Drug concentrations

As stated by the authors, the proposed threshold values for dabigatran or rivaroxaban do not stem from clinical observations and are based largely on the drug’s pharmacokinetics . According to the European Medicine Agency Annexes Summary of Product Characteristics for both drugs , the steady-state concentration of plasma dabigatran expected on the basis of the interindividual variability spans from 117 ng/mL to 275 ng/mL at the peak level, and from 61 ng/mL to 143 ng/mL at the trough level, both taken as the 25th to 75th percentile range . Likewise, the plasma concentrations of rivaroxaban vary from 22 ng/mL to 535 ng/mL at the peak, and from 6 ng/mL to 239 ng/mL at the trough level, both taken as the 90% prediction interval . It is unclear where these values come from and how they have been determined. However, one may assume that they come from the clinical trials and that the method of determination was mass spectrometry. Also, it was not specified whether these values were or not associated with clinical bleeding.

All in all, the above observations point to the fact that threshold values for safe emergency surgery are presently unknown. If this is the case, we wonder whether providing threshold values (although with some caveats) as the authors did in their proposal is not detrimental to patient management as these values may give a false reassurance to those who are not particularly familiar with the effect of DOACs.

PT and aPTT

The threshold presurgical prolongations for PT and aPTT provided by the working group (i.e. a ratio < 1.2) are based largely on considerations stemming from the relative half-life of the circulating drugs and on the relative responsiveness of aPTT or PT reagents stemming from results obtained by spiking in-vitro normal plasmas with known and increasing concentrations of either drug . It should be mentioned that the situation might be different when one investigates plasmas from patients treated with either drug. Such observations are now available. Hawes et al. have in fact investigated plasmas from patients treated with dabigatran. According to their results, the PT and aPTT are often normal in spite of therapeutic dabigatran plasma levels. Furthermore, there was a relatively large between–reagent variability in the response to dabigatran. As an example, when the aPTT was run with a particular reagent, 35% of the aPTT ratios were in the normal range when the dabigatran plasma concentration was up to 100 ng/mL. Rodgers et al. investigated patients on rivaroxaban and found that the recombinant thromboplastin they used was adequately responsive to rivaroxaban. On the contrary, van Veen et al. , while investigating their patients on rivaroxaban with another recombinant thromboplastin, found that it was unresponsive to rivaroxaban. These results clearly demonstrate that providing presurgical cut-off values without taking into proper consideration the responsiveness of the aPTT or PT tests to dabigatran or rivaroxaban may create a dangerous situation for patient management.