Colonization of Infants and Hospitalized Patients with Clostridium difficile and Lactobacilli.

Clostridium difficile is the most frequent agent of nosocomial diarrhea in industrialized countries. Approximately 10%–25% of hospitalized patients become colonized by C. difficile [1]. Among nonhospitalized adults, the rate of carriage of C. difficile in the intestinal tract varies in different countries, ranging from 2% in Sweden to 15% in Japan [2]. Asymptomatic carriage of C. difficile is more common among infants and young children, ranging from 15% to 63% [2]. It is generally agreed that the disruption of the indigenous intestinal microflora is an essential prerequisite for colonization by C. difficile. The presence of lactobacilli as a component of the indigenous intestinal microflora has traditionally been associated with resistance to colonization with intestinal pathogens [3, 4]. However, it is not clear if the counts of lactobacilli differ significantly between persons who are colonized by C. difficile and those who are not. The aim of our study was to compare the prevalence and counts of C. difficile and intestinal lactobacilli in Estonian and Swedish children and to compare the counts of lactobacilli in hospitalized Estonian patients in relation to the presence of C. difficile. We investigated the fecal microflora of 27 healthy Estonian infants and 29 healthy Swedish infants (all 1 year of age) and 34 consecutive patients (66 fecal samples) in the neurological intensive care unit of Tartu University Hospital (Tartu, Estonia). The patients had been hospitalized because of neurological surgery or brain trauma for periods ranging from 2 days to 5 months (median duration, 11 days) before sampling. These patients had been treated with up to nine different antimicrobial agents (median number, four; some had not received any agents). The study was done during two 2-month periods separated by a 10-month interval. To detect C. difficile in these hospitalized patients, the fecal samples were seeded after alcohol shock into cefoxitin-cycloserine-fructose agar and incubated anaerobically for 4 days. In addition, the counts of C. difficile and lactobacilli in all samples from the infants and in 20 randomized samples from the hospitalized patients that were collected in the second period were determined as described previously [5]. Ten (35%) of 29 Swedish infants and only one (4%) of 27 Estonian infants were colonized with C. difficile (P õ .05; figure 1A). Furthermore, counts of C. difficile in feces were significantly higher in Swedish infants than in Estonian infants (Põ .05; figure 1B). However, both the prevalence and counts of lactobacilli were significantly higher among the Estonian infants than among the Swedish infants (P õ .005). None of the 21 hospitalized patients was colonized by C. difficile during the first 2-month observation period, whereas seven of 13 patients were found to be colonized during the second period. The Figure 1. A, Frequency of the occurrence of Clostridium difficile counts of C. difficile varied from 4.3 log cfu/g to 8.1 log cfu/g and lactobacilli in Estonian (h) and Swedish ( ) infants. B, Ranges (median count, 7.0 log cfu/g). The counts of lactobacilli were and medians (j) of counts of C. difficile and lactobacilli in Estonian (h) and Swedish ( ) infants.