The hematologic and plasma iron responses to severe fetal hemorrhage in the ovine fetus

The hematologic and plasma iron responses to severe fetal hemorrhage in the ovine fetus Larry E. Shields, MD,a John A. Widness, MD, b and Robert A. Brace, Phi)"

San Diego, California, and Iowa City, Iowa OBJECTIVE: We previously reported that the ovine fetus does not significantly increase its red blood cell production rate after a 40% loss of blood in spite of a transient elevation in plasma erythropoietin concentration. In this study we hypothesized that, in response to a more severe loss of blood, the ovine fetus would undergo a sustained rise in plasma erythropoietin concentration and an augmentation in its red blood cell mass expansion rate. STUDY DESIGN: Twelve chronically catheterized fetal sheep (six control and six hemorrhaged) were studied beginning at 126 _+1 (SE) days' gestation. Fetal blood volume, plasma volume, red blood cell mass, reticulocyte count, plasma erythropoietin level, and plasma iron level were measured for 10 consecutive days. On days 1, 2, and 3 the hemorrhaged fetuses had an average of 102 + 4 ml per day of blood removed at a rate of 1 ml/min for a total of 305 + 10 ml of blood removed. Statistical analysis was by one- and three-factor analysis of variance. RESULTS: Control animals had a progressive increase in blood volume, plasma volume, and red blood cell mass throughout the 10-day protocol. Reticulocyte counts and plasma iron and erythropoietin levels did not change. In fetuses at 24 hours after the third hemorrhage blood volume averaged 9.3% below (p = 0.03) and plasma volume averaged 16.4% above (p = 0.04) that in the control animals. Thereafter blood and plasma volumes expanded at rates similar to controls. Erythropoietin increased (p < 0.001) but returned to prehemorrhage values by day 7. Posthemorrhage expansion of the red blood cell mass in the hemorrhaged animals was 1.9 times controls (6.8% + 0.9%/day vs 3.5% + 0.5%/day, p = 0.003). Fetal reticulocyte counts remained elevated throughout the posthemorrhage observation period (p < 0.001). The fetal plasma iron concentration decreased (p < 0.0001) and remained depressed. The recovery of red blood cell mass and the 10-day mean plasma iron concentration were highly correlated (p = 0.01, r = 0.91). CONCLUSION: The ovine fetus significantly increases its release of red blood cells in response to a severe hemorrhage. Further, the ability of the fetus to restore its red blood cell mass appears to be dependent on the plasma iron concentration. (AM J OBSTETGYNECOL1996;174:55-61 .)

Key words: Ovine, h e m o r r h a g e , fetus, b l o o d volume, red blood cell mass, iron, erythropoietin

Fetal b l o o d loss may occur by a variety of mechanisms, such as fetomaternal h e m o r r h a g e , subchorionic placental h e m o r r h a g e , or retroplacental h e m o r r h a g e , and is b e c o m i n g increasingly recognized as a cause of adverse perinatal o u t c o m e ) Fetal isoimmunization may also cause large reductions in circulating fetal red blood cells. 2 Large reductions in the fetal red b l o o d cell mass resulting in severe fetal a n e m i a have b e e n associated with intra-

From the Division of Perinatal Medicine, Department of Reproductive Medicine, University of California, San Diego~ and the Department of Pediatrics, University of Iowa) Supported by National Institutes of Health grants No. HD20295 and HL46925. Presented at theForty-firstAnnual Meeting of the Societyfor Gynecologic Investigation, Chicago, Illinois, March 22-25, 1994. Received for publication November 28, 1994; revised May 12, 1995; acceptedJune 5, 1995. Reprint requests: Robert A. Brace, PhD, Department of Reproductive Medicine, University of California, San Diego, La Jolla, CA 920930802. Copyright © 1996 by Mosby-YearBook, Inc. 0002-9378/96 $5.00+ 0 6/1/66804

uterine growth retardation, hydrops retails, and fetal d e a t h ) ' 3, 4 Because of the i m p o r t a n c e of this problem, a n u m b e r of investigators have evaluated the fetal response to acute blood loss. 5-s These studies 5-8 have limited their fetal assessment to 24 hours after h e m o r r h a g e . However, the fetal hematologic response may require several days to develop. To address this, we recently evaluated the fetal hematologic responses over 7 days after an acute h e m o r rhage, g In that study ovine fetuses w h e n subjected to a 34% r e d u c t i o n in red blood cell mass did n o t significantly a u g m e n t their p r o d u c t i o n rate of red blood cells. Additionally, there was only a transient increase in fetal reticulocytes and plasma erythropoietin. In contrast, in severely anemic h u m a n fetuses erythropoietic index values are elevated.2. 10 Thus the small, transient responses in the ovine fetus may have b e e n related to the severity of the fetal h e m o r r h a g e . T h e r e f o r e we hypothesized that in response to a m o r e severe loss of blood (1) the ovine fetus would u n d e r g o a larger and m o r e sustained increase in plasma erythropoietin, (2) the red blood cell mass expan55

56

Shields, Widness, and Brace

January 1996 AmJ Obstet GynecoI

Table I. Gestational age and baseline data between control and h e m o r r h a g e groups on protocol day 1

Gestational age (days) Hematocrit (%) Blood volmne (ml) Plasma protein concentration (gm/dl) Erythropoietin (mU/ml) Plasma iron (lag/dl) Reticulocyte count (%) Arterial pH Arterial Po z (mm Hg) Arterial P c o 2 (ram Hg)

Control (n = 6)

Hemorrhage (n = 6)

Significance

124.2 + 0.8 33 _+1.4 326 + 21 3.4 ± 0.1 37.1 ± 12.6 85.4 + 12.8 1.3 ± 0.5 7.30 ± 0.01 22.5 ± 1.2 51.4 ± 0.8

127.8 ± 0.5 33.6 ± 0.9 315 ± 23 3.5 ± 0.1 12.8 ± 1.5 154.7 ± 24.7 1.6 ± 0.3 7.33 ± 0.01 22.7 ± 0.3 55.2 ± 1.2

p = 0.0034 NS NS NS p = 0.015 p = 0.077 NS p = 0.012 NS p = 0.028

Values are mean + SE. NS, Not significant.

sion rate would increase significantly, and (3) there would be a simultaneous reduction in fetal plasma iron concentration.

Material and methods Twelve time-dated p r e g n a n t ewes with a single fetus were studied. Gestational age was 126 + 1 (SE) days at the start of the e x p e r i m e n t (term 145 to 150 days). With the animal u n d e r general anesthesia maternal vascular catheters were placed in the inferior vena cava and descending aorta by femoral access. Fetal vascular catheters were placed in the descending aorta and inferior vena cava at the level of the fetal d i a p h r a g m t h r o u g h femoral vessels. Amniotic fluid catheters were attached to the fetal skin. All catheters were t u n n e l e d subcutaneously and exteriorized at the maternal flank. All experiments were started a m i n i m u m of 5 days after surgery. Detailed information about animal preparation and m a i n t e n a n c e has b e e n previously described, n T h e experimental protocol was approved by the University of California, San Diego, Animal Subjects Committee, and animal care m e t National Institutes of Health guidelines. T h e 12 animals were divided into 2 groups: conu-ol ( n = 6) and h e m o r r h a g e d ( n = 6). Five of the control animals have b e e n r e p o r t e d in a previous investigation. 9 T h e h e m o r r h a g e d fetuses were subjected to a daily hemorrhage for 3 days. o n the first day 110 + 4 ml (SE) of fetal b l o o d was removed, on the second day 106 +_7 ml, and on the third day 88 + 6 ml (102 + 4 m l / d a y average for the 3 days or 98.6% + 5.2% of day 1 blood volume). Fetal b l o o d was r e m o v e d at a rate of 1 m l / m i n . T h e withdrawal each day was adjusted to achieve a 50% reduction in red blood cell mass after the third h e m o r r h a g e . In both groups blood volume, hematocrit, plasma volume, plasma protein concentration, plasma protein mass, reticulocyte count, heart rate, vascular pressure, arterial pH, and b l o o d gases were m e a s u r e d on protocol days 1, 3, 4, 6, 8, and 10. Fetal plasma erythropoietin and iron levels were m e a s u r e d daily (days 1 to 10). O n days 1 to 3 these m e a s u r e m e n t s were m a d e before the fetal h e m o r r h a g e . Fetal b l o o d volume was m e a s u r e d with technetium-

labeled (99mTc) autologous fetal red b l o o d ceils and indicator dilution techniques. '2 Fetal h e m a t o c r i t was determ i n e d in triplicate, and plasma protein concentration was d e t e r m i n e d by refractometer (American Optical TS meter, Buffalo, N . Y . ) . 7 Fetal red blood cell mass was calculated by multiplying the b l o o d v o l u m e and the fractional hematocrit (Red blood cell m a s s = B l o o d volume x H e m a t o c r i t / 1 0 0 ) . Fetal plasma volume was calculated as the difference between b l o o d v o l u m e and red b l o o d cell mass (Plasma volume = Blood v o l u m e - Red b l o o d cell mass). Plasma protein mass was calculated by multiplying total plasma protein c o n c e n t r a t i o n and plasma v o l u m e (Plasma protein mass = Plasma protein concentration x Plasma volume). Peripheral b l o o d smears were stained with methylene blue, and fetal reticulocyte counts were d e t e r m i n e as a percentage of total fetal red b l o o d cells after a m i n i m u m of 2000 erythrocytes were counted. Because the reticulocyte c o u n t is a percentage of the total red b l o o d cells, the c o r r e c t e d reticulocyte c o u n t was used for analysis. The c o r r e c t e d reticulocyte c o u n t was calculated as Reticulocyte c o u n t × Sample h e m a t o c r i t / P r e h e m o r r h a g e hematocrit. Protocol day 1 was used as the p r e h e m o r r h a g e h e m a t o c r i t value. O n e milliliter of fetal arterial b l o o d was collected daily and centrifuged at 2000 revolutions p e r m i n u t e for 10 minutes~ After centrifugation 0.6 ml of fetal plasma was r e m o v e d and stored at 0 ° C for plasma erythropoietin and iron measurements. Plasma iron and erythropoietin concentrations were d e t e r m i n e d electrochemically and by radioimmunoassay, respectively, as previously described.aS. 14 Fetal arterial and venous b l o o d pressures were r e c o r d e d by onqine c o m p u t e r techniques and corrected for the zero pressure reference of the fetus by continuously subtracting amniotic fluid pressure f r o m the vascular pressures." Fetal heart rate and blood pressures were r e c o r d e d for >1 h o u r during each m o n i t o r i n g session before h e m o r r h a g e and saved on disk as 1-hour averages. Blood gases and p H were m e a s u e r d at 37 ° C and c o r r e c t e d to fetal body t e m p e r a t u r e of 39.5 ° C (Instrum e n t a t i o n Laboratory system p H / b l o o d gas analyzer, Lexington, Mass.). Blood r e m o v e d f r o m the fetus for

Shields, Widness, and Brace 57

Yblume 174, Number 1, Part i .~nJ Obstet Gynecol

i--

.7o;.T

Hemorrhages

25 ! i1) E~

20

>2

10

"oO

20 O~ A

15

0

cO U

5

J

0 -5

++q,

40

: ~ I:_/I - ~: I~

I~

I

I

1

I

-20 -40

. - Control D - Hemorrhaged

I

"

2

r

I =

30

0

.

I]

li

.

i ,

i

i

i

±

~

I

4,

.......... ~..

:: ~_ ~

=g2o

.~113 ~E £ £

-8

g

-12

0"

-16

e~ 0 1 F-

1

2

I-

I

I

I

I

I

I

I

3

4

5

6

7

8

9

10

-20

• -Control ~-Hemorrhaged

Y I--

I-"

1

2

3

I

I

I

I

I

I

I

4

5

6

7

8

9

10

Time (days)

Time (day)

Fig. 1. Change in fetal blood volume (upper panel) and plasma volnme (lower panel) expressed as percent change from prehemorrhage values (mean _+SE). Average of 102 ml of fetal blood was removed on each of 3 days of hemorrhage. The p values represent treatment term of analysis of variance. Asterisk, p < 0.05; two asterisks, p < 0.01 (for daily comparisons of hemorrhaged and controls from post hoc testing).

Fig. 2. Change in fetal red blood cell mass (upper" panel) and hematocrit (lower panel) expressed as change from prehemorrhage values (mean _+SE). Average of 102 ml of fetal blood was removed on each of 3 days of hemorrhage. The p values represent interaction between time and treatment. Three asterisks, p < 0.001 for daily comparisons of hemorrhaged and controls from post hoc testing.

sampling (<3 m l / d a y ) was replaced with fresh heparinized maternal blood. Statistical analysis. All data are expressed as the m e a n _+ 1 SE. To d e t e r m i n e w h e t h e r significant differences existed between the h e m o r r h a g e d and control groups, a three-factor analysis of variance was p e r f o r m e d with time, treatment, and animal as the factors. Differences in responses were d e t e r m i n e d as a significant effect of t r e a t m e n t or a significant interaction between time and treatment. Fisher's least-significant difference for m u h i p l e comparisons was used for post hoc testing if the analysis of variance was significant. Most of the data are expressed as a percentage o f the p r e h e m o r r h a g e values to r e d u c e the interanimal variability and to normalize for potential differences in fetal size. T h e change in blood volume, red blood cell mass, plasma volume, and plasma protein mass are expressed as a p e r c e n t change from protocol day 1 values. Plasma erythropoietin concentrations were log transformed before analysis to normalize their distribution. T h e relationship between the 10-day average fetal plasma iron c o n c e n t r a t i o n and the recovery of the red blood cell mass in the h e m o r r h a g e d animals

(on day 10) was evaluated by regression analysis. T h e 10-day average iron concentration provided the highest correlation. Multivariate regression was used to explore relationships a m o n g several variables simultaneously. For these analyses data f r o m individual fetuses were combined (six fetuses x six p o i n t s / f e t u s or n = 36), and variables were i n c l u d e d only if their regression coefficient was significant. T h e Student t test was used to evaluate differences in baseline physiologic measures between the two groups. Statistical significance was accepted at p < 0.05. Results

Gestational age and baseline physiologic values f r o m protocol day 1 are shown in Table I. T h e h e m o r r h a g e d animals were 4 days older than the control animals (p < 0.05), and there were small but statistically significant differences in baseline pH, Pc%, and erythropoietin levels. T h e control animals had a progressive increase in blood v o l u m e consistent with advancing gestational age (Fig. 1, u p p e r panel). During the 3-day h e m o r r h a g e an

58

Shields, Widness, and Brace

January 1996 A m J Obstet Oynecol

Hemorrhages

Hemorrhages 20

"~

L-

I

i

i

i

I

I

i

t2

¢-

0

ca

+ ++

i

i

i

i

i

i

i

200

16

0

o

400

0 O. 0

12

--E

100 80

:/ ; /

•-Control

~-

n-Hemorrhaged

~/

8o ""0

8

• - Control

E

[] - Hemorrhaged 0

o

0

E

20

P< .001

4

I--

1

40

I-

Iz

I

I

I

I

I

I

I

2

3

4

5

6

7

8

9

10

Time (days)

10

I:

1

2

3

I

I

I

I

[

I

I

4

5

6

7

8

9

10

Time (days)

Fig. 3. Fetal reticulocyte counts corrected for degree of anemia (mean -- SE). The p value represents interaction between time and treatment. Three asterisks, p ~ 0.001 for daily hemorrhaged versus control comparisons.

Fig. 4. Plasmaerythmpoietin levels (mean -+SE, log scale). The p value represents interaction between time and treatment. Asterisk, p< 0.05; two asterisks, p< 0.01; three asterisks, p< 0.001 (for daily hemorrhaged vs control comparisons).

average of 102 + 4 ml/day of fetal blood was removed each day, for a total blood volume withdrawal of 305 +_10 ml (98.6% +_5.2% of blood volume on day 1). On day 4 (24 hours after the third hemorrhage) blood volume in the hemorrhaged group was the same as the prehemorrhage value (day 1) but was 9.3% below the control group. During the 7 days after the hemorrhage (days 4 to 10) blood volume remained below control levels (p= 0.028, Fig. 1, upper panel). After the hemorrhage blood volume expanded at a rate of 8.8 +_2.0 ml/day (3.0%+l.0%/day), which was similar to controls (8.4 +_1.7 ml/day or 2.5% +_0.5%/day). Plasma volume in the control animals expanded in parallel with the expansion of fetal blood volume, increasing at a rate consistent with advancing gestational age. In the hemorrhaged animals plasma volume increased markedly in response to the fetal hemorrhage. Twenty-four hours after the third hemorrhage (day 4) plasma volume was 22.7% +_2.6% (47.7 +_7.3 ml) above the prehemorrhage value (p < 0.05, Fig. 1, lower panel). During this same time plasma volume in the control group increased 5.9% +_2.8% (12.9+_5.9 ml, p<0.01). During the 7 days after the hemorrhage plasma volume in the hemorrhaged group remained above the control group (analysis of variance, p = 0.042 for hemorrhage vs control animals). Similar to the increase in plasma volume, there was a simultaneous increase in the plasma protein mass after hemorrhage. Plasma protein mass (data not shown) in the hemorrhaged group remained

above that of the controls during the 7 days after hemorrhage; however, this difference was of marginal significance (analysis of variance, p = 0.052). Throughout the experiment there was no difference in concentration of plasma proteins between the two groups. Fetal hemorrhage reduced the red blood cell mass from a prehemorrhage value of 105 _+7 ml to 57 +- 6 m124 hours after the third hemorrhage (p < 0.001, Fig. 2, upper panel). This change represented a 47% _+3% reduction in the red blood cell mass. During the 7 days after the hemorrhage (days 4 to 10) the red blood cell mass (relative to day 1 values) in the control group increased at a rate of 3.5% +- 0.5% per day compared with 6.8% + 0.9% per day in the hemorrhaged group (p< 0.001, Fig. 2, upper panel). By the end of the posthemorrhage observation period the red blood cell mass in the hemorrhaged animals returned to prehemorrhage values but was still significantly below (33.8%, p = 0.001) that of the control group. Fetal hematocrit decreased from a prehemorrhage value of 33.6% _+0.9% to its lowest value of 17.9% + 0.6% on day 4 (p < 0.001). During the posthemorrhage observation period fetal hematocrit in the hemorrhaged group increased by 1.5% +- 0.3% per day (i.e., hematocrit percentage points/day) compared with 0.20% +- 0.05% per day in the control group (p < 0.001, Fig. 2, lower panel). Corrected reticulocyte counts increased in response to the fetal hemorrhage from a prehemorrhage mean of 1.6% +_0.3% to a peak of 16.3% +_2.5% on day 4

Shields, Widness, and Brace 59

Volume 174, N u m b e r 1, Part 1

Am J Obstet Gynecol

Hemorrhages

+,++

[

120

. . . . . . .

.....-''""

5O

....'"" ..-'"'"'"

e-

.~

25

100

Q

-.!l

0 O

e"

8 =~

o')

E -~ o.

-25

• - Control D- Hemorrhaged

i

80

-75 -100 I:

1

I:

2

3

4

I

I

I

I

I

I

5

6

7

8

9

10

60 25

t

I

50

75

100

25

Mean Plasma Iron (mg/dL}

Time (days) Fig. 5. Plasma iron levels expressed as change from prehemorrhage values (mean + SE). The p value represents interaction between time and treatment. Asterisk, p< 0.05; two asterisks, p < 0.01; three asterisks, p < 0.001 (for daily hemorrhaged vs controls comparison).

Fig. 6. Linear regression plot (solid line) of red blood cell mass recovery on day 10 (percent of prehemorrhage value) and mean 10-day fetal plasma iron concentration. Dotted lines, 95% confidence interval about regression line. Regression equation is y = 61.2 + 0.376x, r= 0.91, p = 0.01.

(p < 0.001, Fig. 3). Fetal reticulocyte counts in the hemorrhaged animals r e m a i n e d elevated t h r o u g h o u t the posth e m o r r h a g e observation period. T h e r e was n o significant change in the reticulocyte counts in the control group t h r o u g h o u t the 10-day protocol. Fetal plasma erythropoietin concentration increased markedly in response to fetal h e m o r r h a g e (p < 0.001, Fig. 4). F r o m a p r e h e m o r r h a g e level of 12.8-+ 1.5 m U / m l erythropoietin increased to 354 ± 134 m U / m l on day 3 before the third h e m o r r h a g e (Fig. 4). Erythropoietin rem a i n e d elevated for the first 4 days after h e m o r r h a g e (days 4 to 7) and then r e t u r n e d to values that were not significantly different f r o m those in the control animals on days 8 to 10. Plasma iron levels decreased after the fetal h e m o r r h a g e and r e m a i n e d significantly lower than p r e h e m o r r h a g e values for the r e m a i n d e r of the study (p < 0.001, Fig. 5). T h e recovery of the red b l o o d cell mass in the h e m o r r h a g e d animals at the e n d of the e x p e r i m e n t (day 10) was highly correlated to the 10-day average fetal plasma iron c o n c e n t r a t i o n (p = 0.01, r = 0.91, Fig. 6). In the control animals the line relating red b l o o d cell mass on day 10 to m e a n iron c o n c e n t r a t i o n was approximately parallel to that in the h e m o r r h a g e d animals but was shifted 47 units upward on the vertical axis (y = 108.1 + 0.266x, r = 0.61, p = 0.28). T h r o u g h o u t the course of the e x p e r i m e n t there were no differences in fetal arterial and venous blood pressures, fetal heart rate, and fetal arterial p H and Pco~,

between the two groups. Fetal arterial Po 2 decreased significantly in response to the h e m o r r h a g e , from a p r e h e m orrhage value of 22.7 + 0.3 to 17.3 ± 0.4 m m Hg at 24 hours after the third h e m o r r h a g e (p < 0.001, Fig. 7). Arterial Po9 r e m a i n e d depressed during the r e m a i n i n g 6 days of the p o s t h e m o r r h a g e observation p e r i o d and averaged 3.1 m m Hg below that in the control animals and 2.0 m m H g below p r e h e m o r r h a g e values. T h e relationship between plasma erythropoietin, P%, and hematocrit was d e t e r m i n e d by regression analysis on the c o m b i n e d data from all h e m o r r h a g e d animals. T h e increase in erythropoietin was related to either P % (r= -0.754, p < 0.001) or hematocrit ( r = - 0 . 6 9 3 , p < 0.001) as d e t e r m i n e d with bivariate regression. A better fit was f o u n d with multivariate regression: log10 (erythropoietin) = 4 . 7 8 8 0.081~1 x Po 2 - 0.0616x H e m a t o c r i t (r=0.867, p < 0.001). However, the best fit was f o u n d when plasma iron was i n c l u d e d in the regression: log10 (Erythropoietin) = 4.762 - 0.0706 x Po 2 - 0.0835 x H e m a t o c r i t + 0.00383 x Iron ( r = 0.914, p < 0.0001).

Comment T h e primary goal of this investigation was to evaluate the fetal hematologic response to a large reduction in the fetal red blood cell mass. A n u m b e r of previous investigations have evaluated the short-term fetal responses to a graded h e m o r r h a g e , e-< " ' ~20 and only one study from our laboratory e x p l o r e d the fetal responses to r e d u c e d red

60 Shields,Widness,andBrace

January 1996 AmJ Obstet GynecoI

Hemorrhages

+,

+

I

:

A

0

I

!

I

I

i

i

i

'"

~g

P<. 0 0 ~ -2

\

/

•

-4

on,ro,

[] - Hemorrhaged

-6 I:

1

2

3

I

I

I

I

I

I

I

4

5

6

7

8

9

10

Time (days)

Fig. 7. Fetal arterial Po 2 expressed as change from prehemorrhage values (mean ± SE). The p value refers to interaction between time and treatment

blood cell mass over a period of days. 9 In that study reducing red blood cell mass by 34.6% +_2.1% stimulated an initial rise followed by a marginally sustained increase in plasma erythropoietin. However, there was no significant increase in the release of fetal reticulocytes, and the expansion rate for the fetal red blood cell mass was not significantly increased. T h e c u r r e n t 3-day h e m o r r h a g e protocol p r o d u c e d a 47% reduction of the fetal red blood cell mass. Unlike our previous study, the larger reduction in the fetal red blood cell mass of the current study resulted in marked and significantly greater fetal responses. In the h e m o r r h a g e d g r o u p the p o s t h e m o r r h a g e expansion of the red blood cell mass (6.8% _+0.9%/day) was almost twice that in the control group (3.5% _+ 0.5%/day), the fetal reticu!ocyte counts r e m a i n e d elevated t h r o u g h o u t the p o s t h e m o r r h a g e observation period, and the fetal hematocrit significantly increased. Thus, by p r o d u c i n g a larger and m o r e p r o l o n g e d reduction in red blood cell mass, fetal hematologic responses were clearly evident. In our previous study we hypothesized that the lack of a fetal reticulocyte and red blood cell mass responses may have b e e n due to the inability of the ovine fetal b o n e marrow to increase its red blood cell production rate as a result of the rapid rate of growth of the ovine fetus (3% to 4% per day) during this gestational time period. The results of the c u r r e n t study indicate that the ovine fetus can m o u n t a significant hematologic response if given an appropriate stimulus. F u r t h e r study is n e e d e d to determ i n e whether there is a threshold value for hematocrit or volume of blood loss to elicit a fetal response or w h e t h e r there is a graded response on the basis of severity of the hemorrhage.

Previous investigations have suggested that erythropoietin is a p o t e n t stimulus for red b l o o d celI production in the fetus and that the primary stimulus for erythropoietin release is a reduction in oxygen c o n t e n t f r o m either hypoxia or anemia. 2' 8. 20-22 In both this study and in our previous investigation there were large initial increases in fetal plasma erythropoietin concentrations during the immediate p o s t h e m o r r h a g e period. In the current study, howevm; a marked elevation in erythropoietin was sustained for 3 days instead of just 1 day as in the previous study. F u r t h e r m o r e , erythropoietin levels r e m a i n e d significantly elevated c o m p a r e d with control for an additional 3 days after the third h e m o r r h a g e . This o c c u r r e d during a period when the fetus r e m a i n e d both hypoxemic and anemic. During the 3-day h e m o r r h a g e fetal P o 2 progressively decreased, reaching a nadir of 5.3 m m Hg below control values at 24 hours after the third h e m o r rhage. During this same p e r i o d we n o t e d the greatest erythropoietin response. During the r e m a i n i n g 6 days after h e m o r r h a g e arterial P % r e m a i n e d depressed (averaging 2 m m H g below p r e h e m o r r h a g e values), T h e fact that erythropoietin levels r e t u r n e d toward n o r m a l u n d e r these conditions may suggest that the mild degree of hypoxia in this study may not be a sufficient stimulus for a sustained erythropoietin release at a high rate. This finding is similar to that of Kitanaka et al.(-'s who n o t e d that fetal erythropoietin levels r e t u r n e d to baseline after a sustained 10 m m H g reduction in fetal arterial Po9 in the absence of anemia. O n the o t h e r hand, assuming that the hematocrit can be used as a c r u d e index of oxygen content, the results of the c u r r e n t multivariate analysis suggest that both fetal oxygen tension (P%) and oxygen c o n t e n t are i m p o r t a n t determinants of the fetal erythropoietin response to anemia. Further, because the red blood cell mass in the h e m o r r h a g e d fetuses c o n t i n u e d to e x p a n d at a high rate although erythropoietin levels were declining, a sustained elevation in erythropoietin does not appear to be necessary for a c o n t i n u e d accelerated expansion of the fetal red b l o o d cell mass. This is similar to the posmatal response to erythropoietin. 24 Thus the net response to erythropoietin may n o t occur until several days after the m a x i m u m erythropoietin concentration. O n e of the most significant findings of this study was that the recovery of the red blood cell mass in the hemo r r h a g e d animals was significantly correlated with the 10-day m e a n fetal plasma iron concentration. This suggests that the ability of the fetus to respond to a large blood loss d e p e n d s on the availability of iron to the erythron. Additionally, we n o t e d that the fetal plasma iron c o n c e n t r a t i o n decreased significantly after the second h e m o r r h a g e and r e m a i n e d depressed t h r o u g h o u t the p o s t h e m o r r h a g e observational period. Previous work in the oyine fetus has shown that w h e n fetal erythropoiesis is stimulated the fetal plasma iron concentration falls.14 Unlike our findings, in that study plasma iron concentrations r e t u r n e d to baseline as the erythropoietin stimulus

Volume 174, Number 1, Part 1 ,~nj Obstet Gynecol

declined. T h e r e are two possible e x p l a n a t i o n s for the s u s t a i n e d r e d u c t i o n in p l a s m a i r o n t h a t we o b s e r v e d . Most of t h e p l a s m a iron, w h i c h is a l m o s t entirely b o u n d to t h e p l a s m a p r o t e i n t r a n s f e r r i n , was r e m o v e d d u r i n g t h e large h e m o r r h a g e or fetal p l a s m a i r o n was very low as a result of r a p i d fetal utilization. T h e possibility t h a t t h e majority of the p l a s m a i r o n was r e m o v e d a p p e a r s unlikely b e c a u s e t h e fetus actively t r a n s p o r t s i r o n f r o m t h e m a t e r nal to t h e fetal c o m p a r t m e n t 24 a n d b e c a u s e of t h e relatively large a m o u n t s o f i r o n available in i r o n stores. Further, b e c a u s e t h e r e was a significant c o r r e l a t i o n b e t w e e n t h e m e a n fetal p l a s m a i r o n c o n c e n t r a t i o n a n d the fetal r e d b l o o d cell p r o d u c t i o n rate, a m o r e likely e x p l a n a t i o n is t h a t fetal p l a s m a i r o n c o n c e n t r a t i o n s r e m a i n e d dep r e s s e d b e c a u s e o f r a p i d utilization. T h u s p l a c e n t a l i r o n t r a n s p o r t a n d fetal iron stores m a y b e i m p o r t a n t ratelimiting factors in fetal erythropoiesis. In summary, w h e n s u b j e c t e d to a severe h e m o r r h a g e over 3 days, t h e ovine fetus has a t r e m e n d o u s capacity to restore its circulating b l o o d volume, p l a s m a volume, a n d p l a s m a p r o t e i n s b a c k to n e a r - n o r m a l levels, Further, t h e fetus is c a p a b l e of nearly d o u b l i n g t h e e x p a n s i o n rate for its r e d b l o o d cell mass after severe h e m o r r h a g e . E r y t h r o p o i e t i n m a i n t a i n e d at greatly elevated levels does n o t a p p e a r to b e necessary for a c o n t i n u e d a u g m e n t a t i o n o f fetal r e d b l o o d cell p r o d u c t i o n , at least over a p e r i o d of 3 to 4 days after e r y t h r o p o i e t i n levels fall toward n o r m a l . A d e q u a t e fetal p l a s m a i r o n c o n c e n t r a t i o n s may b e t h e rate l i m i t i n g step for a c c e l e r a t e d erythropoiesis. I n spite of these adaptive responses, the h e m o r r h a g e d fetuses r e m a i n e d mildly h}qooxic d u r i n g the 7 day p o s t h e m o r r h a g e r e c o v e r y p e r i o d a n d t h e r e d b l o o d cell mass rem a i n e d far below t h a t in t h e c o n t r o l fetuses. T h e s e o b servations s u p p o r t the c o n c e p t t h a t r e d b l o o d cell transfusion may be n e e d e d to r e s t o r e the o x y g e n - c a r r y i n g capacity in t h e severely a n e m i c fetus. We t h a n k Luis Lizarraga a n d P a m N e w m a n for assist a n c e with daily a n i m a l care a n d m a i n t e n a n c e , Mary C o d d a n d J o h n H o e g e r for assistance with t h e 99mTc u s e d for fetal b l o o d v o l u m e m e a s u r e m e n t s , a n d R o b e r t L. S c h m i d t for analysis of t h e p l a s m a e r y t h r o p o i e t i n a n d iron levels. REFERENCES

1. Sebring ES, Polesky HF. Fetomaternal hemorrhage: incidence, risk factors, time of occurrence, and clinical effects. Transfusion 1990;30:344-57. 2. Nicolaides KH, Thilaganathan B, Mibashan RS. Cordocentesis in the investigation of fetal erythropoiesis. A_,aJ OBsTET G~,'ECOL 1989;161:197-200. 3. Gilbert ~WI, Scioscia AL. Spontaneous fetal-maternal hem-

Shields, Widness, and Brace

4. 5. 6.

7. 8. 9.

10.

11. 12.

13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

61

orrhage resulting in hydrops and elevated maternal serum alpha-fetoprotein levels. J Ultrasound Med 1991;10:645-8. Laube DW, Schauberger CW. Fetomaternal bleeding as a cause for "unexplained" fetal death. Obstet Gynecol 1982; 60:649-51. Adamson J, Hillman R. Blood volume and plasma protein replacement following acute blood loss in normal man. JAMA 1968;205:609-12. Block S, Rose JC, Rawashdeh R, Barnes K. Endocrine responses of fetal lambs to hemorrhage after (Zradrenergic receptor blockade. AM J OBSTET G~ECOL 1988;150:125662. Brace RA. Fetal blood volume responses to acute fetal hemorrhage. Circ Res 1983;52:730-4. Moritz KM, Cooper E, Wintour EM. The effect of hemorrhage on erythropoietin concentration in the mature fetus. J Dev Physiol 1992;17:157-61. Shields LE, Widness JA, Brace RA. Restoration of fetal red blood cells and plasma proteins after a moderately severe hemorrhage in the ovine fetus. A_,aJ OBSTETGY~ECOL1993; 169:1472-8. Nicolaides KH, WarenskiJC, Rodeck CH. The relationship of fetal plasma protein concentration and hemoglobin level to the development of hydrops in rhesus isoimmunization. AMJ OBSXETGYN~COL1985;152:341-4. Brace RA, Cheung CY. Fetal cardiovascular and endocrine responses to prolonged fetal hemorrhage. Am J Physiol 1986;251:R417-24. Brace RA. Blood volume in the fetus and methods for its measurement. In: Nathanielsz PW, ed. Animal models in fetal medicine: IV. Ithaca, New York: Perinatology Press, 1984:19-36. Jaynes P, Kapke GE Sulfasalazine interferes with coulometry of serum iron. Clin Chem 1981;27:202-3, Georgieff MK, Schmidt RL, Mills MM, Radnmr WJ, Widness JA. Fetal iron and cytochrome c status after intrauterine hypoxemia and erythropoietin administration. AmJ Physiol 1992;262:R485-91. Brace RA. Fetal blood volume responses to fetal hemorrhage: autonomic nervous contribution. J Dev Physiot 1987; 9:97-103. Brace RA. Mechanisms of fetal blood volume restoration after slow fetal hemorrhage. AmJ Physiol 1989;256:R1040-3. Brace RA, Cheung CY. Fetal blood volume restoration following rapid fetal hemorrhage, kanJ Physio11990;259:H56773. Davis LE, Hohimer AR. Hemodynamics and organ blood flow in fetal sheep subjected to chronic mlemia. AmJ Physiol 1991;261:R1542-8. Morit.z KM, Clemons G, Tangalakis K, Wintour EM. The effect of chronic and acute hemorrhage on erythropoietin in the neonatal lamb. J Dev Physiol 1992;77:709-17. Mori~ KM, Wintour EM. The effect of graded hemorrhage on erythropoietin production in the immature fetus. Clin Exp Pharmacol Physiol 1992;19:503-8. Widness JA, Teramo KA, Clemons GK, et al. Temporal response of immunoreactive erythropoietin to acute hypoxemia in fetal sheep. Pediatr Res 1986;20:15-9. Zanjani ED, AscensaoJL, McGlave PB, Banisadre M, Ash RC. Studies on the liver to kidney switch of erythropoietin production. J Clin Invest 1981;67:1183-8. Kitanaka T, Alonso JG, Gilbert RD, Siu BL, Clemons GK, Longo LD. Fetal responses to long-term hypoxemia in sheep. Am J Physiol 1989;256:RI 348-54. Gallagher PG, Ehrenkranz RA. Erythropoietin therapy for anemia of prematurity. Clin Perinatol 1993;20:169-91.