The physiological basis of a characteristically low blood flow to the fetal lungs is incompletely understood. To determine the potential role of pulmonary vascular interaction in this phenomenon, simultaneous wave intensity analysis (WIA) was performed in the pulmonary trunk (PT) and left pulmonary artery (LPA) of 10 anesthetized late-gestation fetal sheep instrumented with PT and LPA micromanometer catheters to measure pressure (P) and transit-time flow probes to obtain blood velocity (U). Studies were performed at rest and during brief complete occlusion of the ductus arteriosus to augment pulmonary vasoconstriction (n = 4) or main pulmonary artery to abolish wave transmission from the lungs (n = 3). Wave intensity (dIW) was calculated as the product of the P and U rates of change. Forward and backward components of dIW were determined after calculation of wave speed. PT and LPA WIA displayed an early systolic forward compression wave (FCWis) increasing P and U, and a late systolic forward expansion wave decreasing P and U. However, a marked midsystolic fall in LPA U to near-zero was related to an extremely prominent midsystolic backward compression wave (BCWms) that arose ∼5 cm distal to the LPA, was threefold larger than the PT BCWms (P < 0.001), of similar size to FCWis at rest (P > 0.6), larger than FCWis following ductal occlusion (P < 0.05) and abolished after main pulmonary artery occlusion. These findings suggest that the absence of pulmonary arterial midsystolic forward flow which accompanies a low fetal lung blood flow is due to a BCWms generated in part by cyclical vasoconstriction within the pulmonary microcirculation.
- fetal pulmonary vascular interaction
- fetal pulmonary blood flow
- fetal pulmonary blood pressure
in the fetus, right ventricular (RV) output constitutes 56–67% of the combined ventricular output, while proximal pulmonary blood pressures are equal to or greater than in the aorta (1, 16, 35, 38, 40–42). Despite the high pulmonary pressures, only ∼10% of RV blood entering the pulmonary trunk (PT) passes to the fetal lungs, with the remainder crossing the ductus arteriosus into the descending aorta (12, 35, 36, 42). The physiological basis of this low blood flow to the fetal lungs is incompletely understood, although factors such as a low O2 milieu and pronounced muscularity and reactivity to vasoconstrictors of the pulmonary vasculature have been implicated (5, 12, 36). However, the strikingly different blood flow profiles within the fetal PT, which displays continuous forward flow throughout systole (13, 36–38), and major pulmonary arteries, where forward flow occurs only in early systole (27, 36, 37), suggests that detailed evaluation of hemodynamic interactions between these anatomically proximate sites may provide new insights into the mechanism(s) underpinning a low fetal lung blood flow.
One powerful means of obtaining quantitative and temporal information about specific components contributing to cardiovascular interactions is the relatively new method of wave intensity analysis (WIA), an approach based on the premise that circulatory function is accompanied by the propagation of infinitesimal wavefronts defined by their pressure (P) and velocity (U) effects (4, 31). In the time domain, the product of changes in P and U (“wave intensity”) represents the instantaneous energy carried by the wavefronts (23, 46). Using WIA, we can classify these waves into “forward-running” waves arising from the heart, “backward-running” waves propagating from the vasculature, “compression” waves increasing pressure, and “expansion” waves decreasing pressure (4). Calculation of wave speed enables separation of P and U into forward and backward components and of net wave intensity into the four wave types which may simultaneously exist in an overall profile, namely, “forward compression waves” increasing pressure and velocity, “forward expansion waves” decreasing pressure and velocity, “backward compression waves” increasing pressure but decreasing velocity, and “backward expansion waves” decreasing pressure but increasing velocity (4, 23).
To date, only one study has applied WIA in the fetus, with evaluation primarily of the RV-PT interaction (13). As in the adult (17, 18), the fetal PT WIA was characterized by an initial systolic forward compression wave (FCWis) associated with impulsive RV ejection of blood, and a late-systolic forward expansion wave (FEWls) occurring just prior to pulmonary valve closure (13). However, in contrast to its absence from the adult under normal conditions (17, 18), the fetal PT also displayed a very prominent midsystolic backward compression wave (BCWms) temporally associated with a midsystolic plateau in the flow profile (13). On the basis of its abolition by ligation of the main pulmonary artery and the calculated distance to the wave origin, it was concluded that this BCWms arose from the pulmonary vasculature as a reflection of FCWis (13). These findings are of particular relevance because they suggest that the abrupt midsystolic cessation of flow observed in fetal major pulmonary arteries (27, 36, 37) is related to the presence of a BCWms, even larger than in the PT. If this is the case, however, it is unlikely that vascular reflection alone could underpin such a pulmonary arterial BCWms, as the relative magnitude of the PT BCWms is already double or more that of typical reflected BCWms seen in the fetal (13) or adult ascending aorta (23, 32), or the adult PT in hypoxia (18). An alternative possibility, suggested by the increased vasoreactivity (12) and potent myogenic responses (3, 44) known to occur within the immature pulmonary vasculature, is that vasoconstriction per se also contributes to the genesis of a pulmonary arterial BCWms.
This study, in which simultaneous PT and left pulmonary artery (LPA) WIA was undertaken in anesthetized fetal lambs, therefore had two main aims. The first was to characterize PT-LPA interaction by comparison of wave intensity profiles at these sites, including their contribution to changes in local blood pressure and flow/velocity. The second was to determine the potential role of pulmonary vasoconstriction in generation of a pulmonary arterial BCWms. Studies were performed under resting conditions in all fetuses and in a subgroup of animals during brief occlusion of either the ductus arteriosus to increase pulmonary vasoconstriction via a pressure-induced rise in vessel stretch (44) or the main pulmonary artery to abolish wave transmission from the pulmonary vasculature.
Experiments were approved by the institutional Animal Ethics Committee and conformed to guidelines of the National Health and Medical Council of Australia.
Ten Border-Leicester cross ewes were anesthetized at a gestation of 137 (2) days [mean (SD), term = 147 days] with intramuscular ketamine 5 mg/kg and xylazine 0.1 mg/kg, followed by 4% isoflurane delivered by mask. Animals were placed in a supine position, and the trachea was intubated. Anesthesia was then maintained with isoflurane (2–3%), nitrous oxide (∼30%), and oxygen-enriched air (∼70%) delivered via volume-controlled ventilator (900C Servo, Siemens-Elema, Solna, Sweden), supplemented by an intravenous infusion of ketamine (1–1.5 mg·kg−1·h−1) and midazolam (0.1–0.15 mg·kg−1·h−1). Oxygen saturation was monitored continuously with a cutaneous pulse-oximetry sensor (Oximas Dura-Y, Tyco Healthcare, Pleasanton, CA) applied to the ear. The right common carotid artery was cannulated through a neck incision for monitoring of blood pressure (90308 Multiparameter Monitor, Spacelabs. Medical, Redmond, WA) and for blood gas sampling. On the basis of frequent arterial blood gas analysis (ABL 620, Radiometer, Copenhagen, Denmark), ventilation of the ewe was adjusted to maintain arterial O2 tension at 100–120 mmHg and arterial CO2 tension at 35–40 mmHg.
The pregnant horn of the uterus was exposed through a midline laparotomy, and the fetal head, left forelimb, and upper thorax were exteriorized through a hysterotomy. A multilumen cannula was inserted via the fetal left external jugular vein into the right atrium for fluid administration, and a polyvinyl catheter was passed via the left common carotid artery into the ascending aorta for pressure measurement. A thoracotomy was performed in the 3rd left interspace, and the 3rd and 4th ribs were removed to increase exposure of the heart and great vessels. After incision of the overlying pericardium, 10–14 mm “A series” and 4–6 mm “S series” transit-time flow probes (Transonic Systems, Ithaca, NY) were placed around the PT and LPA, respectively. A cannula with its tip directed toward the heart was inserted into the pulmonary trunk through an adventitial purse-string suture distal to the flow probe on this vessel and connected to a polyvinyl catheter to measure pressure. Through separate purse string sutures, one 2.5 F micromanometer catheter (Millar Instruments, Houston, TX) was inserted into the pulmonary trunk just distal to the flow probe, while another 2.5 F micromanometer catheter was inserted into the pulmonary trunk close to the base of the main pulmonary artery and its tip advanced into the left pulmonary artery. The edges of the pericardial incision were then loosely reapproximated over the left pulmonary artery flow probe.
Aortic and pulmonary trunk blood pressures were measured via the fluid-filled catheters with a transducer (Transpac IV, Abbott Critical Care Systems, Sligo, Ireland), referenced to atmospheric pressure at the level of the left atrium and calibrated against a water manometer before each experiment. Pressure signals from fluid-filled catheters were processed using a transducer amplifier (Transbridge TBM4M, World Precision Instruments, Sarasota, FL). High-fidelity pulmonary trunk and left pulmonary artery pressures were measured by interfacing micromanometers with transducer control units (TCB-500, Millar Instruments). PT and LPA flows were measured with a transit-time flowmeter (model T206; Transonic Systems). All physiological signals were digitized at a sampling rate of 1,000 Hz using an analog-to-digital convertor (iNet-100B, GW Instruments, Somerville, MA) interfaced with programmable acquisition and analysis software (Spike2, Cambridge Electronic Design, Cambridge, UK). No data filtering was employed, apart from application of a 48-Hz low-pass filter at the time of analysis to remove electrical interference from signals. PT and LPA micromanometer signals were calibrated by matching mean values to the mean pressure of the fluid-filled pulmonary trunk catheter.
Hemodynamic variables were allowed to stabilize for 10–15 min after completion of surgery. After withdrawal of blood samples for blood gas analysis, hemodynamic data were recorded onto a computer in all fetuses. In fetal subgroups, data were also recorded before, during, and after a brief (15–20 s) complete occlusion of the main pulmonary artery (n = 3) or ductus arteriosus (n = 4) with a vascular clamp. A brief period of ductal occlusion was specifically chosen because it results in pulmonary vasoconstriction not masked by shear stress-induced vasodilation occurring with longer (>1 min) periods of occlusion (44). Hemodynamic variables rapidly attained a new steady-state level after clamping of either vessel and quickly returned to baseline following removal of the clamp (Supplemental data for this article are available online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology Web site.). At the end of the experiment, animals were killed with a pentobarbital sodium overdose (100 mg/kg), and the position of catheters was checked.
Wave intensity analysis.
As WIA uses pressure and velocity data, PT and LPA blood flows were converted to velocity (U) using cross-sectional area derived from the nominal size of the flow probe (17, 18, 32, 45). After generation of an ensemble average of high-fidelity pressure (P) and U signals from a mean of 66 beats at baseline (range 42–100), 35 beats during occlusion of the ductus arteriosus (range 20–49), and 37 beats during occlusion of the main pulmonary artery (range 25–50), the rates of change of PT and LPA blood pressure (dP/dt) and velocity (dU/dt), and the product of these differentials (wave intensity, dIW), were derived. Note this calculation yielded a “time-corrected” dIW (i.e., dP/dt·dU/dt) that is independent of the digitizing sample rate (8, 21, 32, 34), which contrasts with dIW defined by absolute changes in P and U between samples (i.e., dP·dU) used in a number of previous reports (13, 17, 18, 23, 25, 31, 45, 46). However, the latter can be obtained from time-corrected wave dIW by dividing by the square of the sampling frequency (25).
To separate P, U, and dIW into forward and backward components, wave speed was obtained by derivation of dP/dU and use of the relation ρc = dP/dU (23), where ρ is blood density (assumed as 1,050 kg/m3) and c is wave speed. Using ensemble-averaged P and U data, dP/dU was calculated using least squares linear regression from the P-U slope during early systole, when the contribution of backward-running waves is minimal (23, 24, 32). Hardware-related time lags between P and U data points (18) were corrected by aligning the peak second derivatives of these signals, resulting in a highly linear early-systolic P-U relation for both the PT (R2 = 0.9998 ± 0.0001) and LPA (R2 = 0.9996 ± 0.0004).
As per convention (4), wave direction was referenced to the direction of blood flow, such that waves arising from the right ventricle were defined as forward-running and those propagating from the vasculature as backward-running. Using established methodology (8, 17, 18, 32), the intensity of forward-running waves (dIW+) was calculated as (dP/dt + ρc·dU/dt)2/(4ρc) and that of backward-running waves (dIW−) as −(dP/dt − ρc·dU/dt)2/(4ρc). Waves causing a pressure increase were classified as compression waves and those producing a pressure decrease as expansion waves, with this characteristic defined by the sign of the pressure difference across the respective forward-running wavefront, given by (dP/dt)+ = 1/2 (dP/dt + ρc·dU/dt) and the backward-running wavefront, given by (dP/dt)− = 1/2 (dP/dt − ρc·dU/dt). Thus, a forward-running wave was a compression wave if (dP/dt)+ > 0 and an expansion wave if (dP/dt)+ < 0. Similarly, a backward-running wave was classified as a compression wave if (dP/dt)− > 0 and an expansion wave if (dP/dt)− < 0 (17, 18, 32).
The time interval between wave intensity peaks was obtained from separated WIA profiles and an overall distance to the origin of backward-running waves estimated from the product of wave speed and one-half the time interval between the peaks of the backward and preceding forward compression wave (18, 32). Note that this distance is an approximation, since only a single wave speed (obtained from the PT or LPA) was used in calculations, whereas available information suggests that wave speed varies within different pulmonary arterial segments and within each beat (2).
The cumulative intensity of forward-running (IW+) and backward-running waves (IW−), which is directly related to wave energy, was calculated by integrating the respective dIW over the wave duration (8, 32). In addition, because forward-running compression, forward-running expansion, and backward-running compression wave profiles had major and minor components, total IW for these waves was also calculated. The contribution of waves to P and U was obtained by measuring changes in the forward or backward components of P and U between wave onset and offset.
Statistical analyses were performed using Statistical Package for the Social Sciences ver. 12.0.1 (SPSS, Chicago, IL). Differences in PT and LPA WIA and the effects of vascular occlusion on wave profiles were evaluated using repeated-measures ANOVA. Results are expressed as means (SD), and significance was taken at the P < 0.05 level.
Resting fetal blood gases and hemodynamics.
Ascending aortic pH was 7.282 (0.030), Hb 12.8 (1.3) g/dl, Hb O2 saturation 71 (7) %, Po2 24.7 (2.4) mmHg, Pco2 52.3 (3.6) mmHg, base excess −2.9 (1.8) mmol/l, mean ascending aortic pressure 59 (7) mmHg, mean PT pressure 60 (6) mmHg, and heart rate 142 (19) beats/min.
Resting pressure, velocity and wave intensity profiles.
Systolic blood pressure profiles in the PT and LPA were similar, except that a shoulder in the midportion of the ascending limb was more pronounced in the former (Fig. 1A). As in previous reports, PT (13, 38) and LPA (27, 36, 37) flow profiles displayed an early systolic peak and a late-systolic negative flow, but a midsystolic flow attenuation, occurring in conjunction with rising pressure, was more marked in the LPA (Fig. 1B). Peak positive, mean, and peak negative flows were all greater in the PT. However, peak positive U was similar in the PT and LPA, suggesting that the difference in peak positive flow was largely attributable to differing vessel diameters, while peak negative U was greater in the LPA and mean U was approximately threefold greater in the PT (Fig. 1C, Table 1).
PT WIA displayed an early systolic FCW (FCWis), a large midsystolic BCW (BCWms), and a late-systolic FEW (FEWls). In addition, a small and constant BEW was temporally associated with the shoulder of the ascending limb of the PT pressure pulse, while several smaller midsystolic FCWs (FCWms) evident both before and after the peak of BCWms were associated with further increases in pressure and rebound increases in velocity (Fig. 2, left). WIA in the LPA also demonstrated a FCWis, and FEWls, smaller FCWms, and an inconsistent BEW, present in only 3 of the 10 fetuses. However, the most striking feature at this site was an extremely prominent BCWms of similar size to the preceding FCWis (Fig. 2, right).
Resting wave intensity analysis.
Peak FCWis dIW+ in the PT occurred 6 (3) ms earlier (P < 0.001) and was ∼30% greater (P = 0.03) than in the LPA. FCWis comprised ∼90% of total FCW IW+ and was associated with a similar increase in U at both sites, but ∼25% higher rise in P in the PT (P < 0.001).
Peak FEWls dIW+ in the PT occurred 6 (3) ms earlier (P < 0.001) and was ∼50% greater (P = 0.03) than the LPA. FEWls comprised ∼85% of total FEW IW+ and produced a similar reduction in U at both sites but was associated with a ∼35% greater fall in P in the PT (P < 0.001).
In contrast to the forward-running waves, BCWms peak dIW− in the LPA occurred 4 (4) ms earlier (P = 0.02) and was 3-fold larger (P < 0.001) than in the PT. Furthermore, whereas the magnitude of BCWms IW− was 24 (19) % of FCWis IW+ in the PT, it comprised 95 (33) % of FCWis IW+ in the LPA (P < 0.001), with no significant difference between the amplitudes of LPA BCWms IW− and FCWis IW+ (P > 0.6). Compared with the PT, the LPA BCWms produced a ∼50% greater rise in P (P < 0.001) and double the fall in U (P < 0.001). Using the interval between BCWms peak dIW− and peak FCWis dIW+ in the PT [51 (8) ms] and LPA [41 (12) ms], the calculated origin of BCWms from the measurement site in the PT [8.9 (2.3) cm] was almost double that from the LPA [5.3 (1.7) cm, P = 0.001].
In the PT, peak BEW dIW− occurred 11 (2) ms after peak FCWis dIW+, constituted 3% of the magnitude of peak FCWis dIW+, and had a calculated origin 2.9 (1.9) cm distal to the measurement site.
Effect of transient vascular occlusion.
Ductal occlusion produced similar morphological alterations of wave intensity profiles in the PT (Fig. 3) and LPA (Fig. 4), with the main changes comprising a decrease in the magnitude of FCWis and an increase in the amplitude of FCWms and BCWms. The magnitude of peak BCWms dIW− increased after ductal occlusion in both the PT [from 0.38 (0.18) to 1.28 (0.29) W·m−2s−2 × 106, P < 0.01] and LPA [from 1.95 (0.56) to 3.23 (0.79) W·m−2s−2 × 106, P < 0.05]. Importantly, the amplitude of BCWms IW− was uniformly greater than FCWis IW+ following ductal occlusion, with the BCWms/FCWis ratio increasing from 0.25 (0.19) to 1.25 (0.34) in the PT (P = 0.01) and from 0.92 (0.16) to 1.86 (0.58) in the LPA (P < 0.05).
In the PT, occlusion of the main pulmonary artery was accompanied by disappearance of not only BCWms but also of most of the FCWms occurring after this BCWms, with loss of BEW (Fig. 5). As little or no pulsatile P and U was present, no significant LPA wave intensity profiles were detected after occlusion of the main pulmonary artery (Supplemental data for this article are available online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology Web site.).
Using the novel approach of simultaneous WIA in the fetal PT and LPA, this study has provided new insights into the hemodynamic interaction between these sites, and, in particular, the basis of their markedly different blood flow profiles and the associated low level of blood flow to the fetal lungs.
The most striking finding in our study was an extremely prominent pulmonary arterial midsystolic backward-running compression wave (BCWms), similar in amplitude to the preceding early-systolic forward compression wave (FCWis) under baseline conditions. As evident by its temporal features (Fig. 2), amplitude (Table 2), and effect on U (Table 3), this very large BCWms was responsible for the characteristic abrupt decline in flow to near-zero occurring after a brief initial systolic period of forward flow in major fetal pulmonary arteries (27, 36, 37). This BCWms also made a major contribution to the local systolic pressure profile, producing a rise that was equivalent in magnitude to that of FCWis (Table 3).
In accord with a previous report (13), a prominent BCWms was also present within the fetal PT WIA (Fig. 2) and was responsible for the midsystolic plateau in the PT blood flow/velocity profile. However, the magnitude of this BCWms was only ∼30% that of the LPA BCWms (Table 2), with associated smaller quantitative effects on P and U (Table 3).
Taken together, several results indicate that the LPA BCWms arose from within the lungs and gave rise to the PT BCWms. Thus, the peak of the pulmonary arterial BCWms preceded the PT BCWms peak, while its amplitude exceeded that of the PT BCWms, even when both were enhanced after occlusion of the ductus arteriosus. In accord with the finding obtained from a single fetus (13), occlusion of the main pulmonary artery also abolished the PT BCWms. Finally, the distance of 8.9 cm from the PT to the origin of BCWms, which is similar to a value of 9.4 cm reported previously (13), was appropriately larger than the distance of 5.3 cm calculated from the LPA.
It is widely considered that the mechanism underlying a BCWms is a reflection of the preceding FCWis from “closed-end” reflection sites (18, 20, 22, 23, 25, 31). In accord with this view and using PT WIA alone, it was concluded that the large fetal PT BCWms was a reflection of FCWis from the pulmonary vasculature (13). However, the combined use of PT and LPA WIA in the present study indicated that vascular reflection was not the sole mechanism underlying the fetal LPA BCWms, and by implication, the PT BCWms. Specifically, as the magnitude of the LPA BCWms and its preceding FCWis were similar at rest (Table 2), reflection could only have produced this BCWms if such reflection was near complete, a phenomenon unknown in physiological systems. Even after complete occlusion of the thoracic aorta, for example, the magnitude of BCWms still only increases to 25–40% of FCWis (23, 34). Moreover, the sizes of the LPA and PT BCWms were markedly increased after occlusion of the ductus arteriosus, with both then exceeding the amplitude of FCWis (Figs. 3 and 4). As vascular reflection can return but not itself produce energy, this clearly indicated that a significant mechanism other than reflection was involved in generation of both the fetal LPA and PT BCWms.
As the fetal pulmonary arterial BCWms was equal to or greater than FCWis, a plausible mechanism for this BCWms was that it was in part generated by an impulsive compression wave arising from the pulmonary vasculature during each cardiac cycle. The most likely source of this impulsive compression wave was, in turn, a cyclical midsystolic vasoconstriction that, on the basis of our calculated distance to the origin of the left pulmonary arterial BCWms (5.3 cm), arose from the pulmonary microcirculation. Indeed, strong indirect support for a significant myogenic origin of the pulmonary arterial BCWms is provided by the striking transformation of the fetal pulmonary arterial flow pattern to one resembling the PT profile following infusion of the vasodilator ACh (27, 37). Furthermore, the occurrence of BCWms after FCWis implies that this cyclical pulmonary vasoconstriction was triggered by the preceding FCWis.
In our study, the peak of the pulmonary arterial BCWms occurred ∼40 ms after peak FCWis. Although the specific myogenic pathway(s) and mediator(s) of this BCWms remain to be defined, this temporal feature suggests that the mechanism is distinct from that underlying the classical myogenic vasoconstrictor response occurring in response to increases in intravascular pressure, which is believed to play a role in the autoregulation of blood flow and setting of basal vascular tone (10). Specifically, the latter type of myogenic constriction typically occurs in 5–10 s (7, 26), although an interval of 300 ms has been measured using high-speed video analysis (28).
While abolition of the PT BCWms by occlusion of the main pulmonary artery (Fig. 5) confirmed that this wave resulted from proximal transmission of the pulmonary arterial BCWms, the large reduction (∼65%) in the magnitude of BCWms between the LPA and PT indicated that such transmission was only partial. As dissipation of wave energy (11) was probably quite minor in the short distance between the LPA and PT, it is likely that remaining portions of the LPA BCWms energy entered the right pulmonary artery and ductus arteriosus, and possibly also reflected back down the LPA and its branches.
FCWis is the manifestation of a ventricular impulse generated at the beginning of systole and provides the forward momentum for blood movement from the ventricle into the vasculature (45). The lesser magnitude and later occurrence of FCWis in LPA, compared with PT (Table 2), are both consistent with an RV origin for FCWis.
As well as FCWis, however, smaller FCWms were also detectable in the fetal PT and LPA at baseline. Although not referred to, similar FCWms are clearly evident in published figures of the fetal PT WIA profile (13). As is apparent in Fig. 2, FCWms produces a rebound increase in blood flow/velocity following the characteristic abrupt midsystolic decline related to BCWms. Such a rebound is also present in published recordings of the fetal main and left pulmonary arterial blood flow profiles (27, 36, 37). At least two mechanisms probably contributed to FCWms. FCWms in the vicinity of the FCWis tail region (Fig. 2) were most likely related to additional RV impulsive contractions occurring after FCWis, perhaps reflecting incoordination in the RV systolic contraction pattern secondary to the known structural immaturity of the fetal myocardium (43). However, the observations that FCWms peaks occurring after BCWms became more prominent after ductal occlusion, which also augmented BCWms (Figs. 3 and 4) but were diminished by occlusion of the main pulmonary artery (Fig. 5) raises the possibility that these FCWms were related to proximal reflection of this BCWms and/or transmission of BCWms from the opposite pulmonary artery.
Recent data indicate that FEWls is the vascular manifestation of a ventricular rarefaction (“suction”) wave (45), and as is apparent from Fig. 2, this wave causes flow reversal at the end of systole in the fetal PT and major pulmonary arteries. The larger magnitude and earlier occurrence of FEWls in the PT compared with LPA are consistent with an RV origin for this wave. The explanation for peak negative velocity in the LPA exceeding that in the PT (Table 1), despite FEWls causing a similar decrease in velocity at both sites (Table 3), is that velocity was lower in the LPA just before onset of FEWls due to the much larger BCWms.
Although smaller than observed in the adult (17, 18), a BEW was present in all fetal PT WIA (Fig. 2). In contrast, a BEW was an inconsistent feature in the pulmonary arterial WIA, possibly because it was masked by the very large pulmonary arterial BCWms, which often commenced where a BEW was apparent in the PT WIA. As BEW produced a transient reduction in blood pressure, this wave was responsible for the prominent shoulder in the ascending limb of the PT pressure waveform (Figs. 1 and 2). The disappearance of the PT BEW during occlusion of the main pulmonary artery (Fig. 5) suggests that, as in the adult (17, 18), this wave may have arisen from the presence of an “open-end” reflection site related to an increase in cross-sectional area at downstream branching points.
Three methodological issues require comment. First, separation of net wave intensity into forward and backward components was an essential part of wave intensity analysis in our study, due to the extensive temporal overlapping of forward- and backward-running waves evident in both the fetal PT and LPA (Figs. 2–4). Second, wave speed in the fetal PT (3.6 m/s) was higher than the reported value of 2.6 m/s (13), presumably related to differences in anesthetic regimen and surgical approach. At first glance, a fall in wave speed between the PT and LPA (Table 2) may seem surprising, as other data suggest that wave speed increases from the central to peripheral vasculature (2, 24). However, because wave speed is inversely related to the square root of vessel distensibility (29), our finding could be explained if, as in the adult (15), distensibility increased between the PT and LPA. Finally, one potential limitation of our study was that it was performed under general anesthesia and open-chest conditions, an approach necessary because of the extent of surgical instrumentation required. However, blood gas data were comparable to those of unanesthetized fetuses, while the average level of mean arterial blood pressure in our study (59–60 mmHg) was at the upper end of the normal range reported in chronically instrumented late-gestation fetal lambs (1, 6, 9, 19, 27, 30, 33, 39, 40, 42). In addition, LPA flow (70 ml/min) was close to the range of 71–87 ml/min measured with transit-time flow probes in unanesthetized and anesthetized late-gestation fetuses (14, 19, 33). It is thus unlikely that the qualitative features of our findings were affected by our experimental approach.
Perspectives and Significance
Our observation that the characteristically low lung blood flow of the fetus is accompanied by a very prominent and uniquely large pulmonary arterial BCWms that markedly attenuates midsystolic blood flow/velocity and substantially elevates local blood pressure has major potential implications for both pulmonary physiology and pathophysiology in the perinatal period. As the striking midsystolic decline in fetal pulmonary arterial blood flow to near-zero is not observed postnatally (27, 37), it is tempting to speculate that a marked diminution in the magnitude of the fetal pulmonary arterial BCWms accompanies the fall in pulmonary blood pressures and dramatic rise in pulmonary blood flow associated with lung ventilation and birth (5, 12, 36). On the other hand, continued presence of a large pulmonary arterial BCWms might be one factor contributing to the increased pulmonary blood pressures and reduced lung perfusion accompanying conditions such as persistent pulmonary hypertension of the newborn (12).
This work was supported by the Australia and New Zealand Children's Heart Research Centre.
We thank Magdy Sourial, Dr. Kate Simpson, and Andrew Hattam for their assistance with experimental studies.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society