Predicting Thrombosis and Mortality in Patients With Sickle Cell Disease

Two studies presented at the 2018 ASH Annual Meeting characterized the risk for complications, and offered models for predicting these risks, in patients with sickle cell disease (SCD).

SCD and Thrombotic Risk

First, Andrew Srisuwananukorn, MD, from the University of Illinois (UI) at Chicago, presented findings from an analysis of nearly 1,200 patient records, characterizing the high thrombotic risk among patients with SCD and identifying several laboratory, clinical, and genetic factors that increase this risk.

In the study, data were collected from the electronic medical records of 1,193 patients diagnosed with SCD at UI Health between January 2008 and December 2017.

During a median follow up of 5.6 years (interquartile range = 2.3-9.3 years), 210 patients (17.6%) experienced arterial (n=75) or venous thromboembolic events (VTEs; n=268). The majority of events were deep vein thromboses (DVTs; n=170), and the authors determined that most of the events were catheter-related.

After adjusting for age, sex, and hydroxyurea use, the only factors that were significantly associated with any thrombotic event were:

  • HbSS/Sβ0 thalassemia (odds ratio [OR] = 2.07; p=0.002)
  • higher hospitalization rate (OR=1.07; p<0.001)
  • central venous catheter placement (OR=1.22; p<0.001)
  • baseline creatinine at the initial outpatient visit (OR=1.94; p=0.004)

The authors also focused on VTE, finding similar risk factors, with the addition of:

  • number of pregnancies (OR=1.39; p<0.001)
  • increased systolic blood pressure at the initial outpatient visit (OR=4.91; p=0.04)

Using data from 329 patients for whom genotyping was available, the investigators identified only two genetic risk variants associated with thrombotic events, both in the thrombomodulin protein (OR=1.5; p<0.05 for both).

In the cohort of 174 patients who experienced VTE, 31 (18%) experienced a recurrent VTE in the 10-year time span of this study. “Not surprisingly, most of these patients were started on warfarin (for a duration of 3-6 months), but we were not able to find an association with either type or duration of anticoagulation and risk of recurrent VTE,” Dr. Srisuwananukorn reported.

Although the study involved a large cohort of patients with SCD with a broad age range, the authors noted that the findings are limited by the study’s single-center, retrospective design. Compliance to anticoagulation and thrombotic events occurring outside of the hospital also were difficult to capture, which may have confounded the results.

Dr. Srisuwananukorn noted that researchers are now looking into the risk factors associated with recurrent hospitalizations and other thrombotic complications in patients with SCD.

SCD and Pulmonary Hypertension

Next, in an analysis of long-term data from a cohort study of adults with SCD, researchers confirmed that pulmonary hypertension (PH) is associated with a poor prognosis, identifying a “hyperhemolytic phenotype” of SCD associated with a higher risk of death. The findings were presented by Pablo Bartolucci, MD, PhD, from Henri Mondor Hospital in Creteil, France.

The prospective ETENDARD (Evaluation of the Prevalence of Pulmonary Hypertension in Adult Patients With Sickle Cell Disease) study evaluated the prevalence of PH in 398 outpatients with SCD enrolled between February 2007 and March 2009 at referral centers in France. All patients were homozygous for hemoglobin S or had Sβ0 thalassemia.

In an earlier analysis, researchers estimated the prevalence of PH related to SCD to be 6 percent, and the risk for developing PH was greater in patients who had a tricuspid regurgitation velocity (TRV) ≥2.5 m/s and a mean pulmonary arterial pressure (mPAP) ≥25 mmHg (as measured by right-sided heart catheterization; RHC).

To determine the long-term prognostic value of TRV ≥2.5 m/s and the level of mPAP, the authors analyzed 10-year follow-up data from the ETENDARD cohort. Per study protocol, participants underwent Doppler echocardiography to determine TRV; those with suspected PH (defined as TRV ≥2.5 m/s) then underwent RHC to determine mPAP and confirm PH diagnosis (defined as mPAP ≥25 mmHg).

A total of 289 patients had no suspected PH (group A); among the remaining 109 patients, 98 underwent RHC. Of these, 74 had suspected PH (TRV ≥2.5 m/s and mPAP <25 mmHg; group B) and 24 had confirmed PH (TRV ≥2.5 m/s and mPAP ≥25 mmHg; group C). Dr. Bartolucci noted that the rate of confirmed PH was 6 percent, which was consistent with the earlier analysis.

During a median follow-up of 8.8 years (range not reported), 25 patients were lost to follow-up. Thirty-nine of the remaining patients died, and the mortality rate was highest in patients with confirmed PH:

  • group A: 21/269 (8%)
  • group B: 7/70 (10%)
  • group C: 9/23 (39%)

“While an isolated TRV ≥2.5 m/s on transthoracic echocardiography was found to be a moderate prognostic factor for overall survival,” the authors noted, “our findings reveal highly contrasted outcomes depending on the mPAP values measured by RHC.”

Overall survival was significantly lower in patients with a TRV ≥2.5 m/s (hazard ratio [HR] = 2.5; 95% CI 1.3-4.7; p=0.006 for groups B and C vs. group A). After accounting for mPAP levels, though, there was no statistically significant difference between patients with a TRV <2.5 m/s (group A) and those with TRV ≥2.5 m/s and mPAP <25 mmHg (group B; p=0.54).

Looking at other clinical parameters, the authors found that biologic markers of chronic organ damage, when combined with measures of PH, were associated with a higher risk of death:

  • higher systolic blood pressure (HR=1.03; 95% CI 1.01-1.05; p=0.002)
  • lower estimated glomerular filtration rate (HR=0.99; 95% CI 0.98-1.00; p=0.048)
  • higher lactate dehydrogenase levels (HR=1.48; 95% CI 1.05-2.07; p=0.025)

“Surprisingly, there was no difference in risk of death according to leukocyte count or percentage of hemoglobin F,” Dr. Bartolucci said. There also was no difference in mortality risk according to sex, history of veno-occlusive crises, or tobacco use.

Taking these data together, the authors created a “hyperhemolytic phenotype” of SCD: Patients with pre- and post-capillary PH, leg ulcers, chronic kidney disease, and higher systolic blood pressure have a poorer chance of survival.

Dr. Bartolucci noted that these results need to be replicated in other cohorts, and that future studies should focus on the role of transfusion in patients with higher-risk SCD, according to the factors identified here.

The authors of the first study report no relevant conflicts of interest. The authors of the ETENDARD study report relationships with Novartis, Addmedica, and Global Blood Therapeutics.


Srisuwananukorn A, Raslan R, Zhang X, et al. Clinical, laboratory, and genetic risk factors for thrombosis in sickle cell disease. Abstract #9. Presented at the 2018 ASH Annual Meeting, December 1, 2018; San Diego, CA.

Savale L, Loko G, Lionnet F, et al. Predictive factors for survival in sickle cell disease: a cohort study using Etendard data. Abstract #7. Presented at the 2018 ASH Annual Meeting, December 1, 2018; San Diego, CA.