Thrombolysis for Acute Ischemic Stroke in Patients With Cancer
A Population Study
Background and Purpose—The safety of thrombolysis for acute stroke in patients with cancer is not well established. Our aim is to study the outcomes after thrombolysis in patients with stroke with cancer.
Methods—Patients with acute ischemic stroke who received thrombolysis were identified from the 2009 and 2010 Nationwide Inpatient Sample. Patients with cancer-associated strokes and noncancer strokes were compared based on demographics, comorbidities, and outcomes.
Results—Of the 32 576 strokes treated with thrombolysis, cancer-associated strokes had significantly higher comorbidity indices overall, but fewer vascular risk factors than noncancer strokes. There was no difference in the rates of home discharge and in-hospital mortality, after adjusting for confounders. Subgroup analysis showed that compared with liquid cancers, patients with solid tumors had worse home discharge (odds ratio, 0.178; 95% confidence interval, 0.109–0.290; P<0.001) and higher in-hospital mortality (odds ratio, 3.018; 95% confidence interval, 1.37–6.646; P=0.006) after thrombolysis. Metastatic cancers had poorest outcomes, but intracerebral hemorrhage rates were similar.
Conclusions—Thrombolytic therapy for acute stroke in patients with cancer is not associated with increased risk of intracerebral hemorrhage or in-hospital mortality. However, careful consideration of the cancer subtype may help delineate the subset of patients with poor response to thrombolysis. Prospective confirmation is warranted.
The prevalence of cerebrovascular disease in the setting of malignancy is ≈15%.1 The cause is 2-fold: malignancy-related factors such as hypercoagulability, tumor embolism, and traditional stroke risk factors, such as hypertension, diabetes mellitus, hyperlipidemia, atrial fibrillation, and smoking.2 Cancer treatments, such as radiotherapy3 and chemotherapy,4 also increase stroke risk. Most of the acute ischemic stroke (AIS) trials excluded patients with malignancy, and hence, treatment of AIS in this subset of the population remains controversial.5 To date, there are no large studies published assessing the efficacy of thrombolysis in cancer-associated strokes (CSs). The aim of our study is to evaluate outcomes of thrombolytic therapy in CSs.
Patients admitted with a diagnosis of AIS were selected from the Nationwide Inpatient Sample database for the years 2009 and 2010,6 using the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) codes 433.xx, 434.xx, and 436. The procedure code 99.10 identified patients who received intravenous thrombolysis. Using cancer ICD-9 codes, we divided patients into 3 categories based on the National Comprehensive Cancer Network classification: liquid tumors without metastases (hematologic malignancies), solid tumors without metastases (epithelial tumors gastric, colon, prostate, etc, and sarcomas), and metastatic cancer.7 A weighted score was calculated for comorbidities using the Elixhauser Comorbidity Index. Endovascular therapy for AIS was identified using the procedure code 39.74. The primary outcomes were home discharge and in-hospital mortality. Secondary outcomes were intracerebral hemorrhage (ICH; symptomatic and asymptomatic), pneumonia, venous thromboembolism, gastrostomy, and tracheostomy.
Pearson χ2 and Wilcoxon–Mann–Whitney tests were used for categorical and continuous variables, respectively. Logistic regression models were constructed to assess stroke outcomes after thrombolysis, after adjusting for confounders. Statistical significance was defined as P<0.05. Analyses were performed using SPSS (version 21.0; Chicago, IL). Standard weights provided by Healthcare Cost and Utilization Project were applied.
There were 1 135 574 AIS in 2009 and 2010. Thrombolysis was administered to 807 patients of 50 437 CSs (1.6%), and 31 769 patients of 1 085 137 noncancer strokes (NCSs; 2.9%). Among thrombolysis-treated CSs, 614 (76.1%) patients received intravenous thrombolysis, and 193 (23.9%) received endovascular treatment. Similarly, 25 196 (79.3%) patients received intravenous thrombolysis, and 6573 (20.7%) were treated endovascularly among NCSs. The prevalence of vascular risk factors was lower in CSs, but these patients had higher comorbidities (Table 1). In the regression model, there was no difference in primary outcomes between CSs and NCSs for overall thrombolysis (Table 2). In the subgroup analyses, mortality was not significant in intravenous thrombolysis only (10.9% CSs versus 8.5% NCSs; P=0.313) and endovascular groups (16.1% CSs versus 16.6% NCSs; P=0.404). ICH was also similar in intravenous thrombolysis only (4.1% CSs versus 5.5% NCSs; P=0.476), and endovascular groups (13.5% CSs versus 9.9% NCSs; P=0.276). Subgroup analysis of cancer subtype showed that solid and metastatic tumors were associated with lower home discharge rate and higher in-hospital mortality compared with liquid cancers. Cancer subtype did not influence ICH (Table 3).
There is paucity of data on the safety of thrombolysis for AIS in cancer. We found no significant difference in primary outcomes or ICH between CSs and NCSs after thrombolysis, irrespective of type of thrombolysis or cancer subtype. A small single center study by Masrur et al,8 which compared 44 CSs with 264 NCSs, also obtained similar results. However, outcomes based on cancer subtype were not studied. It is intriguing that solid tumors are associated with higher mortality given the lack of difference in ICH and comorbid conditions. This may be attributed to factors not available in the Nationwide Inpatient Sample database, such as admission stroke severity and cancer staging. Metastasis is the strongest predictor of mortality in patients with evidence of thrombosis9 and accounts for the highest mortality in CSs. The higher prevalence of hypertension and hyperlipidemia in NCSs suggests that nonvascular risk factors are implicated in CSs.
Our study has a few limitations. First, our study included only patients with stroke treated with thrombolysis. The selection bias introduced with the inclusion of only CSs treated with thrombolysis precludes generalization of these results to a wider cancer population with stroke. The Nationwide Inpatient Sample database lacks information on admission stroke severity and stroke subtype. There is no information on modified Rankin Scale at 90 days. Coding error is also a possibility. There were no data on cancer staging and outpatient chemoradiation therapies, which may have influenced outcomes.
Our study suggests that thrombolysis for acute stroke is safe in patients with cancer. ICH rates are not higher in thrombolysis-treated CSs. However, there is heterogeneity among cancer subtypes regarding response to thrombolysis. Prospective studies are warranted to validate these results.
- Received July 31, 2013.
- Revision received August 14, 2013.
- Accepted August 15, 2013.
- © 2013 American Heart Association, Inc.
- 6.↵Healthcare Cost and Utilization Project (HCUP). Overview of the Nationwide Inpatient Sample (NIS). Agency for Research Healthcare and Quality Web site. http://www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed May 15, 2013.
- 7.↵National Comprehensive Cancer Network (NCCN). The difference between solid and liquid tumors. NCCN Web site. http://www.Nccn.Com/component/content/article/54-cancer-basics/1042-liquid-versus-solid-tumors.Html. Accessed May 21, 2013.