A retrospective study of indications and consequences of monitoring direct oral anticoagulant plasma concentrations on patient care in a university hospital: The Retro-AOD study

INTRODUCTION

The use of direct oral anticoagulants (DOAC) is increasing. Specific concentrations are available and have been proven to be reliable and reproducible in optimising patient care. This retrospective, monocentric study aimed to describe the indications and consequences of monitoring DOAC plasma levels on patient care.

MATERIALS AND METHODS

We collected data of patients hospitalised at the Bordeaux University Hospital between January 2017 and December 2018. These included demographics, indications, type, dose of DOAC, standard coagulation tests, creatinine clearance and DOAC plasma concentration using specifically calibrated rivaroxaban and apixaban anti-Xa and dabigatran anti-IIa assays. The date of last DOAC intake, the time between intake and plasma level measurement were also collected and analysed.

RESULTS

A total of 2197 DOAC assays in 1488 patients were obtained in various clinical situations: urgent or elective procedures, context of acute renal failure, suspicion or occurrence of ischemic strokes, intra-cranial and other bleeding sites. Interpretation of these assays led physicians to maintain, postpone or cancel invasive and high haemorrhagic risk procedures in 757, 261 and 56 cases respectively. The remaining 1123 assays were associated with no significant modification of patient care. DOAC plasma concentration was ≤30 ng ml^-1 (sensitivity 85.4%, specificity 73.6%, positive predictive value 71.1%, negative predictive value 86.7%, AUC 0.81) after a last intake of at least 2 days.

CONCLUSIONS

Our study is, to date, the largest report of real-life measurement of specific DOAC plasma level at a single institution. Patient care was not modified in more than half of the assays.

4 hour-stability of anti-Xa activity in unspun citrated tubes

Current guidelines recommend performing laboratory tests aimed at monitoring unfractionated heparin (UFH) treatments within a delay not exceeding 1 to 2 h(s) after sampling when blood is collected into citrated tubes. As such a short delay could be an issue, we evaluated the potential impact of longer delays. For that purpose, two citrated tubes were obtained from patients on UFH: one was centrifuged and tested for anti-Xa activity and aPTT within 1 h after collection (T1 h) and one was stored for 4 h at room temperature (T4 h) before being processed. A total of 123 paired tubes were investigated. Anti-Xa activity was significantly lower at T4 h than at T1 h, with a mean bias, calculated according to Bland-Altman, of 0.05 IU/mL. Considering 0.30 to 0.70 IU/mL as the therapeutic range, there were 12 cases of discrepant test results (9.8%). Most of them being around the lower limit of the therapeutic range had no impact on patients' management. APTT was significantly shortened (p < 0.0001) at T4 h vs. T1 h, with a mean bias of -7.9 s. Considering anti-Xa correlated aPTT therapeutic range, 29 cases of discrepant test results (23.6%) were found, 10% would have induce dosage changes. The concordance between anti-Xa activities measured at T4 h and T1 h was excellent (kappa = 0.813) and good for aPTT (kappa = 0.661). In conclusion, extending the delay between blood collection and measurement of tests prescribed for monitoring UFH therapy up to 4 h was found to lead to a systematic reduction in both anti-Xa activity and aPTT in unspun citrated tubes. As changes at T4 h were limited and had few clinically relevance than the ones observed with aPTT testing, a 4 h-delay was found to be acceptable for anti-Xa activity. The maximum delay for aPTT should remain around 1-2 h as changes were more relevant.

Agreement between activated partial thromboplastin time and anti-Xa activity in critically ill patients receiving therapeutic unfractionated heparin

BACKGROUND

No study supports the use of either aPTT or anti-Xa activity for heparin monitoring in critical care patients. There are no strong data on the agreement between aPTT and anti-Xa. The aims of this study were to: 1. Analyse the agreement between aPTT and anti-Xa in a large population of critically ill patients under unfractionated heparin therapy (UFH), 2. Identify clinical and biological factors associated to agreement or disagreement, and 3. Analyse the impact of anti-Xa availability on the use of aPTT and UFH therapy.

METHODS

Retrospective study in a 35 beds mixed-ICU population between 2006 and 2016 in a University teaching hospital.

INCLUSION CRITERIA

delivery of a UFH dose >15,000 U/24 h during at least one day with one anti-Xa determination.

DATA

demographic variables, aPTT, anti-Xa, laboratory variables, presence of extracorporeal devices (ECD). Pairs of simultaneously dosed aPTT and anti-Xa [aPTT:anti-Xa] were analysed on the basis of their agreement within the sub-therapeutic, therapeutic (aPTT 50-80″, anti-Xa 0.3-0.7 U/ml) or supra-therapeutic ranges.

RESULTS

2283 patient admissions (2085 patients) were analysed. 35,595 [aPTT:anti-Xa] pairs were found. The overall [aPTT:anti-Xa] agreement was 59.6% and lowest (54.3%) in presence of ECD compared to non-ECD patients (61.6%; p < 0.001). Sixteen demographic and biological variables were analysed and were not predictive of [aPTT:anti-Xa] agreement. No significant difference in administered UFH dose was observed after anti-Xa introduction.

CONCLUSION

In this large cohort, the [aPTT:anti-Xa] agreement is <60% and significantly lower in patients with ECD. None of the variables identified as potentially affecting the agreement were predictive. Availability of anti-Xa had neither effect on aPTT use nor on UFH-dose. These results call for a prospective study to determine the optimal UFH-therapy monitoring tool.

The Effect of Obesity on Anti-Xa Concentrations in Bariatric Patients

BACKGROUND

Morbidly obese patients are at increased risk to develop venous thromboembolism (VTE), especially after bariatric surgery. Adequate postoperative thrombosis prophylaxis is of utmost importance. It is assumed that morbidly obese patients need higher doses of low molecular weight heparin (LMWH) compared to normal-weight patients; however, current guidelines based on relative efficacy in obese populations are lacking.

OBJECTIVES

First, we will evaluate the relationship between body weight descriptors and anti-Xa activity prospectively. Second, we will determine the dose-linearity of LMWH in morbidly obese patients.

SETTING

This study was performed in a general hospital specialized in bariatric surgery.

METHODS

Patients were scheduled for a Roux-en-Y gastric bypass with a total bodyweight (TBW) of ≥ 140 kg. Patients (n = 50, 64% female) received a daily postoperative dose of 5700 IU of nadroparin for 4 weeks. Anti-Xa activity was determined 4 h after the last nadroparin administration. To determine the dose linearity, anti-Xa was determined following a preoperative dose of 2850 IU nadroparin in another 50 patients (52%).

RESULTS

TBW of the complete group was 148.5 ± 12.6 kg. Mean anti-Xa activity following 5700 IU nadroparin was 0.19 ± 0.07 IU/mL. Of all patients, 32% had anti-Xa levels below the prophylactic range. Anti-Xa activity inversely correlated with TBW (correlation coefficient - 0.410) and lean body weight (LBW; correlation coefficient - 0.447); 67% of patients with a LBW ≥ 80 kg had insufficient anti-Xa activity concentrations. No VTE events occurred.

CONCLUSIONS

In morbidly obese patients, a postoperative dose of 5700 IU of nadroparin resulted in subprophylactic exposure in a significant proportion of patients. Especially in patients with LBW ≥ 80 kg, a higher dose may potentially be required to reach adequate prophylactic anti-Xa levels.

Comparison of three different anti-Xa assays in major orthopedic surgery patients treated with direct oral anticoagulant

Background

Measurement of edoxaban plasma concentration has been gathering attention in major orthopedic surgery patients receiving edoxaban for the prevention of venous thromboembolism (VTE).

Methods

The anti-Xa activity was measured one hour after edoxaban intake using 3 different assays in 200 patients after major orthopedic surgery.

Results

The anti-Xa activities on Day 8 were significantly higher than those on Day 4 and those on Day 4 were significantly higher than those on Day 1. The anti-Xa activities in two assays closely correlated with each other, but the other anti-Xa assay did not correlated with other two assays. The anti-Xa activities as detected in the three Xa assays were significantly higher in the patients without deep vein thrombosis (DVT) than in those with DVT on Day 4. Additionally, there were no significant differences in the anti-Xa activities of assays A, B and C between patients with and without massive bleeding (MB) on Days 1, 4, 8 and 15.

Conclusion

The results of this study suggest that anti-Xa level could be predictive of the risk of VTE, but not of the risk of massive bleeding.

Empirically Reduced Dosages of Tinzaparin in Patients with Moderate-to-Severe Renal Insufficiency Lead to Inadequate Anti-Xa Levels

BACKGROUND

Due to the higher molecular weight of tinzaparin, the low molecular weight heparin (LMWH) is less dependent on renal excretion than other LMWH preparations. However, several international guidelines recommend the same preemptive dosage reduction for all therapeutic dose LMWHs prescribed in renal insufficient patients, to ensure that there is no accumulation of anticoagulant activity and increased risk of bleeding. This study is aimed at assessing whether a preemptive dosage reduction of tinzaparin in all renal insufficient patients (comprising 25% reduction in patients with Modification of Diet in Renal Disease - estimated glomerular filtration rate (MDRD-eGFR) 30-60 mL/min/1.73 m2 and 50% reduction in patients with MDRD-eGFR <30 mL/min/1.73 m2) leads to adequate anti-Xa levels.

METHODS

We selected the anti-Xa levels of in-hospital patients (≥18 years) with moderate-to-severe renal insufficiency (MDRD-eGFR <60 mL/min/1.73 m2), on therapeutic dosages of tinzaparin. Anti-Xa levels were measured using a chromogenic assay.

RESULTS

Preemptive dosage reduction resulted in a median anti-Xa activity of 0.50 IU/mL (interquartile range [IQR] 0.38-0.60). In 92.3% of patients the anti-Xa level was below the target anti-Xa level of >0.85 IU/mL for therapeutic indications. Unadjusted dosages led to a median anti-Xa activity of 0.74 IU/mL (IQR 0.56-0.92). The preemptive dosage reduction was significantly associated with anti-Xa activity below therapeutic range (p = 0.007). No difference in anti-Xa activity was observed between patients with moderate (0.71 IU/mL, IQR 0.61-0.95) versus severe (0.65 IU/mL, IQR 0.41-1.06) renal insufficiency in whom an unadjusted dose had been administered (p = 0.77). None of the anti-Xa levels were above the upper margin of the presumed therapeutic range of 2.0 IU/mL.

CONCLUSION

In renal insufficient patients, the preemptive dosage reduction of tinzaparin leads to inadequate anti-Xa levels.

Comparison of three different anti-Xa assays in major orthopedic surgery patients treated with fondaparinux

Anti-Xa assays are useful for monitoring the effects of selective anti-Xa drugs, such as fondaparinux, in the prophylaxis of deep vein thrombosis. In the present study, anti-Xa activity was measured using three different assays, Testzym(®) Heparin S, STA(®)-Liquid Anti-Xa and HemosIL(®) Liquid Heparin. Anti-Xa activity in each assay gradually increased from day one after administration to day eight, and still remained on day 15. Although there were significant differences in anti-Xa activity among the three assays, the activity showed significant correlation across assays. There were no significant differences in the anti-Xa activity between patients with and without DVT or between patients with and without massive bleeding on day one before and after administration, day four, day eight and day 15. Anti-Xa activity in each assay was weakly correlated with antithrombin (AT) activity. The AT activity in patients were significantly higher on days four, eight and 15 compared with day one before and after administration, suggesting that AT activity increases following the administration of fondaparinux. The three anti-Xa assay kits tested are useful for monitoring fondaparinux treatment in orthopedic surgery patients.

The relationships among hemostatic markers, the withdrawal of fondaparinux due to a reduction in hemoglobin and deep vein thrombosis in Japanese patients undergoing major orthopedic surgery

BACKGROUND

The relationships among the hemostatic markers, the development of deep vein thrombosis (DVT) and the withdrawal of fondaparinux due to a reduction in the hemoglobin levels were examined.

METHODS

Two-hundred twenty-one Japanese patients who underwent major orthopedic surgery and were treated with 1.5mg of fondaparinux instead of 2.5mg of fondaparinux were studied. Forty-seven of 221 patients discontinued fondaparinux treatment (withdrawal group) and 37 patients developed DVT.

RESULTS

The age, frequency of total knee arthroplasty (TKA), withdrawal of fondaparinux, reduction of hemoglobin and the plasma levels of soluble fibrin (SF), D-dimer and fibrinogen and fibrin degradation product (FDP) on day 1 after the operation were significantly higher in the patients with DVT. Elevated SF, D-dimer or FDP levels were associated with the risk for DVT. The age, frequency of TKA or DVT, anti-Xa activity and the creatinine, FDP and D-dimer levels were significantly higher in the withdrawal group. An anti-Xa level >0.33 mg/l and an elevated D-dimer or FDP level were associated with the risk of withdrawal.

CONCLUSION

The age and SF levels, TKA and withdrawal of fondaparinux were related to the risk of DVT, and the anti-Xa activity, creatinine level and DVT were related to the risk of withdrawal of fondaparinux due to a reduction in hemoglobin.