Arterial blood pressure and blood flow velocity in major cerebral and visceral arteries. I. Interindividual differences

Early Human Development 34 (1993) 227-232

Arterial blood pressure and blood flow velocity in major cerebral and visceral arteries. I. Interindividual differences S.T. Kempley*, H.R. Gamsu Children Nationwide Regional

Neonatal Centre, King’s College Hospital, London, UK

(Received 19 March 1993; revision received 20 April 1993; accepted 10 June 1993)

Abstract

In order to determine the relationship between blood pressure and arterial blood flow velocity in various regional circulations, Doppler ultrasound measurements of blood flow velocity were recorded on the first day of postnatal life from the anterior cerebral (ACA), superior mesenteric, coeliac axis and left renal artery. In 34 ventilated very low birth weight (VLBW) infants, results were correlated with arterial blood pressure and blood gases in a multiple regression model. ACA velocity was correlated with blood pressure (r = 0.70) and P,co* (r = OX%), but there was no relationship between blood pressure and velocity in the other arteries. Repeated measurements were performed at one week of age in 15 infants. Blood flow velocity in the cerebral and renal arteries was related to blood pressure; velocity in the cerebral arteries was inversely correlated with Pao2 and velocity in the coeliac and mesenteric arteries was positively correlated with PaoZ. In VLBW infants on the first day of life, blood flow velocity is related to blood pressure in the cerebral circulation only. Key words: Cerebral

blood flow; Splanchnic

circulation;

Renal artery; Doppler

ultrasound

1. Introduction

Doppler ultrasound studies of newborn infants have shown that the velocity of blood flow in major cerebral arteries varies with birthweight [l], postnatal age [2], mechanical ventilator settings [3,4], spontaneous respiratory effort [5], vasoactive drugs [6], arterial carbon dioxide tension [7], and arterial blood pressure[8]. Some * Corresponding author, Department of Paediatrics, Royal London Hospital, London El lBB, UK. 0378-3782/93/306.00 0 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. SSDI 0378-3782(93)01448-K

228

S. T. Kempley, H. R. Gamsu / Early Hum. Dev. 34 (1993) 227-232

of these factors affect blood flow velocity directly, whereas others act indirectly by altering, for instance, blood pressure, intracranial pressure [3] or blood gases [4]. Doppler ultrasound is now being used to study other parts of the neonatal cardiovascular system and it is possible that blood flow velocity in other arteries may be influenced by some of the factors which also affect the cerebral circulation. We have used Doppler ultrasound to perform concurrent measurements of blood flow velocity in the cerebral, mesenteric, coeliac and renal arteries, in mechanically ventilated very low birth weight (VLBW) infants on the first day of postnatal life. We have attempted to determine the extent to which differences in blood flow velocity between individuals can be related to differences in various characteristics, including blood pressure, for each of these arteries. A multiple regression analysis is used and the analysis is repeated on data obtained from a limited number of infants at 1 week. 2. Methods A single set of measurements was obtained from each of 34 ventilated VLBW infants, less than 24 h old. We restricted our study population to those infants whose weight was appropriate for gestational age, as we have previously demonstrated a specific reduction in mesenteric and coeliac velocity in small-for-gestational-age infants [9]. The main reason for mechanical ventilation was hyaline membrane disease in 15 infants, perinatal bacterial infection in six, transient tachypnoea in one and extreme prematurity in 12. At the time that the measurements were performed, two infants had already received intravenous aminophylline and three were receiving inotropic support in the form of a dopamine infusion. The survival rate was 68%. Birthweight, gestation and other major features of the group are shown in Table 1. All the infants studied had continuous intra-arterial blood pressure recording from umbilical, radial or posterior tibia1 arterial catheters. Blood pressure transducers were used (Medex Inc, UK) and the blood pressure trace and digital recordings were continuously displayed on neonatal monitors (HP78801, HewlettPackard, or Horizon 2000, Mennen). Doppler ultrasound measurements were performed using a duplex ultrasound imaging and Doppler system (Hewlett-Packard Sonos 100 with 7.5 MHz imaging and 5.0 MHz Doppler transducer). Each artery was visualised in real-time and the angle of insonation was measured. A range gate was used to sample Doppler blood flow velocity waveforms from a known portion of the artery, and the resulting waveforms were analyzed manually on-line to obtain the time-averaged mean velocity of the peak velocity envelope over at least four consecutive cardiac cycles. All measurements were corrected for the angle of insonation, which was always less than 45”. For each infant consecutive measurements were made of velocity in the anterior cerebral artery (ACA), coeliac axis, superior mesenteric artery (SMA) and left renal artery. A 3-mm range gate was used to sample signals only from the ascending portion of the ACA, and from the first centimetre of the other arteries. (The reproducibility of measurements from each artery was previously established by performing paired measurements on all four vessels in each of twelve premature newborn subjects who did not form part of this study. The same operator (STK) performed all these measurements, and all of the

ST. Kempley, H.R. Gamsu/ Early Hum. Dev. 34 (1993) 227-232

229

measurements in both parts of the study. The standard deviation of the percentage difference between measurements was 12% for the ACA, 10% for the coeliac, 14% for the SMA and 17% for the left renal artery.) A linear multiple regression analysis was performed separately on the measurements from each of the arteries. The dependent variable was the velocity in the artery, and the independent variables were the mean arterial blood pressure, arterial blood gases (pH, Paco2,P,o& mean airway pressure, peak end-expiratory pressure (PEEP), birthweight, gestation and the age in hours. Independent variables showing no effect on any arterial velocity measurement were then discarded in favour of a condensed model multiple regression analysis, which finally included the effects of blood pressure, P,co* and P,o, on velocity. Some of the infants with low blood pressures received treatment with intravenous colloid infusions. For these infants, only the measurements made before the infusion were included in the regression analysis. The effects of plasma infusion are reported separately [lo]. It was possible to repeat the analysis using the condensed model for data obtained at one week of age from 15 infants from the original group, who were being followed up as part of an investigation into umbilical catheterisation. This group had a mean gestation of 25 weeks (range 23-29) and a mean birthweight of 898 g (range 506-1284). 13 were mechanically ventilated and none were receiving enteral feeds. 3. Results 3.1. Differences between individuals on day 1 Simple correlations which were statistically significant at a 1% level were found between ACA velocity and birthweight (r = 0.44), gestation (r = 0.46) mean blood pressure (r = 0.55) and P&o2 (r= 0.46). The only other significant simple correlation was between coeliac axis velocity and mean airway pressure (r = -0.38, P < 0.05).

In the full model multiple regression analysis the only statistically significant partial correlations were between blood pressure and ACA velocity (r = 0.51, P < O.Ol), and P&o2 and ACA velocity (r = 0.45, P < 0.05). The differences in ACA velocity associated with birthweight and gestation were accounted for by differences in blood pressure and P,co~.In contrast to these findings in the ACA, there was no association between blood pressure and velocity in the mesenteric, coeliac or renal arteries. As a large number of independent variables were included in the full model, producing rather low individual correlation coefficients, the analysis was repeated on a condensed model in which blood pressure, P&O2 and Pao2 were the only independent variables (P,o,was included as the largest non-significant correlation). In this analysis (Table 1) the correlations between blood pressure, P.&o2 and ACA velocity were much stronger, but again there was no relationship between blood pressure and velocity in any of the other three arteries. This condensed model accounted for 59% of the variation in ACA velocity, but only 8% of the variation in coeliac velocity, 3% of the variation in SMA velocity and 9% of the variation in renal artery velocity.

S. T. Kempley, H. R. Gamsu / Early Hum. Dev. 34 (1993) 227-232

230

Table 1 Mean values and ranges for the independant variables, and partial correlation regression analysis carried out on data obtained on the first day Variable (a) Full model Birth weight (kg) Gestation

(weeks)

Age (h) BP (mmHg) MAP (cmH20) PEEP (cmH*O) PH P&O,

(kPa)

Pa02 (kPa)

coefficients

for the multiple

Renal

Mean

Range

ACA

Coeliac

SMA

0.96 26.6 10.1 33 1.1 2.1 1.38 4.8 9.3

0.51-1.49 23-31 l-23 IO-52 2.1-18 o-3 7.1-7.55 2.8-8.5 5.3-13.5

0.07 -0.03 0.11 0.51** -0.05 -0.22 0.02 0.45* -0.33

0.01 0.09 -0.19 0.20 -0.32 0.01 0.04 0.01 -0.18

-0.11 0.19 -0.001 -0.18 -0.19 -0.14 0.14 0.11 0.08

0.05 -0.18 -0.03 -0.03 -0.03 0.05 -0.04 0.19

0.65

0.23

0.19

0.18

0.08 -0.25 0.06

-0.05 -0.16 0.13

-0.17 0.26

0.08

0.03

0.09

Overall r-squared

0.04

(It) Condensed model BP WW)

P&o1 (kPa) Pa% (kPa)

33 4.8 9.3

lo-52 2.8-8.5 5.3-13.5

Overall r-squared

0.70t 0.64t -0.31 0.59

0.04

Dependent variables are velocity of blood flow in the anterior cerebral artery (ACA), coeliac axis, superior mesenteric artery (SMA) and renal artery. Independent variables are birthweight, gestation, age in hours, mean arterial blood pressure (BP), mean airway pressure (MAP), peak end expiratory pressure (PEEP), and arterial pH, carbon dioxide tension (P,co,) and oxygen tension (Pao2). The range of each independent variable is shown. Statistically significant values are indicated (*P < 0.05; **P < 0.01; tP < 0.001)

The regression coefficients for the ACA predicted that a I-mmHg increase in blood pressure would be associated with a 0.39-cm/s increase in ACA velocity, and a l-KPa increase in P&o2 would be associated with a 2.3~cm/s increase in velocity. For an infant with average ACA velocity, a 10% change in velocity would be produced by a 9% change in mean blood pressure or a 10% change in P&02.

Table 2 Partial correlation coefficients for multiple regression analysis carried out on data at one week of age. Dependent variables are velocity of blood flow in the anterior cerebral artery (ACA), coeliac axis, superior mesenteric artery (SMA) and renal artery Variable

Mean

Range

ACA

Coeliac

SMA

Renal

BP (mmHg) P&o2 (kPa) Pa% (kPa)

44 5.1 10.5

19-60 3.9-8.0 6.4-25.0

0.55* 0.04 -0.63*

0.23 -0.04 0.56*

0.10 -0.19 0.66**

0.86t -0.44 -0.39

0.66

0.32

0.49

0.74

Overall

R-squared

Independent variables are mean arterial blood pressure (BP), and arterial carbon dioxide (P&O,) and oxygen tension (Pao2). The range of each independent variable is shown. Statistically cant values are indicated (*P c 0.05; **P < 0.02; tP < 0.001).

tension signifi-

S. T. Kempley. h.R. Gamsu/ Early Hum. Dev. 34 (1993) 227-232

231

3.2. Differences between individuals at 1 week

At one week there was a statistically significant correlation between blood pressure and cerebral artery velocity, and between blood pressure and renal artery velocity (Table 2). There was a significant negative correlation between arterial oxygen tension (Pao2) and cerebral artery velocity, and a positive correlation between P,o, and velocity in both the coeliac and mesenteric arteries. 4. Discussion

We have shown that in ventilated VLBW infants differences between individuals in cerebral artery blood flow velocity can be related to differences in blood pressure and P,co~. In contrast, differences in coeliac, mesenteric and renal artery velocity are not related to differences in blood pressure. We took care to exclude infants with intrauterine growth retardation, in whom we had previously demonstrated a specific reduction in visceral artery blood flow velocity [9]. Although the infants studied were to some extent diagnostically heterogenous, they were representative of appropriately grown infants of this birthweight who require intensive care. Only 3/34 infants were receiving inotropes, making it unlikely that the differences observed between the various circulations was related to drug treatment. Our inability to find a blood pressure-velocity relationship in the visceral arteries could have been the result of greater inaccuracies in the measurement of blood flow velocity in these arteries. This is unlikely, as we found no major differences in the reproducibility of measurements from the different arteries. Our data reflect the relationship between blood pressure and velocity at a single point in time, and we did not attempt to determine whether there were slow physiological variations in blood flow velocity in the visceral arteries, similar to those found in the cerebral circulation illI. The inferences we may draw from these findings depend on assumptions concerning the diameter of the vessels. Volume flow is the product of velocity in the artery and its cross-sectional area. In babies with identical volume blood flow through an artery, velocity may be higher in babies with narrower arteries. Unfortunately, in these very small infants, it is currently impossible to accurately measure the diameter of the arteries studied using ultrasound imaging. Both human and animal data suggest that cerebral artery velocities measured by Doppler ultrasound are proportional to volume flow [ 12,131, but one study proposes that in human neonates cerebral artery velocity may increase without any increase in volume flow [14]. It may be that cerebral artery blood flow velocity was higher in those infants with higher blood pressure, because of large artery vasoconstriction acting to maintain a constant microvascular pressure [ 151. It is surprising to find that renal artery blood flow velocity is unrelated to blood pressure on the first day of life, but this could be a consequence of the fact that the relationship is an indirect one, being mediated by autonomic drive and/or the reninangiotensin system. On the first day of life these systems may be maximally activated by the demands imposed by preterm delivery, so that all we subsequently observe is the autoregulatory ‘escape’ from their influence. By one week of age, when the activity of these systems has decreased [ 161, renal artery velocity is correlated with

232

S. T. Kempley. H. R. Gamsu / Early Hum. Dev. 34 (1993)

227-232

blood pressure. The response to hypoxia may also be mediated by adrenergic mechanisms, and maximal adrenergic activation on day 1 may explain why there is no relationship between Pao2 and velocity at this age, whereas there is evidence of a brain-sparing effect at 1 week of age. Alternatively, chemoreceptors may be set to respond to intrauterine norms of Pao2 on day 1, but by 1 week they could have been ‘reset’ in response to extrauterine oxygen levels. Our data illustrate that although cerebral artery blood flow velocity can be largely predicted from arterial blood pressure and blood gases, this is not the case for other regional circulations. On the first day of life, therefore, measurement of these variables may not identify infants with major disturbances of gut or renal haemodynamics. 5. References 1

Bode, H. and Wais, U. (1988): Age dependence

2

Dis. Child., 63, 606-611 Evans, D.H., Levene, M.I., Shortland, D.B. and Archer, L.N.J. (1988): Resistance index, blood flow velocity and resistance-area product in the cerebral arteries of very low birth weight infants during

3

4

5 6 7

8

9 10 II 12

13

14 15 16

of flow velocities

in basal cerebral

arteries.

Arch.

the first week of life. Ultrasound Med. Biol., 14, 103-l 10 Cowan, F. and Thoresen, M. (1987): The effects of intermittent positive pressure ventilation cerebral arterial and venous blood velocities in the newborn infant. Acta Paediatr. Stand., 239-247

on 76,

Shortland, D.B., Field, D., Archer, L.N.J., Gibson, N.A., Woods, K.L., Evans, D.H. and Levene, MI. (1989): Cerebral haemodynamic effects of changes in positive end expiratory pressure in pretenn infants. Arch. Dis. Child., 64, 465-469 Rennie, J.M., South, M. and Morley, C.J. (1987): Cerebral blood flow velocity variability in infants receiving assisted ventilation. Arch. Dis. Child., 62, 1247-1251 Colditz, P., Murphy, D., Rolfe, P. and Wilkinson, A.R. (1989): Effect of infusion rate of indomethacin on cerebrovascular responses in preterm neonates. Arch. Dis. Child., 64, 8-12 Levene, M.I., Shortland, D., Gibson, N. and Evans, D.H. (1988): Carbon dioxide reactivity of the cerebral circulation in extremely premature infants: effects of postnatal age and indomethacin. Pediatr. Res., 24, 175-179 Ahmann, P.A., Dykes, F.D., Lazzara, A., Holt, P.J., Giddens, D.P. and Carrigan, T.A. (1983): Relationship between pressure passivity and subependymal/intraventricular haemorrhage as assessed by pulsed Doppler ultrasound. Pediatrics, 72, 665-669 Kempley, S.T., Gamsu, H.R., Vyas, S. and Nicolaides, K. (1991) Effects of intrauterine growth retardation on postnatal visceral and cerebral blood flow velocity. Arch. Dis. Child., 66, 1115-I 118 Kempley, ST. and Gamsu, H.R. (1993): Arterial blood pressure and blood flow velocity in major cerebral and visceral arteries. Early Hum. Dev.. 35. In press, Coughtrey, H., Rennie, J.M. and Evans, D.H. (1992): Postnatal evolution of slow variability in cerebral artery blood flow velocity. Arch. Dis. Child., 67, 412-415 Greisen, G., Johansen, K., Ellison, P.H., Fredriksen, P.S., Mali, J. and Friis-Hansen, B. (1984): Cerebral blood flow in the newborn infant: Comparison of Doppler ultrasound and ‘33xenon clearance. J. Pediatr., 104, 41 I-418 Hansen, N.B., Stonestreet, B.S., Rosenkrantz, T.S. and Oh, W. (1983): Validity of Doppler measurements of anterior cerebral artery blood flow: correlation with brain blood flow in piglets. Pediatrics, 72, 526-531 Drayton, M.R. and Skidmore, R. (1987): Vasoactivity of the major intracranial arteries in newborn infants. Arch. Dis. Child., 62, 236-240 Faraci, F.M. and Heistad, D.D. (1990): Regulation of large cerebral arteries and cerebral microvascular pressure. Circ. Res., 66, 8-17 Baumgartner, H., Ritsch, R., Luz, 0.. Schneeberger, J. and Hammerer, 1. (1992): Capillary versus arterial plasma catecholamines as markers for sympatho-adrenal activity in infants. Pediatr. Res., 31, 579-582