Procalcitonin as a Potential Biomarker in the Study of Babesiosis Caused by B. microti

Abstract

Procalcitonin is gaining momentum in the study of protozoal sepsis, but its utility as a biomarker has yet to be fully discovered in human babesiosis. A total of 33 cases of acute babesiosis dating between 2012 and 2019 were retrospectively collected from Stony Brook University Hospital (SBUH) and Stony Brook South Hampton Hospital (SHH), both of which are located on Long Island, NY. Cases were cross-referenced for the need for ICU admission, and the procalcitonin levels were measured by the use of BRAHMS Elecsys assay at SBUH and BRAHMS Architect assay at SHH. Our study demonstrated that the log-transformed procalcitonin levels had a linear correlation with log-transformed maximum parasitemia, which suggests that procalcitonin directly correlates with parasitemia values. Furthermore, when comparing values that predict ICU admission, our ROC analysis of procalcitonin demonstrated similar AUC values to the percentage of parasitemia, suggesting that procalcitonin may assist in determining the severity of disease. We demonstrate that procalcitonin may directly correlate with the parasitemia percentage and have prognostic capabilities, which suggests that procalcitonin may have biomarker potential in human babesiosis.

Keywords: babesiosis; biomarker; parasitology; procalcitonin; tick borne illness.

Figure 1

(A) Linear correlation between log₁₀-parasitemia and log₁₀- procalcitonin levels. The linear correlation of the log-transformed variables was r = 0.556 (p = 0.001), similar to the nonparametric (Spearman) correlation coefficient (ρ = 0.567). The red dots represent patients who were admitted to the intensive care unit. (B) Receiver-operator characteristic curve of parasitemia for prediction of intensive care unit admission (0.75; 95% CI: 0.58–0.89; p = 0.005); optimal cut-off is 3.6%. Receiver-operator characteristic curve of procalcitonin for prediction of intensive care unit admission (C = 0.77; 95% CI: 0.58–0.89; p = 0.005); optimal cut-off is 1.2 ng/mL.