Editorial comment

Abstract

In 1975, Randall Griepp reported the use of deep hypothermia as a technique to make aortic arch surgery possible and survivable. However, profound cooling carried disadvantages (prolonged bypass times, coagulopathy) and the duration of circulatory arrest was constrained by the limited increase in brain ischaemic tolerance. In an attempt to improve this tolerance, retrograde perfusion via the superior vena cava was introduced but achieved little actual brain perfusion, did not alter peri-arrest metabolism and appeared to have limited clinical efficacy. The re-introduction of antegrade brain perfusion established via direct cannulation or via the axillary artery, undoubtedly improved surgical outcomes and also allowed reflection on whether deep hypothermia remained necessary. Various reports have established that deep hypothermia may no longer be an essential adjunct when brain perfusion is not arrested. However, the corpus and spinal cord are also vulnerable and although their ischaemic tolerance is greater than that of the brain, both require protection. The paper by Urbanski et al. [1, in this issue] systematically demolishes the concept that appreciable cooling is a necessary adjunct to all arch surgery. In a truly remarkable series of 347 patients, they demonstrate that an approach to bypass, that tailors the degree of cooling to the anticipated extent of repair and duration of corporeal circulatory arrest, can be rewarded by a hospital mortality and permanent neurological deficit rate of approximately 1% each and no paraplegia. Temporary neurological deficit rates of 2.3% are reported and prolonged ventilation incidence was low. Only five patients required dialysis. How were these prestigious outcomes achieved? In essence, the mainstay of their technique was to use the carotid arteries for arterial return and to limit the duration of non-perfusion of supra-aortic arteries and the distal aorta by prompt re-cannulation and reperfusion using pre-constructed side-limbs of the vascular grafts. This meant that the exclusion period for any arterial territory was reduced. In the cases of total arch replacement, corporeal arrest periods were limited to 34 ± 12 min (range 17–70 min) with circulatory arrest temperatures of 30 °C, by perfusion and then proximal clamping of the main arch graft. Supra-aortic arteries would then be separately reimplanted using further graft limbs, recruiting each into the main perfusion circuit on anastomotic completion. The perfusion apparatus set-up was certainly elaborate but was not dependent upon separate pumps; only single or multiple Y-branching in the circuitry. The less complex procedures required simpler perfusion solutions tolerated at higher temperatures. Using these techniques, lactate generation and other indices of organ dysfunction were modest and did not correlate with temperature or arrest duration. As we do not all reside in Bad Neustadt, can these excellent results be generalised? The authors restrict their report to more elective cases and do not describe their experience with acute dissection, a condition in which hypothermic techniques perhaps remain a mainstay of protection. Moreover, the elaborate methodology requires detailed pre-operative planning and has to be part of coordinated team effort between surgery, anaesthesia and perfusion and cannot be recommended for an occasional practice. Carotid cannulation is a minority, centre-specific technique and has little comparative data demonstrating its position versus alternatives. Nevertheless, Urbanski et al. have set the scene for a new paradigm in aortic surgery demonstrating that the benefits of warmer operations can be accrued by the careful planning of procedures to limit the exclusion periods of critical circulatory territories even within the aortic arch. We wait with interest, whether future contributors will endorse this approach or whether this represents a centre-specific paradox. Journal editors and reviewers will be critical in providing the data to inform future practice in a balanced way and should guard against uni-directional publication bias for favourable outcome studies. As a cardiovascular surgeon, I note the achievements in this report with great interest and I suspect that my own practice will gradually become incrementally warmer as confidence is gained in the techniques described. Hopefully, ultimately, randomised controlled trial data will be forthcoming that will provide re-assurance of which techniques are superior.