Author + information
- Received May 13, 1998
- Revision received September 3, 1998
- Accepted December 4, 1998
- Published online March 15, 1999.
- Karl-Christian Koch, MD∗,
- Juergen vom Dahl, MD, FESC∗,* (, )
- Eduard Kleinhans, MD†,
- Heinrich G Klues, MD, FESC∗,
- Peter W Radke, MD∗,
- Susanne Ninnemann∗,
- Gernot Schulz, MD†,
- Udalrich Buell, MD† and
- Peter Hanrath, MD, FACC, FESC∗
- ↵*Reprint requests and correspondence: Dr. Juergen vom Dahl, Medizinische Klinik I, Universitätsklinikum der RWTH Aachen, Pauwelstrasse 30, D-52057, Aachen, Germany
This study evaluated the effect of the glycoprotein IIb/IIIa (GPIIb/IIIa) antagonist abciximab on myocardial hypoperfusion during percutaneous transluminal rotational atherectomy (PTRA).
PTRA may cause transient ischemia and periprocedural myocardial injury. A platelet-dependent risk of non-Q-wave infarctions after directional atherectomy has been described. The role of platelets for the incidence and severity of myocardial hypoperfusion during PTRA is unknown.
Seventy-five consecutive patients with complex lesions were studied using resting Tc-99m sestamibi single-photon emission computed tomography prior to PTRA, during, and 2 days after the procedure. The last 30 patients received periprocedural abciximab (group A) and their results were compared to the remaining 45 patients (group B). For semiquantitative analysis, myocardial perfusion in 24 left ventricular regions was expressed as percentage of maximal sestamibi uptake.
Baseline characteristics did not differ between the groups. Transient perfusion defects were observed in 39/45 (87%) patients of group B, but only in 10/30 (33%) patients of group A (p < 0.001). Perfusion was significantly reduced during PTRA in 3.3 ± 2.5 regions in group B compared to 1.4 ± 2.5 regions in group A (p < 0.01). Perfusion in the region with maximal reduction during PTRA in groups B and A was 76 ± 15% and 76 ± 15% at baseline, decreased to 56 ± 16% (p < 0.001) and 67 ± 14%, respectively, during PTRA (p < 0.01 A vs. B), and returned to 76 ± 15% and 80 ± 13%, respectively, after PTRA. Nine patients in group B (20%) and two patients in group A (7%) had mild creatine kinase and/or troponin t elevations (p = 0.18). Patients with elevated enzymes had larger perfusion defects than did patients without myocardial injury (4.2 ± 2.7 vs. 2.3 ± 2.5 regions, p < 0.05).
These data indicate that GPIIb/IIIa blockade reduces incidence, extent and severity of transient hypoperfusion during PTRA. Thus, platelet aggregation may play an important role for PTRA-induced hypoperfusion.
Percutaneous transluminal rotational atherectomy (PTRA) is commonly employed in small coronary arteries, long and diffuse, calcified and ostial lesions (1–3). Previous experiences in complex and calcified coronary lesions demonstrated the potential of this technique to remove plaque material (4). However, transient ischemia during PTRA and frequent procedure-related non-Q-wave infarctions have been reported (1–3,5). A scintigraphic approach to assess the frequency, severity and reversibility of perfusion abnormalities during PTRA by serial myocardial perfusion single-photon emission computed tomography (SPECT) with Tc-99m-sestamibi using the advantage of myocardial sestamibi kinetics, which allows delayed imaging of procedure-related hypoperfusion, was recently described by our laboratory (6).
Blockade of the platelet glycoprotein IIb/IIIa (GPIIb/IIIa) receptor during directional atherectomy reduces the incidence of non-Q-wave infarction (7)and reduces creatine kinase (CK) elevations and clinical complications following PTRA (8). Recent in vitro experiments from several groups have shown that PTRA-induced platelet activation and aggregation can be inhibited by GPIIb/IIIa blockade (9,10). These data suggest that myocardial injury following atherectomy procedures is partially platelet-dependent.
This study sought to evaluate quantitatively the effect of GPIIb/IIIa receptor blockade by periprocedural administration of abciximab (Reopro) on myocardial hypoperfusion during PTRA using serial Tc-99m sestamibi SPECT.
Seventy five consecutive patients scheduled for PTRA of complex coronary lesions in the presence of stable angina pectoris symptoms and/or objective signs of ischemia were prospectively studied. The methodological approach has been reported previously describing the first 29 patients (6). The conventional treatment group (group B) consisted of 45 patients including these 29 patients. After 10 additional patients had been included in the conventional treatment group (group B), the following nonselected patients received peri-interventional treatment with abciximab (group A, n = 30), unless contraindications for GPIIb/IIIa blockade were present, in which case the patients (n = 6) were included in group B. The following contraindications for abciximab administration were considered: cerebrovascular complications within the past two years, neurosurgery within the past two months, intracranial tumors, known bleeding diathesis (n = 1), history of internal bleeding (n = 3), severe hypertension, thrombocytopenia, vasculitis, diabetic or hypertensive retinopathy, severe hepatic or renal disease, and previous abciximab administration (n = 2).
The institutional ethical committee approved the study protocol, and patients gave informed consent.
Rotational atherectomy and angiographic analysis
The technique of PTRA has been described previously in detail (1). Incremental burr sizes with steps ≤0.5 mm achieving an attempted burr/artery ratio of ≥0.7 were used. Several slow burr passages (≤30 s) were performed at ≥165,000 rpm with sufficient pauses and saline flushing between each advancement. Burr speed was routinely monitored and decelerations >5,000 rpm were avoided. Adjunctive percutaneous transluminal coronary angioplasty (PTCA) with or without additional stenting was performed as needed for optimal angiographic result at the operator’s discretion. In patients with a lesion of the right coronary artery, a temporary pacemaker was positioned in the right ventricle to prevent the effects of atherectomy-induced bradycardia.
Lesion length and severity were off-line digitally quantified (CAAS, PieMedical, Maastricht, The Netherlands). The degree of calcification was assessed by visual analysis using a three-grade scale: severe calcification: readily apparent on single-frame cineangiography; mild calcification: artery motion required to visualize; no visible calcification (2).
Procedural success was defined as final diameter stenosis of less than 50% in the absence of major complications (death, emergency bypass surgery, and Q-wave myocardial infarction). A non-Q-wave infarction was defined as elevation of creatine kinase (CK) more than three times the upper limit of normal in the absence of pathological Q-waves. “No flow” was defined as delayed antegrade flow (TIMI [Thrombolysis in Myocardial Infarction trial] grade less than 2) in the absence of a clear angiographic explanation of impaired flow (dissection, thrombus, spasm) at or adjacent to the original lesion.
Preprocedural medication included aspirin (100 mg daily) and antianginal drugs (Table 1). All patients were pretreated with ticlopidine (250 mg twice daily) beginning 1 day before the intervention in anticipation of a possible stent deployment. In patients in which no stent was deployed, ticlopidine was discontinued after the intervention. During the procedure, 10,000 IU of heparin were administered through the guiding catheter. Prior to baseline angiography, 0.2 mg nitroglycerin was injected intracoronarily and repeated during the procedure at the discretion of the operator to minimize vessel spasm (Table 1). Except for an intracoronary saline infusion, no specific treatment was given. Intravenous abciximab was administered as a bolus (0.25 mg/kg) ≥10 min prior to PTRA, followed by an infusion of 10 μg/min for 12 h.
SPECT imaging and analysis
For assessment of regional myocardial perfusion, SPECT with Tc-99m sestamibi was performed 1 day before, during PTRA and 2 days after the procedure as described previously (6). The Tc-99m sestamibi (400 to 450 MBq) was administered intravenously to the resting and fasting patient. During rotational atherectomy, intravenous Tc-99m-sestamibi was injected immediately after the last burr passage and intracoronary saline flushing and retrieval of the burr into the guiding catheter. Before SPECT imaging, the patient was allowed to have a light meal to minimize artifacts caused by enteric tracer uptake. The SPECT imaging and the data acquisition were performed 60 to 90 min after tracer injection in every study.
Automated semiquantitative image analysis was performed without knowledge of clinical or angiographic data using a rotating double-head gamma camera with a low-energy all-purpose collimator (ROTA II, Siemens, Germany). Transversal slices were reconstructed (Butterworth filter 5th order, 0.5 cutoff frequency, slice thickness 6.25 mm, matrix 64 × 64) and long and short axis views were obtained using dedicated software (MaxDelta, Siemens, Germany) and hardware (Micro Vax II, Siemens, Germany). For semiquantitative analysis the left ventricular myocardium was divided into 24 regions. Regional myocardial perfusion was expressed as percentage of the region with the maximal tracer uptake in each individual data set. Perfusion below 2 SDs of each regional normal mean value, derived from rest studies in a collective of normal individuals from our institution, was defined as significantly reduced.
Blood samples were drawn at baseline, and 8 h, 16 h and 24 h following the procedure for determination of total CK, CK-MB (creatine kinase MB fraction) and cardiac troponin t (TrT). Both CK and CK-MB were determined using a kinetic method for total CK and immunoinhibition for CK-MB (11). For quantitative measurement of TrT an enzyme-linked-immunosorbent assay was used (12). Creatine kinase was defined as elevated when >100 U/l with a CK-MB fraction >6%. Troponin t was defined as elevated when >0.1 ng/ml.
Data are mean ± SD. Differences in continuous variables were assessed using the Student ttest for paired and unpaired samples, and categorical variables were tested by the Fisher exact test.
Baseline patient, lesion and procedural characteristics
Baseline and procedural characteristics are presented in Table 1. For all baseline characteristics and procedural parameters there was no significant difference between group A and group B. Lesions were complex (type B and type C) and long (mean length 28 mm). Approximately 30% of the lesions were calcified and 40% were collateralized. No patient had a previous transmural Q-wave myocardial infarction in the target vessel territory, whereas approximately 40% of patients in each group had a previous non-Q-wave infarction. Adjunctive PTCA was performed in almost all procedures and adjunctive stents were implanted in 40% of patients. Clinical success rate was high, and only one patient in group B required a reintervention during hospital stay.
Influence of abciximab on myocardial perfusion during PTRA
Visual analysis of SPECT images revealed transient perfusion defects corresponding to the revascularized vessel distribution territory in 39/45 (87%) patients in group B (Table 2). In contrast, only 10/30 patients (33%) treated with abciximab (group A) displayed transient perfusion defects (p < 0.001) (Table 2). Scintigraphic images and regional sestamibi uptake values prior to, during and after PTRA in a representative patient without abciximab administration are illustrated in Figure 1.
Perfusion decreased significantly during PTRA in 3.3 ± 2.5 (0 to 10) regions/patient in group B (Table 2). In contrast, in group A only 1.4 ± 2.5 (0 to 6) hypoperfused regions/patient were observed during PTRA (p < 0.01).
Perfusion in the region with the maximal reduction during PTRA was 76 ± 15% prior to PTRA in group B, decreased significantly to 56 ± 16% (p < 0.001 vs. prior PTRA) during the procedure and normalized again to 76 ± 15 (p < 0.001 vs. during PTRA). In group A, corresponding perfusion values were 76 ± 15% at baseline, 67 ± 14% during PTRA (p < 0.001), and 80 ± 13 (p < 0.001 vs. during PTRA) after the procedure. Minimal myocardial perfusion during PTRA was significantly higher in the patients receiving abciximab than in the control group (67 ± 14% vs. 56 ± 16%, p < 0.01), whereas there was no significant difference at baseline and at follow-up SPECT imaging (Table 2).
One patient in group B experienced a procedure-related Q-wave infarction due to prolonged “no flow” during the procedure and subsequent abrupt vessel closure several hours following PTRA with a maximal CK value of 1170 U/l. Immediate reangiography revealed an occluded left anterior descending artery, and the patient was successfully treated with repeat PTCA and additional stenting.
As assessed by serial measurements of CK and TrT values, nine patients (20%) in group B and two patients (7%) (p = 0.18) in group A had mild CK and/or TrT elevations (range: CK, 105 to 203 U/l; TrT, 0.3 to 1.64 ng/ml). Creatine kinase or TrT release could be detected in 9/49 (18%) patients with a perfusion defect and in 2/26 patients (8%) (p = 0.31) without a perfusion defect. Defect size in patients with positive enzymes was 4.2 ± 2.7 regions, whereas defect size in patients without elevated enzymes was 2.3 ± 2.5 regions (p < 0.05).
Mechanisms of hypoperfusion
Myocardial hypoperfusion during PTRA may lead to transient myocardial dysfunction (1,13)and subsequent clinical sequelae including myocardial enzyme release and non-Q-wave infarctions. Early experimental studies suggested that most ablated particles should pass through the coronary microcirculation because of their small size (14). However, in later experiments by Friedman et al. (15), intracoronary infusion of atheromatous debris resulted in histologically proven isolated microinfarctions. Thus, myocardial hypoperfusion may be caused by obstructing microparticles. Furthermore, Zotz et al. (16)showed that hypoperfusion may be due to microbubbles obstructing the capillary bed. Rotation of the burr at 160,000 rpm in water and in fresh human blood produced bubbles with a mean size of approximately 90 μm and a short half-life.
Recently, platelet activation and aggregation has been proposed as another mechanism for transient hypoperfusion during PTRA, supported by preliminary in vivo and in vitro experiments (9,17). The number of platelet-monocyte complexes in peripheral blood as markers of platelet activation increased significantly after rotational atherectomy (17). Rotation of a burr in vitro in human blood at 180,000 rpm resulted in platelet aggregates ranging from 20 to 60 μm in diameter (9). Both lower rotation speed (140,000 rpm) and addition of the GPIIb/IIIa antagonist abciximab reduced the number of platelet aggregates significantly. In a recent in vitro study, rotablation in human platelet-rich plasma at 150,000 and 180,000 rpm resulted in rapid and extensive platelet aggregation, which was significantly reduced by preincubation with abciximab (10).
Our in vivo study further supports the hypothesis that platelet activation and aggregation is one mechanism leading to transient hypoperfusion. The platelet GPIIb/IIIa receptor is part of the final pathway of platelet aggregation. The reduction of transient ischemia during PTRA by specific blockade of this receptor provides strong evidence for the role of platelets in the pathophysiology of transient hypoperfusion.
However, hypoperfusion was not completely prevented by GPIIb/IIIa blockade, and other mechanisms like peripheral obstruction by ablated particles or microbubbles might be operative as well and responsible for the remaining effect. Although previous investigators have shown that the rotating device produces microbubbles, a causal relationship between microbubbles and myocardial hypoperfusion has not been established. The half-life of the bubbles is reported to be very short (≈10 s) (16), and the question arises whether these microbubbles can result in myocardial hypoperfusion that leads to such long-lasting (mean duration of 2.5 h) postprocedural wall-motion abnormalities as described by Williams et al. (13). In the latter echocardiographic study, hypoperfusion during balloon inflation in the PTCA control group resulted in transient wall-motion abnormalities with a mean recovery time of only 2.7 min.
Myocardial injury with release of myocardial enzymes has been described during PTRA and is probably linked to transient myocardial hypoperfusion. Data from the Evaluation of 7E3 for the Prevention of Ischemic Complication (EPIC) trial evaluating the effects of abciximab for the prevention of ischemic complications in percutaneous interventions demonstrated a significant reduction of non-Q-wave infarctions following directional atherectomy (7). Recently, a preliminary report (8)demonstrated a similar reduction of the incidence and magnitude of CK elevation following PTRA by abciximab.
Concordant with these studies, mild myocardial enzyme elevations were seen in nine patients (20%) of our control group, whereas only two patients (7%) in the abciximab group had enzymatic indices of myocardial injury. Although these differences did not reach statistical significance (this study was not designed to detect a significant difference with regard to clinical end points), these proportions are comparable to the findings of Braden et al. (8), who reported a 30% incidence of elevated CK in patients not receiving specific antiplatelet therapy and a significant reduction to a 17% incidence in abciximab-treated patients in a series of 326 patients.
Patients with perfusion defects had a higher rate of enzyme leakage compared to patients without perfusion defects, but this difference was again (most probably due to the group size) not statistically significant. The transient defect size in patients with periprocedural myocardial enzyme release was significantly larger than in patients without enzymatic signs of myocardial injury.
It can be speculated that transient hypoperfusion during PTRA occurs rather frequently but leads in only a small proportion to irreversible myocardial injury. Therefore, the beneficial effects of abciximab administration on the end point myocardial perfusion is detectable in an even smaller sample size than the effect on the end point CK or TrT release, for which this study was intentionally not designed and would have needed a significantly larger sample size.
The patients in this prospective but observational study were not randomized to abciximab or placebo. However, patients receiving abciximab were recruited consecutively, if no contraindications for abciximab were present and baseline characteristics between the groups were balanced. Because the control group was studied earlier than the treatment group, a learning curve effect due to improvement in interventional technique and growing operator experience might have influenced the results. However, this seems to be of minor importance, as periprocedural factors did not differ between both groups and two experienced operators performed all procedures. Analysis of nuclear images was performed without knowledge of clinical and/or angiographic data and without knowledge of periprocedural treatment. Therefore, any bias with regard to the scintigraphic analysis should be negligible.
We recognize that this study was underpowered to detect a beneficial effect of GPIIb/IIIa blockade on myocardial injury and clinical end points. However, we consider myocardial hypoperfusion during rotational atherectomy, for which we could demonstrate a beneficial effect of abciximab, as the primary and frequent event leading in some patients to definitive myocardial injury. This hypothesis is supported by the finding that myocardial hypoperfusion was significantly more extensive in patients with signs of myocardial injury than in patients without release of myocardial enzymes.
For a definitive answer as to whether peri-interventional GPIIbIIIa blockade reduces clinical events (such as CK or TrT release) during and after rotational atherectomy, a larger, randomized and placebo-controlled trial would be necessary.
Myocardial hypoperfusion during PTRA may lead to myocardial dysfunction with transient wall-motion abnormalities as detected by serial echocardiography during rotational atherectomy (13). In patients with a high-risk profile due to impaired left ventricular systolic function. PTRA may lead via the mechanism of transient hypoperfusion to a further reduction in left ventricular function and subsequent hemodynamic deterioration. Periprocedural therapy with a GPIIb/IIIa receptor antagonist could be used in these selected patients to prevent the potential adverse effects of rotational atherectomy. A similar prophylactic use has been proposed for intra-aortic balloon counterpulsation during rotational atherectomy in selected high-risk patients (18).
As has been shown by Abdelmeguid et al. (19)and in the EPIC study (20), minor CK elevations during coronary interventions might be associated with a worse long-term clinical outcome as compared to patients without CK elevation, although the reasons for this disparity are not clear yet. Periprocedural therapy with abciximab has been shown to reduce transient myocardial ischemia and incidence and magnitude of CK elevation and thus might favorably influence the long-term prognosis.
The results presented here indicate that administration of the platelet GPIIb/IIIa receptor antagonist abciximab reduces incidence, extent and severity of transient hypoperfusion during PTRA as measured by scintigraphic perfusion imaging with quantitative tracer uptake analysis. Thus, platelet activation and aggregation may be one of the major pathophysiologic mechanisms leading to PTRA-induced hypoperfusion.
- creatine kinase
- creatine kinase MB fraction
- glycoprotein IIb/IIIa
- percutaneous transluminal coronary angioplasty
- percutaneous transluminal rotational atherectomy
- single-photon emission computed tomography
- troponin t
- Received May 13, 1998.
- Revision received September 3, 1998.
- Accepted December 4, 1998.
- American College of Cardiology
- Teirstein P.S,
- Warth D.C,
- Najmul H,
- et al.
- Ellis S.G,
- Popma J.J,
- Buchbinder M,
- et al.
- Reisman M,
- Harms V,
- Whitlow P,
- et al.
- Mintz G.S,
- Potkin B.N,
- Keren G,
- et al.
- Bass T.A,
- Whitlow P.A,
- Moses J.W,
- et al.
- Koch K.C,
- Kleinhans E,
- Klues H.G,
- et al.
- Lefkovits J,
- Blankenship J.C,
- Anderson K.M,
- et al.
- Braden G.A,
- Love W.M,
- Applegate R.J,
- et al.
- Reisman M,
- Speck S,
- Petersen J.L,
- et al.
- Williams M.S,
- Coller B.S,
- Väänänen H.J,
- et al.
- Katus H.A,
- Looser S,
- Hallermeyer K,
- et al.
- Williams M.J.A,
- Dow C.J,
- Newell J.B,
- et al.
- Treuth M.G,
- Simon S.I,
- He Z.X,
- et al.
- O’Murchu B,
- Foreman R.D,
- Shaw R.E,
- et al.
- Abdelmeguid A.E,
- Topol E.J,
- Whitlow P.L,
- et al.
- Topol E.J,
- Ferguson J.J,
- Weisman H.F,
- et al.