Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

Background: The COVID-19 pandemic continues to overwhelm intensive care units (ICUs) worldwide, and improved prediction of mortality among COVID-19 patients could assist decision making in the ICU setting. In this work, we report on the development and validation of a dynamic mortality model specifically for critically ill COVID-19 patients and discuss its potential utility in the ICU. Methods: We collected electronic medical record (EMR) data from 3222 ICU admissions with a COVID-19 infection from 25 different ICUs in the Netherlands. We extracted daily observations of each patient and fitted both a linear (logistic regression) and non-linear (random forest) model to predict mortality within 24 h from the moment of prediction. Isotonic regression was used to re-calibrate the predictions of the fitted models. We evaluated the models in a leave-one-ICU-out (LOIO) cross-validation procedure. Results: The logistic regression and random forest model yielded an area under the receiver operating characteristic curve of 0.87 [0.85; 0.88] and 0.86 [0.84; 0.88], respectively. The recalibrated model predictions showed a calibration intercept of -0.04 [-0.12; 0.04] and slope of 0.90 [0.85; 0.95] for logistic regression model and a calibration intercept of -0.19 [-0.27; -0.10] and slope of 0.89 [0.84; 0.94] for the random forest model. Discussion: We presented a model for dynamic mortality prediction, specifically for critically ill COVID-19 patients, which predicts near-term mortality rather than in-ICU mortality. The potential clinical utility of dynamic mortality models such as benchmarking, improving resource allocation and informing family members, as well as the development of models with more causal structure, should be topics for future research.

Epigenetic and genotype-specific effects on the stability of de novo imposed methylation patterns in transgenic mice.

The chloramphenicol acetyltransferase gene under the control of the late E2A promoter of adenovirus type 2 (Ad2) was introduced as transgene into the B6D2F1 mouse strain with mixed genetic background and became extensively de novo methylated. The methylation of this pAd2E2AL-CAT (7-1A) transgene was regulated in a strain-specific manner apparently depending on the site of integration. Transmission of the 7-1A transgene into an inbred DBA/2, 129/sv, or FVB/N genetic background led to a significant loss of methylation in the transgene, whereas C57BL/6, CB20, and Balb/c backgrounds favored the de novo methylation in very specific patterns. The newly established patterns of de novo methylation were transmitted to the offspring and remained stable for many generations, regardless of the heterozygosity of strain-specific DNA sequences present in these mouse strains. Segregation analyses showed a non-mendelian transmission of methylation phenotypes and suggested the involvement of dominant modifiers of methylation. The genotype-specific modifications of the transgene were followed for 11 backcross generations. These observations reflect an evolutionarily conserved mechanism directed against foreign, e.g. viral or bacterial, DNA at least in the chromosomal location of the 7-1A transgene. In seven additional mouse lines carrying the same transgene in different chromosomal locations, strain-specific alterations of methylation patterns were not observed.

Stability of transgene methylation patterns in mice: position effects, strain specificity and cellular mosaicism.

The methylation status of a transgene, which carried the adenovirus type 2 E2A late promoter linked to the chloramphenicol acetyltransferase gene, was studied in three transgenic mouse lines (5-8, 7-1 and 8-1). These lines were analysed over a large number of offspring generations beyond the founder animal. In mating experiments, the influence of the parent-of-origin and strain-specific backgrounds on the transgene methylation patterns were assessed and found to have no effect on the pre-established methylation patterns in mouse lines 5-8 and 8-1. The founder animal 7-1 carried two groups of a total of ten transgenes, which were located on two different chromosomes. These arrays of transgenes could be segregated into separate mouse lines 7-1A and 7-1B. The transgenes of 7-1A animals exhibited cellular mosaic methylation patterns that were demethylated in approximately 10% of the offspring in a mixed genetic background. Upon further transmission of these transgenes in a mixed genetic background, the grandparental methylation patterns were reestablished in most progeny. Mating to inbred DBA/2 mice resulted in maintenance of the demethylated pattern or in further demethylation of the transgenes in approximately 50% of the offspring. In contrast, an equal number of transgenic siblings from matings to C57BL/6 mice showed a return to the original methylation pattern. The mosaic methylation status of this locus was apparently controlled by mouse-strain-specific factors. The methylation patterns of the 7-1B transgenes were not cellular mosaic and remained stable in all offspring, as with lines 5-8 and 8-1. Hence, the strain-dependent and cellular mosaic transgene methylation patterns of 7-1A animals were probably a consequence of the chromosomal integration site of the transgenes (position effect).

A rapid optimized protocol for downward alkaline Southern blotting of DNA.

A simple and efficient 2.5-h Southern blotting procedure is described that uses 0.4 M NaOH to transfer DNA in a downward direction. The resulting blots give signals that are both sharper and 30% stronger than those obtained by conventional upward-transferred blots.