Sickle Cell Diastolic Cardiomyopathy and Mortality Risk: A Novel Echocardiographic Framework for Prognostic Stratification

ABSTRACT

Cardiovascular complications are the leading cause of mortality in sickle cell anemia (SCA) patients. While extensive data have identified diastolic dysfunction (DD) to increase morbidity and mortality, the unique hemodynamic conditions inherent to SCA challenge the current recommendations to assess diastolic function. Thus, there is an urgent need to refine the echocardiographic definition of DD to improve risk stratification and therapeutic strategies in SCA patients. We analyzed data from the French multicentric Etendard cohort and compared them with an age‐ and sex‐matched control group from the Copenhagen City Heart Study (CCHS). We focused on left ventricular diastolic parameters, specifically lateral e′ velocity (e′ lat), E/e′ ratio, and indexed left atrial volume (LAVi), assessing their association with clinical outcomes over a 12‐year follow‐up. Etendard SCA patients (n = 379) had an early impaired diastolic function compared to the CCHS controls (n = 672). This was particularly obvious in young SCA patients (n = 252, age ≤ 38 years) in whom e′ lat was associated with prognosis (p = 0.01), with an optimal cut‐off value below 11 cm/s. Indeed, young SCA patients with DD had a fourfold increased 12‐year mortality rate as compared with SCA patients without DD (16 vs. 4%, p < 0.001). Additionally, e′ lat correlated with 6‐min walk test, NT pro‐BNP levels, diastolic blood pressure, and lactate dehydrogenase levels. In young SCA patients, our data contribute to refine the diagnosis of diastolic dysfunction evaluation. We highlight the prognostic value below 11 cm/s of lateral e′ velocity and its association with key contributors of cardiac impairment such as hemolysis and systemic vasculopathy.

Trial Registration: ClinicalTrials.gov identifier: NCT00434902

Abbreviations

A

peak late diastolic velocity

BMI

body mass index

CO

cardiac output

DT

E wave deceleration time

E

peak early diastolic velocity

E′ lat

early diastolic velocity at the lateral site of the mitral annulus

HF

heart failure

LAV_i

left atrium volume indexed to body surface area

LV

left ventricle

LVEDVi

left ventricular end‐diastolic volume indexed to body surface area

LVEF

left ventricular ejection fraction

LVM_i

left ventricular mass indexed to body surface area

mPAP

mean pulmonary artery pressure

PCWP

pulmonary‐capillary wedge pressure

PH

pulmonary hypertension

PVR

pulmonary vascular resistance

RHC

right heart catheterization

TDI

tissue doppler imaging

TRV

tricuspid regurgitation velocity

1. Introduction

During the last decades, major medical advances in the management of homozygous sickle cell disease (SCD) patients—also called sickle cell anemia (SCA) led to a dramatic increase in healthspan in high‐income countries, allowing the vast majority of them to live through adulthood but reaching a life expectancy 15 years inferior to the general population [1, 2]. However, improved longevity of this population unmasked severe cardiovascular complications, such as pulmonary hypertension (PH) and SCA‐related cardiomyopathy, which have become one of the leading causes of mortality, responsible for more than 20% of deaths, along with a dramatic impact on their quality of life [1, 3, 4, 5, 6, 7, 8]. The early diagnosis of SCA cardiomyopathy is challenging since its initial hallmark is the occurrence of a diastolic dysfunction (DD), which is not recognized by applying the current recommendations in assessing diastolic function [9].

Diastolic dysfunction, a prevalent abnormality occurring during the aging process and cardiometabolic disorders (i.e., obesity, diabetes, hypertension), is a powerful prognostic parameter in the general population [10, 11, 12, 13, 14]. Indeed, accumulated age‐ and metabolic‐related damage to cardiac cells is aggravated by abnormal metabolism, inflammation, and senescence, and results in interstitial fibrosis, cardiac hypertrophy, and myocardial stiffness [4, 13, 14, 15, 16, 17, 18, 19, 20].

SCA‐related cardiomyopathy is complex and results from many intertwined mechanisms such as hemolysis‐induced production of reactive oxygen species (ROS) and oxidation of nitric oxide (NO), microvascular vaso‐occlusions, low‐grade inflammation, telomere attrition, and endothelial dysfunction [21, 22, 23, 24, 25, 26, 27]. These processes lead to the early development of myocardial fibrosis, reduced LV relaxation, and ultimately DD [1, 7, 28, 29, 30, 31, 32]. Diastolic dysfunction combined with increased pulmonary vascular resistance promotes the development of PH and has a dramatic impact on SCA patient's prognosis [1, 8, 23, 31, 33].

Although DD has been widely accepted as the main mechanism of heart failure in SCA and reported as very prevalent (15% to 20% of adult patients) [28, 31, 33], its characterization and echocardiographic definition require further refinement. Indeed, previous investigations to detect DD used the established recommendations validated in the general population, which did not include SCA patients [9, 34]. Thus, these recommendations are misleading to assess DD since they do not take into account the impact of chronic anemia on LV chamber dilation and relaxation parameters [31, 35], which establish SCA cardiomyopathy as a unique model of high‐output heart failure (HF) with preserved ejection fraction (EF) [21, 22, 30, 36]. Given the high prevalence of SCA and its poor prognosis in young individuals, it is now urgent to refine echocardiographic criteria of DD, before the onset of heart failure with the aim to identify SCA patients at risk, intensify follow‐up and treatments, and include them in dedicated randomized trials [37].

To this end, the present study leverages data from the SCA Etendard cohort [8, 22], the Copenhagen City Heart Study (CCHS) cohort for control subjects [38] along with the Drepacoeur cohort [5, 39] to develop a tailored approach for assessing diastolic function in individuals with SCA, with the goal of enhancing prognostic stratification.

2. Methods

2.1. Study Population

Etendard Cohort: We conducted a retrospective analysis of a prospective study. Demographic, clinical, and echocardiographic data of the Etendard cohort were analyzed [8]. Briefly, adult patients aged 18 years or older with sickle cell disease were consecutively enrolled from four French Sickle Cell Disease referral centers. All participants were at steady‐state condition: no blood transfusion and no acute episode 3 months prior to inclusion. Patients with end‐stage renal disease (glomerular filtration rate (GFR) below 30 mL/min), chronic restrictive lung disease (defined by a total lung capacity of less than 70% associated with a chest radiographic pattern of interstitial involvement) or severe liver disease with an international normalized ratio greater than 1.7 were excluded. All participants were interviewed, their medical record was examined, and they underwent a physical examination and a 6‐min walking test (6MWT). Height (cm) and mass (kg) were measured, and body mass index was calculated. The history of past complications of SCA was documented, including stroke, vaso‐occlusive crisis, acute chest syndrome, and chronic leg ulcer. Routine laboratory tests (complete blood count, reticulocyte count, serum chemistry profile, and lactate dehydrogenase) were performed using the biochemistry standards of each participating institution. Some limited follow‐up data were recorded at 3 years, such as TRV, 6MWT, and GFR. Written informed consent was collected from all participating patients. The study (ClinicalTrials.gov Identifier: NCT00434902) was approved by a French National Ethics Committee and was conducted in compliance with the Declaration of Helsinki, Good Clinical Practice guidelines, and local laws and regulations.

The Copenhagen City Heart Study to assess diastolic function in control subjects: To compare echocardiographic diastolic function parameters (namely, e′ lateral velocity, E/e′ ratio, and LAVi) between SCA and control groups, we used data from healthy, non‐SCA individuals recruited in the fourth Copenhagen City Heart Study (CCHS), a longitudinal study focusing on cardiovascular disease and associated risk factors within the general Danish population [40]. This control group was matched on age and sex and had no history of organ failure. Hypertension and diabetes mellitus were defined as previously described [41]. They all underwent a baseline echocardiographic evaluation, including diastolic function evaluation. All subjects gave informed consent to participate, and the study was performed in accordance with the second Helsinki Declaration and approved by the regional ethics committee.

2.2. Echocardiography Evaluation

All patients underwent a transthoracic echocardiography performed by a senior cardiologist (Vivid 7 system (GE Healthcare, Norway) and iE33 system (Phillips Medical Systems, Andover, MA, USA)). Procedures to acquire and analyze images were performed as recommended [42]. Left ventricular (LV) and left atrial (LA) volumes were measured in two‐dimensional apical views by the biplane Simpson rule.

Mitral inflow velocity pattern, peak velocities of E and A waves, and E wave deceleration time (DT) were recorded as recommended [9]. Mitral lateral E′ velocities were measured by TDI. Systolic pulmonary arterial pressure was calculated from tricuspid regurgitation flow using the Doppler method. Valvular heart disease severity was graded as recommended [43]. Right ventricle over LV diameter (RV/LV ratio) was measured from the apical 4‐chamber view. All patients with less than 50% of echocardiography parameters available were excluded from analyses.

2.3. Assessment of Long‐Term Mortality in Etendard Cohort

Patients were included from February 2007 through March 2009. Vital status of all patients was obtained from the French national registry (INSEE) until January 2020. When available, the cause of death was obtained from medical records by investigators blinded to clinical data.

2.4. Method for Defining Diastolic Dysfunction in Young SCA Adults

We investigated the value of e′ lateral velocity (a marker of LV relaxation), E/e′ ratio (indicative of LV filling pressure), and left atrial volume index (LAVi, a hallmark of alteration of LV relaxation and stiffness) [9]. To counteract the confounding influence of age on diastolic parameters, we focused on a young subgroup of Etendard cohort (age < 38 years) [9, 44]. We empirically defined this cutoff as the age below which no significant correlation was found between age and both e′ lat and LAVi values (Figure S1). Next, we assessed the prognostic value of these three parameters over a 12‐year follow‐up in the young SCA population to identify the optimal parameter(s) to define early diastolic dysfunction.

Additionally, we analyzed diastolic function parameters as continuous variables and explored their correlation with established markers of cardiovascular impairment, including the 6MWT and NT pro‐BNP, as well as markers of hemolysis severity (LDH), which are often associated with cardiac complications in SCA [35, 45, 46, 47].

Furthermore, we evaluated the three‐year evolution of three available parameters indicative of cardiovascular deterioration: 6MWT, TRV, and GFR according to the presence of a diastolic dysfunction (previously defined) in SCA patients.

2.5. Statistical Analysis

Data are presented as a percentage for categorical data and mean ± SD or median (interquartile range) for continuous variables, depending on the normality of their distribution. Comparisons between groups for continuous variables were performed using one‐way ANOVA. The chi‐square test was used to compare dichotomous variables. Two‐sided Pearson's correlation analyses were conducted when needed. The distribution of all echocardiographic parameters of interest was assessed using the Kolmogorov–Smirnov test. All data that were not normally distributed were log‐transformed. Although most echocardiographic variables had < 2% missing data, TRV had higher missing data rates (8.5%), as described in the literature [48]. To identify the best markers of DD, a multivariate analysis was carried out using logistic regression to determine those independently associated with 12‐year mortality. The area under the receiver operating characteristic curve was then used to determine the optimal diastolic parameter cut‐off value associated with prognosis. Survival analyses based on time‐to‐event data were performed to assess the prognostic value of the previously identified echocardiographic parameter, and an unadjusted cumulative incidence curve was plotted by the Kaplan Meier method for long‐term mortality. A log‐rank test was performed to assess the significance for group comparison, and the absence of age influence was verified using Cox regression. A p value < 0.05 indicated statistical significance. Statistical analyses were performed using SPSS 26 (IBM, Armonk, New York).

2.6. External Validation

To further evaluate the external validation of our findings, the best markers of DD and the optimal diastolic parameter cut‐off value associated with prognosis obtained from the Etendard cohort was then applied to the Drepacoeur registry [5, 39]. Between 2019 and 2023, SCD adults were prospectively included in the Drepacoeur registry for suspected cardiac involvement, as previously described [39]. All patients underwent a comprehensive cardiovascular evaluation in the day‐hospital, including a clinical exam, echocardiography, and a biology workup. Detailed inclusion criteria are available in supplemental data. Written informed consent was collected from all participating patients; the database was declared to the Commission Nationale de l'Informatique et des Libertés (CNIL n°7830264) and was approved by the Ethics Committee (Protocol 2013/NICB).

3. Results

3.1. Population Characteristics

We screened 445 patients for eligibility in the Etendard cohort. None of those patients had any overt cardiac disease such as ischemic cardiomyopathy, significant valvular heart disease, and metabolic‐related cardiac disorder (no diabetes mellitus). We excluded from the analysis 42 patients (29 for severe renal insufficiency, eight for severe lung disease, and five for severe liver disease), five patients for unstable clinical conditions, 13 patients who did not undergo the complete protocol evaluation, and six patients with incomplete echocardiography to assess diastolic function. All but five patients were homozygous for hemoglobin S, the others being S‐β⁰ thalassemia [22]. Baseline characteristics of the final study population (n = 379) are detailed in Table 1. Briefly, mean age was 34 ± 10 years, and 60% were female. Echocardiography found normal LVEF values (65% ± 7%) in most of the patients except two (0.5%) with LVEF ≤ 45%. Left cardiac cavities were moderately dilated, and right ventricular function was preserved.

TABLE 1.

Clinical, biological, and echocardiographic characteristics of study population.

Characteristics	All cohort (n = 379)	Age < 38 ans (n = 252)	Age ≥ 38 ans (n = 127)	p
Demographic, clinical and prognosis		Young group
Age – yr	34 ± 10	28 ± 5	46 ± 6	< 0.001
BMI – mean – kg/m2	22 ± 3	21 ± 3	22 ± 3	0.01
Female, n (%)	228 (60)	144 (57)	84 (66)	0.1
Systolic BP – mmHg	116 ± 15	115 ± 13	119 ± 17	0.006
Diastolic BP – mmHg	66 ± 11	66 ± 11	67 ± 12	0.2
Hydroxyurea therapy, n (%)	101 (26)	70 (28)	31 (24)	0.5
12‐years mortality, n (%)	41 (11)	15 (6)	26 (20)	< 0.001
Coexisting disorders – n (%)
Active smoking – n = 328	37 (10)	29 (12)	8 (6)	0.02
Hypertension	20 (5)	4 (2)	16 (13)	< 0.001
History of vaso‐occlusive crisis a in the year	172 (45)	133 (53)	39 (31)	< 0.001
History of acute chest syndrome – n = 374	241 (64)	159 (63)	82 (65)	0.6
History of stroke – n = 371	19 (5)	11 (4)	9 (7)	0.3
History of leg ulcers – n = 374	66 (18)	23 (9)	43 (34)	< 0.001
Laboratory data
Hemoglobin – g/dL	8.8 ± 1.5	9.0 ± 1.5	8.4 ± 1.4	< 0.001
Platelet count – G/L	390 ± 140	402 ± 143	359 ± 131	0.005
LDH – Multiple of ULN – n = 362	1.9 ± 0.8	1.7 ± 0.7	2.1 ± 0.9	< 0.001
Total Bilirubin – mg/L	57 ± 41	61 ± 43	52 ± 36	0.03
Aspartate aminotransferase – U/L	40 ± 19	38 ± 17	44 ± 22	0.002
GFR b – mL/min	124 (111; 139)	131 (118; 150)	110 (97; 122)	< 0.001
NT‐ProBNP – Log—n = 280	1.8 ± 0.5	1.7 ± 0.4	2 ± 0.6	< 0.001
Echocardiography
LVEF – % ‐ n = 377	65 ± 7	65 ± 7	65 ± 8	0.6
CO – L/min – n = 366	6.3 ± 1.6	6.4 ± 1.6	6.2 ± 1.6	0.5
LVMind – g/m2	111 ± 32	108 ± 28	118 ± 38	0.002
LAVind– mL/m2	41 ± 14	39 ± 12	44 ± 17	< 0.001
E/A ratio	1.7 ± 0.7	1.9 ± 0.8	1.5 ± 0.5	< 0.001
e′ lat—cm/s	14 ± 4	16 ± 4	12 ± 4	< 0.001
E/e′ ratio	7 ± 3	6.4 ± 2.2	8.4 ± 3.4	< 0.001
TAPSE – cm	27 ± 6	27 ± 6	27 ± 5	0.5
TRV – m/s – n = 346	2.3 ± 0.3	2.3 ± 0.3	2.4 ± 0.4	< 0.001

Note: Ordinal data are reported as mean ± standard deviation when they have a normal distribution, or as median (Q1–Q3) when they do not, and nominal data as absolute value and percentage.

Abbreviations: A, mitral inflow late filling velocity; BMI indicates body mass index; CO, cardiac output; E, mitral inflow early filling velocity; e′ lat, lateral early diastolic myocardial velocity; GFR, glomerular filtration rate; LAV_ind, left atrial volume indexed on body surface; LVEF, indicates left ventricular ejection fraction; LVEV_ind, left ventricular end‐diastolic volume indexed to body surface; LVM_ind, left ventricular mass indexed to body surface; MCV, mean corpuscular volume; TAPSE, tricuspid annular plane systolic excursion; TRV, tricuspid regurgitation velocity; ULN, upper limit of the normal range.

^a Vaso‐occlusive crisis was defined as pain related to sickle cell disease that required consultation at the emergency department or hospitalization.

^b GFR estimation was determined using the BSA adjusted CKD‐EPI formula.

Among the CCHS cohort, 672 non‐SCA individuals were matched on age and sex with Etendard patients, with comparable BMI and similar cardiovascular risk (absence of diabetes and dyslipidemia, slightly more active smokers in CCHS group and more controlled systemic hypertension in Etendard cohort, Table S1).

3.2. Etendard Versus CCHS Cohort Diastolic Function Parameters

We compared established and routinely used left ventricular (LV) diastolic parameters (i.e., e′ lat, E/e′ and LAVi) [9]. Notably, the relevance of these parameters was established by the correlations between hemodynamic (postcapillary wedge pressure and mean pulmonary arterial pressure) and echocardiographic data within a subset of Etendard patients who underwent right heart catheterization, as previously published [8, 22].

Compared to controls, SCA patients showed significant differences in LV diastolic function parameters. They displayed a lower e′ lat velocity value (14 ± 4 vs. 18 ± 2 cm/s, p < 0.001), a higher E/e′ ratio (6.7 ± 1.7 vs. 4.8 ± 0.6, p < 0.001) and a larger LA indexed volume (40 ± 14 vs. 23 ± 4 mL/m², p < 0.001) (Figure 1 and Table S1). Both e′ lat and LAVi correlated with age in the two cohorts. However, SCA patients showed a consistent reduction of e′ lat velocity value through age as compared to the expected e′ lat velocity profile in control subjects of the CCHS cohort. Similarly, LAVi was larger even in the younger SCA individuals and increased significantly over time, Figure 1. Strikingly, 14% of the SCA patients showed an e′ lat velocity value <10 cm/s and 62% a LAVi > 34 mL/m², two major indicators of diastolic dysfunction in non‐SCA patients [9, 34], whereas only 1% and 4% of the CCHF cohort met these criteria, respectively (p < 0.001 for both) (Figure 1F).

FIGURE 1.

Comparison of left ventricular diastolic function parameters between Etendard cohort and a matched subgroup of the Copenhagen City Heart Study. Panels A and B show e′ lat and LAVi evolution through age in both cohorts—Blue line refer as the limit of normal value in non‐SCA populations. Panels C–F compare value or distribution of diastolic function parameters across both cohors. *** stands for p < 0.001. [Color figure can be viewed at wileyonlinelibrary.com]

3.3. Focus on a Young SCA Subgroup to Intercept Early Diastolic Dysfunction

We then focused on a subgroup of 252 young SCA patients < 38 years old (67% of the study population, Figure S1). The description of this subgroup of young SCA patients is provided in Table 1 together with the comparison with the older patients. Briefly, mean age was 28 ± 5 years, 57% were females and 28% were under hydroxyurea treatment. Most patients had a history of recent vaso‐occlusive crisis or acute chest syndrome (53% and 63%, respectively). Mean hemoglobin level was 9.0 ± 1.5 g/dL with biological hallmarks of intravascular hemolysis (increased LDH and total bilirubin, Table 1). They had preserved kidney function with a median GFR of 131 [118; 150] mL/min/1.73 m². LV function was normal in all patients (FEVG = 65% ± 7%) with high cardiac output (6.4 ± 1.6 L/min). Regarding diastolic parameters, e′ lat velocity value was in the range of expected normal values (16 ± 4 cm/s) but LA volume was slightly dilated (LAVi = 39 ± 12 mL/m²). Only a few patients showed echocardiographic patterns of PH (TRV = 2.3 ± 0.3 m/s with 4 (2%) ≥ 3 m/s).

3.4. Refining LV Diastolic Dysfunction Based Upon 12‐Year Prognosis in the Young SCA Group

Overall, 15 patients < 38 years (6%) died during follow‐up. Univariate analysis identified higher E/A and E/e′ ratios along with lower e′ lat velocity values in these patients, Table 2. Multivariate analysis identified e′ lat velocity as the only independent diastolic parameter associated with prognosis (p = 0.01, 95% OR = 0.8 [0.72; 0.96]). The optimal e′ lat velocity cut‐off value of 11 cm/s for prognosis stratification was based upon ROC curve analysis and was further used to define diastolic dysfunction in young SCA patients (Se = 89% and Sp = 50%, AUC = 0.66 ‐ 95% CI [0.52; 0.81], p = 0.03), Figure S2. Survival analysis underlined differences between young SCA patients with an e′ lat < 11 cm/s (Log Rang p = 0.02) even after adjusting for age using Cox regression (p = 0.03), Figure 2. Young SCA patients with diastolic dysfunction had a 12‐year mortality rate four times higher (16% vs. 4%, p < 0.001).

TABLE 2.

Association between diastolic function parameters and 12‐year prognosis in young SCA patients.

12‐years prognosis	Alive (n = 216)	Death (n = 15)	p univariate	p multivariate	OR (95%)
Age – years	28 ± 5	28 ± 5	0.91
Hemoglobin – g/dL	9.0 ± 1.5	8.5 ± 1.1	0.23
CO – L/min	6.3 ± 1.6	7.1 ± 1.5	0.57
LVMi – g/m2	107 ± 27	113 ± 42	0.45
E/A ratio	1.8 ± 0.8	2.4 ± 2.3	0.001	0.06
e′ lat – cm/s	16.0 ± 4	13.5 ± 4	0.04	0.01	0.8 (0.72; 0.96)
E/e′ ratio	6.3 ± 2.1	7.7 ± 2.2	0.02	0.6
LAVi – mL/m2	38 ± 12	44 ± 16	0.14
TRV – m/s	2.3 ± 0.3	2.3 ± 0.3	0.64

Note: Ordinal data are reported as mean ± standard deviation.

Abbreviations: A, mitral inflow late filling velocity; CO, indicates cardiac output; E, mitral inflow early filling velocity; e′ lat, lateral early diastolic myocardial velocity; LAV_i, left atrial volume indexed on body surface; LVM_i, left ventricular mass indexed to body surface; TRV, tricuspid regurgitation velocity.

FIGURE 2.

(A) Kaplan–Meier curves with log rank analysis evaluating e′ lat ≤ 11 cm/s impact on 12‐year mortality. (B) 12‐year survival function plots with e′ lateral wave and 95% individual confidence interval. [Color figure can be viewed at wileyonlinelibrary.com]

Comparison of young SCA patients with and without diastolic dysfunction is shown in Table 3. The two groups were comparable in terms of age, sex ratio, and hydroxyurea therapy. Patients with diastolic dysfunction had higher values of LDH (1.7 ± 0.7 vs. 2.0 ± 0.7 multiple of ULN, p = 0.03), GFR (154 ± 28 vs. 130 ± 22 mL/min, p < 0.001) and NT‐Pro‐BNP (80 [46; 160] vs. 50 [23; 99] ng/mL, p = 0.007). The echocardiography showed that young SCA patients with diastolic dysfunction group had also significantly higher values of LVMi and LAVi. Both groups showed comparable and normal LVEF with similar cardiac output and TRV. Interestingly, e′ lat velocity was significantly correlated with diastolic blood pressure, LDH, 6MWT and NT pro‐BNP, Table S3.

TABLE 3.

Comparison of patients depending on the presence of diastolic dysfunction.

Characteristics	Normal diastolic function (n = 217)	Diastolic dysfunction (n = 33)	p
Demographic and clinical
Age – yr	28 ± 5	28 ± 6	0.7
BSA – m2	1.7 ± 0.2	1.7 ± 0.2	0.5
BMI – kg/m2	21 ± 3	22 ± 4	0.4
Female, n (%)	122 (56)	21 (63)	0.4
Systolic blood pressure – mmHg	115 ± 13	117 ± 14	0.4
Diastolic blood pressure – mmHg	65 ± 11	69 ± 10	0.1
Hydroxyurea therapy, n (%)	63 (29)	6 (18)	0.2
6‐min walking test – m – n = 177&13	533 ± 84	502 ± 59	0.1
Coexisting disorders – n (%)
History of VOC in the year	115 (53)	18 (54)	0.4
History of acute chest syndrome	138 (64)	20 (60)	0.7
History of stroke – n = 243	7 (3)	4 (12)	0.04
History of leg ulcers	16 (7)	7 (21)	0.01
Laboratory data
Hemoglobin – g/dL	9.0 ± 1.5	8.6 ± 1.4	0.14
Platelet count – G/L	397 ± 136	44 ± 183	0.1
GFR – mL/min	130 ± 22	154 ± 28	< 0.001
LDH – Multiple of ULN	1.7 ± 0.7	2.0 ± 0.7	0.03
Total Bilirubin – mg/L	59 ± 41	70 ± 53	0.2
Aspartate aminotransferase – U/L	38 ± 17	33 ± 17	0.12
NT‐ProBNP – ng/mL– Log	1.67 ± 0.4	1.93 ± 0.5	0.007
Echocardiography data
LVEF (Simpson Biplane) – %	65 ± 7	65 ± 8	0.7
CO – L/min – n = 218	6.3 ± 1.7	6.8 ± 1.4	0.18
LVMind – g/m2	106 ± 27	117 ± 34	0.04
e′ lat – cm/s	17 ± 4	9 ± 1	< 0.001
E/e′ ratio	6 ± 2	10 ± 2	< 0.001
LAVind – mL/m2	38 ± 12	44 ± 10	0.02
TAPSE – mm	26 ± 6	29 ± 6	0.045
TRV – m/s – n = 225	2.3 ± 0.3	2.2 ± 0.2	0.6

Note: Vaso‐occlusive crisis was defined as pain related to sickle cell disease that required consultation at the emergency department or hospitalization. GFR estimation was determined using the BSA adjusted CKD‐EPI formula. Data are presented mean ± SD.

Abbreviations: A, mitral inflow late filling velocity; BMI, indicates body mass index; CO, cardiac output; E, mitral inflow early filling velocity; E′, lateral early diastolic myocardial velocity; LAVind, left atrial volume indexed on body surface; LVEDind, left ventricular end‐diastolic volume indexed to body surface; LVEF, indicates left ventricular ejection fraction; LVMind, left ventricular mass indexed to body surface; PAcT, pulmonary acceleration time; TAPSE, tricuspid annular plane systolic excursion; TRV, tricuspid regurgitation velocity; ULN, upper limit of the normal range.

Short term (3‐year) follow‐up information regarding 6MWT, GFR, and TRV was available in 187 (74%) young SCA patients and 177 (70%) for TRV. Young SCA patients with diastolic dysfunction showed a 6MWT distance reduction (507 ± 89 vs. 548 ± 92 m, p = 0.04) and a decline of GFR, Figure S3. Interestingly, 3‐year TRV tended to be higher in the young SCA patients with diastolic dysfunction (2.4 ± 0.5 vs. 2.2 ± 0.4 m/s, p = 0.1), Figure S3.

Causes of death were identified in 14 (93%) patients, Table S2. Notably, young SCA patients with diastolic dysfunction died from heart failure, sudden death, or from unknown cause, while the non‐DD group mostly died from non‐cardiac causes.

The same analysis was performed in the older SCA group, and multivariate analysis found that both e′ lat and TRV velocities were independently associated with prognosis, as previously described (data not shown) [7, 23, 31].

3.5. External Validation in Drepacoeur Cohort

To further explore the relevance of our previous findings, the prognostic impact of an e′ lat < 11 cm/s was assessed on an older and more severe population of SCD patients: the Drepacoeur registry. Main characteristics of this study population are detailed in Table S4. Briefly, mean age was 44 ± 12 years, 48% were females and 65% were under hydroxyurea treatment. They were more severe than the Etendard cohort, with 52% having a history of systemic hypertension, 34% chronic kidney disease, and 18% showing a TRV ≥ 3 m/s on echocardiography. Median follow‐up was 3.1 [2.3; 4.1] years, and 12 (8%) patients died. Remarkably, survival analysis underlined differences between SCA patients with an e′ lat < 11 cm/s (Log Rang p < 0.001) even after adjusting for age using Cox regression (p = 0.001), Figure 3. SCA patients with diastolic dysfunction had short‐term mortality rate 12 times higher (2% vs. 24%, p < 0.001). Furthermore, multivariate analysis integrating echocardiography parameters associated with mortality revealed that e′ lat < 11 cm/s was an independent predictor of mortality in this population, even after adjusting for age (data not shown).

FIGURE 3.

Kaplan–Meier curves with log rank analysis evaluating e′ lat ≤ 11 cm/s impact on short term mortality in Drepacoeur cohort. [Color figure can be viewed at wileyonlinelibrary.com]

4. Discussion

This study represents a first attempt to establish a tailored definition of early diastolic cardiomyopathy within a highly vulnerable population of young SCA adults. Our findings indicate that patterns of mitral annular tissue velocity, particularly the e′ lateral velocity—an indicator of early diastolic relaxation—provide valuable insights into diastolic dysfunction in young SCA patients. Notably, an e′ lateral velocity of ≤ 11 cm/s was linked to poorer cardiovascular outcomes over a three‐year period, emerged as a reliable predictor of severe long‐term adverse events, and was further validated in an older and more severe cohort, with a major impact on short‐term mortality. These results underscore the importance of early identification of high‐risk patients before the development of pulmonary hypertension or heart failure. This approach supports the need for enhanced cardiovascular monitoring, tailored therapeutic strategies for SCA, and the inclusion of these patients in dedicated clinical trials [37].

4.1. Uniqueness of Diastolic Function Parameters in SCA–A Premature Aging Trajectory

Diastolic impairment induced by the development of LV fibrosis is a recognized pathophysiological mechanism in SCA and is strongly associated with mortality [29, 31, 49]. However, the definition of early diastolic cardiomyopathy in the specific context of SCA remains completely undefined. Given their young age, unique hemodynamic features, and adaptative cardiac remodeling, the usual diastolic classification validated in the non‐SCA population cannot be applied [9, 34]. Echocardiographic comparison of the Etendard population with a subgroup of the CCHF cohort, matched on age, sex ratio, BMI, and most cardiovascular risk factors, perfectly illustrates the latter consideration. As compared to controls, SCA patients develop early alteration in LV relaxation as well as major LA dilation. Such an early LV relaxation decline is described in other models of premature cardiac aging, which are induced by metabolic disorders such as obesity or diabetes [12, 50, 51, 52]. Premature biological aging shares multiple pathophysiological mechanisms with SCA, such as chronic low‐grade inflammation, increased oxidative stress, induction of premature senescence, and circulation of pro‐fibrotic factors [25, 45, 53]. Overall, we believe that SCA may serve as a model of accelerated cardiovascular aging, opening new avenues for research focused on targeting cellular senescence within this uniquely challenging disease [25, 54].

4.2. From Prognosis Data to a New Definition of DD in Young SCA Patients

While pulmonary hypertension and echocardiographic measurement of TRV have been the core of cardiovascular risk stratification of SCA patients for the last decades, it reflects an advanced stage of the disease often associated with diastolic dysfunction, elevated LV filling pressure, and increased systemic and pulmonary vascular resistance [8, 23, 24, 33, 55].

Thus, our aim was to identify an earlier stage of diastolic dysfunction before the occurrence of pulmonary hypertension. Indeed, diastolic dysfunction starts with LV relaxation impairment, as observed in aging, systemic hypertension, diabetes, or obesity [9, 12, 18, 51]. Therefore, we explored the value of e′ lat and LAV_i, focusing on a young subgroup of SCA patients (≤ 38 years), for whom age is not impacting these parameters. Interestingly, e′ lat velocity was the only early independent prognostic factor. Indeed, an e′ lat velocity value ≤ 11 cm/s was associated with more than a 4‐fold higher 12‐year mortality risk, reaching 16% (mainly from cardiovascular cause) despite their young age, stable condition, and non‐severe phenotype.

Beyond its association with prognosis and to reinforce underlying pathophysiology, e′ lat velocity correlated with both NT pro‐NBP and 6MWT distance in young SCA patients, both hallmarks of cardiac impairment, especially in SCA [8, 46]. In addition, it is very noteworthy to observe that e′ lat velocity was also associated with hemolysis and inflammation markers, along with increased systemic vascular resistance. These results are in line with previous studies underlining the link between hemolysis, systemic vasculopathy, heart fibrosis, and diastolic dysfunction [29, 30, 45, 56] along with their association with mortality [24, 31, 33, 46, 47].

Furthermore, it is interesting to observe that higher TRV values were not associated with mortality in this young subgroup but in the older one. From a pathophysiological point of view, and considering the fact that this is a low‐risk population, it is only logical that LV relaxation impairment, through the development of myocardial fibrosis, should precede LA dilation and the rise of pulmonary pressures [30, 36, 57]. We therefore provide a simple definition of early diastolic cardiomyopathy in young SCA patients, based on tissue Doppler imaging (e′ lat ≤ 11 cm/s), a simple, widely available, and highly reproducible measurement [9, 34].

4.3. Short Term Follow‐Up Data and External Validation to Reinforce Our Findings

Follow‐up information also strengthens our findings since patients with diastolic dysfunction had a reduced 6MWT distance and GFR at the 3‐year follow‐up, as well as a trend in TRV increase. Altogether, the novelty of our study allows the identification of a young SCA population at risk before the development of heart failure. The external validation of this e′ lat velocity cut‐off was further evaluated in another cohort of older and more severe SCD patients (Drepacoeur registry) [5, 39]. Strikingly, an e′ lat velocity value ≤ 11 cm/s was associated with a 12‐fold higher short‐term mortality risk, reaching 24%, even after adjusting for age and other echocardiography factors.

These patients should be carefully monitored by a cardiologist involved in the management of SCA and closely working with SCA experts. The diagnosis of early diastolic dysfunction should be a starting point to reinforce SCA therapies and include these patients in dedicated clinical trials.

4.4. Limitation

This study has some limitations: First, we have selected a study population, which is in stable clinical conditions, without overt organ damage thus not representing the globality of SCA patients. However, the external validation analysis considerably strengthens our findings showing that it also applies to an older and more severe population, even after adjusting for age. Next, we have limited data regarding follow‐up, for only 3 years. More prolonged monitoring would have enabled the observation of the disease progression to consolidate our conclusions. This limitation was in part counterbalanced by a precise 12‐year prognosis status evaluation. Finally, sample sizes are small among subgroups and can impact statistical power and precision. However, the addition of an independent cohort for external validation significantly strengthens the consistency of our findings.

4.5. Perspective

This is the first study paving the way to define early diastolic heart disease in SCA, with echocardiographic data resulting from two of the largest existing cohorts of SCA patients with extensive cardiovascular evaluation. We identified a new category of younger patients at risk, long before the development of PH and heart failure, enabling early cardiac monitoring and SCD therapeutic reinforcement.

Furthermore, our findings highlight the need to better understand early signs of cardiomyopathy in the pediatric SCA population. Currently, cardiac risk stratification tools in children remain limited, with tricuspid regurgitation velocity (TRV) offering only modest utility in this age group [58, 59, 60, 61]. This underscores the importance of longitudinal studies to track individual trajectories of left ventricular relaxation (e′ lat) throughout childhood and adolescence. Such efforts could help identify patients at higher risk earlier in the disease course, potentially guiding timely initiation of disease‐modifying therapies or consideration of allogeneic hematopoietic stem cell transplantation [62, 63, 64, 65].

4.6. Conclusion

In young SCA adults, echocardiographic evaluation of left ventricular relaxation using Doppler tissue imaging allows a simple and early identification of diastolic cardiomyopathy. Lateral e′ wave is associated with hallmarks of cardiac impairment, hemolysis, systemic vasculopathy, and a value below 11 cm/s dramatically increases 12‐year mortality.

Conflicts of Interest

P.B. received grants from ADDMEDICA, Fabre Foundation, NOVARTIS, and Bluebird in the past 36 months—consulting fees for ADDMEDICA, NOVARTIS, ROCHE, GBT, Bluebird, EMMAUS, HEMANEXT, AGIOS, and honoraria for lectures from NOVARTIS, ADDMEDICA, JAZZPHARMA. P.B. is a member of the NOVARTIS steering committee and cofounder of INNOVHEM.

Supporting information

Data S1. Supporting Information.

Simon T., Savale L., Grundtvig Skaarup K., et al., “Sickle Cell Diastolic Cardiomyopathy and Mortality Risk: A Novel Echocardiographic Framework for Prognostic Stratification,” American Journal of Hematology 100, no. 11 (2025): 1940–1951, 10.1002/ajh.27768.

Data Availability Statement

For original data and upon reasonable request, please contact thomas.dhumieres@aphp.fr.