No abstract

HE first successful intra-thoracic oesoph-T agectomy was performed by Thorek in 1913. Surgery of the oesophagus has gone a long way since then, nevertheless, the mortality and morbidity still remains relatively high; so much so, that there is even a school who consider intra-thoracic resections and anastomosis for cancer of the oesophagus "not worth while" and would rather do some sort of palliative by-pass in the neck. Moreover, it is often taken for granted that anastomotic leaks are inevitable in some of these cases because of the "poor blood supply" of the oesophagus. It was to challenge these statements and attitudes, on anatomical grounds, that the present study was undertaken.Interest in its anatomy, beyond gross description of the oesophagus, lay more or less dormant until Thorek's achievement. Texts in anatomy presented a stereotype account of the structure and its blood and nerve supply; this was static, academic, museum anatomy, rather than living clinical anatomy. Since 1913 many excellent reports have appeared in the literature dealing with the blood supply with special reference to surgical procedures of the oesophagus. Much of this work was done on dogs, others studied the blood supply on mortuary specimens. This report is based on observations on 32 mortuary dissections, the bodies being dead less than thirty-six hours. METHODBasically, the procedure consisted of dissection, cannulation and injection of radiopaque fluid, of vessels supplying the oesophagus, followed by X-ray photography. In some cases, where we were unable to cannulate and inject dye adequately, only photographs were taken. Of the various radiopaque *Formerly visiting Professor of Surgery, University ?Consultant Surgeon, General Hospital, Singapore. of Singapore.fluids tried, a mixture of bismuth subgallate and gelatin in water was the most successful. This was a technique described by Noh1 (1962) in his study of the spread of cancer of the bronchus but we had to modify the actual mixture to suit our purposes because of the comparatively tiny vessels involved. The cannulated vessels were first washed out with tap water and obvious leaks in the specimen ligated. This was followed by injection of the bismuth subgallate-gelatin mixture. The main difficulty with this procedure is that the gelatin often solidifies before the injection is complete, and rupture of the vessels due to excessive pressure results in a deposit of fluid obliterating the detail.In the study of the cervical oesophagus, the superior thyroid artery was cannulated through its external carotid opening and the inferior thyroid vessels either through the thyro-cervical trunk or sometimes through the subclavian artery itself. The aortic branches to the thoracic oesophagus were caanulated after splitting the aorta longitudinally. The oesophageal branch of the left gastric artery was demonstrated by cannulating the main left gastric artery and tying off the gastric branches on the lesser curve. The left inferior phrenic artery was cannulated near its origin and the di...