Neonatal Hyperviscosity: I. Incidence

(1)

Neonatal

Hyperviscosity:

I. Incidence

Frederick H. Wirth, M.D., Karen E. Goldberg, M.S., and Lula 0. Lubchenco, M.D.

troiii the .\Teonatal Unit, Children ‘s Hospital of the King’s Daughters, ,\or/olk, Virginia, (111(1the \t’wi)orH

Service. Dit-i.s’ion of Peri,,atal .fedicine, University of Colorado .Icdical Center, Den n’r

ABSTRACT. Capillary heinatocrits were performed on 790

infants during the first four hours after birth. These infants

were delivered between August 8 and December 7, 1974, at the University of Colorado Medical Center, which is at an

altitude of 1,061 ni above sea level. When the capillary

hematocrit was 7% or greater, venous hematocrit and blood

viscosity were determined. Capillary hematocrits obtained

from warmed heels in the first hour after birth were spuriously high and not consistently related to venous hematocrit. VenOUS polcythemia, defined as a hematocrit of 65% or greater, occurred in 4% of the newborn population. Hperviscosit (> 2 SD above the mean for newborns)

occurred in 5% of the newborn infants. At a venous

hemato-crit of 65% or greater, hyperviscosity was predictable, but

sonic infants with venous hematocrits between 60% and 64% also had hperviscositv of the blood. The incidence of

polcythemia and hyperviscosit was further related to birth weight and gestational age. The infants who were small for gestational age were at highest risk of polycythemia and hperviscositv, followed by infants who were large for gestational age. However, the greatest number of infants with hvperviscosity were term appropriate for gestational age. Preterrri infants with gestational ages of less than 34

weeks were not affected. Pediatrics 63:833-836, 1979, newborn infant, Ilypert-iscosity, venous polycythemia, hein.a-tocrit.

Polycythemia in the newborn is well known

but its incidence has not been established.

Reports of cases of polycythemia with associated

clinical manifestations such as respiratory

dis-tress, abnormal renal function, and central

nervous system disorders have been reported.’

Certain groups of newborn infants are known to

be at increased risk of having high hematocrits,

including infants with intrauterine growth

retar-dation,ui those who have had delayed clamping of

the umbilical cord7 and, occasionally, twins since

twin-to-twin transfusions are not rare.8

Polycy-themia rarely occurs in preterm infants.

Polycythemia is the principal cause of

hyper-viscosity of the blood in newborn infants. Other causes of hyperviscosity are related to an

abnor-mality of red blood cell membranes which result

in lack of deformability.’#{176} Decreased

deformabili-ty may be primary, such as occurs with

spherocy-tosis, or it may be secondary to the effects of

hypothermia, hypoxia, acidosis, or

hypoglyce-mia.9 Elevated plasma protein or lipoprotein

levels, as a cause of hyperviscosity of the blood,

rarely occur in newborn infants.

The purpose of this paper is to present data on

incidence of polycythemia and hyperviscosity in a

newborn population and to relate its occurrence

to birth weight-gestational age categories.

CLINICAL MATERIAL

All infants born in the University of Colorado

Medical Center during a four-month period,

August 8 to December 7, 1974, constituted the

population to be screened for polycythemia.

During this period, 833 infants were delivered

and, of these, 790 had capillary hematocrits

determined within four hours of birth. The 43

who were not screened (5.1% of the births)

included 12 very premature infants who died in

the delivery room or shortly after transfer to the

Intensive Care Nursery, 20 term infants

appro-priate for gestational age, and 1 1 who were from

other birth weight-gestational age categories.

These infants were missed during the institution

of a new screening procedure.

A capillary hematocrit of 70% or greater was

used as the criterion for determining venous

hematocrit and blood viscosity.

The newborn population at the University of

Colorado Medical Center is a relatively high-risk

one, having 1 1% low-birth-weight infants. About

two thirds of the infants are born to parents in low

Received November 10, 1977; revision accepted for pul)hca-tion November 1, 1978.

ADDRESS FOR REPRINTS: (F.H.W.) Neonatal Unit,

(2)

24 26 28 30 32 34 36 38 40 42 44 46

Pre-Term I Term ) Post-Term

Week of Gestation

FIG 2. Incidence of hyperviscosity, ie, viscosity > 2 SD

above mean for newborn infants, is 5%. Hyperviscosity of blood at venous hematocrits < 65% occurred primarily in

infants with weight appropriate for gestational age; 4% of term infants with weight appropriate for gestational age

were found to have hyperviscosity.

834

NEONATAL

HYPERVISCOSITY

NEWBORN PoPuiTIoN SCHEENED FOR PoL-crrHEM IA

Procedures (2nd Iin(Iing.s .\.

Consecutive births 833

Infants screened for polycvtheniia 790 Infants examined for venous hematocrit and 96

viscosity

Infants with venous heniatocrits 65% or 31

greater

Infants with hperviscosit 43

socioeconomic conditions, class V, nearly a third from class IV, and less than 5% in classes I, II, or III.” Denver is at an altitude of 1,610 m above sea level.

The incidence of polycythemia and

hypervis-cosity was related to birth weight-gestational age groups, as defined in the newborn classification

by Battaglia and Lubchenco.12 Gestational age was calculated from the onset of the mother’s last

normal menstrual period and confirmed by

phys-ical examination of the infant during the first hour

after birth. The gestational age based on

menstrual history was accepted unless clinical

criteria clearly indicated a discrepancy of three or

24 26 28 30 32 34 36 38 40 42 44 46

Pre-Term I Term I Post-Term

Week of Gestation

FIG 1. Incidence of venous polycthemia, ie,

hemato-crit 65%, is shown b birth weight-gestational age group-ings. Overall incidence is 4%. Infants small for gestational

age have highest incidence, with term and postterm infants large for gestational age next.

more weeks then the clinical gestational age was

used. The clinical assessment differentiates, with

accuracy, the preterm from the term infant.

However, clinical criteria that accurately define

postterm births are not available and, therefore,

menstrual history was accepted for this popula-tion.

The time of clamping of the cord after birth

was recorded by the nurse in the delivery room.

LABORATORY METHODS

The capillary hematocrit was performed by the

nursing staff in the first one to four hours after

birth while the infant was in an incubator. The

axillary temperature was kept between 36.5 and

37.5 C. A heel stick was done by use of a lancet,

and blood was collected directly into a

heparin-ized microhematocrit tube, spun in the

micro-hematocrit centrifuge for three minutes, and read

from the chart provided by the centrifuge compa-ny. If the hematocrit was 70% or greater, a repeat capillary hematocrit was performed or a venous hematocrit was obtained.

Venous hematocrits were performed by the

microhematocrit method. Since the sample was

also used for determining viscosity, the syringe

at Viet Nam:AAP Sponsored on September 7, 2020

www.aappublications.org/news

(3)

was coated with heparin by using only the amount in the needle and expelling all of it before

performing the venipuncture.

Blood viscosity was measured by the method of

Gross et a1 using a microviscometer run at shear

rates of 5 to 212 sec ‘. The viscosity of

mdi-vidual infants was plotted on the norms

estab-lished by Gross et al. Infants whose values were

more than 2 standard deviations above the mean

for newborns were considered to have

hypervis-cosity.

RESULTS

Capillary hematocrits performed in the first

hour after birth gave spuriously high values; 132

of the

790

infants were found to have capillary

hematocrits of 70% or greater-an incidence of 17%. The capillary hematocrit was repeated two

to four hours later in 24 infants and all were lower than the earlier hematocrit. Furthermore, the venous hematocrit in half of the infants with

capillary hematocrits greater than 70% in the first

hour were less than 65%.

The Table gives the number of consecutive

births, number of infants screened for polycy-themia, and the incidence of venous polycythe-mia and hyperviscosity. A venous hematocrit of

65% or greater occurred in 4% of the population

studied. Because hyperviscosity occurs for reasons

other than an elevated hematocrit, the incidence

of hyperviscosity was higher than polycythemia,

ie, 5%.

The data were then arranged according to birth

weight and gestational age as illustrated in Fig 1.

The highest incidence of venous polycythemia is

found in term and postterm infants who are small

for gestational age, with the next highest being in

the postterm infants who are large for gestational

age.

When the venous hematocrit was 65% or

greater, all infants showed hyperviscosity. A

number of infants with hematocrits less than 65%

also showed hyperviscosity. Since the screening

procedure was based on a high hematocrit, infants

with hyperviscosity due to other causes may have

been missed. Eleven infants identified as having hyperviscous blood had venous hematocrits

between 62% and 64%. One infant with

congen-ital hyperlipidemia had hyperviscous blood with

a hematocrit of 60%. The distribution of infants

with hyperviscosity by birth weight and

gesta-tional age is shown in Fig 2.

The time of clamping of the umbilical cord was

plotted against capillary hematocrits, and no

significant relationship was found (r = .04). The

data for venous hematocrit were similarly

exam-med in relation to cord clamp time and, again, no

correlation was found. However, the cord clamp

times were short, having a mean of 30 seconds,

with only five infants having their cords clamped

later than 60 seconds after birth. The hematocrits

in four of these five infants were high and

hyperviscosity of the blood was demonstrated.

Sex distribution of the infants with hyperviscosity

was 25 girls and 18 boys. The entire population

included 399 male and 391 female infants.

DISCUSSION

A 4% incidence of polycythemia in a newborn

population is impressive because of the number of

infants involved. In a service delivering 100

babies per month, four infants (or one each week)

will require special attention because of the high

hematocrit. Because these infants are usually near

or at term or even postterm, they usually will be

cared for in a low or intermediate care nursery.

There are data which tend to support the high

incidence of polycythemia found in this study.

Gatti and his coworkers2 screened 629 newborn

infants in the first day after birth by performing

capillary hematocrits. The mean capillary

hema-tocrit was 62.9% with 20 infants (3%) having

hematocrits higher than 75%. The number of

infants with hematocrits between 70% and 75% was not given. The venous hematocrits of these

infants were not reported. Mackintosh and

Walk-er’5 determined the blood viscosities in 110

normal newborn infants and found that the mean

values for viscosity was higher than that found in

adults. The high blood viscosity was exaggerated

in infants who were small for gestational age.

Elevated mean hematocrit values in newborns

are reported by other investigators,’1”4”6 but

these data do not give the number of infants at

risk in the population being studied.

The true incidence of hyperviscosity in

newborns was not found. In this study, there were

12 infants with venous hematocrits between 60%

and 64% who had evidence of hyperviscosity. Not

all infants with hematocrits below 65% were

screened for hyperviscosity; therefore, the

mci-dence reported here is a low estimate. At least 1%

of newborn infants have hyperviscosity for causes

other than a high hematocrit. Except for the

infant with hyperlipidemia, the cause of the

hyperviscosity was not clarified. Viscosity of the

blood increases as the hematocrit rises and, at

hematocrits of 65%, all the infants in this study

showed evidence of hyperviscosity. The data of

Bergqvist9 support these findings.

Hyperviscosity is known to be associated with

(4)

836

NEONATAL

HYPERVISCOSITY

Newborns have such red cells. Low pH of the

blood and low body temperature, both common

occurrences in newborns, also cause increased viscosity. It is not certain whether these

condi-tions affect red cell membranes and their deform-ability characteristics or whether pH and cold stress have a direct effect on viscosity. Abnormal

hemoglobin and elevated plasma protein and

plasma lipid levels are rare causes of hyperviscos-ity in newborn infants.

The distribution of cases of polycythemia and

hyperviscosity by birth weight and gestational age clearly shows infants who are small for gestational age to be at highest risk for this condition. It is curious that the next highest incidence is found in infants who are large for gestational age-the opposite in intrauterine growth. One wonders if the etiology of

polycy-themia in the two populations may therefore be different: a large infant with a large placenta would receive a greater volume of blood as a placental transfusion than would the small infant.

The infant who is small for gestational age,

however, may be producing excessive red blood cells because of chronic hypoxia. These hypo-theses must be tested.

Hyperviscosity occurs primarily in term or near-term infants, which makes it important for personnel caring for low-risk infants to be aware

of the high incidence in this population.

SUMMARY

The incidence of polycythemia in a newborn

population was 4%. A venous hematocrit of 65% was invariably associated with hyperviscosity, but

some infants with venous hematocrits between 60% and 64% also had hyperviscosity.

The incidence of hyperviscosity was found to

be 5%. Infants who were small for gestational age were most frequently affected, with infants who were large for gestational age and postterm infants next. Hyperviscosity is a condition most frequently found in low- and medium-risk nurser-ies.

Screening for capillary polycythemia at about four hours after birth is suggested to identify infants at risk. A capillary hematocrit of 70% is a

reasonable level for further investigation, but some infants with hyperviscosity will be missed.

REFERENCES

1. Buckels U, Usher R: Cardiopulmonary effects of placetital transfusion. I Pediatr 67:239, 1965. 2. Gatti RA, Muster AJ, Cole RB, Patti MH: Neonatal

polycythemia with transient cyanosis and cardiores-piratory abnormalities. I P(’(liatr 69: 1063, 1966.

3. Aperia A, Bergqvist C, Broberger 0, et al: Renal function ill newborn infants with high hematocrit

values before and after isovolemic haemodilution.

Acta Paediatr Scand 63:878, 1974.

4. Wood JL: Plethora in the newborn infant associated with cyanosis and convulsions. I P’(liat? 54: 143, 1959.

5. Gross GP, Hathaway WE, McGaughe HR: Hypervis-cosity in the neonate. I Pediatr 82:1004, 1973.

6. Humbert JR, Abelson H, Hathaway WE, Battaglia FC: Polycythemia in small for gestational age infants. I Pediatr 75:812, 1969.

7. Usher R, Shephard M, Lind J: The blood volume of the newborn infant and placental transfusion. Acta

Paediatr 52:497, 1963.

8. Rausen AR, Seki M, Strauss L: Twin transfusion

syndrome. I Pediatr 66:613, 1965.

9. Bergqvist G: Viscosity of the blood in the newborn infant. Acta Paediatr Scand 63:858, 1974.

10. Gross GP, Hathaway WE: Fetal erythrocyte

deformabil-ity. Pediatr Res 6:593, 1972.

11. Myers JK, Bean LL: A Decade Later: A Follow-up of

Social Class and Mental Illness. New York, John Wiley & Sons Inc, 1968, app 3, p 235.

12. Battaglia FC, Lubchenco LO: A practical classification of newborn infants by weight and gestational age. I Pediatr 71:159, 1967.

13. Guest GM, Brown EW, Lahey ME: Normal blood values

in infancy and childhood. Pediatr Gun North Ant

4:357, 1957.

14. Moe PJ: Normal red blood picture during the first three years of life. Aeta Paediatr Scand 54:69, 1965.

15. Mackintosh TF, Walker CHM: Blood viscosity in the newborn. Arch Dis’ Child 48:547, 1973.

16. Gairdner D: The fluid shift from the vascular compart-ment immediately after birth. Arc/i Dis Child

33:489, 1958.

ACKNOWLEDGMENT

This investigation was supported in part by grants-in-aid

No. RR69 from the National Institutes of Health and No.

1-576 from the National Foundation-March of Dimes, and by

yearly contracts from the Department of Health, Education

and Welfare-Maternal Child Health Service through the Colorado State Department of Public Health and the University of Colorado Medical Center.

(5)

1979;63;833

Pediatrics

Frederick H. Wirth, Karen E. Goldberg and Lula O. Lubchenco