Effect of Alteplase Administration on International Normalized Ratio in Patients With Acute Ischemic Stroke

Abstract

Background/Objective:

Alteplase may elevate international normalized ratio (INR) results, although the exact rate of elevation occurrence is not firmly established in the literature. The purpose of this study is to determine the occurrence rate of INR elevation following alteplase administration. We also aimed to determine what factors are independently associated with the development of elevated INR following alteplase administration for ischemic stroke.

Methods:

We conducted a multicenter, retrospective, cohort study of patients who received alteplase for acute ischemic stroke. Patients were screened for baseline INR measurement and a repeat value within 24 hours of alteplase administration. The primary outcome was the percent of patients who experienced ≥0.4-point increase in INR. Secondary outcomes included the rate of adverse bleeding events and identification of factors independently associated with elevated INR following alteplase administration.

Results and Conclusions:

Two hundred and sixty-one patients were included, with 44 (16.9%) patients having an INR increase of 0.4 or more. Patients with an INR increase ≥0.4 experienced a nonstatistically significant increase in bleeding episodes (8.8% vs 18.2%; P = .10). We identified African American race (odds ratio, 3.48, 95% confidence interval, 1.5-7.6; P = .002) as an independent predictor of INR increase ≥0.04. An INR elevation is common following receipt of alteplase for ischemic stroke. Those of African American race were at increased risk of INR elevation; however, more studies are needed to determine whether these patients are at a higher bleeding risk as a result of INR elevation.

Introduction

Each year, approximately 795,000 Americans experience a stroke, 87% of which are ischemic.¹ Thrombolytic therapy with recombinant tissue plasminogen activator (tPA) is recommended to improve functional outcomes.² The 2013 American Heart Association (AHA) guidelines for the early management of acute ischemic stroke recommend measurement of international normalized ratio (INR) prior to alteplase administration to rule out recent use of oral anticoagulation.³ While the 2018 AHA guidelines, however, do not recommend INR measurement prior to initiation of tPA in patients known not to be taking vitamin K antagonists.² Although these guidelines do not make recommendations for measuring INR after alteplase administration, follow-up INR measurements may be used to assess for coagulopathy or to further evaluate bleeding risk following thrombolysis.

Little evidence exists regarding the impact of routine measurement of INR in patients with ischemic stroke, though it may inform providers of bleeding risk. A previous report found that 7.4% of patients had an INR elevation above 1.5 following alteplase.⁴ Although not completely understood, a possible mechanism of increased INR is due to alteplase degrading clot-bound fibrinogen and fibrin.^5,6 Decreased fibrinogen after thrombolysis has been associated with intracerebral hemorrhage, and INR would be a valuable point-of-care test to identify patients at increased risk.⁷ Currently, there is scarce data characterizing the impact of alteplase administration on INR in patients with acute ischemic stroke.⁴ The purpose of this study is to further quantify the relationship between alteplase administration and INR elevation, as well as to identify factors independently associated with INR elevation.

Methods

This was a multicenter, retrospective, cohort study conducted at two urban, academic medical centers. We evaluated for inclusion all patients who received intravenous alteplase for ischemic stroke aged 18 years or older, discharged between October 1, 2011, and August 15, 2016, and had an INR measured prior to and within 24 hours of alteplase administration. Exclusion criteria are listed in Figure 1. The primary objective was to describe the relationship between intravenous alteplase and INR. Investigators at each site extracted all data from the institution’s electronic medical record. The primary aim was to determine the percent of patients who experienced a significant increase in INR, as well as predictors for this outcome. We defined a significant increase in INR as a 0.4-point or greater increase in INR after receipt of alteplase rather than an absolute INR cutoff to account for patients with a slightly elevated baseline INR. For example, a patient with a baseline INR of 1.1 would only require a 36% increase to reach 1.5, while a patient with a baseline INR of 0.9 would require a 67% increase. Our study definition for INR elevation (increase >0.4) was chosen after evaluating the average baseline INR of previous studies and the change needed to satisfy their primary outcome variable of INR >1.5, values we felt could be clinically significant.⁴ Secondary aims were to determine the rate of adverse bleeding episodes between groups. We planned to conduct an analysis to identify factors independently associated with bleeding only if there was a statistical significance found between groups. Institutional review board approval was obtained at each of the participating medical centers.

Figure 1.

Patient enrollment.

Data collected included initial National Institute of Health Stroke Scale (NIHSS) scores, baseline demographics, alteplase dosing and administration information, relevant past medical history and home medications, baseline liver function panel results, inpatient INR measurements prior to and for up to 24 hours following alteplase, and all documented bleeding events within 36 hours of alteplase receipt. Symptomatic intracranial hemorrhage (ICH) events were collected and defined as any new ICH on imaging combined with at least a 4-point increase in NIHSS.⁸ We also collected whether or not patients underwent a neurointerventional procedure prior to follow-up INR measurement. Continuous data were analyzed using the Student t test for parametric data and the Mann-Whitney U test for nonparametric data. Categorical data were analyzed using either the χ² test or the Fisher exact test. Normally distributed data are presented as mean (standard deviation) and non-normally distributed data as median (25%-75% interquartile range). All times are presented as hours: minutes. All tests were 2 tailed, and P < .05 was used to represent statistical significance.

To elucidate risk factors associated with INR elevation after alteplase administration, a univariate analysis was conducted between those who developed an increased INR and those who did not. All variables with P value <0.2 from this analysis were entered into a stepwise multiple logistical regression model. The logistic regression was calibrated based on a nonsignificant Hosmer-Lemeshow goodness-of-fit P value >0.05 and model discrimination was assessed based on the area under the receiver operator characteristic curve. Results were reported as adjusted odds ratio (OR) and corresponding 95% confidence intervals (CI). All analysis was performed using SPSS, version 23.0 for Windows (SPSS, Inc, Chicago, Illinois).

Results

A total of 1938 patients who received alteplase and had a baseline and follow-up INR were screened, and 261 patients were included for analysis. Common reasons for exclusion were lack of a follow-up INR, receipt of warfarin, or a direct-acting oral anticoagulant within the previous 7 days and baseline aspartate aminotransferase/alanine aminotransferase elevation (Figure 1). Patient baseline demographics can be found in Table 1. The median (25%-75% interquartile range) baseline INR among all patients was 1.0 (1-1.1) and increased to a median of 1.2 (1.1-1.3) after alteplase administration. Forty-four (16.9%) patients had an increase in INR greater than 0.4.

Table 1.

Patient Demographics.^a

Characteristics	Increase in INR <0.4, n = 217	Increase in INR >0.4, n = 44	P
Age (years)b	63.5 (14.3)	66.7 (12.0)	.44
Female, n (%)	106 (48.8)	24 (54.5)	.49
African American, n (%)	111 (51.2)	35 (79.5)	.001
Height, cm	170 (163-180)	168 (163-175)	.36
Weight, kg	81.6 (69.3-99.0)	84.5 (70.5-110.4)	.35
Initial NIHSS	7 (4-14)	12 (5-17)	.01
Alteplase dose, mg	73.4 (61.0-90.0)	76.4 (64.0-90.0)	.37
Neurointervention performed, n (%)	30 (14.0)	13 (29.5)	.01
Baseline INR	1 (1-1.1)	1 (1-1.1)	.91
Time from alteplase to follow-up INR, hours: minutes	8:46 (4:45-13:29)	7:40 (4:36-16:12)	.63
Baseline liver function
Total bilirubin, mg/dL	0.5 (0.3-0.7)	0.4 (0.3-0.5)	.15
AST, U/L	21 (15-27)	20 (16-25)	.56
ALT, U/L	24 (18-34)	22 (17-29)	.73
Home medications, n (%)
Aspirin	76 (35.2)	16 (36.4)	.88
P2Y12 antagonist	20 (9.3)	3 (6.8)	.78
Aspirin + dipyridamole	2 (0.9)	0 (0)	.52
Dual-antiplatelet therapy	11 (5.1)	2 (4.5)	>.99
Past medical history
Atrial fibrillation	21 (9.8)	5 (11.4)	.75
Coronary artery disease	49 (22.8)	10 (22.7)	.93
Diabetes mellitus	56 (25.9)	13 (29.5)	.62
Tobacco use	63 (29.4)	8 (18.6)	.15

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; INR, international normalized ratio; NIHSS, National Institutes of Health Stroke Scale.

^aAll data are presented as median (25%-75% interquartile range unless otherwise noted).

^bMean (standard deviation).

There were 3 characteristics found to occur more commonly in those patients with an increased INR: African American race (79.5% vs 51.2%; P = .001), higher initial NIHSS (12 [5-17] vs 7 [4-14]; P = .01), and occurrence of neurointervention (29.5% vs 14%; P = .01). All other baseline characteristics were similar between the two groups. Multivariable logistic regression analysis showed that patients of African American race were more likely to experience an INR increase (OR, 3.48, 95% CI, 1.5-7.6; P = .002; Table 2). There was a nonstatistically significant increase in any bleeding event in patients who had an increase in INR (18.2% vs 8.8%; P = .10, Figure 2). Data comparing those with and without bleeding events are reported in Supplemental Table 1.

Table 2.

Multivariable Logistic Regression Analysis to Identify Predictors of an INR Increase of 0.4 or More.^a

Variable	Odds Ratio	95% Confidence Interval	P
African American race	3.48	1.5-7.6	.002
Neurointervention performed	2.1	1-4.7	.051

Abbreviations: INR, international normalized ratio; NIHSS, National Institutes of Health Stroke Scale.

^aHosmer-Lemeshow goodness-of-fit test = 0.95; area under the curve of receiver operating characteristic 0.68 (0.59-0.76); baseline total bilirubin and tobacco use did not accurately predict outcome. Initial NIHSS was placed into the model separately from neurointerventional procedure due to multicollinearity and also did not accurately predict outcome.

Figure 2.

Bleeding complications.

Conclusion

We found that INR elevations of 0.4 or more are common after alteplase administration. We also noted that those with an INR elevation experienced a bleeding rate nearly double that of those without INR increase. Although this may be clinically significant, we did not find a statistical difference between the two groups.

A previous study analyzed the effect of alteplase on INR for 284 patients with acute ischemic stroke. They defined coagulopathy post-tPA as an INR elevated >1.5, which occurred in 7.4% (21 patients). They also reported fibrinogen levels in a small cohort of patients. Of the 12 patients with fibrinogen levels, 9 patients had low levels, matching their elevated INR.⁴ Although this study was hypothesis generating, it lacked multivariate regression analysis to identify predictors of increased INR in their patient population and did not exclude patients with baseline elevation in INR due to liver dysfunction or use of medications that elevate INR. Although still retrospective, our methodology addressed these baseline characteristics in screening patients, and we identified independent predictors of increased INR after alteplase administration rather than an association between the two. Finally, we elected to utilize a different definition for elevated INR than the previous trial (≥1.5) in an effort to detect significant increases in INR in patients with variable baseline values.

The results of the study by Lee and colleagues did provide information that may help to validate the theory that alteplase may elevate INR via release of fibrinogen degradation products that have been shown to prolong clot formation and impair fibrinolysis in animal studies.^4,5 Lee and colleagues found that one-third of the patients who had increased INR >1.5 also had decreased fibrinogen levels.⁴ Unfortunately, fibrinogen levels were not part of routine care at either institutions during the data collection period and we were unable to confirm the findings of this study. Our results did show an interesting relationship between alteplase administration and INR elevation that reveals a potential new explanation for the increased INR seen after alteplase administration. Those with an INR increase were more likely to have higher NIHSS and undergo a neurointerventional procedure. It is reasonable to infer that these patients may have had a larger clot burden and increased clot degradation products may have contributed to a greater increase in INR. These differences in NIHSS and rates of neurointervention between groups may have also contributed to the higher bleeding rates seen in the increased INR group. Many landmark alteplase trials have shown an association between higher NIHSS and larger clot burden with higher risk of hemorrhagic conversion.^6,8,9 Whether this bleeding is due to increased coagulopathy from the fibrinolysis or a result of reperfusion of ischemic tissue after neurointervention is unknown.

Large population-based studies have shown that African Americans are at a higher risk of coronary artery diseases and stroke than any other race in the United States; however, a distinct explanation for this has not been established.^10,11 It has been hypothesized that differences in endothelial markers and plasma hemostasis may play a role in the development of these chronic diseases.¹² In the Multi-Ethnic Study of Atherosclerosis epidemiologic cohort study of hemostatic factors and endothelial markers across varying racial groups, Lutsey and colleagues concluded that African Americans had the most thrombogenic profile, including elevated levels of factor VIII, d-dimer, and von Willebrand factor.¹² This increased thrombogenicity in the setting of alteplase administration, larger clot breakdown, and greater release of clotting factors may be the source of INR elevation in this patient population. The increased prevalence of INR elevation in patients with African American race seen in our study may further point to a coagulopathy present in this patient population or an association with increased NIHSS and larger clot burden not previously described in the literature.

Our study is not without limitations. First, although we screened nearly 2000 patients, we only had follow-up INR monitoring in only 13.4%. Additionally, we were unable to report fibrinogen levels and its association with the primary outcome as it is not a routinely ordered lab in this patient population at participating institutions. There also could be interpatient variability due to non-standardized timing of follow-up lab studies. Additionally, due to lack of uniform assessments of glucose levels and renal function markers, we were unable to report rates of elevated glucose and acute renal impairment, both of which have been previously identified as risk factors associated with bleeding following tPA administration.¹³ Finally, we are unable to report the use of anticoagulation products that may have been used during any neurointerventional procedures due to documentation limitations at each institution. Our study also has its strengths. Compared to previous literature, we chose to examine the increase in INR from baseline, a definition that has not been previous used. However, we believe this strengthens the results observed and may help accommodate other confounders that may have led to slight elevations in individuals, including patients with a mildly elevated baseline INR.

An INR elevation is common following receipt of alteplase. Those of African American race are at highest risk of INR elevation. Future studies are needed to determine whether postthrombolysis INR elevations lead to increased bleeding events and if there is a predictive value between elevated INR following alteplase administration and bleeding risk.