Background: Selective decontamination of the digestive tract (SDD) is used in intensive care units to prevent nosocomial infections. The antibiotics used are considered safe as they are not absorbed from the gastrointestinal tract. However, tobramycin has been detected in the serum of critically ill patients receiving SDD. The aim of this study was to evaluate the prevalence of SDD-treated patients with tobramycin levels >1 mg/l. The secondary aim was to identify risk factors that may contribute to detectable tobramycin levels. Methods: Patients receiving SDD were asked to participate in the study. Blood was drawn on the day 3, 7, 10 and 14 after the start of SDD. Tobramycin was determined using a Roche immunoassay. Results: Six of the 45 patients (13%; 95% CI 3-23%) had detectable tobramycin levels, none of which had a level >1 mg/l. Patients with detectable tobramycin levels significantly more often had CVVH (p=0.028) and impaired renal function (p=0.043). Three models were evaluated with logistic regression. These models indicated that continuous venovenous haemofiltration and APACHE III (Chi2(2) =16.6, p=0.0002; Nagelkerke R2=0.567), impaired renal function and APACHE III (Chi2(2)=14.8, p=0.0006; Nagelkerke R2=0.517) and abdominal sepsis and APACHE III (Chi2(2)=23.1, p=0.000010; Nagelkerke R2=0.738) were predictors for detectable tobramycin levels. Conclusion Tobramycin was detected in 13% of the patients receiving SDD; none had levels of >1 mg/l. This study is an indication that absorption may be a factor to consider in relation to measurable tobramycin levels. © 2018, Netherlands Society of Intensive Care. All rights reserved.