Role of early complete blood count parameters to predict morbities and mortality in very preterm babies.

Background: Prematurity is an important cause of mortality and morbidity associated with neonatal period. Until today, various markers have been investigated for predicting short and long-term morbidities in preterm infants, however, most of them are difficult to use in clinical practice or the predictive values are very low. The aim of this study is to determine the predictive value of detailed complete blood count parameters in cord blood and early postnatal period for prematurity related morbidities and mortality in premature infants under 32 th gestational weeks. Method: A retrospective descriptive cross-sectional study included 162 infants born before 32nd gestational week in the Dokuz Eylül University Hospital and hospitalized in the neonatal intensive care unit between the dates of January 1, 2015 and December 31, 2018. The maternal and perinatal characteristics of the infants and the morbidity and mortality data in the intensive care process were obtained from the patient files and the electronic information system of the hospital. Respiratory distress syndrome (RDS), patent ductus arteriosus (PDA), intraventricular hemorrhage (IVH), early or late onset neonatal sepsis, necrotizan enterocolitis (NEC), periventricular leukomalasia (PVL), brochopulmonary dysplasia (BPD), retinopathy of prematurity (ROP) were evaluated for prematurity related morbidities. The parameters used in complete blood count (CBC) [(white blood cell count (WBC), absolute neutrophil count (ANC), absolyte lymphocyte count (ALS), monocyte count (MONO), eosinophil count (EOS), basophil count (BAZO), red blood cells count RBC, hemoglobin (HGB), hematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red blood cell distribution width (RDW), platelet count (PLT), mean platelet volume (MPV), plateletcrit (PCT)] were recorded separately for all blood counts of cord blood and first postnatal week, postmenstrual CBC values in 30-36th weeks were recorded separately for each week. CBC parameters of infants who were transfused in the postnatal first week before transfusion were evaluated in order not to affect the results. The relationship between perinatal risk factors and CBC parameters and morbidity and mortality was evaluated with univariate analysis, than the diagnostic value of the parameters which were found statistically significant was investigated by ROC analysis. In the ROC analysis, the predictive values were determined for the parameters with more than 70% of the area under the curve, and these parameters were evaluated by data mining methods for each morbidity and mortality. Results: Median gestational week and birth weight of infants were 28,6 weeks (min-max: 22,1-31,9) and 1130 grams (min-max: 400-2018) respectively. As the gestational week and birth weight decreased, there was a significant increase in morbidity and mortality risk. The most important risk factors affecting mortality and morbidity were related with birth weight, gestational week and chorioamnionitis, duration of invasive mechanical ventilation, extrauterin growth retardation and the number of blood transfusion. Although many CBC parameters were related with certain morbidities in the univariate analysis, a small proportion had reached significant values by ROC analysis. The highest predictive parameters were related to the erythrocyte mass and index of erythrocytes when evaluated in the third day blood count. Red blood cells count (RBC) less than 4 million / mm3 or hematocrit (HCT) below 40% increased the risk of ROP and BPD. Mean corpuscular hemoglobin concentration (MCHC) over 33,5 gr/dl was also associated with increased risk of ROP. In adition, total number of erythrocyte replacement more than 4 times during the treatment process was found to be a significant risk factor for ROP, BPD and mortality. Finally, new models to predict PVL, BPD, ROP and mortality including gestational age, birth weight, type of delivery, chorioamnionitis, extrauterine growth retardation, duration of invasive mechanical ventilation, number of erythrocyte replacement and CBC parameters (HCT, RBC, MCHC) had revealed sufficient sensitiviy and positive predictive values (>85% ) by data mining methods Conclusion: The present study suggests that, in addition to gestation week, birth weight and other well known risk factors, certain paramaters in CBC, related with erythrocyte mass and index predicts the premature complications. These findings may provide a different point of view in evaluating hemogram parameters in clinical practice. If these results can be supported by more comprehensive prospective studies, advanced scoring sytems may be developed.