In the assessment of cases of aortic valve disease with a view to operation, the standard methods of investigation are of limited value. Right heart catheterization gives some indirect information by measuring the cardiac output Haemopericardium sufficient to cause significant increase in the size of the heart shadow was a frequent complication, and many patients developed post-operative pericarditis with substernal pain and some fever. Two patients also complained of dysphagia. For these reasons we have now abandoned this procedure.LEFT VENTRICULAR PUNCTURE.-In order to find an alternative method for measuring the aortic valve gradient we decided to investigate the possibilities of direct puncture of the left ventricle through the intact chest wall. It was realized that there were many possible dangers in this procedure. Insertion of the needle into an irritable hypertrophied ventricle might precipitate ventricular fibrillation and bleeding might occur either from the ventricular cavity or from puncture of a coronary vessel. As we have had considerable experience of puncture of the left ventricle at operation and are familiar with the management of the complications which may arise during cardiac surgery, we felt that it was logical and reasonable for us to investigate the possibilities of this method and that our experience would enable us to deal with any untoward situation which might arise.Puncture of the left ventricle in this way is not an entirely new concept, but it has not previously been used for assessing the gradient across the aortic valve. Previous workers have been concerned with its application to angiocardiography. Nuvoli (1936) first performed left ventricular on 11 May 2018 by guest. Protected by copyright.

The transmembrane pressure is determined by the positive pressure applied to the blood and the negative pressure applied to the other side of the membrane. The blood flow through the filter is a function of pressure in the arterial line of the bypass circuit and the resistance of the haemofiltration circuit. A blood flow of 300-500 ml/min through the filter device and a transmembrane pressure of about 300 mm Hg are considered safe and will not cause appreciable haemolysis.

Our experience with the management of 28 patients with oesophageal perforation is reviewed. The majority of perforations ,followed oesophageal instrumentation. The occurrence of pain, ,fever or cervical crepitus following endoscopy should raise the suspicion of oesophageal perforation. Plain radiographs of the neck, chest and abdomen provided confirmatory evidence o f the presence o f a pecforation in 89 per cent of our cases.Contrast studies of the oesophagus demonstrated the site and extent ofthe leak in 21 of the 26 cases in which they were performed. The overall mortality for the series was 32 per cent. The mortality for thoracoabdominal perforations was nearly three times that for the cervical segment. Instrumental perforations were associated with a lower mortality than spontaneous perforations or those following paraoesophageal surgical procedures. The coexistence of an oesophageal obstruction with a perforation did not have an adverse effect on the outcome. The time lapse between the occurrence of the perforation and surgical intervention had a profound influence on the morbidity and mortality. Early closure of the perforation with drainage was associated with a 25 per cent mortality .for thoracoabdominal perforations and no complications in the survivors. A delay of over 24 h, on the other hand, was associated with a high incidence of septic complications in both cervical and thoracoabdominal perforations, a longer period of druinage and a 50 per cent mortality in the latter group. A plea is therefore made for early surgical intervention in both cervical and thoracoabdominal perforations.

No abstract

The incidence of bronchopleural fistula in 130 patients who had pneumonectomies has been reviewed. Patients were divided into two groups according to