ECG changes during in-situ hypothermic liver resections

Editor, Liver resection is increasingly performed to remove colorectal metastases.1 It provides clinical benefits through complete resection of the neoplastic material.2 Surgical techniques are improving to extend this therapy to more patients and complex situations.3 When major resection with vascular reconstruction is deemed necessary, in-situ hypothermic liver resection becomes an option. It is thought to better preserve liver function when a long ischemic time is needed for parenchyma resection and vascular reconstruction.4 To decrease liver temperature, a topical cooling technique can be used alone or with portal hypothermic perfusion. Such techniques have major implications for the anaesthesiologist, such as potential hemodynamic instability, coagulopathy, hypothermia and hyperkalaemia.5 This article describes the occurrence of ECG changes that were observed during two major liver resections while using a topical cooling technique. When we started using hypothermic liver resection in our centre, liver hypothermia was achieved with topical cooling only. During two of these cases, unexpected ECG changes were observed. The first patient, a 43-year-old man without specific past medical history, was scheduled for open right hepatectomy. After dissection, topical cooling of the liver was induced by placing wet icy cold gauzes both in the retro-hepatic space and over the superior border of the liver. When the surgeon mobilised the liver to start the resection, cold gauzes inserted over the superior border of the liver came into contact with the diaphragm. The patient then developed ST segment depression and profound T wave inversion on DII and V5 derivations. These changes rapidly disappeared when the gauzes were removed, but reappeared quickly after the gauzes were reinserted. Figure 1a shows the preoperative ECG. Figure 1b shows ECG changes 2 min following gauze placement. Figure 1c shows the ECG immediately after removal of the gauzes. A couple of minutes later, the same changes were observed when the gauzes were reinserted in the same space (Fig. 1d). The gauzes placed over the superior border of the liver were then completely removed. A couple of months later, the same changes were observed in a 51-year-old man just after the gauzes over the superior border of the liver had been placed. On the basis of our previous experience, the gauzes were rapidly removed and the ECG normalised (not shown). As the ECG changes were rapidly detected, the contact time between the gauzes and the diaphragm lasted only few minutes. In neither of these patients arrhythmias or hemodynamic instability were noted. Changes (if any) of cardiac function could not be demonstrated, as a transesophageal echocardiography probe was not inserted and not available at the time of the events.Fig. 1: no caption available.The use of cold gauzes over the superior border of the liver during in-situ hypothermic liver resection with topical cooling seems to induce ECG changes. As the liver is very close to the right ventricle, a local effect on epicardial repolarisation might explain the observed changes.6 This technique could have also induced a return of cold blood to the right ventricle, similar to that observed during orthotopic liver transplantation. However, significant temperature changes deep inside the liver might well take a few minutes under topical cooling. ECG changes were observed seconds after placement of gauzes and started to recede very soon after their removal, suggesting a local effect. Indeed, similar but more sustained ECG changes due to possible hypothermic–osmotic damages have been reported in paediatric cardiac patients after topical cooling of the heart with ice.7 In our patients, both the short contact time and the relative protection of the heart offered by the diaphragm likely prevented sustained changes such as arrhythmia or hemodynamic instability. To our knowledge, this is the first report of ECG changes due to local cardiac cooling during non-cardiac surgery. After these observations were made, the cooling technique was modified and cold gauzes placed above the liver are now separated from the diaphragm by gauzes at ambient temperature. Moreover, we now use an intra-hepatic cooling technique with portal hypothermic perfusion of the liver that depends less on the use of cold gauzes. When performing this technique, a porto-veno-venous bypass is first inserted and then both the retro-hepatic space and the superior border of the liver are packed with icy cold gauzes as described above. Total liver vascular exclusion is performed and cold graft preservation solution is infused through the portal vein and drained from a retro-hepatic cavotomy. When surgery is completed, the preservation solution is flushed and the bypass weaned. In our experience, this technique is associated with a lower systemic temperature and a more sudden return of cold blood after unclamping. Nevertheless, no profound ECG changes such as those described previously have been observed with the modified technique which is consistent with a local effect hypothesis. Moreover, portal hypothermic perfusion is not only associated with an intra-hepatic cooling effect which depends less on topical gauzes, but it is the only technique that has been associated with improved clinical outcome.4,5 We suggest that during in-situ hypothermic liver resection, cold gauzes over the superior border of the liver should be separated from the diaphragm to avoid the occurrence of ECG changes. Moreover, a portal hypothermic perfusion technique might be preferred. Acknowledgements The authors would like to thank Dr Marie-Christine Gillon for her assistance in data retrieval and Miss Nadia Morfousse for her clerical work. The Royal College of Physicians and Surgeons of Canada supported this work by a fellowship grant. None of the authors has any conflict of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.