Original Research Paper

Warfarin associated intracerebral hemorrhage in Hong Kong Chinese K. C. Teo1, N. R. Mahboobani1, R. Lee2, C. W. Siu1,3, R. T. F. Cheung1,3, S. L. Ho1,3, K. K. Lau1, K. H. Chan1,3 1

Department of Medicine, Queen Mary Hospital, LKS Faculty of Medicine, The University of Hong Kong, Department of Diagnostic Radiology, Queen Mary Hospital, LKS Faculty of Medicine, The University of Hong Kong, 3Research Center of Heart, Brain, Hormone and Healthy Aging, LKS Faculty of Medicine, The University of Hong Kong

2

Objectives: Warfarin-associated intracerebral hemorrhage (WICH) is a serious neurological condition associated with significant mortality and morbidity. We aimed to study the clinical features and factors that predict clinical outcome of Chinese patients with WICH. Methods: Medical records of patients with spontaneous intracerebral hemorrhage (ICH) admitted to our hospital between July 2001 and June 2010 were reviewed and those with WICH were studied in detail retrospectively. Results: Fifty-one patients with WICH were studied. The mean age was 74.3 ¡ 10.5 years and 52.9% of the patients were female. The mean international normalized ratio (INR) on presentation was 2.9 ¡ 1.0. The median ICH volume was 23.3 (10.4–59.3) ml. The mortality rate at 3–6 months for WICH was 62.0%. Multivariate logistic analysis revealed that an initial ICH volume of . 20 ml (OR 34.4, P 5 0.037) and presence of intraventricular hemorrhage (OR 22.9, P 5 0.046) were independently associated with poor outcome. Supratherapeutic INR (INR . 3.0) on admission (P 5 0.724) and complete correction of INR within 24 hours after admission (P 5 0.486) were not independent predictors of poor outcome. The median ICH volumes did not differ between INR groups (18.2 (9.4–61.1) ml for INR # 3 vs 27.3 (13.7–58.5) ml for INR . 3, P 5 0.718). Neurological deterioration (ND) was documented in 19 (63.3%) of the 30 patients included in a smaller sub-cohort, and was associated with poor neurological outcome (OR 20.7, P 5 0.027). Warfarin was resumed in 7 of the 20 survivors. There were two episodes of recurrent WICH and one episode of ischemic stroke during a mean follow-up duration of 5.4 years. In survivors who were not resumed on warfarin, there were two episodes of recurrent ICH and 12 episodes of ischemic vascular events (nine ischemic strokes) during a mean follow-up duration of 2.6 years. Conclusion: Warfarin-associated intracerebral hemorrhage is a very serious complication of warfarin therapy with high mortality and morbidity. Initial ICH volume, presence of intraventricular hemorrhage, and ND are independent predictors of clinical outcome. Keywords: Warfarin, Intracerebral hemorrhage, Warfarin-associated ICH, Mortality, ICH volume, Intraventricular extension, Neurological deterioration

Introduction Warfarin is the most commonly prescribed anticoagulant for primary and secondary prophylaxis of cardioembolism in patients with atrial fibrillation (AF), chronic rheumatic heart diseases, and prosthetic heart valves.1,2 The major side effect of warfarin is increased risk of hemorrhage including life threatening hemorrhage. Intracerebral hemorrhage (ICH) is the most serious form of warfarin associated hemorrhage accounting for y90% of death in patients with warfarin associated hemorrhage.1,3 Warfarin increases the risk of ICH by 7–10 fold.4 It has been Correspondence to: K. H. Chan, Department of Medicine, 4/F Professorial Block, Queen Mary Hospital, 102 Pokfulam Road, Hong Kong. Email: [email protected]

ß W. S. Maney & Son Ltd 2014 DOI 10.1179/1743132813Y.0000000275

estimated that warfarin associated intracerebral hemorrhage (WICH) accounts for about 10–12% of all ICH, and incidence of WICH was estimated to occur at 2–9 per 100 000 population per year.5 Recent evidence suggests that the proportion of WICH is likely increasing, hence WICH is expected to be a growing problem.6 Although, the risk of WICH increases with increasing intensity of anticoagulation monitored clinically by international normalized ratio (INR), remarkably two-thirds of WICH developed with INR within therapeutic range of 2.0–3.0. Warfarin associated intracerebral hemorrhage has worse prognosis and higher mortality than ICH not associated with anticoagulant therapy. Thirty-day mortality of up to 67% has been reported

Neurological Research

2014

VOL .

36

NO .

2

143

Teo et al.

Warfarin associated intracerebral hemorrhage

for WICH,7 compared to 34–51% for ICH not associated with anticoagulant therapy.8–10 This is partially explained by increased risk and prolonged duration of hematoma expansion with the anticoagulant effect in WICH,11 and hematoma expansion may occur even after full normalization of INR with administration of vitamin K (VAK) and fresh frozen plasma (FFP).12 Reported poor prognostic factors for WICH include INR . 3.0,13,14 large hematoma volume,14–16 intraventricular hemorrhage,17 advanced age,11 and low Glasgow coma scale (GCS) score on presentation.15,16 However, findings in different studies vary. In addition, data of WICH in Asians including Chinese are seriously lacking. We studied the clinical features and factors that predict poor clinical outcome of Chinese patients with WICH.

Patients and Methods Our hospital provides general hospital and specialist service to 0.6 million of the population in Hong Kong. The Accident & Emergency Department is open 24 hours a day on all days to the public. The ambulance service policy in Hong Kong aims to transfer patient with medical emergency to Accident & Emergency Department of the regional government hospital within the shortest distance. Records of all Chinese adults with a diagnosis of nontraumatic ICH at hospital discharge according to the International Statistical Classification of Disease and Related Health Problems during the period from July 2001 to June 2010 were reviewed. Patients with hemorrhagic cerebral infarct, brain tumor, aneurysm, vascular malformation, or head trauma were excluded. Medical records of patients with WICH were studied in detail. Patients who were taking other antithrombotic drugs concomitantly with warfarin, had a normal INR on presentation, or with an estimated glomerular filtration rate of less than 30 ml/minute/1.73 m2 were excluded. Data on demographic characteristics, putative risk factors, hematological and biochemical laboratory findings, and clinical course and outcome were abstracted. The first CT brain scan was reviewed. Intracerebral hemorrhage volumes were calculated by an independent consultant radiologist using General Electric Workstation (Version 4.4). The hematoma was traced manually to obtain the area of the hematoma of each slice of CT. The volume of hematoma was obtained by multiplying the area to individual slice thickness. Complete correction of INR was defined as INR # 1.3 at the last available test within 24 hours after admission, which was our local practice in view of the possible increased bleeding tendency associated with warfarin in Chinese. Patients who were not given

144

Neurological Research

2014

VOL .

36

NO .

2

INR reversal treatment were categorized into the group of incomplete correction of INR. For the clinical outcome, the modified Rankin scale (mRS) was recorded based on the follow-up notes at 3– 6 months. Thirty-five patients (69%) had mRS documented in medical records. For the remaining 16 patients (31%), the mRS was estimated from detailed information concerning their functional status in medical records. Subsequent use of antithrombotic, recurrent ICH, and ischemic vascular events are retrieved. Vascular events included ischemic stroke, systemic embolism, and myocardial infarction. To analyze the early clinical course of WICH, a separate cohort was created. Inclusion criteria for this cohort were: (1) initial examination performed within 24 hours after symptoms onset, and (2) GCS score of § 9. Neurological deterioration (ND) was defined as a decrease of two or more points in the GCS score or death within 24 hours of admission. CT brain findings were not included in this analysis because not all scans were made before ND.

Statistical analysis For univariate analysis, the independent sample t-test was employed for normally distributed continuous variables. The Mann–Whitney U test was used for the skewed variables. Categorical variables were compared using the chi-square test. Outcome models were developed for all WICH patients with good or poor outcome as dependent variable. Good outcome was defined as mRS 0–3 and poor outcome was defined as mRS 4–6.18 To explore the relationship of INR and ICH volume, ICH volume was stratified based on the presenting INR level into two groups: # 3 and . 3. Continuous variables were expressed as either mean¡standard deviation (SD) for parameters with normal distribution or median with interquartile range (IQR) for parameters with skewed distribution. Categorical data was expressed as number with percentages of the subtotal in parentheses. All statistical analysis was performed with SPSS version 19.0, and the P , 0.05 (2 tailed) was considered statistically significant. Variables with P , 0.10 in the univariable analysis were entered into the multiple logistic regression model. Continuous variables were dichotomized using the cut-off identified on receiver operator characteristic (ROC) curve. International normalized ratio on admission was also added into this model with a cut-off of . 3 as we would like to explore whether supratherapeutic INR was an independent predictor for poor outcome. Blood pressure (BP) measurements were not analyzed as an independent predictor of outcome as it is unclear whether abnormal BP was a cause or an effect of the WICH.

Teo et al.

Warfarin associated intracerebral hemorrhage

Table 1 Demographic, and clinical characteristic and outcome of 51 patients with warfarin-associated intracerebral hemorrhage (WICH) Mean age in years Sex Indication of warfarin Valve replacement Valvular AF Non–valvular AF DVT Medical history of hypertension Medical history of diabetes Median GCS on admission Mean INR on admission Site of ICH Lobar Deep cerebral Cerebellum Brainstem Intraventricular hemorrhage Median ICH volume (ml) Intraventricular hemorrhage ICH – event related mortality at 3–6 months# Duration to death Day 1 Day 2–Day 7 Day 8–Day 30 . Day 30 Clinical outcome at 3–6 months# Good neurological outcome (mRS 0–3) Poor neurological outcome (mRS 4–6)

74.3¡10.5 years 24 M (47.1%), 27 F (52.9%) 12 (23.5%) 8 (15.7%) 28 (54.9%) 3 (5.9%) 39/51 (76.5%) 11/51 (21.6%) 14 (7–15) 2.9¡1.0 16 (31.4%) 20 (39.2%) 7 (13.7%) 6 (11.8%) 2 (3.9%) 23.3 (10.4–59.3) 31/51 (60.8%) 31/50 (62.0%) 15 11 3 2

(48.4%) (35.5%) (9.7%) (6.4%)

14 (28.0%) 36 (72.0%)

Number with percentages of subtotal in parentheses is used for categorical variables and mean¡SD or median interquartile range (IQR) for numerical variable. # Excluding one patient for poor documentation of the outcome. AF: atrial fibrillation; DVT: deep vein thrombosis; GCS: Glasgow coma scale; INR: international normalized ratio; ICH: intracerebral hemorrhage; mRS: modified Rankin scale.

Results

Predictors of clinical outcome

A total of 59 WICH patients were identified during the period. A total of eight patients were excluded, three had concomitant use of other antithrombotic drugs and warfarin, two had a normal INR on presentation, and three had an estimated glomerular filtration rate of less than 30 ml/minute/1.73 m2. Data were missing in one patient for smoking history, one for initial BP measurement, two for complete blood count, and four for ICH volume (missing scans). One patient had low anticoagulation ratio for the admission INR and was treated as a missing INR value. One patient was excluded from outcome analysis owing to poor documentation of the outcome.

Table 2 shows clinical characteristic of patients divided into good and poor outcome group. There was no significant difference in the mean INR for the good outcome and poor outcome group (3.1 ¡ 1.4 vs 2.9 ¡ 0.9, P 5 0.562). Poor outcome group has lower first recorded GCS score (11 (5–15) vs 15 (15–15), P 5 0.001), larger ICH volume (53.8 (15.6–75.8) ml vs 11.3 (4.8–15.7) ml, P 5 0.000), higher maximum systolic blood pressure (SBP) at day 1 (185.6 ¡ 37.5 mmHg vs 164.9 ¡ 20.3 mmHg, P 5 0.017), and higher rate of intraventricular hemorrhage (77.8 vs 21.4%, P 5 0.000). Results of the multivariate logistic regression analysis showed that ICH volume of . 20 ml (OR 34.4, P 5 0.037) and presence of intraventricular hemorrhage (OR 22.9, P 5 0.046) were independent predictors for poor neurological outcome (Table 3). Supratherapeutic INR (P 5 0.724) and complete correction of INR within 24 hours after admission (P 5 0.486) were not independent predictors for neurological outcome.

Baseline characteristic and outcome Major clinical and neuroradiological characteristics of patients with WICH are summarized in Table 1. Of the 51 patients who were analyzed, 28 (54.9%) were taking warfarin for non-valvular AF, 12 (23.5%) for valvular replacement, 8 (15.7%) for valvular AF, and 3 (5.9%) for deep vein thrombosis. The mean INR on presentation was 2.9 ¡ 1.0. The median ICH volume was 23.3 (10.4–59.3) ml. The mortality rates at 30 days and 3–6 months were 58.0% and 62.0% respectively. Most of the death (15/31, 48.4%) occurred on day 1 of admission. Only 14 patients (28.0%) had good neurological outcome.

Relationship of INR and ICH volume There was no difference in the median ICH volume by INR categories (18.2 (9.4–61.1) ml for INR # 3 vs 27.3 (13.7–58.5) ml for INR . 3, P 5 0.736).

Neurological deterioration in WICH Thirty patients were included in this cohort. Neurological deterioration occurred in 19 (63.3%)

Neurological Research

2014

VOL .

36

NO .

2

145

Teo et al.

Warfarin associated intracerebral hemorrhage

Table 2 Clinical characteristics of warfarin-associated intracerebral hemorrhage (WICH) patients divided into good and poor outcome group and results of univariable analysis (n 5 50) Parameters

Good outcome (N 5 14)

Poor outcome (N 5 36)

P value

Age, years Female sex History of smokingˆ History of HT History of DM 1st recorded GCS score SBP on adm (mmHg) DBP on adm (mmHg) Mean blood pressure (BP) on adm (mmHg) Hemoglobin on adm (109/l) Platelet count on adm (109/l) Creatinine (mmol/l) INR on adm Time from symptom onset to CT brain (hours) Infratentorial ICH Intraventricular hemorrhage Median ICH volume (ml) Complete correction of INR within 24 hours after admission# Ischemic stroke after correction of coagulopathy SBP D1 (mm Hg) DBP D1 (mm Hg) Maximum SBP D1 (mm Hg) MAP D1 (mm Hg)

70.6¡12.7 6/14 (42.9%) 3/14 (21.4%) 11/14 (78.6%) 2/14 (14.3%) 15 (15–15) 157.4¡23.5 85.1¡15.9 109.2¡17.6 14.0¡1.6 193.4¡53.6 101.8¡23.4 3.1¡1.4 21.3 (11.9–120.3) 2/14 (14.3%) 3/14 (21.4%) 11.3 (4.8–15.7) 5/14 (35.7%) 0/14 (0%) 146.5¡18.2 78.4¡13.9 164.9¡20.3 101.1¡14.3

75.6¡9.5 21/36 (58.3%) 10/35 (28.6%) 27/36 (72.7%) 9/36 (25.0%) 11 (5–15) 165.8¡45.6 83.4¡24.1 110.9¡29.4 13.0¡2.3 223.6¡68.1 103.7¡26.7 2.9¡0.9 5 (3–10.3) 11/36 (30.6%) 28/36 (77.8%) 53.8 (15.6–75.8) 3/28 (10.7%) 3/36 (8.3%) 152.6¡33.8 81.0¡20.3 185.6¡37.5 104.9¡23.9

0.199 0.324 0.609 0.791 0.412 * 0.001 0.519 0.809 0.846 0.153 0.157 0.834 0.562 * 0.000 0.239 * 0.000 * 0.000 0.052 0.265 0.419 0.659 * 0.017 0.582

Number with percentages of subtotal in parentheses is used for categorical variables and mean¡SD or median interquartile range (IQR) for numerical variable. * P value , 0.05 in univariate analysis. ˆ1 missing smoking history. # 8 missing INR value either because of early death or no repeated INR value available. HT: hypertension; DM: diabetes mellitus; GCS: Glasgow coma scale; adm: admission; INR: international normalized ratio; ICH: intracerebral hemorrhage; D1: day 1; SBP: systolic blood pressure; DBP: diastolic blood pressure; MAP: mean arterial pressure.

patients. Age (71.7 ¡ 9.1 vs 77.5 ¡ 4.8, P 5 0.062), 1st recorded GCS (14 (11–15) vs 15 (14–15), P 5 0.146), and INR on presentation (2.7 ¡ 0.8 vs 2.5 ¡ 0.5, P 5 0.547) did not differ between the ND and nonND group. Neurological deterioration group has a higher mean SBP (164.6 ¡ 27.2 mmHg vs 143.4 ¡ 18.1 mmHg, P 5 0.029), higher mean diastolic blood pressure (DBP) (92.1 ¡ 16.7 mmHg vs 74.5 ¡ 14.3 mmHg, P 5 0.007), and higher mean arterial pressure (MAP) (116.3 ¡ 18.9 mmHg vs 97.5 ¡ 15.3 mmHg, P 5 0.009) on day 1. After adjusting to age, sex, and 1st recorded GCS score, ND was associated with poor neurological outcome in patient with WICH (OR 20.7, 95% CI: 1.5–284.4, P 5 0.027).

INR reversal, recurrent ICH, and ischemic vascular events International normalized ratio reversal was attempted in 40 WICH patients, mostly with FFP (87.1%) ¡ intravenous VAK. This was initiated

within 12 hours of symptom onset in 28 (70.0%) patients and within 24 hours of symptom onset in 32 (80.0%) patients. Patients who did not receive correction of INR had worse conscious level on presentation (median GCS 15 (11–15) vs 4 (3–14), P 5 0.001). Three patients had in-patient ischemic stroke after correction of INR (3/40, 7.5%). There were a total of 20 survivors. Seven patients were resumed on warfarin, four were given antiplatelet agents while nine were not given any antithrombotic. Four patients were later prescribed with an antiplatelet agent after recurrent ischemic strokes. Patients who were resumed on warfarin included six for valvular replacement and one for non valvular AF with a CHADS score of six. Mean time to resumption of warfarin from WICH was 13.3 ¡ 5.7 days and the mean follow-up duration was 5.4 ¡ 2.5 years. Only one patient had recurrent WICH (both first and recurrent WICH were lobar ICH) but with two

Table 3 Multivariate logistic regression analysis for independent predictors for poor clinical outcome

ICH volume of . 20 ml Intraventricular hemorrhage First recorded GCS of 15 INR . 3 Complete correction of INR within 24 hours after admission *

Unadjusted OR

Adjusted OR

15.3 12.8 0.08 2.4 0.2

34.4 22.9 0.06 1.8 0.2

95% CI 1.238–956.697 1.059–494.169 0.003–1.113 0.078–39.193 0.005–12.835

P value *

0.037 0.046 0.059 0.724 0.486

*

P value , 0.05 in multivariate logistic regression. CI: confidence interval; OR: odds ratio; ICH: intracerebral hemorrhage; GCS: Glasgow coma scale; INR: international normalized ratio.

146

Neurological Research

2014

VOL .

36

NO .

2

Teo et al.

recurrent episodes and there was one ischemic vascular event (right middle cerebral artery territory infarction with an INR of 1.8). The mean follow-up duration for patients who were not resumed on warfarin was 2.6 ¡ 2.2 years. For patients who were not given any antithrombotic, there were one recurrent ICH and seven ischemic vascular events (all ischemic stroke) during a mean follow-up duration of 2.9 ¡ 2.7 years. For patients who were subsequently given antiplatelet agent, there were one recurrent ICH and five ischemic vascular events (two ischemic strokes, one transient ischemic attack, and two ischemic bowel events) during a mean follow-up duration of 2.5 ¡ 1.5 years.

Discussion Data from our in-hospital registry suggest that about 20–40% of local elderly patients (above 65 years) with AF are treated with warfarin, and the number is increasing over the past 15 years (unpublished data). The demographics, associated medical disorders and sites of hemorrhage of our patients are consistent with other WICH reports.11–17,19–21 The mortality rate of WICH in our cohort at 30 days and 3–6 months were 58.0% and 62.0% respectively, which is on the higher side of reported range of 43.0–67%.7,12,13,15,16,19–22 Only 28.0% of patients with WICH had good neurological outcome with the remaining 10.0% being survivors with severe neurological disabilities. Hence, 72.0% of our WICH patients had poor clinical outcome. This suggests that WICH in Chinese is a very serious form of stroke with high mortality and serious morbidity, which is similar to findings of WICH in Caucasians.7,12,13,15,16,19–22 Our results reveal that an ICH volume of . 20 ml and presence of intraventricular hemorrhage were independent predictors of poor clinical outcome for WICH. These are consistent with most of the previous studies reporting that the large ICH volume and presence of intraventricular hemorrhage are poor prognostic factors.14–17 The overall poor outcome of our patients is consistent with the median ICH volume of 23.3 ml which is more than 20 ml. The initial ICH volume did not differ significantly among patients with different initial INR categories, median ICH volume was 18.2 ml for INR # 3 versus 27.3 ml for INR . 3 (P 5 0.736). In addition, INR . 3 was not an independent predictor of poor clinical outcome. These are contrary to findings of recent studies, predominantly in Caucasians, which suggest that greater anticoagulation intensity is associated with larger initial ICH volume and higher mortality rate.13,14 Flaherty et al. reported that there was a trend toward a larger ICH volume by INR categories (INR , 1.2, 13.4 ml; INR 5 1.2–2.0, 9.3 ml; INR 5 2.1–3.0, 14.0 ml; INR . 3.0, 33.2 ml; P 5 0.10), and higher INR was associated with worse clinical outcome.14 We believe that patients

Warfarin associated intracerebral hemorrhage

with higher INR do have larger ICH volume and worse clinical outcome but statistically insignificant probably because of the small size of our cohort. Similar reason may explain why a supratherapeutic INR (INR . 3) was not associated with poor outcome. In WICH, rapid correction of INR can probably limit hematoma expansion. Hematoma in WICH can expand up to 24 hours in nearly 50% of patients.11,23–25 Older agents like FFP and VAK are effective in correction of INR but take hours to achieve complete correction of INR. Goldstein et al. reported that early time to treatment with VAK and FFP and successful INR correction did not improve clinical outcome of WICH patients.26 A recent study reported that failure to correct INR to # 1.3 one day following initiation of FFP was associated with higher 30-day mortality rate.27 Huttner et al. reported that advanced age, baseline hematoma volume, and occurrence of hematoma growth independently predicted poor outcome in WICH. In addition, they observed that prothrombin complex concentrate (PCC) was associated with a reduced incidence and extent of hematoma growth compared with FFP and VAK but had no significant impact on long-term (one year) clinical outcome. Interestingly, the difference between PCC- and FFP-treated patients disappeared if INR was completely reversed within 2 hours after admission.11 Although shown to be able to correct INR more rapidly, no studies addressing the effect of recombinant Factor VIIa on neurological outcome in WICH are identified.28–31 Complete correction of INR within 24 hours after admission was not associated with good neurological outcome. A plausible explanation is that only very early and aggressive correction of INR within a few hours of WICH onset can significantly limit hematoma expansion and improve clinical outcome.32,33 Late INR reversal may be ineffective to limit hematoma expansion. About 7.5% of our patients developed inpatient ischemic stroke after INR reversal. Hence, the risk of INR reversal must be weighed against the risk of elevated INR in WICH. The 1st recorded GCS was not shown to be a strong independent predictor for neurological outcome (P 5 0.059), but suggested a trend toward an independent predictor for neurological outcome. Similar observation had been reported.11 This was likely because of (1) the small number of our cohort, and (2) the large number of patients who presented with a GCS of $ 9 and developed ND (19/30, 63.3%). Although criteria for ND differed between studies, the frequency of ND in our cohort was much higher than the 21–33% in patients with ICH unrelated to warfarin.34–37 The higher percentage of ND in WICH could be explained by the higher risk of hematoma expansion.11,23–25 As ND was associated with poor

Neurological Research

2014

VOL .

36

NO .

2

147

Teo et al.

Warfarin associated intracerebral hemorrhage

prognosis as shown in our analysis and previous studies,34–36 the time period to the development of ND, may provide a window of opportunity for therapeutic measures to alter the outcome of WICH. In our study, BP recordings were not analyzed for independent predictor of outcome as it is unclear whether the abnormal BP was a cause or an effect of ICH. Patients with ND were documented to have higher mean SBP, DBP, and MAP at day 1. Only the maximum SBP at day 1 was statistically higher in patients with poor neurological outcome (185.6 ¡ 37.5 mmHg vs 164.9 ¡ 20.3 mmHg, P 5 0.017). This may be because of the higher ICH volume in these patients and hence higher maximum SBP owing to Cushing’s reflex. The mean SBP, MAP, and DBP in day 1 were not significantly different between both outcome groups. This was mainly because of the wide variability of BP in the patients of poor outcome group as evident by the wider SD. Blood pressure control in ICH remained controversial. Both ATACH and INT ERACT reported that early intensive lowering of BP down to 140 mmHg was feasible and may reduce hematoma growth but has no effect on clinical outcome.38,39 Recently, acute brain infarcts outside the hematoma as evident by decreased diffusion on diffusion weighted imaging of MRI were reported.18,40 Acute brain infarcts were associated with a larger reduction in BP from baseline during ICH and worse neurological outcome at 3 months.18 The clinical significance of acute brain infarcts in ICH remained unclear. The risk of recurrent ICH after primary ICH was reported to be 1–4% per year.41–46 Recent studies have shown that the risk of recurrent ICH and ischemic stroke was comparable in patients with prior history of ICH.41–43 On the contrary, Bailey et al. reported that recurrent ICH was more common than ischemic stroke, in which about three fourths of recurrent strokes were ICH.44 However, these studies did not study the effect of antithrombotic on recurrent stroke risk. In our cohort, there were disproportionately more ischemic vascular events (n 5 12, with nine ischemic strokes) than recurrent ICH (n 5 2) in patients who were not resumed on warfarin. Such observation could be explained by the fact that these patients were indicated for warfarin and had a high thrombotic risk. Only one out of seven patients who were resumed on warfarin had recurrent WICH (two recurrent episodes) during a mean follow-up duration of 5.4 years. The first and recurrent WICH were lobar ICH. As a result of the limited number of patients, no significant conclusion can be drawn on whether warfarin should be resumed after WICH. The decision should be made with careful evaluation of the risk of recurrent ICH (lobar ICH44 and uncontrolled hypertension47) and the risk of ischemic vascular events.

148

Neurological Research

2014

VOL .

36

NO .

2

Our study is retrospective in nature. Another limitation is that 4 out of 51 ICH volumes were missing. We also did not review any subsequent CT scans to assess hematoma expansion. However, any unmeasured confounder in standardization of care that may affect clinical outcome was minimized as all patients were managed within a single center. In addition, with the emergent and severe nature of WICH, we believe that the catchment of WICH within the catchment area of our hospital in the study period should be complete. In conclusion, our results show that WICH in Chinese is a very serious form of ICH with high mortality and morbidity. ICH volume of . 20 ml, presence of intraventricular hemorrhage, and ND are independent predictors of poor clinical outcome. Large-scale prospective clinical trials are required to confirm the prognostic value of the initial INR, and to address the issues of BP control, rapid correction of INR, and resumption of antithrombotic in WICH.

Disclosures Dr K. C. Teo, Dr N. R. Maboobani, Dr R. Lee, Dr S. L. Ho, Dr R. T. F. Cheung, and Dr K. K. Lau have no disclosures. Dr K. H. Chan has received funding from BayerHealth Care and Merck Serono for research in multiple sclerosis.

Acknowledgements The authors would like to thank Mr Stanley Yeung for statistical assistance. This study is supported by Seed Fund for Basic Science Research from the Faculty of Medicine of the University of Hong Kong.

References 1 Ansell J, Hirsh J, Hylek E, Jacobson A, Crowther M, Palareti G. Pharmacology and management of the vitamin K antagonists: American college of chest physicians evidence-based clinical practice guidelines (8th Edition). Chest. 2008;133(6 Suppl):160S–98S. 2 Hart RG, Aguilar MI. Anticoagulation in atrial fibrillation: selected controversies including optimal anticoagulation intensity, treatment of intracerebral hemorrhage. J Throm Thrombolysis. 2008;25:26–32. 3 Hart RG, Tonarelli SB, Pearce LA. Avoiding central nervous system bleeding during antithrombotic therapy: recent data and ideas. Stroke. 2005;36:1588–93. 4 Hart RG, Boop BS, Anderson DC. Oral anticoagulants and intracranial hemorrhage: facts and hypotheses. Stroke. 1995; 26:1471–77. 5 Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: current practices and unresolved questions. Stroke. 2006;37:256–62. 6 Liotta EM, Prabhakaran S. Warfarin-associated intracerebral hemorrhage is increasing in prevalence in the United States. J Stroke Cerebrovasc Dis. 2013;22:1151–5. 7 Franke CL, de Jonge J, van Swieten JC, Op de Coul AA, van Gijn J. Intracerebral hematomas during anticoagulant treatment.Stroke. 1990;21:726–30. 8 Brott T, Broderick J, Kothari R, Barsan W, Tomsick T, Sauerbeck L, et al. Early hemorrhage growth in patient with intracerebral hemorrhage. Stroke. 1997;28:1–5. 9 Fogelholm R, Murros K, Rissanen A, Avikainen S. Long term survival after primary intracerebral haemorrhage: a retrospective population based study. J Neurol Neurosurg Psychiatry. 2005;76:1534–8.

Teo et al.

10 Flaherty ML, Haverbusch M, Sekar P, Kissela B, Kleindorfer D, Moomaw CJ, et al. Long-term mortality after intracerebral hemorrhage. Neurology. 2006;66:1182–6. 11 Huttner HB, Schellinger PD, Hartmann M, Ko¨hrmann M, Juettler E, Wikner J, et al. Hematoma growth and outcome in treated neurocritical care patients with intracerebral hemorrhage related to oral anticoagulant therapy: complication of acute treatment strategies using vitamin K, fresh frozen plasma, and prothrombin complex concentrates. Stroke. 2006;37:1465– 70. 12 Lee SB, Manno EM, Layton KF, Wijdicks EF. Progression of warfarin-associated intracerebral hemorrhage after INR normalization with FFP. Neurology. 2006;67:1272–74. 13 Rosand J, Eckman MH, Knudsen KA, Singer DE, Greenberg SM. The effect of warfarin and intensity of anticoagulation on outcome of intracerebral hemorrhage. Arch Intern Med. 2004;164:880–4. 14 Flaherty ML, Tao H, Haverbusch M, Sekar P, Kleindorfer D, Kissela B, et al. Warfain use leads to larger intracerebral hematomas. Neurology. 2008;71:1084–89. 15 Berwaerts J, Dijkhuizen RS, Robb OJ, Webster J. Prediction of functional outcome and in-hospital mortality after admission with oral anticoagulant–related intracerebral hemorrhage. Stroke. 2000;31:2558–62. 16 Zubkov AY, Mandrekar JN, Claassen DO, Manno EM, Wijdicks EF, Rabinstein AA. Predictors of outcome in warfarin-related intracerebral hemorrhage. Arch Neurol. 2008;65:1320–25. 17 Biffi A, Battey TW, Ayres AM, Cortellini L, Schwab K, Gilson AJ, et al. Warfarin-related intraventricular hemorrhage: imaging and outcome. Neurology. 2011;77:1840–6. 18 Garg RK, Liebling SM, Mass MB, Nemeth AJ, Russell EJ, Naidech AM. Blood pressure reduction, decreased diffusion on MRI, and outcomes after intracerebral hemorrhage. Stroke. 2012;43:67–71. 19 Hylek EM, Singer DE. Risk factors for intracerebral hemorrhage in outpatients taking warfarin. Ann Intern Med. 1994;120:897–902. 20 Fogelholm R, Eskola K, Kiminkinen T, Kunnamo I. Anticoagulant treatment as a risk factor for primary intracerebral haemorrhage. J Neurol Neurosurg Psychiatry. 1992; 55:1121–24. 21 Flaherty ML, Haverbusch M, Sekar P, Kissela BM, Kleindorfer D, Moomaw CJ, et al. Location and outcome of anticoagulant-associated intracerebral hemorrhage. Neurocrit care. 2006;5:197–201. 22 Secil Y, Ciftci Y, Tokucoglu F, Beckmann Y. Intracranial hemorrhage related with warfarin use and comparison of warfarin and acetylsalicylic acid. J Stroke Cerebrovasc Dis. 2013. Apr 3. [Epub ahead of print]. doi: 10.1016/j.jstrokecerebrovasdis.2013. 03.008. 23 Cucchiara B, Messe S, Sansing L, Kasner S, Lyden P; CHANT Investigators. Hematoma growth in oral anticoagulant related intracerebral hemorrhage. Stroke. 2008;39:2993–96. 24 Flibotte JJ, Hagan N, J O’Donnell, Greenberg SM, Rosand J. Warfarin, hematoma expansion and outcome of intracerebral hemorrhage. Neurology. 2004;63:1059–64. 25 Rowe AS, Turner RM. Coagulation factor VIIa (recombinant) for warfarin induced intracranial hemorrhage. Am J Health Syst Pharm. 2010;67:361–65. 26 Goldstein JN, Thomas SH, Frontiero V, Joseph A, Engel C, Snider R, et al. Timing of fresh frozen plasma administration and rapid correction of coagulopathy in warfarin-related intracerebral hemorrhage. Stroke. 2006;37:151–5. 27 Menzin J, Hoesche J, Friedman M, Nichols C, Bergman GE, Crowther M, et al. Failure to correct international normalized ratio and mortality among patients with warfarin- related major bleeding: an analysis of electronic health records. J Thromb Haemost. 2012;10:596–605.

Warfarin associated intracerebral hemorrhage

28 Aguilar MI, Hart RG, Kase CS, Freeman WD, Hoeben BJ, Garcı´a RC, et al. Treatment of warfarin-associated intracerebral hemorrhage literature review and expert opinion. Mayo Clin Proc. 2007;82:82–92. 29 Freeman WD, Brott TG, Barrett KM, Castillo PR, Deen HG Jr, Czervionke LF, et al. Recombinant Factor VIIa for rapid reversal of warfarin anticoagulation in acute intracranial hemorrhage. Mayo Clinic Proc. 2004;79:1495–500. 30 Sorensen B, Johansen P, Nielsen GL, Sørensen JC, Ingerslev J. Reversal of international normalized ratio with recombinant activated factor VII in central nervous system bleeding during warfarin thromboprophylaxis: clinical and biochemical aspects. Blood Coagul Fibrinolysis. 2003;14:469–77. 31 Brody DL, Aiyagari V, Shackleford AM, Diringer MN. Use of recombinant factor VIIa in patients with warfarin-associated intracranial hemorrhage. Neurocrit care. 2005;2:236–7. 32 Selim MH, Molina CA. The role of hemostatic therapy in anticoagulation-associated intracerebral hemorrhage: intuition versus evidence. Stroke. 2012;43:2539–40. 33 Diringer MN, Zazulia AR. Hemostatic therapy should be used for acute treatment of anticoagulation-related intracerebral hemorrhage. Stroke. 2012;43:2535–6. 34 Mayer SA, Sacco RL, Shi T, Mohr JP. Neurological deterioration in non comatose patients with supratentorial intracerebral hemorrhage. Neurology. 1994;44:1379–84. 35 Leira R, Davalos A, Silva Y, Gil-Peralta A, Tejada J, Garcia M, et al. Early neurological deterioration in intracerebral hemorrhage. Neurology. 2004;63:461–67. 36 Flemming KD, Wijdicks EF, St Louis EK, Li H. Predicting deterioration in patients with lobar haemorrhages. J Neurol Neurosurg Psychiatry. 1999;66:600–5. 37 Qureshi AI, Safdar K, Weil J, Barch C, Bliwise DL, Colohan AR, et al. Predictors of early deterioration and mortality in black Americans with spontaneous intracerebral hemorrhage. Stroke. 1995;26:1764–67. 38 Anderson CS, Huang Y, Wang JG, Arima H, Neal B, Peng B, et al. Intensive blood pressure reduction in acute cerebral haemorrhage trial (INTERACT): a randomized pilot trial. Lancet Neurol. 2008;7:391–9. 39 Qureshi AI, Tariq N, Divani AA, et al. Antihypertensive treatment of acute cerebral hemorrhage trial. Crit Care Med. 2010;38:637–48. 40 Prabhakaran S, Gupta R, Ouyang B, John S, Temes RE, Mohammad Y, et al. Acute brain infarcts after spontaneous intracerebral hemorrhage: a diffusion-weighted imaging study. Stroke. 2010;41:48–53. 41 Hill MD, Silver FL, Austin PC, Tu JV. Rate of stroke recurrence in patients with primary intracerebral hemorrhage. Stroke. 2000;31:123–7. 42 Hanger HC, Wilkinson TJ, Fayez N-Iskander, Sainsbury R. The risk of recurrent stroke after intracerebral hemorrhage. J Neurol Neurosurg Psychiatry. 2007;78:836–40. 43 Zia E, Engstro¨ m G, Svensson PJ, Norrving B, PessahRasmussen H. Three-year survival and stroke recurrence rate in patients with primary intracerebral hemorrhage. Stroke. 2009;40:3567–73. 44 Bailey RD, Hart RG, Benavente O, Pearce LA. Recurrent brain hemorrhage is more frequent than ischemic stroke after intracranial hemorrhage. Neurology. 2001;56:773–7. 45 Vermeer SE, Algra A, Franke CL, Koudstaal PJ, Rinkel GJ. Long term prognosis after recovery from primary intracerebral hemorrhage. Neurology. 2002;59:205–9. 46 Chong BH, Chan KH, Pong V, Lau KK, Chan YH, Zuo ML, et al. Use of aspirin in Chinese after recovery from primary intracranial hemorrhage. Thromb Haemost. 2012;107:241–7. 47 Ariesen MJ, Claus CP, Rinkel GJ, Algra A. Risk factors for intracerebral hemorrhage in the general population: a systemic review. Stroke. 2003;34:2060–5.

Neurological Research

2014

VOL .

36

NO .

2

149

Warfarin associated intracerebral hemorrhage in Hong Kong Chinese.

Warfarin-associated intracerebral hemorrhage (WICH) is a serious neurological condition associated with significant mortality and morbidity. We aimed ...
116KB Sizes 0 Downloads 0 Views