A 3,520-g boy was born at 38 weeks gestation to a 32-year-old G1P1 woman. The mother had a temperature of 101°F at the time of delivery. It was a normal pregnancy, labor, and delivery. The baby was treated with antibiotics because of the mother’s fever. Initial blood count revealed hemoglobin of 18.3 g/dL; white blood cell count of 21,500/mm3 with 43% neutrophils, 4% bands, 4% metamyelocytes, 5% myelocytes, 2% promyelocytes, and 42% lymphocytes; and platelet count of 60,000/mm3. All cultures were negative.
The baby and mother were discharged on the fourth day of life. On day of life 7, the baby was noted to be jaundiced by the pediatrician and was found to have a total bilirubin level of 20.5 mg/dL. The baby had been exclusively breast-feeding without difficulty since discharge. He was referred for evaluation. The repeat complete blood count (CBC) was notable for hemoglobin of 16.3 g/dL; white blood cell count of 16,300/mm3 with 45% lymphocytes, 30% neutrophils, 12% bands, 6% metamyelocytes, 5% myelocytes, and 2% promyelocytes; and platelet count of 55,000/mm3. C-reactive protein measurements on both admissions were normal.
Family history was unremarkable. On examination, the baby was alert and vigorous. Vital signs were unremarkable. The entire physical examination was normal.
Robert Listernick, MD, moderator: Comments?
Aaron Hamvas, MD, neonatologist: Obviously, the first thing we would worry about is an infectious process, even with the original negative cultures. The left shift on the CBC is quite striking. However, the C-reactive protein levels were normal and the baby looks well.
Dr. Listernick: And the persistent thrombocytopenia?
Dr. Hamvas: Once infection has been eliminated as a possibility, I would next consider neonatal alloimmune thrombocytopenia (NAIT). There’s nothing in the history or physical examination to suggest a congenital infection. In addition, the baby doesn’t have any limb abnormalities as might be seen in such syndromes as thrombocytopenia with absent radii.
Dr. Listernick: Should we be worried about bleeding, such as an intraventricular hemorrhage (IVH)?
Dr. Hamvas: This is a nebulous area. Most of the studies, which are mainly retrospective, suggest that the risk of spontaneous IVH increases when the platelet count is below 30,000/mm3. As such, I wouldn’t worry too greatly in this baby.
Dr. Listernick: Let’s talk about NAIT.
Alexis Thompson, MD, pediatric hematologist: In some ways, NAIT is analogous to Rh-associated hemolytic disease of the newborn except that NAIT frequently occurs in the first pregnancy. Essentially, the mother generates antibodies against a paternal-derived antigen on the baby’s platelets, most commonly the human platelet antigen (HPA)-1a.
Dr. Listernick: What is the treatment of NAIT?
Dr. Thompson: As has been said, the risk of bleeding at this level of thrombocytopenia is very low. In this case, we didn’t recommend any treatment. With severe thrombocytopenia, we generally use intravenous immunoglobulin.
Dr. Listernick: I know I’m pretty old, but as a resident I learned that we should transfuse maternal platelets because they don’t have the offending antigen.
Dr. Thompson: That used to be true, but it is extremely cumbersome and may take several days to obtain the maternal platelets, which is too long to be useful. Occasionally, regional blood banks may have a stock of HPA-1a negative platelets that can be used for transfusion.
Dr. Listernick: What’s the natural history of untreated NAIT?
Dr. Thompson: Usually the platelet count returns to normal by age 3 months as the titer of the alloantibodies decline.
Dr. Listernick: Moving forward, the bilirubin on admission declined appropriately. There was no laboratory evidence of a congenital infection. Over the next 4 weeks, the platelet counts hovered between 60,000/mm3 and 80,000/mm3. Physical examination remained normal. At age 3 months he was seen by neurosurgery due to increasing head circumference that progressed from the 65th to the 90th percentile. Weight and length remained on the 50th percentile. Magnetic resonance imaging (MRI) was performed and a diagnosis of benign extra-axial fluid was made.
Jared Green, MD, pediatric radiologist: In the modern era of pediatric neuroimaging, it has been noted that many infants who have rapidly growing heads have dilated subarachnoid spaces. The etiology is not clear, but one theory is that immature arachnoid villi are not able to resorb cerebrospinal fluid (CSF) that is produced continuously.The accumulated CSF expands the subarachnoid space, leading to head growth without increased intracranial pressure due to the open sutures. This is a benign condition that subsides over time.
Dr. Listernick: This occurred as the baby’s platelet count was being followed. He had a repeat MRI 1 month later that was more abnormal.
Dr. Green: The diameter of the extra-axial fluid remained the same, although now the temporal horns of the lateral ventricles were slightly dilated, as was the morphology of the third ventricle. There was concern for developing hydrocephalus, although the etiology was unclear.
Dr. Listernick: Around the same time, in the hematology clinic, he was noted to have mild splenomegaly without hepatomegaly. At that time, CBC demonstrated hemoglobin of 9.2 g/dL; white blood cell count of 16,000/mm3 with 68% lymphocytes, 17% neutrophils, 2% immature neutrophils, 2% basophils, 1% eosinophils, 3% metamyelocytes, 1% myelocytes, and 5 nucleated red blood cells; and platelet count of 125,000/mm3 and reticulocyte count of 3.7%. There always had been concern about the unusual white blood cell differential count, but the focus shifted to the possibility of a storage disease.
Joel Charrow, MD, pediatric geneticist: Storage diseases are certainly worth thinking about in the context of a baby who has splenomegaly and abnormal CBCs. However, in general, it’s lysosomal storage diseases in which we see isolated splenomegaly. Children who have glycogen storage disease should have hepatomegaly as well. Although we sometimes see isolated thrombocytopenia in Gaucher disease, children with storage diseases may have hypersplenism leading to pancytopenia rather than isolated thrombocytopenia. In addition, the marked left shift and all the very immature neutrophils seen in this child’s white blood cell differential count should not be found in storage disorders.
Robert Liem, MD, pediatric hematologist: I basically agree with Joel but I have found that platelets seem to be quite sensitive to the effects of hypersplenism and they may be the first CBC element to become abnormal in such children. The differential count is indeed quite unusual.
Dr. Listernick: Are these diseases on our state newborn screen?
Dr. Charrow: Presently, Illinois screens for five lysosomal storage disorders in the newborn period: Fabry disease, Gaucher disease, type I mucopolysaccharidosis, Niemann Pick disease, and Pompe disease. We perform this screening because we can now offer most of these patients either enzyme replacement therapy or stem cell transplantation.
Dr. Listernick: Gee, you’re taking the fun out of making a clinical diagnosis! OK, moving on, at this point “a diagnostic test was performed.”
Dr. Green: The child underwent a skeletal survey. Almost every visualized bone has a “bone-upon-bone appearance.” For example, looking at the femur, we can’t see the difference between the cortex and the medulla, which should normally have a bright cortical line on the outside with lower density of the medullary cavity. This is the classic appearance of osteopetrosis. In addition, many of the metaphyses are splayed with findings of elongation of the physes and diffuse periostitis, giving the classic appearance of “osteopetrorickets.”
Dr. Listernick: Let’s talk about osteopetrosis.
Dr. Thompson: Osteopetrosis is a disorder of the osteoclasts caused by mutations in one of several different genes. Most of the cases present in infancy and are inherited in an autosomal recessive fashion. It should be remembered that osteoblasts, which are responsible for synthesizing bone matrix, are mesenchymal in origin. However, osteoclasts are hematopoietic in origin and are responsible for bone modeling and remodeling. Mutations in these genes are associated with either an abundance of nonfunctional osteoclasts or a paucity of osteoclasts with impaired differentiation.
Dr. Listernick: How does this lead to the clinical manifestations of osteopetrosis?
Dr. Thompson: Failure of modeling of the bone leads to crowding out of the bone marrow elements and then bone marrow failure. Splenomegaly is the result of extramedullary hematopoiesis. Failure of bone remodeling leads to the persistence of abnormal bone, which causes the neurologic defects seen in osteopetrosis. As the optic foramina and auditory canals get smaller from the persistence of nonremodeled bone, the optic and auditory nerves become damaged. Untreated, these infants become blind and deaf and will likely die from complications of marrow failure.
Dr. Green: In addition, on close examination of a computed tomography scan of the neck, which evaluates bony structures better than MRI, we can visualize a tight craniocervical junction that is leading to the development of noncommunicating hydrocephalous.
Dr. Listernick: To be clear, rickets in osteopetrosis probably occurs due to the inability of the osteoclasts to maintain a normal calcium-phosphorus homeostasis. It’s believed that the serum calcium-phosphorus product is insufficient to mineralize the osteoid, leading to rickets.I know the child received a ventriculoperitoneal shunt to treat the hydrocephalus and he had an ophthalmologic examination in the operating room.
WanWan Xu, MD, pediatric ophthalmologist: Children who have osteopetrosis may lose vision, either from narrowing of the optic foramen leading to optic atrophy or from an associated retinopathy. His optic disks were pink and sharp, but he had a “blonde fundus” indicating that there was less than normal retinal pigment. We sometimes see this in fair-skinned individuals. In addition, his fovea depression was less pronounced than normal. We need to assess his visual acuity when he’s awake using Teller acuity cards. We also are considering an electroretinogram, which would help us to decide whether he has a retinopathy. Visual evoked potentials would help assess optic nerve function.
Dr. Listernick: Can you explain the Teller acuity cards?
Dr. Xu: Teller acuity card testing relies upon the fact that infants preferentially look at patterns of alternating contrast rather than homogeneous areas. These cards have two target areas—one with alternating light and dark stripes and another that is homogeneous. Progressively thinner lines are presented to the child until he or she no longer has a preference between the lines and the homogeneous area. Visual acuity is estimated by the thinnest target lines that the child is able to detect.
Dr. Listernick: So can we offer this child treatment?
Dr. Thompson: Bone marrow examination revealed excess numbers of osteoclasts. Genetic testing will be performed but he almost certainly has one of the variants of autosomal recessive osteopetrosis that leads to nonfunctional osteoclasts. He needs stem cell transplantation as soon as possible.
Reggie Duerst, MD, pediatric oncologist: The role of stem cell transplant is to replace the abnormally functioning osteoclasts with those that function normally, allowing them to remodel bone and open up the marrow space.
Dr. Listernick: The bony overgrowth can actually be reversed?
Dr. Duerst: Yes, but this occurs very slowly. There may be progressive vision or hearing loss until the new osteoclasts are able to do their job. There is some urgency in performing the transplant before the child experiences irreversible neurologic damage.
Dr. Listernick: We’ll keep our fingers crossed. Thanks, everyone.