|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From INSERM U408 (Dr. Stern), INSERM U426 (Dr. Silve), Service de Réanimation Chirurgicale (Drs. Paugam and Mantz), Service de Pneumologie (Drs. Aubier and Crestam), Hôpital Bichat, and Laboratoire de Biologie Endocrinienne (Dr. Bernard), Hôpital Lariboisière (Assistance Publique - Hôpitaux de Paris), Paris, France.
Correspondence to: Bruno Crestani, MD, PhD, Service de Pneumologie, Hôpital Bichat, 46 rue Henri Huchard, 75877 Paris Cedex 18, France; e-mail: bruno.crestani{at}bch.ap-hop-paris.fr
| Abstract |
|---|
|
|
|---|
Design: Prospective study.
Setting: An adult trauma/surgical ICU in an urban teaching hospital.
Patients: A total of 20 patients with ALI receiving mechanical ventilation (patients), and 10 patients without pulmonary disease not receiving mechanical ventilation (control subjects).
Interventions: None.
Measurements and main results: PTHrP was detected in all BAL fluids, and ELF PTHrP concentrations (median; 25% to 75% percentiles) tended to be higher in patients (52.2 nmol/mL; 20.8 to 65.6 nmol/mL) than in control subjects (25.4 nmol/mL; 20.5 to 35.4 nmol/mL; p = 0.18). In patients, ELF PTHrP concentration correlated positively with the PaO2/fraction of inspired oxygen ratio (r = 0.53; p = 0.005), and negatively with lung injury score (r = - 0.44; p = 0.02), radiologic score (r = - 0.40; p = 0.04), and BAL albumin concentration (r = - 0.42; p = 0.02).
Conclusion: PTHrP is present in biologically significant concentrations in the alveolar milieu in humans. In patients with ALI, the PTHrP concentration correlates negatively with the degree of lung injury.
Key Words: acute lung injury alveolar repair alveolar type 2 epithelial cell ARDS cytokine growth factors inflammation intensive care mechanical ventilation parathyroid hormone-related protein
| Introduction |
|---|
|
|
|---|
Recently, parathyroid hormone-related protein (PTHrP), a protein initially described in patients with cancer and hypercalcemia,2 has been shown to be an autocrine/paracrine regulator of AEC2, which express both PTHrP and its receptor.3 Both in vitro and in vivo experimental data obtained in rats indicate that PTHrP stimulates AEC2 differentiation and regulates the number of AEC2 in the lung by independent inhibition of cell proliferation4 and cell apoptosis.5 In vivo, limited data obtained in rats indicate that after hyperoxic lung injury, PTHrP expression in the lung decreases as AEC2 proliferate.6 However, there are no data concerning the expression of PTHrP in the lung in normal individuals or in patients with ALI.
Our hypothesis was that PTHrP was expressed in the lung in subjects without pulmonary disease and that PTHrP concentrations in the lung decreased in patients with ALI because of the need for increased AEC2 proliferation. The aims of this study were as follows: (1) to measure the concentration of PTHrP in the epithelial lining fluid (ELF) from subjects without pulmonary disease and patients with ALI receiving mechanical ventilation sampled within 1 week of the beginning of the disease, and (2) to determine whether PTHrP concentrations correlated with the intensity of lung injury.
| Materials and Methods |
|---|
|
|
|---|
18 mm Hg (when available) or no clinical evidence of congestive heart failure. The exclusion criteria were: HIV infection, end-stage cancer, age < 18 years, PaO2/FIO2
70 mm Hg, current pregnancy, or inclusion in another protocol.
The BAL procedure was performed as previously described.8
All of the patients were intubated and receiving mechanical ventilation when BAL was performed. Briefly, six aliquots of 20 mL of sterile saline solution were injected through a bronchoscope wedged into a pathologic lung segment, and manually gently aspirated. The first aliquot, representative of a bronchial lavage, was discarded, and the other aliquots were pooled and filtered on sterile gauze. Ten milliliters of the BAL fluid were immediately processed for a bacteriologic direct examination and culture. The diagnosis threshold for lung infection was a bacterial growth
104 cfu/mL.9
The remaining of each BAL (at least 10 mL) was kept on ice and handled rapidly in our laboratory.
The day of the BAL, the variables necessary to calculate the simplified acute physiology score (SAPS)-II,10 the number of organ-system failures (OSFs),11 and the lung injury score (LIS)12 were prospectively assessed. For the calculation of the LIS, we did not use the pulmonary compliance as allowed by Murray et al12 in their original article. A radiologic score was obtained with the frontal chest radiograph measuring the density of the lung infiltrates.13 Each lung was divided into two quadrants. For each quadrant, a score from 0 (no infiltrates) to 3 (alveolar condensation) was assessed. When adding the score of the four quadrants, a final score from 0 to a maximum of 12 was obtained.13
The delay between the BAL procedure and the onset of pulmonary infiltrates was recorded for all patients as well as the outcome 30 days (death or survival) after the first BAL. The risk factors associated with the development of ALI were prospectively identified when the patients entered the study.14 The nature of the lung insult was assessed by the physicians in charge of the patients after review of the available clinical and biological data. The physicians in charge of the patients were unaware of the ELF PTHrP concentrations. To assess the prognosis value of ELF PTHrP concentration, we classified patients into survivors or nonsurvivors 30 days after inclusion in the study.
In patients with ALI, we used a PaO2/FIO2 ratio cutoff value < 150 mm Hg to identify the patients with the most severe disease. Such ratio has been previously used to select patients with severe ARDS.15
The systematic collection of blood samples at the time of BAL, as well as the performance of repeated sequential BAL, were refused by our local ethical committee if not needed for clinical purposes. In eight patients, we had the opportunity to obtain aliquots of venous blood samples collected immediately before the BAL procedure in sterile heparinized tubes. Plasma was immediately separated by centrifugation and stored at - 20°C until PTHrP assay.
Control Subjects
We collected BAL fluid from 10 patients not receiving mechanical ventilation who underwent fiberoptic bronchoscopy for unexplained cough (n = 5) or suspicion of hemoptysis (n = 5); 6 of these patients were current smokers. All of the patients had normal chest radiographic findings. In all cases, fiberoptic bronchoscopy did not display any abnormalities. There was no microbiological pathogen identified on the culture of the BAL. The cytologic count of the BAL was in a normal range for all the patients. There was no respiratory disease identified within 3 months of follow-up after the fiberoptic bronchoscopy.
BAL Sample Processing
The cell pellet and the supernatant were separated by centrifugation (250g for 10 min at 4°C). The supernatant was stored at - 80°C with 5% of aprotinin (Trazylol; Bayer Pharma; Sens, France), a protease inhibitor. The cell pellet was washed and suspended in RPMI medium (Gibco, Life Technology; Cergy Pontoise, France) at a final concentration of 106 cells/mL. The differential cell count was performed on a cytocentrifugate smear with a May-Grunwald-Giemsa stain (Labonord; Villeneuve dAscq, France).
Measurement of PTHrP, Urea, and Albumin in BAL Fluid
All measurements were performed on a freshly defrozen, nonconcentrated BAL supernatant. PTHrP was measured in BAL supernatant using a commercially available immunoradiometric assay kit (INCSTAR; Stillwater, MN). This assay detects specifically the 184 PTHrP. The lower limit of detection was 0.2 nmol/mL. The amount of BAL required for the immunoradiometric assay was 200 µL for each test. Each sample was assayed in duplicate, and the final result was the mean of the two assays. We checked that the saline solution with aprotinin used for the BAL did not interact with the PTHrP assay.
The volume of ELF recovered was estimated using urea as an endogenous marker of dilution.16 BAL urea concentration was measured with a multiparametric analyst (Hitachi 947; Hitachi; Tokyo, Japan). BAL albumin concentration was measured with a Hitachi 911 analyst (Hitachi).
The concentration of PTHrP in the ELF was estimated by the following formula:
![]() |
Statistical Analysis
All comparisons were unpaired. The statistical analysis was performed with the statistical software (Sigma Stat; Jandel Scientific; San Jose, CA). Continuous variables were expressed as mean ± SD for normally distributed variables and then compared using a Students t test. The nonnormally distributed variables were expressed as median (25th to 75th percentiles), and compared with the Mann Whitney test. For comparisons between two or more groups, an analysis of variance or Kruskal-Wallis test were used. Correlation between nonnormally distributed variables used the Spearman correlation coefficient (r). All p values
0.05 were considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
ELF PTHrP: Correlation with the Severity of Lung Injury
PTHrP was detectable in all the BAL fluid samples, either from patients with ALI or from control subjects. Median ELF PTHrP concentrations were not statistically different in patients with ALI (52.2 nmol/mL; 20.8 to 65.6 nmol/mL) and in control subjects (25.4 nmol/mL; 20.5 to 35.4; p = 0.18).
In patients with ALI, ELF PTHrP concentration correlated with all the indexes of lung injury that we studied (Fig 1 ). Indeed, ELF PTHrP concentration correlated positively with the PaO2/FIO2 ratio (r = 0.53; p = 0.005), and correlated negatively with LIS (r = - 0.44; p = 0.02), radiologic score (r = - 0.40; p = 0.04), and BAL albumin concentration (r = - 0.42; p = 0.02). By contrast, we found no correlation between ELF PTHrP concentration and the SAPS II score or the number of OSFs. Thus, ELF PTHrP concentrations were the lowest in patients with the most severe lung injury. ELF PTHrP concentrations in patients with severe hypoxemia (27.5 ± 16 nmol/mL) were lower than in patients with less severe hypoxemia (PaO2/FIO2 > 150 mm but < 300 mm Hg) [76 ± 54 nmol/mL; p = 0.01] but not different than ELF PTHrP concentrations measured in control subjects (30.5 ± 17 nmol/mL; Fig 2 ).
|
|
ELF PTHrP: Lack of Prognosis Value
ELF PTHrP concentrations were similar in the 13 patients who died within 30 days after their first BAL (50.9 nmol/mL; 20.8 to 70.1 nmol/mL) and in the 7 patients who survived (52.7 nmol/mL; 17.2 to 101.8 nmol/mL; p = 0.78).
| Discussion |
|---|
|
|
|---|
PTHrP is a protein discovered as a hypercalcemia-inducing product of malignant cells. PTHrP has since been demonstrated to be produced in many normal tissues, where it plays predominantly paracrine and/or autocrine roles and participates in the regulation of tissue and organ development, differentiation, and proliferation. The first 134 amino acids of PTHrP react with the parathyroid hormone (PTH)/PTHrP receptor (PTH receptor 1, which responds to both PTH and PTHrP) and produce the biological effects of PTHrP. The 184 PTHrP that we measured in the ELF contains this 134 sequence and has the same biological action that the 134 PTHrP had in vitro.17 18
PTHrP and the PTH/PTHrP receptor are robustly expressed in fetal lung where PTHrP is produced by epithelial cells, and the PTH/PTHrP receptor is expressed by mesenchymal cells. This hand-in-glove spatial pattern of distribution in developing terminal airways suggests that PTHrP acts as a paracrine factor and is involved in epithelial-mesenchymal interactions.19 In adult rat lung, PTHrP expression has been assigned to AEC2, whereas freshly isolated alveolar macrophages and cultured macrophages did not express PTHrP.20 The PTH/PTHrP receptor was localized to AEC2 by in situ hybridization in the adult rat lung3 and PTHrP has been detected in the pulmonary lavage fluid of adult rat.20 Data are lacking in humans. Our results show that PTHrP is present in biologically significant concentrations in the alveolar ELF in humans both in normal and pathologic conditions, and are coherent with a local production since PTHrP was not detected in plasma.
Among patients with ALI, the patients with the most severe lung injury as assessed by all the indexes that we used (PaO2/FIO2 ratio, LIS, radiologic score, and BAL albumin levels) had the lowest levels of PTHrP, levels similar to those measured in control subjects. The ELF PTHrP concentrations were increased in the patients with less severe ALI. We believe that the timing of the BAL procedure after alveolar injury did not explain this difference between groups since we found no correlation between PTHrP concentration and the delay of the performance of the BAL. Decreased PTHrP levels in the most severe patients could be due to the decreased turnover of the molecule, but we have no data to support this hypothesis. Our findings could rather suggest that after alveolar injury, PTHrP expression by AEC2 cells is increased in the lung and that in the patients with the most severe disease, alteration of AEC2 make them unable to produce PTHrP. This hypothesis is supported by two facts: (1) following cutaneous injury, increased levels of PTHrP are expressed by keratinocytes at wound margins and by the newly restored epidermis21 ; and (2) in the tracheal aspirates fluid of human neonates, PTHrP is associated with various indexes of AEC2 maturation, the lower levels being observed in preterm infants born at < 35 weeks of gestation, and the higher levels being measured in preterm infants exposed to antenatal steroids.
The concentrations of PTHrP that we measured in the ELF (about 50 µM) are well above the concentrations that have been shown to be active in vitro. For example, 10 nm of PTHrP134 maximally stimulated the accumulation of cyclic adenosine monophosphate in fetal lung fibroblasts in vitro,22 and 2.5 nm of 134 PTHrP increased desaturated phosphatidylcholine secretion by rat AEC2 in vitro.3 Animal models suggest that the PTH/PTHrP receptor is operational in vivo in the lung in normal conditions.23 However, in pathologic conditions such as ALI, we do not know if the PTHrP pathway is functional.
The exact function of PTHrP in the adult lung is not known, but available data suggest that PTHrP has a potential role in the alveolar repair process after lung injury. In vitro, PTHrP inhibits the proliferation of AEC2 cells, promotes a differentiated type-2 cell phenotype,3 4 and modulates the apoptosis of AEC2 cells,5 23 probably through the interaction with proteins of the bcl-2 family.24 This is further supported by in vitro data showing that PTHrP acts as a paracrine regulator for keratinocyte growth factor secretion,21 a potent growth factor for AEC2 that is present in the alveolar space in patients with ARDS.8
In conclusion, our study demonstrates that PTHrP, an important mediator of the balance between proliferation and apoptosis of AEC2, is present with biologically significant concentrations in the alveolar space in humans, with a relation with the degree of lung injury.
| Footnotes |
|---|
Supported by a grant from the Progamme Hospitalier de Recherche Clinique (1996).
Received for publication March 21, 2001. Accepted for publication August 8, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. H. Hastings, L. J. Deftos, J.-B. Stern, and B. Crestani Parathyroid Hormone-Related Protein and Lung Injury Chest, October 1, 2002; 122(4): 1494 - 1495. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |