Social History Matters–The Impact of Illicit Drug Use on tPA Use and In-Hospital Mortality in Acute Ischemic Stroke

Aims The objective of this descriptive study was to compare time to medical evaluation, intravenous tissue plasminogen activator (IV tPA) use, and short-term outcomes in illicit drug users compared to non-users presenting with acute ischemic stroke (AIS). Study Design This is a retrospective study performed from our stroke registry using deidentified patient information. Place and Duration of Study Tulane Medical Center Primary Stroke Center (PSC). Consecutive AIS patients presenting to our PSC from July 2008 to December of 2010 were identified from our prospectively collected stroke registry. Methodology Patients were categorized as toxicology positive (TP) or toxicology negative (TN). We compared baseline characteristics, clinical presentation, tPA use, and short-term outcomes in TP and TN patients. Results Two hundred and sixty-three patients met inclusion criteria (median age 63, 35.4% female, 66.5% Black). Nearly 40% of toxicology screens were positive. Stroke severity was similar with the median National Institute of Health Stroke Scale (NIHSS) of 6 in both groups; however, a higher proportion of TN patients were treated with IV tPA (32.1% vs. 21.2%). After adjustment for time from last seen normal to emergency department arrival (LSN-to-ED arrival), the odds of being treated with tPA for TP patients were similar to TN patients (OR 0.69, 95% CI 0.36–1.31, p=0.255). After adjustment for age, NIHSS, glucose, and tPA, the odds of in-hospital mortality in TP patients was 3 times that of TN patients (OR 3.17, 95% CI 1.07–9.43, p=0.038). Conclusion We found that the disparities observed in tPA use were attenuated after adjustment for time from LSN-to-ED arrival, suggesting an area for future intervention. Additionally, we found that TP patients may be at higher risk for in-hospital mortality. Further study on the role of substance abuse in time to ED arrival, tPA use, and outcome in AIS patients is warranted.


INTRODUCTION
The 2011 National Survey on Drug Use and Health found that 6.3% of US adults, age 26 or older currently use illicit drugs [1]. Traditionally, screening for illicit drugs has been performed in younger stroke patients, as drug abuse may be the most common predisposing condition for stroke among patients under 35 years of age [2,3]. National survey data suggest that rates of illicit drug use among adults ages 50 to 59 have been increasing since 2002 [1]. This increase has been attributed to the aging the baby boom cohort in which increased drug use during their youth may be being continued into older age [1].
Little is known about the relationship between illicit drug use and time to emergency department arrival in the setting of acute ischemic stroke (AIS). Further, no study has investigated intravenous (IV) recombinant tissue plasminogen activator (tPA) use in illicit drug users compared to non-users. The objective of this descriptive study was to compare time to medical evaluation, tPA use, and short-term outcomes in illicit drug users compared to non-users presenting with AIS.

Methods
Siemens Dimension Vista system with Flex reagent cartridges were used for urine drug screen, patients with the presence of one or more of the following illicit substances in their urine were categorized as toxicology positive (TP) the metabolites tested on urine toxicology screen are shown [as follows] when applicable: amphetamine [d-amphetamine, lamphetamine, MDA, chloroamphetamine], barbiturates, benzodiazepines, cocaine [benzoylecgonine], methamphetamine, methadone [l-methadone, d-methadone], opiates, phencyclidine (PCP), or tetrahydrocannabinol (THC) [11-nor-9 carboxy-∆9-THC, cannabinol]. The sensitivity and specificity for each compound screened are shown in the supplementary table found in the Appendix. The remaining patients were classified as toxicology negative (TN). Patients who did not have urine toxicology performed were excluded.
We compared baseline characteristics, time from last seen normal (LSN) to emergency department (ED) arrival, stroke severity (as measured by the National Institutes of Health Stroke Scale [NIHSS] score), treatment with intravenous (IV) recombinant tissue plasminogen activator (tPA), and short-term outcomes in TP and TN patients. Short-term neurologic deficits were estimated using the discharge NIHSS. Short-term functional outcomes were assessed using the modified Rankin scale (mRS) score. All NIHSS scores and mRS scores were performed by NIHSS and mRS certified physicians. The proportion of known and unknown LSN times and mean time from LSN to ED arrival were compared by illicit substance.
Categorical data were compared using Pearson Chi-squared (or Fisher exact test where appropriate). Continuous data were compared using the Student's t-test (or Wilcoxon Rank Sum test where appropriate). Logistic regression was used to determine the odds of receiving IV tPA and the odds of in-hospital mortality. Crude and adjusted models were performed. All tests were performed at the α=0.05 level and were two-sided. We did not correct for multiple comparisons, as this was an exploratory study [4]. This cross-sectional study was approved by the institutional review board at the Tulane University.

Statistical Methods
Categorical data are presented as frequencies and were compared using Pearson Chisquared or Fisher exact test where appropriate. Continuous data are presented as medians with ranges and were compared using Wilcoxon Rank Sum test. All tests were performed at the α=0.05 level and were two-sided. The retrospective chart review was approved by the institutional review board at the Tulane University (IRB protocol number 237137-3).

RESULTS
Five hundred and ninety-three patients were screened. Two hundred and sixty-three met inclusion criteria (median age 63, 35.4% female, 66.5% Black). A higher proportion of patients self-reporting a history of substance had a urine drug screen performed (25.3% vs. 8.7% p<0.001). Nearly 40% of toxicology screens performed were positive. Table 1 compares and contrasts TP and TN AIS patients. Table 2 shows the percent of patients with each positive result. There were no significant differences in the age, sex, race, or cardiovascular comorbidities in TP and TN patients (Table 1). A higher proportion of TP patients reported a history of substance abuse (p<0.001) and described themselves as current smokers (p=0.005). Stroke severity was similar with the median NIHSS 6 in both groups; however, a higher proportion of TN patients were treated with IV tPA (32.1% vs. 21.2%, p=0.053). A larger proportion of TN strokes were cardioembolic (26.6% vs. 18.3%, Table 1).

DISCUSSION
Our study found that only 44% of AIS patients routinely received a urine drug screen, similar to a recent report where the proportion ranged from 37-43% [5]. This may be due to the traditional neurologic teaching to consider illicit substance use in cases of stroke in the young. With stroke incidence increasing with age and the assumption being that illicit drug use would decline in older age cohorts, one may conclude that checking for illicit substance abuse would be of lesser importance in the evaluation of older AIS patients. However, recent data suggest that substance abuse rates are rising in older adults, potentially reflecting the aging of the baby boom cohort in which drug use is continuing into older age [1]. This serves as a reminder to neurologists that they may need to reconsider their criteria for performing a urine drug screen [5].
Similar to a recent report where tPA was administered to none of the cocaine positive ischemic stroke patients compared to 11% of cocaine negative patients, our study found that a higher proportion of TP patients were not treated with tPA [5]. Contraindications to tPA administration include SBP > 185 or DBP > 110 mmHg, CT findings suggestive of Intracerebral hemorrhage (ICH) or subarachnoid hemorrhage (SAH), suspicion of SAH, seizure at onset of stroke, recent intracranial or spinal surgery, recent head trauma, stroke within 3 months, major surgery or trauma within previous three months, recent internal bleeding (less than 22 days), platelets < 100,000, heparin use within 48 hours with PTT > 40, INR > 1.7, known bleeding diathesis or major disorder associated with risk of bleeding, or history of intracranial hemorrhage, brain aneurysm, vascular malformation, or brain tumor [6]. Potential reasons for non-treatment may include meeting blood pressure exclusion criteria if on stimulants, or having altered level of consciousness which confounds the initial time to stroke diagnosis. However, this association was no longer present after adjustment for time from LSN to ED, suggesting that time to ED presentation may be confounding the association between illicit drug use and tPA administration. The reasons for delay in seeking care in substance abusers are likely multifactorial and a function of both individual and system barriers. Barriers to care for substance abusers have been well-described and grouped into economic limitations, geographic limitations, lack of integrated services, cultural differences (e.g., language barriers, cultural practices and beliefs), patient physician communication, stigmatization, and lack of trust, respect, and confidentiality [7,8].
Additionally, we found that the proportion of TP patients who experienced in-hospital mortality was higher than that of TN patients. After adjustment for age, NIHSS, glucose on admission, and tPA use, we found that the odds of in-hospital mortality in TP patients in our sample were 3 times that of TN patients. While this is in keeping with previous reports of higher mortality in drug users, it is also possible that the TP and TN groups were inherently different in measures not collected or not assessed by our study.
Our results should be interpreted with caution. Given that we acquired information on exposures and outcomes at the same time, we were unable to make causal inferences. Further, our relatively small sample size may not have allowed us to detect existing differences in groups. While our sample is representative of the population residing in the catchment area of our medical center, these findings may not be generalizable to the US population as a whole.

CONCLUSION
Despite several limitations, our study calls attention to the relevance of substance abuse in AIS patients of all ages. Combined with national survey data and a previous report, our study suggests that performing urine drug screening may be appropriate for stroke patients of all ages [5]. Additionally, it highlights disparities in TP and TN AIS treatment rates, offering one possible explanation for the problem-time to ED arrival, which may be drug-specific. Finally, our study suggests that TP patients may be at higher risk for in-hospital mortality.
Additional study on the role of substance abuse in arrival times, tPA use, and outcome in AIS patients is warranted.

CONSENT
The Tulane Institutional Review Board granted a waiver of informed consent.

ETHICAL APPROVAL
The retrospective chart review was approved by the institutional review board at the Tulane University (IRB protocol number 237137-3).