Les enfants âgés de moins de 18ans présentant une neutropénie sur un hémogramme réalisé au centre hospitalier universitaire (CHU) de Caen, à l’exclusion de ceux ayant reçu une chimiothérapie, ont été inclus pendant 12 mois dans une étude prospective observationnelle à visée descriptive. Le dossier a été suivi au minimum un an après l’inclusion et les données clinico-biologiques ont été recueillies prospectivement, afin de comparer les neutropénies transitoires (<3 mois) et persistantes (>3 mois). Au cours de 55 018 consultations et 13 967 hospitalisations, 8966 hémogrammes ont été effectués et 250 neutropénies répondant aux critères d’inclusion ont été identifiées chez 238 patients, dont 195 ont pu être suivis (suivi médian=12,8 mois). Lorsque la neutropénie était<0,5×10⁹/L (9,2 %), elle était plus fréquemment persistante (RR=3,08, IC 95 %=[1,31–7,22]). En analyse multivariée, seule une neutropénie<0,5×10⁹/L était significativement associée à sa persistance. Les causes infectieuses étaient majoritaires (37,8 %). Ces résultats suggèrent que la majorité des neutropénies découvertes fortuitement en pédiatrie concernent des enfants jeunes, que leur évolution est transitoire et bénigne et qu’elles sont souvent d’origine virales, tandis que les neutropénies<0,5×10⁹/L semblent plus à risque de persister et d’évoluer défavorablement. En conclusion, la découverte d’une neutropénie chez l’enfant n’est pas exceptionnelle, elle doit attirer l’attention du clinicien et nécessite une surveillance clinique, surtout si elle est profonde.

Neutropenia seems to be quite frequent in current pediatric practice and can confuse the clinician since it may result from a severe cause. The aim of this study was to provide a prospective description of episodes of neutropenia in children to assess its clinical relevance in a general pediatric cohort consulting and/or hospitalized in a French university hospital. In this prospective observational and monocentric study conducted from April 2012 to April 2013, we included all the patients under 18 years of age who presented neutropenia (defined as an absolute neutrophil count [ANC] below 1×10⁹/L before 1 year of age and below 1.5×10⁹/L beyond) on a whole blood count (WBC) performed in our hospital. Patients treated with chemotherapy were not included. Medical records were regularly checked for at least 1 year after inclusion, and clinical and biological data were collected prospectively to compare transient episodes of neutropenia (<3 months) with persistent episodes of neutropenia (>3months). Of 55,018 consultations and 13,967 hospitalizations (chemotherapy excluded), 8966 blood counts were performed and 250 episodes of neutropenia were found in 238 patients. Data concerning clinical progression were available in 195 cases of which 136 had at least one subsequent WBC. Two hundred thirty-one episodes corresponded to new episodes, while neutropenia preexisted before inclusion in the others. The median follow-up was 12.8 months. Most episodes of neutropenia occurred in children <2 years of age (52%), with a median age of 22.2 months. Mean ANC was 0.943×10⁹/L (±0.340) and a few episodes of neutropenia were below 0.5×10⁹/L (9.2%). Neutropenia persisted more than 3 months in only 13.2% of cases. When neutropenia was below 0.5×10⁹/L, it significantly persisted (RR=3.08; 95% CI [1.31–7.22]). Other factors associated with persistent neutropenia were thrombocytopenia, monocytopenia, a CRP more than 70mg/L, significant abnormality on the clinical exam, and age over 24 months. However, multivariate analysis showed that only an ANC below 0.5×10⁹/L was significantly associated with persistence. While etiology could not be determined in 32% of cases, neutropenia resulted mostly from infectious causes (37.8%), with other causes being more anecdotal. The majority of infectious episodes of neutropenia were viral (90.3%). Like other studies, this investigation suggests that most episodes of neutropenia concern young children, are transient, are benign and often due to infectious diseases. Although it may not reflect the medullar stock or the real capacity of neutrophils to fight bacterial infections, it seems that neutropenia below 0.5×10⁹/L is more likely to persist and be complicated, as previous studies also suggest. To conclude, neutropenia is not exceptional in children and, even if it often results from viral infections and mostly evolves favorably, the clinician should closely monitor these patients, especially when neutrophils are below 0.5×10⁹/L.