Acute Changes in Cerebral Oxygenation and Cerebral Blood Volume in Preterm Infants During Surfactant Treatment By L. Skov l , L. Hellström- Westas 2 , T.]acobsen l , G. Greisen l and N.

w. Svenningsen2

IDepartment of Neonatology, State University Hospital, Rigshospitalet, Copenhagen, Denmark, and 2Neonatal Intensive Care Unit, Department of Paediatrics, University Hospital, Lund, Sweden

Following administration of surfactant a marked depression in aEEG activity occurs for about 10 minutes; the mechanism of this depression is unknoWll. In view of this, twenty-nine preterm infants were investigated with near infrared spectroscopy (NIRS) to evaluate rapid changes in total cerebral haemoglobin concentration and cerebral oxyhaemoglobin concentration during rescue treatment with natural surfactant. During surfactant instillation there was a short-Iasting hypoxaemia as demonstrated by pulseoximetry as weIl as a considerable fall in arterial blood pressure. With NIRS, tissue hypoxia was demonstrated by a drop in cerebral oxyhaemoglobin concentration. The marked drop in arterial blood pressure occurring immediately following surfactant was not matched by a drop in total cerebral haemoglobin concentration. This suggests that cerebral blood volume and hence cerebral blood flow was maintained. In the following minutes there was an improvement in cerebral oxygenation as indicated by the rise in cerebral oxyhaemoglobin concentration in nearly all the infants. The present study clearly suggests a hyperaemic hyperoxic state in the brain after surfactant rescue treatment and does not explain the transient EEG depression after surfactant treatment as a result of cerebral ischaemia/hypoxia.

Keywords Near infrared spectroscopy Preterm neonates - Surfactant treatment


Introduction Treatment of premature infants with porcine surfactant for severe idiopathic respiratory distress syndrome (lRDS) has resulted in a marked reduction in mortality and morbidity (2). Intraventricular haemorrhage (lVH) and IRDS

Received October 22, 1990; accepted April 2, 1991 Neuropediatrics 23 (1992) 126-130 © Hippokrates Verlag Stuttgart

Immediately following administration of natural surfactant acute changes occur in several physiological parameters. The main concern is a marked depression in EEG activity for about 10 minutes immediately or shortly after administration of surfactant (5, 6). In view of the rapid improvement in arterial oxygen saturation which usually follows the administration of surfactant the cause of this EEG depression is unclear. We, therefore, studied preterm infants with near infrared spectroscopy (NIRS) during administration of surfactant looking for evidence of acute cerebral ischaemia at levels which could be critical to electrophysiologic function.

Materials and methods Twenty-nine newborn infants admitted to the Department of Neonatology of Rigshospitalet in Copenhagen and the Department of Paediatrics of the University Hospital in Lund were studied from August 1989 to May 1990. The inclusion criteria for surfactant therapy were respiratory distress syndrome and requirement of more than 60 % oxygen on intermittent positive pressure ventilation (IPPV). Infants with congenital abnormalities or a history of prolonged rupture of the membranes were excluded. Informed parental consent was obtained in all cases and the study was approved by the Ethical Committees of Copenhagen and Frederiksberg and Lund University respectively. Changes in total cerebral haemoglobin concentration (tHb), and in oxygenated and deoxygenated haemoglobin concentration (Hb0 2 and Hb resp.) were measured by NIRS, as described previously (10). NIRS depends on the relative transparency of the neonate's head to near infrared light, which is attenuated by scattering and absorption in the tissues (8). In our NIR instrument (Radiometer, Denmark) we use three semiconductor laserdiodes with wavelengths 775 nm, 805 nm and 904 nm. The lasers are operated sequentially, and pulsed for 200 ns with a repetition rate of 500 Hz. Optical fiber bundIes are used to transmit and receive the light. By use of the extinction coefficients published by Susan Wray et al (13) and matrix inversion, algorithms were developed for the calculation of changes in oxygenated (Hb0 2 ) and deoxygenated haemoglobin concentration (HB). Changes in total cerebral haemoglobin concentration (tHb) were calculated as the sum of Hb0 2 and Hb. Changes in total blood volume was calculated as the prod-

Downloaded by: Universite de Sherbrooke. Copyrighted material.


are hoth common in preterm infants and these complications of prematurity are strongly associated. Despite this association no reduction in the risk of IVH has been documented following surfactant therapy (11).

Cerebral Haemodynamics During Surfactant Treatment

Neuropediatrics 23 (1992)


Hb02 Hb tHb mMxcm

uct of tHb with a constant divided with the haemoglobin concentration and the interoptode distanc~ (13). In this study we have used the assumption that the optical light pathlength is 4.39 times the interoptode distance obtained by measurement of time of flight of photons in infant brains post-mortem (15). Mean arterial blood pressure (MABP) was recorded from an indwelling arterial catheter if already in situ (25 out of 29) in the other four infants it was performed by oscillometry (Dinamap). Arterial oxygen (TcP0 2 ) and carbon dioxide (TcPC0 2 ) tensions were monitored by means of transcutaneous electrodes (TCM3, radiometer) and arterial blood oxygen saturation (Sa0 2 ) was measured by pulse oximetry (Satlite plus, Datex). All data were stored on computer for later analysis. Amplitude integrated EEG (aEEG) was recorded on paper by a cerebral function monitor (Criticon in Copenhagen, Lectromed Ltd. in Lund) from 21 of the infants (4).

IVH was diagnosed by ultrasonography using 6 MHz sector transducers (Bruel & Kyrer model 1867 Copenhagen, Technicare autoscanner in Lund) and graded according to the system of Papile, O-IV (9). Ultrasonography was performed before surfactant administration if possible and in all infants after. A bolus of 200 mg/kg porcine surfactant in a liquid solution (Curosurf, Stockholm) was instilled into the endotracheal tube via a feeding catheter placed with the tip just above the carina. Some infants received the surfactant by 1/2dose instillation in each lung, and others received the surfactant in one bolus dose lying supine. The infants had NIRS performed and the above mentioned physiological parameters measured continuously for approximately 15 minutes before surfactant instillation and for approximately 60 minutes after instillation. In 19 infants from Copenhagen endotracheal suctioning for L/S ratio was performed before the surfactant was given. For each infant, changes over 5 periods were selected for analysis: 1. The endotracheal suctioning for L/S ratio. 2. The period (0-2 min) during surfactant administration. 3. The period (2-10 min) (ol1owing surfactant administration. 4. The period 15 to 60 minutes after surfactant. 5. The entire study period 75 minutes (Fig. 1).

35 [


4 [ 3




Results The 29 treated babies had a mean birth weight of 1352 g) range 645-2410 g) and mean gestational age of 29.1 weeks (range 25.0 - 34.0 w). The median age at treatment was 13.7 hours (range 4-36 hours), (Table 1). Ultrasound scan was done before in 23 infants, of these had three IVH grade 1-11 and one IVH grade IV. In total had 12 babies IVH later: Six infants IVH grade 1-11 and six infants IVH grade III-IV. Thirteen out of the twenty-nine infants died, seven of them had IVH. 1. Clinical response to surfactant

Fifteen minutes after surfactant instillation there was a reduction in oxygen requirement (Fi0 2 ) in most babies. Before treatment a/A ratio (Pa0 2/[Fi0 2 x 0.95 - PaC0 2]) averaged 0.12 (range 0.04-0.20), and 15 minutes after treatment the ratio was 0.33 (range 0.11-0.69). Most babies rapidly turned pink and in some babies a marked general flushing of the skin (a tolazoline-like reaction) was seen. 2. Changes in aEEG

There was a marked depression in EEG activity for a 10-minute-period immediately or shortly after administration of surfactant, whereas there was no change during endotracheal suctioning. These results are fully reported elsewhere (6). 3. Changes in MABP, TcPC0 2, TcP0 2 andSa02

The transcutaneous oxygen tension and the Sa02 fell equally during the endotracheal suctioning and during the surfactant administration, whereas both parameters increased markedly following the surfactant (Fig. 2). In contrast MABP decreased markedly (mean -8.1 mmHg; 95 % confidence interval (CI95 %) - 10.2 to -6.1 mmHg) during the surfactant administration compared to the endotracheal suctioning (mean -2.0 mmHg; CI95 % -4.14 to 0.14 mmHg). The TcPC0 2 increased slightly following surfactant (mean 0.37 kPa; CI95 % 0.19 to 0.56 kPa) but in most infants we saw an overall fall in TcPC0 2 from before surfactant to the end ofthe study period (mean 0.64 kPa; CI95 % -1.40 to 0.12 kPa).

Downloaded by: Universite de Sherbrooke. Copyrighted material.

Fig. 1 Changes in concentration of Hb0 2 , Hb and tHb (three upper curves) and in Sa0 2 , MABP and TcPC0 2 (three lower curves) during surfactant administration. Period 1: Endotracheal suctioning. Period 2: The surfactant administration. Period 3: The period following su rfacta nt administration. Period 4: Fifteen to sixty minutes following surfactant administration. Period 5: The entire study period. The decrease in Hb0 2 following suctioning is nearly equal to the decrease immediately following surfactant administration. Later, following surfactant, a large rise in Hb0 2 is seen.

Neuropediatrics 23 (1992)

L. Skov et al

Table 1 Summarized data for the infants included in the study from the two hospitals (n BW



a/A 5 min before

a/A 15 min after






900 1050 1560 2120 980 1290 1375 2000 750 800 1055 645 1590 1405 1175 1865 820 1430 2410 1800 960 2000 1085 1930 1050 1180 1310 1865 870

25 28 29


0.144 0.170 0.186 0.071 0.094 0.111 0.065 0.109 0.202 0.158 0.070 0.116 0.147 0.042 0.071 0.202 0.099 0.178 0.153 0.081 0.156 0.060 0.050 0.170 0.100 0.070 0.100 0.104 0.117

0.458 0.297 0.252 0.267 0.350 0.238 0.383 0.632 0.136 0.158 0.317 0.512 0.241 0.148 0.367 0.544 0.462 0.289 0.397 0.177 0.348 0.689 0.109 0.330 0.290 0.120 0.217 0.280 0.307


12 25

33 28 29 29


19 11 15 8 9 4 17 9 21

33 27 27 27 25 32 28 29 31 27 29 34 31 27 32 28 29 27 29 31 32 28


20 13 28 20 5 27 11 36 9 7

6 9 10 9



MABP initial (mm Hg)

MABP change (mm Hg)

tHb change (mM*cm)

42 36

- 3

- 7

o.1 8 -0.11 0.28 -0.11 0.05 0.00 -0.04 -0.12 0.21 0.11 0.04 0.04 0.00 -0.11 0.11 0.11 -0.11 0.00 0.28 0.23 0.04 0.11 -0.04 0.04 -0.04 0.25 0.14 0.14 0.28

48 38 36 36 36 35 35 29 37 44 34 45 41 28 42 41

33 42 48 49 24 24 38 44 39 37

-18 -10 - 2

- 6 -11 -10 - 9

- 5 -10 -11 - 3 -11 -14 - 1 -14 - 6

- 5 - 8 -16 - 8

- 5 - 5 - 6 - 7 -13 - 7






o o o





o o



o o

4 3


o o 2 4

o 2 o o o

o o o







1 1




o o

o o o

o o o o

GA: gestational age, BW: birth weight, a/A: arterial to alveolar oxygen ratio, tHb total haemoglobin, MABP mean arterial blood pressure, US ultrasound image: 0 normal 1,2,3,4, IVH grade I-IV, A: alive, 0: death.

Pa02 (kPa)


8a02 (!Mt)

10 10

.:j ~






time period

time perIod

MABP (mmHg) 1.5

PaC02 (kPa)







time period



time period

Fig. 2 Mean changes (with 95 % confidence interval) in Pa0 2 , Sa0 2 , MABP and PaC0 2 during endotracheal suctioning (period 1) during surfactant administration (period 2) following surfactant (period 3, 4) and across the entire study period (period 5).

Downloaded by: Universite de Sherbrooke. Copyrighted material.


Cerebral H aemodynamics During Surfactant Treatment Hb02 (mmoI/L*cm)

Neuropediatrics 23 (1992)


6. The cerebral blood volume


The cerebral blood volume was calculated from 23 infants from Copenhagen (the interoptode distance was not measured in Lund). In these 23 children the rise in cerebral blood volume just following surfactant administration was 0.14 ml/l00 g (range -0.45 to 0.87 ml/l00 g). Using thecerebral blood volume instead of total haemoglobin did not change the correlations calculated above.

time per iod

Fig. 3

Mean changes (with 95 % confidence interval) in Hb0 21 Hb amT tHb during endotracheal suctioning (period 1) during surfactant administration (period 2) following surfactant (period 3, 4) and across the entire study period (period 5).

The mortality was higher in this study than in the European multicenter study supposed because the postnatal age was higher and the infants more sick. In this study we have searched for evidence of acute cerebral ischaemia, i. e. severe lowering of cerebral blood flow (CBF), as an explanation forthe aEEG changes seen for about 10 minutes after endotracheal surfactant instillation. In the present study, the rescue treatment with surfactant resulted in a short (30 s - 2 min) arterial hypoxaemia as weIl as tissue hypoxia as demonstrated by a drop in cerebral oxyhaemoglobin concentration (Hb0 2 ). The reduction in Hb0 2, however, was not significantly greater than the reduction which occurred during endotracheal suctioning and the latter was not associated with adepression of the aEEG. On the other hand, the considerable fall in MABP immediately following surfactant was much different from the reaction to the suctioning. The arterial hypotension combined with an arterial hypoxaemia, may be the cause of a temporary ischaemic insult of sufficient degree to account for the aEEG changes. However, the total cerebral haemoglobin increased after administration of surfactant suggesting that CBF was maintained.

It should be noted that with NIRS, the CBV is measured and not the CBF. Changes in cerebral blood volume - and cerebral blood flow are proportional in many situations. 4. Changes in Hb0 2 and tHb as measured by The CBV, however, reflects the diameter of veins, arteries and NIRS the number of open capillaries whereas CBF is determined by Cerebral oxyhaemoglobin decreased slightly mean arterial blood pressure, intracranial pressure, blood visduring surfactant administration (mean -0.12 mM*cm; CI95 % cosity as weIl as the diameter of arteries, arterioles and the -0.15 to -0.09 mM*cm) but was thenfollowed by a large in- numbers of open capillaries. Therefore, in some situations it crease (mean 0.18 mM*cm; CI95 % 0.13 to 0.23 mM*cm) may be possible to have an increased CBV, e. g. due to venous (Fig. 3). A similar decrease in oxyhaemoglobin occurred during stasis, without CBF being high. the endotracheal suctioning procedure (mean -0.12 mM*cm, Furthermore, quantification of cerebral blood 95 % CI - 0.15 to -0.09 mM*cm). In the ten minute period fol- volume by NIRS requires that the optical light pathlength is lowing surfactant we saw a significant increase in total haemo- known. We have assumed that the pathlength is proportional to globin the mean values being 0.06 mM*cm (CI95 % 0.002 to 0.11 the distance between the optodes independent of gestational mM*cm). In most infants there was a fall in tHb across the en- age. However, besides possible individual variations, it is likely tire study period (mean -0.04 mM*cm; CI95 % -0.09 to 0.01 that the pathlength increases with increased maturity, due to inmM*cm). creased dendritic arborization and myelination of white matter (1).

5. Changes in tHb versus MABP, pC02, and Sa02 We found no relation between the rise in tHb and the rise in Sa0 2 or the fall in MABP. On the other hand the changes in tHb from the beginning to the end of the study period were significantly related to the changes in pC0 2 (p < 0.01).

Acute effects of intratracheal surfactant instillation on internal carotid blood flow velocity (CBFv) have been examined using the Doppler sonographic technique showing a drop in CBFv at 5 and 10 minutes concurrently with a rise in blood pressure (3). Another study, also using this technique (7), showed no significant changes of cerebral blood velocity. Here, treatment was prophylactic and the blood velocity was measured every 10 minutes and thus rapid changes may have been overlooked.

Downloaded by: Universite de Sherbrooke. Copyrighted material.


130 Neuropediatrics 23 (1992)

In conclusion we confirm a transient, but marked depression of aEEG in all babies treated with surfactant lasting for 2-10 minutes, with a concurrent reduction in mean arterial blood pressure. There was, however, no evidence of cerebral ischaemia since the marked drop in arterial blood pressure occurring immediately following surfactant was not matched by a drop in total cerebral haemoglobin concentration. As it has recently been reported that hypoxanthine concentration increases after surfactant treatment (12) it may be specuIated that hyperaemia during reperfusion with influx of neurotoxins such as oxygen free radicals may be associated with depression of cerebral activity.

Acknowledgments The surfactant used in the present trial was developed by T. Curstedt and B. Robertson, Stockholm, with financial support from the Swedish Medical Research Counsil (Project 3351) and Oscar 11: s Jubileumsfond. The Gerda and Age Haench Foundation, the Dagmar Marschal Foundation, Swedish Medical Research Council grant no. 4732, Lund University Med. Faculty Research Foundation, Greta ö Johan Kocks Foundation, Josef and Linnea Carlsson Research Foundation gave financial support.

3 Cowan, F., M. Silverman, D. Wertheim, A. Whitelaw: The acute effects of surfactant administration on blood pressure and cerebral blood flow velocity in the neonate. Proceedings of the 4th workshop on surfactantreplacement 14 (1989) 4 Greisen, G., L. Hellström-Westas, H. Lou, 1. Rosen, N. W. Svenningsen: EEG depression and germinallayer haemorrhage in the newborn. Acta Paediatr. Scand. 76 (1987) 519-525 5 Hellström- Westas, L., 1. Rosen, N. W. Svenningsen: Cerebral complications detected by EEG monitoring during neonatal intensive care. Acta Paediatr. Scand. Supp!. 360 (1990) 83-86 6 Hellström- Westas, L., A. Bell, L. Skov, G. Greisen, N. W. Swenningsen: Cerebro-electrical depression following surfactant treatment. In press. 7 lorch, G., H. Rabe, M. Garbe, E. Michel, L. Gortner: Acute and protracted effects of intratracheal surfactant application on carotid blood flow velocity, blood pressure and carbondioxide tension in very low birth-weight infants. Eur. J. Pediatr. 148 (1989) 770-773 8 ]@bsis, F. F.: Noninvasive infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 198 (1977) 1264-1267 9 Papile, L.,]. Burstein, R. Burstein, H. KojJler: Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1500 gm. J. Pediatr. 92 (1978) 529-534 10 Pryds, 0., G. Greisen, L. Skov, B. Friis-Hansen: Carbon dioxide related changes in cerebral blood volume and cerebral blood flow in mechanicaly ventilated, preterm neonates. Comparison of near infra-red spectrophotometry and 133-xenon clearance. Pediatr. Res. 27 (1990) 445-449 11 Robertson, B.: Neonatal respiratory distress syndrome and surfactant therapy; a brief review. Eur. Respir. J. 2 (Supp!. 3) (1989) 73-76 12 Saugstad, O. D., K. Gloppestad, T Curstedt, B. Robertson: Washing out of hypoxantine after treatment with curosurf. Proceedings of the 4th workshop on surfactant-replacement 32 (1989) 13 Wray, 5., M. Cope, D. T Delpy,]. S. Wyatt, E. O. R. Reynolds: Characterization of the near infrared absorption spectra of cytochrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochem. Biophys. Acta 933 (1988) 184-192 14 Wyatt,]. 5., M. Cope, D. T Delpy, S. Wray, E. O. R. Reynolds: Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 2 (1986) 1063-1066 15 Wyatt,]. 5., M. Cope, D. T Delphy, P. van der Zee, s. Arridge, A. D. Edwards, E. O. R. Reynolds: Measurement of optical path length for cerebral near-infrared spectroscopy in newborn infants. Dev. Neurosci. 12 (1990) 140-144

References 1 Benaron, D. A., S. Gwiazdowski, C. D. Kurth,]. Steven, B. Chance: Cerebral changes with growth in infants by hemoglobin phase-shift spectroscopy. Pediatr. Res. 27 (1990) 4 (38. Abstract). 2 Collaborative European multicenter study group: Surfactant replacement therapy for severe neonatal respiratory distress syndrome: An international randomized clinical trial. Pediatrics 82 (1988) 683-691

L. Skov, M. D. Dept. of Neonatology State University Hospital Rigshospitalet Blegdamsvej 9 DK-2100 Copenhagen, Denmark

Downloaded by: Universite de Sherbrooke. Copyrighted material.

The rise in total cerebral haemoglobin concentration, found in this study, may be due to a vasodilatation that includes the cerebral vascular bed. The skin flushing, sometimes seen, supports the assumption that peripheral vasodilation may be involved and could explain the fall in blood pressure. Another explanation to the rise in tHb could be a carbon dioxide-induced cerebral vasodilatation. The CBV response to a small (approx 1 kPa) induced change in PaC0 2 can be calculated to 0.06 to 0.08 ml/100 gin preterm infants as earlier reported (10). In the present study, however, the rise in CBV was about 0.14 mi/100 g and the rise in PaC0 2 0.4 kPa only.

L. Skov et al

Acute changes in cerebral oxygenation and cerebral blood volume in preterm infants during surfactant treatment.

Following administration of surfactant a marked depression in aEEG activity occurs for about 10 minutes; the mechanism of this depression is unknown. ...
147KB Sizes 0 Downloads 0 Views