Fig 1.
The perinatal asphyxia model of newborn piglets exposed to hypoxia-reoxygenation.
A total of n = 55 newborn piglets were randomized into three study arms: hypoxia-reoxygenation, hypoxia-reoxygenation and hypothermia, or sham-operated controls. Perinatal asphyxia was mimicked by addition of CO2 aiming at a PaCO2 of 8.0–9.5 kPa until a BE of -20 mmol/l and/or mean blood pressure fell below 20 mm Hg. Blood samples for cfDNA determination were taken at baseline, one hour after stabilization and prior to hypoxia, then repeated after hypoxia, and further at 30, 270, and 570 minutes of reoxygenation. The CSF samples were collected at post-intervention, 570 minutes after the end of hypoxia.
Table 1.
Novel designed primers for the qRT-PCR method.
Fig 2.
Changes in the concentrations of cfDNA in plasma of newborn piglets exposed to hypoxia, stratified by study arm.
The amount of cfDNA in blood plasma of newborn piglets exposed to hypoxia-reoxygenation was determined at baseline (prior to hypoxia), post-intervention (end of hypoxia), and at 30 minutes after the end of hypoxia. a. Mean cfDNA blood concentration and 95% confidence intervals stratified by study arm were determined using the fluorescence-based assay. The p-values from paired t-tests comparing baseline versus end of hypoxia values were p = 0.971 for controls, p = 0.009 for animals of the hypoxia group, and p = 0.193 for the hypoxia and hypothermia group, respectively. The p-values comparing baseline versus 30 minutes post-intervention were p = 0.245 for the control-, p = 0.032 for the hypoxia-, and, p = 0.131 for hypoxia and hypothermia group, respectively. There were statistically significant differences between the study arms comparing the difference from baseline to end of intervention (p = 0.038) and the difference from baseline to 30 minutes post-hypoxia (p = 0.025). b. Mean cfDNA blood concentration and 95% confidence intervals prior to hypoxia, at the end of hypoxia and 30 minutes after the end of intervention for the qRT-PCR method using the novel designed β-globulin primers, stratified by study arm. The p-values from paired t-tests comparing baseline versus end of hypoxia were p = 0.393 for the control-, p = 0.452 for the hypoxia-, and, p = 0.010 for the hypoxia and hypothermia group, respectively. The p-values comparing prior to intervention versus 30 minutes after the end of hypoxia were p = 0.421 for the sham-control animals group, p = 0.212 for the hypoxia group, and p = 0.071 for hypoxia and hypothermia group, respectively. There was no statistically significant difference between the study arms comparing the difference from baseline to end of hypoxia (p = 0.637) and the difference from baseline to 30 minutes after the end of intervention (p = 0.674). The cfDNA concentrations of additional 14 piglets were measured prior to surgery applying the fluorescence method, their mean blood concentration was 25.4 ng/ml (SD = 19.6; 95%CI 14.11–36.72).
Fig 3.
Predicted mixed model of temporal pattern of cfDNA in plasma of newborn piglets, stratified by study arm.
Concentrations of cfDNA were measured in newborn piglets exposed to hypoxia-reoxygenation, hypoxia-reoxygenation in combination with hypothermia, or in sham-operated control animals. The fluorescence method was applied for cfDNA determination. The mean cfDNA blood concentration and 95% confidence intervals from the fluorescence method for the different time points are illustrated and stratified by study arm. Mean blood concentration and 95% confidence intervals are predicted from an univariable linear mixed effects model with random intercept. There were no overall statistically significant effects of study arm (p = 0.838) and time (p = 0.570) on the cfDNA concentration, but there was a significant interaction between study arm and time (p = 0.018).
Fig 4.
Predicted mixed model of mean cfDNA blood concentration of newborn piglets exposed to hypoxia-reoxygenation measured by qRT-PCR and stratified by study arm.
The mean cfDNA blood concentration and 95% confidence intervals from the qRT-PCR method are illustrated for the different time points and stratified by study arm. Mean blood concentration and 95% confidence intervals are predicted from an univariable linear mixed effects model with random intercept. There were no overall statistically significant effects of study arm on the cfDNA concentration (p = 0.9143) or the interaction between study arm and time (p = 0.736). There was a significant effect of time on the cfDNA blood concentration (p = 0.022).
Table 2.
Characteristics and mean cfDNA concentrations in ng/ml at different time points of the study for the included piglets.
Fig 5.
ROC curves for the discrimination between hypoxia and control piglets at time point 30 minutes post-intervention by cfDNA plasma levels.
Receiver operating characteristic (ROC) curves are illustrated for a. cfDNA levels determined by fluorescence method (n = 43) and b. cfDNA levels measured using the qRT-PCR assay (n = 15) for the discrimination of piglets who did or did not undergo hypoxia intervention. The study arm of the group of piglets exposed to hypoxia and hypothermia was excluded from this analysis. The cfDNA by fluorescence method resulted in an area under the curve of 56%, the qRT-PCR method of 46%, respectively (p = 0.143).