Mark Wefers Bettink

Interim analysis; Feasibility study of measurement of mitochondrial function in patients with sepsis 7 135 (negative control) was included. To prevent high Cq values (high Cq values means less accuracy) the DNA samples were used undiluted. For converting the copies/µl DNA sample to copies/µl plasma the following formula was used: c= Q x vF x Vdna / Vplasma where c = copies/µl plasma, Q = copies/µl DNA calculated by qPCR software, vF = dilution factor (undiluted = 1), Vdna = Volume of extracted DNA (final step DNA isolation) (200 µl ), Vplasma = Volume plasma used for DNA isolation (100 µl ) Statistical analysis Descriptive analyses were used to report on baseline characteristics of both groups. For non-normal distributed variables medians and interquartile range (IQR) are shown. For dichotomous or categorical data, percentages are shown. All statistical analyses and figures were performed using GraphPad Prism 7 (La Jolla, California, USA). RESULTS Patients were screened between October 2017 and September 2019. We started the informed consent procedure with 24 patients. Of these eligible patients due to refusal of consent by proxy two control patients and two sepsis patient were not measured. We were not able to obtain final informed consent of four control patients and one sepsis patient. In the control group one measurement of the COMET failed probably due to a problem in the preparation phase. The measurement after 5-7 days were conducted in six patients of the control group and for one patient in the sepsis group. During ICU admittance lactate measurements are 1.2 [0.7;1.4] mmol/liter in the control group and 1.5 [1.2;4.5] mmol/liter in the sepsis group. In our study groups 28 day mortality is 3/6 in the control group and 4/7 in the sepsis group. Baseline characteristics of all patients who underwent at least one measurement are described in Table 1.