Main

Persistent pulmonary hypertension of the newborn is associated with numerous undesirable consequences. Therapy for persistent pulmonary hypertension of the newborn has traditionally attempted to reduce PAP to a greater extent than SAP, eliminating the shunt of blood from venous to arterial circulations. Many attempts have been made to identify an agent that selectively dilates the pulmonary circulation. Almost all have failed. When infused i.v., vasodilator prostaglandins (D2 and E1),α-adrenergic inhibitors (tolazoline), direct NO donors (nitroprusside, nitroglycerin), and agents stimulating NO release (acetylcholine) all tend to dilate the pulmonary and systemic circulations more or less equally(1–6). This nonselective behavior tends to convert a normotensive patient with a right-to-left shunt into a hypotensive patient with the same right-to-left shunt. On balance, this is rarely viewed as a success.

One of the few pharmacologic agents consistently successful selectively reducing PAP in laboratory models of pulmonary hypertension is iNO. NO, thought to be identical (or very closely related) to endothelial derived relaxing factor, is produced in vascular endothelial cells, and acts in a paracrine fashion on neighboring vascular smooth muscle cells to promote vasodilation(7–10). iNO selectively reduces PAP in lambs and piglets with hypoxia-induced pulmonary hypertension(11–14), prematurely delivered lambs with hyaline membrane disease(15), and even some models of nonhypoxia induced pulmonary hypertension(16). iNO has also been used in human infants and adults with pulmonary hypertension, with at times dramatic, at times somewhat mixed results(17–21).

Unfortunately, the technology required to support iNO in human infants is not universally available. We tested whether inhalation of Neb-NP also causes selective vasodilation of PAs. We studied the effects of Neb-NP in piglets with pulmonary hypertension induced by hypoxia and by infusion of GBS. Additionally, we tested whether treatment with dipyridamole, an agent which augments NO-cGMP signal transduction, enhances pulmonary vasodilation induced by inhalation of Neb-NP. Finally, we compared Neb-NP during hypoxia with nebulization of another commonly used vasodilator, tolazoline(Neb-tolazoline).

METHODS

Anesthesia and surgical preparation. Piglets (1-2 wk old; n = 23) received ketamine intraperitoneally (20 mg/kg), and were intubated endotracheally. An ear vein was cannulated for venous access. The piglets were anaesthetized with sodium pentobarbital (20 mg/kg i.v.), and muscle relaxation was achieved with D-tubocurarine (1 mg/kg). Throughout the experiment, the level of anesthesia was assessed by determining the response of heart rate and blood pressure to noxious stimuli, and adjusted by intermittent bolus of pentobarbital, ∼2 mg kg-1 h-1. Warming blankets and overhead heating lamps were used to maintain body core temperature at 37-38°C. The endotracheal tube was replaced by a tracheostomy tube tied snugly with umbilical tapes. A femoral artery catheter was placed to provide access to arterial blood and monitor SAP. A suprapubic cystostomy catheter was placed to establish urinary drainage.

Polyethylene catheters were surgically introduced into the right atrium(via left external jugular vein), and PA (via left lateral thoracotomy). Following our previously described methodology(1), an external electromagnetic flow probe (EP 425; Carolina Medical Electronics, King, NC) was placed around the PA. During each experiment, the PA flow probe was zeroed frequently using the approximation that diastolic blood flow in the PA is zero. In the documented absence of vascular shunts between the systemic and pulmonary circulations, PA blood flow was taken to reflect total body cardiac output.

The PA and right atrium catheters were connected to pressure transducers to determine PAP and central venous pressure. We have previously determined in piglets that left atrial pressure can be estimated as 2 cm greater than central venous pressure, both under conditions of normal PAP as well as under the conditions of hypoxia, GBS infusion, and NP administration(1,4,22,23). Systemic vascular resistance and pulmonary vascular resistance were calculated as the ratio of(SAP - central venous pressure)/cardiac output and (PAP - left arterial pressure)/cardiac output, respectively.

Ventilation and nebulizer circuit. Mechanical ventilation(Sechrist infant ventilator, IV 100B, Sechrist Industries, Anaheim, CA) was used with initial settings of FIO2 = 0.30, pressure maximum, Pmax 18 cm H2O, positive end-expiratory pressure = 3 cm H2O, rate = 15 breaths/min, inspiration time = 0.7 s, flow = 6 L/min. The inspiratory ventilator circuit was modified to include a"Y"-shaped dual nebulizer circuit, which joined the inflow tubing 5 cm from the tracheostomy port. The nebulizer circuit was designed to allow rapid switching from one nebulized agent to another (in our case, from 0.9% NaCl to nitroprusside dissolved in 0.9% NaCl) without altering gas flow or FIO2 in the breathing circuit.

Preparation of GBS organisms. GBS organisms, serotype Ib(graciously provided by Dr. Lawrence Madoff, Channing Laboratories, Harvard Medical School) were incubated in Todd-Hewitt medium overnight to late log phase (∼3 × 109 organisms/mL). Immediately before infusion into the piglet, the organisms were centrifuged (20 min at 3000 ×g), the supernatant was discarded, and the pellet of organisms was resuspended in 0.9% NaCl to one-fourth its original volume. Quantitative assessments of the infusate concentrations were determined retrospectively by serial dilution.

Blood gas analysis. Arterial blood samples, taken at specified intervals (see below), were analyzed for pH, PO2, PCO2, and base excess using a blood gas analyzer (IL 1303, Instrumentation laboratories, Lexington MA) and for Hb and oxygen saturation using a co-oximeter (IL 282, Instrumentation Laboratories). The cooximeter had previously been calibrated for piglet blood in our laboratory using an oxygen scrubbing technique(24). All piglets were allowed to stabilize for approximately 1 h after surgery before the initiation of any experimental protocol.

Experimental protocols. Normoxia/Neb-NaCl (n = 16). During normoxia, 0.9% NaCl (vehicle for subsequent nebulized NP) was aerosolized in a nebulizer chamber at a flow of 4 L/min and introduced into the proximal airway of the ventilator circuit. After 10 min of hemodynamic stability, a blood gas was drawn to ensure adequate oxygen and ventilation during this normoxia/Neb-NaCl condition.

Hypoxia/Neb-NaCl (n = 16). The piglet was then made hypoxic by reducing FIO2 to 0.08 without in any other way altering the ventilator/nebulizer circuit. The piglet was observed continuously for 15 min, by which time hemodynamic and blood gas responses had stabilized. A second blood gas was drawn during this hypoxia/Neb-NaCl condition. In five additional piglets, this hypoxia/Neb-NaCl phase was continued for 1 h, to determine the stability of vascular responses during prolonged hypoxia.

Hypoxia/Neb-NP (n = 16). Upon completion of the hypoxia/Neb-NaCl protocol, the nebulizer gas flow was switched from the NaCl chamber to the NP chamber while FIO2, nebubulizer flow rate, and ventilator settings were held constant. The time required to switch nebulizer circuits was less than 1 s. The dose of NP, 50 mg dissolved in 5 mL of 0.9% NaCl, was chosen after preliminary trials with smaller doses produced less significant reductions in PAP under these experimental conditions. The piglet was observed continuously for 15 min under these conditions, by which time hemodynamic responses had stabilized. A blood gas was drawn at the completion of this hypoxia/Neb-NP condition. In five piglets, the duration of hypoxia/Neb-NP was extended to 60 min in an attempt to assess whether tachyphylaxis or other untoward hemodynamic events might occur.

Hypoxia/Neb-NP/Neb-NaCl (n = 5). After 1 h of the hypoxia/Neb-NP protocol, the nebulizer circuit was switched back from Neb-NP to Neb-NaCl in an attempt to determine the duration of selective reduction in PAP when NP was discontinued while hypoxia persisted. The piglets were observed continuously during hypoxia for an additional 15 min of Neb-NaCl, by which time hemodynamic responses had become stable. A blood gas was drawn at the completion of this protocol.

Dipyridamole/hypoxia/Neb-NP/Neb-NaCl (n = 5). In normoxic piglets, dipyridamole was administered i.v. (0.6 mg/kg over 10 min) while hemodynamic values were monitored continuously. The piglet then received 15 min of the hypoxia/Neb-NaCl protocol, followed by 15 min of hypoxia/Neb-NP, followed by another 15 min of the hypoxia/Neb-NaCl protocol.

Hypoxia/Neb-tolazoline (n = 4). Piglets were made hypoxic according to the hypoxia/Neb-NaCl protocol. These piglets were then treated with Neb-tolazoline (5 mg/mL, dissolved in 0.9% NaCl) and observed for 15 min. A blood gas was drawn at the end of this protocol.

GBS/Neb-NP (n = 6). Piglets who had participated in the hypoxia/Neb-NP protocol were returned to normoxia and subsequently infused with group B streptococci at 4 × 107 organisms kg-1 min-1 for approximately 120 min, until hemodynamically stable pulmonary hypertension had been demonstrated. With the GBS infusion continuing, the normoxia/Neb-NaCl protocol was initiated, and a blood gas was obtained. The nebulizer circuit was then switched, and the piglets received 15 min of normoxia/Neb-NP after which another blood gas was obtained.

Normoxia/Neb-NP (n = 4). Piglets received Neb-NaCl during normoxia followed by 15 min of Neb-NP during normoxia.

Statistical analysis. All statistical comparisons, both within experimental protocols (as a function of time) and across experimental protocols were performed using ANOVA. Where significant statistical differences were determined, subsequent pairwise comparisons were made using the Newman-Keuls test. Statistical significance was accepted at the p < 0.05 level. Data are expressed as mean ± SEM.

RESULTS

Figure 1 presents a tracing of SAP and PAP versus time for a single piglet, before and after the switch from hypoxia/Neb-NaCl to hypoxia/Neb-NP. Within minutes of the onset of Neb-NP, with hypoxia persisting, PAP fell steadily from approximately 28 to 22 mm Hg. SAP was unaffected.

Figure 1
figure 1

Tracing of SAP and PAP vs time for a single hypoxic piglet, before and after the switch from Neb-NaCl to Neb-NP. Within minutes of the onset of Neb-NP, PAP fell steadily, whereas SAP was unaffected.

Full size image

Figure 2 displays SAP and PAP during successive hypoxia/Neb-NaCl and hypoxia/Neb-NP protocols for 16 piglets. During Neb-NaCl, hypoxia induced a significant rise in PAP (+13.9 ± 1.3 mm Hg; p < 0.001), but not SAP. When the circuit was switched to Neb-NP, PAP fell significantly (-8.4 ± 0.9 mm Hg; p< 0.001), whereas SAP remained stable. Cardiac output did not change significantly during the transition from normoxia to hypoxia (726 ± 38 to 735 ± 50 mL/min) or after Neb-NP (709 ± 51 mL/min). Consequently, changes in systemic and pulmonary vascular resistance during these protocols paralleled changes in systemic and pulmonary vascular pressures. Repeated administrations of Neb-NP in four hypoxic piglets did not appear either to augment or diminish the reductions in PAP (-25.6 ± 6.5% versus -30.8 ± 4.5%).

Figure 2
figure 2

Time course of changes in SAP and PAP during successive hypoxia/Neb-NaCl and hypoxia/Neb-NP protocols. Data from 16 piglets. During Neb-NaCl, hypoxia induced a significant rise in PAP, but not SAP. When the circuit was switched to Neb-NP, PAP fell significantly, but SAP was unaffected.

Full size image

Figure 3 displays SAP and PAP for five piglets in whom the hypoxia/Neb-NP protocol was extended for 1 h, after which Neb-NP was replaced by Neb-NaCl while hypoxia continued. Two conclusions can be drawn from these experiments: 1) the selective pulmonary versus systemic activity of Neb-NP did not wane significantly over 1 h; 2) after Neb-NP was discontinued while hypoxia persisted, PAP increased significantly. As a control protocol for this condition, four additional piglets received 1 h of hypoxia/Neb-NaCl without switching to Neb-NP. PAP remained consistently elevated in these piglets (27 ± 2.1 mm Hg at 15 min; 27 ± 3.5 mm Hg at 60 min;p = NS).

Figure 3
figure 3

Time course of changes in SAP and PAP during successive protocols: hypoxia/Neb-NaCl, hypoxia/Neb-NP extended for 1 h, and the subsequent 15-min period after Neb-NP was discontinued while hypoxia continued. Data from five piglets. The selective reduction in PAP induced by Neb-NP did not change between 15 and 60 min. After Neb-NP was discontinued while hypoxia persisted, PAP rose again.

Full size image

Figure 4 reveals the effects of administration of dipyridamole on SAP and PAP during successive normoxia/Neb-NaCl, hypoxia/Neb-NaCl, hypoxia/Neb-NP, and hypoxia/Neb-NaCl protocols. Dipyridamole caused a significant reduction of SAP in the normoxia/Neb-NaCl condition (77.6 ± 5.8 to 44.4 ± 2.7 mm Hg (p < 0.001), whereas cardiac output did not change significantly (682 ± 60 to 604 ± 96 mL/min; p = NS). SAP did not change significantly during any subsequent protocol. There were no significant differences in PAP between piglets who received dipyridamole and those who did not (cf.Fig. 3) under any of these experimental conditions.

Figure 4
figure 4

Time course of changes in SAP and PAP after pretreatment with dipyridamole during successive hypoxia/Neb-NaCl, hypoxia/Neb-NP, and hypoxia/Neb-NaCl protocols. Data from five piglets. SAP fell significantly after dipyridamole during normoxia, and did not recover during any of the subsequent hypoxia protocols. There was no significant difference in PAP between piglets who received dipyridamole and those who did not (cf. Figs. 2 and 3) under any experimental conditions.

Full size image

Figure 5 presents the effects of Neb-tolazoline on SAP and PAP during hypoxia-induced pulmonary hyptertension for four piglets. In contrast to Neb-NP, Neb-tolazoline did not reduce PAP during hypoxia (-1.0± 0.7 mm Hg; p < 0.001 versus Neb-NP). Moreover, SAP fell during Neb-tolazoline (-7.7 ± 3.2 mm Hg;p = 0.05), whereas systemic cardiac output remained unaffected.

Figure 5
figure 5

Time course of changes in SAP and PAP during Neb-tolazoline. Data from four piglets. During hypoxia, Neb-tolazoline did not reduce PAP, but SAP fell significantly.

Full size image

Figure 6 presents the effects of Neb-NP on SAP and PAP for six piglets during GBS-induced pulmonary hypertension. GBS infusion induced more extensive pulmonary vasoconstriction than did hypoxia(ΔPAP: +24.5 ± 1.9 versus +13.9 ± 1.3 mm Hg; p < 0.001). Neb-NP during GBS infusion caused a small but statistically significant reduction in PAP (-3.1 ± 0.7 mm Hg;p < 0.05), although not significantly affecting SAP (-0.5± 3.0 mm Hg; p = NS) nor cardiac output (+2 ± 32 mL/min; p = NS). The selective pulmonary vasodilator effect of Neb-NP was significantly smaller during GBS-induced pulmonary hypertension compared with hypoxia-induced pulmonary hypertension in the same piglets (PAP: -7.6 ± 1.6% versus -28.1 ± 2.4%;p < 0.001).

Figure 6
figure 6

Time course of changes in SAP and PAP during Neb-NP in piglets with GBS-induced pulmonary hypertension. Data from six piglets. PAP fell slightly but significantly during Neb-NP, and SAP was unaffected.

Full size image

Table 1 presents blood gas values under normoxia/Neb-NaCl, hypoxia/Neb-NaCl, hypoxia/Neb-NP (15 min), hypoxia/Neb-NP(60 min), and GBS/Neb-NP. Neb-NP did not significantly impair or promote either oxygenation or ventilation during any of these protocols. After 60 min of hypoxia, a significant degree of metabolic acidosis had developed. Neb-NP had no significant effect on either blood gases or hemodynamics during normoxia in four piglets.

Table 1 Blood gas values under normoxia/Neb-NaCl (n = 16), hypoxia/Neb-NaCl (n = 16), hypoxia/Neb-NP (15 min; n = 16), hypoxia/Neb-NP (60 min; n = 5), and GBS (n = 6)
Full size table

DISCUSSION

We demonstrate here that Neb-NP produced selective pulmonary vasodilation during hypoxia-induced pulmonary hypertension in piglets. Both in timing and magnitude, the reduction of PAP after Neb-NP closely resembled the reduction of PAP after iNO observed in piglets by Neilin et al.(12). The effect began within 2 min of inhalation, and appeared to reach its maximum by roughly 15 min. Thereafter, no increase or decrease of PAP was noted for at least 1 h. The magnitude of selective pulmonary vasodilation effected by Neb-NP or iNO was roughly two-thirds of the entire rise in PAP associated with hypoxia. This relationship appears to hold whether the elevation in PAP was to approximately 25 torr (as usually reported)(11,12) or in the current experiments where we deliberately chose to magnify the amplitude of the effect by pushing PaO2 lower and PAP higher than most hypoxia protocols in piglets(PaO2 ∼ 28 torr; PAP ∼ 30 torr). Because cardiac output was not significantly affected by either hypoxia or Neb-NP, changes in vascular resistance (both systemic and pulmonary) paralleled changes in vascular perfusion pressure in the hypoxia/Neb-NP protocols.

We have little data on the dose-response for Neb-NP in these protocols. In a preliminary series of experiments, we determined that a lower concentration of NP in the nebulizer chamber (5 mg/mL) produced approximately one-third of the reduction in PAP noted with the 50 mg/mL concentration reported here(again with no significant effect on SAP or cardiac output). We were unable to determine the actual amount of NP that reached the piglet from our nebulizer system.

Others have reported that dipyridamole enhanced the selective pulmonary vasodilation associated with iNO(25), consistent with theoretical predictions that dipyridamole, by inhibiting phosphodiesterase, should prolong the stability of intracellular cAMP compounds produced by nitrovasodilators. Dipyridamole did not enhance pulmonary vasodilation induced by Neb-NP in our protocols. In addition, the profound systemic hypotension induced by dipyridamole in the normoxia condition makes clinical application of this agent unlikely. Recently, Dukarm et al.(26) have noted similar untoward effects of dipyridamole on SAP in lambs with pulmonary hypertension.

We have previously reported the effects of i.v. NP in a model of pulmonary hypertension in piglets(4), and demonstrated no significant selectivity of vasodilator effects comparing systemic versus pulmonary circulations. Others(27) have also reported nonselective vasodilation after i.v. NP in lambs, both at normal and elevated PA pressures. In addition, we (and others as well) have reported that tolazoline, whether administered i.v.(1) or endotracheally (as noted here) causes simultaneous pulmonary and systemic vasodilation. Presumably, the kinetics of the "Neb-NP to NO to intracellular GAMP to vasodilation" pathway differ from the kinetics of the 'Nebtolazoline to vasodilation' pathway. On analogy to iNO, the vasodilator kinetics of Neb-NP are likely to be rapid compared with the "stability" of NP in the circulation; that is, Neb-NP gets "degraded" or "inactivated" quickly enough that it is more active at the local site of administration (the lung) than at a distant site (the systemic circulation); whereas, to complete the syllogism, Neb-tolazoline is more stable in the bloodstream and is not degraded in the time frame of a single pass from the pulmonary to systemic circulation. The alternative hypothesis, that the pulmonary circulation is somehow more "NP/NO sensitive" (e.g. more receptors, different receptor characteristics) than the systemic circulation, fails with the finding that i.v. administration of NP does not produce differential pulmonary versus systemic effects.

The question of why Neb-NP was so much more effective in hypoxia-induced versus GBS-induced pulmonary hypertension remains unresolved. In our protocols, Neb-NP worked on every administration (16/16 reported here) to reverse hypoxia-induced pulmonary hypertension (by roughly 66% of the original elevation), whereas it worked only marginally well to reverse GBS-induced pulmonary hypertension (roughly 15%). Berger et al.(16), in a model of NO inhalation in heat-killed GBS-treated piglets demonstrated virtually complete reversal of the pulmonary hypertension. Similar results have been noted by Barrington et al.(28). However, others have not been able to demonstrate such a profound impact of inhaled NO during sepsis or sepsis analog protocols. As an example Zobel et al.(29) demonstrated in a lung washout/thromboxane infusion model of pulmonary hypertension that inhaled NO had virtually no selective impact on the pulmonary circulation. Moreover, Nagoshi et al.(30) could not demonstrate selective reduction in PVR after NO inhalation in septic piglets. At present, we have no explanation for these differences.

In addition to its beneficial effect on PAP, NO has been reported in some, but not all contexts to improve arterial hypoxemia even in the absence of right-to-left shunts(15,21). The presumed mechanism underlying this observation is an intrapulmonary redistribution of blood flow, increasing blood flow to well aerated regions whose perfusion is augmented by NO-induced vasodilation. We noted no improvement in PaO2 after Neb-NP in our hypoxic piglets, who presumably had healthy lungs and minimal V/Q mismatch. In addition, we noted no significant improvement in PaO2 during Neb-NP in the septic piglets, whose lungs had been sufficiently insulted by GBS to have reduced PaO2 significantly, although not to frankly hypoxic levels. In this observation, we support the findings of Berger et al.(16), who found little improvement in PaO2 in septic piglets receiving iNO, despite the presumably widened V/Q distribution induced by GBS infusion.

Methemoglobinemia and tachyphylaxia are two potential concerns when administering iNO, and, by extension, Neb-NP. We observed no tachyphylaxis during 1 h of continuous Neb-NP, nor did we find any reduction in effectiveness when Neb-NP was administered in repeated trials. In addition, others have reported long-term use of iNO (up to several days) without noting complications(17,19,21). Similarly, clinical experience with longer-term use of i.v. NP has suggested that neither tachyphylaxis nor methemoglobinemia is a major clinical side effect in children. Independent of the inhaled agent, continuous nebulization as a mode of drug delivery has been used clinically for pediatric intensive care unit patients in many contexts without undue complications. In support of this, several hours of the Neb-NaCl protocol produced no untoward effects in our piglets, either in terms of altering oxygenation, ventilation, or PAP.

In sum, we believe the most significant finding reported here is that Neb-NP produced prompt, significant, selective reduction of PAP in piglets with pulmonary hypertension, without apparent complications for up to 1 h. Cautious extrapolation of these findings to selected clinical contexts in human infants may be warranted in the future.