Article Text
PDF
Abstract
Background Coil embolization of intracranial aneurysms is being increasingly used; however, thromboembolic events have become a major periprocedural complication.
Objective To determine the safety and efficacy of prophylactic tirofiban in patients with ruptured intracranial aneurysms.
Methods Tirofiban was administered as an intravenous bolus (8.0 μg/kg over 3 min) followed by a maintenance infusion (0.10 μg/kg/min) before stent deployment or after completion of single coiling. Dual oral antiplatelet therapy (loading doses) was overlapped with half the tirofiban dose 2 h before cessation of the tirofiban infusion. Cases of intracranial hemorrhage or thromboembolism were recorded.
Results Tirofiban was prophylactically used in 221 patients, including 175 (79.19%) who underwent stent-assisted coiling and 46 (20.81%) who underwent single coiling, all in the setting of aneurysmal subarachnoid hemorrhage. Six (2.71%) cases of intracranial hemorrhage occurred, including four (1.81%) tirofiban-related cases and two (0.90%) antiplatelet therapy-related cases. There were two (0.90%) cases of fatal hemorrhage, one related to tirofiban and the other related to dual antiplatelet therapy. Thromboembolic events occurred in seven (3.17%) patients (6 stent-assisted embolization, 1 single coiling), of which one (0.45%) event occurred during stenting and six (2.72%) occurred during intravenous tirofiban maintenance. No thromboembolic events related to dual antiplatelet therapy were found.
Conclusions Tirofiban bolus over 3 min followed by maintenance infusion appears to be a safe and efficient prophylactic protocol for the endovascular treatment of ruptured intracranial aneurysms and may be an alternative to intraoperative oral antiplatelet therapy, especially in the case of stent-assisted embolization.
- Aneurysm
- Hemorrhage
Statistics from Altmetric.com
Introduction
Coil embolization has been increasingly used to treat intracranial aneurysms since the publication of the results of the randomized International Subarachnoid Aneurysms Trial.1 However, thromboembolic events (TEs) have become one of the major periprocedural complications associated with the endovascular treatment of intracranial aneurysms, particularly ruptured aneurysms (11.01%),2 for the following reasons: the use of Guglielmi detachable coils,3 ,4 stent implantation in wide-necked aneurysms,5 cerebral vasospasm, and the activation of inflammation and coagulation cascades after subarachnoid hemorrhage (SAH).6–8 Owing to these complications, tirofiban, a non-peptide glycoprotein IIb/IIIa receptor antagonist (GpRA), is increasingly being used in the endovascular treatment of ruptured intracranial aneurysms, especially wide-necked aneurysms. The advantages of this drug are its short plasma half-life (approximately 2 h)9 and rapid recovery of platelet function (within 4 h after drug cessation).10 In addition, Giordano et al11 recently demonstrated that tirofiban could stimulate the migration and proliferation of endothelial cells.
Although tirofiban was proved to be effective in the Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE),12 few studies have assessed both the optimal dosage and the safety and efficacy of its use in ruptured aneurysms. Here we report a new protocol for its prophylactic intraoperative application during the endovascular treatment of ruptured intracranial aneurysms with no oral antiplatelet medications. We also evaluate the safety and efficacy of this new protocol. The present study is the first to assess tirofiban in a large sample of patients with ruptured intracranial aneurysms.
Methods
Patients
After obtaining approval from the institutional review board of our hospital, we retrospectively reviewed our hospital database for patients who had undergone endovascular treatment of intracranial aneurysms in our cerebrovascular interventional therapy center between January 2008 and January 2015. During this 7-year period, 1405 patients who presented with intracranial aneurysms were treated with coil embolization, including 903 (64.27%) patients who were treated in the setting of aneurysmal SAH. Of the 903 (64.27%) patients with ruptured aneurysms, 312 (34.55%) underwent stent-assisted coiling. In 225 patients tirofiban was prophylactically administered without any oral intraoperative medications or postoperative anticoagulation. Four of these patients had a pre-procedure Hunt–Hess grade of >IV and were excluded from the analysis, as it was quite difficult to determine whether they had intracranial hemorrhage (IH) or a TE. Thus, 221 consecutive patients (with 243 treated aneurysms) were included in this research.
Study design
All medical charts including operation record charts, electronic medical records, and imaging data (such as CT, CT angiography (CTA), MRI, MR angiography (MRA), and digital subtraction angiography (DSA) data) were reviewed, and 221 patients (243 treated aneurysms) were selected and their data recorded in detail in another chart. The detailed films were carefully analyzed by two experienced doctors (Z-lW and W-xB) to detect any IH or TE related to this protocol. IH was defined as the extravasation of contrast agents on intraoperative DSA performed after the intravenous bolus of tirofiban, with or without protrusion of the embolization materials outside the limit of the aneurysmal sac,13 or hemorrhage detected on postoperative CT of the head. TE was defined as (1) the presence of an intra-arterial filling defect or the occlusion of the proximal or distal vessel on DSA or (2) the development of TE-related clinical syndromes with or without ischemic infarction on postoperative CT/MRI (vasospasm was excluded). Retroperitoneal, gastrointestinal, or genitourinary bleeding was confirmed by reviewing the medical charts. Thrombocytopenia was defined as a platelet count of <100×109/L. All complications were recorded in detail on a separate sheet, and all the authors guarantee the authenticity of the data.
Periprocedural management
Before the procedure, CTA and/or three-dimensional DSA were performed to evaluate the shape, size, and location of the aneurysm and its anatomical relationship with nearby arteries. The operative risk was determined by a discussion involving all the doctors in our department and an optimal therapeutic strategy was then formulated for each patient. Informed consent was obtained prior to the procedure. All the procedures were performed under general anesthesia, and vital signs (including pulse rate, temperature, respiration rate, and blood pressure) were continuously monitored throughout the procedure. Systemic heparinization (50–70 U/kg) with unfractionated heparin as a bolus followed by intermittent intravenous injections of 1000 U/h was routinely achieved before catheterization to maintain the activated clotting time at 250–300 s. Heparinization was reversed after the procedure owing to concerns about increasing the risk of IH.14 ,15 In addition, in our center, various embolization techniques such as the ‘jailing’ technique16 and the ‘stent-jack’ technique17 are performed in different situations at the discretion of the operators. In this series, Enterprise stents (Codman Neurovascular, Chaska, Minnesota, USA) and Solitaire AB stents (ev3 Neurovascular, Irvine, California, USA) were mostly implanted to assist the embolization. Most aneurysms that required stent-assisted embolization (175 patients) were large or complex with wide necks and not amenable to single coiling or clipping.
During stent-assisted coiling, an 8.0 μg/kg bolus of tirofiban (Grandpharma Company, Wuhan, China) was intravenously administered over a period of 3 min immediately before stent deployment, followed by a maintenance dose of 0.10 μg/kg/min. In single coiling, this was done instantly after the protected embolization (ie, the finish of the satisfactory obliteration) was achieved. After embolization the blood flow status was closely observed to detect any thrombus formation or migration to the distal vessel. If a filling defect was detected, the intravenous bolus of tirofiban was immediately repeated to ensure antegrade flow with no stagnation. If the defect was resolved, the procedure was continued; if not or if the clot became bigger, mechanical means (such as mechanical clot disruption, mechanical thrombectomy, balloon angioplasty, or stent placement) were attempted. During this period, angiography was performed every 5–10 min (for no more than 30 min in total) to detect antegrade flow. If IH occurred during the procedure, embolization was continued without delay until cessation of the extravasation of the contrast agent; additionally, protamine (5 mg/1000 U heparin) was injected at the same time. Intraoperative head CT was performed to check for new IH in every patient in this series. All patients were managed in the neurosurgical intensive care unit in our center for at least 24 h after the procedure. Head CT scans were routinely performed on days 1 and 7 after the operation (or on the day before discharge).
An infusion of 0.10 μg/kg/min tirofiban was maintained for 24 h after the start of the bolus injection. At 2 h before the cessation of tirofiban, dual antiplatelet drug therapy with clopidogrel (loading dose 300 mg) and aspirin (loading dose 300 mg) was overlapped with half the dose of the intravenous infusion, followed by a conventional dosage of 75 mg clopidogrel and 100 mg aspirin daily for 3–6 months. After this period, only one of these drugs (usually aspirin) was continued indefinitely. In patients who underwent single coiling, the same overlapping method was used, but dual oral antiplatelet therapy was usually administered at the conventional dosage for 2–4 weeks and then one of the two drugs was continued for 1–3 months.
Results
This study included a total of 221 patients with 274 intracranial aneurysms (243 treated aneurysms) and SAH; of these patients, 175 (79.19%) were treated with stent-assisted coiling and 46 (20.81%) were treated with single coiling. One-third of all treated aneurysms (33.74%) were located in the terminal segment of the internal carotid artery (table 1), and each of the 221 subjects had at least one irregular ruptured aneurysm. In addition to the aneurysms, the patients had other diseases including hypertension (n=107, 48.42%), diabetes (n=14, 6.33%), and hyperlipidemia (n=5, 2.26%). Our patient cohort consisted of 150 women (67.87%) and 71 men (32.13%) with a mean age of 56±11 years (range 16–81 years). The clinical status of the patients was classified using the Hunt–Hess scale before or on the day of the procedure. The results were as follows: 26 patients (11.77%), grade I; 98 patients (44.34%), grade II; 74 patients (33.48%), grade III; and 23 patients (10.41%), grade IV.
Locations and rates of treated aneurysms in patients with subarachnoid hemorrhage
We detected six (2.71%) cases of IH related to the protocol in this series. Of these, four (1.81%) were related to tirofiban use, including three (1.36%) cases of intraoperative IH and one (0.45%) case of postoperative IH (table 2). The three cases of tirofiban-related IHs (all are SAHs) that occurred during the procedure were also attributable to the protrusion of the embolization materials outside the limit of the aneurysmal sac. Additionally, two cases of IH (0.90%) were secondary to the use of dual antiplatelet therapy in this protocol, one of which resulted in a fatal outcome and the other consisted of new SAH beside the cerebral falx and intraparenchymal multiple microhemorrhages on postoperative CT without any clinical symptoms (patient 6, figure 1). The intraventricular hemorrhages were fatal in two patients, one of whom (patient 4, figure 2) had a tirofiban-related hemorrhage; in the other patient the IH was related to dual antiplatelet therapy. The hemorrhage in the latter patient was also attributable to incomplete embolization caused by an intraprocedural vasospasm. Decompressive hemicraniectomy was performed in one of these patients (patient 4); the family of the other patient refused the therapy. No cases of thrombocytopenia or retroperitoneal, gastrointestinal, or genitourinary bleeding occurred in this series.
Basic characteristics of patients with intracranial hemorrhage
(A) Baseline pretreatment CT showing aneurysmal subarachnoid hemorrhage (SAH). (B) Digital subtraction angiography (DSA) showing an irregular ruptured aneurysm located at the origin of the anterior choroidal artery. (C) Three-dimensional imaging immediately after the operation showing the result of embolization. (D) Head CT on day 7 showing new SAH beside the cerebral falx but (E) no new hemorrhage at the aneurysm treated by stent-assisted coiling. (F) CT on day 14 showing intraparenchymal multiple microhemorrhages due to the dual antiplatelet therapy on the right basal ganglia region and the right lateral ventricle angle.
(A, B) Baseline pretreatment CT angiography showing aneurysmal subarachnoid hemorrhage and digital subtraction angiography showing two aneurysms (D), a ruptured one involving the C7 segment of the left internal carotid artery and an unruptured one located in a branch of the left middle cerebral artery. (E) The ruptured aneurysm was treated but not the unruptured aneurysm. (F) Three-dimensional imaging after the operation showing the result of embolization. (C) Postoperative CT shows that the treated aneurysm has ruptured again on the day after the operation.
There were seven (3.17%) TEs related to this protocol, including six (85.71%) postprocedural TEs and one (14.29%) intraprocedural TE (table 3). None of the TEs were related to the dual antiplatelet therapy. Table 3 shows the characteristics of the patients with TEs and their outcomes at discharge as well as at the 3-month follow-up.
Basic characteristics of patients with thromboembolic events
All cases of bleeding and thromboembolic complications occurred in patients who underwent stent-assisted coiling, except for one TE which occurred in a patient treated with single coiling. In summary, the rates of IH and TE in the stent-assisted group versus the single coiling group were 3.43% vs 0% and 3.43% vs 2.17%, respectively. The total complication rate (both IHs and TEs) was 5.88%.
Patient 7 had an in-stent thrombus which was eventually partially recanalized; this patient was lost to follow-up at 3 months (–); all references to sex and specific age of this patient were removed to anonymize this patient's data (–).
Discussion
The new protocol
Thromboembolic complications are a major concern during and after the endovascular treatment of ruptured aneurysms for various reasons.3 ,4 ,6–8 In one study, as many as 64% of non-ruptured aneurysms treated with stent-assisted embolization were found to be associated with silent TEs on highly sensitive imaging studies such as diffusion-weighted MRI, despite the administration of antiplatelet drugs for 7 days before the operation combined with the use of intraoperative anticoagulation.19 To decrease the morbidity and mortality associated with the procedure, tirofiban, a short-acting antiplatelet drug, is being increasingly used. In the RESTORE trial, tirofiban was administered as a bolus of 10 μg/kg over a 3 min period followed by a 36 h infusion at 0.15 μg/kg/min.12 To decrease the risk of IH in patients with ruptured aneurysms, our cerebrovascular center uses a lower dosage (an intravenous bolus of 8.0 μg/kg over a 3 min period followed by a 24 h infusion at 0.10 μg/kg/min).
Studies have shown that a relatively high level of platelet inhibition can be achieved approximately 5–10 min after the administration of tirofiban,20 ,21 and the inhibition of platelet function by ≥95% at 10 min after the start of therapy is associated with a significant decrease in the incidence of major adverse cardiac events.20 Owing to the pharmacokinetic characteristics of clopidogrel, the rate of ADP-induced platelet aggregation can be inhibited by as much as 38–54% in 2–4 h after a loading dose of 300 mg.22 ,23 Thus, to maintain platelet inhibition at a responding level (>30% inhibition of ADP-induced platelet aggregation),24 we used a loading dose of 300 mg clopidogrel+300 mg aspirin (both administered orally). The drugs were overlapped with the last 2 h of the tirofiban infusion. Amenta et al5 and Chalouhi et al25 routinely administered 600 mg clopidogrel with or without aspirin to patients with ruptured aneurysms during surgery; however, in our series, no oral antiplatelet drugs were administered pre- or intraoperatively.
Complications
Hemorrhagic events
In this series there were six (2.71%) cases of IH, including four (1.81%) that were associated with tirofiban and two that were associated with the dual antiplatelet therapy. However, the three (75%) cases of intraoperative tirofiban-related IH (patients 1–3) were also attributable to the embolization procedure, which might have been the main cause for the extravasation of contrast agents in these cases. In the only case of postoperative tirofiban-related IH, there was no possibility of the hemorrhage being related to the procedure as satisfactory embolization had been achieved in this patient (figure 2).
Both our study and that by Chalouhi et al25 used tirofiban protocols consisting of an intravenous bolus followed by intravenous maintenance. The difference between the two protocols lies in the bolus dose and the period over which the bolus was administered: 8.0 μg/kg over 3 min in this study vs 0.4 μg/kg/min over 30 min in the other study.25 The rates of tirofiban-related IH and fatal hemorrhage (FH) were much lower in our series than in the study by Chalouhi et al25 (IH: 1.81% vs 18.8%; FH: 0.45% vs 12.5%). This difference may be attributable to the different protocols used, the different total dosages (8.0 μg/kg vs 12.0 μg/kg), the period of bolus injection, as well as the additional use of antiplatelet drugs (clopidogrel 600 mg+aspirin 325 mg) during the procedure in the study by Chalouhi et al.25 There have been some concerns about the high risk of IH with the use of abciximab in the setting of neurointerventional procedures, and a multicenter study has found that, after the application of abciximab via various routes in 51 patients, the rates of IH and FH were 17.65% and 7.84%, respectively.26 Moreover, in a similar study including 74 patients with intracranial aneurysms, eptifibatide, another short-acting GpRA, was prophylactically used in 77 procedures at a bolus dose of only 180 μg/kg.9 However, the incidence rates of IH and FH exacerbated by eptifibatide were 6.25% and 3.13%, respectively, among patients with ruptured aneurysms.9 Among the three GpRAs mentioned above, tirofiban seems to be the safest alternative, especially when used with the protocol described in our study. Amenta et al5 routinely used a loading dose of 600 mg clopidogrel without GpRAs in patients with ruptured wide-necked intracranial aneurysms and reported IH and FH incidence rates of 7.7% and 4.6%, respectively. These rates secondary to antiplatelet drugs are much higher than those related to our protocol (2.71% and 0.90%, respectively). Considering all the above findings, we conclude that our protocol without any intraoperative oral antiplatelet therapy is safe for the treatment of ruptured intracranial aneurysms and does not increase and might even decrease the rates of IH and FH.
Thromboembolic events
A total of seven (3.17%) TEs occurred in this series. Ries et al27 demonstrated a significant decrease in TEs in patients with ruptured aneurysms who received 250 mg intravenous acetylsalicylic acid (ASA) (+ASA group: heparin+ASA) during or immediately after the procedure in comparison with the –ASA group (heparin alone) (10.09% vs 20.00%, p=0.047). Compared with the +ASA group in the above study, our series was associated with a low incidence of TEs (heparin+tirofiban, 7/221, 3.17%; heparin+ASA, 11/109, 10.09%). Moreover, the incidence in the stent-assisted group in our series (heparin+tirofiban, 6/175, 3.43%) was also lower than that reported by Amenta et al5 (heparin+600 mg clopidogrel, 5/65, 7.69%). However, in comparison with the results obtained with the revised protocol reported by Chalouhi et al,25 the rate of TEs in the stent-assisted group in our series is a little higher (6/175, 3.43% in our series vs 0%). This difference may be attributable to the fact that no oral antiplatelet drugs were used intraoperatively and heparin was not administered postoperatively in our series. The finding that most TEs (6/7, 85.71%) in our study occurred postoperatively supports the above conjecture. In summary, the protocol in this study appears to be more efficient than routine loading doses of oral antiplatelet drugs used intraoperatively.
When to use tirofiban prophylactically
In this series, tirofiban was prophylactically used in three types of patients: (1) those with ruptured wide-necked aneurysms treated with stent-assisted coiling; (2) those who required only single coiling but in whom the coil protruded into the parent artery necessitating rescue stent implantation; and (3) those who were treated with single coiling and in whom there was a delay in the establishment of antegrade flow, but no filling defects were found in the vessel during the procedure, or those who had a relatively large coil/parent artery interface but did not require stenting.
Problems and expectations
To prevent complications and reduce perioperative morbidity and mortality, we should improve operator skill, shorten operation time, and use sufficient intraoperative anticoagulation as well as antiplatelet therapy. In addition, patients should be closely monitored for antegrade blood flow intraoperatively and clinical manifestations postoperatively.
To the best of our knowledge, this protocol is the first to be reported for use in intracranial procedures, including procedures for the treatment of ruptured aneurysms. Moreover, this is the first study about tirofiban with a rather large number of subjects (221), all of whom had SAH.
However, the standard strategy for the use of tirofiban in ruptured intracranial aneurysms to prevent TEs has not yet been established, and the following questions need to be answered: (1) What is the optimal dosage of both the bolus and maintenance infusion? (2) How long should the period of bolus and maintenance therapy last? (3) What is the relationship between the inhibition of platelet aggregation and the incidence rates of IH and TEs.
There are some limitations to this retrospective study, such as the lack of randomization, a control group, and double-blinding. The subjects included in this series are all from a single center, and the procedures were performed by several experienced doctors. Because the inhibition of platelet aggregation after drug administration was not assessed, the relationship between platelet function inhibition and the incidence of IHs and TEs could not be clarified. Thus, prospective randomized controlled trials with different dosages and durations of drug administration as well as platelet function inhibition rates would be of great interest.
Conclusions
A bolus of tirofiban over 3 min followed by a maintenance infusion for 24 h applied prophylactically with no intraoperative oral antiplatelet drugs appears to be safe and efficient for the endovascular treatment of ruptured aneurysms and may be an alternative to the routine loading doses of oral medications used intraoperatively, especially in the case of stent-assisted embolization.
References
Footnotes
Contributors X-dL, Z-lW, T-xL, W-xB and Y-yW: conception and design. Y-kH and W-xB: literature research. X-dL, G-yZ, Y-yW, Z-lW, T-xL and Y-kH: statistical analysis and drafting the manuscript. All authors were involved in acquisition of data, analysis and interpretation of data, critical revision of the manuscript and approval of the final manuscript.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Ethics approval was obtained from the institutional review board of our hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Unpublished data may be available to academic researchers on a per request basis to the corresponding author.