Effects of maternal administration of 3% carbon dioxide on umbilical artery and fetal renal and middle cerebral artery Doppler waveforms Jean-Claude VeiIle, MD, and Mary Penry, RN, RDMS
Winston-Salem, North Carolina OBJECTIVE: The null hypothesis is that umbilical, middle cerebral, and renal artery pulsed Doppler velocity waveforms in the normal term fetus may be affected during short-term maternal inhalation of 3% carbon dioxide gas mixture. STUDY DESIGN: Seventy-two observations were made on 14 term fetuses before and during maternal 3% carbon dioxide gas mixture inhalation. The umbilical, middle cerebral, and renal arteries of these fetuses were sampled with pulsed Doppl~r velocity, waveforms and recorded on a strip chart at a preset speed of 50 mm/sec. Doppler waveforms' were analyzed for differences in the systolic peak to end-diastolic velocity ratio for these three vascular beds. Peak flow velocity and time velocity integral were also analyzed for the cerebral and renal vascular beds. The data were analyzed with the paired t test. RESULTS: A significant decrease in the systolic-to-diastolic-velocity ratio of the middle cerebral artery occurred with 3% carbon dioxide inhalation (p < 0.02). The other vascular beds had no demonstrable change. CONCLUSION: Transient maternal breathing of 3% carbon dioxide gas mixture selectively causes a decrease in resistance in the fetal cerebral circulation. (AM J OBSTET GYNECOL 1992;167:1668-71.)
Key words: Pulsed Doppler waveforms, maternal carbon dioxide inhalation Respiratory acidosis can occur during the course of normal labor. Most transient episodes of respiratory acidosis are well tolerated by healthy fetuses, I but such biochemical alterations may in turn alter resistance of regional vascular beds in response to a local increase in the concentration of carbon dioxide."' , Maternal inhalation of concentrations of 3% to 5% of carbon dioxide have been found to be safe and relatively well tolerated by patients. During inhalation of such gas mixture fetal breathing movements increase significantly without affecting fetal heart rate in uncomplicated pregnancies. 4 The effects of transient maternal hypercapnia on term human fetal regional circulation has not been previously reported. The purpose of the current study was to determine the effects of maternal inhalation of a 3% carbon dioxide gas mixture on the pulsed Doppler velocity waveforms of three fetal vascular beds: umbilical artery, fetal renal artery, fetal middle cerebral artery. The study goals were as follows: (1) to document the From the Department of Obstetrics and Gynecology, Bowman Gray School of Medicine of Wake Forest University. Supported in part by National Institutes of Health, National Heart, Lung and Blood Institute grant No. HL38296. Presented at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21, 1992. Reprint requests: Jean-Claude Veil/e, MD, Department of Obstetrics and Gynecology, Bowman Gray School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157. 6/6/41726 1668
response to increasing maternal hypercapnia of these vascular beds, (2) to document any regional differences between these vascular beds, and (3) to document and quantify the extent of the response to carbon dioxide. Material and methods Fourteen patients were asked to participate in this study. All were ~ 36 weeks of pregnancy and had no complication during pregnancy. All gave informed consent that was approved by the institutional review board of the hospital. All patients were in the left lateral decubitus position during the entire period of observation. None were taking or had received any medications before the study other than prenatal vitamins. A comprehensive ultrasonographic examination was performed to document fetal anatomy and placental location. The pulsed Doppler sample volume was sequentially placed in the lumen of the umbilical artery at its placental insertion, in the lumen of the renal artery, and in the lumen of the middle cerebral artery. The pulsed Doppler sample volume and the pulsed Doppler angle were kept as small as possible so as to be within the vessel and parallel to the vascular bed that was analyzed. The pulsed Doppler power output was set to the lowest values, and the filter gate was set at 50 mHz. The Doppler time exposure was kept to a minimum. Pulsed Doppler velocity waveforms of these vascular beds were recorded on a strip chart that was preset at a speed of
Carbon dioxide and fetal circulation
Volume 167 Number 6
Table I. Results of Doppler waveforms Vascular bed
I Period 1 I
Period 2
Umbilical systolic- 2.6 ± 0.1 3.0 ± 0.1 to-diastolic ratio Renal systolic-to6.6 ± 0.6 7.2 ± 0.5 diastolic ratio Middle cerebral 4.7 ± 0.4 3.65 ± 0.4 systolic-todiastolic ratio
1669
Table II. Results of Doppler waveforms
ISignificance p NS NS p < 0.02
NS, Not significant. Period 1, Control period before carbon dioxide inhalation; period 2, experimental period during 3% carbon dixoide inhalation.
50 mm/sec." All the Doppler velocity waveforms were obtained during the baseline period while the fetus was not active or making fetal breathing movements. The patient was then asked to place a tightly fitting mask over her nose and mouth for 5 minutes, during which time the patient inhaled room air (acclimatization period). Maternal end-tidal Pco 2 was monitored through an outport from the mask, which was connected to an on-line infrared carbon dioxide analyzer. Mter this initial 5-minute period, a 3% carbon dioxide gas mixture was added for 15 minutes at a flow rate of 5 Vmin. Maternal end-tidal Pc0 2 was again recorded during carbon dioxide inhalation while the Doppler waveforms were obtained. Pulsed Doppler velocity waveforms of the umbilical, renal, and middle cerebral arteries were obtained during the tenth to the fifteenth minute of the carbon dioxide inhalation. Although carbon dioxide stimulated fetal breathing movements, all Doppler velocity waveforms were obtained during apnea. A minimum of three to six pulsed Doppler velocity waveforms per vascular beds were digitized with an electronic handheld pen and a tablet connected to an on-line personal computer (Digisonics, Houston, Tex) and averaged. The ratio of the peak systolic velocity to the minimum end-diastolic velocity was computed for the three vascular beds. Peak flow velocity and the time velocity integral (velocities under the curve) were also computed for the fetal renal and middle cerebral artery Doppler waveforms. A paired t test was performed to determine significance. A p value < 0.05 was used to reject the null hypothesis.
Results A total of 72 observations was obtained on the 14 fetuses before (period 1) and just before the end of (period 2) maternal inhalation of the 3% carbon dioxide gas mixture. Doppler waveforms were obtained during both periods in all 14 fetuses for the umbilical artery, in 13 fetuses for the renal artery, and in nine fetuses for the middle cerebral artery. Our inability to sample all of the vascular beds during both study periods was due to technical difficulty, change in fetal
Vascular bed
Fetal renal artery 40.4 ± 2.9 43.6 ± 2.9 peak flow velocity (cm/sec) Fetal renal artery 7.7 ± 0.6 8.2 ± 0.7 time velocity integral (cm) Middle cerebral ar- 36.9 ± 4.0 38.7 ± 4.3 tery peak flow velocity (cm/sec) Middle cerebral ar- 8.6 ± 0.9 10.4 ± 1.2 tery time velocity integral (cm)
Significance p
NS NS NS NS
NS, Not significant. Period 1, Control period before carbon dioxide inhalation; period 2, experimental period during 3% carbon dioxide inhalation.
position, or fetal breathing movement. Maternal baseline values for end-tidal PC02 varied from 18% to 23% (period I) and between 19% and 36% during the 3% carbon dioxide inhalation (period 2). The mean increase in end-tidal Pco 2 during these two time periods was 11.5% ± 2.4%. Results of the Doppler waveforms from these three vascular beds are shown in Tables I and II and are expressed as mean and standard error. Maternal inhalation of a 3% carbon dioxide gas mixture significantly decreased the pulsed Doppler peak systolic-to-end-diastolic velocity ratio of the middle cerebral artery without significantly changing the peak flow velocity or the time velocity integral. The peak systolic-to-end-diastolic ratio of the Doppler waveform of both the umbilical artery and the fetal renal artery did not change significantly during maternal inhalation of 3% carbon dioxide. Peak flow velocity and time velocity integral of the fetal renal artery Doppler waveform did not change during these two periods of observation. There was no correlation between the peak systolic velocity-to-end-diastolic ratio and maternal end-tidal Pc0 2 (r = - 0.02). The change that occurred in this ratio between the two periods of observation was related to the change in maternal end-tidal Pco 2 (r = 0.69). However, the p value for this correlation, was 0.056.
Comment A significant decrease in the peak systolic-to-enddiastolic ratio of the pulsed Doppler velocity waveforms of the middle cerebral artery was found during maternal inhalation of a 3% carbon dioxide gas mixture. The other two vascular beds studied did not significantly change during maternal hypercapnia. The decrease observed in the middle cerebral artery Doppler waveform was due to an increase in the diastolic component of the Doppler velocity waveform. This suggests that maternal inhalation of 3% carbon dioxide causes a
1670 Veille and Penry
selective decrease in the vascular resistance of the human fetal cerebral circulation. The effect of carbon dioxide on cerebral vasculature is known to have a significant relaxation on the cerebral vascular muscles and consequently influences vascular resistance. 6 These observed effects may be due to direct or local effect of a higher carbon dioxide concentration in the fetal blood. Intracerebral or extracellular cerebral spinal fluid pH changes that also occur when Paco 2 increases may affect cerebral flood flow. 7 Carbon dioxide alone, however, has not been found to be a very potent vasoactive agent. Such effects cause not only a decrease in cerebral vascular resistance but also a curvilinear increase in cerebral blood flow and blood volume. 8 The response to increases in Pac0 2 is complex and unique when compared with other vascular beds. The response to increasing Pac0 2 depends on the degree of maturation of the cerebral circulation, the region of the brain studied, and the type of species used for the experimentation. 9 In the sheep, for example, the cerebrovascular response to carbon dioxide is less pronounced in the fetus than in the adult. However, if one compares the response of specific areas of the brain to carbon dioxide, the results differ. The brain stem is more responsive to changes in blood flow with hypercapnia in the fetal lamb when compared with the adult. The cerebral region of the adult sheep, on the other hand, is much more sensitive to high carbon dioxide concentration than is the fetal lamb. It is postulated that the variability of the response of the cerebral vascular bed to increasing Pac0 2 is due to a difference in the rate of myelination of the brain tissue which occurs at different rates, depending on its region. As a result, metabolic demands and oxygen consumption vary, which in turn influences regional response to Paco 2 and cerebral blood flow. \0 Thus both maturation and regional metabolism influence the response of the cerebral circulation to increasing levels of carbon dioxide. The current study addresses the response of the middle cerebral vascular bed only in term fetuses. The same vascular bed in less mature human fetuses may respond differently. In 11 term healthy newborn infants, Archer et al. II used a 5% carbon dioxide gas mixture to study the response of the anterior cerebral artery by pulsed Doppler. He found that the pulsatility index (pulsatility index == systolic - diastolic/systolic) significantly decreased with rising end-tidal carbon dioxide. This fall in the pulsatility index was caused by an increase in the diastolic velocity. These authors concluded that in healthy newborns hypercapnia caused cerebral vasodilatation, increased cerebral blood flow velocity, and cerebral blood flow. II In our study on term fetuses we found a decrease in the peak systolic-to-end-diastolic velocity ratio during maternal hypercapnia. This was
December 1992 Am J Obstet Gynecol
due to an increase in values in the end-diastolic velocity component of the Doppler waveform. Although both the peak flow velocity and the total velocity integral of the Doppler waveform increased during maternal carbon dioxide inhalation, this increase was not statistically significant. This would suggest that in the term human fetus the decrease in the systolic-to-diastolic ratio we observed is most likely due to a decrease in the downstream resistance, which most likely occurs at the level of the cerebral microvasculature. I2 It is unlikely, however, that the vasodilatory response we observed during hypercapnia is a secondary effect to dilatation of the large cerebral arteries at the base of the brain of these healthy fetuses. If large cerebral arteries had significantly dilated, cerebral blood flow should have increased. In turn, such increase in cerebral blood flow should have increased the total velocity integral (i.e., the area under the curve) in the middle cerebral artery." We could not document any changes in either the time velocity integral or the peak flow velocity within this vascular bed during our two periods of observation. For now, whether the increase in diastolic velocity of the middle cerebral artery is associated with a corresponding increase in cerebral blood flow remains speculative. The current results suggest, however, that in term, healthy human fetuses the middle cerebral artery vasodilates during maternal hypercapnia. This study demonstrates a differential response to hypercapnia among these three vascular beds. It also shows that the higher the maternal concentration of carbon dioxide, the higher the vasodilatory effect of the middle cerebral artery, as found in healthy adults. 12 No significant change in the Doppler velocity waveforms during maternal inhalation of a 3% carbon dioxide gas mixture could be found in the other two vascular beds. The peak systolic-to-end-diastolic velocities ratio in the umbilical artery circulation during maternal hypercapnia did not significantly change. This would suggest that this particular bed does not readily respond to such increases in maternal carbon dioxide. This insensitivity or low sensitivity may reflect the fact that the umbilical circulation is a passive circulation. 14 Similar observations were also found in the ewe during 3% to 10% carbon dioxide administration. In this experimental model no change in the umbilical vascular resistance occurred. Even when the fetal arterial Pco 2 rose to 80 mm Hg, only a trivial but nonsignificant increase in umbilical vascular resistance could be documented. Is The baseline values for the ratio between the peak systolic-to-end-diastolic velocities of the fetal renal artery were elevated, as we previously found in a group of term normal fetuses. 5 This high baseline value for the pulsed Doppler velocity waveforms reflects high resistance of this particular vascular bed. 16 The ratio of peak
Volume 167 Number 6
systolic to end-diastolic velocities and the peak flow velocity and time velocity integral in the fetal renal artery did not change significantly during maternal inhalation of 3% carbon dioxide gas mixture. Like the umbilical vascular bed, the fetal renal artery does not seem to be as sensitive as the middle cerebral artery to maternal hypercapnia. Fetal renal vascular resistance has been reported to increase during progressive hypercapnia in a semichronic lamb preparation during administration of a 20% carbon dioxide gas mixture for 2 to 3 minutes to the ewe. I7 In this experimental model the rise in the resistance of the fetal renal artery was proportional to the rise in the fetal Pac0 2 • The higher the Pac0 2 , the higher the fetal renal artery vascular resistance. Fetuses in which Paco 2 did not rise significantly did not show this increase in fetal renal artery resistance. In the current study neither the peak systolic-to-end-diastolic velocities ratio, the peak flow velocity, nor the time velocity integral of the pulsed Doppler velocity waveform of the human fetal renal artery changed significantly during maternal hypercapnia. This suggests that this vascular bed in the term fetus is not significantly affected by this particular level of maternal hypercapnia. Whether a higher concentration of carbon dioxide gas mixture could have caused a change in the response of this vascular bed remains to be studied. This study shows that maternal inhalation of 3% carbon dioxide selectively affects vascular beds in the normal term human fetus. Although the umbilical and renal vascular beds did not have a change in peak systolic to end-diastolic velocities, vasodilatation occured in the middle cerebral vascular bed as evidenced by an increase in the diastolic velocity of the Doppler waveform. This vascular bed appears to be much more sensitive to maternal hypercapnia. This finding would support our hypothesis that this particular cerebral vascular bed responds differently to maternal hypercapnia than other vascular beds. It would also support the concept of the uniqueness of the response of this vascular bed to hypercapnia, as previously described in experimental models. It would also suggest that the response hypercapnia causes in adults and in other fetal experimental animals is operational in term human fetuses. I3 Whether this uniqueness is found in preterm fetuses or in fetuses with growth abnormality remains to be elucidated.
Carbon dioxide and fetal circulation
1671
REFERENCES
1. Dawes GS. Foetal blood gas tensions and pH. In: Foetal and neonatal physiology. Chicago: Year Book, 1968: 10616. 2. Seldon RE, Peeters LLH, Jones MD ]r, Makowski EL, Meschia G. Redistribution of cardiac output and oxygen delivery in the hypoxemic fetal Iamb. AM] OBSTET GYNECOL 1979;135:1071-8. 3. Cohn HE, Sacks E], Heymann MA, Rudolph AM. Cardiovascular responses to hypoxemia and acidemia in fetal lambs. AM] OBSTET GYNECOL 1974;120:817-24. 4. Ritchie ]WK, Lakhani K. Fetal breathing movements in response to maternal inhalation of 5% carbon dioxide. AM J OBSTET GYNECOL 1980;386-9. 5. Veille JC, Kanaan C. Duplex Doppler ultrasonographic evaluation of the fetal renal artery in normal and abnormal fetuses. AM.J OBSTET GYNECOL 1989; 161: 1502-7. 6. Heistad DD, Kontos HA. Cerebral circulation. In: Volume 3: handbook of physiology. Section 2: the cardiovascular system. Part I: peripheral circulation and organ blood flow, Bethesda, Maryland: American Physiological Society, 1983: 137-82. 7. Lassen NA. Brain extracellular pH: the main factor controlling cerebral blood flow. Scand J Clin Lab Invest 1968; 22:247-51. 8. Reivich M. Arterial Pco 2 and cerebral hemodynamics. Am J Physiol 1964;206:25-35. 9. Rosenberg AA, Jones MD Jr, Traystman RJ, Simmons MA, Molteni RA. Response of cerebral blood flow to changes in pC02 in fetal, newborn, and adult sheep. Am J Physiol 1982;242:H862-6. 10. Himvich HE, Fazekas JF. Comparative studies of the metabolism of the brain of infant and adult dogs. Am ] PhysioI1941;132:454-9. 11. Archer LJ, Evans DH, Paton JY, Levene MI. Controlled hypercapnia and neonatal cerebral artery Doppler ultrasound waveforms. Pediatr Res 1986;20:218-21. 12. Ogawa S, Handa N, Matsumoto M, et al. Carbon dioxide reactivity of the blood flow in human basilar artery estimated by the transcranial Doppler method in normal men: a comparison with that ofthe middle cerebral artery. Ultrasound Med Bioi 1988;14:479-83. 13. Abboub FM. Special characteristics of the cerebral circulation. Fed Proc 1981 ;40:2296-2300. 14. Walker AM, Oakes G, McLaughlin M, Ehrenkranz R, Chez RA. Effects of hypercapnia on uterine and umbilical circulations in conscious pregnant sheep. J AppJ Physiol 1976;41 :727-33. 15. Dawes GS. The umbilical circulation. In: Foetal and neonatal physiology. Chicago: Year Book, 1968:66-78. 16. Vaughn D, Kirschbaum TH, Bersentes T, Dilts PV Jr, Assali NS. Fetal and neonatal response to acid loading in the sheep.] Appl Physiol 1968;24:135-41. 17. Beguin F, Dunnihoo DR, Quilligan EJ. Effect of carbon dioxide elevation on renal blood flow in the fetal lamb in utero. AM J OBSTET GYNECOL 1974;119:630-7.
Winston-Salem, North Carolina OBJECTIVE: The null hypothesis is that umbilical, middle cerebral, and renal artery pulsed Doppler velocity waveforms in the normal term fetus may be affected during short-term maternal inhalation of 3% carbon dioxide gas mixture. STUDY DESIGN: Seventy-two observations were made on 14 term fetuses before and during maternal 3% carbon dioxide gas mixture inhalation. The umbilical, middle cerebral, and renal arteries of these fetuses were sampled with pulsed Doppl~r velocity, waveforms and recorded on a strip chart at a preset speed of 50 mm/sec. Doppler waveforms' were analyzed for differences in the systolic peak to end-diastolic velocity ratio for these three vascular beds. Peak flow velocity and time velocity integral were also analyzed for the cerebral and renal vascular beds. The data were analyzed with the paired t test. RESULTS: A significant decrease in the systolic-to-diastolic-velocity ratio of the middle cerebral artery occurred with 3% carbon dioxide inhalation (p < 0.02). The other vascular beds had no demonstrable change. CONCLUSION: Transient maternal breathing of 3% carbon dioxide gas mixture selectively causes a decrease in resistance in the fetal cerebral circulation. (AM J OBSTET GYNECOL 1992;167:1668-71.)
Key words: Pulsed Doppler waveforms, maternal carbon dioxide inhalation Respiratory acidosis can occur during the course of normal labor. Most transient episodes of respiratory acidosis are well tolerated by healthy fetuses, I but such biochemical alterations may in turn alter resistance of regional vascular beds in response to a local increase in the concentration of carbon dioxide."' , Maternal inhalation of concentrations of 3% to 5% of carbon dioxide have been found to be safe and relatively well tolerated by patients. During inhalation of such gas mixture fetal breathing movements increase significantly without affecting fetal heart rate in uncomplicated pregnancies. 4 The effects of transient maternal hypercapnia on term human fetal regional circulation has not been previously reported. The purpose of the current study was to determine the effects of maternal inhalation of a 3% carbon dioxide gas mixture on the pulsed Doppler velocity waveforms of three fetal vascular beds: umbilical artery, fetal renal artery, fetal middle cerebral artery. The study goals were as follows: (1) to document the From the Department of Obstetrics and Gynecology, Bowman Gray School of Medicine of Wake Forest University. Supported in part by National Institutes of Health, National Heart, Lung and Blood Institute grant No. HL38296. Presented at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21, 1992. Reprint requests: Jean-Claude Veil/e, MD, Department of Obstetrics and Gynecology, Bowman Gray School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157. 6/6/41726 1668
response to increasing maternal hypercapnia of these vascular beds, (2) to document any regional differences between these vascular beds, and (3) to document and quantify the extent of the response to carbon dioxide. Material and methods Fourteen patients were asked to participate in this study. All were ~ 36 weeks of pregnancy and had no complication during pregnancy. All gave informed consent that was approved by the institutional review board of the hospital. All patients were in the left lateral decubitus position during the entire period of observation. None were taking or had received any medications before the study other than prenatal vitamins. A comprehensive ultrasonographic examination was performed to document fetal anatomy and placental location. The pulsed Doppler sample volume was sequentially placed in the lumen of the umbilical artery at its placental insertion, in the lumen of the renal artery, and in the lumen of the middle cerebral artery. The pulsed Doppler sample volume and the pulsed Doppler angle were kept as small as possible so as to be within the vessel and parallel to the vascular bed that was analyzed. The pulsed Doppler power output was set to the lowest values, and the filter gate was set at 50 mHz. The Doppler time exposure was kept to a minimum. Pulsed Doppler velocity waveforms of these vascular beds were recorded on a strip chart that was preset at a speed of
Carbon dioxide and fetal circulation
Volume 167 Number 6
Table I. Results of Doppler waveforms Vascular bed
I Period 1 I
Period 2
Umbilical systolic- 2.6 ± 0.1 3.0 ± 0.1 to-diastolic ratio Renal systolic-to6.6 ± 0.6 7.2 ± 0.5 diastolic ratio Middle cerebral 4.7 ± 0.4 3.65 ± 0.4 systolic-todiastolic ratio
1669
Table II. Results of Doppler waveforms
ISignificance p NS NS p < 0.02
NS, Not significant. Period 1, Control period before carbon dioxide inhalation; period 2, experimental period during 3% carbon dixoide inhalation.
50 mm/sec." All the Doppler velocity waveforms were obtained during the baseline period while the fetus was not active or making fetal breathing movements. The patient was then asked to place a tightly fitting mask over her nose and mouth for 5 minutes, during which time the patient inhaled room air (acclimatization period). Maternal end-tidal Pco 2 was monitored through an outport from the mask, which was connected to an on-line infrared carbon dioxide analyzer. Mter this initial 5-minute period, a 3% carbon dioxide gas mixture was added for 15 minutes at a flow rate of 5 Vmin. Maternal end-tidal Pc0 2 was again recorded during carbon dioxide inhalation while the Doppler waveforms were obtained. Pulsed Doppler velocity waveforms of the umbilical, renal, and middle cerebral arteries were obtained during the tenth to the fifteenth minute of the carbon dioxide inhalation. Although carbon dioxide stimulated fetal breathing movements, all Doppler velocity waveforms were obtained during apnea. A minimum of three to six pulsed Doppler velocity waveforms per vascular beds were digitized with an electronic handheld pen and a tablet connected to an on-line personal computer (Digisonics, Houston, Tex) and averaged. The ratio of the peak systolic velocity to the minimum end-diastolic velocity was computed for the three vascular beds. Peak flow velocity and the time velocity integral (velocities under the curve) were also computed for the fetal renal and middle cerebral artery Doppler waveforms. A paired t test was performed to determine significance. A p value < 0.05 was used to reject the null hypothesis.
Results A total of 72 observations was obtained on the 14 fetuses before (period 1) and just before the end of (period 2) maternal inhalation of the 3% carbon dioxide gas mixture. Doppler waveforms were obtained during both periods in all 14 fetuses for the umbilical artery, in 13 fetuses for the renal artery, and in nine fetuses for the middle cerebral artery. Our inability to sample all of the vascular beds during both study periods was due to technical difficulty, change in fetal
Vascular bed
Fetal renal artery 40.4 ± 2.9 43.6 ± 2.9 peak flow velocity (cm/sec) Fetal renal artery 7.7 ± 0.6 8.2 ± 0.7 time velocity integral (cm) Middle cerebral ar- 36.9 ± 4.0 38.7 ± 4.3 tery peak flow velocity (cm/sec) Middle cerebral ar- 8.6 ± 0.9 10.4 ± 1.2 tery time velocity integral (cm)
Significance p
NS NS NS NS
NS, Not significant. Period 1, Control period before carbon dioxide inhalation; period 2, experimental period during 3% carbon dioxide inhalation.
position, or fetal breathing movement. Maternal baseline values for end-tidal PC02 varied from 18% to 23% (period I) and between 19% and 36% during the 3% carbon dioxide inhalation (period 2). The mean increase in end-tidal Pco 2 during these two time periods was 11.5% ± 2.4%. Results of the Doppler waveforms from these three vascular beds are shown in Tables I and II and are expressed as mean and standard error. Maternal inhalation of a 3% carbon dioxide gas mixture significantly decreased the pulsed Doppler peak systolic-to-end-diastolic velocity ratio of the middle cerebral artery without significantly changing the peak flow velocity or the time velocity integral. The peak systolic-to-end-diastolic ratio of the Doppler waveform of both the umbilical artery and the fetal renal artery did not change significantly during maternal inhalation of 3% carbon dioxide. Peak flow velocity and time velocity integral of the fetal renal artery Doppler waveform did not change during these two periods of observation. There was no correlation between the peak systolic velocity-to-end-diastolic ratio and maternal end-tidal Pc0 2 (r = - 0.02). The change that occurred in this ratio between the two periods of observation was related to the change in maternal end-tidal Pco 2 (r = 0.69). However, the p value for this correlation, was 0.056.
Comment A significant decrease in the peak systolic-to-enddiastolic ratio of the pulsed Doppler velocity waveforms of the middle cerebral artery was found during maternal inhalation of a 3% carbon dioxide gas mixture. The other two vascular beds studied did not significantly change during maternal hypercapnia. The decrease observed in the middle cerebral artery Doppler waveform was due to an increase in the diastolic component of the Doppler velocity waveform. This suggests that maternal inhalation of 3% carbon dioxide causes a
1670 Veille and Penry
selective decrease in the vascular resistance of the human fetal cerebral circulation. The effect of carbon dioxide on cerebral vasculature is known to have a significant relaxation on the cerebral vascular muscles and consequently influences vascular resistance. 6 These observed effects may be due to direct or local effect of a higher carbon dioxide concentration in the fetal blood. Intracerebral or extracellular cerebral spinal fluid pH changes that also occur when Paco 2 increases may affect cerebral flood flow. 7 Carbon dioxide alone, however, has not been found to be a very potent vasoactive agent. Such effects cause not only a decrease in cerebral vascular resistance but also a curvilinear increase in cerebral blood flow and blood volume. 8 The response to increases in Pac0 2 is complex and unique when compared with other vascular beds. The response to increasing Pac0 2 depends on the degree of maturation of the cerebral circulation, the region of the brain studied, and the type of species used for the experimentation. 9 In the sheep, for example, the cerebrovascular response to carbon dioxide is less pronounced in the fetus than in the adult. However, if one compares the response of specific areas of the brain to carbon dioxide, the results differ. The brain stem is more responsive to changes in blood flow with hypercapnia in the fetal lamb when compared with the adult. The cerebral region of the adult sheep, on the other hand, is much more sensitive to high carbon dioxide concentration than is the fetal lamb. It is postulated that the variability of the response of the cerebral vascular bed to increasing Pac0 2 is due to a difference in the rate of myelination of the brain tissue which occurs at different rates, depending on its region. As a result, metabolic demands and oxygen consumption vary, which in turn influences regional response to Paco 2 and cerebral blood flow. \0 Thus both maturation and regional metabolism influence the response of the cerebral circulation to increasing levels of carbon dioxide. The current study addresses the response of the middle cerebral vascular bed only in term fetuses. The same vascular bed in less mature human fetuses may respond differently. In 11 term healthy newborn infants, Archer et al. II used a 5% carbon dioxide gas mixture to study the response of the anterior cerebral artery by pulsed Doppler. He found that the pulsatility index (pulsatility index == systolic - diastolic/systolic) significantly decreased with rising end-tidal carbon dioxide. This fall in the pulsatility index was caused by an increase in the diastolic velocity. These authors concluded that in healthy newborns hypercapnia caused cerebral vasodilatation, increased cerebral blood flow velocity, and cerebral blood flow. II In our study on term fetuses we found a decrease in the peak systolic-to-end-diastolic velocity ratio during maternal hypercapnia. This was
December 1992 Am J Obstet Gynecol
due to an increase in values in the end-diastolic velocity component of the Doppler waveform. Although both the peak flow velocity and the total velocity integral of the Doppler waveform increased during maternal carbon dioxide inhalation, this increase was not statistically significant. This would suggest that in the term human fetus the decrease in the systolic-to-diastolic ratio we observed is most likely due to a decrease in the downstream resistance, which most likely occurs at the level of the cerebral microvasculature. I2 It is unlikely, however, that the vasodilatory response we observed during hypercapnia is a secondary effect to dilatation of the large cerebral arteries at the base of the brain of these healthy fetuses. If large cerebral arteries had significantly dilated, cerebral blood flow should have increased. In turn, such increase in cerebral blood flow should have increased the total velocity integral (i.e., the area under the curve) in the middle cerebral artery." We could not document any changes in either the time velocity integral or the peak flow velocity within this vascular bed during our two periods of observation. For now, whether the increase in diastolic velocity of the middle cerebral artery is associated with a corresponding increase in cerebral blood flow remains speculative. The current results suggest, however, that in term, healthy human fetuses the middle cerebral artery vasodilates during maternal hypercapnia. This study demonstrates a differential response to hypercapnia among these three vascular beds. It also shows that the higher the maternal concentration of carbon dioxide, the higher the vasodilatory effect of the middle cerebral artery, as found in healthy adults. 12 No significant change in the Doppler velocity waveforms during maternal inhalation of a 3% carbon dioxide gas mixture could be found in the other two vascular beds. The peak systolic-to-end-diastolic velocities ratio in the umbilical artery circulation during maternal hypercapnia did not significantly change. This would suggest that this particular bed does not readily respond to such increases in maternal carbon dioxide. This insensitivity or low sensitivity may reflect the fact that the umbilical circulation is a passive circulation. 14 Similar observations were also found in the ewe during 3% to 10% carbon dioxide administration. In this experimental model no change in the umbilical vascular resistance occurred. Even when the fetal arterial Pco 2 rose to 80 mm Hg, only a trivial but nonsignificant increase in umbilical vascular resistance could be documented. Is The baseline values for the ratio between the peak systolic-to-end-diastolic velocities of the fetal renal artery were elevated, as we previously found in a group of term normal fetuses. 5 This high baseline value for the pulsed Doppler velocity waveforms reflects high resistance of this particular vascular bed. 16 The ratio of peak
Volume 167 Number 6
systolic to end-diastolic velocities and the peak flow velocity and time velocity integral in the fetal renal artery did not change significantly during maternal inhalation of 3% carbon dioxide gas mixture. Like the umbilical vascular bed, the fetal renal artery does not seem to be as sensitive as the middle cerebral artery to maternal hypercapnia. Fetal renal vascular resistance has been reported to increase during progressive hypercapnia in a semichronic lamb preparation during administration of a 20% carbon dioxide gas mixture for 2 to 3 minutes to the ewe. I7 In this experimental model the rise in the resistance of the fetal renal artery was proportional to the rise in the fetal Pac0 2 • The higher the Pac0 2 , the higher the fetal renal artery vascular resistance. Fetuses in which Paco 2 did not rise significantly did not show this increase in fetal renal artery resistance. In the current study neither the peak systolic-to-end-diastolic velocities ratio, the peak flow velocity, nor the time velocity integral of the pulsed Doppler velocity waveform of the human fetal renal artery changed significantly during maternal hypercapnia. This suggests that this vascular bed in the term fetus is not significantly affected by this particular level of maternal hypercapnia. Whether a higher concentration of carbon dioxide gas mixture could have caused a change in the response of this vascular bed remains to be studied. This study shows that maternal inhalation of 3% carbon dioxide selectively affects vascular beds in the normal term human fetus. Although the umbilical and renal vascular beds did not have a change in peak systolic to end-diastolic velocities, vasodilatation occured in the middle cerebral vascular bed as evidenced by an increase in the diastolic velocity of the Doppler waveform. This vascular bed appears to be much more sensitive to maternal hypercapnia. This finding would support our hypothesis that this particular cerebral vascular bed responds differently to maternal hypercapnia than other vascular beds. It would also support the concept of the uniqueness of the response of this vascular bed to hypercapnia, as previously described in experimental models. It would also suggest that the response hypercapnia causes in adults and in other fetal experimental animals is operational in term human fetuses. I3 Whether this uniqueness is found in preterm fetuses or in fetuses with growth abnormality remains to be elucidated.
Carbon dioxide and fetal circulation
1671
REFERENCES
1. Dawes GS. Foetal blood gas tensions and pH. In: Foetal and neonatal physiology. Chicago: Year Book, 1968: 10616. 2. Seldon RE, Peeters LLH, Jones MD ]r, Makowski EL, Meschia G. Redistribution of cardiac output and oxygen delivery in the hypoxemic fetal Iamb. AM] OBSTET GYNECOL 1979;135:1071-8. 3. Cohn HE, Sacks E], Heymann MA, Rudolph AM. Cardiovascular responses to hypoxemia and acidemia in fetal lambs. AM] OBSTET GYNECOL 1974;120:817-24. 4. Ritchie ]WK, Lakhani K. Fetal breathing movements in response to maternal inhalation of 5% carbon dioxide. AM J OBSTET GYNECOL 1980;386-9. 5. Veille JC, Kanaan C. Duplex Doppler ultrasonographic evaluation of the fetal renal artery in normal and abnormal fetuses. AM.J OBSTET GYNECOL 1989; 161: 1502-7. 6. Heistad DD, Kontos HA. Cerebral circulation. In: Volume 3: handbook of physiology. Section 2: the cardiovascular system. Part I: peripheral circulation and organ blood flow, Bethesda, Maryland: American Physiological Society, 1983: 137-82. 7. Lassen NA. Brain extracellular pH: the main factor controlling cerebral blood flow. Scand J Clin Lab Invest 1968; 22:247-51. 8. Reivich M. Arterial Pco 2 and cerebral hemodynamics. Am J Physiol 1964;206:25-35. 9. Rosenberg AA, Jones MD Jr, Traystman RJ, Simmons MA, Molteni RA. Response of cerebral blood flow to changes in pC02 in fetal, newborn, and adult sheep. Am J Physiol 1982;242:H862-6. 10. Himvich HE, Fazekas JF. Comparative studies of the metabolism of the brain of infant and adult dogs. Am ] PhysioI1941;132:454-9. 11. Archer LJ, Evans DH, Paton JY, Levene MI. Controlled hypercapnia and neonatal cerebral artery Doppler ultrasound waveforms. Pediatr Res 1986;20:218-21. 12. Ogawa S, Handa N, Matsumoto M, et al. Carbon dioxide reactivity of the blood flow in human basilar artery estimated by the transcranial Doppler method in normal men: a comparison with that ofthe middle cerebral artery. Ultrasound Med Bioi 1988;14:479-83. 13. Abboub FM. Special characteristics of the cerebral circulation. Fed Proc 1981 ;40:2296-2300. 14. Walker AM, Oakes G, McLaughlin M, Ehrenkranz R, Chez RA. Effects of hypercapnia on uterine and umbilical circulations in conscious pregnant sheep. J AppJ Physiol 1976;41 :727-33. 15. Dawes GS. The umbilical circulation. In: Foetal and neonatal physiology. Chicago: Year Book, 1968:66-78. 16. Vaughn D, Kirschbaum TH, Bersentes T, Dilts PV Jr, Assali NS. Fetal and neonatal response to acid loading in the sheep.] Appl Physiol 1968;24:135-41. 17. Beguin F, Dunnihoo DR, Quilligan EJ. Effect of carbon dioxide elevation on renal blood flow in the fetal lamb in utero. AM J OBSTET GYNECOL 1974;119:630-7.
