Gender Differences in acute stroke treatment–The UCSD experience

Abstract

Background and Purpose

To assess the gender differences in patients with acute ischemic stroke treated with and without t-PA. The primary purpose is to evaluate for differences in baseline risk factors, treatment times, and 90-day outcomes. Data regarding gender differences in acute stroke treatment shows a delayed treatment and evaluation in women with stroke, associated with poorer outcome.

Methods

Review of the UCSD SPOTRIAS database from 2001 to 2009. All “code stroke” patients with the admitting diagnosis of acute ischemic stroke were classified based on gender and t-PA treatment (Group 1 with t-PA, Group 2 without).

Results

A total of 848 patients were included, Group 1: 294 patients, baseline NIHSS and age in men was 12±7.9 and 67.6±16.5 years, in women 13.7±7.9 and 72.4±16.5 years. Group 2: 554 patients, baseline NIHSS and age in men was 7.2±8.0 and 68.4±14.0 years, in women 8.0±8.0 and 72.2±14.0 years. Women and men had a similar t-PA treatment rate (women=38%, men=32%). Men were more likely to use tobacco and have a history of CAD and less likely to have a history of AF. In both groups, men had higher rates of a 90-day mRS of 0-1 compared to women (Group 1 35.5% vs 22.6%; Group 2 57.3% vs 47.3%). Multivariable analysis adjusting for observed confounders, namely admission NIHSS, CAD/MI and AF, indicate that this difference was not statistically significant. We found no significant difference between genders when assessing for treatment times in either group.

Conclusions

Treatment times and patient outcomes after treatment for acute ischemic stroke were similar between genders. Previously identified sex differences in stroke treatment times and outcome were not found in our sample.

Introduction

Intravenous tissue plasminogen activator (t-PA) is currently the only approved therapy for acute ischemic stroke. Only 2–3% of acute stroke patients receive this treatment.^1,2 The primary reason for this low rate is delayed presentation to an Emergency Department (ED).³ Because women have a longer life expectancy than men, and age is associated with a higher incidence of stroke, more women than men die of stroke.⁴

Women with acute myocardial infarction receive less medical therapy, experience longer treatment delays, undergo fewer invasive procedures and have a higher risk of adverse outcome than men.^5,6 Data suggests, similarly, that women with ischemic stroke experience similar delays in treatment and lower frequency of thrombolysis.⁷ Women present later to the ED⁸ and are less likely to receive brain imaging within 1 hour of ED arrival.⁹

There is limited data regarding the differences between gender regarding stroke risk factors, treatment times, and outcome after acute ischemic stroke. Previous studies have shown that women are older and more often have a history of atrial fibrillation, less likely to be smokers or have a history of coronary artery disease than men.^8,10–11 Data also shows that women tend to have more severe strokes and are more likely to have a poorer modified Rankin Scale score (mRS) at discharge.¹² Delays in arrival to the Emergency Department may explain why women have a reported lower rate of t-PA treatment.^8,11–13 This could be critical in targeting women for improving stroke symptom awareness and urgency of rapid Emergency Department assessment.

Women have worse functional outcome even 5 years after their first stroke as compared to men.¹⁴ A pooled-analysis of sex-based differences in response to t-PA treatment, demonstrated that women with acute ischemic stroke appeared to have significantly more benefit at 90-days from t-PA than men.¹⁵

We evaluated the UCSD Stroke Center experience regarding gender differences in acute stroke diagnosis and treatment.

Methods

We reviewed all “Code Stroke” calls from January 2001 to April 2009, who were entered into the UCSD SPOTRIAS database and had an admission diagnosis of ischemic stroke. The database allows for collection of treatment times, demographics and outcome data from patients seen at six San Diego Emergency Departments for acute stroke. Of the six hospitals, two are academic facilities, one is a Veteran’s hospital and three are community/private. Five out of the six hospitals are TJC (The Joint Commission) certified Primary Stroke Centers.¹⁴ Patients were excluded from analysis if they were transferred from another hospital, had a stroke while in the hospital, or had a hemorrhagic stroke. Patients were classified by gender and t-PA treatment. Patients who were treated with t-PA are in Group 1 and those without t-PA in Group 2. Univariate comparisons between genders were performed using Wilcoxon Rank-Sum test for continuous variables and the Fisher’s Exact Test for categorical variables, overall. Treatment times were captured and analyzed to assess time of stroke onset to arrival, physician evaluation, treatment decision, CT scan, and t-PA bolus (Group 1). Treatment times were compared between genders for both groups using the Wilcoxon Rank-Sum test. In the subset of patients with 90-day outcomes, day 90 mRS was dichotomized (0–1 vs 2–6). A multivariable logistic regression analysis was performed to control for multiple observed confounding factors. Variables were considered to be confounders and included as covariates in the multivariable model if found to be associated with gender (p<0.1) and moderately associated with response (p<0.15) based on univariate analysis. Discharge destination was compared between genders using the Fisher’s Exact test. A good outcome at hospital discharge was defined as discharge to home or acute and subacute rehabilitation facility. A poor outcome at discharge was defined as discharge to skilled nursing facility, death, and other (hospice). All analyses were done with the statistical software R 2.1.1.

Results

In the time period under study, 2027 Code Stroke alerts were registered. A total of 848 patients had an ED admitting diagnosis of acute ischemic stroke with a decision regarding t-PA and were included in our study. Baseline characteristics are summarized by group in Table 1. There were 294 patients (148 women and 146 men) who were treated with t-PA (Group 1) and 554 patients (242 women and 312 men) without t-PA (Group 2). Women and men had a similar t-PA treatment rate (women 38%, men, 32%). Women in both groups were older than men (Group 1, 72.4±16.5 vs 67.6±16.5, p=0.008; Group 2, 72.2±14.0 vs 68.4±14.0, p=0.001). In both groups, men had a significantly higher rate of coronary artery disease (Group 1, 30.1% vs 14.9%, p=0.02; Group 2, 26.3% vs 17.8%, p=0.02) and a higher rate of tobacco use (Group 1, 33.6% vs 16.2%, p=0.001; Group 2, 29.8% vs 16.5%, p<0.001). In Group 1, men more often had a history of myocardial infarction (21.2% vs 8.1%, p=0.002) and were less likely to have atrial fibrillation (18.5% vs 34.5%, p=0.002). In Group 2, men were significantly more likely to have a history of peripheral arterial disease when compared to women (5.5% vs 0.8%, p=0.04). There were no significant differences between gender in either group when comparing history of diabetes, hypertension or hyperlipidemia. The mean NIHSS on admission was higher for women in both groups (Group 1, 13.7 vs 12, p=0.10; Group 2, 8.0 vs 7.1, p=0.06). In Group 2, more women than men had a pre-stroke mRS>1 (28.9% vs 19.5%, p=0.01). There were no differences in pre-stroke mRS between gender in Group 1 (mRS 2-6 18.9% vs 16.6%, p=0.65). There were no significant differences in discharge destination between gender in Group 1 (p=0.17). In Group 2, however, there was an overall difference in discharge destination (p=0.02). Men were more likely to be discharged home or to acute rehabilitation facilities than to any other discharge destination compared to women (73% vs 62%). Out of the 294 patients included in Group 1, 209 had 90-day outcome data (102 women and 107 men (Table 3)). In Group 1, 35.5% of the men and 22.6% of the women had a good clinical outcome (90-day mRS 0-1). When controlling for observed confounders, namely admission NIHSS, history of coronary artery disease/myocardial infarction and history of atrial fibrillation, the rates of good clinical outcome were not statistically significant (OR=0.58; 95% CI=(0.28,1.18),p=0.13). Out of 554 patients included in Group 2, 194 had 90-day outcome data (91 women and 103 men). In this group, 57.3% of the men and 47% of the women had a good clinical outcome (90-day mRS 0-1). However, when controlling for observed confounding variables, namely admission NIHSS and history of CAD/MI, the comparison was not statistically significant (OR=0.83; 95% CI=(0.43,1.58),p=0.56) (Table 2). We found no significant difference between genders when assessing for treatment times in either group (Table 3).

Table 1.

Baseline Characteristics of Group 1 (with t-PA) and Group 2 (without t-PA)

	Women	Men	P value
Group 1	(N=148)*	(N=146)*
Mean Age-yr	72	68	0.008
Sex -no. (%)	50.3%	49.7%
Risk Factors (%)
Coronary Artery Disease	22(14.9%)	44(30.1%)	0.002
Myocardial Infarction	12(8.1%)	31(21.2%)	0.002
Atrial Fibrillation	51(34.5%)	27(18.5%)	0.002
Diabetes	27(18.2%)	37(25.3%)	0.159
Hypertension	104(70.3%)	96(65.8%)	0.454
Hyperlipidemia	42(28.4%)	48(32.9%)	0.448
Tobacco Use	24(16.2%)	49(33.6%)	0.001
Group 2	(N=242)	(N=312)
Mean Age-yr	72	68	0.001
Sex -no. (%)	44%	56%
Risk Factors (%)
Coronary Artery Disease	43(17.7%)	82(26.3%)	0.019
Myocardial Infarction	27(11.2%)	45(14.4%)	0.308
Atrial Fibrillation	50(20.7%)	59(18.9%)	0.667
Diabetes	55(22.7%)	60(19.2%)	0.342
Hypertension	155(64.1%)	199(63.8%)	>0.999
Hyperlipidemia	77(31.8%)	102(32.7%)	0.855
Tobacco Use	40(16.5%)	93(29.8%)	<0.001

^*The number of women and men vary because not all patients had complete data entered into the database.

Table 3.

Treatment Times

	Women	Men	P value
Group 1	(N=147)	(N=145)
Onset to Arrival (min.)	78±38	64±38	0.951
Onset to Decision	137±41	119±41	0.286
Onset to Bolus	157±51	139±51	0.259
Group 2	(N=231)	(N=303)
Onset to Arrival	348±609	372±609	0.379
Onset to Decision	415±609	440±609	0.539

Table 2.

Dichotomized Day 90 modified Rankin Score

	Women	Men
Group 1	(N=102)	(N=107)
0–1 (%)	23(22.6%)	38(35.5%)
2–6 (%)	79(77.5%)	69(64.5%)
Group 2	(N=91)	(N=103)
0–1 (%)	43(47.3%)	59(57.3%)
2–6 (%)	48(52.8%)	44(42.7%)

When controlling for observed confounding variables, the comparison between genders was not found to be statistically significant.

Discussion

We examined a prospectively collected case series of “Code Stroke” patients from six hospitals regarding gender differences in acute stroke treatment. In our analysis, men with acute stroke were younger, had lower admission NIHSS and were more likely to have a positive smoking history, coronary artery disease and myocardial infarction; they were less likely to have atrial fibrillation. In the group not treated with t-PA, men were more likely to suffer from peripheral arterial disease, had lower admission NIHSS and pre-stroke mRS scores. These gender differences in risk factors were also found by others.^{11, 12} Niewada et al showed, in the International Stroke Trial (IST) cohort, that men, on average, were five years younger than women, less likely to have atrial fibrillation and suffered less severe strokes.¹⁰

While others had found a significant delay in the evaluation and treatment of women with ischemic stroke, ^{8, 15} the data from our center did not show a significant gender difference in t-PA treatment times or time to presentation. Contrary to prior studies ^16,17 , we found similar t-PA rates between genders. In addition, we report a high rate of t-PA administration for both genders. Our findings may be influenced by the fact that all six hospitals were served by one established team of stroke practitioners who were on call 24/7 and were following standardized treatment algorithms. This may help support the need for comprehensive 24/7 stroke teams to ensure consistent stroke case management.

Comparing 90-day mRS, men had a higher proportion of better outcomes. However, after controlling for observed confounders, namely, admission NIHSS, history of coronary artery disease, myocardial infarction and atrial fibrillation, the comparison was not statistically significant. Similarly, Niewada et al, after adjusting for the higher rate of atrial fibrillation and higher NIHSS in women, did not find that gender was an independent predictor of 14-day outcome. ¹⁰ After adjustment for differences in baseline features, they found that women had a lower risk of death by 6 months, but were more likely to be dependent. It is unclear if these differences are due to hormonal changes, socio-economic status, differences in ischemic preconditioning or collateral circulation.

The limitations of our study are that not all Code Stroke patients received a 90-day evaluation and therefore bias may have been introduced into our outcome analysis. We do, however, show close to 100% follow-up on hospital discharge. The conclusions we make may not be applied to other acute stroke populations.

Conclusion

Using a dedicated stroke team and stroke treatment algorithms, no gender differences in treatment times and t-PA utilization were found.