The authors reply
As pointed by Gubensek et al (1), the biocompatibility of this new method heparin-free regional anticoagulation has not been assessed in our study (2), recently published in Critical Care Medicine. The lack of citrate infusion just after the catheter, and thus of efficient anticoagulation in the arterial catheter of the dialyzer, may indeed be sufficient to trigger monocytes, neutrophils, and platelet activation. We are conducting a prospective study comparing various heparin-free regional anticoagulation, including an in-depth assessment of the bioincompatibility. This study will help to better define the optimal anticoagulation of dialysis filter.
From January 2016, this new protocol of heparin-free anticoagulation is now routinely used in our institution. Hence, at September 1, 2017, 190 dialysis sessions were performed with this approach in our renal and ICU units. Premature filter clotting occurred in only five sessions (2.6%; median time 240 min), and no peculiar complications was observed. Due to the risk of severe hypocalcemia, safety is a major concern when regional citrate anticoagulation or calcium-free dialysate is used. In our institutional protocol, ionized calcium is measured before the session in patients at risk of systemic hypocalcemia (for instance, acute pancreatitis, rhabdomyolysis, acute or chronic liver failure) and then at 3–4 hours only if the dialysis time will be extended over 4 hours. Ionic dialysance (ID) is sufficiently accurate for measurement of instantaneous clearance of the small molecules. Therefore, the amount of calcium transfer can be well predicted without the need of ionized calcium measurement in most patients. However, the main pitfall is the discontinuous measurement of the ID occurring every 15–30 minutes. During this time, transient drop of the instantaneous clearance can occur because of catheter dysfunction or severe hypotension. Thus, to reduce the risk of severe hypocalcemia related to mistaken measure of the ID and subsequent inadequate calcium reinjection, we recommend maintaining a minimal calcium infusion rate (i.e., 85% of the initial rate) for at least two ID measures. Whether the use of continuous assessment of the instantaneous clearance by spectrophotometric analysis of uremic toxins available in some dialyzer may allow a more accurate calcium reinjection needs to be addressed.