The prognosis in nonsurgical disease has improved. We have confirmed the previously described good outcome in surgically treated disease. However, we have also demonstrated that the long-term prognosis for patients who have persistent pulmonary hypertension at 3 months after surgery is good. The observed improvements in outcome during the modern treatment era reinforce the importance of identifying patients with this increasingly treatable condition.

The earliest structural change in hypoxia-induced pulmonary hypertension is increased proliferation of adventitial fibroblasts. This fibroproliferative response occurs in acute and chronic hypoxic models, is dependent on p38 mitogen-activated protein (MAP) kinase activation, is selective for the pulmonary circulation, and would seem an important therapeutic target. Simvastatin attenuates pulmonary vascular remodeling in animal models, but additional information regarding mechanisms of action, differential antiproliferative effects and dose responses of available statins is required for appropriate clinical trial design. Our objectives were to determine the effects of statins on acute hypoxia-induced proliferation and p38 MAP kinase activation in pulmonary and systemic artery fibroblasts, to assess the effects of cholesterol intermediates, prenyltransferase and related inhibitors, and to determine the statin's mechanism of action. Atorvastatin, fluvastatin, and simvastatin inhibited adventitial fibroblast proliferation. At low doses (1 microM), this effect was selective for hypoxic (versus serum-induced) proliferation and was also selective for pulmonary (versus systemic) fibroblasts. Complete inhibition of hypoxia-induced p38 MAP kinase activity was achieved at this 1-microM dose. The lipophilic statins exhibited similar potency. The statin effect was reversed by geranylgeranyl pyrophosphate and mimicked by geranylgeranyl transferase and Rac1 inhibitors. Hypoxia-induced p38 MAP kinase activation and proliferation in pulmonary adventitial fibroblasts is dependent on a geranylgeranylated signaling protein, probably Rac1. One micromolar of fluvastatin exhibits a circulation- and stimulus-selective antiproliferative effect on pulmonary artery fibroblasts. The pharmacokinetics of fluvastatin would suggest that its antiproliferative effects may be useful in pulmonary hypertension associated with hypoxia.

Chest radiography (CXR) is the most commonly used imaging modality and deep neural network (DNN) algorithms have shown promise in effective triage of normal and abnormal radiograms. Typically, DNNs require large quantities of expertly labelled training exemplars, which in clinical contexts is a major bottleneck to effective modelling, as both considerable clinical skill and time is required to produce high-quality ground truths. In this work we evaluate thirteen supervised classifiers using two large free-text corpora and demonstrate that bidirectional long short-term memory (BiLSTM) networks with attention mechanism effectively identify Normal, Abnormal, and Unclear CXR reports in internal (n = 965 manually-labelled reports, f1-score = 0.94) and external (n = 465 manually-labelled reports, f1-score = 0.90) testing sets using a relatively small number of expert-labelled training observations (n = 3,856 annotated reports). Furthermore, we introduce a general unsupervised approach that accurately distinguishes Normal and Abnormal CXR reports in a large unlabelled corpus. We anticipate that the results presented in this work can be used to automatically extract standardized clinical information from free-text CXR radiological reports, facilitating the training of clinical decision support systems for CXR triage.

Hypoxic pulmonary hypertension is a worldwide public health problem. Statins attenuate hypoxic pulmonary hypertension in animal models, but the mechanism of action and applicability of these results to human treatment are not established. In hypoxic models, pulmonary artery fibroblast proliferation contributes substantially to pulmonary vascular remodeling. We previously showed that acute hypoxic pulmonary adventitial fibroblast proliferation can be selectively inhibited by statins and p38 mitogen-activated protein (MAP) kinase inhibitors. Here we used complementary chronic hypoxic and acute hypoxic coculture models to obtain necessary preclinical information regarding the utility of fluvastatin in the treatment of chronic hypoxic pulmonary hypertension. The effects of fluvastatin, cholesterol pathway intermediates, and related inhibitors on hypoxic adventitial fibroblast proliferation, p38 MAP kinase phosphorylation, and pulmonary artery smooth muscle cell proliferation were determined, using complementary chronic hypoxic rat and acute hypoxic bovine cell models. Fluvastatin reversed the proliferative phenotypic switch in adventitial fibroblasts from chronic hypoxic animals. This effect was circulation-specific, and implicated a Rac1-p38 MAP kinase signaling pathway. Coculture and conditioned media experiments also implicated this statin-sensitive signaling pathway in the release of pulmonary artery smooth muscle cell mitogens by hypoxic pulmonary adventitial fibroblasts. Treprostinil, sildenafil, and bosentan exerted no effect on the hypoxic fibroblast phenotype. Phenotypic changes (increased proliferation and mitogen release) in pulmonary artery fibroblasts during chronic hypoxia are dependent on a Rac1-p38 MAP kinase signaling pathway. The inhibition of these phenotypic changes with fluvastatin may be therapeutically relevant in high-altitude residents and in patients with hypoxic lung disease.

Background Patients presenting with acute hypercapnic respiratory failure due to exacerbations of chronic obstructive pulmonary disease (AECOPD) are typically managed with non-invasive ventilation (NIV). The impact of low-flow extracorporeal carbon dioxide removal (ECCO2R) on outcome in these patients has not been explored in randomised trials. Methods Open-label randomised trial comparing NIV (NIV arm) with ECCO2R (ECCO2R arm) in patients with AECOPD at high risk of NIV failure (pH < 7.30 after ≥ 1 h of NIV). The primary endpoint was time to cessation of NIV. Secondary outcomes included device tolerance and complications, changes in arterial blood gases, hospital survival. Results Eighteen patients (median age 67.5, IQR (61.5–71) years; median GOLD stage 3 were enrolled (nine in each arm). Time to NIV discontinuation was shorter with ECCO2R (7:00 (6:18–8:30) vs 24:30 (18:15–49:45) h, p = 0.004). Arterial pH was higher with ECCO2R at 4 h post-randomisation (7.35 (7.31–7.37) vs 7.25 (7.21–7.26), p < 0.001). Partial pressure of arterial CO2 (PaCO2) was significantly lower with ECCO2R at 4 h (6.8 (6.2–7.15) vs 8.3 (7.74–9.3) kPa; p = 0.024). Dyspnoea and comfort both rapidly improved with commencement of ECCO2R. There were no severe or life-threatening complications in the study population. There were no episodes of major bleeding or red blood cell transfusion in either group. ICU and hospital length of stay were longer with ECCO2R, and there was no difference in 90-day mortality or functional outcomes at follow-up. Interpretation There is evidence of benefit associated with ECCO2R with time to improvement in respiratory acidosis, in respiratory physiology and an immediate improvement in patient comfort and dyspnoea with commencement of ECCO2R. In addition, there was minimal clinically significant adverse events associated with ECCO2R use in patients with AECOPD at risk of failing or not tolerating NIV. However, the ICU and hospital lengths of stay were longer in the ECCO2R for similar outcomes. Trial registration The trial is prospectively registered on ClinicalTrials.gov: NCT02086084. Registered on 13th March 2014, https://clinicaltrials.gov/ct2/show/NCT02086084?cond=ecco2r&draw=2&rank=8