METHODS
In a single-center, retrospective study approved by the institution’s Research Ethics Committee, patients less than 18 years old who were diagnosed with CE and followed up in a pediatric hematology outpatient unit formed the sample. The paper-based files and electronic medical records of all such patients diagnosed with CE from 2000 to 2020 were evaluated by pediatric hematologists to document demographic characteristics, presenting symptoms, family history, the findings of physical examinations, and the results of laboratory tests. The center’s routine practice also entailed screening full blood counts of each patient’s parents and siblings. Capillary oxygen saturation, venous blood gas, and serum EPO levels, along with the results of Hgb electrophoresis and JAK2 mutation analysis, were recorded as well. Phlebotomy practices and any complications of the disease during follow-up were assessed. During the study, patients who had aged beyond 20 years and been transferred to adult departments were telephoned to gather up-to-date medical data.
Erythrocytosis was defined as having hemoglobin levels exceeding the age- and gender-adjusted 99th percentile or an increase of 2 g/dL in hemoglobin levels from baseline.3 Hgb percentile curves were used to evaluate age- and gender-adjusted values,12 while for diagnosis, full blood counts were performed at least twice at different times. No patients had serum EPO levels exceeding laboratory reference levels. Patients with thrombocytosis, leukocytosis, and/or splenomegaly were excluded from the sample, as were patients with secondary causes of erythrocytosis (e.g., cardiac, pulmonary, renal, and hepatic disorders and malignancies) according to their history, physical examinations, capillary oxygen saturation analysis, venous blood gas measurement, and serum levels. In patients who reported any bleeding-related symptoms, prothrombin time, activated partial thromboplastin time, and bleeding time assays were also performed.
Statistical analysis was performed using IBM SPSS version 22.0. Categorical variables were analyzed as percentages, and quantitative data were recorded as M ± SD and median using descriptive statistics.