Diagnosis of hereditary spherocytosis in the neonate is difficult. Differences in neonatal erythropoiesis, properties of neonatal erythrocytes, and both clinical and laboratory presentation contribute to complexity in diagnosis. Numerous algorithms to aid in diagnosis of neonatal hereditary spherocytosis (HS) have been developed. One of these, the HS index, the mean corpuscular hemoglobin concentration (MCHC)/mean corpuscular volume (MCV) ratio, revealed an index >0.36 was 97% sensitive and >99% specific in diagnosis of dominantly inherited HS in the neonate.1
In this issue of Pediatrics, Weiss et al examine the utility of the HS index in a diverse population of neonates using data extracted from 15 Kaiser Permanente Hospitals in northern California.2 The rationale for this study was to compare results to the original HS index study, which was performed in a primarily white population, and secondarily to determine if the index discriminates with serum bilirubin levels.
Study eligibility was determined by having had a complete blood cell count (CBC) drawn in the first 7 days of life. This CBC was used to calculate the HS index, as opposed to the original study that included all CBCs on HS infants drawn in the first 90 days of life.1 HS infants were identified by International Classification of Diseases, Ninth Revision and 10th Revision codes. Charts of infants who had osmotic fragility, ektacytometry, or eosin-5'-maleimide (EMA) testing performed were also reviewed. Ektacytometry, only available in specialized laboratories, enables diagnosis of various red cell disorders including HS by measuring erythrocyte deformability as a continuous function of osmolality using a laser-diffraction viscometer. EMA is a flow cytometry–based test that measures binding of fluorescently labeled dye to band 3 and other erythrocyte membrane proteins, providing information on membrane protein deficiency as occurs in HS. Total serum bilirubin levels were obtained in the first 30 days of life.
HS prevalence was higher in white and “other” race and/or ethnicities and, not surprisingly, in patients with increased total serum bilirubin levels. HS risk was higher in infants needing phototherapy, requiring readmission for phototherapy, total serum bilirubin levels crossing exchange transfusion threshold, and requiring an exchange transfusion or a red blood cell transfusion in the first month of life. Dichotomized at 0.36, the HS index was 56% sensitive and 93% specific.
Why is HS so difficult to diagnose in the neonate? In older children and adults, diagnosing HS is straightforward with positive family history, splenomegaly, icterus, anemia, reticulocytosis, elevated MCHC, and spherocytosis. However, in neonates, a high index of suspicion is required as presentation is different. Up to a third of HS patients exhibit nondominant inheritance and these often present in the neonatal period.3 Neonates with recessively inherited HS due to α-spectrin defects typically present with marked hemolytic anemia.4 Patients with HS with de novo mutations, 21% of cases in a recent survey,5 often present in the neonatal history without associated family history. This is primarily due to spontaneous mutations in ankyrin and β spectrin, both encoded by large genes with high CpG content, a finding often associated with high de novo mutation rates.6 CpG dinucleotides are sites in DNA where a cytosine nucleotide is followed by a guanine nucleotide that may undergo methylation followed by spontaneous deamination, resulting in a cytosine-to-thymine mutation.
Jaundice is the most common finding in neonates with HS, often requiring phototherapy and sometimes exchange transfusion. Compared with older children and adults, splenomegaly is uncommon.7 Anemia is present in slightly less than half of neonates with HS, often not occurring until the first few weeks of life, but then at a degree of severity necessitating transfusion.8 When present, there is reticulocytopenia relative to the degree of anemia attributed to the relatively low erythropoietic response observed in neonates.8 In some cases, spherocytes are not present on peripheral blood smear. Elevated MCHC is often found in neonates with HS, whereas reported neonatal MCV values have been inconsistent.
Other commonly used HS diagnostic tools (osmotic fragility, ektacytometry, and EMA) all have pitfalls when used in the neonate. Osmotic fragility suffers from lack of specificity, clearly seen in the neonate where an abnormal test cannot discriminate the spherocytes of HS from the spherocytes seen in ABO incompatibility. Also, because of the increased osmotic fragility of neonatal erythrocytes, it is recommended neonatal osmotic fragility curves be used rather than adult curves.9 Similarly, using neonatal samples as normal controls has been considered for ektacytometry and EMA testing because both exhibit different patterns than adult erythrocytes commonly used as controls.10 Diagnostic gene panels continue to gain in popularity, and many clinicians use them as their “go-to” diagnostic test, especially when there are impediments to obtaining osmotic fragility, ektacytometry, or EMA binding testing.11 Although gene panels have limitations such as difficulty in interpretation of variants of uncertain significance and inability to detect heterozygous deletions, they have the ability to diagnose other inherited erythrocyte disorders, such as pyruvate kinase deficiency, when clinical and laboratory findings do not indicate the diagnosis.
This study brings to mind the Mentzer index (MCV/red blood cell) used in discrimination of iron deficiency anemia and thalassemia trait.12 An index of <13 suggests thalassemia trait whereas >13 suggests iron deficiency.13 When carefully scrutinized in varying populations, its utility has come into question.14 However, it is still often used as an adjunctive tool, with user understanding it does not replace evaluation of iron status and measurement of hemoglobin A(2) to differentiate iron deficient patients from those with thalassemia trait.
How should we use the HS index? The utility in the HS index is primarily as an adjunctive bedside tool to help the clinician formulate a differential diagnosis and move forward with additional diagnostic evaluation as indicated. It is not a definitive diagnostic tool. The study by Weiss et al provides valuable information indicating the index may perform less well in nonwhite patients.
The authors end with the cautionary note “RBC indices may vary slightly on the basis of the make and model of hematology analyzer used, and further investigation regarding the typical ranges of HS Index and RDW values among HS neonates using other cell counters may be warranted.” This important comment leads us to a look into the future. As cell analyzers become more and more sophisticated and provide more detailed information on populations of erythrocytes, the dream of making the diagnosis of HS from analyzing parameters of a CBC comes closer to reality.
FUNDING: No external funding.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2020-021642.
complete blood cell count
mean corpuscular hemoglobin concentration
mean corpuscular volume
POTENTIAL CONFLICT OF INTEREST: The author has indicated he has no potential conflicts of interest relevant to this article to disclose.
FINANCIAL DISCLOSURE: The author has indicated he has no financial relationships relevant to this article to disclose.