Lung function decline in children with sickle cell disease treated with hydroxyurea

Sickle cell disease (SCD), the most common form of an inherited hematological disorder, is caused by a genetic variant that leads to the synthesis of an abnormal hemoglobin, HbS. Deoxygenated HbS tends to polymerize which results in a distortion of the red blood cell shape causing vaso-occlusion and end-organ damage. The most severe forms of SCD are homozygous HbSS and HbSβ0–thalassemia. The latter is characterized by compound heterozygosity for HbS and a non-functional beta-thalassemia gene. Milder forms of SCD include compound heterozygous conditions, such as HbSC. SCD is a systemic disease, associated with episodes of acute illness and chronic solid organ damage, including SCD lung disease.1 Prior to therapy with hydroxycarbamide, also called hydroxyurea (HU), in SCD HbSS untreated children suffered from severe and progressive SCD lung disease with annualized declines in pulmonary function testing (PFT) similar to cystic fibrosis.2 Male sex was associated with a more rapid PFT decline in HbSS children not treated with HU, and HbSC genotype was associated with less lung disease compared to HbSS.2 HU is a disease-modifying agent that increases HbF production, total hemoglobin concentration, as well as red blood cell survival, and has been used for SCD therapy for decades. In adults, HU therapy results in a remarkable reduction in morbidity and mortality.3 An earlier, preliminary study in a small pediatric cohort had suggested that HU therapy in SCD children may lead to improvement in PFT decline.4 However, longitudinal data on lung disease progression and risk factors associated with PFT decline in current, larger pediatric cohorts of SCD HbSS treated with HU or HbSC are lacking. The aim of this study therefore was to quantify lung disease progression in a large cohort of SCD HbSS children treated with HU according to current guidelines, and to analyze the potential effects of sex on lung function decline in children with HbSS or HbSC. Longitudinal PFT results were analyzed in a cohort of children with SCD treated at the Hospital for Sick Children, Toronto, Canada, between 1997 and 2022. The study was approved by the local Research Ethics Board (REB# 1000078457). Patients with either HbSS or with HbSC and 5 to 18 years of age attending SCD outpatient clinic who performed PFT as part of routine clinical care were included. Patients were identified by searching the respiratory medicine PFT database for a physician diagnosis of “ sickle cell disease” or “sickle cell anaemia”. To be included in the analysis, children had to have at least two recorded PFTs during the study period that were performed at least three months apart. Clinical visits in follow-up of an episode of acute chest syndrome or labeled as “pulmonary exacerbation” by the treating physician were excluded. Baseline was defined as the first visit with acceptable PFTs by ATS/ERS criteria.5 Forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) measured by spirometry as well as total lung capacity (TLC) obtained by plethysmography were expressed as percent predicted (pp) and z-scores using published GLI equations,6 without correction for race. Demographics, lab results, genetic and clinical characteristics were abstracted from electronic medical records by chart review. Hb analyses were only included in the study if obtained on the same day as PFTs. Statistical analysis was performed using SPSS software V.27.0 (IBM Corporation, NY, USA). For the purpose of the analysis, SCD genotype HbSβ0 (beta thalassemia) was combined with HbSS, as these cause similar disease severity. Sample characteristics at Baseline were calculated and expressed as means and standard deviation (SD). Categorical variables were compared using Pearson's chi-square test. All continuous variables were tested for normal distribution (Kolmogorov–Smirnov test), and differences between groups were analyzed using t-test or Mann–Whitney-U test, as appropriate. To calculate annual PFT decline based on sex disparity, linear regression models were used for males and females and tested for significance by ANOVA. To test for further risk factors, Hb and HbF% were added to the regression models as confounders. All regression models were validated by 1000 iterations of bootstrap. To test the relationship between the slopes a hypothesis test on the difference between regression coefficients was performed. All models were adjusted for the number of visits with PFTs. A p-value <.05 was considered significant and corrected for multiple testing with the Bonferroni method. A total of 795 SCD children who had performed PFTs during the 25-years study period were identified. Seventeen patients were excluded due to misdiagnosing, 301 because of no follow-up PFTs, and 86 SCD children (72 HbSS and 14 HbSC) because they were not treated according to current guidelines recommending routine HU therapy for HbSS, but not HbSC. Overall, 1341 spirometry results from the remaining 408 (229 male, 179 female) patients and 1199 TLC measurements from 377 patients were analyzed. The average observation period was 4.7 (2.7) years. Genotype distribution was 244 for HbSS, 15 HbSβ0, and 149 HbSC; 389 (96%) were of African descent by self-report. All HbSS and HbSβ0, but none of the included HbSC children were treated with HU. For HbSS, the age at inclusion (Baseline) was 11.2 (3.0) years. PFT results were: ppFEV1 74.7 (10.9), FEV1 z-score −2.1 (0.9), ppFVC 76.7 (11.0), FVC z-score −2.0 (1.0), FEV1/FVC% 85.4 (7.4), ppTLC 79.4 (10.3), TLC z-score −1.56 (0.9); Hemoglobin was 9.3 (1.3) g/dL and Hemoglobin F 12.9 (8.1) %. PFTs, as well as total Hb levels, were not different between male and female participants at Baseline, but females had a significantly higher mean (SD) relative (%) HbF compared to male HbSS (14.9 (9.3) vs. 11.5 (6.7); p < .001). Age at Baseline for children with HbSC was 10.5 (2.7) years and average PFT results were within normal limits: ppFEV1 80.7 (9.9), FEV1 z-score −1.6 (0.8), ppFVC 82.2 (9.8), FVC z-score −1.5 (0.85), FEV1/FVC % 86.2 (7.0), ppTLC 85.7 (9.4), TLC z-score −1.1 (0.7), Hb g/dL 11.3 (1.1). HbF levels were not tested routinely in HbSC. The statistical analysis showed no differences in PFTs or Hb levels between male and female HbSC at Baseline. Analysis of the longitudinal PFT data for all HU-treated HbSS revealed a significant annual decline in ppFEV1, ppFVC, and ppTLC (Table 1). There was a sex effect with a more rapid PFT decline in male compared to female HbSS. Similarly, in HbSC, there was a statistically significant but, compared to HbSS less pronounced decline in ppFEV1 (−0.191/year; 95%CI −0.306 to −0.082; p < .001), ppFVC (−0.334; 95%CI −0.451 to −0.215; p < .001) and ppTLC (−0.170; 95%CI −0.385 to −0.126; p < .001). Male sex in HbSC was also associated with a more rapid decline in ppFEV1 compared to female HbSC (−0.196/year; 95%CI −0.349 to −0.040 vs. −0.180/year; 95%CI −0.347 to −0.025; p = .028), ppFVC (−0.449/year; 95%CI −0.590 to −0.296 vs. −0.220/year; 95%CI −0.410 to −0.044; p < .001) and ppTLC (−0.242/year; 95%CI −0.488 to −0.069 vs. −0.105/year; 95%CI −0.358 to −0.005; p < .001). The potential confounders Hb levels and HbF% had no effect on the annual PFT decline for HbSS or HbSC. Post-hoc analysis revealed no difference in PFT decline between subgroups of HbF <8.6% versus HbF ≥8.6% (p = .756) or prepubertal children (age < 12 years) versus adolescents (age ≥ 12 years) (p = .573), while sex disparity was still present in the subgroups. This is the largest analysis of longitudinal pulmonary function data in children with SCD treated according to current guidelines to date. Even though the longitudinal PFT decline was improved compared to an era prior to routine HU use,7 male sex and HbSS genotype continue to be risk factors for more pronounced progression of SCD lung disease. In contrast to this, a recent retrospective study of 62 HbSS children treated with HU reported an improvement in ppFVC.7 The reasons for these discrepancies are not clear, but differences in study design may have contributed, as both earlier studies included only two PFT encounters per patient. For HbSC the annual PFT decline in our current cohort was also less compared to an earlier report from our center on children with SCD not treated with HU,2 likely attributable to improved clinical care of HbSC disease in recent years. The sex disparity observed in our study for both HbSS and HbSC children remains unexplained. Previous works had suggested that higher levels of HbF may protect females from SCD organ manifestations.8 In this cohort, however, PFT decline was not associated with HbF levels or age. Other factors contributing to the observed sex disparity in SCD lung disease progression might include sex hormones, modifier genes, or behavioral characteristics such as non-adherence to therapies. HU has been widely recommended for SCD HbSS, since its approval for SCD by FDA in 1998. In this study, we included data from HbSS only when receiving HU therapy. An unusual benign clinical course, ineligibility for other reasons, or other factors such as non-acceptance might have been reasons why some HbSS children, who were excluded, were not treated according to current guidelines. This may have resulted in a selection bias towards those with more severe disease. Other limitations of this single-center study include the retrospective design. We choose not to explore the potential contribution of VOC or ACS episodes to pulmonary function decline, as these events might have not been accurately charted and mild events not reported to the treating physicians. Also, adherence to therapy was not documented systematically and therefore not included in the analysis. In conclusion, children with sickle cell anemia continue to experience progressive pulmonary function decline despite the use of HU therapy. HbSS genotype and male gender continue to represent risk factors for more severe pulmonary disease despite HU therapy. The author(s) received no financial support for the research, authorship, and/or publication of this article. There are no competing interests for any author. Data are available on reasonable request.