Journal of Applied Physiology AJP citation statistics
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 88: 1397-1406, 2000;
8750-7587/00 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Widness, J. A.
Right arrow Articles by Bard, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Widness, J. A.
Right arrow Articles by Bard, H.
Vol. 88, Issue 4, 1397-1406, April 2000

Adaptive responses during anemia and its correction in lambs

John A. Widness1, Lance S. Lowe1, Edward F. Bell1, Leon F. Burmeister2, Donald M. Mock3, James A. Kistard4, and Harry Bard5

Departments of 1 Pediatrics and 2 Preventive Medicine, College of Medicine, The University of Iowa, Iowa City 52242; and 4 Iowa Statewide Organ Procurement Organization, Iowa City, Iowa 52245; 3 Department of Pediatrics, University of Arkansas for Medical Sciences, Little Rock 72205; and the Arkansas Children's Hospital, Little Rock, Arkansas 72202; and 5 Department of Pediatrics, University of Montreal Research Center, Hôpital Sainte-Justine, Montreal, Quebec, Canada H3T 1C5


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

There is limited information available on which to base decisions regarding red blood cell (RBC) transfusion treatment in anemic newborn infants. Using a conscious newborn lamb model of progressive anemia, we sought to identify accessible metabolic and cardiovascular measures of hypoxia that might provide guidance in the management of anemic infants. We hypothesized that severe phlebotomy-induced isovolemic anemia and its reversal after RBC transfusion result in a defined pattern of adaptive responses. Anemia was produced over 2 days by serial phlebotomy (with plasma replacement) to Hb levels of 30-40 g/l. During the ensuing 2 days, Hb was restored to pretransfusion baseline levels by repeated RBC transfusion. Area-under-the-curve methodology was utilized for defining the Hb level at which individual study variables demonstrated significant change. Significant reciprocal changes (P < 0.05) of equivalent magnitude were observed during the phlebotomy and transfusion phases for cardiac output, plasma erythropoietin (Epo) concentration, oxygen extraction ratio, oxygen delivery, venous oxygen saturation, and blood lactate concentration. No significant change was observed in resting oxygen consumption. Cardiac output and plasma Epo concentration increased at Hb levels <75 g/l, oxygen delivery and oxygen extraction ratio decreased at Hb levels <60 g/l, and venous oxygen saturation decreased and blood lactate concentration increased at Hb levels <55 g/l. We speculate that plasma Epo and blood lactate concentrations may be useful measures of clinically significant anemia in infants and may indicate when an infant might benefit from a RBC transfusion.

transfusion; lactate; hypoxemia; newborn


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

DURING THE EARLY WEEKS AFTER birth, primates and other mammals experience a gradual decline in blood Hb concentration as a result of rapid growth, shortened red blood cell (RBC) survival, increased oxygen availability accompanying the onset of respiration, and a lower tissue oxygen set point (35). In term infants, this fall in Hb typically reaches its nadir between the second and third months of life, with the greatest and earliest decline observed in the smallest, most premature infants. In the intensive care of critically ill newborn infants, blood sampling associated with laboratory testing hastens the onset of neonatal anemia and exacerbates its severity. Among premature infants weighing <1,500 g at birth, daily phlebotomy blood loss of 4-5% of the blood volume is common in the first weeks of life when the severity of cardiopulmonary illness is greatest (4, 7, 28, 33, 36). Because the majority of RBC transfusions are administered during this same period (8, 26, 28, 36, 48), phlebotomy loss as a result of clinical monitoring is felt to be the primary cause of the large number of transfusions received by critically ill term and premature infants (4, 41, 50).

The fundamental basis for transfusion therapy is a decrease in oxygen transport capacity caused by a reduction in circulating RBC mass to the extent that cardiorespiratory status becomes impaired (45). Although the risks associated with RBC transfusion are well described, far less is known about the benefits of transfusing anemic infants at specific Hb levels (26, 41, 45, 50). Because of this uncertainty and because of developmental differences in oxygenation in neonates relative to adults (12), neonatal transfusion practices have varied widely over time and among institutions (28, 36, 48, 50).

Surprisingly, there have been no studies that have examined metabolic and cardiovascular adaptations resulting from slowly evolving anemia and its correction in unsedated neonatal animals in the absence of factors know to perturb oxygen delivery (DO2), e.g., hypoxia and decreased cardiac output (CO). Thus the objective of the present study was to identify clinically applicable indicators of the need for, and urgency of, RBC transfusions before the development of metabolic and cardiovascular decompensation accompanying critically low levels of DO2. We hypothesized that severe phlebotomy-induced isovolemic anemia, and its reversal after RBC transfusion, would result in a clearly defined pattern of adaptive cardiovascular and metabolic responses. Unanesthetized newborn lambs were selected for study because of their hematologic similarities to human infants.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Animals. After we received approval from the local Animal Care and Use Review Committee, neonatal lambs of mixed Dorset and Suffolk stock were obtained from a local breeder. To avoid anemia during the baseline period (34), pregnant ewes were treated with 1 g of iron dextran by intramuscular injection every other week beginning at 100 days of gestation (term = 145 days). Beginning 24-48 h after birth, lambs were housed in an indoor, environmentally controlled facility in which the ambient temperature was maintained at 19°C. Between study sessions, lambs were nursed by their mothers. Except for experimentally induced anemia, all lambs were deemed in good health throughout the study period.

Surgical procedures. Surgery was performed when the lambs were 2-4 days of age. After induction of anesthesia with pentobarbital (13 mg/kg iv), lambs were intubated and mechanically ventilated. Thereafter, anesthesia was maintained with 0.05-2.0% halothane by inhalation. Benzathine penicillin (300,000 U/kg im) was administered before the first skin incision. Through a left inguinal incision, Tygon catheters (ID 0.05 in., OD 0.09 in.; Norton Performance Plastics, Akron, OH) were placed in the femoral artery and vein and advanced with their distal tips estimated to be immediately above the diaphragm. A left anterolateral thoracotomy incision was made parallel to the fourth intercostal space, and the ribs were separated by gentle traction. The pericardial sac was entered, and the great vessels were identified and isolated. After ligation of the ductus arteriosus, a 3-0 Ethicon purse-string suture was placed in the adventitia of the main pulmonary artery ~1 cm from the pulmonic valve. A 14-gauge needle was used to puncture the center of the purse string, and a Tygon catheter (ID 0.04 in., OD 0.07 in.; Norton Performance Plastics) was advanced 3-4 cm distally before being secured to the wall of the pulmonary artery. An appropriately sized 10- to 14-mm transit time Doppler flow probe (Transonic Systems, Ithaca, NY) was placed around the main pulmonary artery to monitor CO. Pleural, muscular, and dermal, but not pericardial, layers were closed separately with the catheters, wires, and a 10-Fr chest tube tunneled subcutaneously and brought out along the left flank through a small incision just below the rib cage. Wire electrodes were implanted subcutaneously for monitoring respiratory movements. Two were placed along opposite midaxillary regions of the chest and the third in the left inguinal region. To ensure postoperative lung expansion after chest closure, negative pressure was applied to the chest tube. All catheters and wires were protected in a cloth pouch secured to skin covering the flank incision and were further protected by an elastic cloth bandage wrapped loosely about the lower thorax and upper abdomen. Surgery lasted <3 h. Estimated blood loss was always <20 ml. After the onset of spontaneous respiration, lambs were extubated, and chest tubes were removed. Within 3 h of the completion of surgery, the lambs were standing and able to nurse.

Experimental protocol. After surgery, the lambs were allowed 2-3 days to recover before initiation of the study. After baseline data were collected on day 0, lambs were studied during a 2-day phlebotomy phase followed by a 2-day transfusion phase (Fig. 1). During the phlebotomy phase, lambs underwent serial phlebotomies with the goal of decreasing the Hb level by ~20% at each step. Each phlebotomy was accomplished over a 10- to 60-min period by isovolemic exchange transfusion with 16.9 ± 2.64 ml/kg of cross-matched adult sheep plasma. A Hb level of 30-40 g/l was chosen as the nadir for study because of the high mortality associated with levels below this (11, 18, 52, 53). During the transfusion phase, anemia was reversed in a stepwise fashion until Hb levels returned to baseline values. Transfusions were performed by using 4.8 ± 0.31 ml/kg of cross-matched adult sheep packed erythrocytes (hematocrit approx 90%) stored for <7 days in citrate phosphate dextrose adenine (CPDA). Transfusions were administered over 1.5-2 h, just before the lamb was returned to its mother but after data collection and blood sampling. Throughout the 4-day study period, benzathine penicillin (300,000 U/kg im) was administered every other day.


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1.   Study design showing degree of anemia that was present over time. Dashed line, mean Hb concentration over time for 10 study lambs. During 2-day phlebotomy phase (P-1 to P-5), lambs were rendered anemic by repeated phlebotomy. During subsequent 2-day transfusion phase (T-1 to T-4), anemia was reversed by repeated administration of autologous adult sheep red blood cell transfusions. Metabolic and cardiovascular indexes were recorded at times indicated by solid circles.

Lambs were weighed daily. Data collection periods during both the phlebotomy and transfusion phases consisted of 2-h sessions during which lambs were removed from their mothers and brought to the laboratory where they were maintained suspended in a cloth sling hung in a transport cart. Each of the 2-h study sessions was begun no sooner than 2 h after the previous phlebotomy or transfusion procedure (see below). Being placed in the sling had a calming effect on the lambs. After a 30- to 60-min period of acclimatization, during which lambs were left undisturbed, continuous cardiovascular data were recorded for ~60 min by using an eight-channel polygraph recorder (Gould, Cleveland, OH) interfaced with a data-acquisition software program (LabTech Notebook Data Acquisition Software, Laboratory Technologies, Wilmington, MA). Directly measured data included heart rate, mean arterial pressure, central venous pressure, and pulmonary artery blood flow. Because the ductus arteriosus had been ligated, pulmonary artery blood flow was used as a surrogate for CO. Stroke volume, systemic DO2, oxygen consumption (VO2), oxygen extraction ratio (ERO2), and peripheral vascular resistance (PVR) were calculated from the directly measured variables as follows
Stroke volume = CO ÷ heart rate (in ml/kg)

<A><AC>D</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB> = CO × Sa<SUB>O<SUB>2</SUB></SUB> × Hb × 1.39 ml O<SUB>2</SUB>/g Hb (in ml O<SUB>2</SUB> ⋅ kg<SUP>−1</SUP> ⋅ min<SUP>−1</SUP>)

<A><AC>V</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB> = (CO × Hb × 1.39 ml O<SUB>2</SUB>/g Hb) 

× (Sa<SUB>O<SUB>2</SUB></SUB> − Sv<SUB>O<SUB>2</SUB></SUB>) (in ml O<SUB>2</SUB> ⋅ kg<SUP>−1</SUP> ⋅ min<SUP>−1</SUP>)

ER<SUB>O<SUB>2</SUB></SUB> = <A><AC>V</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB> ÷ <A><AC>D</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB> (no units)

PVR = (arterial − central venous blood pressure) ÷ CO
where SaO2 is arterial oxygen saturation and SvO2 is central venous oxygen saturation.

Respiratory rate as measured by chest impedance was recorded continuously over a 15- to 20-min portion of the same data collection period by using a Hewlett-Packard respiratory monitor (model 78212D) with a 50- to 60-Hz transducer (Hewlett-Packard, Palo Alto, CA). Rectal temperature was recorded by using a YSI 2600 O2/Temp Meter (Yellow Springs Instruments, Yellow Springs, OH).

All phlebotomies and transfusions took place at the end of each study after all measurements had been successfully completed. Afterwards, the lambs were returned to their mothers for a minimum of 2 h. The only exception to this was on the second study day; after completion of the last phlebotomy and while experiencing the most profound level of anemia, lambs were not returned to their mothers after data collection but, instead, were first transfused and restudied 2 h later. This was done as a precautionary measure to ensure survival.

At the completion of each 60-min recording period, simultaneous systemic and pulmonary arterial blood samples were collected anaerobically in heparinized syringes, placed on wet ice, and analyzed within 5 min of collection for Hb, pH, PCO2, PO2, oxygen saturation, and whole blood lactate. Additional blood was taken each morning with the first sampling for determination of plasma erythropoietin (Epo), whole blood 2,3-diphosphoglycerate (2,3-DPG), and PO2 at which Hb is 50% saturated with oxygen (P50). After each blood transfusion, P50 was again measured. Because endogenous plasma Epo levels require a minimum of several hours to respond to changes in oxygenation before approaching steady-state levels (20), more frequent blood sampling for plasma Epo determinations was deemed inadvisable. Total body circulating RBC volume was measured three times: at the beginning of the study, just before the last phlebotomy, and after the last blood transfusion.

Laboratory determinations. Hb, SvO2, and SaO2 were determined by using an IL-482 CO-oximeter (Instrumentation Laboratories, Lexington, MA). Blood samples were analyzed for PO2, PCO2, and pH by using an IL 1303 blood-gas analyzer and corrected to normal body temperature for lambs, i.e., to 39.5°C. P50 was determined by using the above-mentioned blood-gas analyzer and CO-oximeter and an IL-237 tonometer at PCO2 = 40 Torr. P50 data were corrected to pH of 7.40 and normal lamb body temperature (23). The concentration of 2,3-DPG was determined spectrophotometrically on supernatant fractions of heparinized whole blood precipitated with 5% TCA (2:1) as described by Keitt (22). Lactate was measured on 50-µl whole blood samples by using a YSI model 27 analyzer (Yellow Springs Instruments).

Epo was measured on 100-µl plasma samples in triplicate by using a double-antibody RIA (47). Linear values for sheep Epo are obtained between 10 and 450 mU/ml by using sheep reference standards (EpoConn, Connaught Laboratories, Willowdale, Ontario). For each individual lamb study, samples were run in the same assay. Intra-assay coefficients of variation for three pools of plasma spanning the useful range of the RIA ranged from 4.7 to 11.1%.

Total RBC volume was measured by using autologous erythrocytes with [14C]cyanate (New England Nuclear, Boston, MA) as described by Mock et al. (31). Circulating RBC volume determined with [14C]cyanate agrees almost perfectly with RBC volume determined with 51Cr, i.e., correlation coefficient = 0.99.

Data handling and statistical analysis. The area-under-the-curve methodology of Matthews et al. (29) was combined with the interpolation method of Cilley et al. (11) as the primary statistical method for estimating the Hb level at which study variables demonstrated change. In applying this methodology, the magnitude of response of each study variable was quantitated for individual lambs by measuring the area under the curve for the variable plotted against 5-g/l Hb intervals. This was done separately for both the phlebotomy and transfusion phases. As illustrated for a representative animal with the use of whole blood lactate measured during the phlebotomy phase as an example (Fig. 2), Hb intervals with missing lactate values were conservatively estimated by substituting lactate values from the next higher 5-g/l Hb interval in which a measured lactate value was available. In inspecting Fig. 2, it is evident that the application of this methodology results in underestimation of the Hb concentration at which a change occurs. The only parameters included in the final analyses defining the precise Hb level at which change occurred were those specifically perturbed by anemia, i.e., those demonstrating equivalent and reciprocal area-under-the-curve responses during the phlebotomy and transfusion at the same Hb interval. These parameters included SvO2, CO, ERO2, plasma Epo, whole blood lactate, and VO2. The average values for the area-under-the-curve analyses of the combined phlebotomy and transfusion phases were used in the final analyses. For purposes of graphic illustration, the area-under-the-curve units were converted to SI units.


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2.   Example illustrating how area-under-the-curve data were determined for selected blood lactate during transfusion phase for a representative lamb at 5-g/l Hb intervals. Vertical lines separate discrete 5-g Hb/l intervals integrated to determine areas under the curve. Although baseline period is represented here as a range, Hb interval from 95 to 100 g/l was used as control interval in Dunnett's post hoc test applied for those variables in which repeated measures ANOVA F statistic was significant.

Because RBC volume measurements were taken during only 3 of the 10 study periods, the area-under-the-curve methodology was deemed inappropriate for examining the utility of RBC volume measurements in predicting change in Hb or in testing for associations with other oxygenation parameters. Instead, correlation coefficients of RBC volume and oxygenation parameters demonstrating change in the area-under-the-curve analysis were determined separately for each lamb. The mean group correlation coefficient for each oxygenation parameter was then tested for significance by a one-sample t-test. An identical correlation analysis with oxygenation study parameters was performed by substituting the simultaneously determined Hb concentration. To avoid bias, only Hb data obtained at the same three study periods as the RBC volume measurements were included, i.e., baseline, severe anemia, and complete recovery. The correlation coefficients of those variables demonstrating significant association with both RBC volume and Hb were compared by paired t-test to determine if RBC volume or Hb demonstrated a significantly better correlation.

Results are expressed as means ± SE. A P value <0.05 (two-tailed) was considered statistically significant. Statistical computations were done by using commercially available software (Statview II, Abacus Concepts, Berkeley, CA). Variables demonstrating nonnormal distributions were transformed logarithmically before testing. Single-factor ANOVA with repeated measures was used to test for differences at specified Hb intervals indicated in Fig. 2 or the specified sampling periods indicated in Fig. 1. Dunnett's post hoc testing procedure was performed by using the baseline period data for comparisons with all other periods for significant F tests.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The mean (± SE) age of the lambs at the time of the first phlebotomy study was 5.7 ± 0.2 days. There was no significant change in body weight during the study period; body weight before the first phlebotomy was 5.74 ± 0.25 kg, and the weight after the last transfusion was 5.75 ± 0.33 kg. Autopsies done at the completion of each study demonstrated proper positioning of all catheters and flow probes.

Study periods during which variables demonstrated significant change. Hb concentration decreased from baseline values of 96.9 ± 3.3 g/l to a nadir of 36.0 ± 1.3 g/l on the second day. After the last transfusion on day 4, Hb returned to near baseline levels, i.e., 93.9 ± 1.4 g/l (Fig. 3A). RBC volume measurements taken at baseline, during anemia, and after recovery were 27.1 ± 4.6, 11.1 ± 0.7, and 24.6 ± 1.6 ml/kg, respectively. These changes were of similar proportion to those for Hb at the same study periods.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3.   Mean (± SE) laboratory data in 10 newborn lambs during phlebotomy and transfusion study phases for Hb concentration (A), venous oxygen saturation (SvO2; B), plasma erythropoietin (Epo; C), and blood lactate (D). Analysis by single-factor repeated measures ANOVA revealed significant differences from baseline levels for all 4 measurements. Results of Dunnett's post hoc significance are as follows: * P < 0.05 and dagger  P < 0.01.

Significant progressive (P < 0.05) changes were observed for SvO2, plasma Epo, and blood lactate as anemia intensified. Moreover, reciprocal changes were observed in all three as Hb levels increased during the study's transfusion phase. During the phlebotomy phase, all lambs experienced a decrease in SvO2 from baseline values of 48.1 ± 10.5%. In six, the decrease was >2 SD, and in three of the remaining four the decrease was >1 SD (Fig. 3B). By the end of the phlebotomy phase, 9 of the 10 lambs manifested a progressive >3 SD increase in plasma Epo from baseline, whereas Epo levels in the remaining animal increased by >2 SD (Fig. 3C). The increase observed in blood lactate level was more variable than that in plasma Epo. Seven lambs increased blood lactate levels by >3 SD of baseline, one increased by >2 SD, whereas the remaining two manifested no measurable change in lactate (Fig. 3D).

Arterial pH, PCO2, and PO2 values all fell within normal limits during the baseline period (i.e., 7.39 ± 0.01, 44.6 ± 1.5 Torr, and 80.4 ± 16.3 Torr, respectively) and demonstrated little change thereafter (Fig. 4, A-C). Only a brief, small but significant, decrease in PCO2 occurred coincident with the period of profound anemia during which blood lactate levels increased. Neither pH or PO2 changed significantly during either the phlebotomy or transfusion phases.


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4.   Mean (± SE) arterial pH (A), PCO2 (PaCO2; B), and PO2 (PaO2; C) data in 10 newborn lambs during phlebotomy and transfusion study phases. Analysis by single-factor repeated measures ANOVA revealed no significant differences from baseline levels for arterial pH or PaO2. Results of Dunnett's post hoc significance are as follows: dagger  P < 0.01.

Changes that occurred in the affinity of Hb for oxygen and in 2,3-DPG levels during the course of study did not follow the pattern observed for Hb. The significant decline was first observed in 2,3-DPG just before the first RBC transfusion and continued throughout the remainder of the study (Fig. 5A). This was followed by a significant increase in P50 coincident with the administration of adult sheep blood during the study's transfusion phase (Fig. 5B).


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 5.   Mean (± SE) data of 2,3-diphosphoglycerate (2,3-DPG; A) and PO2 at which Hb is 50% saturated with oxygen (P50; B) in 10 newborn lambs during phlebotomy and transfusion study phases. Analysis by single-factor repeated measures ANOVA revealed significant differences from baseline levels for both measurements. Results of Dunnett's post hoc significance are as follows: * P < 0.05 and dagger  P < 0.01.

Several cardiovascular and respiratory study parameters demonstrated no significant change during the phlebotomy and transfusion phases, whereas others demonstrated marked changes. Although respiratory rate decreased during anemia (Fig. 6A), inspection of individual pneumograms revealed no episodes of apnea, tachypnea, or periodic breathing. Baseline pulmonary arterial pressure, central venous pressure, and body temperature did not change during the course of the study (not shown) and were within expected limits, i.e., 22.8 ± 6.9 mmHg, 3 ± 2 mmHg, and 39.6 ± 0.5°C, respectively. Although tidal volume was not measured, lambs may have breathed more slowly and deeply during the periods of most profound anemia, thereby leading to an increase in minute ventilation. Despite the decrease in respiratory rate, this speculation is consistent with the decrease observed in arterial PCO2. Arterial blood pressure followed a similar pattern to that of respiration (Fig. 6B). PVR decreased significantly as Hb levels fell (data not shown). The nonsignificant trend toward an increase in heart rate (Fig. 6C) combined with a significant rise in stroke volume (not shown) resulted in a gradual, but significant, rise in CO during the phlebotomy phase (Fig. 6D).


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 6.   Mean (± SE) cardiopulmonary status data in 10 newborn lambs during phlebotomy and transfusion study phases for respiratory rate (A), arterial blood pressure (B), heart rate (C), and cardiac output (D). bpm, Beats/min. Analysis by single-factor repeated measures ANOVA revealed significant differences from baseline levels for all variables except heart rate. Results of Dunnett's post hoc significance are as follows: * P < 0.05 and dagger  P < 0.01.

Of the calculated cardiovascular parameters indicative of oxygenation status, DO2 and ERO2 demonstrated significant progressive change as anemia worsened during the phlebotomy phase and reverted back to baseline levels as Hb levels normalized. The pattern of decrease of DO2 levels (Fig. 7A) as Hb concentration fell approximately mirrored the increase in ERO2 levels (Fig. 7B). Despite a pronounced decrease in DO2 and increase in ERO2 as Hb reached its nadir, VO2 did not change (Fig. 7C).


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 7.   Mean (± SE) indicators of oxygenation status data in 10 newborn lambs during phlebotomy and transfusion study phases for systemic oxygen delivery (A), oxygen extraction ratio (B), and oxygen consumption (C). Analysis by single-factor repeated measures ANOVA revealed significant differences from baseline levels for oxygen delivery and for oxygen extraction ratio but not for oxygen consumption. Results of Dunnett's post hoc significance are as follows: * P < 0.05 and dagger  P < 0.01.

Area-under-the-curve analysis: the primary outcome methodology. The area-under-the-curve analysis used in estimating the Hb level at which statistically significant change occurred was reserved only for those oxygenation parameters demonstrating equivalent and reciprocal responses at the same Hb interval during the phlebotomy and transfusion phases. Because the Hb nadir was not identical in all study lambs, the spectrum of Hb intervals included in the area-under-the-curve analyses was indicative of less anemia than was the case with individual lambs. For those oxygenation parameters demonstrating progressive area-under-the-curve change from the baseline Hb interval, the Hb intervals at which statistical significance changes were identified were as follows: CO and plasma Epo increased at Hb <75 g/l; DO2 decreased and ERO2 increased at Hb <60 g/l; SvO2 decreased and blood lactate increased at Hb levels <55 g/l (Fig. 8). VO2 did not demonstrate significant change at the range of Hb levels studied (data not shown). Although each of these parameters demonstrated significant change as anemia worsened and returned to baseline as anemia was corrected, the magnitude of and the progressive nature of change relative to baseline levels was most clearly evident for plasma Epo and blood lactate. Moreover, the variance of these two parameters was equivalent to or smaller than that of the other four oxygenation parameters.


View larger version (72K):
[in this window]
[in a new window]
  		Fig. 8.   Mean (± SE) results of area-under-the-curve analysis for cardiac output (A), plasma erythropoietin (Epo; B), oxygen extraction ratio (C), oxygen delivery (D), SvO2 (E), and blood lactate (F). Single-factor repeated measures ANOVA revealed significant differences from baseline levels for all variables shown. Results of Dunnett's post hoc significance testing with baseline values are as follows: * P < 0.05 and dagger  P < 0.01.

Because the significant rise in P50 after transfusion with adult blood could have exerted an independent effect on the oxygenation parameters, a statistical power analysis was performed to estimate the magnitude of the error encountered in comparing the phlebotomy and transfusion phases of the experiment. To do this, the standard deviation of the mean difference between the two study phases was compared at each of the Hb intervals. The mean value for the mean standard deviations for each interval of the study variables shown in Fig. 8 was derived and multiplied by the appropriate statistical constants when it was assumed that power = 0.80 and alpha  = 0.05 for 10 study subjects. When this value was expressed as a percentage of the mean area under the curve at each Hb interval, the mean variability observed ranged from 19.5 to 26.5% for all of the variables in Fig. 8, except lactate, which was slightly greater, i.e., 33%.

Comparison of total body RBC volume and Hb with oxygenation parameters. In the univariate correlation analysis, both RBC volume and Hb concentration demonstrated significant correlation with several of the oxygenation parameters that demonstrated change in the area-under-the-curve analysis (Table 1). When the degree of association of RBC volume and Hb level was compared, RBC volume was not associated to a more significant degree than Hb with DO2, CO, plasma Epo, SvO2, or ERO2.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Association of red blood cell volume and Hb with oxygenation parameters studied


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The newborn lamb model utilized here simulates, albeit on a shorter time scale, anemia gradually developing in infants as a result of cumulative blood loss for clinical monitoring. The inclusion of events during both phlebotomy and transfusion adds to the specificity, identification, and quantification of the adaptive responses that might serve as indicators of the need for blood transfusion. Few preclinical studies in nonanesthetized newborn animals have examined the relationship of oxygenation parameters in the presence of severe neonatal anemia. None has included the results of plasma Epo, a sensitive and specific indicator of tissue hypoxia, and none has evaluated adaptive responses during both phlebotomy and transfusion over an extended period. In the present study, our hypothesis that a progression of adaptive cardiovascular and metabolic responses would be observed in response to stepwise progressive phlebotomy-induced isovolemic anemia in newborn lambs and their subsequent reversal by transfusion was confirmed. The sequence of responses indicative of progressive tissue hypoxia accompanying a 60% decrease from baseline Hb level was initiated with a fall in CO and a marked rise in plasma Epo. This was followed by an increase in ERO2 and a decrease in DO2. A decrease in SvO2 and an increase in blood lactate were the last changes detected. The severity of anemia was never so marked that a decline in resting VO2 was detected.

Progression of events with worsening isovolemic anemia and its correction. Tissue hypoxia attributable to anemia is difficult to assess at the cellular level in in vivo studies. Nonetheless, the progression we observed in adaptive cardiovascular and metabolic changes as anemia worsened supports the anaerobic threshold hypothesis (11, 17, 18, 52). According to this hypothesis, homeostatic adaptations take place in response to progressive tissue hypoxia to maintain VO2 within fixed, narrowly defined normal limits, thus avoiding cardiovascular collapse and imminent death as tissue DO2 declines to critically low levels. Caution, however, is required to avoid overdrawing conclusions regarding the exact sequence of the changes in the cardiovascular and oxygenation study variables. This is necessary because the calculated parameters are based on those directly measured, e.g., DO2 is determined based on CO, ERO2 is determined based on SvO2, and so forth. Our discussion of the sequence of events needs to be viewed in this context with appropriate allowance for the degree of uncertainty identified in the error estimation performed with the area-under-the-curve analyses.

In the present study, the first of the homeostatic changes accompanying the decline in Hb was increases in CO and plasma Epo. These events occurred as the lambs' Hb levels fell <75 g/l, with both returning to normal after the restoration of Hb levels to baseline during the transfusion phase. Consistent with the findings of Holzman et al. (18) in newborn lambs, the increase we observed in CO was predominantly attributable to an increase in stroke volume, with a lesser, nonsignificant increase in heart rate. The decline in PVR as a result of declining Hb levels, and therefore in decreasing blood viscosity, likely contributed to the increase observed in stroke volume. In clinical studies of human neonates, small but significant changes (<7-15 beats/min) in heart rate have been noted by some (5, 25), but not all (24), investigators as anemia worsens. Unfortunately, these changes in neonatal heart rate have all been within the normal range. Based on the present data, we speculate that this is attributable to less profound anemia than that observed in the present study or to a lesser role of heart rate in controlling CO in the human infant. If the latter is true, continuous heart rate monitoring in human neonates will be of little value in assessing their transfusion needs.

In contrast to the short-term adaptive responses in CO to anemia, the increase in plasma Epo concentration represents the body's long-term Epo adaptation directed toward increasing Hb. Although the precise cellular and molecular hypoxia-mediated events leading to increases in Epo gene expression remain unknown, there is agreement that most Epo production in the adult takes place in the kidney (and to a lesser degree in the liver) (15). In contrast, in the fetus, and perhaps in the young infant (13), the liver is the primary organ of Epo production (15). The progressive increases in plasma Epo first detected in the lambs at Hb levels <75 g/l were in marked contrast to the more modest changes observed in CO and in the other cardiovascular parameters. Between Hb levels of 75 and 40 g/l, the average plasma Epo increased 43-fold relative to baseline (range: 5- to 233-fold), i.e., logarithmically transformed Epo values nearly doubled (Fig. 8B). These increases in plasma Epo concentration are of similar magnitude to those associated with both acute hypoxemia and anemia in fetal sheep (16, 38, 47) and with hypoxemia and anemia of indeterminate duration in human fetuses (30, 43, 49). They differ markedly, however, from the much less marked increases in plasma Epo levels observed among chronically anemic human neonates (9, 24, 25, 30). This difference could be due to the lower oxygen environment of the fetus relative to the newborn.

Compensatory cardiovascular mechanisms in the lambs maintained VO2 at normal levels without a decrease in DO2 or an increase in ERO2 until Hb levels decreased <60 g/l. At the nadir Hb, DO2 had declined to 12 ± 1.3 ml · kg-1 · min-1, a value just above the critical level of 10-12 ml · kg-1 · min-1 at which lactate increases and VO2 decreases in large animals (10, 11, 52). At the lowest Hb levels, ERO2 values reached 0.70-0.80, a value much higher than that reported in critically ill adult humans but similar to those reported for severely anemic fetal and neonatal lambs (3, 14, 17, 42). In humans, Wilkerson et al. (51, 52) reported that ERO2 values exceeding 0.50 were observed only among the critically ill at high risk of death. As a result, elevated ERO2 has been suggested as a valid indicator of the need for blood transfusion (39, 52).

SvO2 and blood lactate levels were the last two oxygenation parameters to demonstrate significant change with worsening anemia. This occurred at Hb levels <55 g/l. Although SvO2 is easily measured in critically ill adults and older children (42a), its determination in newborns is technically challenging and has yet to be demonstrated as safe and effective (46). Blood lactate levels of 4.3 ± 1.0 mM observed at the Hb nadir in the present study were substantially greater than the 1-2 mM reported in human infants with isolated anemia immediately before blood transfusion (1, 19, 37, 40) but similar to levels observed in anemic large animals as VO2 decreases below critical levels of DO2 (11, 27, 52). The relatively low blood lactate levels reported in association with anemia in infants fall in the upper normal range and suggest that RBC transfusions are being administered well in advance of significantly impaired tissue oxygenation. In contrast, anemic infants with blood lactate levels similar to those of our lambs as the lambs' Hb levels approached their nadir merit careful evaluation and strong consideration of transfusion.

Despite the marked increase in blood lactate observed as the nadir of Hb was approached, resting VO2 remained unchanged. Previous acute normovolemic anemia studies have demonstrated that blood lactate levels increase as VO2 declines in close proximity to the point when critically low levels of DO2 are reached (11, 17, 18, 52). Although DO2 in the present study approximated critically low values reported by others, a significant decrease in VO2 in association with the rise in blood lactate was not demonstrated. This could have been due to our failure to directly measure VO2 and DO2 by independent means (11).

Correlation of RBC volume and Hb with indicators of tissue oxygenation. Total circulating RBC volume and Hb were compared for their respective associations with indicators of tissue oxygenation. Although not reaching statistical significance for any of the parameters tested, the correlation coefficients tended to be higher for Hb than for RBC volume. This finding differs from that of Jones et al. (21), who reported that RBC volume measurement provides a more accurate indication of the adequacy of tissue oxygenation than either Hb or hematocrit in human adults and infants. This discrepancy could be due to differences in the severity of illness of the study subjects, to species differences, or to methodological differences.

Changes in oxyhemoglobin affinity. The changes observed in P50 and RBC 2,3-DPG were not equivalent during the phlebotomy and transfusion phases. Because the rise observed in P50 during the transfusion phase will tend to increase DO2 and the release of oxygen to tissues, this disparity in P50 during the two study phases also serves to complicate interpretation of several of the oxygenation parameters, e.g., SvO2 (and therefore ERO2) and arterial and venous PO2. As suggested by the results of the area-under-the-curve error analysis, this could have contributed in making the relationship between Hb and the oxygenation parameters less precise.

In a previous report comparing two groups of newborn lambs made anemic by isovolemic transfusion with RBCs possessing low or high Hb-oxygen affinity (P50 = 32.2 ± 2.5 and 19 ± 1.06 Torr, respectively), Hb levels as low as 40 g/l were tolerated in the low-oxygen affinity group without a decrease in VO2, whereas lambs in the high-oxygen affinity group experienced decreased VO2 (44). Thus the relatively high P50 values (i.e., low-Hb-oxygen affinity) and our finding of no change in VO2 in the lambs in the present study are consistent with the previous report's low-Hb-oxygen affinity lambs.

The changes that took place in P50 and RBC 2,3-DPG during the phlebotomy and transfusion phases likely reflect developmental events and/or were the result of transfusion with adult erythrocytes with their higher P50 and lower RBC 2,3-DPG concentration (2, 27). During the first week after birth, newborn lambs rapidly switch from a high- to low-Hb-oxygen affinity with their P50 increasing from ~18 to 30 Torr (2, 27). This change is mainly due to the rapid rise in RBC 2,3-DPG, which results in a commensurate decrease in oxygen affinity before the switch to adult Hb has been completed. Afterward, as the percentage of adult Hb further increases, levels of DPG decrease, and the postnatal increase in P50 reaches a plateau by 10-14 days (2).

Comparison of anemia in lamb and infants. Although lambs and infants share developmental similarities of the cardiovascular and hematopoietic systems (27, 32), quantitative differences exist that need to be considered in extrapolating from one species to the other. With lambs' higher postnatal P50 values (despite lower 2,3-DPG levels and differences in Hb structure), they may be better adapted for readily releasing oxygen to tissues and thus function more adequately at Hb levels that are normally 20-30% lower than their human infant counterparts (27, 34). Nonetheless, from a qualitative perspective, the sequence of adaptive changes to anemia observed for newborn lambs in the present study is similar to that reported in other newborn and adult animal and adult human studies. Furthermore, it must be reiterated that the present study addresses only the situation of isovolemic anemia in otherwise healthy neonatal lambs. Results of these studies must, therefore, be carefully extrapolated to animals or infants with superimposed acute blood loss, cardiac and/or pulmonary disease, sepsis, or surgery.

In summary, establishment of sound, experimentally based criteria for guiding decisions to administer RBC transfusions to anemic human infants is an important and pressing need. Literature in adults indicating that moderately lower levels of Hb are well tolerated in otherwise healthy individuals has not previously been extended to include newborns. Moreover, there have been no preclinical studies evaluating adaptive responses to anemia over extended periods during both phlebotomy and transfusion phases. Studies in newborn animals provide relevant comparative data yet avoid the ethical difficulties inherent in infant studies. In this study of isovolemic anemia in newborn lambs, we observed that Hb values interpreted in conjunction with adaptive cardiovascular and metabolic indicators of hypoxia, in particular within plasma Epo and blood lactate, provide an informative basis for assessing tissue oxygenation and thus the potential immediate benefit of erythrocyte transfusions. Measurement of blood lactate levels and plasma Epo have advantages over measurements of other responses, as both are easily performed, results of both are readily available in minutes to hours and, as shown by the present data, both demonstrate marked progressive changes in response to the development of anemia that are reversed after transfusion. Although quantitative differences in oxygenation parameters existing between the newborn lambs and human infants preclude precise extrapolation from one species to the other, qualitative similarities exist. Thus the present data in lambs support the assumptions on which infant erythrocyte transfusions are based and merit reexamination. Reasons for the marked elevations in plasma Epo and in blood lactate in anemic and hypoxemic human fetuses, but not neonates, remain uncertain. One possibility is that our present reliance on Hb levels in isolation from other potentially informative laboratory indicators of the need for transfusion in infants, e.g., plasma Epo and blood lactate, is suboptimal. Before firm conclusions can be reached regarding this speculation, studies evaluating parameters of tissue oxygenation such as those included in this and other studies, carried out under a variety of clinical conditions, are needed.


    ACKNOWLEDGEMENTS

We acknowledge the technical contributions of Delores Cordle, Robert Schmidt, Gary Lankford, David Viet, and Barbara Stewart; the secretarial contributions of Mark A. Hart; and helpful comments and critical review of the manuscript by Dr. Ronald G. Strauss.


    FOOTNOTES

This work was supported by National Heart, Lung, and Blood Institute Grant P01-HL-46925 and Medical Research Council of Canada Grant MT-11552.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests and other correspondence: J. A. Widness, Univ. of Iowa Hospitals & Clinics, 200 Hawkins Dr., W222-1 GH, Iowa City, IA 52242-1083 (E-mail: john-widness{at}uiowa.edu).

Received 30 November 1998; accepted in final form 28 December 1999.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Bard, H, Fouron J-C, Chessex P, and Widness JA. Myocardial, erythropoietic, and metabolic adaptations to anemia of prematurity in infants with bronchopulmonary dysplasia. Pediatr Res 132: 630-634, 1997.

2.   Bard, H, Fouron JC, Grothe AM, Soukini MA, and Cornet C. The adaptation of the fetal red cells to newborn lambs to extrauterine life: the role of 2,3-diphosphoglycerate and adult hemoglobin. Pediatr Res 10: 823-825, 1976[ISI][Medline].

3.   Bernstein, D, Teitel DF, and Rudolph AM. Chronic anemia in the newborn lamb: cardiovascular adaptations and comparison to chronic hypoxemia. Pediatr Res 23: 621-627, 1988[ISI][Medline].

4.   Bifano, EM, and Curran TR. Minimizing donor blood exposure in the neonatal intensive care unit. Clin Perinatol 22: 657-669, 1995[ISI][Medline].

5.   Bifano, EM, Smith F, and Borer J. The relationship between determinants of oxygen delivery and respiratory abnormalities in preterm infants with anemia. J Pediatr 120: 292-296, 1991.

7.   Blanchette, VS, Hume HA, Levy GJ, Luban NLC, and Strauss RG. Guidelines for auditing pediatric blood transfusion practices. Am J Dis Child 145: 787-796, 1991[Abstract].

8.   Brown, MS, Berman ER, and Luckey D. Prediction of need for transfusion during anemia of prematurity. J Pediatr 116: 773-778, 1990[ISI][Medline].

9.   Brown, MS, Phibbs RH, Garcia JF, and Dallman PR. Postnatal changes in erythropoietin levels in untransfused premature infants. J Pediatr 103: 612-617, 1983[ISI][Medline].

10.   Cilley, RE, Polley TZ, Zwischenberger JB, Toomasian JM, and Bartlett RH. Independent measurement of oxygen consumption and oxygen delivery. J Surg Res 47: 242-247, 1989[ISI][Medline].

11.   Cilley, RE, Scharenberg AM, Bongiorno PF, Guire KE, and Bartlett RH. Low oxygen delivery produced by anemia, hypoxia, and low cardiac output. J Surg Res 51: 425-433, 1991[ISI][Medline].

12.   Dallman, PR. Anemia of prematurity. Annu Rev Med 32: 143-160, 1981[ISI][Medline].

13.   Dame, C, Fahnenstich H, Freitag P, Hofmann D, Abdul-Nour T, Bartmann P, and Fandrey J. Erythropoietin mRNA expression in human fetal and neonatal tissue. Blood 92: 3218-3225, 1998[Abstract/Free Full Text].

14.   Edelstone, DL, Darby MJ, Bass K, and Miller K. Effects of reductions in hemoglobin-oxygen affinity and hematocrit level on oxygen consumption and acid-base state in fetal lambs. Am J Obstet Gynecol 160: 820-828, 1989[ISI][Medline].

15.   Fisher, JW. Erythropoietin: physiologic and pharmacologic aspects. Proc Soc Exp Biol Med 216: 358-369, 1997[Abstract].

16.   Georgieff, MK, Schmidt RL, Mills MM, and Widness JA. Fetal iron and cytochrome c status following intrauterine fetal hypoxemia and exogenous erythropoietin administration. Am J Physiol Regulatory Integrative Comp Physiol 262: R485-R491, 1992[Abstract/Free Full Text].

17.   Hershenson, MB, O'Rourke PP, Christakis DA, Coopes BJ, and Crone RK. Oxygen extraction in lamb skeletal muscle. Pediatr Res 28: 101-105, 1990[ISI][Medline].

18.   Holzman, IR, Tabata B, and Edelstone DI. Blood flow and oxygen delivery to the organs of the neonatal lamb as a function of hematocrit. Pediatr Res 20: 1274-1279, 1986[ISI][Medline].

19.   Izraeli, S, Ben-Sira L, Harell D, Naor N, Ballin A, and Davidson S. Lactic acid as a predictor of erythrocyte transfusion in healthy preterm infants with anemia of prematurity. J Pediatr 122: 629-631, 1993[ISI][Medline].

20.   Jelkmann, W. Erythropoietin: structure, control of production, and function. Physiol Rev 72: 449-489, 1992[Free Full Text].

21.   Jones, JG, Holland BM, Hudson IRB, and Wardrop CAJ Total circulating red cells vs. haematocrit as the primary descriptor of oxygen transport by blood. Br J Haematol 76: 288-294, 1990[ISI][Medline].

22.   Keitt, AS. Reduced nicotinamide adenine dinucleotide-linked analysis of 2,3-diphosphoglyceric acid: spectrophotometric and fluorometric procedures. J Lab Clin Med 77: 470-475, 1971[ISI][Medline].

23.   Kelman, GR, and Nunn JF. Nomograms for correction of blood PO2, PCO2, pH, and base excess for time and temperature. J Appl Physiol 12: 1484-1490, 1966.

24.   Keyes, WG, Donohue PK, Spivak JL, Jones MDJ, and Oski FA. Assessing the need for transfusion of premature infants and role of hematocrit, clinical signs, and erythropoietin level. Pediatrics 84: 412-417, 1989[Abstract/Free Full Text].

25.   Lachance, C, Chessex P, Fouron J, Widness JA, and Bard H. Myocardial, erythropoetic, and metabolic adaptations to anemia of prematurity. J Pediatr 125: 278-282, 1994[ISI][Medline].

26.   Levy, GJ, Strauss RG, Hume H, Scholz L, Albanese MA, Blazina J, Werner A, Sotelo-Avila C, Barraso C, Blanchette V, Warkentin PI, Pepkowitz S, Mauer AM, and Hines D. National survey of neonatal transfusion practices. I. Red blood cell therapy. Pediatrics 91: 523-529, 1993[Abstract/Free Full Text].

27.   Lister, G, Walter TK, Versmold HT, Dallman PR, and Rudolph AM. Oxygen delivery in lambs: cardiovascular and hematological development. Am J Physiol Heart Circ Physiol 237: H668-H675, 1979[Abstract/Free Full Text].

28.   Maier, RF, Obladen M, Messinger D, and Wardrop CAJ Factors related to transfusion in very low birth weight infants treated with erythropoietin. Arch Dis Child 74: F182-F186, 1996.

29.   Matthews, JNS, Altman DG, Campbell MJ, and Royston P. Analysis of serial measurements in medical research. Br Med J 300: 230-235, 1990.

30.   Millard, DD, Gidding SS, Socol ML, MacGregor SN, Dooley SL, Ney JA, and Stockman JA, III. Effects of intravascular, intrauterine transfusions on prenatal and postnatal hemolysis and erythropoiesis in severe fetal isoimmunization. J Pediatr 117: 447-454, 1990[ISI][Medline].

31.   Mock, DM, Strauss RG, and Lankford GL. [14C]cyanate labeling of sheep red cells: covalent binding to hemoglobin continues in vivo for a day. Pediatr Res 41: 424-429, 1997[ISI][Medline].

32.   Moritz, KM, Lim GB, and Wintour EM. Developmental regulation of erythropoietin and erythropoiesis. Am J Physiol Regulatory Integrative Comp Physiol 273: R1829-R1844, 1997.

33.   Obladen, M, Sachsenweger M, and Stahnke M. Blood sampling in very low birth weight infants receiving different levels of intensive care. Eur J Pediatr 147: 399-404, 1988[ISI][Medline].

34.   Peters, C, Georgieff MK, PdeAlarcon Cook RT, Burmeister LF, Lowe LS, and Widness JA. The effect of chronic erythropoietin administration on iron in newborn lambs. Biol Neonate 70: 218-228, 1996[ISI][Medline].

35.   Ramasethu, J, and Luban NLC Red blood cell transfusions in the newborn. Semin Neonatol 4: 5-16, 1999.

36.   Ringer, SA, Richardson DK, Sacher RA, Keszler M, and Churchill WH. Variations in transfusion practice in neonatal intensive care. Pediatrics 101: 194-200, 1998[Abstract/Free Full Text].

37.   Ross, MP, Christensen RD, Rothstein G, Koenig JM, Simmons MA, Noble NA, and Kimura RE. A randomized trial to develop criteria for administering erythrocyte transfusion to anemic preterm infants 1 to 3 months of age. J Perinatol 9: 246-253, 1989[Medline].

38.   Shields, LE, Widness JA, and Brace RA. The hematologic and plasma iron responses to severe hemorrhage in the ovine fetus. Am J Obstet Gynecol 174: 55-61, 1996[ISI][Medline].

39.   Stehling, L, and Zauder HL. How low can we go? Is there a way to know? Transfusion 30: 1-3, 1990[ISI][Medline].

40.   Stockman, JA, III, Graebe JE, Clark DA, McClellan K, Garcia JF, and Kavey REW Anemia of prematurity: determinants of the erythropoietin response. J Pediatr 105: 786-792, 1984[ISI][Medline].

41.   Strauss, RG. Red blood cell transfusion practices in the neonate. Clin Perinatol 22: 641-655, 1995[ISI][Medline].

42.   Teitel, D, Sidi D, Bernstein D, Heymann MA, and Rudolph AM. Chronic hypoxemia in the newborn lamb: cardiovascular, hematopoetic, and growth adaptations. Pediatr Res 19: 1004-1010, 1985[ISI][Medline].

42a.   Third European Consensus Conference in Intensive Care Medicine. Tissue hypoxia: how to detect, how to correct, how to prevent. Société de Réanimation de Langue Française. The American Throacic Society. European Society of Intensive Care Medicine. Am J Respir Crit Care Med 154: 1573-1578, 1996[ISI][Medline].

43.   Thomas, RM, Canning CE, Cotes PM, Linch DC, Rodeck CH, Rossiter CE, and Huehns ER. Erythropoietin and cord blood haemoglobin in the regulation of human fetal erythropoiesis. Br J Obstet Gynaecol 90: 795-800, 1983[ISI][Medline].

44.   VanAmeringen, MR, Fouron JC, Bard H, LeGuennec JC, and Prosamme J. Oxygenation in anemic newborn lambs with high and low oxygen affinity red cells. Pediatr Res 15: 1500-1503, 1981[ISI][Medline].

45.   Voak, D, Cann R, Finney RD, Fraser ID, Mitchell R, Murphy MF, Napier JAF, Phillips P, Rejman AJ, Waters AH, and Wood JK. British Committee for Standards in Haematology Blood Transfusion Task Force: guidelines for administration of blood products: transfusion of infants and neonates. Transfus Med 4: 63-69, 1994[ISI][Medline].

46.   Whyte, RK. Mixed venous oxygen saturation in the newborn. Can we and should we measure it? Scand J Clin Lab Invest Suppl 203: 203-211, 1990[Medline].

47.   Widness, JA, Garcia JF, Clemons GK, Cavalieri RL, Susa JB, Teramo KA, and Schwartz R. Temporal response of immunoreactive erythropoietin to acute hypoxemia in the sheep fetus. Pediatr Res 20: 15-19, 1986[ISI][Medline].

48.   Widness, JA, Seward VJ, Kromer IJ, Burmeister LF, Bell EF, and Strauss RG. Changing patterns of red blood cell transfusion in very low birth weight infants. J Pediatr 129: 680-687, 1996[ISI][Medline].

49.   Widness, JA, Susa JB, Garcia JF, Singer DB, Sehgal P, Oh W, Schwartz R, and Schwartz HC. Increased erythropoiesis and elevated erythropoietin in infants born to diabetic mothers and in hyperinsulinemic rhesus fetuses. J Clin Invest 67: 637-642, 1981.

50.   Wilimas, JA, and Crist WM. Erythropoietin-not yet a standard treatment for anemia of prematurity. Pediatrics 95: 9-10, 1995[Abstract/Free Full Text].

51.   Wilkerson, DK, Rosen AL, Gould SA, Sehgal LR, Sehgal HL, and Moss GS. Oxygen extraction ratio: a valid indicator of myocardial metabolism in anemia. J Surg Res 42: 629-634, 1987[ISI][Medline].

52.   Wilkerson, DK, Rosen AL, Sehgal LR, Gould SA, Sehgal HL, and Moss GS. Limits of cardiac compensation in anemic baboons. Surgery 103: 665-670, 1988[ISI][Medline].

53.   Winter, H. Myelogram of sheep in post-hemorrhagic anemia. Am J Vet Res 28: 1389-1395, 1967[ISI][Medline].

J APPL PHYSIOL 88(4):1397-1406
8570-7587/00 $5.00 Copyright © 2000 the American Physiological Society



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services