Water-borne transmission of chloramphenicol-resistant salmonella typhi in Mexico (página 2)

TABLE-I NUMBERS OF CASES AND ATTAC ATES OF TYPHOID FEVER BY AGE AND SEX

! Male Female Total

Figures in italic show rate per 100 inhabitants in each age-group.

TABLE I1–ATTACK-RATES FOR SPECIFIC ITEMS OF FOOD AND DRINK

Ate nhabitants for females and 1-6 per 100 inhabitants for males,

Most patients lived in the north-east and east of the town, an area comprising approximately II third of the city, within which the attack-rate for inhabitants was 4-0 per 100 inhabitants as compared to 2-1 per 100 inhabitants for the entire town, Individual blocks had attack-rates which varied from 1 per 100 inhabitants to 16'8 per 100 inhabitants. Of the 83 cases analyzed, 43 (52 %) were clusters with multiple cases per household, and 40 were isolated cases.

During the investigation we learned that because of a defective pump the entire population had been without municipal water throughout May and between June 10 and June 28. During these periods the people had used water from the canal without boiling it or had obtained supplies from trucks bringing water from distant wells.

The canal itself was subsequently studied, having been divided for the purposes of investigation into three segments during its course through the village. The first segment corresponded to that part of the canal most upstream, the second to the middle section of the canal, and the third segment to that part farthest downstream. The banks of the first segment had extensive accumulations of human feces, lesser degrees of accumulation were noted in the second, and still less in the third segment However, in the latter portion of the second serpent, several! Pipes were discovered to be discharging raw sewage into the canal from private multifamily systems.

Analysis of questionnaires completed for both ill and well inhabitants revealed a differential attack-rate of 26-1 per 100 for those drinking water from the canal as compared to 6-8 per 100 for those obtaining water from other sources (table n). In contrast, differential attack-rates for water from other sources and a variety of food and beverages were not significantly different for those who were or were not exposed to these items. Most cases lived in the section of the community adjacent to the third segment of the canal (the segment farthest downstream).

Of the bacteriologica1 specimens co11ected, samples of pork, beef, cheese, and milk were a11 negative. An S. arizona species was isolated from a single specimen of chicharron (fried pig fat). Specimens of water from the canal a11 were heavily contaminated with coli forms; however, no attempts were made to isolate salmone11a from these specimens.

During the survey many of the people questioned reported gastrointestinal illness which roughly coincided with the period of the outbreak. Although no attempt was made to quantify the incidence of this illness, it was the impression of the investigators that there was a general increase in the incidence of gastrointestinal disturbances associated with the outbreak.

DISCUSSION

The Ciudad Cuauhtémoc typhoid out break is of interest as an instance of water-borne transmission of a chloramphenicol/resistant epidemic strain o S. typhi. We suggest that the water of the canal was contaminated by an individual (or individuals) infected with the resistant strain elsewhere in Mexico and that because of the lack of availability of municipal water the community was infected through the ingestion of canal water.

The high rates of attack in the inhabitants of the section of the village adjacent to the portion of the canal most downstream can be partly explained by greater likelihood of exposure. '1'his section is the most densely populated of the village, and the families with the lowest level of income and education live there. When the municipal water supply was interrupted, trucks brought water in. In the most crowded section, however, people had to wait for a long time to get water from the trucks and many decided to use water from the canal instead. The low general attack-rate was probably related to low concentrations of S. typhi in the canal water. The higher attack-rates in women may have be en related to greater likelihood of exposure because women were responsible for obtaining water for the family.

The high mortality-rate was probably related to the lack of medical care and the frequent use of chloramphenicol despite the known resistance of the organism to this agent. Unfortunately we do not know the specific treatment received by those who died.

The fact that all age-groups were affected and that the out break ended as soon as the water supply was re-established supports the hypothesis of the canal water as the vehicle of infection. Unfortunately S. typhi was not isolated from· the water nor was it possible to identify a person (or persons) who served as the source of contamination of the canal.

Although the epidemic was eased by a strain of S. typhi resistant to chloramphenicol, a small proportion of sensitive strains were al so found in outbreaks occurring in other parts of Mexico produced by this strain (Vi degraded approaching phage type A), approximately 10% of the isolates were sensitive to chloramphenicol, most of them belonging to phage types 26 and EJ. These sensitive strains were probably endemic in each community before the outbreaks.3, 8

We thank Dr Paul E. Pierce. and Dr Eugene J. Gangarosa, Center for Disease Control, Atlanta, Georgia, who made suggestions to improve the fieldwork and data interpretation; and Dr José Lopez Franchini, Zacatecas Sate Health Department, for this cooperation.

Requests for reprints should be addressed to A. G. C., Instituto de Salubridad y Enfermedades Tropicales, Carpio 470, México 17, México.

REFERENCES

1. González-Cortes, A., Heredia-Duarte, A., Guzmán ~ Baena, J. Salud Púb. Méx. (in the press).

2. González/Cortes, A., Besudo, D., Velez V. Ibíd. 1972, 14, 697.

3. Besudo, D., Olarte Jorge, Mendoza Hernández P., Galindo, E" Carrillo, l., Gutiérrez-Trujillo, G., Kumate., J. Bol. ofna. Sanit. Panam. 1973, 74, 1.

4. Bernard, R. P. J. Hyg., Camb. 1956, 63, 537.
5. Varela, G., Mendoza-Hernández, P., Vázquez, A., Revta. IInst. Salubr.

Enferm. Trop. Méx. 1956, 18, 33.

6. Gangarosa, J. B., Bennett, V. J., Wyatt, C., Pierce, P., Olarte, J, Mendoza-Hernández, P., Vázquez, V., Bezudo, M. D. J. infec. Dis. 1972, 128,215. 8

7. Anderson, E. S., Smith, H. R. Br. Med. J. 1972, iii, 329.

8. Center for Disease Control Weekly Reports 1972. Morbidity and Mortality, pp. 21, 177, 193,213,239,290, and 327.

PRINTED IN GREAT BRITAIN

The Lancet Office

7, Adam Street, Adelphi, London, W.C.2.

Abel Gonzalez

Cortes Rafael Sanchez

Leyva Marina Hinojosa

Enviado por:

Rafael Fragoso