Renal Impairment Reduces the Efficacy of Thrombolytic Therapy in Acute Ischemic StrokePower A.a · Epstein D.b · Cohen D.c · Bathula R.c · Devine J.c · Kar A.b · Taube D.a · Duncan N.a · Ames D.b
aImperial Renal and Transplant Center and bImperial Hyperacute Stroke Unit, Imperial College Healthcare NHS Trust, and cNorthwick Park Hyperacute Stroke Unit, North West London Hospitals NHS Trust, London, UK Corresponding Author
Background: Renal impairment is a potent risk factor for stroke, which remains a leading cause of death and disability. Thrombolysis for acute ischemic stroke has transformed patient outcomes, although the safety and efficacy of this approach remain poorly characterized in patients with renal dysfunction, who manifest a higher risk of bleeding due to uremia. We therefore examined the impact of renal impairment on clinical outcomes with thrombolysis within the current 4.5-hour therapeutic window. Methods: This retrospective multicenter cohort study (2009–2011) examined 229 stroke patients receiving thrombolysis with alteplase (0.9 mg/kg; mean age 70 ± 13 years; 59% male, 24% diabetic). Sixty-five patients had an estimated glomerular filtration rate (eGFR) <60 ml/min. The primary outcome was the improvement in National Institutes of Health Stroke Scale (NIHSS) score at 24 h. Secondary outcomes included the NIHSS score at 7 days, the incidence of symptomatic and asymptomatic intracranial hemorrhage (ICH), extracranial bleeding and death during the index hospitalization. Univariate and multivariate regression analyses were performed to determine the association between demographic characteristics and comorbid factors of interest and outcomes. eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. Results: There was no significant difference in mean time to thrombolysis between the groups (221 ± 66 vs. 220 ± 70 min from symptom onset; p = 0.9). An eGFR <60 ml/min was independently associated with a statistically significant reduction of the therapeutic effect of alteplase at 24 h on multivariate regression [coefficient –2.3, 95% confidence interval (CI) –3.7 to –0.9; p = 0.002], and this persisted at 7 days (coefficient –3.5, 95% CI –5.3 to –1.7; p < 0.001). On modeling eGFR as a continuous variable, every 10 ml/min decline in eGFR was associated with a 0.40 diminution in NIHSS score improvement with alteplase (95% CI 0.07–0.74; p = 0.02). Older age and a higher presenting NIHSS score were associated with a greater therapeutic effect (p = 0.04 and p < 0.001, respectively). In-patient mortality was 5%, with no significant differences between groups. Renal impairment was not associated with a higher rate of ICH (6.2 vs. 6.7%; p = 0.9). Greater NIHSS score at presentation was the only factor associated with a greater risk of death (odds ratio 1.24, 95% CI 1.10–1.40; p < 0.001) and ICH (odds ratio 1.12, 95% CI 1.03–1.23; p = 0.004). Conclusions: Our results suggest that renal impairment is associated with reduced efficacy of thrombolysis in acute ischemic stroke without any excess hemorrhagic complications. This may relate to diminished fibrinolysis in the uremic milieu or differences in infarct anatomy. Longer-term prospective studies are required to characterize and improve functional outcomes following stroke in a manifestly high-risk group.
© 2013 S. Karger AG, Basel
Stroke remains a major international health issue and constitutes the leading cause of disability in developed countries. It remains one of the leading causes of death in the USA and Europe despite recent improvements in outcome. Ischemic stroke accounts for approximately 80% of events [1,2,3,4]. A number of acute therapies directed at achieving arterial recanalization have been developed. These include pharmacologic thrombolysis delivered either locally with the use of an intra-arterial catheter or systemically and mechanical embolectomy . Since the seminal National Institute of Neurological Disorders and Stroke Recombinant Tissue Plasminogen Activator study in 1995 , the safety and efficacy of systemic recombinant tissue plasminogen activator (or alteplase) for stroke has been confirmed in a number of large randomized controlled trials [7,8]. When delivered within 4.5 h of stroke symptom onset, alteplase results in a significant improvement in neurological deficit [as measured by the National Institutes of Health Stroke Scale (NIHSS)] and functional status compared to placebo . There is a low but appreciable risk of hemorrhagic transformation with alteplase (6–9%), which increases further when thrombolysis is administered more than 6 h after symptom onset [9,10]. The Third International Stroke Trial is currently evaluating outcomes in a wider range of patients (therapeutic window of 6 h, no upper age limit) .
Chronic kidney disease (CKD) is an independent risk factor for stroke . Patients with CKD have a higher incidence of stroke compared to the general population, with rates that increase progressively with each stage of CKD and reach a peak in patients with end-stage renal disease [13,14]. This may relate in part to the increasing prevalence of established atherosclerotic risk factors such as diabetes and hypertension as renal function declines as well as specific pathology such as uremic vascular calcification, although a specific pathophysiological mechanism has not been confirmed. The presence of CKD independently portends a worse prognosis after stroke, although the reasons for this are unclear [15,16]. Dialysis patients in particular have a higher prevalence of primary hemorrhagic stroke (reaching 30% in some series), with particularly poor poststroke survival [14,17,18].
The current therapy of choice for acute ischemic stroke is emergency thrombolysis, which in the London metropolitan area is delivered in dedicated hyperacute stroke units (HASUs). Although renal function per se is not a licensed contraindication to the use of thrombolysis, the efficacy and safety of this therapy in patients with CKD remains unclear as none of the major trials to date have reported renal function in their cohorts. It is known that uremia confers a multifactorial bleeding diathesis that intensifies as renal function declines and is particularly marked in end-stage renal disease . This bleeding risk is further enhanced in hemodialysis patients, who are often exposed to anticoagulation for maintenance of vascular access and hemodialysis circuit patency. Available data have not suggested an increase in adverse outcomes in hemodialysis patients treated with alteplase, although the total number of patients is very small . Importantly, to date there have been very few studies examining the impact of renal impairment on clinical outcome after the use of alteplase for ischemic stroke and they are limited in applicability due to heterogeneity in the definition of renal impairment and the dose of alteplase used. We aimed to determine the effect of renal impairment on the efficacy and safety of alteplase for acute ischemic stroke in a large and multiethnic urban population.
Subjects and Methods
This retrospective cohort study examined electronic and paper records of all patients treated sequentially per protocol with alteplase for acute ischemic stroke in the HASUs of two large urban hospitals for which we provide renal services in west London, UK. A total of 275 patients were identified, of whom 118 were from center 1 (Northwick Park Hospital HASU, 1 August 2009–31 December 2010) and 157 from center 2 (Charing Cross Hospital HASU, 1 December 2009–1 April 2011).
Baseline demographic data (age, gender, ethnicity) were captured as well as the presence of diagnosed comorbidities such as diabetes mellitus, ischemic heart disease (defined as a history of angina, myocardial infarction or coronary revascularization), hypertension, prior cerebrovascular disease (stroke and/or transient ischemic attack), peripheral vascular disease (clinical and/or radiologic evidence of aortic or distal arterial atherosclerotic disease) and atrial fibrillation (paroxysmal or chronic). Renal function was derived using the serum creatinine obtained on admission, and the estimated glomerular filtration rate (eGFR) was calculated using the CKD-Epidemiology Collaboration equation , as follows: GFR = 141 × min(sCr/ĸ, 1)α × max(sCr/ĸ, 1)–1.209 × 0.993age × 1.018 (if female) × 1.159 (if black), where sCr = serum creatinine in milligrams per deciliter, ĸ = 0.7 if female or 0.9 if male, α = –0.329 if female or –0.411 if male, min = the minimum of sCr/ĸ or 1 and max = the maximum of sCr/ĸ or 1. In the absence of prior tests of renal function, imaging and urinalysis results, CKD was defined in this study by an eGFR <60 ml/min/1.73 m2.
Time to treatment was defined as the duration of time between symptom onset and alteplase administration. In keeping with American Heart Association/American Stroke Association guidelines, acute ischemic stroke was defined as an episode of acute neurological dysfunction accompanied by evidence of clinically compatible infarction of central nervous system tissue on imaging . All patients were assessed on arrival to the emergency room by a senior stroke physician and a full clinical history obtained from the patient or their attendant next of kin if they were aphasic. The degree of neurodeficit was recorded using the NIHSS score on admission and 2 and 24 h after treatment, as well as at 7 days in patients who remained within the unit .
A standardized protocol for emergency thrombolysis for stroke existed in both centers. Patients suspected of out-of-hospital acute stroke were transferred rapidly to the emergency department of the dedicated stroke centers and reviewed by the on-call team of stroke physicians or neurologists, and those with possible stroke had emergency, non-contrast-enhanced CT imaging performed. The following assessments were performed in all cases: weight, temperature, blood pressure, heart rate, 12-lead electrocardiogram, capillary blood glucose, Glasgow Coma Scale score, full blood count, coagulation screen, renal and liver function, serum electrolytes and C-reactive protein. Patients were assessed for eligibility for thrombolytic therapy and excluded if they met any of the criteria listed in the Appendix.
Thrombolysis was administered under the direction of the duty stroke consultant in the high-dependency areas of the HASU in each center with appropriate clinical monitoring. Alteplase was prescribed at a dose of 0.9 mg/kg to a maximum dose of 90 mg, with 10% of the total dose administered as a 2-min infusion and the remainder over 60 min. No dose adjustment was made for renal function. All antiplatelet therapy was withheld from admission to 24 h after thrombolysis. The infusion of alteplase was stopped immediately in the event of suspected intracranial hemorrhage (ICH; e.g. fall in Glasgow Coma Scale score, increasing focal neurological deficit, seizures, meningism or papillary asymmetry), and immediate repeat CT scanning and blood tests were performed. Interval CT imaging was repeated in all cases at 24 h per protocol, and all imaging was reviewed by experienced neuroradiologists at the treating center. Patients were monitored in the HASU for 72 h following treatment and transferred thereafter to their nearest stroke unit for ongoing rehabilitation if required, in accordance with established regional guidelines.
The primary outcome measure was the change in NIHSS score from admission to 24 h after thrombolysis. Secondary outcome measures included the incidence of ICH, both symptomatic and asymptomatic, the incidence of extracranial bleeding, discharge destination and death during the index hospitalization.
Ethical approval for this study was granted as part of audit and service development by our institutional review board, which waived the requirement for informed consent.
Descriptive statistics were expressed as the mean ± standard deviation and the median with interquartile range (IQR), as appropriate. Continuous and categorical variables were compared using Student’s t test and the χ2 or Mann-Whitney U test, respectively, as appropriate.
Univariate and multivariate regression analysis was used to examine the effect of factors of interest such as eGFR, age, gender, ethnicity and comorbid conditions on the degree of improvement in NIHSS score. Factors identified on univariate analysis with a p value <0.1 were then examined using a multivariate model with testing for interaction of terms. A backwards selection procedure was then applied to this model to identify risk factors of significance. Similarly, logistic regression analysis was used to examine the association between these factors and the risk of ICH and death. All analyses were performed using Stata 10.0 (www.stata.com). Statistical significance was defined by p < 0.05.
Two hundred and seventy-five patients received alteplase for acute ischemic stroke during the study period. Complete data were available in 229 cases, which formed the cohort under analysis (table 1).
|Table 1. Study cohort characteristics (n = 229)|
Overall, 65 patients (28%) had an eGFR <60 ml/min and formed the CKD subgroup (median eGFR 44 ml/min, IQR 37–52), including 1 patient established on maintenance hemodialysis. CKD patients were significantly older (p < 0.001) and had a greater prevalence of diabetes, ischemic heart disease and hypertension compared to patients with no renal dysfunction (table 2). There were no significant baseline differences in gender, ethnic case mix or established cerebrovascular disease between the subgroups (p = 0.4).
|Table 2. Comparative baseline characteristics of thrombolysed patients according to the presence of renal dysfunction (CKD defined as eGFR <60 ml/min)|
When modeled as a continuous variable, eGFR was positively associated with a greater postthrombolytic improvement in NIHSS score on multivariate analysis [coefficient 0.40 per 10 ml/min, 95% confidence interval (CI) 0.07–0.74; p = 0.02].
There was no significant difference in the time to thrombolytic treatment between the CKD and non-CKD groups (221 ± 66 vs. 220 ± 70 min; p = 0.9). Patients with CKD had higher median NIHSS scores at presentation compared to those with no renal dysfunction [10 (IQR 7–14) vs. 8 (IQR 6–14); p = 0.05]. This difference in median NIHSS scores persisted and was more marked 24 h after treatment [6 (IQR 3–11) vs. 3 (IQR 1–8); p = 0.004].
Higher NIHSS score at baseline (p < 0.001), the presence of ischemic heart disease at baseline (p = 0.03) and increasing age (p = 0.04) were independently associated with a greater improvement in the NIHSS score after thrombolysis on univariate analysis (table 3). These factors remained significant on multivariate analysis, although the effect of ischemic heart disease achieved borderline statistical significance (p = 0.05). Univariate analysis by age band suggested a significantly greater benefit of thrombolysis in the 70- to 80-year age group compared to the younger and older groups (coefficient 1.9, 95% CI 0.03–3.86; p = 0.05). By contrast, the presence of CKD (eGFR <60 ml/min) led to a statistically significant diminution of the 24-hour treatment effect of alteplase (coefficient –2.3, 95% CI –3.7 to –0.9; p = 0.002) on multivariate analysis (table 3). This effect persisted at 7 days following thrombolysis (n = 148; coefficient –3.5, 95% CI –5.3 to –1.7; p < 0.001).
|Table 3. Regression model of the associations between the change in NIHSS score 24 h following thrombolysis and factors of interest|
ICH occurred in 4 patients (6.2%) in the CKD group and 11 (6.7%) in the non-CKD group, with no significant difference in prevalence (p = 0.9). Symptomatic ICH (defined by a deterioration in NIHSS score ≥4) occurred in 3 of 4 cases in the CKD group and 6 of 11 in the non-CKD cohort (p = 0.6). Higher baseline NIHSS score was the only factor associated with an increased risk of ICH on multivariate analysis [odds ratio (OR) 1.12, 95% CI 1.03–1.23; p = 0.004].
Twelve patients died following thrombolysis (5% of the total cohort, 6 in each group). The presence of CKD was not independently associated with a higher mortality risk (p = 0.07). Those who died were significantly older (mean age 77 vs. 70 years; p = 0.01) and had a higher prevalence of diabetes (50 vs. 22%; p = 0.05). A higher NIHSS score at baseline was independently associated with a higher risk of death following thrombolysis (OR 1.24, 95% CI 1.10–1.40; p < 0.001).
Two patients developed extracranial bleeding following thrombolysis. The first (eGFR 39 ml/min) experienced self-limited thoracic and abdominal wall hematomas while the second had recurrent epistaxis and vaginal bleeding that resolved spontaneously (eGFR 79 ml/min).
There was a trend for longer hospitalization in patients with CKD (mean length of stay 11.9 vs. 7.9 days; p = 0.07). CKD was not associated with a lower chance of discharge home following thrombolysis; the admission NIHSS score was the only significant factor influencing this outcome after adjusting for ICH (OR 0.91, 95% CI 0.87–0.96; p < 0.001).
We have demonstrated that renal impairment is independently associated with evidence of diminished neurological improvement in patients treated with alteplase for acute ischemic stroke. Our findings are consistent with data from other series [23,24] and represent the first such analysis of outcomes with the current extended 4.5-hour time frame rather than 3 h as in prior publications [23,24,25].
Patients with renal dysfunction in this study were on average significantly older and had a higher prevalence of coexistent ischemic heart disease and atrial fibrillation, in keeping with other series [24,25]. However, there are significant differences in cohort characteristics which may have influenced outcomes and limit comparisons. Naganuma et al.  report a homogeneous Japanese cohort in contrast to the multiethnic cohort of Agrawal et al.  and ourselves. There was an underrepresentation of diabetic patients in the Japanese cohort (20%)  compared to ours (37%) and that of Agrawal et al.  (50%), where figures are more comparable with a US population. The overall rates of ischemic heart disease in the Japanese series were concomitantly lower compared to our cohort (13 vs. 27%; p < 0.001). Lyrer et al.  provide categorical comorbidity data for their whole cohort but do not stratify according to renal function and provide no objective measure of stroke severity, making it difficult to use their data for comparative analyses.
There were no significant differences in NIHSS scores between subgroups in prior studies, in contrast to our findings [24,25]. Median NIHSS scores at presentation were higher in the studies of Naganuma et al.  and Agrawal et al.  compared with our own (non-CKD group: 12 and 12 vs. 8, respectively; CKD group: 14 and 13 vs. 10, respectively), but this may reflect unrecorded differences in hospital transit times or other confounders.
Retrospective studies such as this are hypothesis generating and cannot prove causality. The borderline and seemingly counterintuitive benefit afforded by a diagnosis of ischemic heart disease may be the result of residual confounding by factors such as the prescription of antiplatelet therapy and lipid-lowering drugs. However, we carried out a post hoc analysis of data from the Northwick Park HASU and found no differences in prescription of antiplatelet agents between groups, although renal dysfunction was associated with more statin prescription (62 vs. 33%; p = 0.02). There is conflicting evidence regarding the influence of statins on outcomes after thrombolysis [26,27], and one study suggests they increase the risk of ICH .
The mechanism whereby renal impairment may attenuate the improvement of neurodeficit with alteplase is unclear. Renal impairment may lead to clinically significant differences in the efficacy of thrombolysis by modulating thrombus mechanics. Studies have shown that patients with end-stage renal disease have more compact, rigid clots with smaller fibrin network pore sizes and greater resistance to fibrinolysis with alteplase in vitro compared to those from people with normal renal function [29,30]. CKD is associated with greater subclinical cerebrovascular disease burden , and this may compromise collateral blood flow around infarcts, a factor known to influence recanalization rates [32,33]. It is possible that patients with renal impairment have larger infarct volumes and/or proportionately smaller ischemic penumbrae that are responsive to reperfusion . There are no published data to confirm or refute any of these hypotheses at present.
Renal impairment was not associated with a higher risk of ICH in our study. Interestingly, despite using a lower dose of alteplase (0.6 mg/kg), Naganuma et al.  reported higher rates of ICH (27%) in patients with renal dysfunction, but the reasons for this are not clear. We described lower rates of ICH than those reported in the Safe Implementation of Treatment in Stroke-International Stroke Thrombolysis Registry study, where rates of any ICH reached 14–16% using 0.9 mg/kg alteplase . This variation may relate to greater experience with alteplase in the present era, variability in pretreatment blood pressure and glucose profiles  or confounding from occult factors affecting patient selection. The association between NIHSS score at baseline and the risk of ICH that we described is in keeping with prior international registry data .
The overall mortality rate of 5% in our study compares favorably with international data showing rates of 7% in groups treated within 3 and 3–4.5 h  and is compatible with prior outcomes in patients with CKD .
The strengths of this study compared to prior work include the use of the CKD-Epidemiology Collaboration equation to estimate GFR and reduce misclassification, homogeneous treatment protocols and applicability by virtue of reporting on up-to-date clinical practice. However, we are mindful of its limitations, whereby reported associations cannot infer causation and remain prone to multiple confounders such as selection bias, confounding by indication and variables that were not captured, such as admission glucose level or blood pressure, which are known to influence outcomes with thrombolysis. In keeping with prior studies, we examined serum creatinine on admission but are aware that the calculated eGFR does not necessarily reflect premorbid eGFR and can be altered by the effects of acute stroke such as volume depletion. Although this forms the largest European cohort to date, the sample size remains relatively small compared to international stroke registries, and we cannot rule out center-specific treatment effects that may have biased outcomes. This study did not analyze longer-term outcome data (e.g. at 3 months), and the medium- as well as long-term effects of thrombolytic therapy in patients with CKD remain uncharacterized. Although relatively objective, the NIHSS score was the sole outcome measure and cannot account for valuable information gained by assessing functional status using validated scores such as the modified Rankin scale.
The burden of stroke in patients with advanced renal dysfunction is gradually being evaluated. At a time when increasing numbers of patients with CKD are receiving thrombolysis for acute stroke, the safety-efficacy profile of this therapy remains poorly described. It is unclear whether any modulation of alteplase delivery is required for these patients. A better understanding of the mechanisms underlying the negative effect of renal impairment on outcomes may allow for evidence-based interventions to improve stroke management in this high-risk group.
A portion of this work was presented at the annual meeting of the American Society of Nephrology, 8–13 November 2011, Philadelphia, Pa., USA.
Dr. Albert Power, MD
Imperial Renal and Transplant Center, Imperial College Healthcare NHS Trust
4th Floor, Hammersmith House, Hammersmith Hospital, DuCane Road
London W12 0HS (UK)
Received: August 17, 2012
Accepted: October 11, 2012
Published online: February 14, 2013
Number of Print Pages : 8
Number of Figures : 0, Number of Tables : 3, Number of References : 35
Vol. 35, No. 1, Year 2013 (Cover Date: February 2013)
Journal Editor: Hennerici M.G. (Mannheim)
ISSN: 1015-9770 (Print), eISSN: 1421-9786 (Online)
For additional information: http://www.karger.com/CED