Журнал инфектологии (Oct 2022)

Predictive value of complete blood count for early diagnosis of mixed infection of the non-erythema migrans form of Lyme borreliosis and tick-borne encephalitis

  • E. N. Ilyinskikh,
  • E. N. Filatova,
  • Yu. I. Bulankov,
  • V. N. Nekrasov,
  • A. V. Reshetova,
  • E. V. Portnyagina,
  • S. V. Aksyonov,
  • O. V. Voronkova,
  • E. A. Motlokhova,
  • I. N. Ilyanova

DOI
https://doi.org/10.22625/2072-6732-2022-14-4-69-76
Journal volume & issue
Vol. 14, no. 4
pp. 69 – 76

Abstract

Read online

Objective: is to develop a model for early diagnosis of mixed infection of the non-erythema migrans form of Lyme borreliosis and tick-borne encephalitis using the assessment of the complete blood count and the blood leukocyte indices of patients in the first week of the disease.Materials and methods. The retrospective clinical study involved Group 1 of 27 patients with the mixed infection of the non-erythema migrans form of Lyme borreliosis and the febrile form of tick-borne encephalitis and Group 2 of 29 patients with the monoinfection of the non-erythema migrans form of Lyme borreliosis, who were hospitalized no later than in the 7th day of the disease. The average age of patients in Groups 1 and 2 was 50.6±3.4 and 49.9±2.3. We analyzed 14 parameters of the complete blood count as well as calculated the leukocyte intoxication index and the body resistance index. Statistical significance assessment was carried out using the chi-square test and ROC analysis. Logistic regression model was developed using STATISTICA 12.0 modules.Results. The levels of the band and polymorphonuclear neutrophils (<0,001 и p=0,002) and the leukocyte intoxication index (p<0,001) were significantly higher and the levels of the body resistance index (p<0.001), lymphocytes (p<0.001) and platelets (p=0.004) were lower in Group 1 than in Group 2. Informative predictors of mixed infection included the body resistance index (AUC=0.77), leukocyte intoxication index (AUC=0.75), the band and polymorphonuclear neutrophils (AUC=0.74), lymphocytes (AUC=0.77), and platelets (AUC=0.70). Logistic regression model has a “very good” predictive value (AUC=0.85) and include two parameters: body resistance index and platelets (×109/L).Conclusion. The developed model has a “very good” predictive value for early diagnosis of the mixed infection of the non-erythema migrans form of Lyme borreliosis and tickborne encephalitis before laboratory diagnosis confirmation.

Keywords