Thrombolysis in Postoperative Stroke

Nicolas Voelkel, MD; Nikolai Dominik Hubert, MSc; Roland Backhaus, MD; Roman Ludwig Haberl, MD; Gordian Jan Hubert, MD


Stroke. 2017;48(11):3034-3039. 

In This Article


In our study, stroke patients with IVT after surgery had a 6.7% risk of SSH. Serious hemorrhage only occurred in 3.0% of patients. The study group represents patients who were selected for IVT treatment on the basis of clinical judgment and risk–benefit assessment.

Previous case series have shown variable complication rates. A study on 8 patients reported no hemorrhages.[3] In another study on 13 patients who had surgery or trauma, 2 (15%) developed systemic hemorrhages after IVT.[4] And in 13 patients after cardiac surgery, 3 patients developed hemothorax or a drop in hemoglobin level after receiving intra-arterial tPA.[5] These differences may be because of not only low number of patients studied but also the variable bleeding risk of each type of surgery and of each individual intraoperative situation.

In our study, non-ICHs were found in 11.2%. Published data have found 6.2%[11] and 9.6%[12] in off-label IVT patients and up to 23% in the NINDS trial (National Institute of Neurological Disorders and Stroke).[13] So, despite preceding surgery, overall non-ICHs did not differ largely from previous nonsurgical patients.

The rate of ICH was 9.7% and similar to previous publications on patients with IVT.[13–15] No hemorrhage was symptomatic according to Safe Implementation of Thrombolysis criteria.[16] One of the 7 other bleeding complications (not in surgical area and not intracranial) was fatal because of a massive gastric bleed. In this case, the operation (clipping of cerebral artery aneurysm) was 73 days ago, so it is unlikely that surgery had influenced this complication. Mortality in our study population was similar to overall mortality of patients with IVT in Bavaria, Germany (8.4%).[17]

Among patients who did not experience SSH after IVT was one with carotid endarterectomy 18 days before, 2 patients with pacemaker implantations 10 and 12 days before, 1 with colonoscopy and clipping of polyps on day 1, and 1 with a biopsy of a pancreatic tumor 13 days before IVT. The study population included several patients with various abscess openings 1 to 8 days before IVT and patients with appendectomy and inguinal hernia closure 16 to 22 days before IVT. There were 2 patients with decompression of Carpal tunnel syndrome 5 and 6 days before IVT. And there were 4 patients with intramuscular injections 1 to 4 days before IVT, 7 patients receiving knee arthroscopy 5 to 25 days before IVT, and 3 patients receiving a tooth extraction 3 to 6 days before IVT. All of these patients did not experience SSH.

Risk of SSH was significantly increased when surgery had been performed recently (within 10 days before IVT). This result possibly reflects the course of surgical wound healing and general health improvement after surgery. No significant difference was found between patients with major and minor operations, with a tendency toward higher bleeding risk after major surgery. Results may be limited by the difficulties classifying major and minor surgery and by not taking into account the individual operative situation. For this study, we chose a classification that reflects postoperative bleeding risk because our primary end point was SSH. This approach does not necessarily reflect potential damage and loss of function resulting from a hemorrhage, for example, although cataract operation may be considered minor, hemorrhage in this surgical area may result in severe handicap (permanent loss of vision), whereas superficial bleeding in major orthopedic surgery may not have a long-term negative effect. Further research is warranted with functional outcome over time as primary end point and classification based on the risk of decreasing functional outcome. The elevated bleeding risk in patients with both recent and major surgery found in our population should be studied further. A previous study that had administered tPA to patients early after major cardiac surgery found the bleeding rate was also high (23%).[5]

Mechanical thrombectomy as a treatment option for patients with large-vessel occlusion is likely to be of even lower risk for SSH if systemic thrombolytic agents are not administered. However, data on mechanical thrombectomy in postoperative stroke patients are rare.[18]

Our study has several limitations. First, it was a retrospective design with no direct control group; therefore, we do not know the bleeding risk in patients with postoperative stroke who did not receive IVT. Moreover, patients included in the study were recommended IVT after a clinical risk–benefit assessment. This study group is therefore highly selected and not representative for all patients with postoperative stroke, as we do not know the risk of SSH after IVT in patients where clinical risk–benefit assessment does not lead to thrombolysis treatment. Second, because this is a rare condition and the total rate of bleeding complications are low, the number of our study patients and low numbers of primary end points limit the statistical analysis and the interpretation of these results. Rate of major versus minor surgery was unequally distributed in recent and nonrecent subgroups, reflecting what was assumed to be a less thorough documentation of minor nonrecent surgeries in our medical records. However, adjusting for this parameter in our logistic regression model did not change significance or direction of results. Third, we did not have a long-term clinical outcome, so we are unable to determine, whether thrombolysis was beneficial in these patients, whether bleeding complications had a long-term negative effect on outcome, or whether the positive effect of IVT outweighed the complications. Bleeding in surgical area may require invasive interventions and transfusions, but may not necessarily influence overall clinical outcome. Fourth, all data come from the registry of a TeleStroke Network. This structure may have influenced the decision on whether to treat a patient with postoperative stroke and thus may have influenced the composition of the study population. Finally, because the variability of bleeding risk in different types of operations is high, generalizability to all patients with postoperative stroke is limited.

To conclude, we found an overall low risk of SSH in selected patients who received thrombolysis for postoperative stroke. SSH was higher when surgery was recent and major. In these patients, individual risk assessment is warranted before application of recombinant tPA.