Search for a command to run...
Sickle-cell disease (SCD) has various effects on the cardiovascular system. Pulmonary hypertension (PH) and diastolic dysfunction have been well described in adults, in whom tricuspid regurgitant velocities (TRV) and systolic and diastolic dysfunction measured by echocardiogram are markers of a poor prognosis and are associated with increased morbidity and mortality [1]. Functional alterations have been described on the echocardiograms of children, but their clinical significance and prognostic value remain unclear. The sickle-cell anemia (SCA) genotype subgroup (HbSS and HbSβ0 thalassemia) is associated with greater left ventricular (LV) dilation and hypertrophy beginning early in childhood [2]. The pathogenesis of cardiac alterations in SCA remains unclear but it involves anemia resulting in a high-output state leading to cardiac remodeling and microvascular dysfunction related to repeated vaso-occlusive (VO) events, and nitrite oxide scavenging following chronic intravascular hemolysis. In our recently published regional newborn SCD cohort study, we described substantial persistent morbidity due to acute complications and chronic organ damage in the SCA subgroup, providing a rationale for the earlier introduction of disease-modifying therapies (DMTs) [3]. However, the ability of DMTs to normalize echocardiographic alterations has been little studied in children with SCA. Our main objective was to study the impact of DMTs, such as hydroxyurea (HU), chronic transfusion programs (TPs), and hematopoietic stem-cell transplantation (HSCT), on cardiac functional parameters, using prospectively collected longitudinal echocardiographic data from our SCD cohort. This study was approved by our institutional ethics committee (no. 2021-05-05). We included only children diagnosed with SCD by newborn screening. Echocardiographic parameters, including cardiac index (CI), left heart parameters (LV end-diastolic diameter [LVEDD], LV mass [LVM], LV ejection fraction [LVEF]), left atrial diameter (LAD), and TRV were recorded prospectively during regular check-ups as part of standard patient management, making longitudinal evaluations of echocardiographic changes possible. We standardized the comparison across ages and body surface area by converting LVEDD, LVM, and LAD to Z-scores. We also monitored patient biological and treatment characteristics to assess changes following DMT. In accordance with national guidelines, at our regional center, HU is introduced after the recurrence of VO complications and/or low hemoglobin (Hb) levels, and TP, mainly for stroke prevention. Our center also specifically offers HSCT to patients with cerebral vasculopathy or frequent VO complications with a human leukocyte antigen-identical sibling [3]. Statistical analyses were performed using STATA v18. Echocardiographic parameters are reported as mean ± SD. Genotype groups were compared with Student's t test or Wilcoxon–Mann–Whitney test. Pre- and post-HSCT comparisons used paired t tests or Wilcoxon signed-rank tests, as appropriate. From 1999 to 2020, 511 children aged 1–22 years—408 with severe SCA genotypes and 103 with milder genotypes (HbSC and HbSβ+ thalassemia)—had at least one echocardiogram. In total, 1488 echocardiograms were analyzed. Mean age at first echocardiogram was 6.3 ± 4.1 years (range, 1.0–19.5), with a mean of 2.9 ± 1.8 echocardiograms per child over a period of 4.9 ± 4.4 years (range, 0–16.8). We first investigated the natural course of left heart parameters in patients with SCD, by restricting our analysis to data collected before any DMT. The differences between the SCA and milder subgroups are shown in Figure 1. All the comparisons were statistically significant, indicating that LVEDD, LVM, LAD, and CI were higher in the SCA subgroup, across the pediatric age spectrum. Conversely, LVEF and TRV did not differ significantly between genotype groups (data not shown). LV dilation and mass were significantly greater in the SCA subgroup, with this difference becoming apparent very early, in children as young as 3 years of age, followed by LA dilation, detected by 4 years of age. Our results are consistent with the only available longitudinal analysis, showing that LV dilation, followed by LV hypertrophy, was the first abnormalities observed, with onset at a young age and cumulative incidence progressively increasing during childhood. This study identified the SCA genotype, severe anemia and hemolysis as risk factors associated with the development of echocardiographic abnormalities [2]. We evaluated the impact of therapeutic intensification on echocardiographic parameters by limiting our analysis to children with SCA. We compared values obtained close to the date of treatment modification: the last value obtained before and the first value obtained after treatment intensification. We also evaluated parameters at the last check-up. We investigated the mechanisms contributing to changes in echocardiographic parameters after therapeutic intensification by analyzing Hb, HbF, HbA, and HbS levels, reticulocyte, absolute leukocyte, and neutrophil counts (ALC and ANC), total bilirubin levels and lactate dehydrogenase (LDH) activity immediately before and after therapeutic intensification. The first cardiac assessment after HU initiation (n = 53) revealed no significant change in left heart parameters. At the last check-up, we observed a significant decrease only in LVEDD (median Z-score + 0.8 after versus + 1.9 before). Conversely, after TP (n = 45), all left heart parameters improved significantly: LVEDD (median Z-score +0.9 after versus +1.6 before); LVM (median Z-score +2.3 after versus +4.6 before); and LAD (median Z-score +1.9 after versus +2.6 before) (Table S1A,B). Our data for HU therapy is consistent with two other longitudinal studies showing an improvement in LV dilation after HU initiation [4, 5]. One study evaluating 60 children described, in addition, a reduction of LV hypertrophy after HU initiation [6]. As in our study, a longer duration of treatment with HU was associated with a greater improvement in LV dilation [4]. HU dose was not recorded in any of these studies, including ours. In our study, laboratory tests following the introduction of HU showed significant increases in %HbF and significant decreases in ALC, but the increase in mean corpuscular volume (MCV) was only moderate (mean 87.6 ± 13.7 after vs. 75.3 ± 9.2 fL before HU) suggesting suboptimal dosing. By contrast, Dhar reported an increase in MCV of 20 fL after HU treatment, suggesting a higher dosage [4]. Three cross-sectional studies investigating the impact of TP have yielded conflicting results. One study showed no difference in LV dilation and hypertrophy between children with and without transfusion [2]. Two other studies reported a significant decrease in both LVEDD and LVM in children on TPs relative to age-matched children without transfusions [1, 6]. In a recent longitudinal study in which children were already on a TP at the first echocardiographic assessment, no change was observed over the 2-year study period [5]. In our cohort, 64 children with both pre- and post-HSCT echocardiograms underwent HSCT at a median age of 7 years (range 4.0–19.0 years), mostly from a genoidentical sibling and with the same myeloablative conditioning regimen based on busulfan and cyclophosphamide chemotherapy (Table S2). Following HSCT, successful engraftment was observed in all patients, as shown by the levels of donor-type hemoglobin (median HbS 32%, range [0%–42%]). As expected, Hb levels (median 13.1 g/dL, range [8.8–16.5 g/dL]) normalized after transplantation. All patients underwent a TP before transplantation and most received HU before HSCT. The pretransplantation echocardiographic parameters were therefore collected from patients on HU or TP. We observed a significant decrease in LVEDD, LVM, and LAD between the value recorded immediately before HSCT and that obtained at the first check-up after transplantation (n = 64, mean interval after HSCT: 2.3 ± 1.7 years). This decrease was maintained at the most recent check-up (n = 33, mean interval after HSCT: 10.5 ± 2.6 years) (Table 1). One previous study reported significant improvements in cardiac morphology (LVEDD, LVM, and LAD) in addition to diastolic function and TRV, 2 years after HLA-matched sibling HSCT for SCD. This study differed from ours in several respects (adult patients mean age: 29 vs. 8.7 years in our study), reduced-intensity conditioning (all 123 patients vs. 3/64 in our study), and more severe disease, with a high prevalence of TRV ≥ 2.5 m/s (40% in their study) [7]. In our younger cohort, at baseline, five children had a high TRV ≥ 2.5 m/s (2.7; 2.6 and 3 children with 2.5 m/s). At the most recent check-up after HSCT, maximal TRV values were 2.5 m/s in two children, including the child with a pretransplant TRV of 2.7 m/s and another child with a pretransplant TRV of 2.23 m/s. Finally, we compared posttransplantation echocardiographic parameters with those of age-matched children with HbSC disease, from the ages of 4 to ≥ 16 years. After transplantation, LVEDD, LVM, LAD, and CI decreased to levels similar to those observed in HbSC disease for children of all ages, suggesting that echocardiographic alterations were reversed following HSCT (Table 2). In conclusion, we report the largest prospective longitudinal study of left heart echocardiographic parameters in children with SCD. In the absence of DMT, LV dilation and hypertrophy, and LA dilation are significantly greater in the SCA subgroup than in the subgroup with milder disease. LA dilation has recently been shown to be associated with atrial arrhythmia, a risk factor for stroke in adults, suggesting that LA dilation in children may be predictive of future cardiovascular morbidity [8]. Prospective longitudinal studies from childhood to adulthood are warranted, to confirm that the cardiac changes that develop in childhood are risk factors for overt cardiac disease in adulthood. Both HU and TP significantly improved CI, but the impact of TP on cardiac function was stronger and occurred earlier than that of HU in our cohort. Both treatments similarly increased Hb, but only TP raised HbA, whereas only HU increased HbF and reduced ALC. These results suggest that myocardial oxygen delivery may be less efficient under HU due to persistent abnormal red blood cell rheology, and that higher HbA levels may better prevent SCD-related cardiac remodeling than elevated HbF or reduced leukocyte counts. Interestingly, a longer duration of treatment with HU and higher doses has been shown to be associated with greater cardiac improvement [4]. The absence of information about HU dose at our center is one limitation of this study. An escalation to the maximal tolerated dose (MTD) was recommended from 2015, but was rarely implemented before, as reflected by the moderate myelosuppression observed on HU (mean ANC: 3.5 ± 2.5 x 109/L on HU at last check-up; Table S1A). Further investigations should assess whether earlier HU use, longer treatment, or MTD escalation yield greater cardiac benefits. By contrast, all left heart parameters improved considerably after HSCT. These changes were associated with a normalization of hemoglobin levels and hemolytic parameters. As long-term severe anemia has been shown to lead to cardiac fibrosis, LV dysfunction, and congestion [1], the reversal of echocardiographic alterations after HSCT may prevent overt cardiac disease in SCA patients. Our results further support early genoidentical HSCT in this population [3]. C.R. and C.P. cared for patients, designed the study, collected and interpreted the data, wrote the article, and took final responsibility for the decision to submit for publication. A.L., C.V., and M.-E.L. performed and interpreted echocardiograms. A.K. cared for patients, collected and interpreted data. C.A. cared for patients and collected data. E.B. collected data. S.B. carried out the statistical analysis. B.K. provided biological data and red blood cell units. I.H., J.-H.D., M.F., N.D., C.Pa., and C.D. cared for the patients. All the authors reviewed the paper and approved the final manuscript. We thank Dr. F. Bernaudin for designing the database, collecting data, and caring for study patients. We thank her for her important scientific contribution to the field of sickle cell disease. We thank Dr. F. Lionnet and Dr. N. Hammoudi for their constructive advice throughout the study, and for their contributions to research on the cardiovascular manifestations of sickle-cell disease. We thank all the patients and their families for participating in this study, and all the nurses from the various inpatient and outpatient units for their dedication. We thank Julie Sappa, from Alex Edelman & Associates, for English editing assistance. C.P. reports honoraria and expert consultancy for Theravia. The other authors declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study. Data S1: ajh70163-sup-0001-Supinfo.docx. Table S1: Echocardiographic and biological parameters immediately before and after treatment intensification, (A) with hydroxyurea (HU), or (B) a chronic transfusion program (TP), in patients with the HbSS and HbSβ0 genotypes. Table S2: Characteristics of the patients who underwent hematopoietic stem-cell transplantation (HSCT). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.