Hypoxic Gene Expression of Donor Bronchi Linked to Airway Complications after Lung Transplantation

Abstract

Rationale: Central airway stenosis (CAS) after lung transplantation has been attributed in part to chronic airway ischemia; however, little is known about the time course or significance of large airway hypoxia early after transplantation.

Objectives: To evaluate large airway oxygenation and hypoxic gene expression during the first month after lung transplantation and their relation to airway complications.

Methods: Subjects who underwent lung transplantation underwent endobronchial tissue oximetry of native and donor bronchi at 0, 3, and 30 days after transplantation (n = 11) and/or endobronchial biopsies (n = 14) at 30 days for real-time polymerase chain reaction of hypoxia-inducible genes. Patients were monitored for 6 months for the development of transplant-related complications.

Measurements and main results: Compared with native endobronchial tissues, donor tissue oxygen saturations (Sto2) were reduced in the upper lobes (74.1 ± 1.8% vs. 68.8 ± 1.7%; P < 0.05) and lower lobes (75.6 ± 1.6% vs. 71.5 ± 1.8%; P = 0.065) at 30 days post-transplantation. Donor upper lobe and subcarina Sto2 levels were also lower than the main carina (difference of -3.9 ± 1.5 and -4.8 ± 2.1, respectively; P < 0.05) at 30 days. Up-regulation of hypoxia-inducible genes VEGFA, FLT1, VEGFC, HMOX1, and TIE2 was significant in donor airways relative to native airways (all P < 0.05). VEGFA, KDR, and HMOX1 were associated with prolonged respiratory failure, prolonged hospitalization, extensive airway necrosis, and CAS (P < 0.05).

Conclusions: These findings implicate donor bronchial hypoxia as a driving factor for post-transplantation airway complications. Strategies to improve airway oxygenation, such as bronchial artery re-anastomosis and hyperbaric oxygen therapy merit clinical investigation.

Keywords: angiogenic proteins; bronchial diseases; cell hypoxia; lung transplantation; oximetry.

Figure 1.

Absolute tissue oxygen saturations (Sto₂) of native (open circles) and donor (solid circles) (A) upper lobe, (B) first subcarina, and (C) lower lobe bronchi in individual patients at 0, 3, and 30 days after single or bilateral lung transplantation. N = 6–11 patients per group per time point. Bars represent means. *P < 0.05 calculated by the unpaired Student’s t test. There was also a trend for reduced absolute Sto₂ levels in donor lower lobe bronchi at 30 days (71.5 ± 1.8%, n = 11 vs. 75.6 ± 1.6%, n = 8; P = 0.065).

Figure 2.

Differential endobronchial tissue oxygen saturation (Sto₂). (A) Sto₂ levels were measured in single orthotopic lung transplantation (SOLT) patients only. Paired measurements (SOLT donor lung vs. native contralateral lung) were collected from donor and native upper lobes, subcarinae, and lower lobes at 0 (black bars), 3 (gray bars), and 30 (white bars) days after single lung transplantation. The contralateral Sto₂ difference (donor – native) was calculated for upper lobe, subcarina, and lower lobe sites and analyzed for statistical significance (*P < 0.05) by the Wilcoxon signed-rank test against a hypothetical value of 0. The contralateral Sto₂ difference was significantly less than 0 in the upper lobes at 0 and 30 days, and there was a trend in the lower lobes at 30 days (−2.9 ± 1.3, n = 8; P = 0.069). N = 5–8 patients per group per time point. Bars represent mean ± SEM. (B) Sto₂ levels were measured in both SOLT and bilateral orthotopic lung transplantation (BOLT) patients. Paired measurements were collected in each patient at the native main carina (above the anastomosis) and the donor upper lobe, subcarina, and lower lobe tissues (below the anastomosis) at 0 (black bars), 3 (gray bars), and 30 (white bars) days after SOLT or BOLT. The anastomosis Sto₂ difference (donor – main carina) was calculated for upper lobe, subcarina, and lower lobe tissues and analyzed by the Wilcoxon signed-rank test against a hypothetical value of 0. Differences were significantly less than 0 in the upper lobes and subcarinae at 0 and 30 days post-transplantation (*P < 0.05). N =  7–11 patients per group. Bars represent mean ±  SEM.

Figure 3.

Real-time polymerase chain reaction of hypoxia-inducible genes in native (open circles) and donor (solid circles) endobronchial tissues of individual subjects. N = 5–14 per group. Bars represent median. *P < 0.05 is calculated by the Wilcoxon signed-rank test (VEGF-A, FLT1, KDR, and HMOX1), or by the Mann-Whitney U test if transcript data were missing (TIE2 and VEGF-C). Because of limited RNA sample, not every transcript could be measured for every patient. See Figure E2 for additional genes tested.

Figure 4.

Median ratios of donor to native gene expression for paired samples. N = 5–14 per group. Bars represent interquartile range. *P < 0.05 is calculated by the Wilcoxon signed-rank test against a hypothetical value of 1. There was also a trend for increased TIE2 (TEK) and VEGFC expression (both P = 0.062).

Figure 5.

Comparison of donor/native gene expression ratios and clinical outcomes. Gene expression ratios in patients who developed prolonged respiratory failure (gray bars, n = 3–5) or did not (black bars, n = 2–9) (A), with time to hospital discharge (DC) ≥15 days (gray bars, 4–8) or <15 days (black bars, 1–6) (B), with extensive airway necrosis (gray bars, n = 2–5) or no/limited airway necrosis (black bars, n = 2–9) (C), and with central airway stenosis (gray bars, n = 1–2) or no stenosis (black bars, n = 4–12) (D). VEGFC and TGFB1 polymerase chain reactions were not performed in the patients that developed central airway stenosis owing to lack of remaining cDNA sample. Bars represent mean ± SEM. Patients with prolonged hospital stays after transplantation also displayed trends for higher VEGFA (3.1 ± 0.7, n = 8, vs. 1.5 ± 0.4, n = 6, P = 0.076) and KDR (VEGFR2) expression (3.4 ± 1.3, n = 5, vs. 0.7 ± 0.08, n = 4, P = 0.10). n represents number of transcripts measured per patient and ranges from 1 to 12 per group. Because of limited RNA sample, not every transcript could be measured for every patient. *P < 0.05 is calculated by the unpaired Student’s t test.