Search

Aim Three different methods are widely available in neonatal units for the measurement of neonatal Haematocrit (HCT); centrifugation (spun micro-haematocrit; SHCT), automated calculation (laboratory full blood count; LHCT) and conductivity (blood gas analyzer haematocrit; BGHCT). All three methods are often used interchangeably or simultaneously. However, limited studies suggest that SHCT values are higher than others due to plasma trapping. This study compares neonatal HCT measurements by the three different techniques using statistical methods. Methods Retrospective review of data extracted from a point of care testing logbook kept in our neonatal unit (for SHCT and BGHCT) and electronic laboratory records. Venous samples that concurrent testing with more than one methodology was done were included. Capillary and arterial samples were excluded. Results from each method were compared using a paired sample t-test and agreement was examined by Bland-Altman analysis. Results A total of 137 samples from 127 unique neonatal patients were identified. The mean and median HCT values for each method are shown in table 1. A strong correlation (R2= 0.875) between LHCT and SHCT values, as well as between LHCT and BGHCT values (R2=0.903) was seen. However, SHCT values were higher than LHCT values with a mean difference of 5.19% (4.53, 5.86). A paired t-test found this difference to be statistically significant; t (126) =15.5, p 65% which are consistent with polycythemia. The Bland-Altman difference plot for LHCT and SHCT methods demonstrated a good agreement between the two, but also defined a bias of 5.2%, with an agreement range of -2.18 to 12.57. Conclusion Although there is good correlation between all the test values, our study confirms that SHCT values are consistently higher than the others. Further studies are needed to investigate the clinical impact of this observed difference between methodologies, in neonatal pratice and particularly in the management of neonatal polycythemia.