Salt intake and stomach cancer: some contrary evidence.

Considerable evidence has accumulated linking high salt intake to the risk of developing stomach cancer (1, 2). In addition to a geographic correlation of areas showing high salt intake with those having high stomach cancer mortality (3) and evidence incriminating preserved, often salty foods (4, 5), case-control studies in which individuabs have been asked about the routine addition of salt to food (4, 6), preference for salty food (7, 8), or salt consumption (9, 10) have often shown the association of higher salt intake with stomach cancer, in study groups in the United States and other countries. We recently completed a case-control study of stomach cancer based on persons who had had serum placed in frozen storage in 1964-1969 at multiphasic health checkups. This showed that prior infection with He!icobacter py!ori was associated with considerably increased risk (11). In the self-administered checkup questionnaire there was one question about salt intake: “Do you usually salt food before tasting it?” This permitted an attempt to replicate findings of other studies implicating the self-reported routine addition of salt. Partially allaying our skepticism about assessing salt intake by this single crude question, subjects of another study who answered yes showed an increased risk of developing hypertension (12). The study group consisted of 200 examinees randomly selected from 246 who subsequently developed stomach cancer by the end of 1988, according to cornputer-stored records obtained from the Kaiser Permanente hospitalization files and the San Francisco Bay Area Surveillance, Epidemiology, and End Results Registry (13) and 200 comparison subjects who did not develop cancer and who were individually matched to the cases for age (born in same 5-year interval, e.g., 1900-1904), date of examination (same month and year), sex, race, and location of examination (Oakland or San Francisco). Review of available pathology slides led to subclassification by histological type and location of the tumor (1 1). There was no evidence that routine salting led to increased risk. In fact, for all subjects with available data and for intestinal-type adenocarcinoma, risk was lower in the routine salters (Table 1). This finding persisted when salt use was entered into a logistic regression analysis along with He!icobacter py!ori infection as independent variables, using all subjects with serum (relative Table 1 Estimated relative risk and 95% confidence interval of developing stomach cancer according to whether or not subject reported usually salting food before tasting it