Portal de Periódicos da CAPES

Automated Determination of Neutrophil Volume as Screening Test for Late-Onset Sepsis in Very Low Birth Infants

2010; Lippincott Williams & Wilkins; Volume: 29; Issue: 3 Linguagem: Inglês

10.1097/inf.0b013e3181c37fb4

1532-0987

Francesco Raimondi, Teresa Ferrara, Letizia Capasso, Maria Sellitto, Francesca Landolfo, Antonia Romano, Ernesto Grimaldi, Francesco Scopacasa,

Streptococcal Infections and Treatments

To the Editors: The diagnosis of sepsis is often a difficult issue for the neonatologist. Clinical signs and symptoms are poorly specific, and the quest for the optimal laboratory assay is still open. C-reactive protein (CRP) requires serial determinations to be of practical value,1 interleukins have not yet entered the routine clinical use, total white blood cells or absolute neutrophil counts alone have not enough sensitivity or specificity. Current hematology analyzers can determine white blood cell subpopulations based on their physical characteristics (eg, volume and granularity). In particular, mean neutrophil volume (MNeV) and its standard deviation (neutrophil distribution width) have been previously used to diagnose sepsis in an adult population with encouraging results.2,3 We investigated these same parameters in screening for late-onset neonatal sepsis, defined as a systemic infection, validated by a positive blood culture, diagnosed beyond the third day of life. For the purpose, we used an LH 750 analyzer (Beckman Coulter, Fullerton, CA) which can rapidly process a sample of about 8000 cells. A total of 120 consecutive, inborn very-low-birth-weight infants were screened for sepsis, using blood culture, complete blood count, absolute neutrophil count, I/T ratio, and CRP. Sixty babies were asymptomatic. Forty infants had clinical signs or symptoms of sepsis but a negative blood culture. Twenty infants were clinically suspected of sepsis and had a positive blood culture. We isolated Gram positive bacteria (5 Staphylococcus epidermidis, 1 other coagulase-negative Staphylococcus, and 1 Rothia mucilaginosa), Gram negative bacteria (1 Klebsiella oxytocica, 2 Pseudomonas aeruginosa, and 1 Enterobacter cloacae), and fungi (2 Candida albicans, 1 Candida glabrata, and 5 Candida parapsilosis). For each assay a receiver-operator curve was calculated and its area under the curve expressed to choose the optimal cut-off. Having chosen a threshold value, we calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value. Table 1 summarizes the tests performance. On a single determination, MNeV did better than any individual test with a negative predictive value of 98.9%, useful in excluding bloodstream infection. Since the MNeV PPV was considerably lower, this may limit the assay power to predict sepsis. However, among cases with a negative blood culture, the isolation of a pathogen from the urine or the central lines—done at discretion of the attending neonatologist—was significantly more frequent in the group with a positive MNeV (data not shown). Pooling together all symptomatic (proven and probable sepsis) infants with at least 1 positive culture, the PPV was 90%. When taken together, CRP and MNeV represent a powerful rapid combination both to suspect or exclude late onset sepsis. Neutrophil distribution width did not prove to be of any practical help in our neonatal series, possibly because the resting neutrophil population in the newborn is already more heterogeneous in size and shape than in the adult.TABLE 1: Sensitivity, Specificity Positive Predictive Value (PPV), Negative Predictive Value (NPV), and Area Under the ROC Curve for C-Reactive Protein, I/T Ratio, NeDW, Neutrophil Count, MNeV, White Blood Cells at the Optimal Cut-Off ValuesAs a conclusive remark, we acknowledge that further studies are warranted before MNeV can be recommended. A multicenter trial is desirable to address the issue of intercenter determination variability and to add power to data analysis. Francesco Raimondi, MD Teresa Ferrara, MD Letizia Capasso, MD Maria Sellitto, MD Francesca Landolfo, MD Antonia Romano, MD Division of Neonatology Department of Pediatrics Ernesto Grimaldi, MD Francesco Scopacasa, MD Department of Biochemistry and Medical Biotechnology Università "Federico II" Napoli, Italy