Background <p>A review of the literature reveals that there is no descriptive epidemiological study based on the source of candidaemia in intensive care units(ICUs). In addition to that the number and scope of studies evaluating breakthrough candidaemia in ICUs are limited. The primary objective of this study was to analyse the epidemiology of candidaemia in our ICUs based on the source of each case, and the secondary aim was to evaluate breakthrough candidaemia.</p> Methods <p>141 adult patients(≥ 18 years) who had <i>Candida</i> spp. growth in their blood cultures 48&#xa0;h after admission to an ICU were included in the study. Statistical comparisons were made between groups for abdominal, central venous catheter and primary candidaemia. The level of significance was set at α = 0.05.</p> Results <p>In the classification based on the sources of candidaemia, 52 (37%) were abdominal, 32 (23%) were central venous catheter-related, 55 (39%) were primary candidaemia and 2 (1%) were urinary tract-related. Total parenteral nutrition use rates were significantly lower in primary candidaemia than in abdominal candidaemia (<i>p</i> &lt; 0.001). Steroid use rates were significantly higher in primary candidaemia than in abdominal candidaemia (<i>p</i> = 0.021). Mortality rates on days 14 and 28 did not differ significantly according to the source of infection. Beakthrough candidaemia rates were significantly higher in abdominal candidaemia. The acute physiology and chronic health evaluation II (APACHE-II) score at ICU admission was significantly higher in the breakthrough candidaemia(<i>p</i> = 0.037).</p> Conclusions <p>A significant proportion of candidemia cases in our intensive care units were non–catheter-related candidemia. Breakthrough candidaemia was found to be more common in patients with high APACHE-II scores at ICU admission and in candidaemia originating from the abdomen.</p>