Preliminary clinical trials with praziquantel in Schistosoma japonicum infections in the Philippines

Bulledn of the World Health Organization, 57 (5): 793-799 (1979) Preliminary clinical trials with praziquantel in Schistosoma japonicum infections in the Philippines A. T. SANTOS,1 B. L. BLAS,2 J. S. NOSENAS,3 G. P. PORTILLO,3 O. M. ORTEGA,4 M. HAYASHI,5 & K. BOEHME6 Praziquantel, a new antischistosomal compound, was tested for tolerance and efficacy against placebo in two double-blind clinical trials in Philippine patients infected with Schistosoma japonicum. The compound was given orally at a dose of 3 x20 mg/kg at intervals of 4 hours to a total of 82 patients-some without advanced disease and some with hepatosplenic involvement. A total of 43 patients received placebo. In a single-blind trial, 42 patients were given a single oral dose of 50 mg/kg. Monitoring of vital organ functions included comprehensive laboratory tests and serial electrocardiograms. In 38 patients with hepato- splenic involvement due to advanced stages of infection, serial electroencephalograms were additionally recorded. No toxic effects were observed in any of these examinations. Undesirable side effects occurred in 53% of the patients given 3 x20 mg/kg and in 70% after a dose of 1 x 50 mg/kg. They consisted mainly of abdominal discomfort, fever, sweating, and occasionally giddiness, but in general were transient and mild. At 6 months post-treatment, 60 of 75 patients treated with 3 x20 mg/kg and 29 of 41 treated with 1 x 50 mg/kg were completely negative for eggs. At 12 months post-treatment, 25 of 33 and 14 of26 patients in the two treatment groups were cured. Thus the divided dosage gave a superior therapeutic result. Praziquantel proved to be free ofmajor toxicity, and was well tolerated, highly effective, and easy to administer. Confirmation of results in extended trials may soon permit large-scale treatment. The initial clinical trials of praziquantel in the Philippines were planned to assess the tolerance and efficacy of the drug given as a one-day oral treat- ment against Schistosoma japonicum infections in Philippine patients. The trials formed part of a multicentre series coordinated by the World Health Organization. A common trial design was used and case documentation sheets were standardized as far as the species of infecting parasite and local condi- tions permitted (1). I Executive Director, Schistosomiasis Control and Research Project, Palo, Leyte (SCRPPL) Department of Health, Manila, Philippines. 2 Deputy Project Director, SCRPPL. 3 Medical Officer, SCRPPL. 4 Bethany Hospital, Tacloban City. 5 Director, Department of Neuropsychiatry, Kofu City Hospital, Yamanashi, Japan. 6 Bayer AG, Pharma Research Centre, Department of Medical Biometry, Wuppertal, Federal Republic of Germany. MATERIALS AND METHODS Initially, individuals living in highly endemic areas within a 20 km radius of the Schistosomiasis Control and Research Project, Palo, Leyte, were screened parasitologically and infected persons were referred to the outpatient department of the Research Pro- ject Laboratory. Entrants to these trials were selected from among these persons according to general and specific criteria (1). Patients selected for the trial were informed of the nature of the trials to be carried out with the drug and therefore of the necessity for hospitalization during treatment and for subsequent parasitological follow-up examinations for one year. Before treat- ment, written consent was obtained from patients of legal consenting age and also from the parents or guardians of minors. Two therapeutic studies were conducted consecu- tively; although they followed a similar trial design, 3856 793 794 A.T. SANTOS ET AL. they differed in the characteristics of the patients entering each trial. For an early appreciation of possible drug-related effects in the different stages of infection, the second trial was performed in patients with signs of hepato- splenic involvement. Special attention to this condi- tion appeared essential since it was known that another schistosomicide, niridazole, may cause se- vere adverse reactions in such patients, as a result of unchanged drug by-passing the liver after absorption from the gastrointestinal tract and reaching the central nervous system. The findings on drug toler- ance are therefore described by presenting the results of each trial in a synoptic way; patients without advanced disease are denoted as "A-pa- tients" and those with hepatosplenic involvement as " B-patients ". Case collection Only patients with a geometric mean egg count per gram of faeces (EPG) of at least 100 for A- patients and 50-100 for B-patients were selected. Egg counts were made from two aliquots of two consecutive stool samples using the Kato-Katz technique (2). Egg viability was confirmed by a positive hatching test in all patients. Patients of both sexes who qualified parasitologi- cally for the trial were then transferred to hospital for full physical examination including an electrocar- diogram for all patients and an electroencephalo- gram for B-patients only. I Electroencephalograms (EEGs) were recorded with a Nihon-Kohden 13-channel electroencephalo- graph before medication and 24 hours after the first dose. If alterations were seen or if readings were difficult to assess at that time, EEGs were repeated 7-8 days later and/or 2-4 weeks after treatment. A- patients stayed in hospital for 3 days, B-patients for 10 days. Children under the age of 6 years were excluded from the trial as were adults who had a personal or family history of CNS disease, mental abnormality, drug or alcohol abuse, had received antischistosomal treatment with any other drug during the last 6 months, or were pregnant or lactating. The patients finally selected for inclusion in the trial were allocated at random to one of three treatment groups: praziquantel at a dosage of 3 x a Details of the biochemical and haematological monitoring tests have been described in accompanying papers (1, 3, 4, 5). 20 mg/kg given at 4-hour intervals, placebo given similarly, or praziquantel at a single oral dose of 50 mg/kg. Assessment of efficacy The degree of infection was determined by two consecutive faecal examinations, with egg counts on two aliquots from each sample. Egg viability was confirmed by the Faust-Meleney hatching technique (6). For assessment of drug efficacy, two aliquots of three consecutive daily stools were examined by the Kato-Katz technique for S. japonicum eggs 1,2, 3,6, and 12 months after treatment in A-patients and 1, 2, 3, and 6 months after treatment in B-patients. Hatching tests were also performed. Characteristics of patients For each of the two classes of patient, age, weight, and height were examined for the three dosage groups by variance analysis, and the comparability of the dosage groups within each class could thus be established (Table 1). Further details of the analyti- cal programmes used and the results obtained are available on request.b Of the 30 A-patients initially given placebo, 26 were treated with praziquantel at 3 X 20 mg/kg after the 6-month follow-up examination and all of the 13 B-patients were so treated after the last EEG examination. Investigations showed that 86 (95.5 %) of the 90 A-patients and 25 (65.8%) of the 38 B-patients were infected by one or more of the following intestinal parasites: Ascaris lumbricoides, A-59/86 (68.6%), B-15/25 (60.0%); Trichuris trichiura, A-81/86 (94.2%), B-24/25 (96.0%); and hook- worm, A-20/86 (23.2%), B-25/25 (100%). 15 patients (13.5 %) had triple infections. Clinically, hepatomegaly was seen in 32 (35.5 %) of 90 A-patients, splenomegaly in 7 (7.8%), and anaemia in 68 (75.5 %). These signs were present in every B-patient, but in various degrees. Very severe hepatomegaly was present in 19 (50.0%) of 38 patients, severe splenomegaly in 20 (52.6%), and anaemia in 18 (47.4%). For the clinical classifica- tion of these symptoms, severity was graded 1-4. The degrees 3 and 4 were regarded as very severe. 'From Dr Kathrin Boehme, Bayer AG, Pharma Research Centre, P.O. Box 101709, 5600 Wuppertal 1, Federal Republic of Germany. PRAZIQUANTEL IN S. JAPONICUM INFECTONS IN THE PHILIPPINES Table 1. Pretreatment variables in drug- and placebo-treated patients MeanMean Proportion Group geometr-icPatient group No. Mean bMoedayn Mean Op mean of individual and dose treated age wheight Of geometric mean EPGs;(years ± SD) (kg ± SD) (cm SD) maes and standard factor' Group A 3x20 mg/kg 30 21 ± 9 41 ± 11 148 ± 15 26/30 326; 2.07 placebo, 3 dosea 30 24 ± 13 42 ± 10 149 ± 12 23/30 288; 2.12 1x50 mg/kg 30 26 ± 13 41 ± 8 151 ± 11 22/30 505; 2.31 Group B 3x20 mg/kg 13 15 ± 6 27 ± 9 131 ± 12 11/13 137; 2.25 placebo, 3 doses 13 15 ± 7 28 ± 9 130 11 13/13 103; 1.74 1x50 mg/kg 12 14 ± 2 26 ± 4 131 9 12/12 242; 3.09 'EPG = no. of eggs per gram of stool; "standard factor" is the same mathematical quantity for geometric means as is the " standard deviation" for arithmetic means. RESULTS Table 2. Overall frequency of side effects in patients infected with S. japonicum and treated with praziquan- tel or placebo Tolerance: adverse reactions Regimen The overall frequency of unwanted side effects is Group ofpatients praziquantel, praziquantel, placebo, 3 doses shown in Table 2, and the individual frequencies and 3X20 mg/kg 1 x50 mg/kg types of adverse reaction within dosage groups are shown in Table 3. A comparison of the overall A-patients 16/30 (53.3%) 21/30 (70.0%) 4/30 (13.3%) frequencies of side effects after treatment with those B-patieonts(hepatosplenic 8/13 (61.5%) 7/12 (58.3%) 1/13 (7.7%) occurring after placebo indicates clearly the relation- involvement ship of the side effects to treatment. Yet a careful Table 3. Actual frequencies of main symptoms present before or first seen after treatment with praziquantel or placebo A-patients B-patients praziquantel praziquantel placebo, praziquantel praziquantel placebo, Symptom 3x20 mg/kg 1 x 50 mg/kg 3 doses 3x20 mg/kg 1 x 50 mg/kg 3 doses(n= 30) (n= 30) (n= 30) (n= 13) (n= 12) (n= 13) before after before after before after before after before after before after abdominal discomfortd 52 9 54 14 60 1 27 5 23 5 28 0 recurrent diarrhoea 12 0 10 0 11 0 7 0 9 0 11 0 fever 14 5 11 9 11 4 8 3 8 2 6 0 giddiness, drowsiness, dizziness 19 2 30 2 23 1 16 0 11 0 7 0 headache 21 0 19 1 19 0 6 1 4 0 9 1 lassitude, tiredness 10 0 15 1 17 1 10 0 8 0 8 0 sweating 13 1 9 7 9 0 8 2 2 1 9 0 urticaria 0 0 0 0 0 0 1 0 0 1 0 0 'Abdominal discomfort includes the following symptoms, which often appeared simultaneously: abdominal pain, nausea, vomiting, anorexia, and dyspepsia. 795 796 A.T. SANTOS ET AL. enquiry revealed that many symptoms had existed before treatment and therefore could possibly have been caused by the infection (Table 3). It was often difficult to decide whether an observed reaction was drug-related. Abdominal discomfort, fever, sweat- ing, and dizziness or drowsiness were the symptoms that occurred most frequently for the first time after treatment. It is interesting to note that where abdominal discomfort, and especially abdominal pain, was not present before medication, it was characterized by sudden onset usually beginning a few minutes to 1 /2 hours after intake of only the first dose of the drug. Occasionally, pain was described as crampy or colicky. In patients treated with 3X 20 mg/kg, 1 (1.8%) of 56 A-cases and 2 (7.7%) of 26 B-cases developed severe abdominal pain after medication; all had complained of the same pain before treat- ment, but of less severity. In the groups given 1 x 50 mg/kg, this adverse reaction was observed in 3 (10%) of 30 A-patients and 1 (8.3%) of 12 B- patients. Rapid relief was obtained after administra- tion of an antispasmodic. Additional medication was necessary also for patients in whom high fever occurred. An antipyretic was given to 1 (3.9%) of 26 B-patients receiving 3 x 20 mg/kg as well as to 2 (6.7%) of 30 A-patients and 2 (16.7%) of 12 B- patients receiving 1 x 50 mg/kg. In 3 of these 5 cases, however, fever had been noticed before medication. In general, all complaints and adverse reactions were of mild to moderate intensity, of short duration, and did not require additional treat- ment. Tolerance: electrocardiographic findings Electrocardiograms before and 24 hours after medication were recorded in 85 patients treated with praziquantel. Comparison of pre- and post- treatment readings did not reveal any significant changes, and no signs of cardiotoxicity were found. Tolerance: electroencephalographic findings EEGs were recorded before medication in all 38 B-patients, 24 hours after medication in those re- ceiving a single dose, and 24 hours after the first dose in those receiving three doses; EEGs were repeated if indicated. Three patients treated with 3 x 20 mg/kg and 2 patients given placebo had slight alterations in the EEG 24 hours after intake of the first dose. Of these, 2 of the drug-treated patients and 1 of the placebo cases had normal EEG findings one week later. It remains debatable whether these obser- vations are causally related to praziquantel treat- ment. Detailed evaluation of all EEGs recorded before and after treatment indicated that praziquan- tel treatment in patients with marked hepatospleno- megaly did not produce any major adverse reactions in the CNS. Tolerance: laboratory findings For biometrical assessment of laboratory findings, it had to be clarified first whether the distribution of variables within each group was normal.c Since this was not the case, all values of the following tests were logarithmically transformed for both A- and B- patients: total serum bilirubin; leukocytes; eo- sinophils, serum alanine aminotransferase (SGPT); serum aspartate aminotransferase (SGOT); fasting blood sugar (A-patients only). The statistical anal- yses of the changes after treatment are shown in Table 4 for both A- and B-patients. In the A- patients, significant differences between drug- and placebo-treated patients were found only in the post-treatment values of neutrophils and lympho- cytes. These findings may reflect the impact of the drug on the parasites. Increases in mean SGPT and/or SGOT values after treatment in A-patients occurred in all groups, whether given drug or placebo but were without clinical relevance. In B-patients, increases in SGPT values were seen after all dosage regimes and also in the placebo group. A larger rise in mean post-treatment SGOT values after a dosage of 3 x 20 mg/kg also appeared to be without clinical significance. Clinical efficacy Improvement of impaired health generally began 1 week after treatment. Medical examinations were carried out at the same time as the parasitological follow-ups, 1, 2, 3, 4, and approximately 6 and 12 months after treatment, and relief of pretreatment complaints was confirmed in all patients treated with praziquantel. Complete symptomatic relief was ob- tained in 86.7% of patients treated with 3 x 20mg/kg and in 80% of those treated with 1 x 50 mg/kg. No patient in either dosage group reported or showed any exacerbation of a pretreatment complaint. cFurther details are available on request from Dr Kathrin Boehme, Bayer AG, Pharma Research Centre, P.O. Box 101709, 5600 Wuppertal 1, Federal Republic of Germany. PRAZIQUANTEL IN S. JAPONICUM INFECIlONS IN THE PHILIPPINES Table 4. Differences between pre- and post-treatment means of laboratory findings in A- and B-patients, separately praziquantel praziquantel placebo P .3x20 mg/kg lx 5Omg/kg A-patients (n = 30) (n = 30) (n = 30) total serum bilirubin' +8%b +14%b - 1% 0.36 SGPT' +27%b + 6%b + 9% 0.28 SGOT' + 4%b +19%b +11% 0.35 fasting blood sugar' + 4%b + 2%b - 1% 0.57 haemoglobinI (g/I) 0 - 1.0 - 4.0 0.21 leukocytes' -13% - 9% - 5% 0.58 eosinophilsa -13% -18% - 5% 0.63 neutrophils' (%) + 8.2 + 6.4 - 2.5 0.003 lymphocytes' (%) - 7.0 - 5.6 + 2.9 0.0002 B-pationts (n = 13) (n = 12) (n = 13) total serum bilirubin' + 5% +20% + 1% 0.55 SGPT' +37% +22% +13% 0.67 SGOT' +39% +13% -12% 0.13 haemoglobinI (g/i) + 4.0 - 3.0 - 3.0 0.13 leukocytes' -13% -10% -17% 0.97 eosinophils' -15% - 4% - 9% 0.96 neutrophils' (%) + 4.4 + 6.1 + 3.8 0.77 lymphocytesI (%) - 5.3 - 3.3 - 0.3 0.47 ' Geometric mean. b No information was available for 1 patient before or after treatment. CArithmetic mean. Among patients given placebo, however, complete relief was not noted in any case. Partial symptomatic relief was seen in 16.7 % and a worsened health status in 40 %. Parasitological efficacy Assessment of parasitological efficacy was based on repeated quantitative faecal egg counts using the Kato-Katz technique, by the glycerin-concentration method, and qualitatively by microscopic examina- tions and hatching tests for egg viability. Thus for patients with all examinations negative 6 months after treatment, a cure was assumed to be highly probable and those with persistently negative results to 12 months after treatment were regarded as cured. The parasitological observations made for A- and B-patients are summarized in Table 5. High cure rates were seen after both dose regimes. Overall, 60 (80%) of 75 patients given 3 X 20 mg/kg were completely egg-negative at 6 months and 25 (76%) of 33 were egg-negative at 12 months after treatment. After a single dose of 50 mg/kg, 29 (71 %) of 41 had no eggs in the excreta at 6 months and 14 (54%) of 26 had none at 12 months. None of the patients treated with placebo were egg-negative at 6 months. The mean reduction in egg excretion at this time in placebo-treated patients apparently reflects the biological fluctuations over time in untreated infections. Particularly noteworthy was the high degree of reduction of egg output in those treated patients in whom egg excretion persisted; 96% at 6 months and 95% at 12 months in those given 3 x 20 mg/kg. Although curative in schistosomiasis, praziquantel was found to be without effect against Ascaris, Trichuris, or hookworm species. 797 A.T. SANTOS ET AL. Table 5. Results of parasitological follow-up examinations in patients infected with S. japonicum treated with praziquantel or placebo After approximately 6 months After approximately 12 months Doe No. ofDrug (mg/kg) patients No. of Percentage Overall per- Patients Percentage Overall per-treated patients probably centage egg examined probably centage egg examined cured' reduction rate cured" reduction rate A-patients praziquantel 3X20 56b 50 84.0 96.5 33b 75.8 95.0 praziquantel 1 x 50 30 29 65.5 93.0 26 53.8 58.0 placebo 30 24 0.0 42.4 treated with praziquantel after 6 months B-patients praziquantel 3X20 260 25 72.0 89.7 praziquantel 1 X50 12 12 83.3 97.1 placebod 13 - - - ' For definition of cure see text. b Including 26 patients first given placebo as control group and treated after 6 months. Including 13 patients first given placebo as control group and treated after the last EEG examination. d Parasitological follow-up was not applicable. ' Parasitological follow-up for more than 6 months was not scheduled for B-patients. DISCUSSION Since all drugs previously available for the treat- ment of S. japonicum infections in Philippine pa- tients either caused frequent severe adverse reac- tions or necessitated prolonged treatment, they were unsuitable for large-scale treatment. Any new compound effective against S. japonicum must be investigated with great care in order to test its usefulness for large-scale treatment. Thus, special attention was paid to the testing of the tolerance of praziquantel since experience has shown that infec- tion with S. japonicum more often develops into severe disease than may be the case with schis- tosomiasis due to S. haematobium or S. mansoni. Conseqently, tolerance was first tested intensively in Japanese patients with only light chronic infections with S. japonicum, and the results obtained were encouraging (5). Adverse reactions were seen more frequently and their intensity was of a higher order when Philippine patients with a geometric mean EPG value of at least 100 were treated. While fever was not observed in any of the Japanese patients, it was present before treatment in 58 (54.3 %) of 128 Philippine patients and appeared for the first time after treatment in an additional 23 (17.9%) pa- tients. Analysis did not reveal any relationship between its occurrence and faecal egg output, body weight, or other variables. The different types and degrees of adverse reaction observed in Philippine and Japanese patients may be attributable more to the degree and stage of infection than to ethnic differences in the two population samples. Because of the known frequency of cerebral involvement and portal-systemic circulatory anas- tomoses in S. japonicum infections, the frequency of adverse reactions was compared in patients with and without hepatosplenic involvement, and in those with advanced stages of infection, neurological func- tion was carefully monitored by electroence- phalography before and shortly after praziquantel treatment. There was no difference in the frequency of adverse reactions and no evidence of toxicity to the central nervous system was found. Both observa- tions suggest that the compound will be suitable for use in the field in patients at all stages of infection. Confirmation of these results in extended trials is highly recommended, in particular to determine the optimum dose. Since the logistic aspects of chemotherapeutic control campaigns against schistosomiasis are of particular importance in developing countries, the oral treatment of S. japonicum infection on a single day is a very important advance. Further trials are, however, essential to investigate the minimum number of doses necessary to provide both max- imum safety and antischistosomal efficacy. 798 PRAZIQUANTEL IN S. JAPONICUM INFECTIONS IN THE PHILIPPIS 799 ACKNOWLEDGEMENTS We are grateful to Dr Julio Dolorico, Director of the Bethany Hospital in Tacloban City, to his assistant Dr Romualdo Cabalona, and to all other members of the hospital staff for their kind and very effective cooperation. We further thank the technicians Mrs Adelaida J. Flores and Mr Felix Luangco, as well as the drivers Mr Avelino Rojas, Mr Felimon Villas, Mr Bonifacio Rosales, and Mr Jorge Ilagan for their untiring services without which it would not have been possible to successfully terminate these trials. We are also indebted to Mr Domingo Bernaba and Mrs Lili Bautista for their clerical support and to Dr Tirso C. Banzon for his kind assistance in drafting this manuscript. We also wish to express our appreciation and thanks to Dr Dietrich H. G. Wegner, Bayer AG, Federal Republic of Germany for his valuable suggestions and advice concerning the trial design and for supplying the drug. RtSUMI ESSAIS CLINIQUES PRItLIMINAIRES DU PRAZIQUANTEL DANS LE TRArTEMENT DE L'INFECTION A s. JAPONICUM AUX PHILIPPINES Le praziquantel, nouveau compose schistosomicide ad- ministr6 par voie orale, a fait l'objet d'essais de tolerance et d'efficacit6 par rapport a un placebo lors de deux essais cliniques en double insu realis6s chez des Philippins souf- frant d'une infection a Schistosoma japonicum. Au total 128 malades ont particip6 a ces essais: 82 ont recu 3 doses de 20 mg/kg a intervalle de quatre heures, 42 ont recu une dose orale unique de 50 mg/kg et 43 ont d'abord recu un placebo puis, a l'exception de 4 d'entre eux qui n'ont pas pu suivre de traitement ult6rieur, ont ete places, apres avoir ainsi servi de temoins, dans le'groupe soumis a 3 x 20 mg/kg (39 malades). Chez 38 malades, on observait une atteinte hepatosplenique due a l'etat avance de l'infection. La surveillance des fonctions vitales n'a r6vele aucun effet toxique sur le systeme cardiovasculaire (examine par ECG en serie), sur le systeme nerveux central (examine a plusieurs reprises par 6lectroencephalographie dans tous les cas d'infection avanc6e), et sur les systemes hemato- poietique, h6patique et renal. Les effets secondaires indesirables etaient plus fr6quents et plus prononces a la dose d'1 X 50 mg/kg (21 cas sur 30, 70,0%); qu'a la dose de 3 x 20 mg/kg (16 cas sur 30, 53,0%). Ils consistaient principalement en gene abdomi- nale, fievre, sudation et parfois vertiges. En ge'neral, ces symptomes etaient moder6s, passagers, et n'exigeaient pas de traitement supplementaire. L'efficacite antischistosomique a 6te v6rifiee au bout de 1, 2, 3, 4 et environ 6 et 12 mois apres le traitement. Six mois apres le traitement, I'efficacite a ete verifiee chez a) 75 malades ayant requ 3 x 20 mg/kg, b) 41 malades ayant requ 1 x 50 mg/kg, et c) 24 malades ayant requ un placebo. A ce moment, les aeufs avaient disparu chez 80,0% et 70,7% respectivement des malades ayant requ le compose, lesquels ont ete consideres comme probablement gueris. Aucune guerison n'a ete observee chez les malades ayant requ le placebo. Douze mois apres le traitement, 1'efficacite a e verifi6e chez a) 33 malades ayant requ 3 x 20 mg/kg dont 75,8% ont ete gueris, etchez b) 26maladesayantrequ 1 x 50mg/kg, dont 53,8% ont e gu6ris. Le taux global de r6duction du nombre d'aeufs etait, respectivement, de 95% et 58%. La dose fractionnee a donc donne de meilleurs r6sultats. A l'heure actuelle, aucun autre compose utilise pour le traitement des infections a S. japonicum ne s'est r6vele a la fois si d6pourvu de toxicite, si bien toler6, si efficace et d'administration aussi aisee. Si les resultats decrits peuvent etre confirm6s par des essais elargis, la possibilite de traitement a grande echelle des malades souffrant d'infec- tion a S. japonicum semble pouvoir se realiser dans un proche avenir. REFERENCES 1. DAVIS, A. & WEGNER, D. H. G. Bulletin of the World Health Organization, 57: 767-771 (1979). 2. KATZ, N. ET AL. Revista do Instituto de Medicina Tropical de Sao Paulo, 14: 397-400 (1972). 3. DAVIS, A. ET AL. Bulletin ofthe World Health Organiza- tion, 57: 773-779 (1979). 4. KATZ, N. ET AL. Bulletin of the World Health Organiza- tion, 57: 781-785 (1979). 5. IsHIzAKi, T. ET AL. Bulletin of the World Health Organi- zation, 57: 787-791 (1979). 6. FAUST, E. C. & MELENEY, H. E. American journal of hygiene, Monograph series, No. 3, pp. 1-339 (1924).