The neuroradiological findings and its outcomes of intracerebral hemorrhage (ICH) were compared between the non–vitamin K antagonist oral anticoagulant (NOAC) therapy and warfarin therapy. In the latest 3 years, 13 cases of nonvalvular atrial fibrillation on NOAC therapy were admitted for ICH. For comparison, 65 age- and gender-comparable patients with ICH on warfarin therapy were recruited. Three NOACs had been prescribed: dabigatran (n = 4), rivaroxaban (n = 2), and apixaban (n = 7). The average ages were 76 ± 9 and 78 ± 8 years in the warfarin (n = 65) and NOAC groups (n = 13), respectively. There was no difference in the clinical features, including the CHADS2 score or HAS-BLED score: 2.62 ± 1.31 versus 2.62 ± 1.33, or 1.09 ± 0.43 versus 1.00 ± 0.41, for the warfarin and NOAC groups, respectively. The volume of ICH <30 ml was found in 84.6% of the patients on NOACs, but it was found in 53.8% of the patients on warfarin (p = 0.0106). The expansion of hematoma was limited to 7 patients (10.8%) of the warfarin group. A lower hospital mortality and better modified Rankin Scale were observed in the NOAC group than in the warfarin group: 1 (7.7%) versus 27 (41.5%; p = 0.0105) and 3.2 ± 1.4 versus 4.5 ± 1.6 (p = 0.0057), respectively. In conclusion, ICH on NOAC therapy had smaller volume of hematoma with reduced rate of expansion and decreased mortality compared with its occurrence on warfarin.
Anticoagulant therapy of nonvalvular atrial fibrillation (NVAF) using non–vitamin K antagonist oral anticoagulants (NOACs) have been shown to be superior or not inferior in efficacy compared with warfarin, but direct comparisons of the neuroradiological findings of intracerebral hemorrhage (ICH) between the patients on warfarin and NOAC therapy are limited. We analyzed patients with ICH occurring during anticoagulation therapy using warfarin or NOACs, and the clinical features, the neuroradiological findings, and the outcomes were compared.
Methods
This study was approved by the Institutional Board of Review at Tachikawa Medical Center. NOACs became available in 2011 in this country, and 13 patients with NVAF with ICH developing during NOAC therapy had been referred to the neurosurgery department of our hospital, Tachikawa General Hospital, in 2015 and 2016. Intracranial hematoma caused by trauma and patients with epidural or subdural hematoma or subarachnoid hemorrhage were excluded from this study. NOACs were prescribed according to the recommendations of each drug considering age, body weight, and renal function.
We had had 78 patients with NVAF who developed ICH during warfarin therapy in the latest 6 years. Of these, the patients outside the age strata of the patients on NOAC were excluded, and remaining 65 patients were studied for comparison: 42 patients had ICH before the NOAC era and 23 patients had ICH during the same period of time as the NOAC group. All had been directly admitted and treated in our hospital. The target prothrombin time international normalized ratio (PT-INR) was set according to the Japan Guideline of treatment of NVAF: at 1.6 to 2.6 of PT-INR for patients with age ≥70 years. All patients were transferred directly to our hospital and admitted.
ICH was diagnosed by brain computed tomography (CT), and warfarin or NOACs were stopped after admission. Warfarin was reversed immediately by intravenous administration of vitamin K. Blood pressure >160 mm Hg on admission was lowered by intravenous nicardipine, a calcium antagonist, to 140 mm Hg or below.
The volume of hematoma was calculated from the slices of CT images in each patient, and the expansion of hematoma was checked in all patients by repeated brain CT within 24 hours. When neurological deterioration was observed, brain CT was performed at that timing. Surgery for aspiration and/or drainage was indicated on the basis of the repeated brain CT finding.
Blood was withdrawn for compete blood counts, blood chemistry, and serological testing. Prothrombin time (PT) and activated partial thromboplastin time were measured routinely, and PT-INR was checked in the patients on warfarin therapy. For risk evaluation of stroke or hemorrhage, CHADS2 and HAS-BLED scores were used.
The clinical and neuroradiological findings and the clinical outcomes were compared between patients with ICH on warfarin or NOAC therapy. The volume of hematoma, expansion rate of ICH, the hospital mortality and the functional outcomes by the modified Rankin Scale were compared between the 2 groups at the time of discharge.
The continuous data are presented as means ± SD, and the categorical variables are expressed as absolute numbers or percentages. Statistical comparisons between groups were performed using the t test or analysis of variance for continuous variables or Pearson’s chi-square test for categorical variables. The distribution pattern of ICH was also compared by the Pearson’s chi-square test (2 × 7 contingency table). JMP software (Statistical Discovery Software, version 5.0.1 J; SAS Institute, Cary, North Carolina) was used to complete the statistical analysis. A 2-sided test with p <0.05 was considered statistically significant.
Results
Three NOACs had been prescribed in 13 patients: dabigatran at 110 mg twice a day (n = 2) or 150 mg twice a day (n = 2), rivaroxaban at 10 mg/day (n = 2), and apixaban at 2.5 mg twice a day (n = 5) or 5 mg twice a day (n = 2). Of the 65 patients with ICH on warfarin therapy, 42 patients who had ICH before the NOAC era showed the same clinical features as 23 patients who were treated in NOAC era. Although the type of AF, PT-INR, and PT were different between the patients on NOACs and warfarin therapy, other clinical features including CHADS2 and HAS-BLED scores were not different between the 2 groups ( Table 1 ).
| Warfarin (n= 65) | NOACs (n=13) | P-Value | |
|---|---|---|---|
| Age (years) | 76.1±8.9 | 77.5±8.0 | 0.5697 |
| Men | 40 (61.5 %) | 8 (61.5%) | 1.0000 |
| Chronic atrial fibrilation | 56 (86.2%) | 5 (38.5%) | 0.0005 |
| Creatinine clearance (ml/min) | 56.6±28.8 | 57.7±27.7 | 0.9258 |
| Systolic blood pressure (mmHg) | 163±21 | 159±21 | 0.5684 |
| Diastolic blood pressure (mmHg) | 92±18 | 82±23 | 0.1336 |
| Congestive heart failure | 30 (46.2%) | 3 (23.1%) | 0.1133 |
| Hypertension | 50 (76.9%) | 9 (69.2%) | 0.5635 |
| Age >75 years | 37 (56.9%) | 9 (69.2%) | 0.4037 |
| Diabetes mellitus | 15 (23.1%) | 9 (69.2%) | 1.0000 |
| Previous stroke | 19 (29.2%) | 6 (46.2%) | 0.2435 |
| Antiplatelets | 22 (33.8%) | 2 (15.4%) | 0.1652 |
| PT-INR | 2.24±0.73 | – | – |
| Activated partial thromboplastin time (sec) | 40.3±9.4 | 36.4±7.0 | 0.1541 |
| Prothrombin time (sec) | 25.0±7.4 | 1.09±0.43 | <0.0001 |
| CHADS2 score | 2.62±1.3 | 2.6±1.3 | 1.0000 |
| HAS-Bled score | 1.09±0.43 | 1.00±0.41 | 0.8088 |
Warfarin and NOACs were withdrawn after admission, and vitamin K was administered to all patients to reverse the PT-INR value to 1.0 or below. Systolic blood pressure ≥160 mm Hg was treated with nicardipine (5 on warfarin and 3 on NOAC).
The location distribution of ICH was not different between the 2 groups (p = 0.4831, Figure 1 ). Each site of bleeding was associated with death in the warfarin group, but only one patient on NOAC died from brain stem bleeding. The volume of ICH at the time of admission was larger in the warfarin group than the NOAC group but nonsignificantly: 41.0 ± 44.2 ml (median = 32.0 ml) versus 20.7 ± 35.1 ml (median = 5.0 L; p = 0.0762, Table 2 ). However, the volume of ICH was <30 ml was more often in the NOAC group ( Table 2 , Figure 1 ). In the warfarin group, the volume of ICH was not different before (n = 42) and during NOAC era (n = 23): 38.2 ± 43.0 ml versus 51.7 ± 53.7 ml, respectively (p = 0.2945).
| Warfarin (n=65) | NOACs (n=13) | P-Value | |
|---|---|---|---|
| Volume of hematoma (mean±SD) | 41.9±46.9 | 20.7±35.1 | 0.0762 |
| Hematoma <30ml | 35 (53.8%) | 11 (84.6 %) | 0.0106 |
| Hematoma expansion | 7 (10.8 %) | 0 | 0.1140 |
| Hospital mortality | 27 (41.5 %) | 1 (7.7 %) | 0.0105 |
| modified Rankin scale | 4.5±1.6 | 3.2±1.4 | 0.0057 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
