SummaryThis study was carried on 88 hospital patients in whom aSalmonellawas isolated from stools (78 cases), from blood (28 cases), from urine (4 cases) or from other media (7 cases). The different groups ofSalmonellaefound were: A (1 case), B (52 cases), C (22 cases), D (12 cases, including 10 ofS. typhosa), E (1 case). As in other countries, we see thus in Belgium a decrease in the number of cases of typhoid fever, largely compensated by an important increase of otherSalmonellainfections.This group of 88 patients is not representative of the adult population of the country, and certaiidy does not reflect the exact incidence of salmonellosis in Brussels. The mean age of the group was 52,9 years. Generally, the admission to hospital was due either to social reasons or to seriousness of the disease. The proportion of patients free from any previous affection (cancers, digestive or cardiac diseases, diabetes, etc.) was only 40%.Gastrectomized patients seem particularly sensitive toSalmonella. They constituted 18 % of our series.The four common clinical forms of salmonellosis were encountered : 1) acute gastroenteritis (73 %), 2) septicemia (19 %)3) focal infections (2 %) and 4) asymptomatic form (6 % Usually, infections caused byS. typhosawere characterized by fever without severe digestive disorders, those caused bySalmonellaeof C group by a gastroenteritis without fever and those due to bacilli from A, B and D (exceptS. typhosa)groups by a febrile gastroenteritis.In most cases, and even in the septicemic forms, the blood leukocyte count and the sedimentation rate were only moderately increased, sometimes normal. In one third of the cases, blood urea was higher than 1 g/L.In the 10 cases of typhoid fever, chloramphenicol proved very effective. On the other hand, in the gastrointestinal form of salmonellosis, stool cultures did not become negative in 22 % of the cases, after administration of chloramphenicol. This observation confirms the poor efficacity of antibiotics in this form of salmonellosis.On the total of 88 patients, 13 died. This high rate of mortality seems to be due 1) to their great age, 2) to the high incidence of associated chronic diseases.The autopsy of the 13 deceased (1 typhoid and 12 paratyphoid) did not show the different lesions considered as characteristic of typhoid fever.Concluding this analysis of cases of salmonellosis observed in hospitals, two facts appear striking : the great number of gastrectomized subjects and the importance of the mortality in our series. The first observation should incite physicians to ask more often a stool culture for gastrectomized patients, suffering from diarrhea. The high mortality observed in this group of necessarily selected patients underlines that aSalmonellainfection, usually mild, can be the fatal complication of an associated chronic disease.The great importance taken by salmonellosis in our country should incite the public health authorites to organize an efficient service of detection and control of this type of infection.