Chronic lung injury in preterm lambs: abnormalities of the pulmonary circulation and lung fluid balance.

Chronic lung disease of early infancy, or bronchopulmonary dysplasia, is a frequent complication of prolonged mechanical ventilation after premature birth. Pulmonary hypertension and edema are common features of this condition, which is often attributed to long-term, repetitive overinflation of incompletely developed lungs. The overall objective of this work was to examine the effects on the pulmonary circulation and lung fluid balance of different ventilation strategies using large versus small inflation volumes in an animal model of bronchopulmonary dysplasia. We studied 16 newborn lambs that were delivered prematurely (124+/-3 d gestation, term = 147 d) by cesarean section and mechanically ventilated for 3 to 4 wk. Ten lambs were ventilated at 20 breaths/min, yielding a tidal volume of 15+/-5 mL/kg, and six lambs were ventilated at 60 breaths/min, yielding a tidal volume of 6+/-2 mL/kg. All lambs received surfactant at birth and had subsequent surgery for closure of the ductus arteriosus and catheter placement to allow serial measurements of pulmonary vascular resistance and lung lymph flow. Chronic lung injury, documented by serial chest radiographs and postmortem pathologic examination, developed in all lambs irrespective of the pattern of assisted ventilation. Pulmonary vascular resistance, which normally decreases during the month after birth at term, did not change significantly from the first to the last week of study. Lung lymph flow, an index of net transvascular fluid filtration, increased with time in lambs that were ventilated at 20 breaths/min, but not in lambs ventilated at 60 breaths/min. Lymph protein concentration decreased with time, indicative of increased fluid filtration pressure, without evidence of a change in lung vascular protein permeability. Postmortem studies showed interstitial lung edema, increased pulmonary arteriolar smooth muscle and elastin, decreased numbers of small pulmonary arteries and veins, and decreased capillary surface density in distal lung of chronically ventilated lambs compared with control lambs that were killed either 1 d (same postconceptional age) or 3 wk (same postnatal age) after birth at term. Thus, chronic lung injury from prolonged mechanical ventilation after premature birth inhibits the normal postnatal decrease in pulmonary vascular resistance and leads to lung edema from increased fluid filtration pressure. These abnormalities of the pulmonary circulation may contribute to the abnormal respiratory gas exchange that often exists in infants with bronchopulmonary dysplasia.

Chronic lung injury in preterm lambs. Disordered respiratory tract development.

The cause of chronic lung disease of early infancy, often called bronchopulmonary dysplasia (BPD), remains unclear, partly because large-animal models that reliably reproduce BPD have not been available. We developed a model of BPD in lambs that are delivered prematurely and ventilated for 3 to 4 wk after birth to determine whether the histopathology of chronic lung injury in premature lambs mimics that which occurs in preterm infants who die with BPD, and to compare two ventilation strategies to test the hypothesis that differences in tidal volume (VT) influence histopathologic outcome. The two ventilation strategies were slow, deep ventilation (20 breaths/min, 15 +/- 2 ml/kg body weight VT; n = 5) or rapid, shallow ventilation (60 breaths/min, 6 +/- 1 ml/kg body weight VT; n = 5). Lambs were delivered at 125 +/- 4 d gestation (term = 147 d), treated with surfactant, and mechanically ventilated with sufficient supplemental oxygen to maintain normal arterial oxygenation (60 to 90 mm Hg). Quantitative histologic analysis revealed lung structural abnormalities for both groups of experimental lambs compared with lungs of control term lambs that were < 1 d old (matched for developmental age; n = 5) or 3 to 4 wk old (matched for postnatal age; n = 5). Compared with control lambs, chronically ventilated preterm lambs had pulmonary histopathology characterized by nonuniform inflation patterns, impaired alveolar formation, abnormal abundance of elastin, increased muscularization of terminal bronchioles, and inflammation and edema. Slow, deep ventilation was associated with less atelectasis, less alveolar formation, and more elastin when compared with rapid, shallow ventilation. We conclude that prolonged mechanical ventilation of preterm lambs disrupts lung development and produces pulmonary histopathologic changes that are very similar to those that are seen in the lungs of preterm infants who die with BPD. This chronic lung disease is not prevented by surfactant replacement at birth, does not appear to require arterial hyperoxia, and is influenced by VT.

Chronic lung injury in preterm lambs: disordered pulmonary elastin deposition.

Prolonged mechanical ventilation of premature neonates is often associated with abnormal morphological development of the lung and chronic lung disease, sometimes called bronchopulmonary dysplasia (BPD). Impaired alveolar development is a hallmark of this disease. To better understand the effects of mechanical ventilation on lung elastin expression, we studied lung tissue from 10 preterm lambs (gestation = 125 days; term = 148 days) mechanically ventilated for 3-4 wk at a respirator rate of 20 breaths/min and tidal volume of 15 +/- 5 ml/kg (n = 5) or 60 breaths/min and tidal volume of 5 +/- 2 ml/kg (n = 5). Histopathology showed increased elastin accumulation and abnormal morphological development in the ventilated groups. Postmortem lung desmosine content was increased significantly in the 20 breaths/min group. Tropoelastin mRNA expression was increased in both ventilated groups. In situ hybridization localized increased tropoelastin mRNA expression to sites of accumulated elastin in extended alveolar walls with scant, attenuated secondary crests. Lung collagen content, as assessed by the amount of hydroxyproline in lung tissue, was similar to controls. These data suggest that excessive production and accumulation of elastin is associated with chronic lung injury from prolonged mechanical ventilation after premature birth.

Methods for retrovirus-mediated gene transfer to fetal lung.

With the advent of improved vectors for DNA delivery, somatic gene therapeutic approaches have expanded rapidly in the last few years. The vast majority of applications include ex vivo and in vivo protocols in patients postnatally. Nonetheless there is increasing interest and compelling reasons to consider prenatal application of somatic gene therapy (1,2. In the current chapter, we will review theoretical, ethical, and experimental support for in utero gene therapy and then outline the methodology and large animal model we are currently using to consider retrovirus-mediated gene transfer to fetal lung. In this latter regard, the candidate inherited disorder is cystic fibrosis and the reader is referred to Chapters 1 and 12 in this volume.

Retrovirus-mediated gene transfer in lungs of living fetal sheep.

In utero somatic gene transfer may be a useful therapeutic strategy for a variety of inherited disorders. In the present study, we demonstrate transgene expression in the airways of fetal lamb lungs, 2-3 weeks after injection of Moloney murine leukemia retrovirus based vectors containing cDNA for beta-galactosidase (lacZ) or human interleukin receptor antagonist protein (IRAP), into the fluid filled future airspace of fully catheterized twin fetal lambs (104-117 days gestational age; term 147 days). Expression of lacZ or IRAP was limited to the twin that received the respective vector and was apparent, at light microscopic level, in the epithelium and submucosal space of proximal airways, and to a lesser extent, in the respiratory epithelium of the distal airways. These data demonstrate for the first time that transfer of foreign DNA to fetal lung can be accomplished. These findings support the use of retroviral vectors for somatic lung DNA transfer and suggest that inherited disorders such as cystic fibrosis may be approached therapeutically via gene transfer, in utero.

Developmental changes in lung epithelial ion transport and liquid movement.

During fetal life, the lungs are filled with liquid that flows from the pulmonary circulation across the epithelium in response to the osmotic force generated by Cl- secretion of airway and distal lung epithelial cells. As birth approaches, net Cl- secretion across the respiratory tract epithelium decreases, and this is associated with a reduction in the flow of liquid into the lung lumen. The cause for this change is unknown, but several recent studies indicate that it may be related to alterations in the hormonal milieu to which the lung epithelium is exposed late in gestation. The switch from placental to pulmonary gas exchange at birth requires rapid removal of liquid from the lung lumen. During labor and the immediate postnatal period, the pulmonary epithelium changes from a predominantly Cl-(-)secreting membrane to a predominantly Na(+)-absorbing membrane, with resultant reversal of the direction of flow of lung liquid. There is considerable evidence that this change reflects an active metabolic process involving increased Na(+)-K(+)-ATPase activity in lung epithelial cells, which drives liquid from the lung lumen into the interstitium, with subsequent absorption into the pulmonary circulation. This Na(+)-K(+)-ATPase-dependent process persists in the bronchopulmonary epithelium of the mature lung and probably has an important role in clearance of alveolar edema associated with heart failure or lung injury.

Developmental differences in rabbit lung epithelial cell Na(+)-K(+)-ATPase.

Previous studies showed that ouabain-sensitive rubidium (86Rb+) uptake by rabbit lung epithelial cells increases at birth, followed by a more gradual postnatal increase, reaching adult values by age 30 days. To see whether these changes in ouabain-sensitive cation transport were the result of changes in Na+ pump number or turnover rate, we measured binding of [3H]ouabain and ouabain-sensitive 86Rb+ uptake by freshly isolated lung epithelial cells harvested from near-term fetal, newborn, and adult rabbits. Ouabain-sensitive 86Rb+ uptake by fetal cells was 1/4 that of newborn cells and 1/10 that of adult cells. The maximal number of ouabain binding sites (Umax) was the same for fetal and newborn cells but almost threefold greater for adult cells. Na+ pump turnover rate, determined from ouabain-sensitive 86Rb+ uptake and Umax, was four times greater in newborn and adult cells than it was in fetal cells. Thus the increase in 86Rb+ uptake at birth could be explained by an increase in Na(+)-K(+)-adenosinetriphosphatase (ATPase) turnover rate, whereas the postnatal increase in 86Rb+ uptake could be accounted for by an increase in the amount of Na(+)-K(+)-ATPase per cell.

Cation transport in lung epithelial cells derived from fetal, newborn, and adult rabbits.

Recent studies done with fetal and adult sheep and with monolayers of cultured rat alveolar type II cells suggest that active transport of Na+ across the lung epithelium may contribute to liquid absorption from air spaces, an essential component of the normal switch from placental to pulmonary gas exchange at birth. The goals of this work were 1) to study the ontogeny of cation transport in lung epithelial cells derived from fetal, newborn, and adult rabbits and 2) to determine the influence of premature birth, air breathing, labor, and postnatal lung maturation on K+ uptake in these cells. We harvested granular pneumonocytes by tracheal instillation of proteolytic enzymes followed by centrifugation of the dispersed cells over a discontinuous density gradient of metrizamide. This procedure yielded 65-90% granular pneumonocytes, of which more than 80% excluded vital dye. Using freshly isolated cells, we measured uptake of 86Rb+, which mimics transmembrane movement of K+, in the presence or absence of 10(-4) M ouabain and in the presence or absence of 5 X 10(-4) M furosemide or bumetanide. In adult rabbit studies, 86Rb+ uptake was twice as fast in lung epithelial cells (98 +/- 7 nmol X 10(6) cells-1 X h-1) as it was in alveolar macrophages (51 +/- 6 nmol X 10(6) cells-1 X h-1). Ouabain inhibited 55-60% of the uptake by pneumonocytes, and "loop" diuretics inhibited an additional 15-20%. The rate of 86Rb+ uptake in fetal cells was less than 10% (6 +/- 1 nmol X 10(6) cells-1 X h-1) of the rate in adult cells; ouabain inhibited 80-85% of 86Rb+ uptake in fetal cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Common physiologic patterns in general surgical patients: hemodynamic and oxygen transport changes during and after operation in patients with and without associated medical problems.

An examination of perioperative cardiorespiratory parameters in surviving and nonsurviving patients has identified several physiologic responses to the stress of surgical trauma, namely enhancement of circulatory performance and oxygen transport. When specific subsets of surgical illness are examined, it was found that these responses often began from different plateaus of baseline function. Because these responses are greater in survivors than nonsurvivors and this relationship is fairly consistent among a variety of surgical illnesses, it would seem that duplicating or augmenting these responses would be a rational goal for therapeutic intervention that would lead to improved patient outcome.

Probability of survival as a prognostic and severity of illness score in critically ill surgical patients.

A new quantitative method for measuring the prognosis and severity of illness in terms of probability of survival was developed from 224 studies in an index population of 220 critically ill surgical patients. Patients were selected preoperatively to eliminate pre-existing cardiac disease, cirrhosis, nutritional debility, shock or sepsis, in order to evaluate the physiologic relationships of surgical trauma to outcome free of confounding associated medical disorders. The empirically derived numeric severity index was calculated from the probability of survival for each of 28 hemodynamic and oxygen transport variables at each time period after surgery. The score correctly indicated patient outcome in 96% of the index population and 94% of an independent, prospective population. The survivors' score consistently predicted survival within 21.6 +/- 4.4 (SEM) h after the end of surgery. The severity score of those who died consistently predicted nonsurvival within 37 +/- 11 (SEM) h after the end of surgery. We conclude that the score provides a useful, objective, physiologic measure of the severity of illness and prognosis.

Cardiorespiratory effects of pneumatic trousers in critically ill patients.

Although pneumatic antishock trousers (PT) are widely used in prehospital and emergency care, little is known about their cardiorespiratory effects in critically ill patients. To examine this issue, we measured hemodynamic and oxygen metabolism variables in ten critically ill patients. All patients were studied with PTs uninflated, after five minutes of PT inflation to 40 mm Hg, and five minutes after PT deflation. Significant increase in mean arterial pressure, systemic vascular resistance, and pulmonary artery pressures were present after PT inflation. No significant changes in cardiac index, stroke index, arterial or mixed venous blood gas values, or oxygen delivery were found. There was a downward trend in VO2 that was on the border of statistical significance. Regression analysis of cardiorespiratory variables on blood volume demonstrated no physiologic effects of external counterpressure in hypovolemic, hypervolemic, or normovolemic patients. We concluded that PT inflation increases BP through its effects on peripheral resistance. No significant autotransfusion effect was present, and there was a suggestive impairment in oxygen metabolism.

Sequential hemodynamic and oxygen transport abnormalities in patients with acute pancreatitis.

Physiologic abnormalities were evaluated by sequential hemodynamic and oxygen transport measurements in 33 patients with acute pancreatitis. The hypotensive crisis, which was defined as the lowest mean arterial pressure (MAP), was used as the common temporal reference point to align the data in a coherent fashion. The data of the 48-hour periods before and after the hypotensive crisis then were evaluated. The patients were divided into normotensive and hypotensive groups and the latter were divided into survivors and nonsurvivors to evaluate the severity of the disorder and to identify the patterns representative of survival and death. The physiologic abnormalities of the hypotensive patients include decreased systemic vascular resistance index (SVRI) and compromised cardiac function. The latter was demonstrated during the hypotensive episode by significantly reduced left ventricular stroke work index (LSWI), despite increases in heart rate (HR), central venous pressure (CVP), pulmonary capillary wedge pressure (WP), and cardiac index (CI). The normotensive group had increased oxygen consumption (Vo2), oxygen delivery (Do2), pulmonary shunt (Qsp/Qt), LSWI, normal SVRI, and high CI. The hypotensive nonsurvivors had lower MAP, LSWI, SVRI, Do, and hematocrit as well as higher pulmonary vascular resistance index (PVRI) and Qsp/Qt than did the survivors. These findings do not support myocardial depression as the primary cardiovascular abnormality in acute pancreatitis, but rather suggest the decreased vascular tone from flow maldistribution in the peripheral microcirculation limits tissue oxygenation in the face of increased metabolic requirements of the hypercatabolic state.

Edema formation in the lungs and its relationship to neonatal respiratory distress.

Pulmonary edema is an important feature of many newborn lung diseases, including respiratory distress from severe perinatal asphyxia, heart failure, hyaline membrane disease, pneumonitis from group B beta-hemolytic streptococcus, and chronic lung disease (bronchopulmonary dysplasia). Neonatal pulmonary edema often results from increased filtration pressure in the microcirculation of the lungs. This occurs during sustained hypoxia, in left ventricular failure associated with congenital heart disease or myocardial dysfunction, following excessive intravascular infusions of blood, colloid, fat, or electrolyte solution, and in conditions that increase pulmonary blood flow. Low intravascular protein osmotic pressure from hypoproteinemia may predispose infants to pulmonary edema. Hypoproteinemia is common in infants who are born prematurely. Large intravascular infusions of protein-free fluid further decrease the concentration of protein in plasma and thereby facilitate edema formation. Lymphatic obstruction by air (pulmonary interstitial emphysema) or fibrosis (long-standing lung disease) also may contribute to the development of edema. Bacteremia, endotoxemia, and prolonged oxygen breathing injure the pulmonary microvascular endothelium and cause protein-rich fluid to accumulate in the lungs. The risk of neonatal pulmonary edema can be reduced by several therapeutic measures designed to lessen filtration pressure, increase plasma protein osmotic pressure, and prevent or reduce the severity of lung injury.

Edema formation in the newborn lung.

Pulmonary edema is an important cause of respiratory distress in newborn infants. It occurs with severe perinatal asphyxia, heart failure, hyaline membrane disease, persistent patency of the ductus arteriosus, pneumonitis from group B beta-hemolytic streptococcus, and chronic lung disease (bronchopulmonary dysplasia). Neonatal pulmonary edema often develops from increased pressure in the microcirculation of the lungs. This may occur in conjunction with sustained hypoxia; left ventricular failure associated with congenital heart disease or myocardial dysfunction; following excessive intravascular infusions of blood, colloid, fat, or electrolyte solution and in conditions that increase pulmonary blood flow. Low intravascular protein osmotic pressure from hypoproteinemia may predispose infants to pulmonary edema. Hypoproteinemia is common in infants who are born prematurely. Large intravascular infusions of protein-free fluid further decrease the concentration of protein in plasma and thereby facilitate edema formation. Lymphatic obstruction by air (pulmonary interstitial emphysema of fibrosis (chronic lung disease) also may contribute to the development of edema. Bacteremia, endotoxemia, and prolonged oxygen-breathing injure the pulmonary microvascular endothelium and cause protein-rich fluid to accumulate in the lungs. Epithelial protein leaks may develop when the transpulmonary pressure needed to inflate the lungs increases because of high surface tension at the air-liquid interface. Fibrin clots from in some of the air spaces, which in combination with atelectasis and edema constitute the pathologic features of hyaline membrane disease. The risk of neonatal pulmonary edema can be reduced by several therapeutic measures designed to lessen fluid filtration pressure, increase plasma protein osmotic pressure, and prevent or reduce the severity of lung injury.

Lung fluid dynamics in awake newborn lambs.

We measured steady-state lung lymph flow, lymph protein flow, and simultaneous pulmonary vascular pressures in 12 1-wk-old unanesthetized lambs and compared these measurements to those of previous studies, performed under similar conditions, on nine awake adult sheep. The purpose of these experiments was to compare newborn and adult sheep with respect to transvascular filtration of fluid and microvascular permeability to plasma proteins. We prepared the lambs surgically to isolate and collect lung lymph and measure average pulmonary arterial and left atrial pressures, allowing at least 2 days for the lambs to recover from surgery before studies began. Lambs had higher pulmonary arterial and left atrial pressures, lower lymph and plasma protein concentrations, and 57% more lymph flow per gram of dry bloodless lung than sheep; the difference in protein flow between lambs and sheep was not significant. Protein concentration in lymph relative to that in plasma was significantly lower in lambs than in sheep; but the ratio of albumin concentration to globulin concentration in both lymph and plasma was almost identical in the two groups of animals. Extravascular lung water per gram of dry bloodless lung was greater in lambs (4.82+/-0.11 g) than in sheep (4.45+/-0.08 g), but there was no histologic evidence of pulmonary edema in either group of animals. These findings suggest that lambs have more transvascular filtration of fluid per unit lung mass than sheep, but that microvascular sites for protein exchange do not differ appreciably in lambs and sheep. To test this conclusion, we measured steady-state lymph flow in three lambs before and after raising pulmonary microvascular pressure by rapid intravenous infusion of saline. Lymph flow increased as a function of the net transvascular driving pressure (hydraulic pressure gradient-protein osmotic pressure gradient). This response was almost identical to that of four sheep with pulmonary microvascular pressure augmented by inflation of a balloon in the left atrium. In eight lambs we measured the time for intravenously injected (125)I-albumin to equilibrate in lymph at half the specific activity of plasma: the protein tag equilibrated faster than in sheep. This difference could be explained partly by the higher pulmonary arterial and left atrial pressures of lambs than sheep, and possibly by the presence of more microvascular sites for protein exchange relative to the volume of distribution of protein in the lung of the younger animals.

Effect of acute hemorrhagic shock on pulmonary microvascular fluid filtration and protein permeability in sheep.

In an attempt to further delineate the pathophysiology of the shock-lung syndrome, we studied the effect of hemorrhagic shock on the pulmonary microcirculation in the adult sheep with lung lymph flow and lymph protein transport as indices of fluid filtration and vascular permeability to protein. We noted that lymph flow remained at baseline levels during shock, despite significant decreases in pulmonary artery and left atrial pressures. This suggests that microvascular pressure is maintained by an increase in pulmonary venous resistance. We demonstrated that lymph protein transport does not increase during shock, which indicates that a change in pulmonary vascular permeability to protein does not occur.