Safety of High-dose Tirofiban During Coronary Angioplasty

Background The acute occlusion due to intrastent thrombosis represented a major event causing acute myocardial infarction, cardiac death and necessity for a new procedure or coronary by-pass intervention.

To avoid such complications different high anticoagulation regimens (heparin, antiplatelets agents and warfarin) were employed. Even if effective, such treatments were accompanied by a higher cost of side effects due to major bleeding (hemorrhagic stroke, retroperitoneal bleeding, gastrointestinal bleeding, need for transfusion, site access complications).

Pre-treatment with aspirin and ticlopidine was found to be very effective reducing intrastent acute thrombosis. Since several days of pre-treatment were required it became a limiting factor for interventional procedures forcing physicians to adopt a two-stage strategy that separated the diagnostic from interventional time increasing the hospital's costs and the patient's risks.

Beside that, high doses of heparin were still required during the procedures. The introduction of GP IIb/IIIa inhibitors was initially reserved to acute coronary syndrome, later it was shown how they reduce the composite incidence of death, myocardial infarction and the need for target vessel revascularization after percutaneous coronary intervention.

As major benefits arising from the use of those drugs, interventional procedures could be carried out at end of diagnostic procedure and major bleedings were reduced as require a lower heparin regimen.

Nowadays the most recent guidelines for the management of patients with acute coronary syndrome strongly recommend the use of IIb/IIIa inhibitors when percutaneous coronary intervention is performed. Although several randomized trials with different IIb/IIIa inhibitors have demonstrated the usefulness of this therapeutic strategy a number of unsolved issues concerning which agent should be used and most appropriate timing and dosage remain to be explored.

The use of tirofiban (10µ/Kg bolus followed by a 0.15µ/Kg/min infusion) during PCI has been evaluated with controversial results. The TARGET trial showed that tirofiban provides a significantly less protection during PCI than abciximab, and the RESTORE trial found that tirofiban did not significantly reduce the combined endpoint in comparison with heparin alone. The sub-therapeutic inhibition of GP IIb/IIIa binding activity has been invoked, in the first hour of tirofiban treatment, as plausible explanation, in fact with a 10µ/Kg bolus as employed in the TARGET study about 60% of platelet inhibition was achieved in the first hour.

Schneider first proposed an high dose bolus (25µ/Kg bolus followed by a 0.15µ/Kg/min 18-h infusion) to improve the efficacy of tirofiban during percutaneous coronary interventions and the extent of average inhibition increased during first hour to 95% that is similar with that achieved by abciximab. This observation was followed by several papers (with short series of patients) comparing the efficacy and safety of high dose bolus of tirofiban and abciximab that showed similar efficacy for both IIb-IIIa inhibitors.

Among those the works of Danzi (Brescia I), Bolognese (EVEREST sudy ) (Arezzo I) and Gunasekara (Brisbane Au). In all these studies the number of enrolled patients was slightly enough to compare tirofiban and abciximab for their efficacy but it is still uncertain if high dose bolus of tirofiban could suffer of major side effects due to bleeding. Even if in all those studies the safety of high dose bolus of tirofiban was disclaimed, the study populations were too small for conclusive data. Actually no data are available in the literature on larger series of patients treated with high dose bolus of tirofiban.

1. Anna Hospital Report After the preliminary results showing a similar effect using abciximab and high dose bolus of tirofiban, from September 2002, we started high dose regimen of tirofiban with good therapeutic result and no evidence of major side effects. Since January 2003 high dose bolus of tirofiban was then adopted as routine regimen for all patients undergoing to percutaneous coronary interventions considering that treatment as having the best efficacy to cost ratio.

The data of all those patients (about 2000) were collected in our files and could be retrieved for a safety study with regard to the major side effect.

Study Endpoints

The primary endpoint is mortality. Secondary endpoints are: incidence of major bleeding and the rate of site access complication. Major bleeding is defined as cerebrovascular, (emorrhagic stroke), retroperitoneal bleeding, gastrointestinal bleeding, need for transfusion. Site access complications is defined as pseudoaneurysm, arteriovenous fistula, major hematoma (decrease in hematocrit level more than 15%) and need for surgical repair. Sub-groups of patients referred to our institution to perform a rescue-PCI few hours after thrombolyses failure and treated with high dose bolus of tirofiban will be also be investigated as patient with high-risk of bleeding side effects.

Postdischarge clinical outcomes were ascertained by means of a hospital visit or information collected through the referring cardiologist.

Statistical Analysis

Data obtained from about 2000 patient's files will be written on a dedicated data form and then they will be collected by means of a friendly user, dedicated program, compiled for that study and then exported in a data file that will be analysed by BMDP package.

Continuous variables are expressed as mean value ± SD, and discrete variables are expressed as absolute values and percentages. Clinical and instrumental variables will be compared using the Student's t-, chi square and Fisher's exact test. For groups and and subgroups analise of variance as well logistic regression will be performed. P values of <0.05 will be considered statistically significant.

We planned to perform random controls to check the quality of data collection and interinal analyse when 500th patient's data will be collected.

We think that even with the limitations of a retrospective study, such an investigation may be of high interest and thus set the base, for future, to collect and perform a prospective analyse of our data, in comparison (or in random allocation, with a 3 to 1 design) with different agent(s).