Preliminary data that Dr. Florin presented at PAS found that the median CRP, PCT, and proADM did not differ by gender, race, ethnicity, or insurance status. “In addition, there were not significant differences in the distribution of disease severity by biomarker performed, with approximately 27% of patients being classified as mild, 66% as moderate, and 7% as severe,” he said.
The median CRP was 2.4 ng/mL in those with mild disease, 2.5 ng/mL in those with moderate disease, and 6.25 ng/mL in those with severe disease, with the difference between the two subclasses of nonsevere disease and moderate disease and severe disease reaching statistical significance (P = .002). The median PCT was 0.16 ng/mL in those with mild disease, 0.26 ng/mL in those with moderate disease, and 0.49 ng/mL in those with severe disease, with the difference between the two subclasses of nonsevere disease and moderate disease and severe disease reaching statistical significance (P = .047). Meanwhile, the median proADM was 0.53 ng/mL in those with mild disease, 0.59 ng/mL in those with moderate disease, and 0.81 ng/mL in those with severe disease, with the difference between the two subclasses of nonsevere disease and moderate disease and severe disease also reaching statistical significance (P less than .0001).
Next, the researchers performed logistic regression of each biomarker individually and in combination. They found that and had the best ability to discriminate those developing severe vs. nonsevere disease (area under the receiving operating curve of 0.72, vs. 0.67 and 0.60, respectively). When CRP and PCT markers were combined with proADM, they were no longer associated with severe disease, while a strong association with proADM remained significant.
Dr. Florin acknowledged certain limitations of the study, including the fact that requiring collection of blood samples may have resulted in an enrollment bias toward patients receiving phlebotomy or IV line placement in the ED. “In addition, the children in the moderate-severity group are likely more heterogeneous than the other two severity groups,” he said. “Finally, given that this is a single-center study, we had a relatively small number of outcomes for some of the individual severity measures, which may have limited power and precision.”