Evaluation of the relationship between serum eosinophil and D-dimer levels and demographic and clinical characteristics in bullous pemphigoid patientsDear Editor, Peripheral eosinophilia is reported in 50%-60% of BP patients in various studies, with findings suggesting that both peripheral and lesional eosinophil counts correlate with the disease's severity. 1 This retrospective study evaluated the eosinophil count, D-dimer level and venous thromboembolism (VTE) frequency and their relationship with clinical features in 207 patients diagnosed with BP between 2005 and 2022 (Table 1). Serum eosinophil levels were available for 196 patients, and 46.9% of them were found to have eosinophilia (normal range: 0-0.7/μL). Similar to findings in a recent study, eosinophilia was associated with older age at diagnosis, palmoplantar involvement and greater disease severity compared to normal eosinophil levels, which were associated with younger age at diagnosis and mucosal involvement. 2 No statistically significant difference in mortality was observed between those with and without eosinophilia. Given that patients with eosinophilia are generally older at disease onset and at a higher risk for age-related systemic diseases, it may not be appropriate to establish a direct link between eosinophil count and mortality. Similar to previous researches, BPDAI global score (the sum of total activity score with skin lesions and mucosal involvement and damage score), pruritus score and D-dimer levels were found to be significantly higher in patients with serum eosinophilia than in those without (Table 2). [1][2][3][4] Imaging studies identified thrombosis in 11 (22.4%) and pulmonary embolism in 12 (48%) patients, with 3 patients exhibiting both conditions. In total, thrombotic events were detected in 20 (36%) of the 56 patients investigated. However, it is noteworthy that this rate appears to be relatively low (≤20%) among clinically suspected individuals, as indicated by previous studies. 5-7 D-dimer level is affected by diverse factors. While highly sensitive (>95%) for VTE detection with a 500-ng/mL threshold, its clinical utility diminishes with age. 8 Age-adjusted thresholds combined with clinical assessment improve negative predictive value (NPV) in patients ≥50 years. 9 In our study, NPV for VTE was 93.5% with a sensitivity of 84.2% using ageadjusted thresholds. Conducting our study at a single reference centre may have led to the inclusion of more severe and treatment-resistant patients. The retrospective nature of the study limited the application of standardized criteria