Does every millilitre of blood drawn from a newborn matter? Analysis of selected red blood cell and biochemical parameters in relation to anaemia in newborns

INTRODUCTION

Anaemia is a disorder caused by a decrease in the total number of red blood cells (RBC) and/or a decrease in the level of haemoglobin (Hb) as compared to the normal range for age and sex. It is also manifested by a reduced ability of RBC to carry oxygen. Circulating blood volume in newborns is about 80 ml/kg body weight. The main causes of anaemia include blood loss, reduced red blood cell production, and an increased in the number of destroyed erythrocytes. The condition may also be physiological or result from iatrogenic blood collection. Iatrogenic anaemia is of particular concern to paediatricians and neonatologists. It most often affects premature infants in intensive care units. This group of patients is particularly susceptible to electrolyte and acid-base imbalances and infectious diseases, the diagnosis and treatment of which frequently require laboratory evaluation, resulting in significant blood loss [1]. Anaemia of prematurity is the result of a physiological process that occurs in all newborns; however, it is aggravated by features such as reduced red cell mass at birth, shorter red cell lifespan, low iron stores, rapid weight gain, and increased iatrogenic losses during laboratory sampling. It has far-reaching consequences, including the transfusion of red cell preparations. Proper transfusion in children of all ages is essential to balance the benefits against its risks. These risks include transfusion of the wrong blood component due to errors, such as patient misidentification or unpredictable acute post-transfusion reactions. Recent studies suggest that a significant percentage of paediatric patients receive only one trans-fusion upon admission to the hospital, meaning that many transfusions could be avoided [2]. Anaemia in this group of patients has been proven to lead to poorer patient outcomes, and red cell concentrate transfusions are used to manage potential complications associated with low Hb levels [3]. Despite strict guidelines, the procedure still entails some risks. It should be emphasised that although in the past there was little evidence on which to base clinical decisions about transfusion in paediatrics and neonatology, a number of studies have been published in recent years that contribute to decision-making in this area [2].
The purpose of this study was to evaluate the effect of the number of taken blood samples on the total red blood cell count and to analyse its correlation with the development of iatrogenic anaemia among newborns.

MATERIAL AND METHODS

The study was conducted among 227 randomly selected newborns of both sexes (104 girls, 123 boys; mean age 14 days) hospitalised between January 2018 and March 2019. On admission to the hospital, after initial the laboratory tests, pneumonia (n = 202), hyperbilirubinaemia (n = 84), and umbilical cord inflammation (n = 38) were mainly diagnosed. It was a retrospective study conducted in the Department of Paediatrics, Neonatal Pathology, and Metabolic Bone Diseases. On admission to the Clinic, laboratory tests such as morphology and basic biochemical were collected in new-borns according to standard procedures. The capacity of the tube for a single morphology was 1.2 ml, while for bio-chemical tests it was 2.7 ml. The range of biochemical parameters included C-reactive protein (CRP), ionogram, glucose, urea, creatinine, and total bilirubin. Because this was a retrospective analysis of the documentation, the work did not require approval from the Bioethics Committee. Consent for the study was obtained from the parents, and the hospital and clinic authorities. No identifiable patient data were used in the study. Statistical analysis was performed using Statis-tica version 13. The results were expressed as mean ± standard deviation for continuous variables, and as numbers and percentages for categorical variables. The Shapiro-Wilk test was used to assess the normality of the distribution of the quantitative parameters in the study. The groups were compared using Student’s t-test or the non-parametric Mann-Whitney U-test, depending on the distribution of variables, and the chi-square test or Fischer’s exact test. Pearson’s cor-relation coefficient or Spearman’s rank correlation coefficient were also used for ordinal and quantitative variables. In all the analyses, a test probability value of p < 0.05 was considered statistically significant. Full data are available from the authors of the paper. The tests were performed by a diagnostic laboratory with the use of certified equipment. The following methods were applied: blood count – automatic method on a Sysmex analyser; biochemistry – enzymatic method on a Cobas Pro instrument. Data on the normal ranges of the evaluated parameters for newborns are shown in Table 1.

RESULTS

Among the study subjects, 89% (n = 202) of the newborns admitted to the hospital were diagnosed with pneumonia, 37% of the patients (n = 84) had hyperbilirubinaemia, 16.7% (n = 38) had umbilical cord inflammation, 7% (n = 16) were premature infants, and 6.2% (n = 14) presented transient feeding difficulties. The mean number of tests done per hospitalisation in all the newborns was 14.1 (M = 12), and the mean volume of blood taken per sample was 2.9 ml per kg body weight, or about 1.1% of circulating blood. The more samples were collected, the higher the observed decreases in parameters such as RBC, Hb, haematocrit (Hct), and mean corpuscular volume (MCV) (p < 0.001). Despite the small mean volume of blood collected for testing (M = 2.9 ml), its increase correlates with higher decreases in RBC count, Hb, Hct, and average platelet volume (in all p < 0.0001), as shown in Figures 1-3. The patients were divided into 6 groups according to the length of hospitalisation, as presented in Table 2. As for the number of tests taken, there were higher decreases in white blood cells (WBC) in group 1 and higher decreases in RBC, Hb, and Hct in group 2 (p < 0.001), which was also associated with taking larger volumes of blood, and higher decreases in Hb and Hct in group 6 (p = 0.046). Additionally, a statistically significant decrease in potassium was observed (p < 0.001). Thid proves that neonates with a lower body weight present more significant decreases in RBC, Hb, and Hct parameters (p < 0.001). Larger decreases in Hb, Hct, and RBC were shown in the newborns diagnosed with hyperbilirubinaemia (p < 0.001), who did not have more blood samples taken for testing (p = 0.23). On hospital admission, 3.1% (n = 7) of the newborns had a low Hb value, 3.5% (n = 8) a low Hct level, and 22.5% (n = 51) had a low RBC count. On the other hand, at hospital discharge, 4.8% (n = 11) of the newborns had a low Hb value, 6.2% (n = 14) a low Hct level, and 32.2% (n = 73) a low RBC count. During hospitalisation, the mean decrease in Hb was 2.1 g/dl (p < 0.00001), Hct 6.1% (p < 0.00001), RBC 0.51 million/mm³ (p < 0.00001), MPV 2.2 fl (p < 0.00001), and WBC (mean decrease 1870 thousand). Due to a greater amount of blood taken for tests in newborns with hyperbilirubinaemia (p < 0.00001), these patients are more likely to have lover red blood cell parameters, as shown in Figure 4 (data are for newborns hospitalised for 7-9 days, n = 134).

DISCUSSION

Laboratory tests are one of the most important aspects of medical decision-making in clinical practice. Analysing the literature on anaemia in newborns, we may find articles on premature infants, hospitalised in intensive care units, who are most affected by anaemia [4, 5]. The lack of guidelines in the practice of blood sampling is often a topic of contention because it is up to the physician to decide how often blood tests are ordered. Often young and inexperienced doctors are afraid of overlooking a change in a patient’s condition and therefore want to rule out every possibility, even if most tests do not provide any information relevant for the clinical picture. Such an approach may be the cause of iatrogenic anaemia [3]. Another issue discussed in the literature is the use of new methods and medical equipment aimed at reducing the collection of large volumes of blood sample taken for testing. Such methods involve continuous non-invasive monitoring of total Hb concentration. In newborns, they offer reliable Hb values that are comparable to the more traditional method of invasive venous blood sampling [6]. Another scientific study compared the results of haematological parameter determinations performed using a standard laboratory method and a “point-of-care” micromethod to validate the latter method in neonatal clinical practice. One hundred and fifty venous or capillary blood samples were collected from newborns born at term and preterm infants. The agreement between the values obtained with the 2 analysers was high for each of the parameters considered, i.e. Hb, Hct, and red and WBC [7]. German researchers conducted a single-centre prospective cohort study of newborns born at term and preterm infants in which they compared Hb levels obtained using a transcutaneous spectroscopic device against venous or capillary blood samples taken from newborns. Eighty infants were included in the study, and a total of 313 spectroscopic recordings were made over 2 hours (181 capillary, 142 venous). The correlation coefficient R was 0.96 for capillary/spectroscope pairs and 0.99 for vein/spectroscope pairs [8]. Another issue concerns transfusion therapy and potential administration of erythropoietin (EPO). Several multicentre studies have documented a moderate but statistically significant reduction in the number of required red blood cell transfusions in treated infants as compared to controls. Previous studies have unequivocally shown that EPO increases both erythropoiesis and reticulocyte counts. EPO can reduce the need for red blood cell transfusion in newborns, but its effect appears to be relatively small, whether given early or late. EPO is suggested to play a broader neuroprotective role, but risks include the development of retinopathy of prematurity (ROP) associated with pathological neovascularisation [9]. There is no consensus on the optimal Hb or platelet concentration thresholds for transfusions. A growing number of studies suggest that strict guidelines for blood transfusion should be introduced [2]. To this end, better markers are necessary to assess the need for transfusion itself, as well as to implement effective preventive measures, such as delayed cord clamping and minimising iatrogenic blood loss [10]. Our study shows that the more blood is drawn, the greater the declines in morphology regarding morphotic elements such as red and WBC, Hb, and Hct. Newborns with a low birth weight are most prone to rapid declines in the above-mentioned parameters. Researchers came to similar conclusions when they conducted a study on 20 newborns with a birth weight of less than 1500 grams during their first 4 weeks of hospitalisation. Blood was drawn from 1 to 13 times per day in each patient. The average blood loss ranged from 7 to 51 ml per kilogram of body weight over the 4 weeks – or 5% to 45% of the calculated total blood volume. It was further found that of the total blood taken, about 25% was surplus to the needs of analytical procedures [11]. Another study assessing the cumulative amount of iatrogenic blood loss in extreme preterm infants in the first month of life showed that extreme preterm infants lose almost one-third of their total blood volume in the first month of life as a result of blood loss caused by multiple draws for laboratory tests and procedures. Blood loss was higher in the most extreme preterm infants (30.2 ml/kg at 24 weeks versus 15.9 ml/kg at 27 weeks), and the median number of punctures per infant was 47 [12]. It is also worth analysing a study that focused on identifying factors that minimise blood loss caused by blood draws in a paediatric intensive care unit. Phlebotomy resulted in an average blood volume loss of 2.5 ±1.4 ml per collection, 7.1 ±5.3 ml per day, and 34 ±37 ml per paediatric intensive care unit stay, of which 1.4 ±1.1 ml per collection, 3.8 ±3.6 ml per day, and 23 ±31 ml represented excess blood. This excess represented 210 ±174% of the volume required by the laboratory and 110% of the excess collection. Blood drawn from central venous catheters had significantly higher excess volumes, at 254 ±112%, as compared to volumes drawn from arterial catheters, at 168 ±44%, and peripheral intravenous catheters, at 143 ±39%, p < 0.001 [13]. The results of the study clearly indicate that every microliter of blood drawn from a newborn requires immediate attention of the attending physicians. Consideration should be given to how and to what extent laboratory testing can be reduced so that blood sampling occurs as infrequently as possible. Most newborns experience a significant decrease in red blood cell parameters during hospitalisation, which consequently leads to an increased risk of developing anaemia. Implementing less invasive methods and using smaller blood volumes should lead not only to cost savings, but also to a reduction in iatrogenic anaemia for departments and clinics that recognise this problem. In cases where it is necessary to monitor the patient, the capacity of the tubes can be reduced, and for this purpose blood can be drawn for basic tests such as bilirubin or CRP, or possibly morphology alone without a blood smear. Bilirubin concentration can also be measured using a transcutaneous method with a bilirubinometer, because the results are comparable, thus reducing the risk of another blood draw [14]. Additionally, it is worth revisiting the reasons for hospitalisations and minimising overdiagnosis of certain disease entities to prevent unnecessary testing.

CONCLUSIONS

Hospitalisation in the neonatal period and associated diagnostic blood sampling can be a factor in the development of anaemia. Indications for laboratory tests in this age group should be carefully considered.

Disclosures

This research received no external funding.
Institutional review board statement: Not applicable.
The authors declare no conflict of interest.

References