When transplanting a kidney from donor to recipient, there is an unavoidable period in which the kidney receives no blood flow. This is referred to as ischemia. While the kidney can be put on ice, this period of ischemia inevitably leads to the kidney sustaining some damage, especially for kidneys from deceased donors, which can have longer ischemic times. When the transplant operation is complete, a donor’s kidney will be fully connected to the recipient’s blood supply. However, even once connected to the recipient’s circulation, we felt that there was often still room to optimize patient hemodynamics in order to improve blood flow to the newly transplanted, vulnerable kidney.

We thus aimed to characterize features of intraoperative anesthesia courses in order to determine whether such parameters (blood pressures, vasopressor doses, IV fluids doses) were associated with delayed graft function (DGF). This is a postoperative adverse event where, despite a successful surgery, the recipient needs to receive dialysis within a week postoperatively due to the graft being slow to recover function.

In this study, we looked at the anesthesia records for over 600 kidney transplant recipients from a single academic medical center (Vancouver General Hospital, 2014-2020). This yielded granular time series data on the blood pressures throughout the surgery, as well as the different kinds and doses of medications and fluids administered, in 5-minute intervals. Information on special intraoperative time points such as anastomosis (surgical connection of the kidney to the recipient blood supply) were also obtained. We then used a variety of methods to analyze the data.

One hundred forty-two DGF events were captured in our cohort (62/294 neurologically determined death; 80/167 donation after cardiac death; 11/210 live donation). Our first analysis was to graph mean arterial pressures throughout the surgeries, from start to end of anesthesia. We noted that in patients who ended up having DGF, diastolic blood pressure and mean arterial pressure throughout the operation were substantially lower. Next, we closely investigated average postanastomosis blood pressures, as this is the time point in the operation where the new kidney is exposed to recipient hemodynamics. We found that the rate of DGF rose precipitously in deceased-donor kidneys from 25% to 50% when the postanastomosis average blood pressure dropped below 85 mm Hg. Finally, we performed a regression analysis where we checked the interaction of vasopressor and fluid dosing on postanastomosis blood pressure and DGF incidence. We found that recipients of deceased-donor kidneys did better when increased vasopressors and fluids were used when the blood pressures were low, below 75 mm Hg. However, DGF rates rose in patients when high doses of pressors were used when postanastomosis mean arterial pressure was already above 75 mm Hg. This suggests that high blood pressures in surgeries may be optimal (>85 mm Hg), but the use of higher and higher doses of vasopressors and fluids beyond 75 mm Hg may be suboptimal.

Overall, we have collected a highly granular data set in transplant anesthesia. With this, we have been able to take a detailed look into intraoperative hemodynamics and optimal decision-making in kidney transplant surgeries. Looking forward, we plan to further interrogate the data, and also to assimilate the surgical data with patient factors, in order to create bespoke, personalized anesthesia targets, uniquely optimizing individual patients for their new kidneys.