Reply: Practice Guideline for Pulmonary Hypertension in Sickle Cell: Direct Evidence Needed before Universal Adoption

From the Authors:

Dr. Hassell and colleagues raise concerns about our recommendations to treat, with hydroxyurea or transfusions, patients with sickle cell disease (SCD) who are at high risk of death based on an elevated tricuspid regurgitant jet velocity (TRV), high levels of serum N-terminal prohormone brain natriuretic peptide (NT-proBNP), or pulmonary hypertension (PH) identified by right heart catheterization (1). Although they point out important caveats concerning the age of screening and risks associated with blood transfusions, we disagree with their positions on hydroxyurea therapy in patients with SCD at the highest risk for death.

Although the use of hydroxyurea and chronic transfusions have not been specifically tested in placebo-controlled trials in patients with SCD with PH, and likely will never be tested based on lack of equipoise, the use of indirect evidence to inform judgments is acceptable and appropriate in guideline development, as long as it is transparently described and the quality of evidence adjusted accordingly. Hydroxyurea decreases the frequency of vasoocclusive events and the acute chest syndrome and improves survival for HbSS patients (2). It is safe and effective even in children with SCD and has been effective in small cohorts of patients with HbSC disease (3, 4). HbSS adults with an elevated TRV, NT-proBNP level, or PH diagnosed by right heart catheterization suffer an increased mortality risk ranging from 3- to 15-fold, established by more than eight independent prospective cohort studies (see full citations in Guidelines document, Reference 1). Furthermore, the risk of death with PH increases during vasoocclusive events and acute chest syndrome (5). Considering the extensive safety and efficacy data supporting hydroxyurea therapy, even in very young children with SCD (4), it is difficult to argue a position that would deny this therapy for the group of patients with SCD at highest risk of death.

Other concerns raised by Dr. Hassell and colleagues about the age of screening, the potential risks associated with chronic transfusion therapy, and potential limitations of single determinations of the levels of NT-proBNP are all important caveats to recognize and consider in judgments regarding intervention in patients with SCD at high risk of death. The literature supporting the use of chronic transfusions in SCD is not nearly as strong as that for hydroxyurea, and this reflects the weak recommendation for consideration of transfusions for only those hydroxyurea-intolerant patients at highest mortality risk. Although hydroxyurea therapy has not been as well studied in other SCD genotypes, including HbSC disease, the mortality risk of an elevated TRV or PH documented by right heart catheterization in these patients is similar to that observed in HbSS disease (6).

As the Guidelines point out, increased mortality risk in SCD is observed only in adults 18 years of age or older with an elevated TRV or NT-proBNP level, so these are the patients who should be considered for intensification of SCD-specific therapies, such as hydroxyurea or chronic transfusions. The optimal frequency of mortality risk assessment in the asymptomatic adults is unknown, but those on our committee advocated performing this every 1 to 3 years in those 18 years of age and older.

Although we have learned much about PH in SCD over the last decade, many questions remain, and their answers will better inform the future versions of these guidelines. The current version reflects the consensus of many individuals from the fields of adult and pediatric pulmonary, cardiology, and hematology, with integration of the available literature with our own clinical experiences. We would welcome working with the authors of the letter to move this field forward and address the major morbidity and mortality risks facing our patients with an elevated TRV, NT-pro-BNP level, and/or PH.