OBJECTIVES

We developed a diagnostic score to differentiate congenital from noncongenital neutropenia at the time of diagnosis using reliable data collected at the first visit of a patients with neutropenia.

STUDY DESIGN

In a pilot retrospective study, we included 120 patients diagnosed with chronic neutropenia; 61 had congenital and 59 had noncongenital neutropenia. We reviewed patient medical charts and collected the initial complete blood count (CBC) and other reliable data. We used logistic regression to determine the probability that the neutropenia was congenital.

RESULTS

On the initial CBC, the degree of neutropenia had no predictive value; only monocytosis >1.5 × 10 /l, hemoglobin <90 g/l, or mild thrombocytopenia <150 × 10 /l suggested congenital neutropenia. The most predictive factors for congenital neutropenia were a medical history (consanguinity and patient history of neutropenia), severe infections, and oral stomatitis or gingivitis at the time of diagnosis. The age at diagnosis had limited predictive value.

CONCLUSION

A diagnosis of congenital neutropenia may be reliably suspected based only on information from the CBC, some basic information from patient and parent interviews, and a clinical examination. A pilot score with six factors that could be readily, reliably collected, should facilitate the diagnosis of congenital neutropenia.