Assessment of prognostic factors for differential diagnostics between mono- and mixed infection of the febrile form of tick-borne encephalitis

Abstract

Objective: is to assess clinical and laboratory prognostic factors to develop a differential diagnostic model between the monoinfection of tick-borne encephalitis febrile form and the mixed infection of tick-borne encephalitis with Lyme borreliosis non-erythemal form at the onset of the disease.Materials and methods. The clinical examination involving 56 patients with tick-borne encephalitis febrile form (mean age: 46.1±3.1 years) and 27 patients with the mixed infection of tick-borne encephalitis with Lyme borreliosis non-erythemal form (mean age: 47.2±3.2 years) has resulted in the assessment of 65 clinical and laboratory parameters in the first week of the disease including 14 indicators of standard and extended hemogram profiles and 6 blood leukocyte indices. Pearson’s goodness-of-fit test was used for statistical analysis. The predictive values of the parameters were determined by the odds ratio and ROC analysis with AUC. The logistic regression model was developed using STATISTICA 12.0.Results. To make differential diagnosis between mono- and mixed infection at the onset of the disease the following hematological parameters with “average” or “good” predictive values can be used: band neutrophil count (AUC=0.65), the index of leukocytes and erythrocyte sedimentation rate ratio (AUC=0.66), erythrocyte sedimentation rate (AUC=0.70), neutrophil granularity intensity (AUC=0.66), neutrophil reactivity intensity (AUC=0.72) and reactive lymphocytes count (AUC= 0.72). A logistic regression model with a “very good” predictive value (AUC=0.83) is developed which includes the following four predictors: band neutrophil count, erythrocyte sedimentation rate, NEUT-RI and NEUT-GI in peripheral blood.Conclusion. The model is allowed to make a differential diagnosis between the mono- and the mixed infection of tick-borne encephalitis with good sensitivity and specificity values in the first week of disease.

Keywords

• tick-borne encephalitis
• febrile form
• mixed infection
• lyme borreliosis
• reactive neutrophils
• reactive lymphocytes
• blood leukocyte indices
• logistic regression model