- Email: [email protected]
Increased arterial stiffness is an independent risk factor for hemorrhagic transformation in ischemic stroke undergoing thrombolysis
International Journal of Cardiology 243 (2017) 466–470
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Increased arterial stiffness is an independent risk factor for hemorrhagic transformation in ischemic stroke undergoing thrombolysis☆ Maurizio Acampa a,⁎, Silvia Camarri a, Pietro Enea Lazzerini b, Francesca Guideri a, Rossana Tassi a, Raffaella Valenti a, Alessandra Cartocci c, Giuseppe Martini a a b c
Stroke Unit, Department of Neurological and Sensorineural Sciences, Azienda Ospedaliera Universitaria Senese, “Santa Maria alle Scotte” General Hospital, Siena, Italy Department of Medical Sciences, Surgery and Neurosciences, University of Siena, Italy Department of Economics and Statistics, University of Siena, Siena, Italy
a r t i c l e
i n f o
Article history: Received 25 February 2017 Accepted 27 March 2017 Keywords: Ischemic stroke Hemorrhagic transformation Thrombolysis Arterial stiffness Blood pressure Arterial hypertension
a b s t r a c t Background: Hemorrhagic transformation (HT) is a multifactorial phenomenon and represents a possible complication of ischemic stroke, especially after thrombolytic treatment. Increased arterial stiffness has been associated with intracranial hemorrhage, but there is no evidence of association with HT after thrombolytic therapy. The aim of our study is to investigate a possible link between arterial stiffness and HT occurrence after thrombolytic therapy in patients with ischemic stroke. Methods: We enrolled 258 patients (135 males, 123 females; mean age: 73 ± 12 years) with acute ischemic stroke undergoing intravenous thrombolysis or/and mechanical thrombectomy. All stroke patients underwent neuroimaging examination, 24-h heart rate and blood pressure monitoring and brain CT-scan after 24–72 h to evaluate HT occurrence. The linear regression slope of diastolic on systolic blood pressure was obtained and assumed as a global measure of arterial compliance, and its complement (1 minus the slope), named arterial stiffness index (ASI), has been taken as a measure of arterial stiffness. Results: Out of 258, HT occurred in 55 patients. ASI was significantly higher in patients with HT than in patients without HT (0.70 ± 0.12 vs 0.62 ± 0.14, p b 0.001). Logistic regression model showed ASI as independent predictors of HT (OR: 1.9, 95% CI: 1.09–3.02, for every 0.2 increase of ASI): in particular, OR was 5.2 (CI: 2.22–12.24) when ASI was N0.71, in comparison with ASI lower than 0.57. Conclusions: Our results point to arterial stiffness as a novel independent risk factor for HT after ischemic stroke treated with thrombolysis, suggesting a particularly high bleeding risk when ASI is N 0.71. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Hemorrhagic transformation (HT) is a bleeding in the ischemic brain tissue and represents a possible complication of ischemic stroke, especially after thrombolytic treatment, occurring in 13%– 43% of patients [1]. HT is a complex and multifactorial phenomenon [2], not fully understood and only partly predictable. Known risk factors include age, blood glucose level, low platelet count, high National Institute of Health Stroke Scale score (NIHSSs), size and location of ischemic area, poor collateral vessels, thrombolytic agent used and time window allowed for the initiation of the therapy [1]. Blood
☆ The authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation ⁎ Corresponding author at: U.O.C. Stroke Unit, Policlinico ‘S. Maria alle Scotte’, viale Bracci, n. 1, 53100 Siena, Italy. E-mail address: [email protected] (M. Acampa).
http://dx.doi.org/10.1016/j.ijcard.2017.03.129 0167-5273/© 2017 Elsevier B.V. All rights reserved.
pressure (BP) may also play a role in HT pathogenesis: there is a consensus to maintain BP below 180/105 mm Hg for the first 24 h after thrombolytic therapy [3], but the early BP management after thrombolysis remains inconclusive. Acute hypertensive response in ischemic stroke is associated with poor outcomes and may also be associated with increased aortic stiffness [4]. Arterial stiffness has independent predictive value for cardiovascular events [5]; in particular, in acute ischemic stroke, high arterial stiffness index values have been observed [6]. In our previous study, we suggested a link between deep intracerebral hemorrhage and arterial stiffening that represents a possible pathogenic factor modifying arterial wall properties and contributing to vascular rupture in response to intravascular pressure acute elevation [7]. However, no studies have evaluated the relationship between stiffness and HT occurrence after thrombolytic therapy. The aim of our study is to investigate a possible link between arterial stiffness (evaluated by means of arterial stiffness index [ASI]) and HT occurrence after thrombolytic therapy in patients with ischemic stroke.
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470 2. Materials and methods We enrolled 258 patients (135 males, 123 females; mean age 73 ± 12 years), admitted consecutively to the Stroke Unit Department of Siena University Hospital for acute ischemic stroke and submitted to intravenous thrombolysis or/and mechanical thrombectomy. Neurological status at admission was assessed by using NIHSSs; 3 severity levels were defined: mild (NIHSSs b 8), moderate (NIHSSs: 8–16), and severe (NIHSSs N 16). All patients underwent neuroimaging examination (brain computerized tomography with angio-CT scan and/or brain magnetic resonance imaging), extracranial and transcranial arterial ultrasound examination, transthoracic echocardiography, and 12-lead resting ECG. A 24-h heart rate (HR) and BP monitoring was conducted for all the subjects recording the following parameters: systolic blood pressure (SBP), diastolic blood pressure (DBP), mean blood pressure (MBP), pulse pressure (PP), and heart rate (HR). In all patients, high-sensitivity C-reactive protein (hsCRP) level, creatinine, potassium, sodium plasma levels, HbA1c percentage, serum total-cholesterol level, and LDL- and HDL-cholesterol levels were measured. In patients with age b 60 years (n = 44) transesophageal echocardiography and screening for hypercoagulable state were performed. Stroke subtypes were determined according to the ASCOD classification [8]. In all patients a brain CT scan was performed after 24–72 h after thrombolysis to evaluate the occurrence of hemorrhagic transformation. The modified Rankin scale (mRS) was assessed at the time of presentation (preadmission mRS) and at 90 days by a stroke-trained physician, to evaluate 3-month clinical outcome. The study was approved by the Ethics Committee of the University Hospital of Siena, Italy. 2.1. ASCOD classification Every patient was graded into one of the 5 predefined phenotypes: A (atherosclerosis), S (small-vessel disease), C (cardiac pathology), O (other cause), and D (dissection), assigning a
467
degree of likelihood of causal relationship to every potential disease (1 “potentially causal”, 2 “causality is uncertain”, 3 “unlikely causal but the disease is present”, 0 “absence of disease”, and 9 “insufficient workup to rule out the disease”). The ASCOD system allowed us to weigh the potentially causal relationship between every disease detected and the ischemic stroke. According to this system, when the patient was classified as “degree 1” in one of the 5 phenotypes of ASCOD, the stroke etiology was respectively “atherosclerosis”, “small vessel disease”, “cardioembolic”, “other causes” (polycythemia, thrombocytemia, systemic lupus, disseminated intravascular coagulation, antiphospholipid syndrome, Fabry's disease, sickle cell disease, ruptured intracranial aneurysm, severe hyperhomocysteinemia, Horton's disease, cerebral inflammatory angiitis, and Moyamoya disease) and “arterial dissection”. Instead, the etiology of stroke was classified as “cryptogenic” when the patients were grade 0 (absence of disease), 9 (insufficient workup), or 2 and 3 (being unable to establish a direct causal relationship between these diseases and the ischemic stroke).
2.2. Hemorrhagic transformation According to European Cooperative Acute Stroke Study (ECASS) radiological classification [9], hemorrhagic transformation has been classified in the following different types: hemorrhagic infarct type 1 (HI1; small petechiae along the periphery of the infarct), hemorrhagic infarct type 2 (HI2; confluent petechiae within the infarcted area without a space-occupying effect), parenchymal haematoma type 1 (PH1; bleeding b30% of the infarcted area with a mild space-occupying effect), parenchymal haematoma type 2 (PH2; bleeding N30% of the infarcted area with a significant spaceoccupying effect), and remote parenchymal hemorrhage (PHr; bleeding in brain areas remote from infarcted tissue).
Table 1 Demographic characteristics of the patients of the study.
Age (years) Women/men Neurological deficit (at admission) Patients number with NIHSSs b 8 Patients number with NIHSSs 8–16 Patients number with NIHSSs N 16 ASCOD phenotype Atherosclerosis, n (%) Small vessel disease, n (%) Cardioembolic, n (%) Other, n (%) Dissection, n (%) Cryptogenic, n (%) Cardiovascular risk factors Hypertension, n (%) Diabetes mellitus, n (%) Hypercholesterolemia, n (%) Atrial fibrillation, n (%) Previous CAD, n (%) Previous stroke, n (%) Smoking, n (%) Antihypertensive drugs, (%) ACE/angiotensin II receptor inhibitors Diuretics Beta-blockers Calcium channel blockers Laboratory parameters Glucose (mg/dl) Platelets (n/mm3) INR Creatinine (mg/dl) sodium potassium Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HbA1c (%) C reactive protein (mg/dl) Hemodynamic parameters SBP (mm Hg) DBP (mm Hg) MBP (mm Hg) PP(mm Hg) HR (beats/m)
Pt. without HT (n = 203)
Pt. with HT (n = 55)
P value
72.8 ± 13 94:109
75.2 ± 11.6 29:26
0.24 0.44
83 (40.9%) 61 (30%) 59 (29.1%)
6 (10.9%) 17 (30.9%) 32 (58.2%)
b0.001 0.76 b0.001 0.37
49 (24%) 6 (3%) 60 (29.5%) 3 (1.5%) 6 (3%) 79 (39%)
18 (33%) 1 (2%) 18 (33%) 1 (2%) 3 (5%) 14 (25%)
143 (70%) 40 (19.7%) 97 (47%) 67 (33%) 19 (9%) 41 (20%) 41 (20%) 127 (62%) 89 (43%) 63 (31%) 52 (25%) 29 (14%)
44 (80%) 13 (23.6%) 29 (52%) 23 (41%) 9 (16%) 10 (18%) 7 (12.7%) 40 (72%) 25 (45%) 19 (34%) 18 (32%) 6 (10%)
0.18 0.6 0.5 0.3 0.15 0.85 0.24 0.27 0.88 0.63 0.31 0.66
131 ± 45 219,774 ± 82,590 1.09 ± 0.2 0.9 ± 0.3 140 ± 3.8 4 ± 0.4 185 ± 42 114 ± 37 6 ± 0.8 2.01 ± 4.49
123 ± 29 194,509 ± 55,198 1.08 ± 0.2 1.02 ± 0.6 140 ± 5.9 3.9 ± 0.4 186 ± 43 116 ± 35 5.9 ± 0.8 2.77 ± 6.4
0.5 0.01 0.4 0.25 0.5 0.3 0.8 0.6 0.3 0.11
133 ± 15 69 ± 9 90.8 ± 10.2 64 ± 15 74 ± 13
142 ± 18 72 ± 11 95.9 ± 12.6 70 ± 16 72 ± 15
0.0007 0.12 0.008 0.005 0.5
Data are expressed as mean ± SD. NIHSSs: National Institutes of Health Stroke Scale score. CAD: coronary artery disease. ACE: angiotensin converting enzyme. SBP: systolic blood pressure. DBP: diastolic blood pressure. MBP: mean blood pressure. PP: pulse pressure. HR: heart rate.
468
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
2.3. Arterial stiffness index (ASI) BP monitoring was performed using validated oscillometric devices (Bedside Monitor Life Scope I BSM-2303K, International Div., Nihon Kohden Corp., Tokyo, Japan), in the first 24 h from the admission in the hospital. BP was recorded for 24 h every 30 min, between 06:00 h to midnight, and hourly, from midnight to 06:00 h. For inclusion in the study, at least 80% of valid BP and heart rate measurements for each subject were needed. By plotting the individual values of SBP and DBP measurements, obtained through 24-h non-invasive monitoring, the linear regression slope of DBP on SBP was obtained and assumed as a global measure of arterial compliance, and its complement (1 minus the slope), named ASI, has been taken as a measure of arterial stiffness [10]. 2.4. Statistical analysis All results are presented as means ± SD. Normal distribution of quantitative variables was preliminary tested using the Anderson-Darling test to select parametric or nonparametric inferential statistical methods. Homoscedasticity was tested using Bartlett test for normal distribution. Unpaired t-test was performed to compare SBP, MBP and HR in patients with and without HT. Mann-Whitney test was performed to compare age, serum potassium, serum sodium, HbA1c, C reactive protein, creatinine, total cholesterol, LDL-cholesterol, PP, glucose levels, platelet count, and INR in both groups. Welch's test was performed to test DBP. The two-sided Fisher's exact test was performed to evaluate statistical correlation between categorical variables (cardiovascular risk factors and antihypertensive drugs) evaluated in both groups of patients. The χ2 test was performed to evaluate ASCOD phenotypes; an extension of Fisher's exact test was performed to evaluate NIHSS scores. A p value below 0.05 was considered statistically significant. Logistic regression model was constructed to determine the predictors of HT. All significant variables of statistical descriptive analysis were entered initially. The statistical criterion for considering retention of a variable in the model was P b 0.05. Akaike's information criterion was used to select the best subset of predictor variables and corresponding OR were obtained. Statistical analysis was performed using R Statistical Software (version 3.3.2 for Windows, R Foundation for Statistical Computing, Vienna, Austria).
3. Results Out of 258 patients, HT occurred in 55 patients enrolled in the study, with the following radiologic sub-types: HI in 39 patients (HI1 in 27 patients and HI2 in 12 patients), PH in 14 patients (PH1 in 6 patients and PH2 in 8 patients), and PHr in 2 patients. Demographic characteristics of both groups of patients (with and without HT) are depicted in Table 1. Severity level of neurological deficit is higher in patients with HT; in particular, in severe deficit (NIHSSs N 16) the percentage of HT patients was significantly higher than that of non-HT patients, whereas in mild deficit (NIHSSs b 8) the percentage of HT patients was significantly lower than that of non-HT patients. Patients with moderate neurological deficit (NIHSSs between 8 and 16) were similar about percentage of HT. There were no differences between patients with and without HT in regard to age, sex, ASCOD phenotype, cardiovascular risk factors, and antihypertensive drugs used (Table 1). No differences were found regarding the main laboratory parameters evaluated in both groups (hs-CRP, total cholesterol, HDL, LDL, HbA1c, creatinine plasma levels, sodium, potassium, glucose, and INR), except for platelet count (194,509 ± 55,198 in the HT group vs 219,774 ± 82,590/μl in the non-HT group; p = 0.01) (Table 1). 3.1. ASI and other hemodynamic parameters In patients with HT, ASI was significantly higher in the HT group than in the non-HT group (0.70 ± 0.13 vs 0.62 ± 0.14, p = 0.0001) (Fig. 1); in particular, ASI of patients with HI and PH were higher than ASI of the non-HT patients' group (0.69 ± 0.13 and 0.73 ± 0.11 vs 0.62 ± 0.14, p = 0.02) (Fig. 2, a); these results were observed in adults (18–65 years) as well as in older adults (N 65 years) (Fig. 2, b and c). SBP, MBP and PP were significantly higher in patients with HT than in patients without HT (142 ± 18 vs 133 ± 15 mm Hg, p = 0.0007; 95.9 ± 12.6 vs 90.8 ± 10.2 mm Hg, p = 0.008; 70 ± 16 vs 64 ± 15 mm Hg, p = 0.005); whereas DBP and HR were similar in both groups (Table 1).
Fig. 1. Comparison of ASI in patients with and without HT. A: All patients (p b 0.001). B: Adult patients (age between 18 and 65 years) (p = 0.005). C: Patients over 65 years (p = 0.004). HT: hemorrhagic transformation. Unpaired t-test. ***p b 0.001, **p b 0.01.
3.2. Type of treatment Systemic thrombolysis was administered to 154 patients (60%), whereas mechanical thrombectomy was performed in 50 patients (19%) and a combined treatment (systemic thrombolysis + mechanical thrombectomy) was performed in 54 patients (21%). HT occurrence was significantly more frequent in patients treated with mechanical thrombectomy or combined treatment than in patients submitted to systemic thrombolysis (see Fig. 1S in supplementary data). There were no significant differences among the 3 groups in baseline demographic (Table 2); however, patients treated with mechanical thrombectomy or combined treatment had a higher severity level of neurological deficit, according to NIHSSs. Furthermore, in the “mechanical thrombectomy group” treatment started later than in the other 2 groups (Table 2).
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
469
Fig. 2. Comparison of ASI in patients without HT and with different types of hemorrhagic transformation. HT: hemorrhagic transformation. HI: hemorrhagic infarct. PH: parenchymal haematoma. PHr: remote parenchymal hemorrhage. Unpaired t-test, **p = 0.002.
3.3. Logistic regression and multivariable model Logistic regression model was constructed to determine the predictors of HT. All significant variables of statistical descriptive analysis were entered initially. These variables were the following: SBP, MBP, PP, ASI, type of treatment (systemic thrombolysis, mechanical thrombolysis, and combined thrombolysis), platelet count, and severity level of neurological deficit according to NIHSSs. Akaike's information criterion showed that the best subset of predictor variables for HT were the following: ASI (OR 1.81 for 0.2, 95% CI 1.09–3.02), MBP (OR 1.57 for 10 mm Hg, 95% CI 1.06–2.18), NIHSS (OR 4.63 for moderate vs mild level of neurological deficit, CI 1.55–17.15 and OR 8.05 for severe vs mild level of neurological deficit, CI 2.86–28.91) and platelet count (OR 0.94 for 10,000 platelets/μl, CI 0.89–1). In particular, when patients were classified into tertile groups by ASI, we observed that, in comparison with the first tertile (ASI b 0.57), the second (ASI: 0.57–0.71) and the third tertile (ASI N 0.71) showed an OR for HT of 2.4 (CI 0.97–5.91) and 5.2 (CI 2.22–12.24), respectively.
3.4. Outcome The functional outcome was evaluated at three months by means of mRS in patients with and without HT (see Fig. 2S in supplementary data). The distribution of mRS scores in the two groups at 90 days
shows a higher rate of death in the HT group in comparison to the non-HT group (18% vs 7%; p = 0.03). Furthermore, the proportion of patients with favorable functional outcome (mRS 0–2) was significantly higher in the non-HT group than in the HT group (mRS 0–2, 61% vs 30%; p = 0.0001). 4. Discussion The novel findings of our study are the following: 1) ASI is higher in patients with HT than in patients without HT, and 2) ASI represents an independent predictor of HT in patients with ischemic stroke treated with thrombolysis. In our study, ASI was measured during the first 24 h from the onset of symptoms, in order to evaluate the risk of HT occurrence in the next 24–72 h after initiation of treatment. Our results showed that ASI values N0.71 determine a risk of HT 5.2 fold higher in comparison with ASI values lower than 0.57. Furthermore, logistic regression analysis highlighted ASI as a predictor of HT, independently from other wellknown factors such as SBP, DBP or MBP. In fact, ASI explores the relationship of SBP and DBP, based on the premise that SBP and DBP are less tightly coupled in patients with higher arterial stiffness than in those with lower arterial stiffness. Our results suggest that arterial stiffness may represent an important factor contributing to HT in patients with ischemic stroke, irrespective of pathogenesis of stroke. Arterial stiffness is closely associated with small vessel diseases [6],
Table 2 Characteristics of stroke patients according to the type of treatment.
Age (years) Gender (F:M) NIHSSs at admission Patients with NIHSS score b 8 Patients with NIHSSs 8–16 Patients with NIHSS N 16 Time from the onset of symptoms (min)
Systemic thrombolysis (n: 154)
Mechanical thrombolysis (n: 50)
Combined therapy (n: 54)
P value
73.8 ± 13 70:84
72.9 ± 11 27:23
72.6 ± 14 26:28
0.45 0.57 b0.001
78 (50.7%)a,b 51 (33.1%) 25 (16.2%)a,b 178 ± 48a
8 (16%) 8 (16%) 34 (68%) 288 ± 172.5b
3 (5.6%) 19 (35.2%) 32 (59.2%) 172 ± 68
NIHSSs: National Institutes of Health Stroke Scale score. Kruskall-Wallis test. Chi squared test. a Versus mechanical thrombolysis group, b Versus combined therapy.
b0.001
470
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
but our results showed that, according to ASCOD phenotypes, the number of patients with small vessel disease is similar in both, the HT and non-HT groups. Following acute ischemic stroke, there is a breakdown of the blood– brain barrier resulting in friable intracranial vasculature, with the subsequent increased risk of HT. In healthy subjects, the aorta is highly compliant and a part of the pulsatile energy is stored in the arterial wall rather than being transmitted to the distal vasculature: this is a protective mechanism, limiting transmission of excessive pulsatility into the microcirculation. Instead, in patients with arterial stiffness, the arteries require a greater amount of force to expand and pulsatile energy is highly transmitted into the microcirculation, where it may cause damage [11]. The vasculature of the brain is particularly sensitive to excessive pressure and flow pulsatility, because it is characterized by high flow, which requires low microvascular impedance [12]. Therefore, in subjects with arterial stiffness the low-impedance brain circulation is exposed to greater pressure fluctuations with subsequent damage. This mechanism may explain, at least in part, the possible role of arterial stiffness in the pathogenesis of HT after ischemic stroke. Furthermore, cerebral autoregulation is impaired after an ischemic stroke, resulting in a relatively linear relation between systemic arterial pressure and cerebral perfusion pressure [13]. It is possible also that a high pulsatility may cause an altered reperfusion of ischemic territory through collateral circulation; a recent study has shown that high blood pressure variability may lead to instability of cerebral perfusion resulting in damage of ischemic territory and derangement of blood brain barrier [14]. Furthermore, our study showed a relationship among HT and different factors such as low platelet count, high NIHSS score (expression of severe neurologic deficit), and high mean blood pressure values, confirming the results of previous studies [15–18]. Surprisingly, according to Akaike's information criterion, the final “best” subset of predictor variables for HT excluded the variable “type of treatment”. Given that patients treated with mechanical or combined treatment had higher NIHSSs than patients treated with systemic thrombolysis, it is possible that HT occurrence in these subjects was already explained by the higher NIHSS scores. It is possible that a higher HT rate in patients treated with mechanical thrombectomy or combined treatment is due to more severe neurological deficit and higher time from symptom onset to treatment. In particular, regarding the combined treatment, we have reported the time from symptom onset to the first treatment (systemic thrombolysis); for this reason time to treatment with mechanical thrombectomy resulted higher than time to combined treatment. 5. Conclusions During the past decade, increased aortic stiffness has emerged as an important risk factor for target organ damage and cardiovascular disease events. Arterial stiffness index, based on 24-h BP monitoring, represents a simple and unexpensive tool providing valuable prognostic information about the risk of HT in patients with ischemic stroke undergoing thrombolytic therapy. In addition to known risk factors for bleeding after thrombolysis, we propose to evaluate ASI as a further bleeding risk factor for hemorrhage, suggesting caution when ASI values are N 0.71. Furthermore, the role of ASI as a predictor of HT could be also important in the group of patients with cardioembolic stroke, in particular regarding the choice of the optimal time to start anticoagulant therapy. As early HT is a major concern in the acute phase of stroke associated to atrial fibrillation, ASI N 0.71 (increasing of 5.2 fold the risk of HT) suggests delaying initiation of oral anticoagulation. Further studies will be
necessary to prove this hypothesis in a group of cardioembolic patients undergoing anticoagulant therapy. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2017.03.129. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. References [1] J. Zhang, Y. Yang, H. Sun, Y. Xing, Hemorrhagic transformation after cerebral infarction: current concepts and challenges, Ann. Transl. Med. 2 (2014) 81. [2] E.S. Sussman, E.S. Connolly Jr., Hemorrhagic transformation: a review of the rate of hemorrhage in the major clinical trials of acute ischemic stroke, Front. Neurol. 4 (2013) 69. [3] E.C. Jauch, J.L. Saver, H.P. Adams Jr., A. Bruno, J.J. Connors, B.M. Demaerschalk, P. Khatri, P.W. McMullan Jr., A.I. Qureshi, K. Rosenfield, P.A. Scott, D.R. Summers, D.Z. Wang, M. Wintermark, H. Yonas, Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association, Stroke 44 (2013) 870–947. [4] M. Kwarciany, D. Gąsecki, K. Kowalczyk, A. Rojek, S. Laurent, P. Boutouyrie, M. Skrzypek-Czerko, W.M. Nyka, K. Narkiewicz, B. Karaszewski, Acute hypertensive response in ischemic stroke is associated with increased aortic stiffness, Atherosclerosis 251 (2016) 1–5. [5] C. Palombo, M. Kozakova, Arterial stiffness, atherosclerosis and cardiovascular risk: pathophysiologic mechanisms and emerging clinical indications, Vasc. Pharmacol. 77 (2016) 1–7. [6] A. Tuttolomondo, R. Di Sciacca, D. Di Raimondo, A. Serio, G. D'Aguanno, A. Pinto, G. Licata, Arterial stiffness indexes in acute ischemic stroke: relationship with stroke subtype, Atherosclerosis 211 (2010) 187–194. [7] M. Acampa, F. Guideri, I. Di Donato, R. Tassi, G. Marotta, G. Lo Giudice, P. D'Andrea, G. Martini, Arterial stiffness in patients with deep and lobar intracerebral haemorrhage, J. Stroke 16 (2014) 184–188. [8] P. Amarenco, J. Bogousslavsky, L.R. Caplan, G.A. Donnan, M.E. Wolf, M.G. Hennerici, The ASCOD phenotyping of ischemic stroke (updated ASCO phenotyping), Cerebrovasc. Dis. 36 (2013) 1–5. [9] W. Hacke, M. Kaste, E. Bluhmki, M. Brozman, A. Dávalos, D. Guidetti, V. Larrue, K.R. Lees, Z. Medeghri, T. Machnig, D. Schneider, R. von Kummer, N. Wahlgren, D. Toni, Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke, N. Engl. J. Med. 359 (2008) 1317–1329. [10] Y. Li, J.G. Wang, E. Dolan, P.J. Gao, H.F. Guo, T. Nawrot, A.V. Stanton, D.L. Zhu, E. O'Brien, J.A. Staessen, Ambulatory arterial stiffness index derived from 24-hour ambulatory blood pressure monitoring, Hypertension 47 (2006) 359–364. [11] G.F. Mitchell, M.A. van Buchem, S. Sigurdsson, J.D. Gotal, M.K. Jonsdottir, Ó. Kjartansson, M. Garcia, T. Aspelund, T.B. Harris, V. Gudnason, L.J. Launer, Arterial stiffness, pressure and flow pulsatility and brain structure and function: the Age, Gene/Environment Susceptibility—Reykjavik study, Brain 134 (Pt11) (2011) 3398–3407. [12] L.L. Cooper, G.F. Mitchell, Aortic stiffness, cerebrovascular dysfunction, and memory, Pulse 4 (2016) 69–77. [13] S.L. Dawson, R.B. Panerai, J.F. Potter, Serial changes in static and dynamic cerebral autoregulation after acute ischemic stroke, Cerebrovasc. Dis. 16 (2003) 69–75. [14] K. Liu, S. Yan, S. Zhang, Y. Guo, M. Lou, Systolic blood pressure variability is associated with severe hemorrhagic transformation in the early stage after thrombolysis, Transl. Stroke Res. 7 (2016) 186–191. [15] V. Larrue, R.R. von Kummer, A. Muller, E. Bluhmki, Risk factors for severe hemorrhagic transformation in ischemic stroke patients treated with recombinant tissue plasminogen activator: a secondary analysis of the European-Australasian Acute Stroke Study (ECASS II), Stroke 32 (2001) 438–441. [16] C.S. Kidwell, J.L. Saver, J. Carneado, J. Sayre, S. Starkman, G. Duckwiler, Y.P. Gobin, R. Jahan, P. Vespa, J.P. Villablanca, D.S. Liebeskind, F. Vinuela, Predictors of hemorrhagic transformation in patients receiving intra-arterial thrombolysis, Stroke 33 (2002) 717–724. [17] A.K. Gilligan, R. Markus, S. Read, V. Srikanth, T. Hirano, G. Fitt, M. Arends, B.R. Chambers, S.M. Davis, G.A. Donnan, Baseline blood pressure but not early computed tomography changes predicts major hemorrhage after streptokinase in acute ischemic stroke, Stroke 33 (2002) 2236–2242. [18] J.H. Lee, K.Y. Park, J.H. Shin, J.K. Cha, H.Y. Kim, J.H. Kwon, H.G. Oh, K.B. Lee, D.E. Kim, S.W. Ha, K.H. Cho, S.I. Sohn, M.S. Oh, K.H. Yu, B.C. Lee, S.U. Kwon, Symptomatic hemorrhagic transformation and its predictors in acute ischemic stroke with atrial fibrillation, Eur. Neurol. 64 (2010) 193–200.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Increased arterial stiffness is an independent risk factor for hemorrhagic transformation in ischemic stroke undergoing thrombolysis☆ Maurizio Acampa a,⁎, Silvia Camarri a, Pietro Enea Lazzerini b, Francesca Guideri a, Rossana Tassi a, Raffaella Valenti a, Alessandra Cartocci c, Giuseppe Martini a a b c
Stroke Unit, Department of Neurological and Sensorineural Sciences, Azienda Ospedaliera Universitaria Senese, “Santa Maria alle Scotte” General Hospital, Siena, Italy Department of Medical Sciences, Surgery and Neurosciences, University of Siena, Italy Department of Economics and Statistics, University of Siena, Siena, Italy
a r t i c l e
i n f o
Article history: Received 25 February 2017 Accepted 27 March 2017 Keywords: Ischemic stroke Hemorrhagic transformation Thrombolysis Arterial stiffness Blood pressure Arterial hypertension
a b s t r a c t Background: Hemorrhagic transformation (HT) is a multifactorial phenomenon and represents a possible complication of ischemic stroke, especially after thrombolytic treatment. Increased arterial stiffness has been associated with intracranial hemorrhage, but there is no evidence of association with HT after thrombolytic therapy. The aim of our study is to investigate a possible link between arterial stiffness and HT occurrence after thrombolytic therapy in patients with ischemic stroke. Methods: We enrolled 258 patients (135 males, 123 females; mean age: 73 ± 12 years) with acute ischemic stroke undergoing intravenous thrombolysis or/and mechanical thrombectomy. All stroke patients underwent neuroimaging examination, 24-h heart rate and blood pressure monitoring and brain CT-scan after 24–72 h to evaluate HT occurrence. The linear regression slope of diastolic on systolic blood pressure was obtained and assumed as a global measure of arterial compliance, and its complement (1 minus the slope), named arterial stiffness index (ASI), has been taken as a measure of arterial stiffness. Results: Out of 258, HT occurred in 55 patients. ASI was significantly higher in patients with HT than in patients without HT (0.70 ± 0.12 vs 0.62 ± 0.14, p b 0.001). Logistic regression model showed ASI as independent predictors of HT (OR: 1.9, 95% CI: 1.09–3.02, for every 0.2 increase of ASI): in particular, OR was 5.2 (CI: 2.22–12.24) when ASI was N0.71, in comparison with ASI lower than 0.57. Conclusions: Our results point to arterial stiffness as a novel independent risk factor for HT after ischemic stroke treated with thrombolysis, suggesting a particularly high bleeding risk when ASI is N 0.71. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Hemorrhagic transformation (HT) is a bleeding in the ischemic brain tissue and represents a possible complication of ischemic stroke, especially after thrombolytic treatment, occurring in 13%– 43% of patients [1]. HT is a complex and multifactorial phenomenon [2], not fully understood and only partly predictable. Known risk factors include age, blood glucose level, low platelet count, high National Institute of Health Stroke Scale score (NIHSSs), size and location of ischemic area, poor collateral vessels, thrombolytic agent used and time window allowed for the initiation of the therapy [1]. Blood
☆ The authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation ⁎ Corresponding author at: U.O.C. Stroke Unit, Policlinico ‘S. Maria alle Scotte’, viale Bracci, n. 1, 53100 Siena, Italy. E-mail address: [email protected] (M. Acampa).
http://dx.doi.org/10.1016/j.ijcard.2017.03.129 0167-5273/© 2017 Elsevier B.V. All rights reserved.
pressure (BP) may also play a role in HT pathogenesis: there is a consensus to maintain BP below 180/105 mm Hg for the first 24 h after thrombolytic therapy [3], but the early BP management after thrombolysis remains inconclusive. Acute hypertensive response in ischemic stroke is associated with poor outcomes and may also be associated with increased aortic stiffness [4]. Arterial stiffness has independent predictive value for cardiovascular events [5]; in particular, in acute ischemic stroke, high arterial stiffness index values have been observed [6]. In our previous study, we suggested a link between deep intracerebral hemorrhage and arterial stiffening that represents a possible pathogenic factor modifying arterial wall properties and contributing to vascular rupture in response to intravascular pressure acute elevation [7]. However, no studies have evaluated the relationship between stiffness and HT occurrence after thrombolytic therapy. The aim of our study is to investigate a possible link between arterial stiffness (evaluated by means of arterial stiffness index [ASI]) and HT occurrence after thrombolytic therapy in patients with ischemic stroke.
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470 2. Materials and methods We enrolled 258 patients (135 males, 123 females; mean age 73 ± 12 years), admitted consecutively to the Stroke Unit Department of Siena University Hospital for acute ischemic stroke and submitted to intravenous thrombolysis or/and mechanical thrombectomy. Neurological status at admission was assessed by using NIHSSs; 3 severity levels were defined: mild (NIHSSs b 8), moderate (NIHSSs: 8–16), and severe (NIHSSs N 16). All patients underwent neuroimaging examination (brain computerized tomography with angio-CT scan and/or brain magnetic resonance imaging), extracranial and transcranial arterial ultrasound examination, transthoracic echocardiography, and 12-lead resting ECG. A 24-h heart rate (HR) and BP monitoring was conducted for all the subjects recording the following parameters: systolic blood pressure (SBP), diastolic blood pressure (DBP), mean blood pressure (MBP), pulse pressure (PP), and heart rate (HR). In all patients, high-sensitivity C-reactive protein (hsCRP) level, creatinine, potassium, sodium plasma levels, HbA1c percentage, serum total-cholesterol level, and LDL- and HDL-cholesterol levels were measured. In patients with age b 60 years (n = 44) transesophageal echocardiography and screening for hypercoagulable state were performed. Stroke subtypes were determined according to the ASCOD classification [8]. In all patients a brain CT scan was performed after 24–72 h after thrombolysis to evaluate the occurrence of hemorrhagic transformation. The modified Rankin scale (mRS) was assessed at the time of presentation (preadmission mRS) and at 90 days by a stroke-trained physician, to evaluate 3-month clinical outcome. The study was approved by the Ethics Committee of the University Hospital of Siena, Italy. 2.1. ASCOD classification Every patient was graded into one of the 5 predefined phenotypes: A (atherosclerosis), S (small-vessel disease), C (cardiac pathology), O (other cause), and D (dissection), assigning a
467
degree of likelihood of causal relationship to every potential disease (1 “potentially causal”, 2 “causality is uncertain”, 3 “unlikely causal but the disease is present”, 0 “absence of disease”, and 9 “insufficient workup to rule out the disease”). The ASCOD system allowed us to weigh the potentially causal relationship between every disease detected and the ischemic stroke. According to this system, when the patient was classified as “degree 1” in one of the 5 phenotypes of ASCOD, the stroke etiology was respectively “atherosclerosis”, “small vessel disease”, “cardioembolic”, “other causes” (polycythemia, thrombocytemia, systemic lupus, disseminated intravascular coagulation, antiphospholipid syndrome, Fabry's disease, sickle cell disease, ruptured intracranial aneurysm, severe hyperhomocysteinemia, Horton's disease, cerebral inflammatory angiitis, and Moyamoya disease) and “arterial dissection”. Instead, the etiology of stroke was classified as “cryptogenic” when the patients were grade 0 (absence of disease), 9 (insufficient workup), or 2 and 3 (being unable to establish a direct causal relationship between these diseases and the ischemic stroke).
2.2. Hemorrhagic transformation According to European Cooperative Acute Stroke Study (ECASS) radiological classification [9], hemorrhagic transformation has been classified in the following different types: hemorrhagic infarct type 1 (HI1; small petechiae along the periphery of the infarct), hemorrhagic infarct type 2 (HI2; confluent petechiae within the infarcted area without a space-occupying effect), parenchymal haematoma type 1 (PH1; bleeding b30% of the infarcted area with a mild space-occupying effect), parenchymal haematoma type 2 (PH2; bleeding N30% of the infarcted area with a significant spaceoccupying effect), and remote parenchymal hemorrhage (PHr; bleeding in brain areas remote from infarcted tissue).
Table 1 Demographic characteristics of the patients of the study.
Age (years) Women/men Neurological deficit (at admission) Patients number with NIHSSs b 8 Patients number with NIHSSs 8–16 Patients number with NIHSSs N 16 ASCOD phenotype Atherosclerosis, n (%) Small vessel disease, n (%) Cardioembolic, n (%) Other, n (%) Dissection, n (%) Cryptogenic, n (%) Cardiovascular risk factors Hypertension, n (%) Diabetes mellitus, n (%) Hypercholesterolemia, n (%) Atrial fibrillation, n (%) Previous CAD, n (%) Previous stroke, n (%) Smoking, n (%) Antihypertensive drugs, (%) ACE/angiotensin II receptor inhibitors Diuretics Beta-blockers Calcium channel blockers Laboratory parameters Glucose (mg/dl) Platelets (n/mm3) INR Creatinine (mg/dl) sodium potassium Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HbA1c (%) C reactive protein (mg/dl) Hemodynamic parameters SBP (mm Hg) DBP (mm Hg) MBP (mm Hg) PP(mm Hg) HR (beats/m)
Pt. without HT (n = 203)
Pt. with HT (n = 55)
P value
72.8 ± 13 94:109
75.2 ± 11.6 29:26
0.24 0.44
83 (40.9%) 61 (30%) 59 (29.1%)
6 (10.9%) 17 (30.9%) 32 (58.2%)
b0.001 0.76 b0.001 0.37
49 (24%) 6 (3%) 60 (29.5%) 3 (1.5%) 6 (3%) 79 (39%)
18 (33%) 1 (2%) 18 (33%) 1 (2%) 3 (5%) 14 (25%)
143 (70%) 40 (19.7%) 97 (47%) 67 (33%) 19 (9%) 41 (20%) 41 (20%) 127 (62%) 89 (43%) 63 (31%) 52 (25%) 29 (14%)
44 (80%) 13 (23.6%) 29 (52%) 23 (41%) 9 (16%) 10 (18%) 7 (12.7%) 40 (72%) 25 (45%) 19 (34%) 18 (32%) 6 (10%)
0.18 0.6 0.5 0.3 0.15 0.85 0.24 0.27 0.88 0.63 0.31 0.66
131 ± 45 219,774 ± 82,590 1.09 ± 0.2 0.9 ± 0.3 140 ± 3.8 4 ± 0.4 185 ± 42 114 ± 37 6 ± 0.8 2.01 ± 4.49
123 ± 29 194,509 ± 55,198 1.08 ± 0.2 1.02 ± 0.6 140 ± 5.9 3.9 ± 0.4 186 ± 43 116 ± 35 5.9 ± 0.8 2.77 ± 6.4
0.5 0.01 0.4 0.25 0.5 0.3 0.8 0.6 0.3 0.11
133 ± 15 69 ± 9 90.8 ± 10.2 64 ± 15 74 ± 13
142 ± 18 72 ± 11 95.9 ± 12.6 70 ± 16 72 ± 15
0.0007 0.12 0.008 0.005 0.5
Data are expressed as mean ± SD. NIHSSs: National Institutes of Health Stroke Scale score. CAD: coronary artery disease. ACE: angiotensin converting enzyme. SBP: systolic blood pressure. DBP: diastolic blood pressure. MBP: mean blood pressure. PP: pulse pressure. HR: heart rate.
468
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
2.3. Arterial stiffness index (ASI) BP monitoring was performed using validated oscillometric devices (Bedside Monitor Life Scope I BSM-2303K, International Div., Nihon Kohden Corp., Tokyo, Japan), in the first 24 h from the admission in the hospital. BP was recorded for 24 h every 30 min, between 06:00 h to midnight, and hourly, from midnight to 06:00 h. For inclusion in the study, at least 80% of valid BP and heart rate measurements for each subject were needed. By plotting the individual values of SBP and DBP measurements, obtained through 24-h non-invasive monitoring, the linear regression slope of DBP on SBP was obtained and assumed as a global measure of arterial compliance, and its complement (1 minus the slope), named ASI, has been taken as a measure of arterial stiffness [10]. 2.4. Statistical analysis All results are presented as means ± SD. Normal distribution of quantitative variables was preliminary tested using the Anderson-Darling test to select parametric or nonparametric inferential statistical methods. Homoscedasticity was tested using Bartlett test for normal distribution. Unpaired t-test was performed to compare SBP, MBP and HR in patients with and without HT. Mann-Whitney test was performed to compare age, serum potassium, serum sodium, HbA1c, C reactive protein, creatinine, total cholesterol, LDL-cholesterol, PP, glucose levels, platelet count, and INR in both groups. Welch's test was performed to test DBP. The two-sided Fisher's exact test was performed to evaluate statistical correlation between categorical variables (cardiovascular risk factors and antihypertensive drugs) evaluated in both groups of patients. The χ2 test was performed to evaluate ASCOD phenotypes; an extension of Fisher's exact test was performed to evaluate NIHSS scores. A p value below 0.05 was considered statistically significant. Logistic regression model was constructed to determine the predictors of HT. All significant variables of statistical descriptive analysis were entered initially. The statistical criterion for considering retention of a variable in the model was P b 0.05. Akaike's information criterion was used to select the best subset of predictor variables and corresponding OR were obtained. Statistical analysis was performed using R Statistical Software (version 3.3.2 for Windows, R Foundation for Statistical Computing, Vienna, Austria).
3. Results Out of 258 patients, HT occurred in 55 patients enrolled in the study, with the following radiologic sub-types: HI in 39 patients (HI1 in 27 patients and HI2 in 12 patients), PH in 14 patients (PH1 in 6 patients and PH2 in 8 patients), and PHr in 2 patients. Demographic characteristics of both groups of patients (with and without HT) are depicted in Table 1. Severity level of neurological deficit is higher in patients with HT; in particular, in severe deficit (NIHSSs N 16) the percentage of HT patients was significantly higher than that of non-HT patients, whereas in mild deficit (NIHSSs b 8) the percentage of HT patients was significantly lower than that of non-HT patients. Patients with moderate neurological deficit (NIHSSs between 8 and 16) were similar about percentage of HT. There were no differences between patients with and without HT in regard to age, sex, ASCOD phenotype, cardiovascular risk factors, and antihypertensive drugs used (Table 1). No differences were found regarding the main laboratory parameters evaluated in both groups (hs-CRP, total cholesterol, HDL, LDL, HbA1c, creatinine plasma levels, sodium, potassium, glucose, and INR), except for platelet count (194,509 ± 55,198 in the HT group vs 219,774 ± 82,590/μl in the non-HT group; p = 0.01) (Table 1). 3.1. ASI and other hemodynamic parameters In patients with HT, ASI was significantly higher in the HT group than in the non-HT group (0.70 ± 0.13 vs 0.62 ± 0.14, p = 0.0001) (Fig. 1); in particular, ASI of patients with HI and PH were higher than ASI of the non-HT patients' group (0.69 ± 0.13 and 0.73 ± 0.11 vs 0.62 ± 0.14, p = 0.02) (Fig. 2, a); these results were observed in adults (18–65 years) as well as in older adults (N 65 years) (Fig. 2, b and c). SBP, MBP and PP were significantly higher in patients with HT than in patients without HT (142 ± 18 vs 133 ± 15 mm Hg, p = 0.0007; 95.9 ± 12.6 vs 90.8 ± 10.2 mm Hg, p = 0.008; 70 ± 16 vs 64 ± 15 mm Hg, p = 0.005); whereas DBP and HR were similar in both groups (Table 1).
Fig. 1. Comparison of ASI in patients with and without HT. A: All patients (p b 0.001). B: Adult patients (age between 18 and 65 years) (p = 0.005). C: Patients over 65 years (p = 0.004). HT: hemorrhagic transformation. Unpaired t-test. ***p b 0.001, **p b 0.01.
3.2. Type of treatment Systemic thrombolysis was administered to 154 patients (60%), whereas mechanical thrombectomy was performed in 50 patients (19%) and a combined treatment (systemic thrombolysis + mechanical thrombectomy) was performed in 54 patients (21%). HT occurrence was significantly more frequent in patients treated with mechanical thrombectomy or combined treatment than in patients submitted to systemic thrombolysis (see Fig. 1S in supplementary data). There were no significant differences among the 3 groups in baseline demographic (Table 2); however, patients treated with mechanical thrombectomy or combined treatment had a higher severity level of neurological deficit, according to NIHSSs. Furthermore, in the “mechanical thrombectomy group” treatment started later than in the other 2 groups (Table 2).
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
469
Fig. 2. Comparison of ASI in patients without HT and with different types of hemorrhagic transformation. HT: hemorrhagic transformation. HI: hemorrhagic infarct. PH: parenchymal haematoma. PHr: remote parenchymal hemorrhage. Unpaired t-test, **p = 0.002.
3.3. Logistic regression and multivariable model Logistic regression model was constructed to determine the predictors of HT. All significant variables of statistical descriptive analysis were entered initially. These variables were the following: SBP, MBP, PP, ASI, type of treatment (systemic thrombolysis, mechanical thrombolysis, and combined thrombolysis), platelet count, and severity level of neurological deficit according to NIHSSs. Akaike's information criterion showed that the best subset of predictor variables for HT were the following: ASI (OR 1.81 for 0.2, 95% CI 1.09–3.02), MBP (OR 1.57 for 10 mm Hg, 95% CI 1.06–2.18), NIHSS (OR 4.63 for moderate vs mild level of neurological deficit, CI 1.55–17.15 and OR 8.05 for severe vs mild level of neurological deficit, CI 2.86–28.91) and platelet count (OR 0.94 for 10,000 platelets/μl, CI 0.89–1). In particular, when patients were classified into tertile groups by ASI, we observed that, in comparison with the first tertile (ASI b 0.57), the second (ASI: 0.57–0.71) and the third tertile (ASI N 0.71) showed an OR for HT of 2.4 (CI 0.97–5.91) and 5.2 (CI 2.22–12.24), respectively.
3.4. Outcome The functional outcome was evaluated at three months by means of mRS in patients with and without HT (see Fig. 2S in supplementary data). The distribution of mRS scores in the two groups at 90 days
shows a higher rate of death in the HT group in comparison to the non-HT group (18% vs 7%; p = 0.03). Furthermore, the proportion of patients with favorable functional outcome (mRS 0–2) was significantly higher in the non-HT group than in the HT group (mRS 0–2, 61% vs 30%; p = 0.0001). 4. Discussion The novel findings of our study are the following: 1) ASI is higher in patients with HT than in patients without HT, and 2) ASI represents an independent predictor of HT in patients with ischemic stroke treated with thrombolysis. In our study, ASI was measured during the first 24 h from the onset of symptoms, in order to evaluate the risk of HT occurrence in the next 24–72 h after initiation of treatment. Our results showed that ASI values N0.71 determine a risk of HT 5.2 fold higher in comparison with ASI values lower than 0.57. Furthermore, logistic regression analysis highlighted ASI as a predictor of HT, independently from other wellknown factors such as SBP, DBP or MBP. In fact, ASI explores the relationship of SBP and DBP, based on the premise that SBP and DBP are less tightly coupled in patients with higher arterial stiffness than in those with lower arterial stiffness. Our results suggest that arterial stiffness may represent an important factor contributing to HT in patients with ischemic stroke, irrespective of pathogenesis of stroke. Arterial stiffness is closely associated with small vessel diseases [6],
Table 2 Characteristics of stroke patients according to the type of treatment.
Age (years) Gender (F:M) NIHSSs at admission Patients with NIHSS score b 8 Patients with NIHSSs 8–16 Patients with NIHSS N 16 Time from the onset of symptoms (min)
Systemic thrombolysis (n: 154)
Mechanical thrombolysis (n: 50)
Combined therapy (n: 54)
P value
73.8 ± 13 70:84
72.9 ± 11 27:23
72.6 ± 14 26:28
0.45 0.57 b0.001
78 (50.7%)a,b 51 (33.1%) 25 (16.2%)a,b 178 ± 48a
8 (16%) 8 (16%) 34 (68%) 288 ± 172.5b
3 (5.6%) 19 (35.2%) 32 (59.2%) 172 ± 68
NIHSSs: National Institutes of Health Stroke Scale score. Kruskall-Wallis test. Chi squared test. a Versus mechanical thrombolysis group, b Versus combined therapy.
b0.001
470
M. Acampa et al. / International Journal of Cardiology 243 (2017) 466–470
but our results showed that, according to ASCOD phenotypes, the number of patients with small vessel disease is similar in both, the HT and non-HT groups. Following acute ischemic stroke, there is a breakdown of the blood– brain barrier resulting in friable intracranial vasculature, with the subsequent increased risk of HT. In healthy subjects, the aorta is highly compliant and a part of the pulsatile energy is stored in the arterial wall rather than being transmitted to the distal vasculature: this is a protective mechanism, limiting transmission of excessive pulsatility into the microcirculation. Instead, in patients with arterial stiffness, the arteries require a greater amount of force to expand and pulsatile energy is highly transmitted into the microcirculation, where it may cause damage [11]. The vasculature of the brain is particularly sensitive to excessive pressure and flow pulsatility, because it is characterized by high flow, which requires low microvascular impedance [12]. Therefore, in subjects with arterial stiffness the low-impedance brain circulation is exposed to greater pressure fluctuations with subsequent damage. This mechanism may explain, at least in part, the possible role of arterial stiffness in the pathogenesis of HT after ischemic stroke. Furthermore, cerebral autoregulation is impaired after an ischemic stroke, resulting in a relatively linear relation between systemic arterial pressure and cerebral perfusion pressure [13]. It is possible also that a high pulsatility may cause an altered reperfusion of ischemic territory through collateral circulation; a recent study has shown that high blood pressure variability may lead to instability of cerebral perfusion resulting in damage of ischemic territory and derangement of blood brain barrier [14]. Furthermore, our study showed a relationship among HT and different factors such as low platelet count, high NIHSS score (expression of severe neurologic deficit), and high mean blood pressure values, confirming the results of previous studies [15–18]. Surprisingly, according to Akaike's information criterion, the final “best” subset of predictor variables for HT excluded the variable “type of treatment”. Given that patients treated with mechanical or combined treatment had higher NIHSSs than patients treated with systemic thrombolysis, it is possible that HT occurrence in these subjects was already explained by the higher NIHSS scores. It is possible that a higher HT rate in patients treated with mechanical thrombectomy or combined treatment is due to more severe neurological deficit and higher time from symptom onset to treatment. In particular, regarding the combined treatment, we have reported the time from symptom onset to the first treatment (systemic thrombolysis); for this reason time to treatment with mechanical thrombectomy resulted higher than time to combined treatment. 5. Conclusions During the past decade, increased aortic stiffness has emerged as an important risk factor for target organ damage and cardiovascular disease events. Arterial stiffness index, based on 24-h BP monitoring, represents a simple and unexpensive tool providing valuable prognostic information about the risk of HT in patients with ischemic stroke undergoing thrombolytic therapy. In addition to known risk factors for bleeding after thrombolysis, we propose to evaluate ASI as a further bleeding risk factor for hemorrhage, suggesting caution when ASI values are N 0.71. Furthermore, the role of ASI as a predictor of HT could be also important in the group of patients with cardioembolic stroke, in particular regarding the choice of the optimal time to start anticoagulant therapy. As early HT is a major concern in the acute phase of stroke associated to atrial fibrillation, ASI N 0.71 (increasing of 5.2 fold the risk of HT) suggests delaying initiation of oral anticoagulation. Further studies will be
necessary to prove this hypothesis in a group of cardioembolic patients undergoing anticoagulant therapy. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2017.03.129. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. References [1] J. Zhang, Y. Yang, H. Sun, Y. Xing, Hemorrhagic transformation after cerebral infarction: current concepts and challenges, Ann. Transl. Med. 2 (2014) 81. [2] E.S. Sussman, E.S. Connolly Jr., Hemorrhagic transformation: a review of the rate of hemorrhage in the major clinical trials of acute ischemic stroke, Front. Neurol. 4 (2013) 69. [3] E.C. Jauch, J.L. Saver, H.P. Adams Jr., A. Bruno, J.J. Connors, B.M. Demaerschalk, P. Khatri, P.W. McMullan Jr., A.I. Qureshi, K. Rosenfield, P.A. Scott, D.R. Summers, D.Z. Wang, M. Wintermark, H. Yonas, Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association, Stroke 44 (2013) 870–947. [4] M. Kwarciany, D. Gąsecki, K. Kowalczyk, A. Rojek, S. Laurent, P. Boutouyrie, M. Skrzypek-Czerko, W.M. Nyka, K. Narkiewicz, B. Karaszewski, Acute hypertensive response in ischemic stroke is associated with increased aortic stiffness, Atherosclerosis 251 (2016) 1–5. [5] C. Palombo, M. Kozakova, Arterial stiffness, atherosclerosis and cardiovascular risk: pathophysiologic mechanisms and emerging clinical indications, Vasc. Pharmacol. 77 (2016) 1–7. [6] A. Tuttolomondo, R. Di Sciacca, D. Di Raimondo, A. Serio, G. D'Aguanno, A. Pinto, G. Licata, Arterial stiffness indexes in acute ischemic stroke: relationship with stroke subtype, Atherosclerosis 211 (2010) 187–194. [7] M. Acampa, F. Guideri, I. Di Donato, R. Tassi, G. Marotta, G. Lo Giudice, P. D'Andrea, G. Martini, Arterial stiffness in patients with deep and lobar intracerebral haemorrhage, J. Stroke 16 (2014) 184–188. [8] P. Amarenco, J. Bogousslavsky, L.R. Caplan, G.A. Donnan, M.E. Wolf, M.G. Hennerici, The ASCOD phenotyping of ischemic stroke (updated ASCO phenotyping), Cerebrovasc. Dis. 36 (2013) 1–5. [9] W. Hacke, M. Kaste, E. Bluhmki, M. Brozman, A. Dávalos, D. Guidetti, V. Larrue, K.R. Lees, Z. Medeghri, T. Machnig, D. Schneider, R. von Kummer, N. Wahlgren, D. Toni, Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke, N. Engl. J. Med. 359 (2008) 1317–1329. [10] Y. Li, J.G. Wang, E. Dolan, P.J. Gao, H.F. Guo, T. Nawrot, A.V. Stanton, D.L. Zhu, E. O'Brien, J.A. Staessen, Ambulatory arterial stiffness index derived from 24-hour ambulatory blood pressure monitoring, Hypertension 47 (2006) 359–364. [11] G.F. Mitchell, M.A. van Buchem, S. Sigurdsson, J.D. Gotal, M.K. Jonsdottir, Ó. Kjartansson, M. Garcia, T. Aspelund, T.B. Harris, V. Gudnason, L.J. Launer, Arterial stiffness, pressure and flow pulsatility and brain structure and function: the Age, Gene/Environment Susceptibility—Reykjavik study, Brain 134 (Pt11) (2011) 3398–3407. [12] L.L. Cooper, G.F. Mitchell, Aortic stiffness, cerebrovascular dysfunction, and memory, Pulse 4 (2016) 69–77. [13] S.L. Dawson, R.B. Panerai, J.F. Potter, Serial changes in static and dynamic cerebral autoregulation after acute ischemic stroke, Cerebrovasc. Dis. 16 (2003) 69–75. [14] K. Liu, S. Yan, S. Zhang, Y. Guo, M. Lou, Systolic blood pressure variability is associated with severe hemorrhagic transformation in the early stage after thrombolysis, Transl. Stroke Res. 7 (2016) 186–191. [15] V. Larrue, R.R. von Kummer, A. Muller, E. Bluhmki, Risk factors for severe hemorrhagic transformation in ischemic stroke patients treated with recombinant tissue plasminogen activator: a secondary analysis of the European-Australasian Acute Stroke Study (ECASS II), Stroke 32 (2001) 438–441. [16] C.S. Kidwell, J.L. Saver, J. Carneado, J. Sayre, S. Starkman, G. Duckwiler, Y.P. Gobin, R. Jahan, P. Vespa, J.P. Villablanca, D.S. Liebeskind, F. Vinuela, Predictors of hemorrhagic transformation in patients receiving intra-arterial thrombolysis, Stroke 33 (2002) 717–724. [17] A.K. Gilligan, R. Markus, S. Read, V. Srikanth, T. Hirano, G. Fitt, M. Arends, B.R. Chambers, S.M. Davis, G.A. Donnan, Baseline blood pressure but not early computed tomography changes predicts major hemorrhage after streptokinase in acute ischemic stroke, Stroke 33 (2002) 2236–2242. [18] J.H. Lee, K.Y. Park, J.H. Shin, J.K. Cha, H.Y. Kim, J.H. Kwon, H.G. Oh, K.B. Lee, D.E. Kim, S.W. Ha, K.H. Cho, S.I. Sohn, M.S. Oh, K.H. Yu, B.C. Lee, S.U. Kwon, Symptomatic hemorrhagic transformation and its predictors in acute ischemic stroke with atrial fibrillation, Eur. Neurol. 64 (2010) 193–200.