abortion, and stillbirth, may be increased. For these reasons, and because chloroquine has not been found to have any harmful effects on the fetus when used in the recommended doses for malaria prophylaxis, pregnancy is not a contraindication to malaria prophylaxis with chloroquine or hydroxychloroquine. However, because no chemoprophylactic regimen is completely effective in areas with CRPF, women who are pregnant or likely to become so should avoid travel to such areas.

The safety of FansidarR during pregnancy has not been completely established. Experimental data demonstrating the teratogenic effect of pyrimethamine in laboratory animals has resulted in restrictions in the licensing of compounds containing pyrimethamine. However, pyrimethamine, alone and in combination with sulfonamides, has been used for nearly 30 years to treat pregnant women with toxoplasmosis (another protozoal parasitic infection). While caution must be exercised when extrapolating from accumulated case reports of women treated for this infection, it is difficult to implicate pyrimethamine as a cause of fetal abnormalities. Thus, while the teratogenic effect in animals cannot be ignored, published data do not substantiate the inference that pyrimethamine is a human teratogen.

Sulfadoxine is a sulfonamide antimicrobial that, when administered during the last trimester of pregnancy, theoretically could compete with bilirubin for plasma proteins and exacerbate neonatal jaundice. It is unclear, however, whether this specific sulfa congener poses any risk to the newborn.

Doxycycline, a tetracycline, is generally contraindicated for malaria prophylaxis during pregnancy. Adverse effects of tetracyclines on the fetus include discoloration and severe dysplasia of the teeth and inhibition of bone growth. In pregnancy, therefore, tetracyclines would be indicated only if required to treat life-threatening infections due to multidrug-resistant P. falciparum.

Primaquine should not be used during pregnancy because the drug may be passed transplacentally to a G6PD-deficient fetus and cause life-threatening hemolytic anemia in utero. Whenever radical cure or terminal prophylaxis with primaquine is indicated, chloroquine should be given once a week until delivery, at which time the decision to give primaquine may be made.

Prophylaxis While Breastfeeding

Very small amounts of antimalarial drugs are secreted in the breast milk of lactating women. The amount of drug transferred is not thought to be harmful to the nursing infant; however, more information is needed. Because the quantity of antimalarials transferred in breast milk is insufficient to provide adequate protection against malaria, infants who require chemoprophylaxis should receive the recommended dosages of antimalarials (Table 1).

Chemoprophylaxis for Children

Children of any age can contract malaria. Consequently, the indications for prophylaxis are identical to those described for adults. Doxycycline is contraindicated for children less than 8 years of age, and Fansidar® is contraindicated for infants less than 2 months of age.

Chloroquine phosphate, which is manufactured in the United States in tablet form only, tastes quite bitter. Pediatric doses should be calculated carefully according to body weight. Pharmacists can pulverize tablets and prepare gelatin capsules with calculated pediatric doses. Mixing the powder in food or drink may facilitate the weekly administration of chloroquine to children. Alternatively, chloroquine in suspension is widely available overseas.

OVERDOSE OF ANTIMALARIAL DRUGS CAN BE FATAL. THE MEDICATION SHOULD BE STORED IN CHILDPROOF CONTAINERS OUT OF THE REACH OF CHILDREN.

Health Information for International Travel 1988 will soon be published by the Center for Prevention Services, CDC. This document includes the above recommendations for the prevention and presumptive treatment of malaria in travelers. In addition, it includes the chemoprophylactic regimen recommended for each country and the risk of malaria in each country. It will be a useful reference to health professionals, travel agencies, international businesses, and other agencies that advise international travelers concerning malaria and other health risks they may encounter when visiting foreign countries. This publication will be available from the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402, telephone (202)783-3238, as DHHS publication no. (CDC)88-8280.

Reported by: Malaria Br, Div of Parasitic Diseases, Center for Infectious Diseases; Div of Quarantine, Center for Prevention Svcs, CDC.

Reference

1. Centers for Disease Control. Agranulocytosis associated with the use of amodiaquine for malaria prophylaxis. MMWR 1986;35:165-6.

✩U.S. Government Printing Office: 1988-530-111/81500 Region IV

Over the past several years, CDC has received an increasing number of reports of hepatitis A outbreaks involving drug abusers. These outbreaks have occurred in many areas of the United States, including Alaska, Oregon, Washington State, northern and southern California, Oklahoma, upstate New York, and Connecticut. A variety of drugs have been used: in Oregon, northern California, and Oklahoma, intravenous (IV) amphetamines have been most commonly implicated; in one locality in southern California, a new form of heroin, referred to as "black tar" because of its color and consistency, has been linked with transmission; in upstate New York, IV cocaine has been the primary drug. In several areas, cases have occurred among people who only smoked marijuana. Outbreaks in upstate New York and northern California and data from the Viral Hepatitis Surveillance Program (VHSP)* are summarized below to illustrate this trend.

Upstate New York

Since December 1, 1986, hepatitis A outbreaks predominantly involving drug abusers have been reported in Monroe, Cortland, Onondaga, and Chemung counties in upstate New York. In Monroe County, 87 cases of physician-diagnosed hepatitis A were reported to local health authorities between December 1, 1986, and May 31, 1987. An average of nine cases had been reported for the same period of the previous 2 years. Twenty-four (28%) of these patients were IV drug abusers without other identifiable risk factors for hepatitis A. Eight additional patients were sexual or household contacts of these 24 patients. Information about the specific drugs used was not available for all patients; however, local drug enforcement officials believe that cocaine is the primary drug used intravenously in Monroe County.

Thirty-eight cases of hepatitis A occurred in Cortland County in 1987. Twenty-two (58%) of the patients were known or suspected drug abusers. Eleven of these cases occurred between March 25 and April 18. About 1 month before becoming ill, these 11 patients had attended two different social gatherings at which seven of them had used IV cocaine and shared needles. No food was consumed nor were beverages shared at these gatherings, and no other risk factors for hepatitis A could be identified.

*A nationwide reporting system in which patients serologically confirmed to have hepatitis A are interviewed to identify probable sources of illness.

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES/PUBLIC HEALTH SERVICE

Hepatitis A - Continued

Increases in hepatitis A also occurred in Onondaga and Chemung Counties during 1987. Fifty (38%) of the 131 cases of hepatitis A in Onondaga County in 1987 were among known or suspected drug abusers; 70% of the 50 patients used only IV cocaine. Thirteen cases occurred between October and December in Chemung County. Four (31%) of these patients were drug abusers; all used IV cocaine only. Anecdotal information suggested that drug abusers in Chemung County had recently obtained cocaine from persons in Onondaga County.

Northern California

In June 1987, an outbreak of hepatitis A among patrons of a restaurant in a northern California county was reported to local health authorities. Investigation revealed that a restaurant cook with a history of IV drug abuse had been diagnosed with hepatitis A several weeks earlier. A review of all serologically confirmed cases of hepatitis A in the county between January 1 and June 30, 1987, was subsequently conducted. Thirty (42%) of the 71 cases identified were associated with the foodborne outbreak originating at the restaurant (Figure 1). Thirty-three of the remaining 41 patients were contacted either directly or through friends or family. Twenty-four (73%) of these 33 patients were IV drug abusers and did not have other risk factors for hepatitis A. Eleven (46%) of the IV drug abusers were male; all were white; and they ranged from 21 to 39 years of age. Twelve of the drug abusers admitted to IV drug use within 6 weeks before onset of hepatitis A. All twelve had injected "crank" (an amphetamine derivative). Twelve of the IV drug abusers admitted to either casual or intimate contact or sharing needles or drug paraphernalia with at least one other IV drug abuser who contracted hepatitis A during the same period. Viral Hepatitis Surveillance Program

Data from VHSP indicate an increasing association between drug abuse and hepatitis A in the United States. Between 1982 and 1986, the percentage of persons with hepatitis A who admitted to previous IV drug use rose steadily from 4% to 19%. During this period, overall hepatitis A rates were relatively constant, and the proportion of patients with hepatitis A who had other identifiable risk factors remained stable. It should be noted, however, that only one-third of patients reported FIGURE 1. Cases of hepatitis A, by week of onset

northern California, 1987

to the MMWR Morbidity Surveillance System were interviewed through VHSP and that only a modest proportion of such persons are routinely asked about IV or other drug use.

Reported by: SA Jenkerson, MSM, JP Middaugh, MD, State Epidemiologist, Alaska Dept of Health and Social Svcs. SL Pittman, RS Hill, MD, Napa County Health Dept, Napa; AG Redeker, MD, Rancho Los Amigos Medical Center, Los Angeles; RR Roberto, MD, California Dept of Health Svcs. JL Hadler, MD, MPH, State Epidemiologist, Connecticut State Dept of Health Svcs. KM Bell, MD, Monroe County Dept of Health, Rochester; J Miller, MD, MPH, Onondaga County Health Dept, Syracuse; J Feuss, Cortland County Health Dept, Cortland; C Benjamin, Chemung County Health Dept, Elmira; S Kondracki, M Toly, DL Morse, MD, MS, State Epidemiologist, New York State Dept of Health. B Gildon, GR Istre, MD, State Epidemiologist, Oklahoma State Dept of Health. J Polder, RN, MPH, LR Foster, MD, MPH, State Epidemiologist, Oregon Dept of Human Resources. JM Kobayashi, MD, State Epidemiologist, Washington Dept of Social and Health Svcs. Div of Field Svcs, Epidemiology Program Office; Hepatitis Br, Div of Viral Diseases, Center for Infectious Diseases, CDC.

Editorial Note: In the United States, transmission of hepatitis A has traditionally been associated with crowding, poor personal hygiene, improper sanitation, and, less commonly, contamination of food or water. Recognized risk factors include intimate or close contact with persons with hepatitis A, foreign travel to developing countries, and contact with children in day-care centers.

The association of drug use and hepatitis A has been recognized only recently. Well-documented outbreaks of hepatitis A among drug abusers have been reported in Scandinavian countries (1,2). In seroprevalence studies of antibodies against hepatitis A virus (HAV) in Denmark, drug abusers have had antibody rates four times those of the general population (3).

Two possible explanations for the association between hepatitis A and drug use have been proposed: 1) HAV may be transmitted by injection or ingestion* of contaminated drugs (common-source spread), or 2) transmission may result from direct person-to-person contact. The culture of fecal coliforms from marijuana confiscated during one investigation (Alaska Department of Health and Social Services, unpublished data) raises the possibility that direct contamination of drugs could be a factor in some of these outbreaks. Drugs could become contaminated with fecal material containing HAV at the cultivation site (e.g., through use of human feces as fertilizer) or during transport, preparation, or distribution (e.g., through smuggling in condoms concealed in the rectum [4] or in baby diapers). However, the pattern of occurrence of the cases by dates of onset in each of the outbreaks and the diversity of drugs involved argue against a single common-source mode of transmission. Nevertheless, sustained common-source transmission is possible if contaminated drugs were distributed among persons who then used them at different times. Person-to-person transmission of HAV between drug abusers could result from sharing needles, from sexual contact, or from generally poor sanitary and personal hygiene conditions, which have often been observed among drug abusers. Isolated instances of bloodborne transmission resulting from transfusions from donors who had given blood during the incubation period of viral infection have been reported. Due to the relatively short viremic phase of HAV infection, however, bloodborne

*By tasting the drug to assess quality, for example.

'HAV could not, however, be isolated from the marijuana by tissue culture (CDC, unpublished data).