Aspirin causes bronchoconstriction in aspirin-intolerant asthma (AIA) patients by triggering cysteinyl-leukotriene (cys-LT) production, probably by removing PGE2-dependent inhibition. To investigate why aspirin does not cause bronchoconstriction in all individuals, we immunostained enzymes of the leukotriene and prostanoid pathways in bronchial biopsies from AIA patients, aspirin-tolerant asthma (ATA) patients, and normal (N) subjects. Counts of cells expressing the terminal enzyme for cys-LT synthesis, LTC4 synthase, were fivefold higher in AIA biopsies (11.5+/-2.2 cells/mm2, n = 10) than in ATA biopsies (2.2+/-0.7, n = 10; P = 0. 0006) and 18-fold higher than in N biopsies (0.6+/-0.4, n = 9; P = 0. 0002). Immunostaining for 5-lipoxygenase, its activating protein (FLAP), LTA4 hydrolase, cyclooxygenase (COX)-1, and COX-2 did not differ. Enhanced baseline cys-LT levels in bronchoalveolar lavage (BAL) fluid of AIA patients correlated uniquely with bronchial counts of LTC4 synthase+ cells (rho = 0.83, P = 0.01). Lysine-aspirin challenge released additional cys-LTs into BAL fluid in AIA patients (200+/-120 pg/ml, n = 8) but not in ATA patients (0. 7+/-5.1, n = 5; P = 0.007). Bronchial responsiveness to lysine-aspirin correlated exclusively with LTC4 synthase+ cell counts (rho = -0.63, P = 0.049, n = 10). Aspirin may remove PGE2-dependent suppression in all subjects, but only in AIA patients does increased bronchial expression of LTC4 synthase allow marked overproduction of cys-LTs leading to bronchoconstriction.

Our results point to a deficient PGE(2) production under proinflammatory conditions in asthmatic airways. This could weaken local defensive mechanisms and promote cysteinyl leukotriene overproduction.

We have shown that inhalation of lysine aspirin enhances leukotriene production in the lungs of patients with aspirin-induced asthma (AIA). To assess the specificity of this reaction, we compared two well-matched groups of patients: eleven with AIA versus 14 asthmatics tolerant to aspirin (ATA). All subjects underwent bronchoalveolar lavage (BAL) with saline followed immediately by instillation of 10 mg of lysine aspirin, into a right middle lobe segmental bronchus, which was lavaged 15 min later. At baseline the two groups did not differ with respect to BAL fluid concentrations of cyclooxygenase products, peptido-leukotrienes, histamine, tryptase, interleukin-5 (IL-5), eosinophil cationic protein (ECP), or eosinophil number. Fifteen minutes after aspirin instillation, there was a statistically significant rise in peptido-leukotrienes, IL-5, and eosinophil number in AIA, but not in ATA, but not in ATA patients. In the former, but not in the latter group, mean histamine concentrations rose in response to aspirin, approaching the level of statistical significance. Tryptase and ECP levels showed no significant change. Aspirin significantly depressed PGE2 and thromboxane B2 (TXB2) in both groups, however PGD2, PGF2 alpha, and 9 alpha, 11 beta-PGF2 decreased only in ATA patients. A characteristic disturbance in eicosanoid balance, produced by aspirin in patients intolerant to this drug, might explain precipitation of asthma attacks.

We have recently shown that oral aspirin provocation leads to an increase in LTE4 and a reduction in 11-dehydro-TXB2 levels in urine of patients with aspirin induced-asthma (AIA). To test the hypothesis that cyclooxygenase inhibition and an enhancement of cysteinyl-leukotriene production occurs in the lungs of patients with AIA, we examined the eicosanoid levels in bronchoalveolar lavage fluid obtained 30 min after lysine-aspirin or placebo inhalation in 10 patients with AIA. Eosinophil cationic protein (ECP) levels were determined to evaluate eosinophil activation. Six asthmatics nonsensitive to aspirin (NA) underwent challenge with placebo. The dose of lysine-aspirin inhaled by patients with AIA was equal to that which had produced > or = 20% fall in FEV1. Compared with NA, patients with AIA had: (1) eicosanoid levels, particularly PGE2 and TXB2, elevated and (2) higher number of eosinophils and ECP. The overproduction of eicosanoids could be related to a distinct eosinophilic inflammation in airways of patients with AIA. Inhalation of lysine-aspirin had no effects on 12-HETE and 15-HETE levels, but it markedly depressed cyclooxygenase products and significantly enhanced leukotriene production in the lungs of patients with AIA.

Objectives: This prospective study aimed to assess the diagnostic yield of the combined approach -endobronchial (EBUS) and endoscopic (EUS) ultrasound-guided needle aspiration (combined ultrasound-needle aspiration (CUS-NA)) in the radiologically normal mediastinum in non-small-cell lung cancer (NSCLC) staging. Methods: CUS-NA was performed simultaneously under local anaesthesia and sedation in consecutive NSCLC patients with mediastinal nodes that were not enlarged on CT (stage IA-IIB). All patients with negative CUS-NA subsequently underwent the transcervical extended bilateral mediastinal lymphadenectomy (TEMLA) as a confirmatory test. Results: A total of 120 NSCLC patients underwent CUS-NA between 1 January 2008 and 31 December 2008. There were 318 mediastinal nodes biopsied (158 EBUS-NA -stations: 2R -2, 2L -1, 4R -34, 4L -33 and 7 -88 and 160 EUS-NA -stations: 4L -57, 7 -101 and 9 -2). CUS-NA revealed metastatic lymph node involvement in 19 of 120 patients (16%) and in 31 of 318 biopsies (10%). The prevalence was 22%. In 99 patients with negative CUS-NA, who underwent subsequent TEMLA, metastatic nodes were diagnosed in nine patients (8%) in 11 stations: 2R -2, 4R -4, 4L -1, 5 -3 and 7 -1. In all but one patient there were 'minimal N2' only. Diagnostic sensitivity, specificity, total accuracy, positive predictive value (PPV) and negative predictive value (NPV) of CUS-NA for normal mediastinum was 68% (95% confidence interval (CI): 48-84), 98% (95% CI: 92-100), 91% (95% CI: 86-96), 91% (95% CI: 70-99) and 91% (95% CI: 83-96), respectively. The sensitivity of CUS-NA was significantly higher than with EBUS-NA alone ( p = 0.04) and higher, close to the level of significance than with EUS-NA alone ( p = 0.07). The NPV of all techniques was high and that of CUS-NA was significantly higher than EBUS-NA alone and EUS-NA alone ( p = 0.01, p = 0.03). No complications of CUS-NAwere observed. Conclusions: In the radiologically normal mediastinum, CUS-NA is a highly effective and safe technique in NSCLC staging and, if negative, a surgical diagnostic exploration of the mediastinum may be omitted. #