Gastroesophageal re¯ux is a major postoperative problem in esophageal patients with cancer, and the principal cause is resection of the lower esophageal sphincter. Two new antire¯ux operations to solve this problem were investigated. The number of patients studied was 139, with a male to female ratio of 5. The reservoir technique was applied to the ®rst 50 patients and the globe technique was used in the remaining 89. Hospital mortality was 9.35%. Patient satisfaction from a re¯ux standpoint was excellent in 91.4%, good to fair in 6.5%, and poor in 2.1%. Postoperative barium swallow at the ®rst, third, and sixth months showed no re¯ux in 93% of cases. Postoperative preanastomotic mean pressure was 14.2 mmHg. Postoperative mucosal biopsies revealed a remarkable reduction in esophagitis. The radiologic, manometric, and histologic ®ndings as well as the patient satisfaction rate suggest that these antire¯ux operations are suitable and eective for patients undergoing esophageal resection and intrathoracic esophagogastric anastomosis.

Gastroesophageal reflux is a major postoperative problem in esophageal patients with cancer, and the principal cause is resection of the lower esophageal sphincter. Two new antireflux operations to solve this problem were investigated. The number of patients studied was 139, with a male to female ratio of 5. The reservoir technique was applied to the first 50 patients and the globe technique was used in the remaining 89. Hospital mortality was 9.35%. Patient satisfaction from a reflux standpoint was excellent in 91.4%, good to fair in 6.5%, and poor in 2.1%. Postoperative barium swallow at the first, third, and sixth months showed no reflux in 93% of cases. Postoperative preanastomotic mean pressure was 14.2 mmHg. Postoperative mucosal biopsies revealed a remarkable reduction in esophagitis. The radiologic, manometric, and histologic findings as well as the patient satisfaction rate suggest that these antireflux operations are suitable and effective for patients undergoing esophageal resection and intrathoracic esophagogastric anastomosis.

One hundred and six patients who left hospital after Starr-Edwards ball valve replacements have beenfollowedfor a period of 6 months tO 4 years. All patients were maintained on oral anticoagulant therapy so that a prothrombin ratio of approximately I 7:I was attained. Four patients suffered minor nose bleeds. During the period offollow-up there have been 4 patients who have suffered embolic phenomena, one ofwhich provedfatal. The site of thrombusformation in thefatal case was the left atrium and not the prosthetic valves. The relatively low incidence of embolic and haemorrhagic complications can be added to the satisfactory immediate and late function of the valves themselves. Minor leaks have developed around 7 per cent of surviving aortic valve replacements but no mitral or tricuspid valve shows evidence of regurgitation. No case of valve failure has occurred.

No abstract

Left and right atrial pressures were monitored in 100 patients undergoing open heart surgery for acquired valvular disease of the heart at St. Thomas' Hospital. The right atrial pressure bore no constant relationship to the left atrial pressure in 61 % of the patients and there was no reliable way of deducing left atrial pressure from the right. We believe that it is imperative to measure both right and left atrial pressures if post-operative management after cardiac valve surgery is to be effective. There has been no instance of pulmonary oedema or postperfusion syndrome in our practice since its introduction.The maintenance of an optimum cardiac output is the main aim of post-operative care in a cardiac surgical patient. Cardiac output is a function of stroke volume and heart rate. After cardiac surgery the heart rate can usually be kept at an adequate level by pacemaking or by chronotropic drugs such as isoprenaline. The stroke volume depends in part on the filling pressure and in part on myocardial contractility, which can be improved by inotropic agents but is largely determined by the pre-operative condition of the patient.The stroke volume of the right ventricle is directly related to the central venous pressure (the right atrial pressure), and the output of the left heart is similarly related to the left atrial pressure. The value of the measurement of central venous pressure in regulating blood transfusion in the post-operative care of a cardiac surgery patient has been accepted (Sykes, 1963;Braimbridge and Ghadiali, 1965;Kirklin and Rastelli, 1965) and is universally employed. The importance of measuring the left atrial pressure has been recognized (Fishman, Hutchinson, and Roe, 1966;Keddie, Provan, and Austen, 1966) but has not been generally adopted, mainly because it is believed that right and left atrial pressures are closely related. This may be true in health but in acquired valvular disease, where one side of the heart may be more affected than the other, the normal rela-'Based on a paper presented at the annual meeting of the Society of Thoracic and Cardiovascular Surgeons of Great Britain and Ireland at Belfast (September 1969) tionship between right and left atrial pressures may not exist.At St. Thomas' Hospital we have been using right and lef.t atrial presures as a guide to the postoperative management of blood or fluid replacement in patients undergoing valve surgery. It is the purpose of this communication to analyse the data on the inter-relationship of right and left atrial pressures in these patients. METHODS AND MATERIALSOne hundred consecutive patients undergoing valve replacement or repair with the aid of cardiopulmonary bypass constitute the clinical material. The diagnoses and surgical procedures carried out in these patients are shown in Table I