Life threatening haemoptysis : a clinical and radiological study.

Abstract

The investigation and management of patients with life threatening haemoptysis is a common clinical problem in South African Hospitals. Establishing the aetiology and origin of the haemorrhage and treating these patients is both difficult and expensive in terms of human and financial resources. The purpose of this study was to identify common local aetiologies for severe haemoptysis, review the investigation and treatment of these patients at Wentworth Hospital, Durban and to formulate a plan of management. Retrospective and prospective studies of consecutive patients treated at Wentworth Hospital were performed. In the prospective study a new embolic material gelatin linked acryl microspheres (embospheres) was used for bronchial artery embolization (BAE). The study demonstrated a change in the spectrum of aetiologies of haemoptysis, from bronchiectasis following tuberculosis to destructive pneumonias. The chest radiograph was always the initial imaging investigation but was found to be inaccurate in detecting the origin of the bleeding. High resolution computed tomography of the lungs (HRCT) was the single best investigation to detect the cause and origin of the haemoptysis. HRCT detected focal bronchiectasis and intracavitatory aspergillomas that were undetected on the chest radiograph. Pleural thickening detected on CT was a good indicator of the presence of transpleural collaterals. The major limitation with HRCT was that it could not be performed if the patient was too dyspnoeic to cooperate during the scan. The role of bronchoscopy appears limited in patients with severe haemoptysis to those patients who are potential surgical candidates. I found that bronchoscopy was not accurate in detecting the source of bleeding in the few patients in which it was performed. Bronchial arteriography remains the gold standard in the detecting the source of haemorrhage. Bleeding sites were detected on angiography in the presence of focal hypervascularity, neovascularity and the presence of broncho-pulmonary shunts. Bronchial arteries were hypertrophied in bronchiectasis but were normal in size in some patients who had acute pneumonias. Bronchial artery embolization was the treatment of choice for severe haemoptysis in the patients studied. The use of gelatin cross linked micro spheres has significantly improved the initial success rate following the procedure with less complications compared to the use of polyvinyl alcohol particles (PVA). It is important to identify systemic transpleural collaterals at arteriography and to embolize them to reduce recurrent haemoptysis. Patients with aspergillomas responded well to embolization. Recurrent haemoptysis remains the major limitation of BAE but is reduced with the use of micro spheres as embolic agents and thorough embolization of systemic collaterals on the affected side. Surgical resection was an option for a limited number of patients with focal disease in one lung and good respiratory reserve. The major limitation of the study was the absence of long term follow up to detect those patients with late recurrent haemoptysis.