SUMMARYBackground: On-demand therapy with esomeprazole is effective for long-term treatment of non-erosive gastrooesophageal reflux disease, but it has not been evaluated in erosive gastro-oesophageal reflux disease. Aims: To compare endoscopic and symptomatic remission over a 6-month period when patients with healed erosive gastro-oesophageal reflux disease are treated with esomeprazole 20 mg, either once daily or on-demand. Methods: Patients with verified erosive reflux oesophagitis of Los Angeles grades A-D were enrolled. Following 4-8 weeks treatment with esomeprazole 40 mg daily, those who were endoscopically healed and had symptom control during the last week were randomized to maintenance therapy for 6 months with esomeprazole 20 mg, taken either once daily or on-demand.

No abstract

Esomeprazole had no clinically relevant effect on the overall 24-h incidence of PONV. However, esomeprazole significantly reduced the total amount of vomit during 24-h post-operatively. This may be of value in patients with an increased risk of pulmonary aspiration.

We read with interest the review article by Komori et al. (Br J Surg 1991; 78: 1027-30). They emphasize the important role of the endothelium in modulating the underlying vascular smooth muscle by releasing endothelium-derived relaxing factor (EDRF), which is a potent vasodilator and inhibitor of platelet aggregation. In addition, they speculate on the possible relationship between decreased EDRF production by a functionally injured endothelium and the development of both vasospasm and intrinsic graft lesions.However, their article concentrates on reversed vein grafts where the effects of surgical preparation of the graft on EDRF production are well established', but they make no mention of the in situ technique. In femorodistal bypasses to below the knee this technique is the method of choice because it has the advantage of a better size-match between artery and vein graft at each anastomosis. Despite the fact that the vein is often fully mobilized' and a valvulotome is passed along its lumen to destroy the valves, it has been suggested that this technique leads to superior endothelial preservation when compared with the reversed method3.We have recently investigated endothelial and smooth muscle cell injury during preparation of both reversed and in situ vein grafts4. Whilst we agree that EDRF production is reduced after preparation of reversed vein grafts, it is absent after preparation of in situ grafts because of severe loss of endothelial cells and functional impairment of any remaining cells. In addition, preparation of reversed and in situ vein grafts also leads to injury to the smooth cells of the media.It is likely that cellular injury plays a role in the subsequent development of intrinsic lesions in infrainguinal vein grafts which are an important cause of failure in the first postoperative year. Recently a link has been established between endothelial and smooth muscle cell injury during preparation of reversed vein grafts and intimal hyperplasia in an organ culture system'. Therefore the importance of endothelial loss and absent EDRF production associated with the in situ technique warrants further study. . Surgical preparation impairs release of endothelium-derived relaxing factor from human saphenous vein. Ann Thorac Surg 1989; 48: 41 7-20. Beard JD. Wyatt M, Scott DJA, Baird RN, Horrocks M. The non-reversed vein femorodistal bypass graft: a modification of the standard in-situ technique. Eur J Vasc Sury 1989; 3: 55-60. Cambria RP, Megerman J, Abbott WM. Endothelial preservation in reversed and in-situ autogenous vein grafts. Ann Surg 1985; 4. Sayers RD, Watt PAC, Muller S, Bell PRF, Thurston H. Structural and functional smooth muscle injury after surgical 2. 3. 202: 50-5. preparation of reversed and non-reversed (in situ) saphenous vein bypass grafts. Br J Surg 1991; 78: 1256-8. Angelini GD, Soyombo AA, Newby AC. Smooth muscle cell proliferation in response to injury in an organ culture of human saphenous vein. Eur J Vasc Surg 1991; 5 : 5-12. 5.