T Abiramalatha1, S Santhanam1, J J Mammen2, G Rebekah3, M P Shabeer1, J Choudhury2, S C Nair2. 1. Department of Neonatology, Christian Medical College, Vellore, India. 2. Department of Transfusion medicine and Immunohematology, Christian Medical College, Vellore, India. 3. Department of Biostatistics, Christian Medical College, Vellore, India.
Abstract
OBJECTIVE: The objective of this study was to determine changes in neutrophil volume conductivity scatter (VCS) parameters and their distribution widths (DW) in neonatal sepsis and to estimate their optimal cutoff levels using receiver operating characteristic (ROC) curves. STUDY DESIGN: In a cohort of neonates evaluated for sepsis, blood counts and blood culture were performed initially, with repeat counts and C-reactive protein (CRP) done after 24 to 48 h. Neutrophil VCS parameters from both the initial and repeat blood counts were analyzed. Babies were classified as having blood culture-positive sepsis, probable sepsis (clinical course consistent with sepsis and CRP-positive, but culture-negative) and no sepsis (clinical course not compatible with sepsis, culture- and CRP-negative). RESULTS: A total of 600 babies were included: 240 (40%) babies in the sepsis group and 360 (60%) babies in the control group. All the neutrophil VCS parameters and their DWs (except for low angle light scatter in the repeat counts) were significantly different between the two groups, with an area under curve in the ROC curve of >0.6 for most parameters. The five most significant VCS parameters (mean neutrophil volume (MNV), median angle light scatter (MALS), lower median angle light scatter (LMALS), MNV-DW and ALL-DW) had around 65 to 75% sensitivity and specificity. A combination of leukopenia, thrombocytopenia, MNV and LMALS had a likelihood ratio (LR)+ of 15.3 and LR- of 0.17. With a pre-test probability of 40%, post-test probability increased to 91% for a positive test and decreased to 10% for a negative test. A prospective validation study was performed recruiting an additional 60 babies, which showed similar results, assuring that the cutoffs were robust. CONCLUSION: Neutrophil VCS parameters cannot be considered as stand-alone tests to diagnose or rule out neonatal sepsis, but can be used in combination with other hematological screening tests to improve the diagnostic accuracy of the neonatal sepsis screen.
OBJECTIVE: The objective of this study was to determine changes in neutrophil volume conductivity scatter (VCS) parameters and their distribution widths (DW) in neonatal sepsis and to estimate their optimal cutoff levels using receiver operating characteristic (ROC) curves. STUDY DESIGN: In a cohort of neonates evaluated for sepsis, blood counts and blood culture were performed initially, with repeat counts and C-reactive protein (CRP) done after 24 to 48 h. Neutrophil VCS parameters from both the initial and repeat blood counts were analyzed. Babies were classified as having blood culture-positive sepsis, probable sepsis (clinical course consistent with sepsis and CRP-positive, but culture-negative) and no sepsis (clinical course not compatible with sepsis, culture- and CRP-negative). RESULTS: A total of 600 babies were included: 240 (40%) babies in the sepsis group and 360 (60%) babies in the control group. All the neutrophil VCS parameters and their DWs (except for low angle light scatter in the repeat counts) were significantly different between the two groups, with an area under curve in the ROC curve of >0.6 for most parameters. The five most significant VCS parameters (mean neutrophil volume (MNV), median angle light scatter (MALS), lower median angle light scatter (LMALS), MNV-DW and ALL-DW) had around 65 to 75% sensitivity and specificity. A combination of leukopenia, thrombocytopenia, MNV and LMALS had a likelihood ratio (LR)+ of 15.3 and LR- of 0.17. With a pre-test probability of 40%, post-test probability increased to 91% for a positive test and decreased to 10% for a negative test. A prospective validation study was performed recruiting an additional 60 babies, which showed similar results, assuring that the cutoffs were robust. CONCLUSION: Neutrophil VCS parameters cannot be considered as stand-alone tests to diagnose or rule out neonatal sepsis, but can be used in combination with other hematological screening tests to improve the diagnostic accuracy of the neonatal sepsis screen.