Growth of pulmonary microvasculature in ventilated preterm infants

Abstract

Rationale: Density-based morphometric studies have demonstrated decreased capillary density in infants with bronchopulmonary dysplasia (BPD) and in BPD-like animal models, leading to the prevailing view that microvascular development is disrupted in BPD.

Objective: To perform a comprehensive analysis of the early and late effects of ventilation on pulmonary microvascular growth in preterm infants.

Methods: Postmortem lung samples were collected from ventilated preterm infants who died between 23 and 29 wk ("short-term ventilated") or between 36 and 39 wk ("long-term ventilated") corrected postmenstrual age. Results were compared with age-matched infants or stillborn infants ("early" and "late" control subjects). Microvascular growth was studied by anti-platelet endothelial cell adhesion molecule (PECAM)-1 immunohistochemistry, quantitative stereology, analysis of endothelial cell proliferation, and Western blot analysis of pulmonary PECAM-1 protein levels.

Measurements: Measurements were made of capillary density, volume of air-exchanging parenchyma, volume of microvascular endothelial cells, Ki67 labeling index of endothelial cells, and PECAM-1/actin protein levels.

Main results: Lungs of long-term ventilated infants showed a significant (more than twofold) increase in volume of air-exchanging parenchyma and a 60% increase in total pulmonary microvascular endothelial volume compared with late control subjects, associated with 60% higher pulmonary PECAM-1 protein levels. The marked expansion of the pulmonary microvasculature in ventilated lungs was, at least partly, attributable to brisk endothelial cell proliferation. The microvasculature of ventilated lungs appeared immature, retaining a saccular architectural pattern.

Conclusions: The pulmonary microvasculature of ventilated preterm infants displayed marked angiogenesis, nearly proportionate to the growth of the air-exchanging lung parenchyma. These results challenge the paradigm of microvascular growth arrest as a major pathogenic factor in BPD.

Figure 1.

Lung histology. (A) Early control lung showing large-sized, simple airspaces, relatively wide septa, and focal early secondary crest formation (asterisks), characteristic of late canalicular/early saccular stage of lung development (infant born at 24 wk gestation, lived for 2 h). (B) Short-term ventilated lung showing widening and increased cellularity of the septa, as well as focal hemorrhages within the air spaces (infant born at 23 wk, lived for 7 d, ventilated). (C) Late control lung showing complex gas-exchanging parenchyma with abundant secondary crests (asterisks) and thin alveolar septa, consistent with late saccular/early alveolar stage of lung development (stillborn at 38 wk). (D) Long-term ventilated lung showing simple, large-sized air spaces with hypercellular and thickened septa (infant born at 27 wk, lived for 12 wk, ventilated). Hematoxylin–eosin staining; original magnification, ×400.

Figure 2.

PECAM-1 immunohistochemistry. (A) Early control lung showing abundant capillary structures within the septa, often arranged in a double subepithelial capillary network (infant born at 24.5 wk gestation, lived for minutes). (B) Short-term ventilated lung showing abundant capillary structures of varying sizes randomly scattered within the widened septa (infant born at 23.5 wk, lived for 6 d, ventilated). (C) Late control lung showing a complex capillary pattern, characterized by a double or focally single capillary network with numerous outsproutings toward secondary crests (infant born at 36 wk, lived for 3 d). (D) Long-term ventilated lung showing abundant, intensely immunoreactive capillary structures within the thickened septa. The capillaries are prominently subepithelial, mostly arranged in a dual parallel pattern, and show few branching points (infant born at 26 wk, lived for 12 wk, ventilated, clinical diagnosis of BPD). PECAM-1 immunohistochemistry; 3,3′-diaminobenzidine tetrachloride (DAB) with hematoxylin counterstain; original magnification, ×400.

Figure 3.

Ki-67 immunolabeling. (A) Early control lung showing scattered Ki-67–positive nuclei within the interstitium and in adlumenal (epithelial) cells (infant born at 24 wk, lived for 2 h). (B) Short-term ventilated lung showing markedly increased numbers of Ki-67–positive nuclei, especially within the interstitium (infant born at 23 wk, lived for 7 d, ventilated). (C) Late control lung showing relatively few Ki-67–positive nuclei in interstitium and epithelium (stillborn at 36 wk). (D) Long-term ventilated lung showing more Ki-67–positive nuclei than late control lungs, evenly distributed over interstitium and epithelium (infant born at 27 wk, lived for 11 wk, ventilated). (E) Lung proliferative index of early control, early (short-term) ventilated, late control, and late (long-term) ventilated infants. Values represent means ± SD of at least five patients per group. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control; **p < 0.01 versus control. Ki-67 immunohistochemistry; DAB with hematoxylin counterstain; original magnification, ×400.

Figure 3.

Ki-67 immunolabeling. (A) Early control lung showing scattered Ki-67–positive nuclei within the interstitium and in adlumenal (epithelial) cells (infant born at 24 wk, lived for 2 h). (B) Short-term ventilated lung showing markedly increased numbers of Ki-67–positive nuclei, especially within the interstitium (infant born at 23 wk, lived for 7 d, ventilated). (C) Late control lung showing relatively few Ki-67–positive nuclei in interstitium and epithelium (stillborn at 36 wk). (D) Long-term ventilated lung showing more Ki-67–positive nuclei than late control lungs, evenly distributed over interstitium and epithelium (infant born at 27 wk, lived for 11 wk, ventilated). (E) Lung proliferative index of early control, early (short-term) ventilated, late control, and late (long-term) ventilated infants. Values represent means ± SD of at least five patients per group. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control; **p < 0.01 versus control. Ki-67 immunohistochemistry; DAB with hematoxylin counterstain; original magnification, ×400.

Figure 4.

PECAM-1 and Ki-67 double immunofluorescence. (A) Early control lung showing Ki-67–positive endothelial cells (arrows) and nonendothelial cells (same infant as in Figure 3A). (B) Short-term ventilated lung showing overall increase in Ki-67 labeling, both in PECAM-positive endothelial cells (arrows) and in nonendothelial cells (same infant as in Figure 3B). (C) Late control lung showing relatively low proliferative activity. Shown is a single nonendothelial Ki-67–positive cell. Inset: Focal intense endothelial cell proliferation associated with secondary crest formation (same stillborn as in Figure 3C). (D) Long-term ventilated lung showing increased Ki-67 labeling compared with late control, both in endothelial cells (arrows) and in nonendothelial cells (same infant as in Figure 3D). (E) Endothelial cell proliferative index. Values represent means ± SD of at least five patients per group. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control; **p < 0.01 versus control. Ki-67 (fluorescein isothiocyanate, green) and PECAM-1 (Cy3, red) double immunofluorescence; original magnification, ×400.

Figure 4.

PECAM-1 and Ki-67 double immunofluorescence. (A) Early control lung showing Ki-67–positive endothelial cells (arrows) and nonendothelial cells (same infant as in Figure 3A). (B) Short-term ventilated lung showing overall increase in Ki-67 labeling, both in PECAM-positive endothelial cells (arrows) and in nonendothelial cells (same infant as in Figure 3B). (C) Late control lung showing relatively low proliferative activity. Shown is a single nonendothelial Ki-67–positive cell. Inset: Focal intense endothelial cell proliferation associated with secondary crest formation (same stillborn as in Figure 3C). (D) Long-term ventilated lung showing increased Ki-67 labeling compared with late control, both in endothelial cells (arrows) and in nonendothelial cells (same infant as in Figure 3D). (E) Endothelial cell proliferative index. Values represent means ± SD of at least five patients per group. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control; **p < 0.01 versus control. Ki-67 (fluorescein isothiocyanate, green) and PECAM-1 (Cy3, red) double immunofluorescence; original magnification, ×400.

Figure 5.

Western blot analysis of pulmonary PECAM-1 protein levels. (A) Western blot analysis of PECAM-1 protein expression in whole lung homogenates. Bands appropriately sized for PECAM-1 (∼ 100 kD) were detected. Actin (42 kD) served as internal loading control. (B) Densitometry of PECAM-1 Western blot analysis. Values represent means ± SE. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control. IOD = integrated optical density.

Figure 5.

Western blot analysis of pulmonary PECAM-1 protein levels. (A) Western blot analysis of PECAM-1 protein expression in whole lung homogenates. Bands appropriately sized for PECAM-1 (∼ 100 kD) were detected. Actin (42 kD) served as internal loading control. (B) Densitometry of PECAM-1 Western blot analysis. Values represent means ± SE. Dotted bars, control; black bars, ventilated. *p < 0.05 versus control. IOD = integrated optical density.