1310 little valued ginseng leavesand to prepare chemically standardised ginseng root extracts. These compounds and extracts could then be made available to investigators. Barna is incorrect to claim that there has been only one wellcontrolled ginseng study in man. The definition of what constitutes a well-controlled experiment may be controversial, but controlled human studies do exist for an effect of ginseng on hyperlipidaemiaz2 on psychological3 or physical4 stress, for example. There are hundreds of other in vitro, cell, laboratory animal and human ginseng publications. The publications are themselves controversial-as demonstrated by those that d05,6 or do not7,8 support an effect on physical stress in animals. Extensive though our knowledge of ginseng chemistry,9 pharmacology, 10 or pharmacokineticsil may appear, most ginseng specialists cannot say if (or how) ginseng affects man. I know of no proven case of ginseng having cured anyone. Evaluation of existing studies is not going to establish ginseng as a useful therapeutic agent. What we need now are long-term, controlled human studies of ginseng to identify predictable beneficial or harmful physiological effects. If such effects were to be found we might be better able to find out how ginseng works. Department of
Medicinal and Pharmacognosy,
College of Pharmacy, University of Minnesota, Minneapolis, Minnesota 55455, USA
E. JOHN STABA
1. Chen SE, Staba EJ. American ginseng I: Large scale isolation of ginsenosides from leaves and stems. J Nat Prod (Lloydia) 1978; 41: 361-66. 2. Yamamoto M, Uemura T, Nakama S, Uemiya M, Kumagai A. Serum HDL-cholesterol-increasing and fatty liver-improving actions of Panax ginseng in high cholesterol diet-fed rats with clinical effect on hyperlipidemia in man. Am J Chinese Med 1983; 11: 96-101. 3. Johnson A, Jiang NS, Staba EJ. Whole ginseng effects on human response to demands for performance. In: Proceedings ofthe 3rd international Ginseng Symposium (Sept 8-10, 1980). Seoul: Research Institute, Office of Monopoly, 1980: 237-44. 4. Knapik JJ, Wright JE, Welch MJ, et al. The influence of Panax ginseng on indices of substrate utilization during repeated, exhaustive exercise in man. Fed Proc 1983; 42: 336 (abstr). 5. Avakian EV, Sugimoto RB, Taguchi S, Horvath SM. Effect of Panax ginseng extract on energy metabolism during exercise in rats. Planta Med 1984; 48: 151-54. 6. JFL, Does Eleutherococcus improve mental and physical powers of Soviet athletes? Pediatrics 1981, 67: A46. 7. Lewis WH, Zenger VE, Lynch RG. No adaptogen response of mice to ginseng and Eleutherococcus. Ethanopharmacol J 1983; 8: 209-14. 8. Martinez B, Staba EJ. The physiological effects of Aralia, Panax, and Eleutherococcus on exercised rats. Japan J Pharmacol 1984; 35: 79-85. 9. Tanaka O, Kasai R. Saponins of ginseng and related plants. Progr Chem Organic Natural Products 1984; 46: 1-76. 10. Kaku T, Miyata T, Uruno T, Sako I, Kinoshita A. Chemicopharmacological studies on saponins of Panax ginseng CA Meyer II: Pharmacological part. Arzneimittel Forschung 1975; 25: 539-47. 11. Chen SE, Sawchuk RJ, Staba EJ. American ginseng III. Pharmacokinetics of ginsenosides in the rabbit. Europ J Drug Metab Pharmacokinetics 1980; 5: 161-68.
INTRAVENOUS IMMUNOGLOBULIN VERSUS ORAL CORTICOSTEROIDS IN ACUTE IMMUNE THROMBOCYTOPENIC PURPURA IN CHILDHOOD
SIR,-Dr Imbach and colleagues’ study (Aug 31, p 464) is flawed by not having an untreated group of patients to permit comparison of the results of observation alone with those of active treatment. If, as I suspect, the long-term and medium-term results would have been the same for the no treatment and treatment groups-the intravenous administration of corticosteroids or immunoglobulins could be considered as unwarranted interference because any treatment can be associated with complications while observation alone is not. I was surprised that this study was acceptable to an ethics committee. Imbach et al concede that more than 80% of children with immune thrombocytopenic purpura (ITP) have a spontaneous remission and that the death rate from cerebral haemorrhage is less than 1 %. For such a benign disease it is difficult to justify treatments that may have serious side-effects. , Imbach et al state that their study "shows that for acute ITP in childhood IgG can justifiably be used as first therapy". This trial shows nothing of the sort. What it does show is that infusion of high doses of IgG can accelerate the recovery of platelet counts as steroids do-but it does not answer the far more important question of whether any form of active interference is justified in the treatment of ITP in childhood. In my department, where neither steroids nor
been used in the treatment of acute ITP, spontaneous remission has been observed in over 80% of cases; there have been no deaths from ITP during the acute phase, and the frequency of chronic ITP (as defined by a platelet count of less than 30x 109/1) has been about 4%. Our patients have avoided the side-effects of steroids and the unpleasantness of daily intravenous infusions. Treatment has also been far less expensive.
Department of Clinical Haematology and Oncology, Royal Children’s Hospital,
Parkville, Victoria 3052, Australia
**This letter has been shown to Dr Imbach and colleagues, whose reply follows.-ED. L.
SIR,-A prospective, randomised studyl showed that corticosteroids, compared with placebo, significantly accelerated the initial rise in platelet count and reduced the critical time period during which most life-threatening haemorrhages occur. This was the main reason why we had no untreated group in our study. Infusions of
IgG not only accelerated the recovery of the platelet count as corticosteroids did but also induced a more rapid recovery than corticosteroids did in a subgroup of patients. The risk of life-threatening haemorrhage and the speed of recovery are relevant. If children with severe acute ITP are admitted to hospital, as they often are in Switzerland, the cost of therapy might well be compensated for by shorter admissions. Practical considerations apart, the effects of IgG infusions may provide a better understanding of pathogenic mechanisms involved in this and other diseases. Department of Paediatrics, University Hospital, 3010 Bern, Switzerland; and Institute for Clinical and Experimental Cancer
P. IMBACH S. BARANDUN H. P. WAGNER
University of Bern, Tiefenauspital, Bern
1. Sartorius JA. Steroid treatment of idiopathic thrombocytopenia purpura Am J Pediatr Hematol/Oncol 1984; 6: 165-69.
ANTIMALARIAL EFFECTS OF RIBOFLAVIN DEFICIENCY and colleagues (Nov 9, p 1040) suggest a novel malaria chemotherapy by interfering with riboflavin metabolism. They draw attention to our studies on the effect of riboflavin deficiency on rodent plasmodial but they overlook the important fact that, although riboflavin deficiency almost completely suppressed parasite development, most of the rats died. Furthermore, Seeler and Ott,2 who worked with less virulent, avian parasites, also reported increased mortality in riboflavin-deficient birds. Before suggesting that riboflavin deficiency may be beneficial against malaria the adverse effects of the deficiency must be
explained. I have suggested that these effects on parasite and host are not specific for riboflavin deficiency but are characteristic of a deficiency of nutrients with antioxidant properties.3Dutta et al are incorrect to.claim that the prevalence and severity of infection is usually made worse by malnutrition, and this is not true for malaria. Deficiencies of tocopherol4 and ascorbic acids5 (andp-aminobenzoic acid, thiamin, protein and even plain starvation) all depress parasite development in experimental hosts. Tocopherol and ascorbic acid are of particular interest in the light of the riboflavin work for all three deficiencies may involve
interaction with the
parasite/host system. The mechanism by which riboflavin deficiency.. suppresses malaria parasite development may well be linked to the coenzyme role of flavine-adenine dinucleotide in the regeneration of glutathione (GSH). The malaria parasite needs GSH to infect and grow within the red cell;and, as the substrate for glutathione peroxidase, GSH is also necessary to counteract the oxidant stress generated by the parasite,4to maintain membrane integrity and ensure parasite survival. Red cell GSH is reduced in riboflavindeficient rats but not in man.8 Tocopherol is also vitally important to protect polyunsaturated fatty acids in erythrocyte membranes against oxidants.4Ascorbic acid may play a part both in the