TRANSACTIONS OF THE
Hydrocortisone New Guinea
ROYALSocwr~ or TROPICALMEDICINEAND HYGIENE(19911,
in chloramphenicol-treated
113
85, 113-l 16
severe typhoid
fever in Papua
Stephen J. Rogerson’*, Veronica J. Spooned, Tom A. Smith* and John Richens’ ‘Goroka Base Hospital, Goroka, Papua New Guinea; ‘Papua New Guinea Institute of Medical Research, Goroka, Papua New Guinea Abstract We investigated the effect of hydrocortisone on mortality and complications in chloramphenicoltreated severe typhoid fever (STF) in Goroka, Papua New Guinea. Of 374 culture-positive patients, 146 formed a retrospective comparison group, of whom 41 had STF. Of 228 patients in the intervention group, 58 had STF. Patients without STF had low mortality (2.5%) with standard treatment. In the intervention group, hydrocortisone was used in two dosage schedules, 100 mg for 12 doses (23 patients) and 400 mg for 12 doses(23 natients). There was no difference in mortality b&w&n steroid-treated and comparison STF patients (44.8% versus 43.9%) or in complications, and we conclude that moderate dosesof steroids are not beneficial in severe typhoid fever. Introduction The prevalence and severity of typhoid fever have recently increased markedly throughout the highlands of Papua New Guinea. In Goroka, RICHENS (1988) found an in-hospital mortality of 13.2% over a two-year period. Similarly high mortalitv rates ‘have been reported in many d&eloping countries (OSUNTOKUN et al.. 1972; KHOSLA, 1982: B~~MSMA, 19881. Corticosteroids have been used in physiological dosesin the treatment of typhoid fever since the 1950s (SMADEL et al., 1951; WISSEFMAN et al., 1954). To our knowledge, there has been no controlled trial of the effect on mortality of such doses. HOFFMAN et al. (1984) ip Jakarta reported that very high doses of dexamethasone (11 mgikg over 48 h) markedly decreasedmortality in severe typhoid fever (STF). This therapy is expensive, and is controversial, being based on results of earlier studies using high doses of corticosteroids in Gram-negative shock (SHEAGREN, 1981), an indication which has now been discredited (BONE et al.. 1987; VETERANS ADMINISTRATION SYSTEMIC SEPSISCOOPERATIVE STUDY GROUP, 1987).
On the basis of published recommendations (HOOK, 1984) and the advice of our ethical committee it was considered inappropriate to carry out a dacebo-controlled trial. Instead. we cornDared the in-hospital mortality of hydrocortisone-treated STF natients with that of STF natients admitted immediatelv before the introdu&ion of hvdrocortisone therapy.- Two dosage schedules of -100 mg and 400 ma, 4 times dailv for 12 doses, were evaluated. These are close to current recommended doses and are appropriate for use in the developing world. High-dose dexamethasone was not available for evaluation. *Present address: Regional Infectious Diseases Unit, Fazakerlev Hosnital. Liveruool. UK. Addres$ for correiponden’ce and offprints: Dr S. Roger~2, 17 Oakdale Road, Waterloo, LIverpool, L22 9QS,
In order to confirm that our results were not confounded by secular trends in the pattern of disease we also compared the symptoms and outcomes of all typhoid patients for the period during which the treatment was offered with those of previous patients. Materials and Methods All adult patients with a discharge diagnosis of typhoid fever admitted to the medical ward, Goroka Base Hospital, between 3 December 1987 and 12 March 1989 formed the intervention group. The comparison group included all typhoid patients admitted from 1 Ianuarv 1986 to 2 December 1987. Analysis was cor&ned to those patients with positive blood and/or bone marrow culture for Salmonella twhi, durina their first admission. --Severe typhoid was defined using the scheme of HOFFMAN et al. (1984) as showine one or more of the following signs:‘ (i) ‘delirium, -markedly confused thinking and speech; (ii) obtundation, appearing unconscious but rouseable and able to respond to commands; (iii) stupor, responding only to painful stimuli; (iiv) coma, no response to painful stimuli; and (v) shock. We defined shock as systolic blood pressure
Hydrocortisone New Guinea
ROYALSocwr~ or TROPICALMEDICINEAND HYGIENE(19911,
in chloramphenicol-treated
113
85, 113-l 16
severe typhoid
fever in Papua
Stephen J. Rogerson’*, Veronica J. Spooned, Tom A. Smith* and John Richens’ ‘Goroka Base Hospital, Goroka, Papua New Guinea; ‘Papua New Guinea Institute of Medical Research, Goroka, Papua New Guinea Abstract We investigated the effect of hydrocortisone on mortality and complications in chloramphenicoltreated severe typhoid fever (STF) in Goroka, Papua New Guinea. Of 374 culture-positive patients, 146 formed a retrospective comparison group, of whom 41 had STF. Of 228 patients in the intervention group, 58 had STF. Patients without STF had low mortality (2.5%) with standard treatment. In the intervention group, hydrocortisone was used in two dosage schedules, 100 mg for 12 doses (23 patients) and 400 mg for 12 doses(23 natients). There was no difference in mortality b&w&n steroid-treated and comparison STF patients (44.8% versus 43.9%) or in complications, and we conclude that moderate dosesof steroids are not beneficial in severe typhoid fever. Introduction The prevalence and severity of typhoid fever have recently increased markedly throughout the highlands of Papua New Guinea. In Goroka, RICHENS (1988) found an in-hospital mortality of 13.2% over a two-year period. Similarly high mortalitv rates ‘have been reported in many d&eloping countries (OSUNTOKUN et al.. 1972; KHOSLA, 1982: B~~MSMA, 19881. Corticosteroids have been used in physiological dosesin the treatment of typhoid fever since the 1950s (SMADEL et al., 1951; WISSEFMAN et al., 1954). To our knowledge, there has been no controlled trial of the effect on mortality of such doses. HOFFMAN et al. (1984) ip Jakarta reported that very high doses of dexamethasone (11 mgikg over 48 h) markedly decreasedmortality in severe typhoid fever (STF). This therapy is expensive, and is controversial, being based on results of earlier studies using high doses of corticosteroids in Gram-negative shock (SHEAGREN, 1981), an indication which has now been discredited (BONE et al.. 1987; VETERANS ADMINISTRATION SYSTEMIC SEPSISCOOPERATIVE STUDY GROUP, 1987).
On the basis of published recommendations (HOOK, 1984) and the advice of our ethical committee it was considered inappropriate to carry out a dacebo-controlled trial. Instead. we cornDared the in-hospital mortality of hydrocortisone-treated STF natients with that of STF natients admitted immediatelv before the introdu&ion of hvdrocortisone therapy.- Two dosage schedules of -100 mg and 400 ma, 4 times dailv for 12 doses, were evaluated. These are close to current recommended doses and are appropriate for use in the developing world. High-dose dexamethasone was not available for evaluation. *Present address: Regional Infectious Diseases Unit, Fazakerlev Hosnital. Liveruool. UK. Addres$ for correiponden’ce and offprints: Dr S. Roger~2, 17 Oakdale Road, Waterloo, LIverpool, L22 9QS,
In order to confirm that our results were not confounded by secular trends in the pattern of disease we also compared the symptoms and outcomes of all typhoid patients for the period during which the treatment was offered with those of previous patients. Materials and Methods All adult patients with a discharge diagnosis of typhoid fever admitted to the medical ward, Goroka Base Hospital, between 3 December 1987 and 12 March 1989 formed the intervention group. The comparison group included all typhoid patients admitted from 1 Ianuarv 1986 to 2 December 1987. Analysis was cor&ned to those patients with positive blood and/or bone marrow culture for Salmonella twhi, durina their first admission. --Severe typhoid was defined using the scheme of HOFFMAN et al. (1984) as showine one or more of the following signs:‘ (i) ‘delirium, -markedly confused thinking and speech; (ii) obtundation, appearing unconscious but rouseable and able to respond to commands; (iii) stupor, responding only to painful stimuli; (iiv) coma, no response to painful stimuli; and (v) shock. We defined shock as systolic blood pressure
