Main

Meconium-stained amniotic fluid is present in 8 to 15% of all pregnancies, but MAS occurs in only 2 to 6% of these neonates(1). Airway obstruction, pneumonitis, hypoxemia, and pulmonary hypertension characterize this disease. MAS still remains a significant cause of neonatal morbidity and mortality and 30% of infants with MAS require mechanical ventilation, whereas severe cases need invasive therapies such as extracorporeal membrane oxygenation(2,3). Recently reported mortality from MAS ranges from 4% to as high as 40%(4,5).

After meconium is inhaled, it migrates into the peripheral airways where it leads to mechanical obstruction, hyperinflation, atelectasis, and a high risk for air leaks(1). Surfactant dysfunction has also been implicated in the pathogenesis of MAS(6,7). As early as 6 h after exposure to meconium, there is evidence of chemical pneumonitis, which is well established by 24 to 48 h(8,9). Nonetheless, the pathogenesis of the meconium-induced chemical pneumonitis still remains unclear. Glucocorticoids such as dexamethasone are potent anti-inflammatory agents that could have a role in attenuating the chemical pneumonitis observed in MAS. However, there are few studies of the effects of glucocorticoids in MAS, and their results have been controversial(10–13). Further, the use of glucocorticoids before the onset of severe chemical pneumonitis has not been extensively studied. Thus, we hypothesized that early administration of dexamethasone would significantly decrease oxygen requirements and would improve lung function in a piglet model of MAS.

METHODS

Meconium preparation. First-pass meconium was collected from healthy term newborns under sterile conditions and was cultured to exclude bacterial contamination. The meconium was pooled and diluted with normal saline to a final concentration of 20% by weight. This meconium slurry was stored in 20-mL aliquots and frozen at -20°C until used.

Animal preparation. The protocol was approved by the Animal Welfare Committee. Mixed-breed piglets, 4 to 6 d old, had a tracheostomy performed under halothane anesthesia, using a 3-French endotracheal tube (Sheridan Catheter Corporation, Argyle, NY). Then, each piglet was placed supine under a radiant warmer, and mechanical ventilation was started with a Sechrist neonatal ventilator (Sechrist Industries, Anaheim, CA). The piglets were sedated with an initial dose of 15 mg/kg of pentobarbital and maintained with 5 to 10 mg/kg as needed to keep the heart rate below 200 beats per minute. Also, the animals received pancuronium bromide 0.1 mg/kg as needed to prevent spontaneous respiration. A femoral arterial catheter was placed to monitor blood pressure (Hewlett Packard Neonatal, Andover, MA) and blood gases (Gem Premier, Mallinckrodt Sensor Systems, Ann Arbor, MI). In addition, ETCO2 and Hb saturation were monitored continuously using a Nellcor Ultracap (Nellcor Inc., Haywood, CA). Dynamic compliance and mean airway pressure were measured continuously using a BICORE CP-100 neonatal monitor (BICORE Monitoring Systems, Irvine, CA).

Ventilation protocol. All animals were started at a ventilation rate of 20 breaths per minute, PIP of 20 cm H2O, PEEP of 3 cm H2O, FiO2 of 0.21, inspiratory time of 0.5 s, and flow of 8 L/min. These settings resulted in acceptable chest wall movement and arterial blood gases. The ventilator settings were modified in a standardized manner to attempt to maintain an ETCO2 of 35-40 mm Hg and Hb saturation of 92-96%. These parameters were correlated periodically with determinations of arterial blood gases. An ETCO2 of 35-40 mm Hg corresponded to similar PaCO2 values (r2 = 0.95), whereas a Hb saturation of 92-96% resulted in a PaO2 of 65-80 mm Hg (r2 = 0.95). For Hb saturations <92%, the PEEP was increased first to a maximum of 6 cm H2O followed by increases of FiO2 up to 1.0 if needed. For Hb saturation values above 96%, the FiO2 was decreased first and then the PEEP to as low as baseline values. For ETCO2 above 40 mm Hg, the tidal volume was increased initially to a maximum of 12 mL/kg, and then the ventilator rate was increased up to 40 breaths per minute. For ETCO2 values below 35 mm Hg, the ventilatory rate was decreased first, and then the tidal volume was reduced to baseline if needed.

The OI and VEI were calculated as follows: (Equation 1)

Study protocol. After the animals were stabilized on the ventilator, 4 mL/kg of the 20% meconium slurry was instilled into the airways via the tracheostomy over 30 min. This ensured a relatively uniform distribution of meconium in both lungs and was tolerated well by the piglets. Ventilator settings were adjusted according to the aforementioned ventilation strategy from that point onward throughout the duration of the study. Two hours after instillation of meconium, the piglets were randomly assigned to receive either 0.5 mg/kg of dexamethasone or an equal volume of normal saline intravenously. A second dose of dexamethasone/placebo was given 6 h after the first dose (8 h after administration of meconium). Twelve hours after the meconium instillation, the animals were killed using an infusion of high-dose pentobarbital and potassium chloride. The lungs were removed and inflated via the tracheostomy with 12mL/kg of formalin to a pressure of 25 cm H2O and were used for histologic examination.

Data analysis. Data are expressed as mean and SEM. Comparisons were done by 1-way ANOVA for repeated measures with post hoc testing by the Student-Newman-Keuls test. A p value of <0.05 was considered significant.

RESULTS

A total of 10 piglets were studied, five in the control and five in the dexamethasone group. Their mean age and weight were comparable (5 ± 0.3 versus 5 ± 0.4 d and 2.62 ± 0.1 versus 2.52 ± 0.2 kg, respectively).

Before meconium instillation, the oxygen and ventilatory requirements to maintain acceptable blood gases were similar between both groups (Fig. 1). Administration of meconium 2 h after the start of the study protocol resulted in significantly increased oxygen and ventilatory requirements to maintain blood gases in the predetermined range and a marked deterioration of lung compliance in all piglets (Figs. 1 and 2). There were no differences between the two groups at this time point.

Figure 1
figure 1

Oxygen requirement, PaO2, and PaCO2 in the two groups. Data are expressed as mean ± SEM. *Indicates a significant differences between the groups (p < 0.05).

Full size image
Figure 2
figure 2

OI, VEI, and lung compliance in the two groups. Data are expressed as mean ± SEM. *Indicates a significant difference between the groups (p < 0.05).

Full size image

Dexamethasone or placebo was administered 2 h after instillation of meconium. Piglets exposed to dexamethasone exhibited a rapid and sustained decrease in oxygen requirements, which remained significantly different from the placebo group throughout the rest of the study period (Fig. 1). At the end of the protocol, the dexamethasone-treated group needed an FiO2 of 0.27 ± 0.01 to maintain acceptable PaO2 values compared with the control group that still required an FiO2 of 1.0 ± 0.0 (p < 0.0001). Because the adjustments in ventilator settings were based primarily on a range of ETCO2 and Hb saturation values, there were small but significant differences in PaO2 and PaCO2 between groups (Fig. 1). These were only observed beyond administration of the initial dose of dexamethasone or placebo.

Dexamethasone administration resulted in a sustained improvement in lung function in that group of piglets. They had a lower need for ventilatory support compared with the placebo group within a few hours of the initial dose of dexamethasone. This is shown better by comparing the OI and VEI, which incorporate the values of arterial blood gases in their calculation and account for some of their variability (Fig. 2). Lung compliance improved after administration of the first dose of dexamethasone and remained higher than in the control group thereafter.

There were no differences between the two groups in heart rate and blood pressure measurements throughout the study (Fig. 3). Also, the amount of sedation given to piglets from both groups was comparable (data not shown). Upon histologic examination of the lungs, there were no overt differences in the degree of inflation or atelectasis between the two groups (Fig. 4).

Figure 3
figure 3

Heart rate and mean arterial blood pressure (MABP) in the two groups. Data are expressed as mean ± SEM.

Full size image
Figure 4
figure 4

Representative lung sections stained with hematoxylin and eosin from control (A) and dexamethasone-treated (B) piglets. Magnification × 200.

Full size image

DISCUSSION

The pathophysiology of MAS seems to be multifactorial. The initial acute changes in both oxygenation and lung compliance are most likely due to the obstructive nature of particulate meconium(8). Distal migration of meconium is believed to cause alterations in surfactant function and also induction of inflammatory processes leading to chemical pneumonitis(6–8). A better understanding of MAS has been derived from studies in animal models of this disease. Although in most of these studies MAS has been produced primarily by administration of meconium as a bolus, we observed a comparable worsening of lung function in our newborn piglets after infusion of a 20% meconium slurry for 30 min via a tracheostomy(13–15).

Although the mechanisms and mediators involved in the chemical pneumonitis associated with MAS have not been extensively studied, several attempts to attenuate this inflammatory response with glucocorticoids have been made. Earlier studies by Frantz et al.(10) in a rabbit model and Yeh et al.(11) in human newborns failed to show any benefit of glucocorticoid administration in MAS. However, van den Anker and van Loenen(12) who studied newborns and Soukka et al.(13) who used a piglet model of MAS did show a beneficial effect of glucocorticoids. These studies used glucocorticoids with different antiinflammatory potencies as well as various time periods for administration of their initial dose. Moreover, some of these studies were done in animals beyond the neonatal period, and they may respond differently to the inflammatory effects of meconium. Since our primary goal was to study an intervention that could potentially be used in clinical trials, we elected to give dexamethasone because of its strong antiinflammatory effect and extensive use in neonatal intensive care nurseries for treatment of inflammatory lung diseases(16,17). Dexamethasone was given 2 h after the instillation of meconium to allow sufficient time for the baseline lung injury to develop.

In our study, the dexamethasone-treated group showed a rapid and sustained improvement in oxygenation and lung function. Multiple mechanisms may be involved in this beneficial effect of dexamethasone. There is evidence to suggest that there is a surfactant dysfunction in MAS(14–20). This could account for the worsening lung compliance and hence the increase in oxygen requirements and ventilatory pressures observed in this disease. These abnormalities of surfactant function may be secondary to displacement of some of its components by the fatty acids of meconium or alterations in surfactant morphology(7,14,15,18–20). In addition, the inflammatory changes associated with the chemical pneumonitis of MAS, which include a migration of neutrophils to the lungs, activation of alveolar macrophages, and influx of plasma proteins, may affect surfactant function further(21,22). By virtue of its potent antiinflammatory properties, dexamethasone could potentially attenuate the lung injury and surfactant inhibition secondary to meconium. Because we observed a rather rapid onset of the dexamethasone response, it is unlikely that this reflects the known effects of glucocorticoids on synthesis of the phospholipids and specific proteins of surfactant(23,24). Further, at the end of the study, we found no obvious histologic differences between the two groups in terms of lung inflation and atelectasis. This suggests that the higher oxygen and ventilator pressure requirements observed in the control piglets were not due to under- or overinflation of the lungs.

Other abnormalities seen in MAS include increases in lung water content and pulmonary hypertension(8,9). These changes are probably secondary to alterations in vascular permeability and pulmonary vasomotor tone induced by products derived from arachidonic acid metabolism(25). By blocking the effects of lipoxygenase and phospholipase A2, dexamethasone may have decreased vascular permeability and progressively improved lung compliance(26,27). Moreover, Soukka, et al.(13) showed that pretreatment of 10-wk-old pigs with methylprednisolone attenuated the pulmonary hypertensive response to meconium. The combination of an improved lung compliance and decreased ventilation-perfusion mismatch may explain the marked improvement in oxygenation observed after dexamethasone administration in our study.

We did not observe significant changes in heart rate or blood pressure after induction of MAS or administration of dexamethasone. Thus, it is unlikely that variations in cardiac output explain the improvements in oxygenation observed after initiation of the therapy with glucocorticoids.

In conclusion, we have shown that, in a newborn piglet model of MAS, early treatment with dexamethasone significantly improved oxygenation and lung function. Our data along with preliminary results from a clinical study in newborns with MAS(28) suggest that early administration of glucocorticoids with potent antiinflammatory effects may be of benefit in the treatment of neonates with MAS. However, this hypothesis needs to be tested in a well-designed, controlled clinical trial.