clinicians and nurses were not aware which drug had been given. The children were carefully rehydrated. A lumbar puncture was performed on all comatose children. A small (22G) teflon catheter was inserted into an antecubital vein for repeated blood sampling and patency was maintained with heparin-saline. Blood samples (1 ml) were taken at 30 rain and after 1, 2.5, 4, 5, 6, 8,12,16, and 24 h. Plasma was separated and stored at - 20°C until analysis. Vital signs and neurological observations were recorded hourly. A subjective assessment was made by the clinician at each observation in the first 4 h following the injection as to whether the conscious children be-
108
F. ter Kuile et al.: Phenobarbitone absorption in cerebral malaria
Table 1. Clinical details of 21 children with severe malaria treated with intramuscular phenobarbitone (7 mg/kg) or placebo
Age (years) Weight (kg) Coma On admission: Core temperature (°C) Pulse rate (beats- min-1) SystolicBP (mmHg) Haematocrit (%) Geometric mean parasite count (gl-1) Recovery (h): Until afebrile Untilaparasitaemic Until fully conscious
Phenobarbitone (n = 12) Mean (SD) Range 4.5 (2.4) 1.7-11 12.7 (5.0) 9-23 8 37.9 136 101 18.6
(1.2) (15) (19) (8.9)
5.75-104 36.6 (20.1) 54.0 (24.1) 14.4 (11.7)
Placebo (n = 9) Mean (SD) 6.0 (3.8) 16.1 (5.2) 4
36.2-40.5 100-166 80-140 8-38
38.3 124 103 24.9
591-8.1.105
9.33-104
5.36.103-3.32 •105
24.5 (9.2) 40.2 (3.8) 8.7 (6.0)
9-36 36-46 3-10
3-72 26-90 1-35
(0.8) (20) (23) (8.6)
Range 2-11 11-24
37.0-39.4 100-166 70-150 13-40
None of the differences was statistically significant came more sleepy, or in cerebral malaria, whether the modified Glasgow coma scale score changed (Molyneux et al. 1989).
Phenobarbital assay Phenobarbital was measured using an EMIT (Syva, Palo Alto, CA) assay on a Cobas FARA centrifugal analyser (Hoffman - la Roche, Basle, Switzerland).
Statistical analysis e
Comparison of means was performed by the Student t-test, and comparison of proportions by Fisher's exact test.
Results
Clinical details Twenty one children (11 boys, 10 girls) with severe malaria were recruited into the studies. Details of clinical parameters are given in Table 1. Twelve children received phenobarbitone, and in 11 of them plasma phenobarbitone levels were analysed (the samples went astray in the remaining case). Nine children received an identical placebo. Five of the 12 (42%) children in the phenobarbitone group, and 2 out of 9 (22%) in the placebo group had been given diazepam (0.3 mg-kg -1 parenterally or rectally) because of convulsions less than 2 h before they entered the study (P : 0 . 3 ) . Eight of 12children (67%) receiving phenobarbitone and 4 of 9 (44%) patients receiving placebo, respectively, were in unrousable coma (cerebral malaria) on admission (P = 0.3). None of the patients was hypoglycaemic. One placebo recipient, who was deeply comatose on admission had repeated convulsions and died I h after starting treatment. None of the 20 surviving patients had subsequent convulsions and all made an uncomplicated recovery.
There was no difference in clinical, biochemical or parasitological responses to the two treatment regimens. The mean (SD) time to regain consciousness fully was short in both groups; 14.4 (11.6)h in children who received phenobarbitone, compared to 8.7 (6.0) h in placebo recipients (P = 0.45). Of the patients surviving cerebral malaria, 5/8 (63%) who received phenobarbitone, and 0/3 who received placebo, had a drop in coma score within 4 h following the onset of treatment. Of the conscious children 3/4 who received phenobarbitone were estimated to be more sleepy after the injection, compared to 1/5 conscious controls. Thus, 8/12 phenobarbitone recipients compared to 1/8 surviving placebo recipients showed a deterioration in the level of consciousness or coma following the injection, P = 0.035, but in none of them was the deterioration considered clinically important.
Phenobarbitone concentrations Intramuscular phenobarbitone was well absorbed. There was relatively little variance between patients in the plasma concentrations. Mean peak concentrations of 34.2 [range 29.3-42.6] gmol.1 ~ were reached in a median (range) of 4 (2.5-12) h. Thereafter plasma concentrations tended to plateau. All values exceeded 28 gmol-1 -~ (i. e. 80% of the peak level) by 4 h. At 1 h and at 2.5 h the mean (SD) plasma phenobarbitone concentrations were 72 (19)% and 91 (11)% of the eventual peak level respectively. The profile of plasma concentrations is shown in Fig, 1.
Discussion In severe falciparum malaria, convulsions often mark the onset of coma, and in cerebral malaria, seizures are a common cause of aspiration pneumonia and a deteriora-
109
F. ter Kuile et al.: Phenobarbitone absorption in cerebral malaria 44 4O v,
36
E o
x22
O
£ k0
Fig. 1. Profile of plasma phenobar-
(3_
0
'
I
I
I
]
4
8
12
16
20
bitone concentrations following intramuscular injection of 7 rag. kg 1 in eleven children with severe malaria 214
Time (h)
tion in neurological status ( W H O 1990). Fits are also common in conscious infants and young children with malaria. Protracted seizures in children with severe malaria are associated with residual neurological deficit in survivors (Brewster et al. 1990). The indications for and the optimum drug treatment and dose for anticonvulsant prophylaxis and treatment in severe malaria remain to be determined. Phenobarbitone has the advantages of simplicity; because of its slow elimination a single dose provides anticonvulsant 'cover' for several days. It is stable, cheap, easy to administer, widely available and its anticonvulsant properties have been well-characterised. In an earlier prospective evaluation of anticonvulsants in cerebral malaria (White et al. 1988), a small dose of phenobarbitone (3.5 m g . k g -1) was studied because of initial uncertainty whether the additional sedative effect might prove harmful in patients already comatose with cerebral malaria. However no deleterious effects were noted. Fits are a particular problem in childhood malaria, and the seizures may be refractory to anticonvulsant treatment. It was decided, therefore, to evaluate a higher dose (7 mg-kg -1) here, i.s. one that although was still considerably lower than the usually recommended rapid loading dose of 20 mg. kg- 1. Phenobarbitone was absorbed quickly after intramuscular injection, reaching a peak concentration within 4 h. A similar pattern of rapid absorption has been observed in children with febrile convulsions (Pearce et al. 1977) and those given intramuscular phenobarbitone as a premedication (Brachet-Liermain et al. 1977). As would be expected from the slow metabolic clearance of the drug, plasma concentrations changed relatively little thereafter in the first 24 h (the average duration of coma in childhood cerebral malaria) when anticonvulsant cover would be most needed. However, with a dose of 7 m g . k g -~ the levels achieved were lower than those usually recommended for optimum anticonvulsant effect (45110 gmol. 1-1) (Clayton and Round 1984). The mean peak
plasma concentration of 34.2 gmol- 1-1 may be compared with mean peak values of 53.2~tmo1.1-1 (Pearce et al. 1977) and 60 gmol. 1 1 (Brachet-Liermain et al. 1977) following an intramuscular dose of 10 mg. kg- 1 in children with febrile convulsions, and 80 ~tmol. 1-~ after 15 mg. kg- 1given to healthy children. The pharmacodynamic properties of anticonvulsant drugs have not been defined in severe malaria. A small dose of phenobarbitone (3.5 rag. kg 1) prevented convulsions in adults (White et al. 1988), and a larger dose (5 rag. kg -1) produced no deterioration of coma grade in 50 Gambian children with cerebral malaria, (White NJ, unpublished observations). In the present series of 21 children with severe malaria (12 of whom had cerebral malaria) none of those treated with phenobarbitone subsequently had a convulsion, but they were more likely to have a fall in coma score, or to become more drowsy than the placebo-treated controls. Despite this, phenobarbitone did not prolong the duration of coma or delay recovery. The number of children studied was too small to make confident statements either about anticonvulsant efficacy or adverse effects. However, it can be stated with confidence that intramuscular phenobarbitone was well absorbed in severe and cerebral malaria and that the relationship between dose and initial blood concentrations was similar to that in other febrile conditions. Further studies are needed in a larger series of children with cerebral malaria to determine the optimum prophylactic- anticonvulsant dose of intramuscular phenobarbitone.
Acknowledgements. We thank the staff of the hospital in Shoklo and the nurses and doctors of Mddecins Sans Frontieres for their help and support, and Nucharee Cholvilai. E ter Kuile is very grateful to 'Stichting de Drie Lichten ~and the 'Professor Stokvis Reisfonds' for financial support. The study was part of the Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme funded by the Wellcome Trust of Great Britain.
110
References Brachet-Liermain A, Goutieres F, Aicardi J (1975) Absorption of phenobarbital after the intramuscular administration of single doses in infants. J Pediatr 87:624%26 Brewster D, Kwiatkowski D, White NJ (1990) Neurological sequelae of cerebral malaria in childhood. Lancet II: 1039-1043 Clayton BE, Round JM (1984) Chemical pathology and the sick child. Blackwell, Oxford, p 422 Molyneux M, Taylor TE, Wirima JJ, Borgstein A (1989) Clinical features and prognostic indicators in paediatric cerebral malaria: a study of 131 comatose Malawian children. Q J Med 71:441-459 Nosten E Imvithaya S, Vincenti M, Delmas G, Lebihan G, Hausler B, White NJ (1987) Malaria on the Thai-Burmese border; treatment of 5192patients with mefloquine-sulfadoxine-pyrimethamine combination. Bull WHO 65:891-896 Pearce JL, Sharman JR, Forster RM (1977) Phenobarbital in the acute management of febrile convulsions. Pediatrics 60:569-572 Waller D, Krishna S, Craddock C, Brewster D, Jammeh A, Kwiatkowski D, Karbwang J, Molunto R White NJ (1990) The pharma-
E ter Kuile et al.: Phenobarbitone absorption in cerebral malaria cokinetic properties of intramuscular quinine in Gambian children with severe falciparum malaria. Trans R Soc Trop Med Hyg 84:488-491 White NJ (1986) Pathophysiology in clinics in tropical medicine and communicable disease, vol 1. Malaria. In: Strickland GT (ed) Pathophysiology. Saunders, London, pp 55-90 White NJ, Looareesuwan S, Chanthavanich R Phillips RE, Warrell D A (1988) Single dose phenobarbitone prevents convulsions in cerebral malaria. Lancet II: 64-66 World Health Organization (1990) Severe and complicated malaria. Trans R Soc Trop Med Hyg 84 [Suppl 2]: 1%5
Dr. N. J. White Faculty of Tropical Medicine Mahidol University 420/6 Rajvithi Road Bangkok 10400 Thailand
@ Springer-Verlag 1992
Short communications
Absorption of intramuscular phenobarbitone in children with severe falciparum malaria E ter Kuile 1' 2, 4, F. Nosten 1, 4, T. Chongsuphajaisiddhi 1, P. Holloway s, L. Maelankirri 4, and N. J. White 1' 3 Faculty of Tropical Medicine, Mahidol University, Bangkok 10400, Thailand 2 Unit of Infectious Diseases and Tropical Medicine, Academic Medical Center, University of Amsterdam, The Netherlands Nuffield Department of Clinical Medicine, University of Oxford, U. K. 4 Shoklo Malaria Research Unit, Mae Sod, Thailand Received: May 7, 1991/Accepted in revised form: May 22, 1991
Summary. T h e absorption of intramuscular p h e n o b a r b i tone 7 mg. k g - 1was studied in 11 K a r e n children aged bet w e e n 1.7 and 11 y with severe falciparum malaria. Eight of the children were comatose. Clinical findings were c o m p a r e d with those in 9 further children with severe malaria of similar age range (four of w h o m were unconscious), w h o received an identical placebo. O n e child, w h o had received placebo, had r e p e a t e d convulsions and died 1 h after admission to hospital. T h e r e m a i n d e r m a d e an u n c o m p l i c a t e d recovery. T h e r e were no convulsions subsequent to treatment, although the study was too small to assess anticonvulsant efficacy. T h e r e was no observable toxicity, but p h e n o b a r b i t o n e recipients had a significant t e n d e n c y to d e e p e n in their level of c o m a or to b e c o m e sleepy within the 4 h after drug administration. P h e n o b a r bitone was rapidly absorbed, reaching a m e a n (range) p e a k c o n c e n t r a t i o n of 34.2 [29.342.6] g m o l - 1 - ~ in a median (range) of 4 (2.5-12) h. These values are c o m p a r a b l e to those previously r e p o r t e d in healthy children and in children with febrile convulsions. I n t r a m u s c u l a r p h e n o barbitone is well a b s o r b e d in children with severe malaria; the o p t i m u m prophylactic anticonvulsant dose remains to be determined. Key words: P h e n o b a r b i t o n e , Cerebral malaria; E falciparum, kinetics, drug absorption, children
Convulsions are an i m p o r t a n t complication of severe malaria, particularly in childhood. T h e y m a y cause death f r o m aspiration p n e u m o n i a or neurological sequelae in survivors (Brewster et al. 1990). In a recent, double blind, placebo-controlled trial, a single small dose (3.5 mg- kg -1) of intramuscular p h e n o b a r b i t o n e was shown to reduce the incidence of convulsions in adults with cerebral malaria (White et al. 1988). H o w e v e r most cases of cerebral m a laria occur in children, and seizures are a particular problem in this age group. Severe malaria is associated with dist u r b a n c e of microvascular b l o o d flow, skeletal muscle damage, and alteration in the disposition of the antimalarial drugs (White 1986). T h e objective of the study was
to determine the rate o f p h e n o b a r b i t o n e absorptioni.s, following intramuscular injection and the plasma concentration profile during c o m a in children with severe malaria.
Patients and methods
Patients Children of the Karen ethnic minority aged 18 months to 12 y and presenting with severe malaria (WHO 1990) were studied. The investigation was part of a placebo controlled double blind study of phenobarbitone anticonvulsant prophylaxis in children with severe malaria and forms part of a prospective study of severe malaria in childhood. The children were admitted to a specially constructed hospital located in the largest of a series of camps for displaced persons, situated in a malarious area of the Thai-border 110 km north of the town of Mae Sod. The epidemiology and clinical pattern of malaria in this area has been described previously (Nosten et al. 1987). Children were included only if their parents gave fully informed consent. The studies were approved by the Ethical Committee of the Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand.
Clinical procedures On admission a blood smear was performed to confirm the diagnosis, baseline blood samples ( < 5 ml) were taken and an infusion of 0.9% saline was started. Patients were randomised in pairs to receive either quinine dihydrochloride (20 mg salt, kg- ~) given by intravenous infusion over 4 h, or by split intramuscular injection to the anterior thighs (Waller et al. 1990). Patients were also independently randomised to receive either phenobarbitone sodium (Gardenal, May & Baker, 200mg-m1-1) or an identical placebo. The drugs were drawn up in a 1 ml syringe, and a volume corresponding to 7 mg. kg- 1was given by single intramuscular injection to the anterior thigh when quinine treatment was started. The clinicians and nurses were not aware which drug had been given. The children were carefully rehydrated. A lumbar puncture was performed on all comatose children. A small (22G) teflon catheter was inserted into an antecubital vein for repeated blood sampling and patency was maintained with heparin-saline. Blood samples (1 ml) were taken at 30 rain and after 1, 2.5, 4, 5, 6, 8,12,16, and 24 h. Plasma was separated and stored at - 20°C until analysis. Vital signs and neurological observations were recorded hourly. A subjective assessment was made by the clinician at each observation in the first 4 h following the injection as to whether the conscious children be-
108
F. ter Kuile et al.: Phenobarbitone absorption in cerebral malaria
Table 1. Clinical details of 21 children with severe malaria treated with intramuscular phenobarbitone (7 mg/kg) or placebo
Age (years) Weight (kg) Coma On admission: Core temperature (°C) Pulse rate (beats- min-1) SystolicBP (mmHg) Haematocrit (%) Geometric mean parasite count (gl-1) Recovery (h): Until afebrile Untilaparasitaemic Until fully conscious
Phenobarbitone (n = 12) Mean (SD) Range 4.5 (2.4) 1.7-11 12.7 (5.0) 9-23 8 37.9 136 101 18.6
(1.2) (15) (19) (8.9)
5.75-104 36.6 (20.1) 54.0 (24.1) 14.4 (11.7)
Placebo (n = 9) Mean (SD) 6.0 (3.8) 16.1 (5.2) 4
36.2-40.5 100-166 80-140 8-38
38.3 124 103 24.9
591-8.1.105
9.33-104
5.36.103-3.32 •105
24.5 (9.2) 40.2 (3.8) 8.7 (6.0)
9-36 36-46 3-10
3-72 26-90 1-35
(0.8) (20) (23) (8.6)
Range 2-11 11-24
37.0-39.4 100-166 70-150 13-40
None of the differences was statistically significant came more sleepy, or in cerebral malaria, whether the modified Glasgow coma scale score changed (Molyneux et al. 1989).
Phenobarbital assay Phenobarbital was measured using an EMIT (Syva, Palo Alto, CA) assay on a Cobas FARA centrifugal analyser (Hoffman - la Roche, Basle, Switzerland).
Statistical analysis e
Comparison of means was performed by the Student t-test, and comparison of proportions by Fisher's exact test.
Results
Clinical details Twenty one children (11 boys, 10 girls) with severe malaria were recruited into the studies. Details of clinical parameters are given in Table 1. Twelve children received phenobarbitone, and in 11 of them plasma phenobarbitone levels were analysed (the samples went astray in the remaining case). Nine children received an identical placebo. Five of the 12 (42%) children in the phenobarbitone group, and 2 out of 9 (22%) in the placebo group had been given diazepam (0.3 mg-kg -1 parenterally or rectally) because of convulsions less than 2 h before they entered the study (P : 0 . 3 ) . Eight of 12children (67%) receiving phenobarbitone and 4 of 9 (44%) patients receiving placebo, respectively, were in unrousable coma (cerebral malaria) on admission (P = 0.3). None of the patients was hypoglycaemic. One placebo recipient, who was deeply comatose on admission had repeated convulsions and died I h after starting treatment. None of the 20 surviving patients had subsequent convulsions and all made an uncomplicated recovery.
There was no difference in clinical, biochemical or parasitological responses to the two treatment regimens. The mean (SD) time to regain consciousness fully was short in both groups; 14.4 (11.6)h in children who received phenobarbitone, compared to 8.7 (6.0) h in placebo recipients (P = 0.45). Of the patients surviving cerebral malaria, 5/8 (63%) who received phenobarbitone, and 0/3 who received placebo, had a drop in coma score within 4 h following the onset of treatment. Of the conscious children 3/4 who received phenobarbitone were estimated to be more sleepy after the injection, compared to 1/5 conscious controls. Thus, 8/12 phenobarbitone recipients compared to 1/8 surviving placebo recipients showed a deterioration in the level of consciousness or coma following the injection, P = 0.035, but in none of them was the deterioration considered clinically important.
Phenobarbitone concentrations Intramuscular phenobarbitone was well absorbed. There was relatively little variance between patients in the plasma concentrations. Mean peak concentrations of 34.2 [range 29.3-42.6] gmol.1 ~ were reached in a median (range) of 4 (2.5-12) h. Thereafter plasma concentrations tended to plateau. All values exceeded 28 gmol-1 -~ (i. e. 80% of the peak level) by 4 h. At 1 h and at 2.5 h the mean (SD) plasma phenobarbitone concentrations were 72 (19)% and 91 (11)% of the eventual peak level respectively. The profile of plasma concentrations is shown in Fig, 1.
Discussion In severe falciparum malaria, convulsions often mark the onset of coma, and in cerebral malaria, seizures are a common cause of aspiration pneumonia and a deteriora-
109
F. ter Kuile et al.: Phenobarbitone absorption in cerebral malaria 44 4O v,
36
E o
x22
O
£ k0
Fig. 1. Profile of plasma phenobar-
(3_
0
'
I
I
I
]
4
8
12
16
20
bitone concentrations following intramuscular injection of 7 rag. kg 1 in eleven children with severe malaria 214
Time (h)
tion in neurological status ( W H O 1990). Fits are also common in conscious infants and young children with malaria. Protracted seizures in children with severe malaria are associated with residual neurological deficit in survivors (Brewster et al. 1990). The indications for and the optimum drug treatment and dose for anticonvulsant prophylaxis and treatment in severe malaria remain to be determined. Phenobarbitone has the advantages of simplicity; because of its slow elimination a single dose provides anticonvulsant 'cover' for several days. It is stable, cheap, easy to administer, widely available and its anticonvulsant properties have been well-characterised. In an earlier prospective evaluation of anticonvulsants in cerebral malaria (White et al. 1988), a small dose of phenobarbitone (3.5 m g . k g -1) was studied because of initial uncertainty whether the additional sedative effect might prove harmful in patients already comatose with cerebral malaria. However no deleterious effects were noted. Fits are a particular problem in childhood malaria, and the seizures may be refractory to anticonvulsant treatment. It was decided, therefore, to evaluate a higher dose (7 mg-kg -1) here, i.s. one that although was still considerably lower than the usually recommended rapid loading dose of 20 mg. kg- 1. Phenobarbitone was absorbed quickly after intramuscular injection, reaching a peak concentration within 4 h. A similar pattern of rapid absorption has been observed in children with febrile convulsions (Pearce et al. 1977) and those given intramuscular phenobarbitone as a premedication (Brachet-Liermain et al. 1977). As would be expected from the slow metabolic clearance of the drug, plasma concentrations changed relatively little thereafter in the first 24 h (the average duration of coma in childhood cerebral malaria) when anticonvulsant cover would be most needed. However, with a dose of 7 m g . k g -~ the levels achieved were lower than those usually recommended for optimum anticonvulsant effect (45110 gmol. 1-1) (Clayton and Round 1984). The mean peak
plasma concentration of 34.2 gmol- 1-1 may be compared with mean peak values of 53.2~tmo1.1-1 (Pearce et al. 1977) and 60 gmol. 1 1 (Brachet-Liermain et al. 1977) following an intramuscular dose of 10 mg. kg- 1 in children with febrile convulsions, and 80 ~tmol. 1-~ after 15 mg. kg- 1given to healthy children. The pharmacodynamic properties of anticonvulsant drugs have not been defined in severe malaria. A small dose of phenobarbitone (3.5 rag. kg 1) prevented convulsions in adults (White et al. 1988), and a larger dose (5 rag. kg -1) produced no deterioration of coma grade in 50 Gambian children with cerebral malaria, (White NJ, unpublished observations). In the present series of 21 children with severe malaria (12 of whom had cerebral malaria) none of those treated with phenobarbitone subsequently had a convulsion, but they were more likely to have a fall in coma score, or to become more drowsy than the placebo-treated controls. Despite this, phenobarbitone did not prolong the duration of coma or delay recovery. The number of children studied was too small to make confident statements either about anticonvulsant efficacy or adverse effects. However, it can be stated with confidence that intramuscular phenobarbitone was well absorbed in severe and cerebral malaria and that the relationship between dose and initial blood concentrations was similar to that in other febrile conditions. Further studies are needed in a larger series of children with cerebral malaria to determine the optimum prophylactic- anticonvulsant dose of intramuscular phenobarbitone.
Acknowledgements. We thank the staff of the hospital in Shoklo and the nurses and doctors of Mddecins Sans Frontieres for their help and support, and Nucharee Cholvilai. E ter Kuile is very grateful to 'Stichting de Drie Lichten ~and the 'Professor Stokvis Reisfonds' for financial support. The study was part of the Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme funded by the Wellcome Trust of Great Britain.
110
References Brachet-Liermain A, Goutieres F, Aicardi J (1975) Absorption of phenobarbital after the intramuscular administration of single doses in infants. J Pediatr 87:624%26 Brewster D, Kwiatkowski D, White NJ (1990) Neurological sequelae of cerebral malaria in childhood. Lancet II: 1039-1043 Clayton BE, Round JM (1984) Chemical pathology and the sick child. Blackwell, Oxford, p 422 Molyneux M, Taylor TE, Wirima JJ, Borgstein A (1989) Clinical features and prognostic indicators in paediatric cerebral malaria: a study of 131 comatose Malawian children. Q J Med 71:441-459 Nosten E Imvithaya S, Vincenti M, Delmas G, Lebihan G, Hausler B, White NJ (1987) Malaria on the Thai-Burmese border; treatment of 5192patients with mefloquine-sulfadoxine-pyrimethamine combination. Bull WHO 65:891-896 Pearce JL, Sharman JR, Forster RM (1977) Phenobarbital in the acute management of febrile convulsions. Pediatrics 60:569-572 Waller D, Krishna S, Craddock C, Brewster D, Jammeh A, Kwiatkowski D, Karbwang J, Molunto R White NJ (1990) The pharma-
E ter Kuile et al.: Phenobarbitone absorption in cerebral malaria cokinetic properties of intramuscular quinine in Gambian children with severe falciparum malaria. Trans R Soc Trop Med Hyg 84:488-491 White NJ (1986) Pathophysiology in clinics in tropical medicine and communicable disease, vol 1. Malaria. In: Strickland GT (ed) Pathophysiology. Saunders, London, pp 55-90 White NJ, Looareesuwan S, Chanthavanich R Phillips RE, Warrell D A (1988) Single dose phenobarbitone prevents convulsions in cerebral malaria. Lancet II: 64-66 World Health Organization (1990) Severe and complicated malaria. Trans R Soc Trop Med Hyg 84 [Suppl 2]: 1%5
Dr. N. J. White Faculty of Tropical Medicine Mahidol University 420/6 Rajvithi Road Bangkok 10400 Thailand
