Incidence and Risk Factors for New and Recurrent Infarcts in Adults With Sickle Cell Disease

Abstract

Background

Most adults with sickle cell disease will experience a silent cerebral infarction (SCI) or overt stroke. Identifying patient subgroups with increased stroke incidence is important for future clinical trials focused on stroke prevention. Our 3‐center prospective cohort study tested the primary hypothesis that adults with sickle cell disease and SCIs have a greater incidence of new stroke or SCI compared with those without SCI. A secondary aim focused on identifying additional risk factors for progressive infarcts, particularly traditional risk factors for stroke in adults.

Methods and Results

This observational study included adults with sickle cell disease and no history of stroke. Magnetic resonance imaging scans of the brain completed at baseline and >1 year later were reviewed by 3 radiologists for baseline SCIs and new or progressive infarcts on follow‐up magnetic resonance imaging. Stroke risk factors were abstracted from the medical chart. Time‐to‐event analysis was utilized for progressive infarcts. Median age was 24.1 years; 45.3% of 95 participants had SCIs on baseline magnetic resonance imaging. Progressive infarcts were present in 17 participants (17.9%), and the median follow‐up was 2.1 years. Incidence of new infarcts was 11.95 per 100 patient‐years (6.17–20.88) versus 3.74 per 100 patient‐years (1.21–8.73) in those with versus without prior SCI. Multivariable Cox regression showed that baseline SCI predicts progressive infarcts (hazard ratio, 3.46 [95% CI, 1.05–11.39]; P=0.041); baseline hypertension was also associated with progressive infarcts (hazard ratio, 3.23 [95% CI, 1.16–9.51]; P=0.025).

Conclusions

Selecting individuals with SCIs and hypertension for stroke prevention trials in sickle cell disease may enrich the study population with those at highest risk for infarct recurrence.

Nonstandard Abbreviations and Acronyms

SCD

sickle cell disease

SCI

silent cerebral infarcts

Research Perspective.

What Is New?

• This study identified a new risk factor, hypertension, as well as confirmed a known risk factor, prior silent cerebral infarcts, associated with the development of a new stroke or silent cerebral infarcts in adults with sickle cell disease.

• Traditional stroke risk factors were uncommon in this young adult cohort with sickle cell disease; 1 risk factor (current smoker, hypertension, chronic kidney disease, diabetes, history of deep vein thrombosis or pulmonary embolism) was present in 26% and 2 or more risk factors were present in only 14.7%; thus further study in a larger cohort is necessary.

What Question Should Be Addressed Next?

• Clinical trials focused on secondary prevention of stroke and silent cerebral infarction are the next step in adults with sickle cell disease; patients with prior silent cerebral infarcts and hypertension are a particularly high‐risk group for infarct progression and should be included in trials.

Sickle cell disease (SCD) occurs in an estimated 100 000 children and adults in the United States. ¹ Silent cerebral infarcts (SCI) and overt strokes contribute significantly to morbidity in SCD, resulting in functional impairment, challenges with school and job performance, and premature death. Approximately 60% of adults with the most severe form of SCD, sickle cell anemia, will experience 1 or more cerebral infarctions (≈50% with silent cerebral infarcts and up to 10% with overt strokes) by 30 years of age without primary prevention. ² , ³ The 2020 American Society of Hematology SCD Cerebrovascular guidelines recommend magnetic resonance imaging (MRI) of the brain to screen for SCI at least once in childhood and once in adulthood, though the guidelines do not comment on how often scans should be repeated if SCI are found. ⁴ Six National Institutes of Health–funded randomized controlled trials have identified therapies to prevent SCI and overt strokes in children with sickle cell anemia, including monthly blood transfusion therapy, for preventing initial strokes ⁵ , ⁶ , ⁷ , ⁸ and recurrent SCI, ⁹ and hydroxyurea for preventing initial strokes in low‐ and middle‐income countries. ¹⁰ Despite the observation that at least 95% of children with SCD in high‐income countries reach adulthood, ¹¹ , ¹² no clinical trials have established therapeutic approaches for adults with SCD. For adults with SCD, there are no adequate evidence‐based guidelines for primary stroke prevention strategies.

Secondary stroke prevention strategies in adults with SCD are primarily based on treatment initiated in children with SCIs or strokes that are continued through young adulthood ⁷ , ⁹ and do not take into consideration comorbidities that may influence the clinical outcomes in adults with SCD. Understanding the contribution of established stroke risk factors in the general population, including smoking, hypertension, diabetes, and renal disease, for adults with SCD will provide the requisite data for phase III clinical trials focused on secondary stroke prevention in adults with SCD. Hypertension is the strongest modifiable risk factor for stroke in adults without SCD ¹³ and thus, is a risk factor of particular interest.

We conducted a multicenter prospective observational study to test the primary hypothesis that the incidence of new stroke or SCI would be greater in adults with SCD with SCI versus those without SCI at baseline. A secondary objective was to identify potentially modifiable risk factors for new or recurrent cerebral infarcts.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request. The 3 study sites included Vanderbilt University Medical Center (Nashville, TN), Washington University Medical Center (St. Louis, MO), and the University of Alabama (Birmingham, AL). Site selection was based on the following criteria: 1) estimated number of patients at each site with a history of SCD; 2) presence of an established transition system between pediatric and adult SCD clinics at each site; and 3) prior experience of each site in conducting clinical trials involving patients with SCD. The institutional review boards of all 3 sites approved the study. Written informed consent was obtained from all study participants.

Inclusion criteria consisted of a confirmed diagnosis of sickle cell disease (sickle cell anemia, Sβ0 thalassemia, Sβ+thalassemia, sickle cell hemoglobin C) on hemoglobin analysis or other confirmatory documentation of phenotype; ≥18 years of age, followed regularly (at least 2 visits per year) in the sickle cell centers, demonstrated adherence with follow‐up visits for ≥3 years, and agreement to a standard of care entry/exit MRI/MRA of the brain. Exclusion criteria comprised contraindications to MRI, including individuals with MRI‐incompatible foreign metal objects. For this analysis, participants with a history of overt stroke were excluded because they are known to have a high rate of stroke progression. Childhood transcranial Doppler ultrasound data were typically unavailable and thus were not part of the inclusion or exclusion criteria.

Participants underwent standard of care MRI/MRA of the brain ≥1.5 Tesla strength at the time of study entry (baseline) and the same magnet strength 1 year or more later. If participants received an MRI/MRA scan within 18 months before study enrollment, as per the study protocol and informed consent, this scan was included as the baseline MRI/MRA. Enrollment was from February 2018 to February 2021. We required completion of the second MRI of the brain by September 2022. Participants enrolled in an existing single‐center observational neuroimaging study of adults with SCD were enrolled in this multicenter cohort with their consent. Participants did not receive any specific interventions for stroke prevention per study protocol; care was determined by the treating hematologist at each site. Stroke risk factor data, medical history, and neurological examination data were extracted from the medical chart. Hypertension was defined as a diagnosis of hypertension in the medical chart with treatment provided or offered. Systolic blood pressure obtained at the clinic visit closest to the baseline study MRI was also recorded. At baseline and every 12 months, a brief, stroke‐free questionnaire was administered by phone or completed via chart review to assess for the possibility of any new stroke or focal neurological deficits that may have developed in the interim.

Imaging Protocol

Infarct was defined as MRI of the brain with a T2‐weighted image or fluid attenuation inversion recovery image showing an infarct of at least 3 mm in the axial plane or a similar lesion using 3D imaging of the brain. Coronal imaging, used in pediatric SCD trials, was not required because it is not part of the standard neuroimaging protocol for adults at any of the 3 centers. Time‐of‐flight MRA of the brain was performed. Cervical MRA was not performed.

Radiology Review and Adjudication of MRIs

Two independent radiologists blinded to clinical information adjudicated initial and subsequent MRIs assessing for the primary outcome, new infarcts at least 3 mm in size and progression of previously identified infarcts that increased by at least 3 mm along any linear dimension on MRI when compared with baseline MRI, defined for this article as progressive infarcts. MRA of the brain: each major intracranial artery (the internal carotid, middle cerebral, anterior cerebral and posterior cerebral artery) was graded as follows: normal, stenosis 25% to 49%, 50% to 74%, 75% to 99%, or occlusion. Disagreements were resolved via review by a third radiologist and consensus.

Sample Size Calculation

Based on preliminary, single‐center data with a hazard ratio of 5.27 for recurrent infarcts in those with SCI and a 2‐year follow‐up, we calculated a sample size of 31 participants (at least 12 with a SCI) to achieve 85% power. To allow for imprecision in our estimate and subgroup analysis, our goal was to recruit a minimum of 60 participants from 3 sites, with at least 24 participants with SCI followed for a mean of 2 years.

Statistical analysis was performed using SPSS version 29.0 (IBM, Armonk, NY) by the study statistician (MR). The corresponding author had full access to all the data in the study and takes responsibility for its integrity and the data analysis. Participant characteristics were described by mean and SD for continuous measures, or median and interquartile range, depending on the distribution, and by number and percent for categorical measures. A 2‐sided P value ≤0.05 was considered statistically significant. Time‐to‐event analysis with a log‐rank test assessed those with or without SCI at baseline for a progressive infarct. Cox regression was used to assess the association of age at first MRI, sex, hemoglobin, hypertension, and infarct status at first MRI with a new or progressive infarct. Additional Cox regressions examined other covariates, but the number of participants and patients with progressive infarcts limited the number of covariates that could be included in a model.

Results

We enrolled 116 participants with 2 MRIs completed at least 1 year apart. Twenty‐one participants were excluded from this analysis for a history of baseline overt stroke; therefore, 95 participants were evaluable for this study. Demographics and SCD‐related stroke risk factors are presented in Table 1 and traditional stroke risk factors are presented in Table 2. The median age of participants was 24.1 years (interquartile range, 20.8–29.3); the prevalence of SCIs on baseline MRI was 45.3%. The median number of SCIs per participant among participants with SCI at baseline was 3.0 (interquartile range, 1.0–6.0). As expected, the frequency of SCI was greater in those with sickle cell anemia: 47.7% (41/86) of individuals with sickle cell anemia/Sβ0Thal had SCI at baseline, compared with 22.2% (2/9) with sickle cell hemoglobin C/Hbβ+Thal, but this was not statistically significant (P=0.267). Intracranial stenosis >50% of vessel diameter was found in 4 patients with sickle cell anemia who had an SCI at baseline.

Table 1.

Baseline Characteristics of Participants by Infarct Status (n=95)

Variable	Complete cohort (n=95)	No infarct (n=52)	Silent cerebral infarct (n=43)	P value*
Age, y, median (IQR)	24.1 (20.8–29.3)	23.0 (20.6–28.0)	26.2 (21.7–33.2)	0.030
Sex, female, n (%)	45 (47.4)	26 (50.0)	19 (44.2)	0.572
Phenotype, n (%)				0.177†
SS/ Sβ0Thal	86 (90.5)	45 (86.5)	41 (95.3)
SC/SBThal+	9 (9.5)	7 (13.5)	2 (4.7)
Time from baseline MRI to last MRI, y, median (IQR)	2.1 (1.6–3.3)	2.1 (1.6–3.6)	2.1 (1.5–3.0)	0.335
New infarct or progression of existing infarct	17 (17.9)	5 (9.6)	12 (27.9)	0.021
Intracranial stenosis >50%	4 (4.4)	0 (0.0)	4 (9.3)	0.046†
Hemoglobin, g/dL, mean (SD)	9.4 (1.7)	9.4 (1.7)	9.3 (1.7)	0.778
Hematocrit, %, mean (SD)	27.0 (4.9)	27.0 (4.7)	27.0 (5.3)	0.970
White blood cell count, 109/L, mean (SD)	10.9 (5.0)	10.7 (4.2)	11.1 (5.9)	0.683
Platelets, 103/mm3, mean (SD)	387.6 (170.6)	375.9 (177.2)	401.8 (163.4)	0.464
Reticulocytes, median (IQR) (n=93)	9.4 (5.0–14.2)	9.2 (4.7–13.9)	9.9 (5.4–15.4)	0.490
Absolute neutrophil count, mm3, mean (SD)	5889.9 (3724.5)	5495.5 (2860.3)	6366.7 (4548.2)	0.259
MCV, fL, mean (SD) (n=94)	92.8 (12.6)	93.1 (13.1)	92.4 (12.2)	0.783
Creatinine, mg/dL, mean (SD)	0.61 (0.23)	0.60 (0.20)	0.63 (0.26)	0.423
BMI, kg/m2, mean (SD)	24.0 (4.9)	24.5 (5.7)	23.4 (3.8)	0.279
Systolic blood pressure, mm Hg, mean (SD)	117.7 (12.6)	117.2 (11.8)	118.4 (13.7)	0.665
Diastolic blood pressure, mm Hg, mean (SD)	70.0 (9.9)	69.3 (9.4)	70.9 (10.6)	0.445
Oxygen saturation, %, median (IQR) (n=93)	98.0 (96.0–99.0)	97.0 (96.0–99.0)	100.0 (98.0–100.0)	0.328
Aspirin, n (%)	3 (3.2)	2 (3.8)	1 (2.3)	1.000†
Disease‐modifying treatment, n (%)				0.110
No treatment	13 (13.7)	9 (17.3)	4 (9.3)
Hydroxyurea only	60 (63.2)	35 (67.3)	25 (58.1)
Transfusion, with or without hydroxyurea	22 (23.2)	8 (15.4)	14 (32.6)

BMI indicates body mass index; IQR, interquartile range; MCV, mean corpuscular volume; and MRI, magnetic resonance imaging.

^*χ² test for categorical variables, t test for means, Mann–Whitney test for medians.

^† Fisher exact test. Note that n is provided in the left column when data were missing, and n was <95.

Table 2.

Baseline or New Risk Factors Before Follow‐Up MRI by Infarct Status (n=95)

Variable, n (%)	Complete cohort (n=95)	No infarct (n=52)	Silent cerebral infarct (n=43)	P value*
History of current smoking (n=93)	11 (11.8)	7 (13.7)	4 (9.5)	0.532
History of hypertension	17 (17.9)	7 (13.5)	10 (23.3)	0.215
Antihypertensive medication	7 (7.4)	1 (1.9)	6 (14.0)	0.044†
Chronic kidney disease	9 (9.5)	0 (0.0)	9 (20.9)	<0.001†
History of deep vein thrombosis	12 (12.6)	3 (5.8)	9 (20.9)	0.033†
History of pulmonary embolism	6 (6.0)	3 (5.8)	3 (7.0)	1.00†
History of venous thromboembolism‡	17 (17.9)	5 (9.6)	12 (27.9)	0.021
Diabetes	2 (2.1)	0 (0.0)	2 (4.7)	0.202†
High cholesterol (n = 92)	0 (0.0)	0 (0.0)	0 (0.0)	NA
Number of risk factors				0.015
None of the above risk factors	56 (58.9)	36 (69.2)	20 (46.5)
One of the above risk factors	25 (26.3)	13 (25.0)	12 (27.9)
Two or more of the above risk factors	14 (14.7)	3 (5.8)	11 (25.6)

MRI indicates magnetic resonance imaging; and NA, not applicable.

^*χ² test unless otherwise noted.

^† Fisher exact test.

^‡ Venous thromboembolism includes patients with either pulmonary embolism or deep vein thrombosis.

Progressive infarcts were present in 17 of 95 participants (17.9%) over a median follow‐up of 2.1 years (interquartile range, 1.6–3.3), which included 234 person‐years; 16 participants had a new or enlarged SCI and 1 had an overt stroke (Table 3). This represents an incidence of new infarcts of 11.95 per 100 patient‐years (95% CI, 6.17–20.88) for participants with SCI at baseline compared with 3.74 per 100 patient‐years (95% CI, 1.21–8.73) without prior SCI. The incidence rate ratio between the 2 groups was 3.19 (95% CI, 1.05–11.57), P=0.040. A Kaplan–Meier analysis (Figure) showed a difference in time to progression between those with and without an SCI at baseline (log‐rank test, P=0.022). Of the 4 participants with intracranial stenosis and SCI at baseline, 2 (50%) had progressive infarcts, 1 had an SCI, and 1 had an overt stroke.

Table 3.

Comparison of Participants With and Without Progressive Infarcts (n=95)

Variable	No progressive infarcts (n=78)	Progressive infarcts (n=17)	P value*
Age, y, median (IQR)	23.5 (20.8–28.7)	27.9 (24.2–38.7)	0.033
Sex, female, n (%)	36 (46.2)	9 (52.9)	0.612
Hemoglobin phenotype, SS/ Sβ0Thal, n (%)	71 (91.0)	15 (88.2)	0.661†
Silent cerebral infarct at baseline	31 (39.7)	12 (70.6)	0.021
Hemoglobin, g/dL, mean (SD)	9.4 (1.6)	9.0 (2.0)	0.363
History of current smoking (n=93), n (%)	7 (9.1)	4 (25.0)	0.092
History of hypertension, n (%)	10 (12.8)	7 (41.2)	0.012†
Antihypertensive drug, n (%)	3 (3.8)	4 (23.5)	0.018†
Chronic kidney disease, n (%)	5 (6.4)	4 (23.5)	0.051†
History of deep vein thrombosis, n (%)	8 (10.3)	4 (23.5)	0.218†
History of pulmonary embolism, n (%)	5 (6.4)	1 (5.9)	1.00†
Diabetes, n (%)	1 (1.3)	1 (5.9)	0.327†
Number of risk factors			0.009†
None of the above risk factors, n (%)	51 (65.4)	5 (29.4)
One of the above risk factors, n (%)	19 (24.4)	6 (35.3)
Two or more of the above risk factors, n (%)	8 (10.3)	6 (35.3)

IQR indicates interquartile range.

^*χ² test for categorical variables, t test for means, Mann–Whitney test for medians.

^† Fisher exact test.

Figure . Kaplan–Meier curve shows time to new infarcts by baseline infarct status.

Adults with sickle cell disease and silent cerebral infarcts at baseline (n=52) were more likely to have new infarcts than those with no infarcts at baseline (n=43) at a median of 2.1 years of follow‐up, log‐rank test P=0.022. SCI indicates silent cerebral infarction.

We assessed progressive infarcts by treatment type (transfusion, hydroxyurea, or no treatment): 10/60 (16.7%) progressed on hydroxyurea, 4/22 (18.2%) progressed on transfusion, and 3/13 (23.1%) progressed with no treatment. The majority (63.2%) of the cohort was on hydroxyurea, 23.2% were receiving regular transfusions, and 13.7% were not receiving disease‐modifying therapy. No participants were prescribed voxelotor or crizanlizumab.

On univariable analysis of traditional stroke risk factors (Table 2), only chronic kidney disease was significantly different in those with and without infarcts at baseline (P <0.001). Hypertension was documented in 17 participants, 10 with SCI and 7 without SCI; 12 (71%) of those with a diagnosis of hypertension were receiving an antihypertensive medication, including all patients with chronic kidney disease. Two patients were diagnosed with type 2 diabetes, and both were receiving treatment. No patients had documented hyperlipidemia or were receiving a treatment for dyslipidemia. One of the following traditional risk factors was present in 25 patients (26.3%): current smoker, hypertension, chronic kidney disease, diabetes, deep vein thrombosis, and pulmonary embolism. Fourteen patients (14.7%) had 2 or more of these risk factors: 11 with infarcts (25.6%) and 3 without infarcts (5.8%) at baseline. Notably, of those with infarct progression, 2 or more of these risk factors were present in 35.3% versus 10.3% in the group that did not progress (P=0.009).

We assessed for any site effects in SCI prevalence or progression. The frequency of SCIs at the 3 sites from the highest to the lowest enrolling site was 51.7%, 28.6%, and 57.1%; these differences were not significant (P=0.105). Infarct progression at the sites, in the same order, was 21.7%, 10.7%, and 14.3%, and these differences were not significant (P=0.482).

Our primary model, a multivariable Cox regression with age, sex, hemoglobin, and hypertension, showed that SCI at baseline remains a significant predictor of progressive infarcts with a hazard ratio of 3.46 (95% CI, 1.05–11.39; P=0.041). Hypertension at baseline was also associated with progressive infarcts with a hazard ratio of 3.23 (95% CI, 1.16–9.51; P=0.025) (Table 4). All participants with chronic kidney disease had an SCI at baseline that confounded multivariable modeling.

Table 4.

Primary Cox Regression Model for Progressive Infarcts

Variable	Hazard ratio	95% CI	P value
Age	1.023	0.964–1.087	0.453
Sex, female	1.009	0.346–2.942	0.987
Hemoglobin	0.884	0.679–1.153	0.364
Hypertension	3.320	1.159–9.508	0.025
Silent cerebral infarct at baseline	3.462	1.052–11.392	0.041

As a secondary analysis for data exploration, we ran 2 additional models. A multivariable Cox regression with age, sex, hemoglobin, and presence of 2 or more traditional stroke risk factors (current smoker, hypertension, chronic kidney disease, diabetes, history of deep vein thrombosis or pulmonary embolism) demonstrated that SCI at baseline was no longer significant with a hazard ratio of 2.61 (95% CI, 0.84–8.12, P=0.098), and the presence of multiple stroke risk factors was not statistically significant with a hazard ratio of 3.19 (95% CI, 0.99–10.20, P=0.05) (Table S1). A multivariable Cox regression with age, sex, hemoglobin, and treatment type (none, hydroxyurea, or transfusion) showed that, overall, treatment type was not a predictor of stroke progression (P=0.576) (Table S2). Similarly, there was also no difference in time to infarct progression by treatment type (Figure S1).

Of note, 15 participants met inclusion criteria, were enrolled, but did not complete the second MRI of the brain. These participants were not analyzed with the cohort. The only significant differences in this group were that only 66% had phenotype sickle cell anemia and 7 (46.7%) were not receiving disease‐modifying therapy.

Discussion

In this prospective, multicenter cohort study of young adults with SCD, we report the incidence and risk factors for new SCI among adults with SCD, 45% of whom had SCI at baseline. Similar to prior work in adults with SCD, preexisting SCI was a statistically significant risk factor for new cerebral infarct. ¹⁴ Hypertension was also a risk factor for new SCI in our larger cohort, while older age was not found to predict future infarction.

Some risk factors with high biological plausibility were simply uncommon. For example, intracranial stenosis has been reported in 15% of adults with SCD. ¹⁵ However, many of these adults will have had an overt stroke. In our study population that excluded overt stroke, only 4 participants had intracranial stenosis on MRA, but 4 of 4 with intracranial stenosis >50% had SCI at baseline and 2 had progressive infarcts. Intracranial stenosis is a risk factor for new SCI and stroke.

The contribution of traditional stroke risk factors to the development of new cerebral infarcts in adults with SCD has been underexplored, despite the large impact these risk factors have on stroke and small‐vessel disease progression in adults without SCD. In this relatively young cohort, with a median age of 24 years, stroke risk factors were uncommon. Hypertension was present in 17.9%, and 2 or more risk factors (current smoker, hypertension, chronic kidney disease, diabetes, history of deep vein thrombosis or pulmonary embolism) were present in only 14.7%. When a stroke occurs in a young person, we often suspect a confluence of risk factors as the cause; however, this was not confirmed in our cohort. The presence of 2 or more stroke risk factors was associated with the presence of SCI at baseline (11/43, 25.6%) versus no infarcts (3/52, 5.8%), P=0.015. Number of risk factors was associated with infarct progression on univariable analysis (P=0.009), but this was not confirmed on multivariable analysis. Current American Society of Hematology Guidelines suggest 1 MRI in childhood and 1 MRI in adulthood to assess for SCI ⁴ ; however, patients with multiple stroke risk factors may warrant serial monitoring. In children, SCIs are known to have an impact on cognition and thus are not silent; covert infarcts or infarcts without focal neurological deficits may be more appropriate terminology.

While hypertension was a risk factor for infarct progression in SCD, the measured blood pressure at baseline did not differ between those with and without infarcts. The retrospective nature of this study did not allow for serial, standardized BP measurements. Thus, a diagnosis of hypertension was accepted when coded in the medical chart. However, not all participants identified as having hypertension had an antihypertensive drug on their medication list. BPs are generally believed to be lower in SCD than in the non‐SCD population, ¹⁶ , ¹⁷ and our data support this with relatively low measured baseline BP values. Future studies could help develop BP treatment guidelines in SCD to determine whether hypertension treatment prevents stroke and SCI. All 9 participants with chronic kidney disease had an SCI at study entry; the most likely mechanism of SCI in this scenario is relative hypertension. In SCD, hypertension has been associated with SCI but compared with the general population, BPs in SCD are lower. Thus, hypertension may go unrecognized or be attributed to pain; as a result, treatment may be delayed. Another possible mechanism for infarcts related to chronic kidney disease is more substantial anemia related to the combination of SCD and kidney disease. Our sample size is not large enough to assess whether hypertension, chronic kidney disease, and other risk factors interact to hasten SCI progression, but attention to screening and treatment is warranted.

The rate of new cerebral infarcts in our study, 11.95 events per 100 patient‐years for adults with SCI at baseline versus 3.74 per 100 patient‐years for those without SCI at baseline, is consistent with prior studies and emphasizes that stroke prevention studies and new treatments are needed for adults with SCD, particularly those with SCI. ¹⁴ The lack of a treatment effect for participants on transfusion compared with hydroxyurea was interesting. In children with sickle cell anemia and SCI, the incidence of progressive infarcts is 4.8 per 100 person‐years if untreated, ⁹ 2.9 per 100 patient years when treated with hydroxyurea, ¹⁸ , ¹⁹ and 2.0 per 100 person‐years when treated with regular blood transfusions. ⁹ Small sample size and the observational nature of the study must be taken into consideration. Treatments were reported, but compliance was not monitored closely as part of the study. Additionally, our study spanned the COVID‐19 pandemic when regular transfusions were challenging to complete, and some patients with SCD on transfusion were started on low‐dose hydroxyurea to allow for longer time frames between transfusions. ²⁰

Prospective enrollment, dedicated sickle cell centers, and integrated SCD care from childhood to adulthood to provide consistent disease‐modifying therapy in most patients were strengths of our study. Review of neuroimaging by radiologists familiar with identifying and quantifying silent cerebral infarcts was also a strength. Limitations of this work include a median follow‐up duration of 2.1 years and a sample size of 95 adults, which while larger than many other published adult SCD cohorts followed for new infarcts is still small for detailed analysis when only 17 (17.9%) had new infarcts seen on follow‐up MRI. The requirement for 3 years of regular follow‐up in a dedicated SCD clinic for study inclusion may have biased our results toward a healthier population and could make our results less generalizable. The requirement for 2 MRIs of the brain was also challenging during the pandemic, when many individuals with SCD preferred virtual clinic visits. While venous thromboembolism (deep venous thrombosis and pulmonary embolism) is listed as potential risk factors for stroke, for venous clots to reach the arterial circulation there must be a patent foramen ovale or other right‐to‐left intracardiac or intrapulmonary shunt. Given the uncertain relationship with new infarcts and our limited sample size, these risk factors were not included in our multivariable models. No patients were diagnosed with hypercholesterolemia, and none were on lipid‐lowering medications, but actual lipid profile data are often collected through primary care not SCD clinics and thus were not available.

While the current life expectancy for adults with SCD is a median of 48 years, ²¹ this should continue to increase with time and improved therapies. Traditional stroke risk factors may be more important to consider with advancing age. A recent scientific statement from the American Heart Association highlighted the importance of risk factor control for brain health, ²² while a recent review suggested SCD is a premature aging syndrome. ²³ Presumably, treatment of hypertension and control of stroke risk factors will help prevent new infarcts in adults with SCD. Future clinical trials of secondary stroke prevention in adults with SCD should consider both SCD‐specific therapies combined with treatment for hypertension.

Sources of Funding

L.C.J. was funded by National Institutes of Health (NIH) (K24HL147017). A.L.F. was funded by NIH R01HL129241. A.A.K. received research funding from Global Blood Therapeutics for this research project. J.K. received research funding from Global Blood Therapeutics for this research project. Global Blood Therapeutics had no role in the study design, data analysis, or manuscript preparation.

Disclosures

Dr Jordan receives minor royalties from UpToDate, an evidence‐based clinical information resource, for chapters she has written on stroke in sickle cell disease. Dr DeBaun and his institution sponsored 2 externally funded research investigator‐initiated projects. Global Blood Therapeutics provided funding for the cost of the clinical studies but was not a cosponsor of either study. Dr DeBaun did not receive any compensation for the conduct of these 2 investigator‐initiated observational studies. Dr DeBaun was a member of the Global Blood Therapeutics advisory board for a proposed randomized controlled trial for which he receives compensation, is on the steering committee for a Novartis‐sponsored phase II trial to prevent priapism in men, was a medical advisor in developing the CTX001 Early Economic Model, provided medical input on the economic model as part of an expert reference group for the Vertex/CRISPR CTX001 Early Economic Model in 2020, and consulted for the Forma Pharmaceutical company about sickle cell disease in 2021 and 2022. The remaining authors have no disclosures to report.

Supporting information

Tables S1–S2

Figure S1