How to cite this article: Henry E, Walker D, Wiedmeier SE, Christensen RD. 2007. Hematological abnormalities during the first week of life among neonates with Down syndrome: Data from a multihospital healthcare system. Am J Med Genet Part A 143A:42–50.
Hematological abnormalities during the first week of life among neonates with Down syndrome: Data from a multihospital healthcare system†
Version of Record online: 12 DEC 2006
Copyright © 2006 Wiley-Liss, Inc.
American Journal of Medical Genetics Part A
Volume 143A, Issue 1, pages 42–50, 1 January 2007
How to Cite
Henry, E., Walker, D., Wiedmeier, S.E. and Christensen, R.D. (2007), Hematological abnormalities during the first week of life among neonates with Down syndrome: Data from a multihospital healthcare system. Am. J. Med. Genet., 143A: 42–50. doi: 10.1002/ajmg.a.31442
- Issue online: 21 DEC 2006
- Version of Record online: 12 DEC 2006
- Manuscript Accepted: 18 JUL 2006
- Manuscript Received: 15 MAY 2006
- complete blood count;
- Down syndrome;
Various hematological abnormalities have been reported among neonates with Down syndrome. Thrombocytosis, thrombocytopenia, polycythemia, neutrophilia, transient myeloproliferative disorder (TMD), and congenital leukemia have all been reported. The two largest case series previously reported involved 63 and 31 cases. To acquire hematological data from a larger case series, we obtained all CBCs done during the first week after birth on all neonates with Down syndrome cared for in an Intermountain Healthcare (IHC) hospital with a date of birth between January 1, 2001 and December 31, 2005. During this period, 145,522 live births were recorded at 18 hospitals. Down syndrome was recognized in 226 (1 in 644). One hundred fifty-eight (70%) of these had one or more CBCs obtained before the seventh day (144 hr). Neonates who did versus did not have a CBC in the first week had a similar gestational age, birth weight, percentage who were LGA and SGA, and length of stay. Neutrophilia was the most common hematological abnormality detected, with 80% of absolute neutrophil counts above the upper limit of normal for age. Six percent (9/158) had blasts identified on the blood film and three, where this was persistent, were referred to the pediatric hematology service for further evaluation. The next most commonly detected abnormality was thrombocytopenia, with 66% of platelet counts <150,000/µl, and with 6% of counts <50,000/µl. The mean platelet volume did not correlate with the platelet count, but tended to run slightly large (9.2 ± 1.3 fl), with 24% of values above 10 fl. Only one had a platelet transfusion. Polycythemia was the next most common hematological abnormality detected, with 33% of hematocrit values above 65% or hemoglobin concentrations above 22 g/dl. Six had a reduction transfusion. One patient had significant anemia (hematocrit <15%) and received an erythrocyte transfusion. One had neutropenia associated with an infection after bowel surgery. Neutrophilia, thrombocytopenia, and polycythemia were the most common hematological abnormalities observed among neonates with Down syndrome. Anemia, thrombocytosis, and neutropenia were not more common than among neonates who do not have Down syndrome. Hematological abnormalities were so common in this group that it seems reasonable to recommend that one or more CBCs be obtained on all neonates with Down syndrome. © 2006 Wiley-Liss, Inc.
Down syndrome occurs in one of 600–800 births, or in 5000–6000 neonates annually in the USA [Centers for Disease Control and Prevention (CDC, 2006)]. Various hematological abnormalities have been reported to be present, during the first days of life, among neonates with Down syndrome. The most common of these are polycythemia, thrombocytopenia, and neutrophilia [Miller and Cosgriff, 1983; Hord et al., 1995; Sola and Christensen, 1999; Luchtman-Jones et al., 2005; Dixon et al., 2006]. However, the expected incidence of these abnormalities in neonates with Down syndrome is unclear. The largest series assessing hematological abnormalities in neonates with Down syndrome was reported by Miller and Cosgriff 1983 and involved 61 neonates. The largest series reporting the incidence of thrombocytopenia among neonates with Down syndrome was by Hord et al. 1995 and involved 25 neonates.
Intermountain Healthcare (IHC) is a not for profit corporation operating 18 hospitals in the western United States. About 30,000 births occur annually in the IHC hospitals. Devlin and Morrison 2004 recently reported that 90% of neonates with Down syndrome are identified during the initial hospitalization. On that basis we estimated that 35–45 neonates/year in the IHC system would be identified, before the initial hospital discharge, as having Down syndrome. We speculated that 5 years of IHC archived records would contain about 200 neonates with Down syndrome, and that a report of their hematological findings would constitute the largest series to date dealing with these issues.
Data were collected as a deidentified limited data set from archived IHC electronic records of all newborn infants where the diagnosis of Down syndrome was made either prenatally or during the first month after birth. Records were included if a karyotype analysis confirmed the diagnosis of trisomy 21, translocation trisomy 21, or mosaic trisomy 21. When Down syndrome was suspected in the newborn nursery, and later confirmed with a karyotype, those CBCs were included in the study. When Down syndrome was suspected in the newborn nursery but the karyotype was later found to be normal, those CBCs were not included.
The information collected for analysis was limited to demographics and the data contained in the CBCs and differential cell counts. Data were collected only for CBCs obtained <7 days after birth, and only on neonates with a date of birth from January 1, 2001 through December 31, 2005, who within the first month of life had a karyotype analysis confirming the diagnosis of Down syndrome. The program used for data collection is a modified subsystem of “clinical workstation.” The 3M Company (Minneapolis, MN) approved the structure and definitions of all data points for use within the program. The data were collected from the electronic medical record, case mix, and laboratory systems. Trained and designated clinical personnel enter additional data. Data are managed and accessed by authorized data analysts. The Intermountain Healthcare Institutional Review Board approved this study.
Between January 1, 2001 and December 31, 2005, 145,522 live births were recorded at the I8 IHC hospitals. During this period 226 neonates were identified as having Down syndrome (prevalence estimate, 1/644). Of these, 158 had one or more CBCs obtained during the first 144 hr after birth and recorded in the IHC electronic database. Karyotype analysis of these 158 neonates included; 142 with trisomy 21, three with translocation trisomy 21, and one with mosaic trisomy 21. The chromosomal abnormality of the remaining 12 was not specified in the electronic records, although all were stated to have “trisomy 21.” All 158 lived through the first week.
Demographic features of the 226 neonates with Down syndrome are listed in Table I. The group of 158 who had a CBC obtained during the first week, and the group of 68 who did not, were of similar gestational age, birth weight, LGA/SGA proportion, gender, and length of hospital stay. The number of CBCs obtained from each of the 158 neonates during the first week after birth ranged from 1 to 8 (median, 3). The length of initial hospitalization ranged from 1 to 77 days (median 7, mean 9.4). Five had initial hospitalizations of 50 days or more. One of these had duodenal atresia and pulmonary hypertension (62 days), one had obstructive uropathy and congenital heart disease (58 days), one had pulmonary hypertension and chylothorax and was treated with ECMO (56 days). Two had congenital heart disease and imperforate anus; one was hospitalized 56 days and the other 77 days.
|Feature||Neonates with Down syndrome (n = 226)||Neonates with Down syndrome who had one or more CBC's during the first week (n = 158)||Neonates with Down syndrome who had no CBC's during the first week (n = 68)|
|Gestational age (weeks) ± SD||37.4 ± 1.9||37.4 ± 1.9||37.3 ± 1.7|
|Birth weight (g) mean ± SD||2894 ± 618||2885 ± 638||2912 ± 573|
|Percent who were LGAa||3% (7/226)||3% (5/158)||3% (2/68)|
|Percent who were SGAb||29% (65/226)||30% (48/158)||25% (17/68)|
|Gender (% male)||58%||61%||53%|
|Length of initial hospital stay (days) (median and range)||5, 1–77||5, 1–77||5, 1–53|
Blood neutrophil concentrations of the 158 who had a CBC obtained are shown in Figure 1. Using the criteria of Manroe et al. 1979, 80% of the blood neutrophil counts fell in the neutrophilic range, 16% in the normal range, and 4% in the neutropenic range. Only one patient had a neutrophil count <1,000/µl. That patient had Hirschprung disease with a bowel perforation and peritonitis on the fifth day of life. Blasts were identified on the differential count of nine neonates. These nine had a range of 1–11 CBCs performed during the first 144 hr of life. The percentage of blasts on the differential counts ranged from 2% to 82%. Three were referred to pediatric hematology for evaluation and follow-up.
Platelet counts obtained on the 158 neonates are shown in Figure 2. Five hundred fifty-six platelet counts were obtained, with values ranging from 20,000 to 490,000/µl. Sixty-six percent of the counts were <150,000/µl. Thirty-three were in the “mild thrombocytopenia” range (101,000–149,000/µl), 27% were in the “moderately severe thrombocytopenia” range (50,000–100,000/µl) and 6% were in the “severe thrombocytopenia” range (<50,000/µl). The mean platelet volume did not correlate with platelet count (Fig. 2), but tended to be slightly large (9.15 ± 1.3 fl), with 24% of values above 10 fl, and only 5% of values below 7.5 fl. Only one patient received platelet transfusions. This neonate was delivered at 34 weeks gestation, had transient myeloproliferative disorder (TMD), and received four platelet transfusions during the first 2 weeks of life because of persistent thrombocytopenia of undetermined etiology.
Hematocrit and hemoglobin values are shown in Figure 3 (panels A and B). One neonate had severe anemia recognized on the first day, explained by gastrointestinal bleeding. The patient received erythrocyte transfusions and the problem did not recur. Fourteen percent of hematocrits were above 70% or hemoglobins were above 23 g/dl. Thirty-three percent of hematocrits were above 65% or hemoglobin above 22 g/dl. Six patients underwent a reduction transfusion. Nucleated erythrocyte counts (Fig. 3, panel C) ranged from 0 to about 500/µl (Fig. 3, panel C). By 3 days (72 hr) all RNBC concentrations were less than 100/µl. The mean corpuscular volume (MCV, Fig. 3, panel D) (110 ± 7 fl) was similar to expected (108 ± 8 fl, [Christensen, 1999]).
Correlations were sought, to determine whether the neonates with one hematological abnormality were more likely to have a second or third. If thrombocytopenia was present, the odds ratio for having polycythemia was 3.39 (95% confidence limits, 1.50–7.67). No correlation was found between the presence of neutrophilia and either thrombocytopenia or polycythemia.
When a neonate is recognized as having Down syndrome, or is suspected of having Down syndrome, certain medical issues arise. These include the possibility of congenital heart malformations, gastrointestinal tract anomalies, and hematological abnormalities [Jones, 2005]. The incidence of the various hematological abnormalities associated with Down syndrome, and the likelihood that any of these will constitute clinically significant problems during the first days of life, has not been clearly delineated. Although a number of reports have been published describing hematological abnormalities in neonates with Down syndrome, these have been based on rather small series of cases. We judged that it would facilitate expectant management of neonates with Down syndrome, during their initial hospitalization, if physicians caring for them had information on the frequency and severity of the most common hematological abnormalities. To obtain this information we used electronically archived data from an 18 hospital healthcare system to retrieve all CBCs obtained during the first days of life of 158 consecutive neonates with Down syndrome.
The most common hematological abnormality we detected was neutrophilia, which was present in the majority of cases (Table II). The upper limit of normal for blood neutrophil concentration during the first week of life was defined by Manroe et al. 1979. Although this reference is over two decades old, the upper limit for neutrophil concentration among neonates has not likely changed in this interval. In fact, subsequent studies by the same group [Mouzinho et al., 1994; Engle et al., 1997] included additional cases and involved additional analyses, but found the upper limit of normal for blood neutrophil concentration in the first week of life to be that we used in this study, namely the upper limit as defined in the original Manroe study [Manroe et al., 1979].
|Hematological problem||Proportion with this problem recognized||Proportion receiving treatment or specific, extended evaluation for this problem*|
In 6% of cases blasts were identified on the blood film. In three the blast count increased over the first days and the patients were referred to pediatric hematology for further evaluation, and were found to have transient TMD. Hitzler and Zipursky 2005 noted that about 10% of neonates with Down syndrome have TMD, which has sometimes been termed “transient leukaemia.” TMD in neonates with Down syndrome is characterized by circulating blasts with morphologic and phenotypic features of leukemic cells [Zipursky et al., 1997]. Although this condition is transient, a low percentage develops severe or even fatal complications, such as hepatic fibrosis or hydrops [Suda et al., 1987; Zipursky et al., 1997; Bonno et al., 1998; Schwab et al., 1998]. In addition, about 20% of patients who recover from TMD develop acute megakaryoblastic leukemia (AMKL) within 4 years [Massey, 2005]. Mutations in the GATA1 gene have been detected in nearly all patients with Down syndrome who develop either TMD or AMKL [Ahmed et al., 2004]. GATA1 encodes a zinc-finger transcription factor essential for normal erythroid and megakaryocytic differentiation. Although many questions remain about GATA1 mutations and Down syndrome, this mechanism may be part of the explanation for their hematopoietic dysregualtion.
The next most common hematological abnormality we observed was thrombocytopenia, which, like neutrophilia, occurred in the majority. Hord et al. 1995 reported that 28% of neonates with Down syndrome had platelets counts <100,000/µl in the first 48 hr. Similarly, we found 33% of platelet counts in the first week to be <100,000/µl. However, we observed that another 33% of platelet counts were between 100,000 and 149,000/µl, making the incidence of thrombocytopenia 66%. Kivivuori et al. 1996 prospectively followed platelet counts of 25 neonates with Down syndrome during their first year and determined that the thrombocytopenia is generally short-lived, resolving over the first weeks, and generally replaced thereafter by thrombocytosis.
The third most common hematological abnormality we detected was polycythemia, with 14% of hematocrits over 70%, and 33% of hematocrits over 65%. Weinberger and Oleinick 1970 reported that 15% of neonates with Down syndrome had polycythemia. Subsequent studies confirmed polycythemia as common in neonates with Down syndrome, with a prevalence ranging between 8% and 64% [Lappalainen and Kouvalainen, 1972; Miller and Cosgriff, 1983; Hord et al., 1995; David et al., 1996]. Widness et al. 1994 reported elevated erythropoietin concentrations in umbilical cord blood of neonates with Down syndrome, speculating that hypoxemia in utero resulted in the elevated erythropoietin and consequent polycythemia. Nucleated red blood cell concentrations are generally elevated in neonates who have had in utero hypoxemia and elevated erythropoietin concentrations [Ghosh et al., 2003], but NRBC were not particularly high in our patients [McCarthy et al., 2006]. Starc 1992 and later David et al. 1996 reported high MCVs among patients, of all ages, with Down syndrome. David et al. 1996 found no underlying cause for the macrocytosis and postulated an abnormality of folate metabolism. We found MCVs to be within the general range expected among neonates without Down syndrome [Christensen, 1999].
Previous reports [Weinberger and Oleinick, 1970; Lappalainen and Kouvalainen, 1972; Miller and Cosgriff, 1983; Hord et al., 1995; David et al., 1996; Kivivuori et al., 1996] did not list anemia or neutropenia as common hematological problems among neonates with Down syndrome. Our findings concur with those reports. We speculate that the cases of anemia and neutropenia we observed were not more common than would be expected among a group of 158 neonates without Down syndrome.
We recognize that our study has several shortcomings. First, the anatomic site of the blood drawn for testing is not listed in our electronic records, and therefore the results we report surely contain some venous and some capillary sources, perhaps with some arterial sources as well. Capillary values tend to run somewhat higher than vascular sources when assessing a neonate's hemoglobin and hematocrit [Starc, 1992; Christensen, 1999; Kayiran et al., 2003]. Arterial blood tends to have a lower leukocyte count than simultaneously obtained venous or capillary samples [Christensen and Rothstein, 1979]. Platelet counts, and determinations of the MPV and MCV, do not vary significantly with the anatomic source [Christensen, 1999].
A second shortcoming is that CBCs were not drawn on all 226 neonates with Down syndrome, and thus our results might have a selection bias. For instance, the 158 on whom a CBC was obtained might be a somewhat different group than the 68 where no CBC was drawn. However, we speculate that the failure to obtain a CBC in the group of 68 was largely discretionary on the part of the pediatrician, because we found no evidence of demographic differences or length of initial hospital stay between those who did versus those who did not have a CBC obtained in the first week.
Third, in neonates with Down syndrome, changes occur in CBC values over the first months and years, yet we only sought values obtained in the first week. We did this to focus on the hematological abnormalities encountered in the newborn nursery, but in so doing we ignored important subsequent changes.
A fourth shortcoming is that our study does not clarify the underlying reasons why neonates with Down syndrome tend to have a low blood concentration of platelets, but high concentrations of neutrophils and erythrocytes. However, we identified one association, namely between polycythemia and thrombocytopenia. This association has also been observed among infants of diabetic mothers, where it has been postulated that hyperviscous blood results in thrombi, thus lowering the platelet count [Oppenheimer and Esterly, 1965].
We hope this report will assist those caring for neonates with Down syndrome, by illustrating how common neutrophilia, thrombocytopenia, and polycythemia are in this population. In our healthcare system, we obtained a CBC on only 158 of 226 (70%) neonates with Down syndrome. However, because hematological abnormalities are so common in newborn infants with Down syndrome, and since some of these neonates will need special treatments or tests on the basis of a hematological abnormality, it seems reasonable to recommend that one or more CBCs should be performed on all neonates with Down syndrome during their initial hospitalization. For instance, obtaining a CBC might facilitate diagnosing polycythemia, which could trigger treatment before further neurological damage has occurred. Moreover, the presence of neutrophilia and blasts may not always be transient and these children might need additional monitoring [Dixon et al., 2006]. We maintain that anticipatory care of neonates with Down syndrome should include awareness of their hematological abnormalities.
The authors thank Diane K. Lambert, RN, research nurse at McKay-Dee Hospital Center, Ogden, UT for valuable assistance, and Catherine Coverston, PhD, RNC, Associate Dean, Undergraduate Affairs, Brigham Young University College of Nursing, Provo, UT, for helpful discussions about the project.
- 2004. Natural history of GATA1 mutations in Down syndrome. Blood 103: 2480–2489. , , , , , , , , , , , , , .
- 1998. Thrombopoietin level is inversely related to blast count, not platelet number, in Down syndrome neonates with transient myeloproliferative disorder. Am J Hematol 58: 267–272. , , , , , , , , , , , , .
- Centers for Disease Control and Prevention (CDC) 2006. Improved national prevalence estimates for 18 selected major birth defects—United States, 1999–2001. Morb Mortal Wkly Rep 54: 1301–1305.
- 1999. Expected hematologic values for term and preterm neonates. In: ChristensenRD, editor. Hematologic problems of the neonate. Philadelphia: W.B. Saunders. p 122-123. .
- 1979. Pitfalls in the interpretation of leukocyte counts of newborn infants. Am J Clin Path 72: 608–611. , .
- 1996. Hematological studies in children with Down syndrome. Pediatr Hematol Oncol 113: 271–275. , , , , , , , , .
- 2004. Accuracy of the clinical diagnosis of Down syndrome. Ulster Med J 73: 4–12. , .
- 2006. Clinical manifestations of hematologic and oncologic disorders in patients with Down syndrome. Am J Med Genet Part C 142C: 149–157. , , .
- 1997. Circulating neutrophils in septic preterm neonates: Comparison of two reference ranges. Pediatrics 99: E10. , , , , , .
- 2003. Prediction of perinatal asphyxia with nucleated red blood cells in cord blood of newborns. Int J Gynaecol Obstet 81: 267–271. , , , .
- 2005. Origins of leukaemia in children with Down syndrome. Nat Rev/Cancer 5: 11–20. , .
- 1995. Thrombocytopenia in neonates with trisomy 21. Arch Pediatr Adolesc Med 149: 824–825. , , .
- 2005. In: Smith's recognizable patterns of human malformation. 6th edn. Philadelphia: WB Saunders Co. p 8–10. .
- 2003. Significant differences between capillary and venous complete blood counts in the neonatal period. Clin Lab Haematol 25: 9–16. , , , .
- 1996. Peripheral blood cell counts in infants with Down's syndrome. Clin Genet 49: 15–19. , , .
- 1972. High hematocrits in newborns with Down's syndrome: A hitherto undescribed finding. Clin Pediatr 11: 472–474. , .
- 2005. Hematologic problems in the fetus and neonate. In: MartinRJ, FanaroffAA, WalshMC. editors. Fanaroff and Martin's neonatal-perinatal medicine. 8th edn. Philadelphia: WB Saunders. p 1314–1315. , , .
- 1979. The neonatal blood count in health and disease. I. Reference values for neutrophilic cells. J Pediatr 95: 89–98. , , , .
- 2005. Transient leukemia in newborns with Down syndrome. Pediatr Blood Cancer 44: 29–32. .
- 2006. Umbilical cord nucleated red blood cell counts: Normal values and the effect of labor. J Perinatol 26: 89–92. , , , , .
- 1983. Hematological abnormalities in newborn infants with Down syndrome. Am J Med Genet 16: 173–177. , .
- 1994. Revised reference ranges for circulating neutrophils in very-low-birth-weight neonates. Pediatrics 9: 76–82. , , , .
- 1965. Thrombosis in the newborn: Comparison between infants of diabetic and non-diabetic mothers. J Pediatr 67: 549–554. , .
- 1998. Down syndrome, transient myeloproliferative disorder, and infantile liver fibrosis. Med Pediatr Oncol 31: 159–165. , , .
- 1999. In: ChristensenRD, editor. Hematologic problems of the neonate. Philadelphia: W.B. Saunders. p 281–291. , .
- 1992. Erythrocyte macrocytosis in infants and children with Down syndrome. J Pediatr 121: 578–581. .
- 1987. Differentiation of blast cells from a Down's syndrome neonate with transient myeloproliferative disorder. Blood 69: 508–512. , , , , , , , , .
- 1970. Congenital marrow dysfunction in Down's syndrome. J Pediatr 77: 273–279. , .
- 1994. Elevated erythropoietin levels in cord blood of newborns with Down's syndrome. Biol Neonate 66: 50–55. , , , .
- 1997. Leukemia and/or myeloproliferative syndrome in neonates with Down syndrome. Semin Perinatol 21: 97–101. , , , , .