Recently, Dominik et al.  reported on their experience with cardiopulmonary bypass (CPB) and deep hypothermic circulatory arrest (DHCA) in a series of 21 patients with RCC and atrial tumour thrombus. After careful reading, we feel that several issues described in their paper need further discussion.
The authors should be congratulated for their excellent long-term outcomes. The use of standardized metrics for reporting surgical complications has allowed a more direct comparison between surgical series. The most commonly used system is the one described by Clavien et al. . In this article the authors only reported on in-hospital mortality. Of interest is the non-fatal sequel which is frequently experienced by those patients who undergo CPB and DHCA. In the future we hope that the authors will provide these data using a standardized means of reporting so that we, and others, can compare between surgical series. Only then can the authors conclude that DHCA would be considered ‘the safest procedure’.
Cardiopulmonary bypass, with or without DHCA, has been shown to be a useful but potentially dangerous procedure with a wide range of complications, including haematological, hepatic, renal, neurological and septic complications, among others .This is the main reason why we avoid CPB whenever possible. We agree with the idea that this procedure is absolutely necessary in particular situations but, in our opinion, it is not necessary in the majority of cases . In this way, our group has been refining the approach to RCC with tumour thrombus for more than 10 years now. The main basis supporting our work has been focused on the translation of the use of surgical manoeuvres derived from the field of organ transplantation into urological oncology [3-6]. As widely described in our previous reports [3-6], the use of these techniques provides the surgeon with a vast therapeutic arsenal to deal with cases of RCC with tumour thrombus. In our opinion, the technique of incising the diaphragm and clamping the heart from a total intra-abdominal approach  would definitely play a role in increasing the safety of these procedures and thus, improving the complication rates observed.
We read of the authors' criticism of and scepticism about the ‘milking’ manoeuvre that we commonly use in the management of level IV tumour thrombi. The cross-clamping of the right atrium above the upper thrombus burden is performed rapidly to avoid the migration of dislodged fragments into the pulmonary arteries. Before atrial cross-clamping, a quick trial is attempted to assess the haemodynamic tolerability. In cases where preload descends rapidly, a rapid infuser is available. Further, this is performed under transoesophageal echocardiography (TOE) guidance. We have not observed any case of intra-operative embolism using this approach [3-6]. In our opinion, this manoeuvre allows the safe excision of level IV tumour thrombi in most instances avoiding the use of CPB or veno-venous bypass.
Piggy-back liver dissection, Pringle manoeuvre and sequential inferior vena cava (IVC) clamping (total hepatic vascular exclusion mimicking the anhepatic phase of liver transplantation), allow the dissection and excision of a given tumour thrombus placed at any location inside the IVC under superb visual conditions. This manoeuvre also minimizes potential bleeding during the thrombectomy and thus improves surgeon visibility.
The authors assert that this manoeuvre can be considered a ‘blind’ procedure. To our knowledge, this assumption implies the absence of direct or indirect vision inside the atrium/caval lumen during its performance. It is true that this technique does not provide direct visual information, but in fact it can be followed in real-time with indirect visual control by monitoring the sequence of events with intra-operative TOE. This imaging technique can be used in order to change our surgical strategy, i.e. for CPB onset, if it is considered necessary for the overall safety of the operation [3-6].