Acta PEdiatr Scand 79: 391-396, 1990
Cerebral Blood Flow Reactivity in Spontaneously Breathing, Preterm Infants Shortly after Birth 0. PRYDS, G. E. ANDERSEN and B. FRIIS-HANSEN From the Department of Neonatology, University Hospital of Copenhagen, Copenhagen, Denmark
ABSTRAm. Pryds, O., Andersen, G. E. and Friis-Hansen, B. (Department of Neonatology, University Hospital of Copenhagen, Copenhagen, Denmark). Cerebral blood flow reactivity in healthy, preterm infants shortly after birth. Acta Paediatr Scand 79:391,1990. In 18 spontaneously breathing, preterm infants (mean gestational age 30.3 weeks) cerebral blood flow (CBF) was investigated twice, 2 and 3 hours after birth when spontaneous changes in arterial carbon dioxide tension (P,C02) and mean arterial blood pressure (MABP) were expected. Transcutaneous oxygen tension (TcO,)was kept normal by adjusting the inspiratory oxygen fraction. In 12 infants, plasma adrenaline and noradrenaline were constant throughout the study. Changes in CBF infinity (CBF,) were significantly related to changes in P,C02 (p=O.OOOl) whereas neither changes in MABP nor Tc02 reached a significant association to changes in CBF, (p=0.67 and p= 1.0, respectively). The calculated CBF,-C02 reactivity of 28.9 % per kPa P,C02 (95% confidence interval 16.1-43.0) is comparable to findings in older newborns and healthy adults. Only one of 18 infants developed germinal layer haemorrhage (grade I) in spite of the hypercapnic state which was observed during the first hours of life. Periventricular leucomalacia was not detected. It is suggested that the cerebral blood flow is well regulated within physiological variations of P,C02 and MABP in the healthy, preterm newborn even shortly after birth. Key words: cerebral blood flow,carbon dioxide reactivity, newborn infants.
Impaired cerebral autoregulation is thought to be of major importance for the development of haemorrhagic and ischemic lesions in the preterm newborn. Thus, pressure-passive cerebral blood flow has been reported in distressed newborns implying risk for hyper- or hypoperfusion of the brain (1). Moreover, after the identification of hypercapnia as a risk factor for intracranial haemorrhage care has been taken in many units to keep PaC02within normal ranges by early institution of mechanical ventilation (2). Artificial ventilation, one the other hand, may be accompanied by several complications which are associated with cerebral lesions and bad outcome, i.e. pneumothorax. Although of greatest importance, sparse information is available on the regulation of the cerebral circulation in spontaneously breathing, preterm infants shortly after birth. There are several methodological and ethical problems involved in studies addressing this question in healthy newborns. Therefore, we designed the investigation to include serial measurements of the cerebral blood flow (CBF) at a time when arterial carbon dioxide and blood pressure were expected to have changed spontaneously. In the healthy newborn, the arterial carbon dioxide tension (PaC02)decreases within the first few hours after birth from a hypercapnic towards a normocapnic level, whereas mean arterial blood pressure (MABP) tends to rise. In contrast arterial oxygen tension is kept constant by adjusting the inspiratory oxygen fraction. We report the findings in 18 spontaneously breathing, preterm infants investigated 2 and 3 hours after birth.
392 0. Pryds et al.
Acta Paediatr Scand 79
MATERIAL AND METHODS Eighteen preterm infants with a mean gestational age of 30.3 weeks (range 27 to 33 weeks) and a mean birthweight of 1430 g (SD 425 g) were studied. The delivery was uncomplicated; 16 infants were delivered by caesarian section and 2 infants vaginally. Apgar scores at 5 min were above 7 in all (mean 9.2) and umbilical artery pH averaged 7.33 (SD 0.04). The infants were treated with nasal continuous positive airway pressure (CPAP) and monitored with transcutaneous oxygen and carbon dioxide electrodes (TCM3, Radiometer) instituted shortly after birth. Mean inspiratory fraction of oxygen was 0.30 (range 0.21 to 0.60). After being placed in the ventral position in the incubator care was taken to disturb the infants minimally within the first two hours of life, postponing blood-sampling, X-rays etc. Two hours after birth the cerebral blood flow was measured. At the same time mean arterial blood pressure was recorded by oscillometry (Dynamap) and a capillary blood sample was obtained for blood gas analysis and measurement of blood glucose. Infants experiencing hypoglycaemia, defined as blood glucose less than 1.7 mmol/l were excluded from the study as hypoglycaemia is known to have pronounced effect on the CBF (3). One hour later the same procedure was repeated. Most of the infants (n= 12) served as controls in another study (4) which was why plasma adrenaline and noradrenaline were measured at both CBF studies by use of a single isotope derivate technique (5, 6). After the two measurements of CBF, and on the following days an ultrasonographic examination of the brain was done (6 MHz, Briiel & Kjaer) and any haemorrhage (GLH) graded according to Papile et al. (7). Periventricular leucomalacia (PVL) was defined as periventricular echodensities resolving into cystic lesions. Cerebral blood flow was determined by the intravenous '33Xeclearance technique, validated in preterm infants with respiratory distress, as previously described (8). In short, 0.5-1.0 mCi/kg '33Xe was injected into a peripheral vein and the clearance recorded by scintillators placed over one fronto-parietal region and the thorax, respectively. Expired air was removed from the incubator. Fifteen minutes recording time was used in order to improve the accuracy at low CBF levels (8) and CBF infinity (CBF,) was calculated from the time when the activity in the lung had decreased to 15Yo of the peak activity. This level was used to reduce the effect of scattering from the airways. The blood-brain partition coefficient was set to 0.8 ml/g and adjusted for haemoglobin concentration (9). CBF, was calculated using the Obrist 2-compartment analysis modified to adjust for increased recirculation of the tracer. CBF, represents the weighted mean of gray and white matter flow (10). Since the newborn's head is small and the scintillation geometry allows count from a volume of 100-200 ml, CBF, is considered to represent global cerebral blood flow and is presented as m1/100g x min-'. Before each of the two CBF measurements the background activity and presence of remaining tissue activity were measured (at 112 and 5 min, respectively) and accounted for in the subsequent CBF, calculation. Statistics. The CBF, values were slightly positively skewed and transformed logarithmically to obtain homogeneity of variance (1 1). Thus, a log-linear relation was assumed between CBF, and PaC02,TcOl and MABP, respectively, implying that the calculated CBF, response was percentual (transforming back, the antilogarithm to the regression coefficient is equal to the CBF,-reactivity). Changes in CBF, were analyzed introducing changes in PaC02,Tc02 and MABP as single and multiple variates. A variable with 18 levels, one for each infant was included to test only the intraindividual variability (12). Besides, the Wilcoxon test for paired data and the Spearman rank correlation test were used. The statistical analysis were performed using the program SPSS (Chicago). The study was approved by the Ethics Committee of Greater Copenhagen and parental consent obtained for each infant.
RESULTS Clinical data of CBF,, PaC02and MABP for the two measurements are presented in Table 1. There was no significant covariation between changes in PaC02,TcOz or MABP indicating that these variables were independent of each other in the statistical analysis. The intra-individual variation in CBF, was 13.1 Yo after having accounted for changes in PaC02.In 12 infants plasma adrenaline and noradrenaline averaged 0.244 ng/ml (range 0.01-0.95) and 0.77 ng/ml (range 0.25-1.84) two hours
CBF in healthy, preterm infants shortly after birth 393
Acta Paediatr Scand 19
n
ig 14
7P,C02
6
5
(KPA)8
9
10
Fig. 1. CBF, (log scale) vs. P,C02 in 18 spontaneously breathing, preterm infants investigated twice, 2 and 3 hours after birth.
after birth. The hormone levels remained constant throughout the study period. Changes in CBF, were significantly related to changes in P,C02 (p=O.OOOl); (Fig. 1). Accordingly, CBF, increased by a mean of 28.9 O/o per kPa P,C02 (95 Yo confidence interval 16.1-43.0 Yo) and this estimate was not altered after introduction of changes in MABP and/or Tc02 in the regression analysis. In contrast, neither changes in MABP (p=0.67) nor Tc02 (p= 1.0) reached a significant association to changes in CBF, (Fig. 2). The calculated CBF,-MABP and CBF,-Tc02 reactiviTable 1. Clinical data of 18 spontaneous breathing infants investigated 2 and 3 hours after birth 2 hours
3 hours
ID
PaC02 (kPa)
MABP (mmHg)
CBF, (m1/100 g/min)
PaC02 (kPa)
MABP (mmHg)
CBF, (m1/100 g/min)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
9.4 7.1 7.3 5.3 8.0 8.6 8.0 6.1 7.2 7.6 6.4 5.1 7.7 7.3 6.1 7.8 5.1 4.0
35 33 51 34 39 45 32 42 39 33 36 38 26 52 29 40 36 43
28.5 17.6 31.4 20.7 15.0 16.5 18.7 13.6 16.4 16.0 11.5 10.8 11.8 16.3 9.5 12.0 10.1 10.9
9.8 6.8 6.3 5.4 6.7 8.5 8.1 6.0 6.4 8.0 6.7 5.0 5.3 6.4 6.8 7.4 5.1 4.0
31 35 41 41 40 45 31 46 37 32 37 42 25 48 30 40 39 44
26.2 13.0 18.0 17.3 8.3 13.9 15.7 12.4 13.8 14.1 12.4 9.8 8.1 11.9 10.5 10.2 9.6 9.2
.
394 0. Pryds et al.
Acta Paediatr Scand 79
108T
0
0
0
0
125
30
35
48
45
50
55
MABP (MHHG)
Fig. 2. CBF, (log scale) vs. MABP in 18 spontaneously breathing, preterm infants investigated twice, 2 and 3 hours after birth.
ties were 0.0% per mmHg (95 O/o confidence interval -4.2-4.3) and -6.2% per kPa Tc02 (95 Yo confidence interval - 54.9-95.0), respectively. There was no significant change in CBF, between the two measurements after having adjusted for the effect of P,CO2. After the study only one infant was intubated and treated in ventilator because of persisting hypercapnia. Another infant developed subependymal haemorrhage (grade I) whereas the remaining infants all had consistently normal brain ultrasonography. PVL was not detected. DISCUSSION In the normal adult brain cerebral vessels are sensitive to change in carbon dioxide, oxygen and arterial blood pressure as well as substrate deficiency (13). Thus in normal individuals, CBF increases during hypercapnia, hypoxia and hypoglycaemia but due to the pressure-flow autoregulation CBF remains constant at different blood pressures within a certain range. There is reason to believe that the fetus and preterm infant have an identical capability unless insults resulting in vasoparalysis have occurred (14). No prior study has tested the cerebral autoregulation in spontaneously breathing, healthy preterm infants within the first hours after birth. In an identical group of infants Winberg et al., using Doppler ultrasonography, demonstrated a 29 O/o reduction in mean blood flow velocity (MFV) between 1 and 5 h of life. Assuming a constant vessel diameter, this decrease in MFV reflects a parallel decrease in CBF. Unfortunately, these investigators did not account for changes in blood gases (15). The present data indicate that the spontaneously breathing newborn is born with a CBF-C02 reactivity which is comparable to that reported in older newborn infants and adults (1 6-20). Thus, the hypothesis of a general cerebral vasoconstriction in the immediate postpartal period causing an attenuation of the C 0 2 sensitivity shortly after birth (21) seems not to be valid for spontaneously breathing, healthy newborns. In our infants the pressure-flow autoregulation was intact within the small physio-
Acta Paediatr Scand 79
CBF in healthy, preterm infants shortly afer birth 395
logical variations of MABP taking place during the observation period indicating that MABP in the present range of 26 to 52 mmHg may be within the autoregulatory plateau. That CBF was unaffected by the small changes in Tc02 tensions may be due to the even smaller changes in haemoglobin saturation and blood oxygen concentration (22). Although the regulation of CBF is multifactorial some confounding variables may be excluded. Firstly, in 12 infants plasma adrenaline and noradrenaline remained at constant levels during the two CBF measurements. Secondly, substrate deficiency in terms of hypoglycaemia was not present and thirdly, spontaneous closure of an open ductus arteriosus might only have underestimated the CBF-C02 reactivity due to a subsequent improvement of the central circulation. Only one of 18 infants developed minor intracerebral haemorrhage. The remaining 17 had normal ultrasonography without signs of intracerebral haemorrhage or PVL. Therefore, it seems as if the cerebral blood flow is well regulated within physiological variations of P,C02 and MABP in the healthy, preterm newborn. ACKNOWLEDGEMENTS The Dagmar Marschal Foundation and the Gerda and Age Haensch Foundation gave financial support. The Statistical Research Unit, University of Copenhagen provided the statistical model.
REFERENCES 1 . Lou HC, Lassen NA, Friis-Hansen B. Low cerebral blood flow in hypotensive perinatal distress. Acta Neurol Scand 1977; 56: 343-50. 2. Cooke RWI. Factors associated with periventricular haemorrhage in very low birthweight infants. Arch Dis Child 1981; 56: 425-31. 3. Pryds 0, Greisen G, Friis-Hansen B. Compensatory increase of CBF supports the cerebral metabolism in preterm infants during hypoglycaemia. Acta Paediatr Scand 1988; 77: 632-37. 4. Pryds 0, Christensen NJ, Friis-Hansen B. Increased CBF and plasma epinephrine in hypoglycemic, preterm infants. Pediatrics. 1990; 85. 5. Christensen NJ, Vestergaard P, Serrensen T, Rafaelsen OJ. Cerebrospinal fluid adrenaline and noradrenaline in depressed patients. Acta Psychiatr Scand 1980; 61: 178-82. 6. Christensen NJ. Plasma noradrenaline and adrenaline in patients with thyreotoxicosis and myxoedema. Clin Sci 1973; 45: 163-71. 7. Papile L, Burstein J, Burstein R et al. Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1500 g. J Pediatrics 1978; 92: 529-34. 8. Greisen G, Pryds 0. Intravenous 133Xeclearance in preterm neonates with respiratory distress. Internal validation of CBF, as a measure of global cerebral blood flow. Scand J Clin Lab Invest 1988; 48: 673-78. 9. Greisen G. Cerebral blood flow in preterm infants during the first week of life. Acta Paediatr Scand 1986; 75: 43-51. 10. Obrist W, Thompson HK, Wang HS et al. Regional cerebral blood flow estimated by 1'33Xenoninhalation. Stroke 1975; 6: 245-56. 1 1 . Godfrey K. Simple linear regression in medical research. In: Bailar JC, Mosteller F, eds. Medical uses of statistics. Massachusetts, USA: NEJM Books, 1986: 170-201. 12. Smith EO. Analysis of repeated measures designs (Editorial). J Pediatrics 1987; I 1 1 : 723-25. 13. Heistad DD, Kontos HA. Cerebral circulation: Shepherd JT, Abboud FM, eds. Handbook of physiology: The cardiovascular system, section 2. Bethesda: American Physiological Society, 1983; 3: 137-82. 14. Purves MJ, James IM. Observation on the control of blood flow in sheep fetus and newborn lamb. Circ Res 1969; 26: 651-67.
396 0. Pryds et al.
Acta Paediatr Scand 79
15. Sonesson SE. Intracranial Doppler flow velocimetry in newborn infants. Dissertation. Stockholm: University of Stockholm, 1988: 43-44. 16. Kety SS, Schmidt CF. The effects of altered tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 1948; 27: 487-92. 17. Lassen NA. Control of cerebral circulation in health and disease. Circ Res 1974; 34: 749-60. 18. Rahilly PM. Effects of 2% carbon dioxide, 0.5%carbon dioxide, and 100%oxygen on cranial blood flow of the human neonate. Pediatrics 1980; 66: 685-89. 19. Costeloe K, Smyth DPL, Rolfe P, Tizard JPM. A comparison between electrical impedance and strain gauge plethysmography for the study of cerebral blood flow in the newborn. Pediatr Res 1984; 18: 290-95. 20. Greisen G, Trojaborg W. Cerebral blood flow, P,C02 changes, and visual evoked potentials in mechanically ventilated, preterm infants. Acta Paediatr Scand 1987; 76: 394-400. 21. Levene M, Shortland D, Gibson N, Evans DH. Carbon dioxide reactivity of the cerebral circulation in extremely premature infants: Effects of postnatal age and indomethacin. Pediatr Res 1988; 24: 175-79. 22. Nijima S, Shortland DB, Levene MI, Evans DH. Transient hyperoxia and cerebral blood flow velocity in infants born prematurely and at full term. Arch Dis Child 1988; 63: 1 126-30.
Submitted Jan. 17, 1989. Accepted July 13, 1989 (0.P.) Department of Neonatology, 5024, Rigshospitalet DK-2100 Copenhagen 0 Denmark
Cerebral Blood Flow Reactivity in Spontaneously Breathing, Preterm Infants Shortly after Birth 0. PRYDS, G. E. ANDERSEN and B. FRIIS-HANSEN From the Department of Neonatology, University Hospital of Copenhagen, Copenhagen, Denmark
ABSTRAm. Pryds, O., Andersen, G. E. and Friis-Hansen, B. (Department of Neonatology, University Hospital of Copenhagen, Copenhagen, Denmark). Cerebral blood flow reactivity in healthy, preterm infants shortly after birth. Acta Paediatr Scand 79:391,1990. In 18 spontaneously breathing, preterm infants (mean gestational age 30.3 weeks) cerebral blood flow (CBF) was investigated twice, 2 and 3 hours after birth when spontaneous changes in arterial carbon dioxide tension (P,C02) and mean arterial blood pressure (MABP) were expected. Transcutaneous oxygen tension (TcO,)was kept normal by adjusting the inspiratory oxygen fraction. In 12 infants, plasma adrenaline and noradrenaline were constant throughout the study. Changes in CBF infinity (CBF,) were significantly related to changes in P,C02 (p=O.OOOl) whereas neither changes in MABP nor Tc02 reached a significant association to changes in CBF, (p=0.67 and p= 1.0, respectively). The calculated CBF,-C02 reactivity of 28.9 % per kPa P,C02 (95% confidence interval 16.1-43.0) is comparable to findings in older newborns and healthy adults. Only one of 18 infants developed germinal layer haemorrhage (grade I) in spite of the hypercapnic state which was observed during the first hours of life. Periventricular leucomalacia was not detected. It is suggested that the cerebral blood flow is well regulated within physiological variations of P,C02 and MABP in the healthy, preterm newborn even shortly after birth. Key words: cerebral blood flow,carbon dioxide reactivity, newborn infants.
Impaired cerebral autoregulation is thought to be of major importance for the development of haemorrhagic and ischemic lesions in the preterm newborn. Thus, pressure-passive cerebral blood flow has been reported in distressed newborns implying risk for hyper- or hypoperfusion of the brain (1). Moreover, after the identification of hypercapnia as a risk factor for intracranial haemorrhage care has been taken in many units to keep PaC02within normal ranges by early institution of mechanical ventilation (2). Artificial ventilation, one the other hand, may be accompanied by several complications which are associated with cerebral lesions and bad outcome, i.e. pneumothorax. Although of greatest importance, sparse information is available on the regulation of the cerebral circulation in spontaneously breathing, preterm infants shortly after birth. There are several methodological and ethical problems involved in studies addressing this question in healthy newborns. Therefore, we designed the investigation to include serial measurements of the cerebral blood flow (CBF) at a time when arterial carbon dioxide and blood pressure were expected to have changed spontaneously. In the healthy newborn, the arterial carbon dioxide tension (PaC02)decreases within the first few hours after birth from a hypercapnic towards a normocapnic level, whereas mean arterial blood pressure (MABP) tends to rise. In contrast arterial oxygen tension is kept constant by adjusting the inspiratory oxygen fraction. We report the findings in 18 spontaneously breathing, preterm infants investigated 2 and 3 hours after birth.
392 0. Pryds et al.
Acta Paediatr Scand 79
MATERIAL AND METHODS Eighteen preterm infants with a mean gestational age of 30.3 weeks (range 27 to 33 weeks) and a mean birthweight of 1430 g (SD 425 g) were studied. The delivery was uncomplicated; 16 infants were delivered by caesarian section and 2 infants vaginally. Apgar scores at 5 min were above 7 in all (mean 9.2) and umbilical artery pH averaged 7.33 (SD 0.04). The infants were treated with nasal continuous positive airway pressure (CPAP) and monitored with transcutaneous oxygen and carbon dioxide electrodes (TCM3, Radiometer) instituted shortly after birth. Mean inspiratory fraction of oxygen was 0.30 (range 0.21 to 0.60). After being placed in the ventral position in the incubator care was taken to disturb the infants minimally within the first two hours of life, postponing blood-sampling, X-rays etc. Two hours after birth the cerebral blood flow was measured. At the same time mean arterial blood pressure was recorded by oscillometry (Dynamap) and a capillary blood sample was obtained for blood gas analysis and measurement of blood glucose. Infants experiencing hypoglycaemia, defined as blood glucose less than 1.7 mmol/l were excluded from the study as hypoglycaemia is known to have pronounced effect on the CBF (3). One hour later the same procedure was repeated. Most of the infants (n= 12) served as controls in another study (4) which was why plasma adrenaline and noradrenaline were measured at both CBF studies by use of a single isotope derivate technique (5, 6). After the two measurements of CBF, and on the following days an ultrasonographic examination of the brain was done (6 MHz, Briiel & Kjaer) and any haemorrhage (GLH) graded according to Papile et al. (7). Periventricular leucomalacia (PVL) was defined as periventricular echodensities resolving into cystic lesions. Cerebral blood flow was determined by the intravenous '33Xeclearance technique, validated in preterm infants with respiratory distress, as previously described (8). In short, 0.5-1.0 mCi/kg '33Xe was injected into a peripheral vein and the clearance recorded by scintillators placed over one fronto-parietal region and the thorax, respectively. Expired air was removed from the incubator. Fifteen minutes recording time was used in order to improve the accuracy at low CBF levels (8) and CBF infinity (CBF,) was calculated from the time when the activity in the lung had decreased to 15Yo of the peak activity. This level was used to reduce the effect of scattering from the airways. The blood-brain partition coefficient was set to 0.8 ml/g and adjusted for haemoglobin concentration (9). CBF, was calculated using the Obrist 2-compartment analysis modified to adjust for increased recirculation of the tracer. CBF, represents the weighted mean of gray and white matter flow (10). Since the newborn's head is small and the scintillation geometry allows count from a volume of 100-200 ml, CBF, is considered to represent global cerebral blood flow and is presented as m1/100g x min-'. Before each of the two CBF measurements the background activity and presence of remaining tissue activity were measured (at 112 and 5 min, respectively) and accounted for in the subsequent CBF, calculation. Statistics. The CBF, values were slightly positively skewed and transformed logarithmically to obtain homogeneity of variance (1 1). Thus, a log-linear relation was assumed between CBF, and PaC02,TcOl and MABP, respectively, implying that the calculated CBF, response was percentual (transforming back, the antilogarithm to the regression coefficient is equal to the CBF,-reactivity). Changes in CBF, were analyzed introducing changes in PaC02,Tc02 and MABP as single and multiple variates. A variable with 18 levels, one for each infant was included to test only the intraindividual variability (12). Besides, the Wilcoxon test for paired data and the Spearman rank correlation test were used. The statistical analysis were performed using the program SPSS (Chicago). The study was approved by the Ethics Committee of Greater Copenhagen and parental consent obtained for each infant.
RESULTS Clinical data of CBF,, PaC02and MABP for the two measurements are presented in Table 1. There was no significant covariation between changes in PaC02,TcOz or MABP indicating that these variables were independent of each other in the statistical analysis. The intra-individual variation in CBF, was 13.1 Yo after having accounted for changes in PaC02.In 12 infants plasma adrenaline and noradrenaline averaged 0.244 ng/ml (range 0.01-0.95) and 0.77 ng/ml (range 0.25-1.84) two hours
CBF in healthy, preterm infants shortly after birth 393
Acta Paediatr Scand 19
n
ig 14
7P,C02
6
5
(KPA)8
9
10
Fig. 1. CBF, (log scale) vs. P,C02 in 18 spontaneously breathing, preterm infants investigated twice, 2 and 3 hours after birth.
after birth. The hormone levels remained constant throughout the study period. Changes in CBF, were significantly related to changes in P,C02 (p=O.OOOl); (Fig. 1). Accordingly, CBF, increased by a mean of 28.9 O/o per kPa P,C02 (95 Yo confidence interval 16.1-43.0 Yo) and this estimate was not altered after introduction of changes in MABP and/or Tc02 in the regression analysis. In contrast, neither changes in MABP (p=0.67) nor Tc02 (p= 1.0) reached a significant association to changes in CBF, (Fig. 2). The calculated CBF,-MABP and CBF,-Tc02 reactiviTable 1. Clinical data of 18 spontaneous breathing infants investigated 2 and 3 hours after birth 2 hours
3 hours
ID
PaC02 (kPa)
MABP (mmHg)
CBF, (m1/100 g/min)
PaC02 (kPa)
MABP (mmHg)
CBF, (m1/100 g/min)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
9.4 7.1 7.3 5.3 8.0 8.6 8.0 6.1 7.2 7.6 6.4 5.1 7.7 7.3 6.1 7.8 5.1 4.0
35 33 51 34 39 45 32 42 39 33 36 38 26 52 29 40 36 43
28.5 17.6 31.4 20.7 15.0 16.5 18.7 13.6 16.4 16.0 11.5 10.8 11.8 16.3 9.5 12.0 10.1 10.9
9.8 6.8 6.3 5.4 6.7 8.5 8.1 6.0 6.4 8.0 6.7 5.0 5.3 6.4 6.8 7.4 5.1 4.0
31 35 41 41 40 45 31 46 37 32 37 42 25 48 30 40 39 44
26.2 13.0 18.0 17.3 8.3 13.9 15.7 12.4 13.8 14.1 12.4 9.8 8.1 11.9 10.5 10.2 9.6 9.2
.
394 0. Pryds et al.
Acta Paediatr Scand 79
108T
0
0
0
0
125
30
35
48
45
50
55
MABP (MHHG)
Fig. 2. CBF, (log scale) vs. MABP in 18 spontaneously breathing, preterm infants investigated twice, 2 and 3 hours after birth.
ties were 0.0% per mmHg (95 O/o confidence interval -4.2-4.3) and -6.2% per kPa Tc02 (95 Yo confidence interval - 54.9-95.0), respectively. There was no significant change in CBF, between the two measurements after having adjusted for the effect of P,CO2. After the study only one infant was intubated and treated in ventilator because of persisting hypercapnia. Another infant developed subependymal haemorrhage (grade I) whereas the remaining infants all had consistently normal brain ultrasonography. PVL was not detected. DISCUSSION In the normal adult brain cerebral vessels are sensitive to change in carbon dioxide, oxygen and arterial blood pressure as well as substrate deficiency (13). Thus in normal individuals, CBF increases during hypercapnia, hypoxia and hypoglycaemia but due to the pressure-flow autoregulation CBF remains constant at different blood pressures within a certain range. There is reason to believe that the fetus and preterm infant have an identical capability unless insults resulting in vasoparalysis have occurred (14). No prior study has tested the cerebral autoregulation in spontaneously breathing, healthy preterm infants within the first hours after birth. In an identical group of infants Winberg et al., using Doppler ultrasonography, demonstrated a 29 O/o reduction in mean blood flow velocity (MFV) between 1 and 5 h of life. Assuming a constant vessel diameter, this decrease in MFV reflects a parallel decrease in CBF. Unfortunately, these investigators did not account for changes in blood gases (15). The present data indicate that the spontaneously breathing newborn is born with a CBF-C02 reactivity which is comparable to that reported in older newborn infants and adults (1 6-20). Thus, the hypothesis of a general cerebral vasoconstriction in the immediate postpartal period causing an attenuation of the C 0 2 sensitivity shortly after birth (21) seems not to be valid for spontaneously breathing, healthy newborns. In our infants the pressure-flow autoregulation was intact within the small physio-
Acta Paediatr Scand 79
CBF in healthy, preterm infants shortly afer birth 395
logical variations of MABP taking place during the observation period indicating that MABP in the present range of 26 to 52 mmHg may be within the autoregulatory plateau. That CBF was unaffected by the small changes in Tc02 tensions may be due to the even smaller changes in haemoglobin saturation and blood oxygen concentration (22). Although the regulation of CBF is multifactorial some confounding variables may be excluded. Firstly, in 12 infants plasma adrenaline and noradrenaline remained at constant levels during the two CBF measurements. Secondly, substrate deficiency in terms of hypoglycaemia was not present and thirdly, spontaneous closure of an open ductus arteriosus might only have underestimated the CBF-C02 reactivity due to a subsequent improvement of the central circulation. Only one of 18 infants developed minor intracerebral haemorrhage. The remaining 17 had normal ultrasonography without signs of intracerebral haemorrhage or PVL. Therefore, it seems as if the cerebral blood flow is well regulated within physiological variations of P,C02 and MABP in the healthy, preterm newborn. ACKNOWLEDGEMENTS The Dagmar Marschal Foundation and the Gerda and Age Haensch Foundation gave financial support. The Statistical Research Unit, University of Copenhagen provided the statistical model.
REFERENCES 1 . Lou HC, Lassen NA, Friis-Hansen B. Low cerebral blood flow in hypotensive perinatal distress. Acta Neurol Scand 1977; 56: 343-50. 2. Cooke RWI. Factors associated with periventricular haemorrhage in very low birthweight infants. Arch Dis Child 1981; 56: 425-31. 3. Pryds 0, Greisen G, Friis-Hansen B. Compensatory increase of CBF supports the cerebral metabolism in preterm infants during hypoglycaemia. Acta Paediatr Scand 1988; 77: 632-37. 4. Pryds 0, Christensen NJ, Friis-Hansen B. Increased CBF and plasma epinephrine in hypoglycemic, preterm infants. Pediatrics. 1990; 85. 5. Christensen NJ, Vestergaard P, Serrensen T, Rafaelsen OJ. Cerebrospinal fluid adrenaline and noradrenaline in depressed patients. Acta Psychiatr Scand 1980; 61: 178-82. 6. Christensen NJ. Plasma noradrenaline and adrenaline in patients with thyreotoxicosis and myxoedema. Clin Sci 1973; 45: 163-71. 7. Papile L, Burstein J, Burstein R et al. Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1500 g. J Pediatrics 1978; 92: 529-34. 8. Greisen G, Pryds 0. Intravenous 133Xeclearance in preterm neonates with respiratory distress. Internal validation of CBF, as a measure of global cerebral blood flow. Scand J Clin Lab Invest 1988; 48: 673-78. 9. Greisen G. Cerebral blood flow in preterm infants during the first week of life. Acta Paediatr Scand 1986; 75: 43-51. 10. Obrist W, Thompson HK, Wang HS et al. Regional cerebral blood flow estimated by 1'33Xenoninhalation. Stroke 1975; 6: 245-56. 1 1 . Godfrey K. Simple linear regression in medical research. In: Bailar JC, Mosteller F, eds. Medical uses of statistics. Massachusetts, USA: NEJM Books, 1986: 170-201. 12. Smith EO. Analysis of repeated measures designs (Editorial). J Pediatrics 1987; I 1 1 : 723-25. 13. Heistad DD, Kontos HA. Cerebral circulation: Shepherd JT, Abboud FM, eds. Handbook of physiology: The cardiovascular system, section 2. Bethesda: American Physiological Society, 1983; 3: 137-82. 14. Purves MJ, James IM. Observation on the control of blood flow in sheep fetus and newborn lamb. Circ Res 1969; 26: 651-67.
396 0. Pryds et al.
Acta Paediatr Scand 79
15. Sonesson SE. Intracranial Doppler flow velocimetry in newborn infants. Dissertation. Stockholm: University of Stockholm, 1988: 43-44. 16. Kety SS, Schmidt CF. The effects of altered tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 1948; 27: 487-92. 17. Lassen NA. Control of cerebral circulation in health and disease. Circ Res 1974; 34: 749-60. 18. Rahilly PM. Effects of 2% carbon dioxide, 0.5%carbon dioxide, and 100%oxygen on cranial blood flow of the human neonate. Pediatrics 1980; 66: 685-89. 19. Costeloe K, Smyth DPL, Rolfe P, Tizard JPM. A comparison between electrical impedance and strain gauge plethysmography for the study of cerebral blood flow in the newborn. Pediatr Res 1984; 18: 290-95. 20. Greisen G, Trojaborg W. Cerebral blood flow, P,C02 changes, and visual evoked potentials in mechanically ventilated, preterm infants. Acta Paediatr Scand 1987; 76: 394-400. 21. Levene M, Shortland D, Gibson N, Evans DH. Carbon dioxide reactivity of the cerebral circulation in extremely premature infants: Effects of postnatal age and indomethacin. Pediatr Res 1988; 24: 175-79. 22. Nijima S, Shortland DB, Levene MI, Evans DH. Transient hyperoxia and cerebral blood flow velocity in infants born prematurely and at full term. Arch Dis Child 1988; 63: 1 126-30.
Submitted Jan. 17, 1989. Accepted July 13, 1989 (0.P.) Department of Neonatology, 5024, Rigshospitalet DK-2100 Copenhagen 0 Denmark
