Correspondence
Prediction of Adverse Outcomes in Children with Sickle Cell Disease
N Engl J Med 2000; 342:1612-1613May 25, 2000
- Article
To the Editor:
In a well-designed and informative study, Miller and colleagues (Jan. 13 issue)1 demonstrated that among children with sickle cell disease who were younger than two years of age, dactylitis, severe anemia, or an elevated leukocyte count predicted an adverse outcome later in childhood. An adverse outcome was defined as any one of the following: frequent pain, recurrent acute chest syndrome, stroke, or death. Using the three predictors, they developed a prognostic model to identify children who might benefit from high-risk therapies.
We believe the model would be more clinically useful if only death and stroke were included as adverse outcomes. There is much less impetus for primary prevention of the acute chest syndrome, which is rarely fatal in children,2 or recurrent pain. High-risk therapies, such as stem-cell transplantation, could be safely delayed until frequent pain or recurrent acute chest syndrome develops and a trial of hydroxyurea therapy has been initiated, since hydroxyurea has been shown to be effective as prophylaxis against these two complications3 but has not yet been evaluated for the prevention of stroke.
The receiver-operating-characteristic curve shown in Figure 2 of the paper shows the trade-off between sensitivity and the false positive rate (1 – specificity) as a function of the probability of severe disease. The authors suggest a probability of 36 percent as a threshold for classifying children as being at high risk. At this cutoff, the specificity is high (91 percent), but the sensitivity is only 23 percent. This threshold, which minimizes the false positive rate, is appropriate when one is considering patients for high-risk therapies such as stem-cell transplantation, since it reduces the number of children who would unnecessarily receive this therapy. An alternative cutoff might also be used effectively in conjunction with transcranial Doppler ultrasonography to identify children at high risk for stroke. With a cutoff whose purpose is to identify children at high and moderate risk (a threshold of 9 percent), the sensitivity for identifying children who will have adverse outcomes, including stroke, is increased to almost 95 percent, albeit at a reduced specificity. This approach would eliminate the need for ultrasonography in all low-risk children (44 percent of the total group) and thereby reduce costs and improve the positive predictive power of ultrasonography.
3 ReferencesJohn Horan, M.D.
Norma Lerner, M.D.
University of Rochester, Rochester, NY 146421
Miller ST ,Sleeper LA ,Pegelow CH , et al. Prediction of adverse outcomes in children with sickle cell disease. N Engl J Med 2000;342:83-89
Full Text | Web of Science | Medline2
Castro O ,Brambilla DJ ,Thorington B , et al. The acute chest syndrome in sickle cell disease: incidence and risk factors: the Cooperative Study of Sickle Cell Disease. Blood 1994;84:643-649
Web of Science | Medline3
Charache S ,Terrin ML ,Moore RD , et al. Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia. N Engl J Med 1995;332:1317-1322
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We are pleased that Drs. Horan and Lerner have found our prognostic model provocative and potentially useful in clinical practice. The value of the study certainly depends on the acceptance of all our end points as severe, and there was consensus among our authors and several reviewers that this constellation of factors indeed does identify a group with severe disease. These end points are similar to those used by others to justify therapeutic intervention with hydroxyurea or marrow transplantation, and there have been separate analyses of all four of the complications that we chose to indicate the presence of severe disease. In our study patients included in the group with severe disease on the basis of frequent pain and recurrent acute chest syndrome had substantial morbidity (a mean of 41.7 and 16.9 episodes, respectively, during the nine years of follow-up); an increased rate of both these complications has been associated with early death in adult patients with sickle cell disease.1-3
An alternative threshold for classifying children as at high risk might be useful, depending on the intervention being considered; this is why we included the receiver-operating-characteristic curve in the report. A cutoff point of 9 percent would indeed have a high sensitivity — that is, virtually all children destined to have a severe course would be in this group. Other screening tests might then be used to attempt to identify and exclude the lower-risk children.
Currently, transcranial Doppler ultrasonography is the most sensitive way to identify children with sickle cell disease who are at increased risk for stroke. Although our analysis was based on a composite end point, 36 percent of the children with severe disease in our study did indeed have a stroke, and anemia,4,5 an elevated leukocyte count,5 and frequent acute chest syndrome have previously been associated with an increased risk of stroke. A study assessing whether an approach similar to ours but focused solely on the prediction of stroke would enhance the predictive value of transcranial Doppler ultrasonography deserves consideration.
5 ReferencesScott T. Miller, M.D.
State University of New York–Downstate Medical Center, Brooklyn, NY 11203Lynn A. Sleeper, Sc.D.
Laura E. Enos, M.S.
New England Research Institutes, Watertown, MA 024721
Platt OS ,Thorington BD ,Brambilla DJ , et al. Pain in sickle cell disease: rates and risk factors. N Engl J Med 1991;325:11-16
Full Text | Web of Science | Medline2
Castro O ,Brambilla DJ ,Thorington B , et al. The acute chest syndrome in sickle cell disease: incidence and risk factors. Blood 1994;84:643-649
Web of Science | Medline3
Platt OS ,Brambilla DJ ,Rosse WF , et al. Mortality in sickle cell disease: life expectancy and risk factors for early death. N Engl J Med 1994;330:1639-1644
Full Text | Web of Science | Medline4
Ohene-Frempong K ,Weiner SJ ,Sleeper LA , et al. Cerebrovascular accidents in sickle cell disease: rates and risk factors. Blood 1998;91:288-294
Web of Science | Medline5
Balkaran B ,Char G ,Morris JS ,Thomas PW ,Serjeant BE ,Serjeant GR . Stroke in a cohort of patients with homozygous sickle cell disease. J Pediatr 1992;120:360-366
CrossRef | Web of Science | Medline
