Main

The structural and functional pulmonary immaturity which characterizes HMD renders the premature newborn uniquely susceptible to lung injury from oxygen toxicity and volutrauma(1, 2). Other factors implicated in the evolution of chronic lung disease in premature infants with HMD include increased lung vascular permeability(3–5) neutrophil-mediated inflammation(6–9) and altered production of inflammatory cytokines(10). Although surfactant replacement therapy has dramatically improved the management of HMD, its effects on the incidence of chronic lung disease have been less striking, attesting to the multifactorial nature of lung injury in prematurity(11–13).

Pulmonary hypertension complicates the course of infants with severe HMD(14–17) and may increase mortality despite surfactant therapy(18). Although recent studies have shown that iNO may improve oxygenation and decrease right-to-left extrapulmonary shunting in premature infants with severe HMD(19, 20), potential risks of iNO treatment in this setting include increased oxidative lung injury and inflammation, and augmented pulmonary edema due to increased pulmonary blood flow. In addition to its effects on pulmonary hemodynamics and gas exchange during inhalation, endogenous NO may regulate vascular permeability and neutrophil adhesion in the microcirculation(21). However, its effects in the immature lung have not been studied. Although recent reports suggest that iNO may decrease capillary permeability in acute lung injury(22, 23), the immature lung is particularly susceptible to lung protein leak during mechanical ventilation(24). Increasing pulmonary blood flow with iNO in the premature subject could also have potential adverse effects. For example, conditions associated with increased pulmonary blood flow in the premature infant (patent ductus arteriosus) cause increased lung neutrophil accumulation(25). Although low dose iNO may be a safe and effective therapy in term neonates with severe persistent pulmonary hypertension of the newborn(26, 27), treatment of the premature subject with iNO must also take into account decreased antioxidant host defenses and the predisposition to bronchopulmonary dysplasia in the immature lung(1).

The effects of iNO on pulmonary vascular permeability and lung leukosequestration in premature infants with HMD are unknown, and the potential application of this new therapy in premature infants warrants careful scrutiny for adverse consequences. We hypothesized that low dose iNO(≤20 ppm) might have a cytoprotective effect resulting in decreased lung vascular protein leak and pulmonary neutrophil accumulation. To test these hypotheses, we delivered ovine fetuses at 78% of term gestation (115 d of gestation) in two separate protocols to measure the effects of iNO on pulmonary vascular protein permeability and lung neutrophil accumulation.

METHODS

Mixed-breed (Columbia-Rambouillet) pregnant ewes were used in this study. All procedures and protocols were reviewed and approved by the Animal Care and Use Committee at the University of Colorado Health Sciences Center. This study includes two separate protocols that examine the effects of iNO on pulmonary vascular endothelial permeability (protocol 1) and lung neutrophil accumulation (protocol 2) in this premature (115 d of gestation, 0.78 term; term = 147 d) ovine model of severe HMD(28).

Surgical preparation. Ewes were sedated with i.v. pentobarbital sodium (2-4 g total dose) and anesthetized with intrathecal tetracaine hydrochloride (1% solution, 3 mg). Under sterile conditions a uterine incision was made, and the left fore limb of the fetal lamb was delivered. A skin incision was made in the axilla of the left fetal fore limb after local infiltration with lidocaine (1% solution, 2-3 mL). Polyvinyl catheters (20 gauge; Martech Medical Products, Lansdale, PA) were advanced into the ascending aorta through the axillary artery and into the superior vena cava through the axillary vein. A left thoracotomy was performed, exposing the heart and great vessels. A catheter was inserted into the main pulmonary artery by direct puncture through purse-string sutures. This catheter was guided into position with a 14-gauge i.v. placement unit (Angiocath; Travenol, Deerfield, IL), and secured by tightening the purse-string suture as the introducer was withdrawn. The main pulmonary artery catheter was inserted between the ductus arteriosus and the pulmonic valve. A left atrial catheter was inserted in the medial portion of the left atrial appendage. An ultrasonic flow transducer (6 mm, Transonic Systems Inc., Ithaca, NY) was placed around the left pulmonary artery to measure blood flow to the left lung.

Study design-general. After stabilization of physiologic parameters, pancuronium was administered to the fetus (0.1 mg/kg), the fetal head was exteriorized, and a tracheotomy was performed with placement of an endotracheal tube (3.0 mm inside diameter). All animals were treated with exogenous surfactant (Infasurf, kindly provided by E. A. Egan M.D.) at an estimated dose of 3 mL/kg (105 mg of phospholipid/kg) before the first breath. Mechanical ventilation was initiated with a continuous flow, time-cycled, pressure-limited, neonatal ventilator at the following settings: PIP, 35 cm H2O (3.4 kPa); PEEP, 6 cm H2O (0.6 kPa); rate, 30 breaths/min; inspiratory time, 1.0 s; and Fio2 = 1.00. After 30 min of mechanical ventilation, the umbilical cord was ligated, and a 5% dextrose solution was infused to provide 10 mL/h crystalloid and 1 mg/kg/h pentobarbital. Mechanical ventilator settings were modified during the course of studies based on results of arterial blood gas samples. Changes in PIP were determined by measurements of Paco2. If Paco2 was <35 mm Hg (4.7 kPa), then the PIP was reduced to 30 cm H2O (2.9 kPa). If subsequent measurements of Paco2 were <35 mm Hg (4.7 kPa), then the PIP was reduced to 25 cm H2O (2.5 kPa). The maximum PIP delivered was 35 cm H2O (3.4 kPa). If Paco2 was greater than 45 mm Hg (6 kPa), then the ventilator rate was increased to a maximum of 60 breaths/min. The inspiratory time was then decreased to maintain an inspiratory to expiratory ratio of ≤1.0. PEEP was changed according to Pao2. If Pao2 was <100 mm Hg (13.3 kPa), PEEP was maintained at 6 cm H2O (0.6 kPa). With Pao2 > 100 mm Hg (13.3 kPa), PEEP was decreased to 5 cm H2O (0.5 kPa). If Pao2 was >200 mm Hg (26.7 kPa), PEEP was decreased to 4 cm H2O (0.4 kPa). Main pulmonary artery, aortic, and left atrial catheters were connected to a Gould-Statham P23 ID pressure transducer. Calibrations of pressure transducers were performed with a mercury column manometer. QLPA and heart rate were measured using the ultrasonic flow transducer. Pulmonary vascular resistance was calculated as: [(pulmonary artery pressure - left atrial pressure)/QLPA]; mm Hg·(mL/min)-1. Blood samples for pH, Po2, Pco2, percent oxygen saturation, and percent methemoglobin were withdrawn anaerobically into Natelson glass pipettes and analyzed at 39.5 °C using a Radiometer OSM3 blood gas analyzer (Copenhagen, Denmark). The NO gas (Scott Specialty Gases, Plumbsteadville, PA) used in these experiments was certified at a concentration of 450 ppm NO (chemiluminescence method) with less than 1% contamination by other oxides of nitrogen. Concentrations of iNO and NO2 were measured with an electrochemical sensor calibrated against a reference NO tank.

Experimental design. Protocol 1. Effects of iNO treatment on pulmonary hemodynamics, gas exchange, and lung vascular endothelial permeability after 1 h (iNO n = 5, control n = 5) and 3 h (iNO n = 7, control n = 7) of mechanical ventilation in the preterm lamb. After at least 1 h of recovery and stabilization of physiologic parameters, baseline hemodynamics and arterial blood gas measurements were recorded. An endotracheal tube was placed by tracheostomy, and surfactant was instilled before the first mechanical ventilator breath. In the iNO treatment groups, NO(20 ppm) was administered throughout the 1- or 3-h study period. All animals were treated with Fio2 = 1.00 (due to blending of the NO mixture with inspired gases, measured Fio2 for the NO treatment group was 0.94).

Lung vascular permeability was measured using a modification of the method of Stelzner et al.(29) This method employs125 I-labeled albumin and 51Cr-tagged RBC to measure extravascular protein accumulation and provides a sensitive determination of lung vascular permeability, which is independent of pulmonary blood flow. Six milliliters of blood for radiolabeling were withdrawn from the fetus after hemodynamic monitoring catheters are placed, and 6 mL of 0.9 N saline solution were infused as replacement fluid. Red blood cells were labeled with 10 μCi of 51Cr. At 1 h before the end of the experiment, the labeled RBC and125 I-labeled albumin (40 μCi) were injected into a venous catheter. Lung blood volume was determined by simultaneous counting of the RBC activity of the lungs and a blood sample. At the end of the experiment, animals were killed with T-61 euthanasia solution (American Hoechst, Summerville, NJ). Blood was withdrawn for scintillation counting (to determine 51Cr and125 I counts) in an Auto-Gamma 5650 gamma counter employing a 3-inch through-hole NaI crystal. Fetal weight was recorded and appropriate catheter placement verified. The left lung was excised and divided into segments which were placed into preweighed scintillation vials for determination of51 Cr and 125I activity. The lung leak index was calculated in the following manner: 1) extravascular lung protein activity = whole lung protein activity - intravascular lung protein activity; 2) intravascular lung protein activity = (blood protein counts/blood weight)× (RBC counts in lung/RBC counts blood/blood weight); and 3) lung blood weight = RBC counts lung × (blood weight/RBC counts blood).

The results are then expressed as an index normalizing extravascular activity for intravascular protein activity and lung blood content, the protein leak index. Protein leak index = extravascular lung protein activity/blood protein activity (counts/g blood)/lung blood weight.

The lung leak index was also measured in two fetal animals at 115 d of gestation. After hysterotomy, a fetal hind limb was exposed and a catheter was placed in the saphenous vein for injection of 125I-labeled albumin and51 Cr-tagged RBC to measure extravascular protein accumulation, as described above. No other instrumentation was performed in these animals. One hour after injection of the isotopes, the animals were killed, and lung leak measurements were performed as described above. Lung wet/dry weight was measured using a segment of the right lower lobe, which was blotted dry and weighed (wet weight), then desiccated at 50 °C and weighed daily until three consecutive weights were unchanged (dry weight).

Protocol 2. Effects of iNO treatment on lung neutrophil accumulation during 4 h of mechanical ventilation (iNO treatment group, n = 4; controls, n= 4; fetal animals, n = 6). A separate group of animals was used in the second protocol to avoid radioisotope contamination of the MPO assay and to eliminate the potential effects of thoracotomy on lung neutrophil accumulation. After recovery from surgery, baseline arterial blood gas measurements were recorded (PAP and QLPA were not measured in protocol 2). An endotracheal tube was placed, surfactant instilled, and mechanical ventilation initiated (as described above). In the NO treatment group, iNO (5 ppm) was administered throughout the 4-h study period. All animals were treated with Fio2 = 1.00 (due to blending of the NO mixture with inspired gases, measured Fio2 for the NO treatment group was 0.99). After the study, animals were killed with T-61 euthanasia solution(American Hoechst, Summerville, NJ). Fetal weight was recorded, and appropriate catheter placement was verified. Fetal animals were killed before delivery by injection euthanasia solution into the umbilical vein through the hysterotomy.

A segment of the left lower lobe was placed in 10% formalin for subsequent histology. Tissue samples for light microscopy were stained with hematoxylin/eosin and naphthol AS-D chloroacetate solution for leukocyte esterase enhancement (Sigma Chemical Co. Diagnostics, St. Louis, MO).

A segment of the right lower lobe was snap frozen in liquid nitrogen for subsequent determination of MPO activity. MPO activity was determined in whole lung tissue using a RIA as previously described(30, 31). Lung tissue was homogenized for 30 s in 4 mL of 20 mmol/L potassium phosphate buffer, and centrifuged for 30 min at 40 000 × g, 4°C (Sorvall RC-5B centrifuge, Dupont Instruments Inc., Irving, TX). The pellet was resuspended in 4 mL of 50 mM potassium phosphate buffer, containing 0.5 g/dL centrimonium bromide. Resuspended pellets were frozen at -70 °C until the MPO assay was performed. Frozen samples were thawed, sonicated for 90 s at full power (ultrasonic cell disrupter; Kontes, Vineland, NJ), incubated in a 60 °C water bath for 2 h, and centrifuged for 10 min at maximum speed (Beckman Microfuge 12; Beckman Instruments, Irvine, CA). Supernatant (0.1 mL) was added to 2.9 mL of 50 mM potassium phosphate buffer, pH 6.0, containing 0.167 mg/mL o-dianisidine and 5 × 10-4 M hydrogen peroxide; absorbance of 460-nm visible light was measured for 3 min (Beckman DU7 spectrophotometer; Beckman Instruments). MPO activity per g of wet lung was then calculated as: MPO activity (units/g of wet lung) = (A460)(13.5)/lung weight (g); where A460 is the change in absorbance of 460 nm light from 1 to 3 min after the initiation of the reaction. The coefficient 13.5 was empirically determined such that 1 U of MPO activity is the amount of enzyme that will reduce 1 fmol peroxide/min(32).

Statistical analysis. Statistical comparisons of within-group continuous variables were performed using one-way repeated measures ANOVA. Where significant differences were identified, post hoc analysis was performed using Student-Neuman-Keull's test. The analysis of differences between the NO treatment and control groups was performed using two-way ANOVA for repeated measures (with time and treatment group identified as independent variables and the interaction between them analyzed). Comparisons of responses to each intervention (between treatment groups) at the hourly time points, wet/dry data and lung leak indexes were performed using the t test. The MPO data were analyzed using ANOVA, and post hoc analysis was performed using Student-Neuman-Keull's test. The level of statistical significance was set at p < 0.05; results are reported as mean± SE.

RESULTS

Protocol 1. Effects of iNO treatment on pulmonary hemodynamics, gas exchange, and lung vascular endothelial permeability after 1 and 3 h of mechanical ventilation. After 1 h of mechanical ventilation, there was no significant difference in hemodynamic measurements or gas exchange between the iNO and control groups (PVR = 0.16 ± 0.03 versus 0.23± 0.05 mm Hg·(mL/min)-1, Pao2 = 161 ± 35versus 161 ± 44 mm Hg; respectively). Wet/dry lung weight(9.06 ± 0.15 versus 9.33 ± 0.20) and lung leak index (Fig. 1) were not different between iNO treatment and control groups, respectively. Treatment with iNO for 3 h decreased PVR and improved oxygenation compared with controls (Table 1). Ventilator settings were not different between the iNO and control groups over the 3-h study period (PIP = 30.1 ± 1.0 cm H2O iNO versus 30.9 ± 1.1 cm H2O control, p = 0.6; mean airway pressure = 15.5 ± 1.4 cm H2O iNO versus 17.9 ± 0.7 cm H2O, p = 0.15). Wet/dry lung weight(8.03 ± 0.10 iNO versus 8.49 ± 0.16 control), extravascular lung water (8.46 ± 0.17 g/g of dry lung iNO versus 8.62 ± 0.22 g/g of dry lung control), and lung leak index (Fig. 1) were not different between the iNO treatment group and the control group. The value for fetal lung leak index was low (1.1 ± 0.2), in contrast to a >3-fold increase with ventilation.

Figure 1
figure 1

Lung leak index for 1- and 3-h mechanical ventilation studies in protocol 1. Proteinaceous pulmonary edema was similar between groups, despite a marked increase in pulmonary blood flow in the iNO-treated animals after 3 h of mechanical ventilation.

Full size image
Table 1 Aterial blood gas and hemodynamic data for 3-h studies in protocol 1; control group (n = 7) vs NO treatment group (n = 7)
Full size table

Protocol 2. Effects of iNO treatment on lung neutrophil accumulation during 4 h of mechanical ventilation. After 4 h of mechanical ventilation, Pao2 was higher in the iNO group compared with controls(119 ± 17 mm Hg versus 41 ± 5 mm Hg, p < 0.05). Lung neutrophil accumulation was apparent on lung histology in the control animals, as shown in Figure 2. Lung MPO activity increased from undetectable levels in normal fetal animals to 0.53 ± 0.20 U/g of lung in controls. Low dose iNO (5 ppm) markedly reduced MPO activity (0.11 ± 0.04 U/g of lung, p < 0.05versus control) (Fig. 3).

Figure 2
figure 2

Photomicrography of lung in a control (left) and NO-treated animal (right). Leukocyte esterase staining demonstrates marked neutrophil accumulation in air spaces and interstitium in a control animal, in contrast to the NO-treated lung (magnification,×20).

Full size image
Figure 3
figure 3

Lung MPO activity in premature lambs with severe HMD. Low dose iNO decreased neutrophil accumulation compared with controls.

Full size image

DISCUSSION

The susceptibility of the immature lung to injury from mechanical ventilation and hyperoxia compromises effective adjuvant therapies for HMD, such as surfactant replacement therapy. When injured, protein leak occurs from the vascular space into the alveoli, which decreases the surface-active properties of surfactant(4). This proteinaceous pulmonary edema causes atelectasis and ventilation/perfusion abnormalities, worsening gas exchange, and increasing PVR. In this study, we report that low dose iNO increases pulmonary blood flow and improves gas exchange in premature lambs with severe HMD, without increasing pulmonary edema, and decreases lung neutrophil accumulation.

The effects of low dose iNO on early neutrophil accumulation may have important clinical implications in HMD. There is recent evidence that endogenous NO production modulates neutrophil-endothelial cell interactions in the microcirculation(23), and the neutrophil plays an important role in the inflammatory cascade, which contributes to lung injury and the evolution of the most important sequel of HMD, chronic lung disease(7). Sequestration of neutrophils in the lung is an early step in a complex inflammatory response mediated through the elaboration of oxyradicals, proteases, phospholipases, and lipid compounds(12). Therapies that reduce neutrophil accumulation in the lung in HMD could potentially modify the early inflammatory process that amplifies acute lung injury and contributes to the development of chronic lung disease(33).

Conventional mechanical ventilation with high inspired oxygen concentrations after delivery of premature lambs at 115 d of gestation leads to increased PVR, decreased pulmonary blood flow, and worsening gas exchange, despite a good initial response to exogenous surfactant(28). Early and sustained treatment with low dose NO in this model attenuates the decline in gas exchange and pulmonary perfusion during mechanical ventilation(28, 34). After 3-4 h of mechanical ventilation in the 115-d premature lamb, low dose iNO increases pulmonary blood flow, without increasing pulmonary edema, and decreases lung neutrophil accumulation in severe experimental HMD.

Lung neutrophil accumulation is a critical component in lung injury. In this study, neutrophil sequenstration in the lung was quantitated using an assay for MPO activity. MPO activity provides a sensitive measurement of neutrophil accumulation in the lung, which correlates well with lung lavage and histopathologic measurements(33, 35–37). However, MPO activity does not necessarily reflect neutrophil activation. Neutrophil activation is a time-dependent phenomenon and was not determined because of the brief period of mechanical ventilation(33). The relatively short period of mechanical ventilation is a potential limitation of this study. The 115-d premature lamb has severe disturbances in gas exchange despite surfactant treatment, thus limiting the duration of ventilation. In studies of more mature lambs (127-133 d of gestation) mechanically ventilated for 6-8 h, vascular endothelial protein leak was greater in nonsurfactant-treated animals with HMD when compared with controls(38, 39). It is possible that the period of mechanical ventilation in our experiments was too brief to determine differences in lung leak and that iNO treatment at low doses could decrease lung leak over more prolonged periods of ventilation in animals with less severe disease. Pulmonary blood flow was not significantly different between the groups until 3 h of mechanical ventilation, and differences in pulmonary edema may not have been apparent over this time period. However, despite the markedly edematous lung of the 115-d gestation lamb, low dose iNO increased pulmonary blood flow by 3 h of mechanical ventilation and decreased lung neutrophil accumulation.

We speculate that low dose iNO may attenuate the early inflammatory cascade induced by pulmonary leukosequestration in severe HMD, thus favorably modulating one important factor contributing to the development of chronic lung disease in prematurity. However, the effects of iNO treatment on chronic lung disease in infants with HMD is unknown and requires randomized, controlled, clinical trials.