Analysis of the Prevalence and Risk Factors of Adult Neutropenia in a Large Cohort in Japan

Background: The prevalence of neutropenia is estimated to be 0.5% to 25% based on the definition of neutrophil count less than 1,500 /μL. The high variety mostly depends on the race. Most of the estimation have been based on the values obtained from one measurement of neutrophil count. The prevalence of "chronic" neutropenia is largely unknown, particularly in adults, with only a few small epidemiological studies of adult chronic neutropenia being found. Here we analyzed the medical-check data from over 30,000 Japanese adults, and report the prevalence and characteristics of neutropenia in Japanese adults referring to chronic cases.

Methods: Data of medical-checks including neutrophil count from 2010 to 2014 were collected. The entire cohort included 33,255 adults. Neutropenia was defined as neutrophil count less than 1,500 /μL. Anemia was defined as hemoglobin level less than 13 g/dL in men and 12 g/dL in women, and thrombocytopenia as platelet count less than 100,000 /μL. Data of past histories and comorbidities were based on the self-report by the participants. When simple neutropenia (not necessarily chronic) was analyzed, the minimum neutrophil count was used for those who received more than one medical-checks, and other data at the time of the minimum neutrophil count were adopted. For the statistical analysis for comparison of subjects with and without neutropenia, Student's t-test or Wilcoxon's rank sum test were used for continuous variables, and chi-square test was used for categorical variables. Univariate and multivariate regression analyses were performed for testing the effect of the candidate factors on the neutrophil count. For the analysis of the risk factors of neutropenia, multivariate logistic regression analysis was used. For the analysis of chronic neutropenia, only those who received at least 2 medical-checks with measurement of neutrophil count were included. Those who showed neutropenia at least twice were defined as having chronic neutropenia.

Results: After exclusion of the cases with past histories of hematological malignancies, 33,201 cases were included. Among them, 1,391 (4.18%) cases had neutropenia. Frequencies of collagen diseases, hypothyroidism, and under-treatment hyperthyroidism were higher in the neutropenic group than the non-neutropenic group. When the subjects with these diseases were excluded, 1,348 (4.12%) out of 32,695 cases had neutropenia. Among them, frequencies of anemia and thrombocytopenia in the neutropenic group were higher than in the non-neutropenic group (18.7% vs 6.2%, and 0.7% and 0.1%, respectively, p < 0.001). In order to analyze isolated neutropenic cases, the subjects with anemia and/or thrombocytopenia were further excluded from the cohort. As the result, 28,615 cases were analyzed. Among them, 1,083 (3.78%) cases were diagnosed to have neutropenia, in which 1,025 (94.6%) cases had mild neutropenia (1,000 - 1,499 /μL) and 58 (5.4%) had moderate neutropenia (500 - 999 /μL). There were no cases of severe neutropenia (< 500 /μL). The neutropenic group showed older age (52.2 vs 50.2 years old), female predominance (64.2% vs 37.1%), lower bone mass index (BMI) (21.5 vs 23.4), lower smoking habit (23.3% vs 46.5%), lower drinking habit (30.8% vs 38.3%) (p < 0.001 for all). Diabetes mellitus (3.0% vs 4.9%, p = 0.005), hypertension (10.6% vs 15.7%, p < 0.001), and hyperuricemia (1.7% vs 3.3%, p = 0.003) were less frequent, and pollen allergy (6.3% vs 4.2%, p < 0.001) and viral hepatitis (1.1% vs 0.3%, p < 0.001) were more frequent in the neutropenic group. In multivariate regression analysis, female, drinking habit, pollen allergy, and viral hepatitis were associated with decrease in neutrophil count, whereas higher BMI and smoking habit with increase in neutrophil count. Risk factors of neutropenia were female with adjusted odds ratio (OR) 2.18 (p < 0.001), drinking habit (OR 1.31, p < 0.001), pollen allergy (OR 1.32, p = 0.034), and viral hepatitis (OR 3.98, p < 0.001). Higher BMI and smoking habit were negative factors of neutropenia (OR 0.40 and 0.52, respectively, p < 0.001).Among the 18,279 subjects who received at least 2 medical checks with measurement of neutrophil count, the number of chronic neutropenia was 294 (1.61%).

Conclusion: The prevalence of isolated neutropenia in Japanese cohort was 3.78%, and chronic neutropenia was seen in 1.61%. Risk factors of neutropenia were female, drinking habit, pollen allergy, and viral hepatitis.