INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

PULMONARY VASOREACTIVITY has long been known to vary greatly from one species to another and, within a species, from one individual to another (13). However, less attention has been paid to baseline interspecies or interindividual differences in pulmonary vascular resistance (PVR). Eldridge et al. (8) recently showed in humans that subjects with the strongest hypoxic pulmonary vasoconstriction also had an increased baseline PVR. We reported that minipigs compared with dogs not only have a higher PVR [as assessed by multipoint mean pulmonary arterial pressure (Ppa)-pulmonary blood flow () plots] in hyperoxia, and more so in hypoxia, but also react to hypoxia by an increase in characteristic impedance (Zc) and that these species differences persist after inhibition of hypoxic vasoconstriction by the inhalation of nitric oxide (17). These findings were explained by structural differences not only in flow-resistive properties of peripheral pulmonary arterioles but also in elastic properties of more proximal pulmonary arteries (17).

Because differences in pulmonary vascular reactivity appear to be associated with structural differences, we thought it of interest to compare PVR (evaluated by multipoint Ppa vs. plots) and pulmonary vascular impedance (PVZ) spectra in minipigs and in dogs after induction of embolic pulmonary hypertension. Both Ppa vs. plots and PVZ spectra have been previously reported in dogs with acute embolic pulmonary hypertension (5-7, 9-11, 19, 23), but no such studies have been reported in pigs. Because PVZ varies with body size (18) and (17), we compared minipigs and dogs of the same weight and at the same level of , which was controlled by a manipulation of venous return. We hypothesized that, at the same level of Ppa or , minipigs would present with a higher pulmonary arterial elastance and wave reflection, leading to an increased pulsatile opposition to pulmonary arterial .

	METHODS

Top Abstract Introduction Methods Results Discussion References

Preparation. Six mongrel dogs (21-36 kg) and six weight-matched minipigs were included in the present study, which adhered to the "Guide for the Care and Use of Laboratory Animals" [DHEW Publ. No. (NIH) 86-21, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20892] approved by the American Physiological Society. The minipigs were premedicated with intramuscular ketamine sodium (20 mg/kg), midazolam (0.1 mg/kg), and 0.25 mg atropine. In the dogs, anesthesia was induced with propofol (10 mg/kg) and atropine (0.25 mg iv). Thereafter, all the animals were anesthetized with sufentanyl (2-3 µg · kg¹ · h¹) and midazolam (0.1 mg · kg¹ · h¹) and paralyzed with pancuronium bromide (0.2 mg · kg¹ · h¹). They were ventilated by using an Elema 900 B ventilator (Siemens, Solna, Sweden) via a cuffed endotracheal tube, the inspired O₂ fraction being 0.4, adjusted up to 0.6 to maintain arterial PO₂ between 100 and 200 Torr, the respiratory rate of 10 strokes/min, and the tidal volume of 15-20 ml/kg, adjusted to maintain arterial PCO₂ between 35 and 45 Torr. Any metabolic acidosis was corrected by slow intravenous administration of sodium bicarbonate. Throughout the experiment, normal saline was infused at a rate of 10 ml · kg¹ · h¹. The temperature was maintained at 37-38 °C by means of a heating blanket. A standard lead electrocardiogram was used for the monitoring of heart rate (HR).

Measurements. Ppao and Psa were measured by using disposable pressure transducers (Gould-Spectramed, Binchoven, The Netherlands). The vascular pressure and flow signals were displayed by using a monitoring system (Sirecust 404, Siemens, Erlangen, Germany) and recorded on a 6-channel Gould recorder (model 2600S, Gould, Instruments Division, Cleveland, OH). The pressure transducers were zero referenced at midchest. The zero flow was adjusted to the end-diastolic value, assumed to be zero. All pressures and flows were measured at end expiration. The system phase shift between pressure and flow was found to be negligible by cross correlation, and thus no correction factor was applied. Arterial and mixed venous blood gases were measured immediately after the samples were drawn by an automated analyzer (ABL 2, Radiometer, Copenhagen, Denmark) and corrected for temperature.

Protocol. As soon as the animals were in a stable state as estimated by stable HR, Psa, Ppa, and , baseline hemodynamics were assessed. Thereafter, by stepwise inflations of the inferior vena cava balloon, was adjusted first between 2 and 2.5 l/min, then ~1.5 l/min. At each , Psa and HR were recorded, and Ppa and flow signals were sampled for PVZ calculations. Arterial and mixed venous blood gases were measured at the highest . The same procedure was repeated 30 min after injection of autologous blood clots. For this purpose, a 250-ml blood sample, collected before thoracotomy, was allowed to clot in a beaker and was cut into 3- to 5-mm pieces for injection (6). A large-bore polyethylene cannula (ID 3 mm) was inserted into the left external jugular vein, and blood-clot pieces were injected by an irrigation syringe over 30 min. Embolization was carried out progressively until, in a first step, Ppa reached 40-45 mmHg and, in a second step, reached 50-55 mmHg. At the low- as well as at the high-Ppa level the animals were allowed to stabilize for 30 min without changing the inspired O₂ fraction. The Ppa stabilized between 20 and 25 min at a level that was, in general, 3-5 mmHg below the value achieved at the end of embolization. Thereafter, at each level of Ppa, arterial as well as mixed venous blood gases were measured at the highest flow, and hemodynamics were recorded at the highest flow as well as at each step down to a flow rate of ~1.5 l/min.

Statistical analysis. Results are expressed as means ± SE. A linear regression analysis was performed on each of the three-point (Ppa  Ppao)/ plots. To obtain composite (Ppa  Ppao)/ plots for each experimental situation, (Ppa  Ppao) values interpolated from individual regression analysis were averaged at 1.0-l/min intervals of from 1.5 to 3.5 l/min. The blood-gas and hemodynamic data were analyzed by a two-way repeated-measures analysis of variance. When the F-ratio of the analysis of variance reached a P < 0.05 level, comparisons between baseline and embolism measurements were made by using the Scheffé test, and pairwise comparisons between weight-matched dog and pig measurements were made by using the U-test, with P < 0.05 considered as statistically significant (29).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Baseline hemodynamics. Before embolization, HR, Psa, Ppao, , arterial pH (pH_a), and blood gases were not different in dogs and in weight-matched minipigs (Table 1). However, the minipigs had a higher (Ppa  Ppao) gradient at all levels of studied (Fig. 1), with higher slopes but not pressure intercepts of (Ppa  Ppao)/ plots (Table 1). At identical (2.2 l/min), the minipigs had a higher Z₀ and Z₁, no different Z_c, a more negative phase angle, and a first minimum frequency ( f_min) of the PVZ spectrum shifted toward a higher frequency, and a higher Ws and a lower Wosc-to-WT ratio (Wosc/WT) (Table 2). The ETR (Table 2) and global (i.e., Rc) (Table 2) and time-domain wave-reflection indexes (Table 3) were not different.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Composite plots of mean pulmonary arterial pressure (Ppa)  pulmonary arterial occluded pressure (Ppao) vs. cardiac output at baseline (solid line), at 1st level of embolization (dashed line), and at 2nd level of embolization (dotted-dashed line) in 6 dogs and 6 weight-matched minipigs. At baseline, minipigs presented with higher Ppa  Ppao at all levels of flow compared with dogs. Pulmonary embolism increased Ppa  Ppao in both animal species at all levels of flow but more in minipigs than in dogs at highest levels of flow. * P < 0.05, (*) P = 0.05, dogs vs. minipigs. P < 0.01, baseline vs. embolism.

Effects of embolism. The injection of autologous blood clots increased Ppa and decreased , decreased arterial PO₂, mixed venous PO₂, and pH_a, and increased arterial PCO₂ in both the dogs and minipigs (Table 1). Embolization shifted (Ppa  Ppao)/ plots to higher pressures (Fig. 1), with increased slopes in both species, but increased intercepts in dogs only (Table 1). Accordingly, at the rates of 2.5 and 3.5 l/min, the (Ppa  Ppao) gradients were higher in minipigs than in dogs (Fig. 1). In both species at a same of 2.0-2.2 l/min, embolization increased Z₀, Z₁, f_min, and Rc, whereas Zc increased in minipigs only (Table 2). The phase angle of the first harmonic became more negative in dogs, whereas it remained unchanged in minipigs. Embolization increased both the Ws and the Wosc components of WT in both species. However, Wosc/WT decreased in dogs and remained unchanged in minipigs. The ETR decreased in both species but was relatively better preserved in minipigs. The pooled PVZ spectra in dogs and minipigs are shown in Fig. 2. At the standardized of 2.0-2.2 l/min, progressive pulmonary embolization increased low-frequency pulmonary impedance and shifted the PVZ spectrum to the right more in minipigs than in dogs. Embolization affected the time-domain wave reflection indexes mainly by an increase in the amplitude of the reflected wave, and a decreased time to peak of the backward wave, with no effect on time to foot of the backward wave, time to positive backward wave, (except a slight decrease in dogs), and time to end of positive backward wave (Table 3). Embolism increased systolic Ppa and pulse Ppa in both species but increased the difference between measured systolic pressure and forward pressure, and between measured pulse pressure and forward pressure in the dogs only (Table 3).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Pulmonary vascular impedance spectra in dogs (n = 6) and in minipigs (n = 6) at baseline (solid line), at 1st level of embolization (dashed line), and at 2nd level of embolization (E2; dotted-dashed line). Pulmonary blood flow was between 2.0 and 2.2 l/min. Minipigs, compared with dogs, had a higher low-frequency impedance and a more negative 1st-harmonic phase angle at baseline. After embolization, minipigs still had a higher low-frequency impedance, but 1st-harmonic phase angle was not any more different. NS, not significant. E2 vs. baseline for impedance moduli and phase <3 Hz.  E2 vs. baseline for impedance moduli <6 Hz. * P < 0.05, dogs vs. minipigs.

Comparison of dogs and pigs after embolism. At the same standardized of 2.2 l/min and at identical Ppa level of 40 mmHg, minipigs presented with higher low- and high-frequency impedance and an f_min shifted toward higher frequencies (Fig. 3). PVZ spectra differences between dogs and minipigs are summarized in Table 4. At identical and Ppa, minipigs presented with higher Z₁, Zc, and ETR; an f_min shifted toward higher frequencies; and a lower Wosc/WT. However, none of the time-domain indexes of wave reflection were different (Table 3), even after correction for systolic time intervals.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Pooled pulmonary vascular impedance spectra in embolic pulmonary hypertension in 6 dogs (solid lines) and in 6 minipigs (dashed lines) at same flow (2.2 l/min) and mean Ppa (40 mmHg). Minipigs presented with higher low-frequency impedance, higher characteristic impedance, and a 1st minimum frequency shifted toward higher frequencies. * P < 0.05, dogs vs. minipigs.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study shows that, compared with dogs, the pulmonary circulation in minipigs is characterized by a higher baseline PVR and PVZ spectrum indexes of a higher pulmonary arterial elastance and wave speed but by no different Zc or indexes of wave reflection. At the same severity of embolic pulmonary hypertension, as assessed at identical levels of Ppa and , the PVZ spectrum in minipigs presents with an increase in frequency-dependent oscillations but still with no different indexes of wave reflection, thus probably explained by a relatively more important increase in pulmonary arterial elastance. As indicated by lower Wosc/WT and higher ETR in minipigs, these differences allow for a better energy transfer from the right ventricle to the hypertensive pulmonary circulation.

Several studies have shown that pigs present with a stronger pulmonary vasoreactivity to hypoxia than many other species, including dogs (13). Tucker et al. (27) exposed seven species to hypobaric hypoxia for several weeks and observed that pulmonary hypertension developed in the following order of decreasing severity: calves and pigs (severe), rats and rabbits (moderate), and sheep, guinea pigs, and dogs (mild). There are data suggesting that species or individuals with a stronger pulmonary vascular reactivity also present with a higher basal PVR. Attinger and Cahill (1) reported that the flow resistance of the pulmonary vasculature in pigs was ~12 times as great as that in the human lung and 2-3 times greater than in the canine lung. Eldridge et al. (8) found a higher baseline PVR in subjects with an enhanced hypoxic pulmonary vasoconstriction. We reported that minipigs, compared with weight-matched dogs, have an increased slope of (Ppa  Ppao)/ plots and an increased pulmonary vasoreactivity to hypoxia (17). In agreement with these previous studies, the slope of (Ppa  Ppao)/ plots was greater in minipigs than in dogs before, as well as after, embolization of the pulmonary circulation.

Species with increased PVR and reactivity have been reported to present an increased content of collagen, elastin, and smooth muscle cells of conductive elastic pulmonary arteries and an increased medial thickness of resistive pulmonary arteries and arterioles (ID 30-1,000 µm) (15, 27). In addition, network factors could play a role. It has been shown in a simple branching network model that the slope of the pressure-flow curve is sensitive to pulmonary arterial reactivity because it is related to both resistance of the main pulmonary artery and to the distribution of cross-sectional area associated with branching (16). No detailed morphological studies of the porcine pulmonary circulation are available in the literature allowing direct comparison of main pulmonary arterial diameter and area ratios of bifurcations between pigs and dogs. In the present study, baseline Zc appeared, on average, not different in dogs and minipigs, confirming our previous observations (17). Zc is determined by the ratio between inertance and compliance of the proximal pulmonary arterial tree (20). Because arterial elastance would be expected to be higher in pigs than in dogs (1, 15), the cross-sectional area of the proximal porcine pulmonary arterial tree could therefore only be greater, implying a lower proximal pulmonary arterial resistance. Increased slope of (Ppa  Ppao)/ plots and increased PVR in minipigs, therefore, more likely reflect an increased flow resistance at the periphery of the pulmonary arterial tree. Except for an increase in the Z₁, frequency oscillations in the impedance spectrum did not appear different at baseline between dogs and minipigs, and indexes of wave reflection were not different as well. These findings suggest that branching should not play a major role in baseline differences between dogs and pigs in slope of pressure-flow relationships, PVR, and reactivity.

Experimental pulmonary embolism in dogs has been reported to shift (Ppa  Ppao)/ plots to higher pressures, with variable increases in slopes and extrapolated pressure intercepts (6, 7, 19). The present findings are in keeping with these previous results, which can be explained by an increased resistance and a variable contribution of increased compliance of small resistive pulmonary arteries by reference to a viscoelastic model of the pulmonary circulation (19). The effects of pulmonary embolism on (Ppa  Ppao)/ plots have not been previously reported in pigs. In the present study, embolization of the porcine circulation increased the slope of (Ppa  Ppao)/ plots, with no change in pressure intercepts. A more predominant effect on slope of (Ppa  Ppao)/ plots in pigs compared with dogs may be explained by either a more important increase in resistance and/or a lesser increase in compliance at the periphery of the porcine pulmonary arterial tree.

The effects of embolic pulmonary hypertension on PVZ have been described in several studies in dogs. Embolization with 150- to 200-µm-diameter glass beads or up to 5-mm-diameter blood clots have been reported to increase Z₀, shift the f_min of Ppa/ moduli to higher frequencies, and increase low-frequency phase angle negativity, with either no change or a decrease in Zc (5, 9-11, 23). Similar changes were observed in dogs in the present study, whereas an increase in Zc occurred in minipigs. Zc has been reported to increase in a more chronic canine model of embolic pulmonary hypertension by injection of 3- to 4-mm-diameter acrylic beads (9). Acute proximal obstruction of the pulmonary arterial tree in dogs increases Zc (5, 10, 11). The increase in Zc in chronic embolic pulmonary hypertension can be explained, at least partially, by a remodeling of the vessel wall because of the prolonged exposure to an elevated Ppa. The increase in Zc in acute proximal pulmonary arterial obstruction compared with distal obstruction appears, at least partially, to be humorally mediated. The administration of norepinephrine decreases pulmonary arterial distensibility at normal as well as at high intravascular pressures (25). Stimulation of the stellate ganglion in dogs increases Zc without change in PVR (24). Platelet-derived vasoactive substances, particularly serotonin, have been reported to increase pulmonary vascular tone in acute pulmonary embolism (26). The serotonin antagonist ketanserin (which also has some ₁-adrenergic-blocking effects) blocks the Zc increase induced by ensnarement of the left main pulmonary artery (10). In the present study, the increase in Zc after induction of embolic pulmonary hypertension in minipigs may be explained by a lesser proximal distensibility, either structural or related to a more important release of humoral mediators such as catecholamines or serotonin.

In the present study, the PVZ spectrum pattern at baseline was similar in dogs and in minipigs, with a steep fall from a relatively high value at 0 Hz, a f_min at 2-4 Hz, followed by a first maximum at 6-8 Hz, and a negative phase angle at low frequencies. The canine PVZ spectra matched those of a theoretical pulsatile flow model that used experimentally measured morphometric and elasticity data and model-derived mean pressure-flow conditions of the canine pulmonary vascular tree, as recently reported by Gan and Yen (12). The porcine PVZ spectra presented with a higher Z₀ and Z₁, a f_min slightly displaced to higher frequencies, a more negative phase angle at low frequencies, but no different Zc. In both species, the fluctuations of the impedance modulus at high frequencies were small, confirming previous observations in dogs (4, 5, 9-11, 17, 18, 20, 23) and in minipigs (17). Thus, both in dogs and minipigs, the PVZ pattern is compatible with the existence of a functionally discrete reflecting site, representing a myriad of individual reflecting sites, at the periphery of the pulmonary arterial tree (2, 18, 20). It is of interest that both the global reflection coefficient Rc and time-domain-derived indexes of wave reflection were not different in the dogs and in the minipigs. Thus f_min displaced to higher frequencies and a more negative low-frequency phase angle in the minipigs have to be explained solely on the basis of an increased pulmonary arterial elastance and increased wave speed. (18, 20).

Embolic pulmonary hypertension was associated with a marked increase in high-frequency oscillations in the PVZ spectrum in pigs, but not in dogs, suggesting the existence of multiple reflection sites in the embolized porcine pulmonary circulation (2). Rc increased with pulmonary embolism but was never different between dogs and minipigs. However, Rc has been shown to be relatively insensitive to the distribution of arterial cross-sectional area and compliance (3). We therefore considered more sensitive time-domain arterial reflection indexes as reported by Ha et al. (14). Some of these indexes were affected by embolism in the sense of an earlier return of reflected waves in both dogs and minipigs, but no index of wave reflection was different between the species. Thus differences in impedance spectra between dogs and minipigs are to be explained solely on the basis of a higher elastance, probably with a higher wave speed along the porcine pulmonary arterial tree.

At baseline, minipigs, compared with dogs, presented a higher Ws, in relation to a higher Ppa at the same , no different ETR, and a smaller Wosc/WT. After embolism, Ws and Wosc increased and ETR decreased in both species, but Wosc/WT decreased only in dogs. However, in minipigs, ETR was better preserved and Wosc/WT remained lower than in dogs. The observation that embolic pulmonary hypertension is associated with a low Wosc/WT is in keeping with previous canine studies on the effects of pulmonary embolism with either 3- to 4-mm-diameter acrylic beads (9) or 150- to 200-µm glass beads (5, 23). Along the same line, patients with pulmonary hypertension secondary to mitral stenosis present with a decreased Wosc/WT (22). An increase in HR decreases Wosc/WT (21), but this cannot explain the differences in Wosc/WT between dogs and minipigs in the present study. Proximal pulmonary arterial constriction decreases Ws and Wosc by the same amount, leaving the Wosc/WT unchanged, probably because of the earlier return of reflected waves from a most proximal reflection site (5, 10, 11). A lower Wosc/WT indicates that less of the WT is wasted in pulsations, and, accordingly, Wosc/WT can be taken as an index of arterial efficiency (18, 20). Our results confirm previous suggestions (10) that a higher proximal pulmonary arterial elastance and a higher Zc improve the coupling between the right ventricle and the hypertensive pulmonary circulation.