Poster Session II: Sickle cell disease
Amongst patients with sickle cell disease (SCD) the leading cause of death is the acute chest syndrome (ACS). This pneumonia-like complication frequently occurs during or shortly after a vaso-occlusive crisis (VOC). In pediatric patients hospitalized for VOC, incentive spirometry has demonstrated to prevent the development of ACS.
This study was designed to determine if a comparable effect of incentive spirometry can be demonstrated in adult patients with SCD. Furthermore, we aimed to validate the ability of the Bartolucci score to identify patients at risk of ACS and assessed the value of procalcitonin as a potential biomarker for ACS. In addition, clinical characteristics and laboratory results were determined to identify potential risk factors.
In this multicenter prospective randomized trial, we included consecutive adult patients (≥18 yr) admitted for VOC presenting with chest or back pain above the diaphragm. Patients were randomly assigned to spirometry or control group. Patients presenting with ACS were excluded. A chest radiograph was performed 5 days after admission, or sooner when clinically indicated, in order to diagnose pulmonary abnormalities. ACS was defined as a new infiltrate/atelectasis combined with pulmonary symptoms. At presentation, procalcitonin plasma levels were assessed and the Bartolucci risk score was calculated to determine to the risk of developing ACS for each patient. In addition, clinical and laboratory parameters were compared between patients with and without ACS during admission.
In total 66 episodes of hospitalization for VOC in 48 patients were included. Median age was 26 years and 46 of the hospitalizations concerned patients with a severe genotype (HbSS/HbSβ0 thalassemia) versus 20 hospitalization with a mild genotype (HbSC/HbSβ+thalassemia). The overall incidence of ACS in this study cohort was 19.7%. In the spirometry group, ACS was diagnosed in 5/34 (14.7%) hospitalizations compared to 8/32 (25%) hospitalizations in the control group (OR 0.5 [0.15-1.8]; P = .293). Twelve of the 13 ACS episodes occurred in patients with a severe genotype. The Bartolucci risk score could be calculated for 50 hospitalizations. The scores area under the curve (AUC) was 0.747 (P = .013), with a negative predictive value (NPV) of 94% and a positive predictive value (PPV) of 31%. No difference in procalcitonin plasma levels were found between patients with and patients without ACS (0.52 ± 1.56 μg/ml versus 0.56 ± 1.44 μg/ml, respectively). At baseline, hemoglobin levels were significantly lower while LDH plasma levels, leukocyte and platelet counts were significantly higher in ACS hospitalizations as compared to non-ACS hospitalizations. Patients who developed ACS showed significantly more documented fever during admission (61.5% vs 17.0%) and a longer length of hospital stay (median 10.0 days vs 4.5 days).
Incentive spirometry did not significantly reduce the development of ACS in this prospective study in adult patients with SCD admitted with VOC and pain above the diaphragm. Procalcitonin plasma levels and the Bartolucci score could not accurately identify patients that at risk to develop ACS, but a low score appeared to be a reliable tool to identify patients with a low risk of ACS.