- Email: [email protected]
Reduced procedural risk for coronary catheter interventions in carriers of the coagulation factor VII-Gln353 gene
Journal of the American College of Cardiology © 2000 by the American College of Cardiology Published by Elsevier Science Inc.
Vol. 36, No. 5, 2000 ISSN 0735-1097/00/$20.00 PII S0735-1097(00)00925-6
Interventional Cardiology
Reduced Procedural Risk for Coronary Catheter Interventions in Carriers of the Coagulation Factor VII-Gln353 Gene Przemyslaw M. Mrozikiewicz, MD,* Ingolf Cascorbi, MD, PHD,* Sabine Ziemer, MD,† Michael Laule, MD,‡ Christian Meisel, MD,* Verena Stangl, MD,‡ Wolfgang Rutsch, MD,‡ Klaus Wernecke, PHD,§ Gert Baumann, MD,‡ Ivar Roots, MD,* Karl Stangl, MD‡ Berlin, Germany We have focused on the role of coagulation factor VII (FVII) Arg353Gln polymorphism as a risk predictor of complications following percutaneous transluminal coronary angioplasty (PTCA), directional coronary atherectomy (DCA), and stenting. BACKGROUND The FVII Arg353Gln mutation decreases FVII activity, and presence of the Gln353 allele could be protective against thrombus formation during catheter interventions. METHODS A total of 666 consecutive patients with coronary artery disease who had undergone PTCA (n ⫽ 280), DCA (n ⫽ 104), or stenting (n ⫽ 282) were followed up for a 30-day composite end point, which included need for target vessel revascularization, myocardial infarction, and death. The Arg353Gln polymorphism of FVII was determined by PCR/RFLP assay. RESULTS Carriers of the Gln353 allele had significantly lower levels of total FVII activity (FVIIc, ⫺20.7%, p ⬍ 0.001) and of activated circulating FVII (FVIIa, ⫺32.7%, p ⫽ 0.03) compared with Arg353/Arg353. The composite end point occurred in 43 patients: 4 were heterozygous Arg353/Gln353, and 39 were homozygous Arg353/Arg353. The incidence of the composite end point was 2.5% in carriers of the Gln353 allele and 7.7% in Arg353/Arg353 homozygotes (p ⫽ 0.013). This corresponds to a 72% risk reduction in carriers of the Gln353 allele (relative risk: 0.28; 95% confidence interval: 0.09 – 0.81; p ⫽ 0.02). CONCLUSIONS The Gln353 allele of FVII is associated with substantial risk reduction in adverse events that complicate coronary catheter interventions. With the perspective of active site-blocked activated FVII (FVIIai) as conjunctive medication, the results suggest that the FVII genotype should be taken into due consideration in assessment of FVIIai medication and of its dosage. (J Am Coll Cardiol 2000;36:1520 –5) © 2000 by the American College of Cardiology OBJECTIVES
There is emerging evidence that the extrinsic coagulation pathway plays a major role in the pathophysiology of intravascular thrombus formation after tissue injury (1,2). This pathway is initiated when tissue factor (TF), a membrane-bound glycoprotein receptor, is exposed to flowing blood (3,4). Circulating factor VII (FVII), a singlechain zymogen, is the only known TF ligand (3). Binding to TF leads to an enhanced cleavage of FVII at Arg152 resulting in formation of activated FVII (FVIIa), a twochain serine protease, which activates coagulation factors X and IX, ultimately leading to thrombin generation (1–5). See page 1526 Increased FVII activity represents a risk factor for thrombotic events in the course of coronary artery disease (CAD) (6). Several polymorphisms influencing FVII activity have been recently identified. A single G10,976A-point mutation From the *Institute of Clinical Pharmacology, †Institute of Laboratory Medicine and Pathobiochemistry, ‡Department of Internal Medicine, Cardiology, §Institute of Biometry, Charite´ University Medical Center, Humboldt University of Berlin, Berlin, Germany. Manuscript received December 3, 1999; revised manuscript received April 28, 2000, accepted June 26, 2000.
(7) in exon 8 of the FVII gene causes an exchange of arginine to glutamine (Arg353Gln), which in turn results in decreased FVII activity by 20% to 30% (6 –9). It has been suggested that the Gln353 allele protects against myocardial infarction (MI) (6). Thus, presence of the Gln353 allele may consequently also be protective in other situations in which thrombus formation is a fundamental pathophysiological mechanism, as it is for adverse events complicating coronary catheter interventions. We, therefore, focused on the role of FVII Arg353Gln polymorphism as a risk predictor of complications following percutaneous transluminal coronary angioplasty (PTCA), directional coronary atherectomy (DCA), and stenting.
METHODS Study population and operational definitions. The 666 subjects participating in the present study were consecutive CAD patients receiving coronary catheter interventions at the Charite´ University Medical Center between October 1995 and January 1997. The design of this study has been described in detail elsewhere (10). Coronary artery disease was defined on the basis of angiographic criteria as stenosis ⱖ50% in a major coronary artery or major branch,
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
Abbreviations CAD ⫽ DCA ⫽ FVII ⫽ FVIIa ⫽ FVIIai ⫽ FVIIc ⫽ MI ⫽ nt ⫽ PTCA ⫽ TF TIMI TVR
and Acronyms coronary artery disease directional coronary atherectomy factor VII activated FVII active site-blocked activated FVII total FVII activity myocardial infarction nucleotide percutaneous transluminal coronary angioplasty ⫽ tissue factor ⫽ thrombolysis in myocardial infarction ⫽ target-vessel revascularization
with classification according to the number of affected arteries. Angiograms were analyzed by quantitative angiography. To ascertain the occurrence of the composite end point (in particular, periprocedural MI), we routinely performed surveillance of post-procedure creatine kinase, MB fraction levels at least twice for all patients. The CK levels greater than twice normal were established as criteria for suspecting MI. In addition, at least one post-procedural electrocardiogram was carried out. In case of MI suspicion on the basis of clinical, enzymatic and/or ECG evidence (11), angiography was again immediately carried out. The final diagnosis of MI was confirmed only in combination with angiographical findings. The 30-day composite clinical end point encompassed the following: need for target-vessel revascularization (TVR), MI, and death. Percutaneous transluminal coronary angioplasty was performed by standard techniques. Indications for DCA followed commonly accepted criteria, with target lesions located in the proximal sections of the three major coronary branches. Predilation was used to facilitate passage of the 7F Simpson atherocath. In each pass, 8 to 12 cuts were performed and repeated if necessary. In a manner analogous to the OARS (Optimal Atherectomy Restenosis Study) protocol (12), post-procedural PTCA at low inflation pressures (4 atm) was performed to reduce residual stenosis to a level ⬍20% (i.e., optimal DCA). Elective indications for stent implantation were de novo lesions (lesion length ⬍30 mm; vessel diameters ⱖ2.75 mm), restenosis, venousbypass graft lesions, and residual stenosis after PTCA or DCA ⱖ30%. Bail-out procedures were performed in typeD3 F dissections, for flow deterioration of at least 1 thrombolysis in myocardial infarction trial (TIMI) degree after PTCA or DCA, and in cases of acute vessel occlusion (i.e., TIMI flow grades 0 and 1). One coil stent type (Wiktor stent) and four tubular stent types (Palmaz-Schatz, NIR, Multi-link, Micro AVE) were used. The median number of stents/attempted lesion was 2.0 (1.2 to 2.8). Stents were inserted with use of high inflation pressures
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
1521
(⬎12 atm), and balloons were slightly oversized on the basis of visual estimate of the angiogram. We administered aspirin and 10,000 U of heparin to all patients before intervention and, as required, repeated doses of 2,500 U heparin to maintain activated clotting time ⬎250 s. Patients undergoing stent deployment received an additional 500 mg of ticlopidine orally after the procedure and for the following four weeks. All patients provided written consent according to the study protocol approved by the Charite´ Hospital Ethics Committee. Genotyping of FVII Arg353Gln, FVII ⴚ323 nt 0/10-bp promoter, and glycoprotein IIIa A1/A2 polymorphism; FVII determinations. Genomic DNA was amplified with primers 5⬘-GGG AGA CTC CCC AAA TAT CAC and 5⬘-ACG CAG CCT TGG CTT TCT CTC (7). Amplification conditions were 35 cycles for 30 s at 94°C, 30 s at 60°C, and 60 s at 72°C. Amplicons (312 bp) were digested with MspI. Fragments of 206 bp, 67 bp, and 39 bp were detected in presence of the Arg353 allele (G10,976), and 273-bp and 39-bp bands indicated the Gln353 allele (A10,976). The FVII ⫺323 nt 0/10-bp promoter polymorphism was genotyped according to the procedure described by Marchetti et al. (13). Genotyping of the A1/A2 polymorphism of glycoprotein IIIa (GP IIIa) has been described earlier (10). For determination of FVIIa and total FVII activity (FVIIc), venous blood samples were collected in tubes with sodium citrate solution between 7 and 8 AM, after overnight fasting of at least 10 h. After drawing, the samples were centrifuged for 15 min at a relative centrifugation force of at least 2,500. Total FVII activity was determined with FVII Deficient Plasma (Immuno AG, Vienna, Austria) and thromboplastin (Immunoplastin HIS, Immuno AG, Vienna, Austria) (14). Activated FVII (FVIIa) was determined by a commercial kit (STACLOT VIIa-rTF, Diagnostica Stago, Asnie`res-sur-Seine, France) (15). Levels of fibrinogen, von Willebrand factor, C-reactive protein C, and plasminogen activator inhibitor were determined by commercial kits. Statistical analysis. Frequencies of genotypes were compared by chi-square (2) test or the Fisher exact test. Values of FVIIc and FVIIa of different genotypes were compared by Mann-Whitney U-test. In an initial step we made univariate comparison of the genotypes—stratified by device (PTCA, DCA, and stent)—with respect to baseline characteristics, as well as common and procedural risk factors (Table 1). In a second step we developed a logistic regression in which the composite end point was included as dependent variable and the parameters listed in Table 4 were independent variables. Possible correlations between GPIIIa and FVII genotypes were tested using Pearson’s exact chi-square test. All statistical tests were calculated using SPSS 9.0.
1522
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
Table 1. Comparison of Baseline Characteristics, Common and Angiographic Risk Factors Stratified by Genotype and Device PTCA
DCA 353
N Age (yrs) Male/female (%) Smoking (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Acute MI (%) Unstable angina (%) Recanalization (%) Vessel diameter (mm) Lesion type A (%) B1 (%) B2 (%) C (%) Lesion length (mm) Vessel affected One (%) Two (%) Three (%) Residual stenosis (%) MLD pre (mm) post (mm)
353
Stenting 353
353
Arg353/Arg353
Arg /Gln ⴙ Gln353/Gln353
Arg353/Arg353
Arg /Gln ⴙ Gln353/Gln353
Arg353/Arg353
Arg353/Gln353 ⴙ Gln353/Gln353
212 60.7 (55.3–67.8) 78.3/21.7 47.6 22.6 50.5 50.5 2.8 15.1 15.1 2.6 (2.4–3.1)
68 59.8 (54–66.8) 75.0/25.0 51.5 20.6 55.9 61.8 2.9 10.3 13.2 2.6 (2.2–3.1)
79 59.2 (52.3–63.8) 84.8/15.2 53.2 17.7 51.9 51.9 8.9 22.8 2.5 3.3 (2.8–3.6)
25 57.4 (53.6–65.1) 76.0/24.0 36.0 36.0 56.0 52.0 2.4 28.0 1 3.5 (2.9–3.7)
218 61.2 (55.9–66.4) 73.4/26.6 41.7 22.5 60.1 53.2 6.0 17.4 15.6 3.1 (2.7–3.5)
64 59.2 (53.2–65.4) 79.7/20.3 56.3 18.8 54.7 51.6 12.0 15.6 10.9 3.1 (2.8–3.5)
24.1 10.8 48.6 16.5 8.2 (6.0–13.0)
33.8 5.9 41.2 19.1 8.0 (5.0–12.0)
36.7 5.1 37.9 20.3 7.0 (5.6–9.0)
40.0 4.0 24.0 32.0 7.0 (4.0–10.0)
2.3 6.9 64.2 26.6 8.0 (6.0–11.0)
15.8 15.6 53.1 15.5 8.0 (6.3–12.1)
21.2 44.3 34.4 22.5 (18.1–26.3)
27.9 51.5 20.6 20 (16.3–28.8)
43.0 35.4 21.5 12 (0.1–20.0)
32.0 52.0 16.0 15 (0.1–20.0)
28.9 38.1 33.0 3.8 (0.1–8.5)
29.7 35.9 34.4 3.2 (0.2–6.5)
0.8 (0.7–1.8) 1.7 (1.5–2.5)
1.0 (0.4–1.2) 2.0 (1.9–2.5)
0.9 (0.6–1.2) 2.5 (1.2–3.1)
1.0 (0.7–1.0) 2.7 (2.3–3.9)
0.5 (0.1–1.0) 3.2 (3.0–3.4)
0.6 (0.1–1.2) 3.0 (2.6–3.3)
Continuous values are expressed as median and 25th and 75th percentiles. MLD ⫽ minimal lumen diameter. There were no significant differences between the genotypes (Arg353/Arg353 vs. Arg353/Gln353 ⫹ Gln353/Gln353) within the three intervention groups.
RESULTS 353
Arg Gln polymorphism. Table 1 depicts demographic and angiographic baseline characteristics for the 666 patients studied according to genotype. The prevalence of common atherogenic and procedural risk factors did not differ between both groups (Table 1). This also held true when both groups were further stratified according to device. Genotype frequencies of Arg353/Arg353 homozygotes, Arg353/Gln353 heterozygotes, and Gln353/Gln353 homozygotes were 76.4% (95% confidence interval [CI], 73.0 –79.6%), 22.2% (19.1–25.6%), and 1.4% (0.6%–2.6%), respectively, being in accordance with Hardy-Weinberg equilibrium. The corresponding predicted values were 76.6%, 21.8%, and 1.6%. Allelic frequency of the Gln353 mutation was 12.5% (10.7%–14.4%). Correlation to FVII ⴚ323 nt 0/10-bp promoter and glycoprotein IIIa A1/A2 polymorphism. We further investigated the functionally relevant ⫺323 nt 0/10-bp promoter polymorphism (16) in our intervention group. The
genotype frequencies were 0-bp/0-bp 75.7% (predicted: 75.9%), 95% CI: 72.2–78.9%, 0-bp/10-bp 22.9% (predicted: 22.4%), 95% CI: 19.8 –26.4%, and 10-bp/10-bp 1.4% (predicted: 1.7%), 95% CI: 0.6%–2.4%. In 96.9% of patients the ⫺323 nt 0/10-bp promoter polymorphism was linked with the Arg353Gln polymorphism, indicating almost complete linkage disequilibrium. Furthermore, because the glycoprotein IIIa A1/A2 polymorphism has been described as inherited risk factor during stenting (17), we controlled our results for GP IIIa genotypes. In our results we found no significant correlations between FVII and GP IIIa genotypes (p ⫽ 0.47). Arg353Gln polymorphism and FVIIc and FVIIa. To determine FVIIa and FVIIc in a representative sample of the patient population, the total of 666 patients was randomly mixed, and then every second patient, up to number 359, was sampled. Allelic and genotype frequencies of this sample did not differ from those of the total population (data not shown).
Table 2. Correlation Between FVII Genotypes, FVIIc and FVIIa
N FVIIc (%) FVIIa (mU/ml)
Arg353/Arg353
Arg353/Gln353
p Value
Gln353/Gln353
p Value
Arg353/Gln353 ⴙ Gln353/Gln353
p Value
264 135 (111, 201) 55 (25, 313)
90 107 (89, 144) 38 (17, 275)
⬍ 0.001 0.05
5 86 (66, 160) 28 (7, 152)
0.04 0.12
95 107 (86, 144) 37 (16, 260)
⬍ 0.001 0.03
Values are given as median (25th, 75th percentile). All comparisons refer to Arg353/Arg353.
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
1523
Table 3. 30-Day Composite End Point and FVII Genotypes Arg353/Gln353 ⴙ Gln353/Gln353
OR (95% CI)
p Value
470 (92.3%) 39 (7.7%)
153 (97.5%) 4 (2.5%)
0.32 (0.08–0.90)
0.013
199 (93.9%) 13 (6.1%)
66 (97.1%) 2 (2.9%)
0.46 (0.05–2.1)
0.24
73 (92.4%) 6 (7.6%)
25 (100.0%) 0 (0.0%)
198 (90.8%) 20 (9.2%)
62 (96.9%) 2 (3.1%)
353
Arg Total (n ⫽ 666) No complications End point PTCA (n ⫽ 280) No complications End point DCA (n ⫽ 104) No complications End point Stent (n ⫽ 282) No complications End point
/Arg
353
—
0.27
0.32 (0.04–1.38)
0.09
Composite end point: need for TVR, MI, and death.
As shown in Table 2, FVIIc levels were lower by 20.7% (p ⬍ 0.001) and 36.3% (p ⫽ 0.04), respectively, lower in heterozygous and homozygous carriers of Gln353, compared with Arg353/Arg353. Similarly, FVIIa was reduced by 30.9% (p ⫽ 0.05) and 49.0% (p ⫽ 0.12), respectively. In the group reaching the composite end point there were no significant differences in FVIIc and FVIIa levels between homozygous Arg353/Arg353 and carriers of Gln353. FVII Arg353Gln polymorphism and 30-day composite end point. As Table 3 shows, 280 patients underwent PTCA; 104, DCA; and 282, stenting. The composite end point occurred in 43 patients, of which 42 occurred within the first seven days in the hospital. Immediate repeat angiography was performed for these patients. The only case of death was sudden and out-of-hospital, after 14 days. Of these 43 patients, 4 were heterozygous Arg353/Gln353 and 39 were homozygous Arg353/Arg353. No Gln353/Gln353 homozygote was found in the group with complications; we therefore calculated the combined risk for both Arg353/ Gln353 and Gln353/Gln353. The incidence of the composite end point was 2.5% in carriers of the Gln353 allele and 7.7% in Arg353/Arg353 homozygotes (odds ratio [OR], 0.32; 95% CI, 0.08 – 0.90; p ⫽ 0.013). There was also a tendency to Gln353-allele-dependent risk reduction when patients were further subgrouped by device (Table 3). Moreover, comparison for each single end point between carriers of the Gln353 allele and Arg353/Arg353 homozygotes revealed, respectively, significant risk reduction for TVR: 3 (1.9%) versus 27 (5.3%), p ⫽ 0.05; trends for MI: 1 (0.6%) versus 11 (2.2%), and death: 0 (0%) versus 1 (0.2%). To control for potential confounding in the relationship between the Arg353 polymorphism and ischemic complications, we employed logistic regression in the next step, with the composite end point as dependent variable and the risk factors listed in Table 4 as independent variables. As shown in Table 4 with a relative risk of 0.28 (95% CI: 0.09 – 0.81, p ⫽ 0.02) for the Arg353 polymorphism, this logistic regression model confirmed the marked risk reduction established in univariate analysis (Table 3). Furthermore, lesion type C proved to be associated with higher risk when
compared with types A, B1, and B2, and von Willebrand factor demonstrated a slight association with excess risk (Table 4). FVII Arg353Gln polymorphism and acute coronary syndromes. In our intervention group the Arg353Gln polymorphism was not associated with protection against acute MI (OR: 1.33; 95% CI: 0.80 –2.24; p ⫽ 0.28) or unstable angina (OR: 0.89; 95% CI: 0.60 –1.32; p ⫽ 0.58). We have also examined the question of association in the overall population of the study, with 2,000 cases and controls (10). Here, as well, we determined no associations with the 578 patients with a history of MI (OR: 0.82; 95% CI: 0.61–1.12; p ⫽ 0.13), the 91 with acute MI (OR: 1.47; 95% CI: 0.90 –2.41; p ⫽ 0.14), or the 144 with unstable angina (OR: 1.06; 95% CI: 0.69 –1.63; p ⫽ 0.74). Table 4. Relative Risk for the Composite End Point in Logistic Regression Analysis Variables FVII genotype (Arg/Arg vs. Arg/Gln ⫹ Gln/Gln) Gender Age Diabetes Hypercholesterolemia Hypertension Smoking Angina (unstable) Infarction (acute) No. of diseased vessels (1 vs. 3) No. of diseased vessels (2 vs. 3) Lesion type (A vs. C) Lesion type (B1 vs. C) Lesion type (B2 vs. C) Recanalization Reference vessel diameter Fibrinogen CrP PAI-1 vWF
RR 0.28 0.69 0.99 0.89 1.68 1.61 1.00 0.97 2.05 0.81 0.63 0.11 0.40 0.41 0.53 0.72 1.10 1.00 0.99 1.0035
95% CI
p Value
0.09–0.81
0.02
0.31–1.50 0.95–1.03 0.40–2.01 0.83–3.37 0.81–3.19 0.50–2.09 0.42–2.43 0.33–12.85 0.35–1.87 0.26–1.49 0.01–0.91 0.13–1.27 0.19–0.88 0.18–1.54 0.40–1.27 0.81–1.50 0.99–1.01 0.96–1.03 1.0009–1.0061
0.35 0.62 0.78 0.15 0.18 0.96 0.99 0.44 0.63 0.29 0.04 0.12 0.02 0.24 0.25 0.52 0.93 0.71 0.01
CrP ⫽ C-reactive protein; PAI-1 ⫽ plasminogen activator inhibitor; vWF ⫽ von Willebrand factor.
1524
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
DISCUSSION Findings. The present study provides first evidence that the Gln353 allele of coagulation factor VII is associated with substantial risk reduction (in our findings, by two thirds) following coronary catheter interventions. It was a crucial finding of this study that the risk reduction was fairly consistent across the component end points as well as across the three different types of catheter interventions. In the large sample size of 666 patients, 43 patients suffered from major complications during the first 30 days and reached the composite end point. This small absolute number of complications may explain why the reduction in risk achieved significance only in the context of our total patient population. Because we had hypothesized that the Arg353Gln polymorphism associated with a reduction of FVIIa could reduce the vascular events such as thrombosis, we were slightly surprised to find no reduction of risk for acute coronary syndromes or for chronic MI. Conversely, these results of ours did concur with case-control studies with large populations (18) as well as with the results of the prospective Framingham Study (19). Potential limitations. The surprising results for the group as a whole (i.e., the substantial reduction of procedural risk among carriers of the Gln353 allele) must be interpreted with the necessary circumspection. Discussion of the possibility of coincidence in the finding data, or of selection bias in the patient group, is necessary with respect to Gln353 allelic frequency and the complication rate. The following, however, speaks against the possibility of selection bias: the fact that the Gln353 allelic frequency of 12.5% disclosed in our study lies within the reported range for Middle European Caucasians (7,9). Moreover, the prevalence of the composite end point of 6.5% as determined by us corresponds to comparable figures achieved by others with consecutive patients (20). From the standpoint of work being performed today, our DCA frequency of 16% appears to be very high. This level may be justified by considering that our interventions took place basically in 1996, at the beginning of the stent era, and that this percent figure is in line with data published in a comparable registry (20). Possible mechanisms. During catheter interventions the protective endothelial lining of the arterial wall is disrupted, and such catheter-induced endothelial injury may trigger thrombogenesis. It has been evidenced by our findings (Table 2) that the Gln353 mutation possesses functional importance, suggesting a gene-dose effect that consists in a FVIIc reduction of 21% in heterozygous and of 36% in homozygous Gln353 carriers. An analogous phenomenon could be observed in FVIIa levels, even though the 49% reduction in Gln353 homozygotes did not reach significance. These phenotypical findings accord well with the literature (6 –9) and may explain the observed protective effect of the Gln353 allele. Because a ⫺323 nt 0/10-bp polymorphism in the pro-
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
moter region has been described as functionally relevant in lowering FVII levels (16), we examined this aspect in our study group. The almost complete disequilibrium between this promoter polymorphism and the Arg353Gln polymorphism determined in our study accords with findings of other groups (21). By contrast, these “within-gene” allelic interactions do not in the clinical context of our study allow differentiation between the effects of each single polymorphism on decreased FVII levels. Instead, we can represent the decreased levels only as cumulative effects. Although many factors influence FVII concentration and activity, presence of the Gln353 allele is not necessarily associated with lower FVII levels, as demonstrated in the group of Gln353 allele carriers presenting with complications. By contrast to functional variations of FVIIc or FVIIa levels, the FVII genotype is a constant predictor of the lifelong tendency of individual FVII activity. Our observation (i.e., risk reduction for thrombotic events) is in line with findings by Iacoviello and colleagues (6), who described a protective effect against MI in carriers of the Gln353 allele. By contrast, several studies failed to demonstrate any protection (18,22). Clinical implications. Our results could well have two clinically important implications: first, that FVII polymorphism has an impact on the development of catheter complications and may serve as a risk predictor. Second, with the perspective of active site-blocked activated FVII (FVIIai) (23) as conjunctive medication during catheter interventions, our findings suggest that the FVII genotype should be taken into due consideration in use, dosage, and assessment of FVIIai effectiveness. Acknowledgment This work was supported in part by a grant of the German Federal Ministry of Education, Science, Research, and Technology (No. 01 GG 9845-5). Reprint requests and correspondence: Dr. Karl Stangl, Medizinische Klinik, Kardiologie, Angiologie, Pneumologie, Charite´, Campus Mitte, Humboldt Universita¨t zu Berlin, Schumannstrasse 20/21, D-10117 Berlin, Germany. E-mail: [email protected].
REFERENCES 1. Furie B, Furie BC. The molecular basis of blood coagulation. Cell 1988;53:505–18. 2. Rapaport SI, Rao LVM. Initiation and regulation of tissue factordependent blood coagulation. Arteriosler Thromb 1992;12:1111–21. 3. Nemerson Y. Tissue factor: then and now. Thromb Haemost 1995; 74:180 – 4. 4. Rao LV, Rapaport SI. Cells and the activation of factor VII. Haemostasis 1996;26 Suppl 1:1–5. 5. Muller YA, Ultsch MH, Kelley RF, de Vos AM. Structure of the extracellular domain of human tissue factor: location of the factor VIIa binding site. Biochemistry 1994;33:10864 –70. 6. Iacoviello L, Di Castelnuovo A, de Knijff P, et al. Polymorphisms in the coagulation factor VII gene and the risk of myocardial infarction. N Engl J Med 1998;338:79 – 85. 7. Green F, Kelleher C, Wilkes H, Temple A, Meade T, Humphries S. A common genetic polymorphism associated with lower coagulation
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
8.
9.
10.
11.
12. 13. 14.
15.
factor VII levels in healthy individuals. Arterioscler Thromb 1991;11: 540 – 6. Ghaddar HM, Folsom AR, Aleksic N, et al. Correlation of factor VIIa values with factor VII gene polymorphism, fasting and postprandial triglyceride levels, and subclinical carotid atherosclerosis. Circulation 1998;98:2815–21. Hanault M, Arbini AA, Lopaciuk S, Carew JA, Bauer KA. The Arg353Gln polymorphism reduces the level of coagulation factor VII. In vivo and in vitro studies. Arterioscler Thromb Vasc Biol 1997;17: 2825–9. Laule M, Cascorbi I, Stangl V, et al. A1/A2 polymorphism of glycoprotein IIIa and association with excess procedural risk for coronary catheter interventions: a case-controlled study. Lancet 1999; 353:708 –12. Nomenclature and criteria for diagnosis of ischemic heart disease. Report of the Joint International Society and Federation of Cardiology/World Health Organization Task Force on Standardization of Clinical Nomenclature. Circulation 1979;59:607–9. Simonton CA, Leon MB, Baim DS, et al. ‘Optimal’ directional coronary atherectomy: final results of the Optimal Atherectomy Restenosis Study (OARS). Circulation 1998;97:332–9. Marchetti G, Patracchini P, Papacchini M, Ferrati M, Bernardi F. A polymorphism in the 5⬘ region of coagulation factor VII gene (F7) caused by an inserted decanucleotide. Hum Genet 1993;90:575– 6. Morrissey JH, Macik BG, Neuenschwander PF, Comp PC. Quantitation of activated factor VII levels in plasma using a tissue factor mutant selectively deficient in promoting factor VII activation. Blood 1993;81:734 – 44. Hellstern P, Beeck H, Fellhauer A, Fischer A, Faller-Stockl B. Measurement of factor VII and of activated factor VII in healthy
16. 17. 18. 19.
20. 21.
22.
23.
1525
individuals and in prothrombin complex concentrates. Thromb Res 1997;86:493–504. Pollak ES, Hung HL, Godin W, Overton GC, High KA. Functional characterisation of the human factor VII 5⬘ flanking region. J Biol Chem 1996;271:1738 – 47. Walter DH, Schachinger V, Elsner M, Dimmeler S, Zeiher AM. Platelet glycoprotein IIIa polymorphisms and risk of coronary stent thrombosis. Lancet 1997;350:1217–9. Lane DA, Grant PJ. Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease. Blood 2000;95:1517–32. Feng D, Tofler GH, Karson MG, et al. Factor VII gene polymorphism, factor VII levels, and prevalent cardiovascular disease: the Framingham Heart Study. Arterioscler Thromb Vasc Biol 2000;20: 593– 600. King SB III, Yeh W, Holubkov R, et al. Balloon angioplasty versus new device intervention: clinical outcomes. A comparison of the NHLBI and NACI registries. J Am Coll Cardiol 1998;31:5558 – 66. Di Castelnuovo A, D’Orazio A, Amore C, Falanga A, Donati MB, lacoviello L. The decanucleotide insertion/deletion polymorphism in the promoter region of the coagulation factor VII gene and the risk of myocardial infarction. Thromb Res 2000;98:9 –17. Lane A, Green F, Scarabin PY, et al. Factor VII Arg/Gln353 polymorphism determines factor VII coagulant activity in patients with myocardial infarction (MI) and control subjects in Belfast and in France but is not a strong indicator of MI risk in the ECTIM study. Atherosclerosis 1996;119:119 –27. Golino P, Ragni M, Cirillo P, et al. Antithrombotic effects of recombinant human, active site-blocked factor VIIa in a rabbit model of recurrent arterial thrombosis. Circ Res 1998;82:39 – 46.
Vol. 36, No. 5, 2000 ISSN 0735-1097/00/$20.00 PII S0735-1097(00)00925-6
Interventional Cardiology
Reduced Procedural Risk for Coronary Catheter Interventions in Carriers of the Coagulation Factor VII-Gln353 Gene Przemyslaw M. Mrozikiewicz, MD,* Ingolf Cascorbi, MD, PHD,* Sabine Ziemer, MD,† Michael Laule, MD,‡ Christian Meisel, MD,* Verena Stangl, MD,‡ Wolfgang Rutsch, MD,‡ Klaus Wernecke, PHD,§ Gert Baumann, MD,‡ Ivar Roots, MD,* Karl Stangl, MD‡ Berlin, Germany We have focused on the role of coagulation factor VII (FVII) Arg353Gln polymorphism as a risk predictor of complications following percutaneous transluminal coronary angioplasty (PTCA), directional coronary atherectomy (DCA), and stenting. BACKGROUND The FVII Arg353Gln mutation decreases FVII activity, and presence of the Gln353 allele could be protective against thrombus formation during catheter interventions. METHODS A total of 666 consecutive patients with coronary artery disease who had undergone PTCA (n ⫽ 280), DCA (n ⫽ 104), or stenting (n ⫽ 282) were followed up for a 30-day composite end point, which included need for target vessel revascularization, myocardial infarction, and death. The Arg353Gln polymorphism of FVII was determined by PCR/RFLP assay. RESULTS Carriers of the Gln353 allele had significantly lower levels of total FVII activity (FVIIc, ⫺20.7%, p ⬍ 0.001) and of activated circulating FVII (FVIIa, ⫺32.7%, p ⫽ 0.03) compared with Arg353/Arg353. The composite end point occurred in 43 patients: 4 were heterozygous Arg353/Gln353, and 39 were homozygous Arg353/Arg353. The incidence of the composite end point was 2.5% in carriers of the Gln353 allele and 7.7% in Arg353/Arg353 homozygotes (p ⫽ 0.013). This corresponds to a 72% risk reduction in carriers of the Gln353 allele (relative risk: 0.28; 95% confidence interval: 0.09 – 0.81; p ⫽ 0.02). CONCLUSIONS The Gln353 allele of FVII is associated with substantial risk reduction in adverse events that complicate coronary catheter interventions. With the perspective of active site-blocked activated FVII (FVIIai) as conjunctive medication, the results suggest that the FVII genotype should be taken into due consideration in assessment of FVIIai medication and of its dosage. (J Am Coll Cardiol 2000;36:1520 –5) © 2000 by the American College of Cardiology OBJECTIVES
There is emerging evidence that the extrinsic coagulation pathway plays a major role in the pathophysiology of intravascular thrombus formation after tissue injury (1,2). This pathway is initiated when tissue factor (TF), a membrane-bound glycoprotein receptor, is exposed to flowing blood (3,4). Circulating factor VII (FVII), a singlechain zymogen, is the only known TF ligand (3). Binding to TF leads to an enhanced cleavage of FVII at Arg152 resulting in formation of activated FVII (FVIIa), a twochain serine protease, which activates coagulation factors X and IX, ultimately leading to thrombin generation (1–5). See page 1526 Increased FVII activity represents a risk factor for thrombotic events in the course of coronary artery disease (CAD) (6). Several polymorphisms influencing FVII activity have been recently identified. A single G10,976A-point mutation From the *Institute of Clinical Pharmacology, †Institute of Laboratory Medicine and Pathobiochemistry, ‡Department of Internal Medicine, Cardiology, §Institute of Biometry, Charite´ University Medical Center, Humboldt University of Berlin, Berlin, Germany. Manuscript received December 3, 1999; revised manuscript received April 28, 2000, accepted June 26, 2000.
(7) in exon 8 of the FVII gene causes an exchange of arginine to glutamine (Arg353Gln), which in turn results in decreased FVII activity by 20% to 30% (6 –9). It has been suggested that the Gln353 allele protects against myocardial infarction (MI) (6). Thus, presence of the Gln353 allele may consequently also be protective in other situations in which thrombus formation is a fundamental pathophysiological mechanism, as it is for adverse events complicating coronary catheter interventions. We, therefore, focused on the role of FVII Arg353Gln polymorphism as a risk predictor of complications following percutaneous transluminal coronary angioplasty (PTCA), directional coronary atherectomy (DCA), and stenting.
METHODS Study population and operational definitions. The 666 subjects participating in the present study were consecutive CAD patients receiving coronary catheter interventions at the Charite´ University Medical Center between October 1995 and January 1997. The design of this study has been described in detail elsewhere (10). Coronary artery disease was defined on the basis of angiographic criteria as stenosis ⱖ50% in a major coronary artery or major branch,
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
Abbreviations CAD ⫽ DCA ⫽ FVII ⫽ FVIIa ⫽ FVIIai ⫽ FVIIc ⫽ MI ⫽ nt ⫽ PTCA ⫽ TF TIMI TVR
and Acronyms coronary artery disease directional coronary atherectomy factor VII activated FVII active site-blocked activated FVII total FVII activity myocardial infarction nucleotide percutaneous transluminal coronary angioplasty ⫽ tissue factor ⫽ thrombolysis in myocardial infarction ⫽ target-vessel revascularization
with classification according to the number of affected arteries. Angiograms were analyzed by quantitative angiography. To ascertain the occurrence of the composite end point (in particular, periprocedural MI), we routinely performed surveillance of post-procedure creatine kinase, MB fraction levels at least twice for all patients. The CK levels greater than twice normal were established as criteria for suspecting MI. In addition, at least one post-procedural electrocardiogram was carried out. In case of MI suspicion on the basis of clinical, enzymatic and/or ECG evidence (11), angiography was again immediately carried out. The final diagnosis of MI was confirmed only in combination with angiographical findings. The 30-day composite clinical end point encompassed the following: need for target-vessel revascularization (TVR), MI, and death. Percutaneous transluminal coronary angioplasty was performed by standard techniques. Indications for DCA followed commonly accepted criteria, with target lesions located in the proximal sections of the three major coronary branches. Predilation was used to facilitate passage of the 7F Simpson atherocath. In each pass, 8 to 12 cuts were performed and repeated if necessary. In a manner analogous to the OARS (Optimal Atherectomy Restenosis Study) protocol (12), post-procedural PTCA at low inflation pressures (4 atm) was performed to reduce residual stenosis to a level ⬍20% (i.e., optimal DCA). Elective indications for stent implantation were de novo lesions (lesion length ⬍30 mm; vessel diameters ⱖ2.75 mm), restenosis, venousbypass graft lesions, and residual stenosis after PTCA or DCA ⱖ30%. Bail-out procedures were performed in typeD3 F dissections, for flow deterioration of at least 1 thrombolysis in myocardial infarction trial (TIMI) degree after PTCA or DCA, and in cases of acute vessel occlusion (i.e., TIMI flow grades 0 and 1). One coil stent type (Wiktor stent) and four tubular stent types (Palmaz-Schatz, NIR, Multi-link, Micro AVE) were used. The median number of stents/attempted lesion was 2.0 (1.2 to 2.8). Stents were inserted with use of high inflation pressures
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
1521
(⬎12 atm), and balloons were slightly oversized on the basis of visual estimate of the angiogram. We administered aspirin and 10,000 U of heparin to all patients before intervention and, as required, repeated doses of 2,500 U heparin to maintain activated clotting time ⬎250 s. Patients undergoing stent deployment received an additional 500 mg of ticlopidine orally after the procedure and for the following four weeks. All patients provided written consent according to the study protocol approved by the Charite´ Hospital Ethics Committee. Genotyping of FVII Arg353Gln, FVII ⴚ323 nt 0/10-bp promoter, and glycoprotein IIIa A1/A2 polymorphism; FVII determinations. Genomic DNA was amplified with primers 5⬘-GGG AGA CTC CCC AAA TAT CAC and 5⬘-ACG CAG CCT TGG CTT TCT CTC (7). Amplification conditions were 35 cycles for 30 s at 94°C, 30 s at 60°C, and 60 s at 72°C. Amplicons (312 bp) were digested with MspI. Fragments of 206 bp, 67 bp, and 39 bp were detected in presence of the Arg353 allele (G10,976), and 273-bp and 39-bp bands indicated the Gln353 allele (A10,976). The FVII ⫺323 nt 0/10-bp promoter polymorphism was genotyped according to the procedure described by Marchetti et al. (13). Genotyping of the A1/A2 polymorphism of glycoprotein IIIa (GP IIIa) has been described earlier (10). For determination of FVIIa and total FVII activity (FVIIc), venous blood samples were collected in tubes with sodium citrate solution between 7 and 8 AM, after overnight fasting of at least 10 h. After drawing, the samples were centrifuged for 15 min at a relative centrifugation force of at least 2,500. Total FVII activity was determined with FVII Deficient Plasma (Immuno AG, Vienna, Austria) and thromboplastin (Immunoplastin HIS, Immuno AG, Vienna, Austria) (14). Activated FVII (FVIIa) was determined by a commercial kit (STACLOT VIIa-rTF, Diagnostica Stago, Asnie`res-sur-Seine, France) (15). Levels of fibrinogen, von Willebrand factor, C-reactive protein C, and plasminogen activator inhibitor were determined by commercial kits. Statistical analysis. Frequencies of genotypes were compared by chi-square (2) test or the Fisher exact test. Values of FVIIc and FVIIa of different genotypes were compared by Mann-Whitney U-test. In an initial step we made univariate comparison of the genotypes—stratified by device (PTCA, DCA, and stent)—with respect to baseline characteristics, as well as common and procedural risk factors (Table 1). In a second step we developed a logistic regression in which the composite end point was included as dependent variable and the parameters listed in Table 4 were independent variables. Possible correlations between GPIIIa and FVII genotypes were tested using Pearson’s exact chi-square test. All statistical tests were calculated using SPSS 9.0.
1522
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
Table 1. Comparison of Baseline Characteristics, Common and Angiographic Risk Factors Stratified by Genotype and Device PTCA
DCA 353
N Age (yrs) Male/female (%) Smoking (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Acute MI (%) Unstable angina (%) Recanalization (%) Vessel diameter (mm) Lesion type A (%) B1 (%) B2 (%) C (%) Lesion length (mm) Vessel affected One (%) Two (%) Three (%) Residual stenosis (%) MLD pre (mm) post (mm)
353
Stenting 353
353
Arg353/Arg353
Arg /Gln ⴙ Gln353/Gln353
Arg353/Arg353
Arg /Gln ⴙ Gln353/Gln353
Arg353/Arg353
Arg353/Gln353 ⴙ Gln353/Gln353
212 60.7 (55.3–67.8) 78.3/21.7 47.6 22.6 50.5 50.5 2.8 15.1 15.1 2.6 (2.4–3.1)
68 59.8 (54–66.8) 75.0/25.0 51.5 20.6 55.9 61.8 2.9 10.3 13.2 2.6 (2.2–3.1)
79 59.2 (52.3–63.8) 84.8/15.2 53.2 17.7 51.9 51.9 8.9 22.8 2.5 3.3 (2.8–3.6)
25 57.4 (53.6–65.1) 76.0/24.0 36.0 36.0 56.0 52.0 2.4 28.0 1 3.5 (2.9–3.7)
218 61.2 (55.9–66.4) 73.4/26.6 41.7 22.5 60.1 53.2 6.0 17.4 15.6 3.1 (2.7–3.5)
64 59.2 (53.2–65.4) 79.7/20.3 56.3 18.8 54.7 51.6 12.0 15.6 10.9 3.1 (2.8–3.5)
24.1 10.8 48.6 16.5 8.2 (6.0–13.0)
33.8 5.9 41.2 19.1 8.0 (5.0–12.0)
36.7 5.1 37.9 20.3 7.0 (5.6–9.0)
40.0 4.0 24.0 32.0 7.0 (4.0–10.0)
2.3 6.9 64.2 26.6 8.0 (6.0–11.0)
15.8 15.6 53.1 15.5 8.0 (6.3–12.1)
21.2 44.3 34.4 22.5 (18.1–26.3)
27.9 51.5 20.6 20 (16.3–28.8)
43.0 35.4 21.5 12 (0.1–20.0)
32.0 52.0 16.0 15 (0.1–20.0)
28.9 38.1 33.0 3.8 (0.1–8.5)
29.7 35.9 34.4 3.2 (0.2–6.5)
0.8 (0.7–1.8) 1.7 (1.5–2.5)
1.0 (0.4–1.2) 2.0 (1.9–2.5)
0.9 (0.6–1.2) 2.5 (1.2–3.1)
1.0 (0.7–1.0) 2.7 (2.3–3.9)
0.5 (0.1–1.0) 3.2 (3.0–3.4)
0.6 (0.1–1.2) 3.0 (2.6–3.3)
Continuous values are expressed as median and 25th and 75th percentiles. MLD ⫽ minimal lumen diameter. There were no significant differences between the genotypes (Arg353/Arg353 vs. Arg353/Gln353 ⫹ Gln353/Gln353) within the three intervention groups.
RESULTS 353
Arg Gln polymorphism. Table 1 depicts demographic and angiographic baseline characteristics for the 666 patients studied according to genotype. The prevalence of common atherogenic and procedural risk factors did not differ between both groups (Table 1). This also held true when both groups were further stratified according to device. Genotype frequencies of Arg353/Arg353 homozygotes, Arg353/Gln353 heterozygotes, and Gln353/Gln353 homozygotes were 76.4% (95% confidence interval [CI], 73.0 –79.6%), 22.2% (19.1–25.6%), and 1.4% (0.6%–2.6%), respectively, being in accordance with Hardy-Weinberg equilibrium. The corresponding predicted values were 76.6%, 21.8%, and 1.6%. Allelic frequency of the Gln353 mutation was 12.5% (10.7%–14.4%). Correlation to FVII ⴚ323 nt 0/10-bp promoter and glycoprotein IIIa A1/A2 polymorphism. We further investigated the functionally relevant ⫺323 nt 0/10-bp promoter polymorphism (16) in our intervention group. The
genotype frequencies were 0-bp/0-bp 75.7% (predicted: 75.9%), 95% CI: 72.2–78.9%, 0-bp/10-bp 22.9% (predicted: 22.4%), 95% CI: 19.8 –26.4%, and 10-bp/10-bp 1.4% (predicted: 1.7%), 95% CI: 0.6%–2.4%. In 96.9% of patients the ⫺323 nt 0/10-bp promoter polymorphism was linked with the Arg353Gln polymorphism, indicating almost complete linkage disequilibrium. Furthermore, because the glycoprotein IIIa A1/A2 polymorphism has been described as inherited risk factor during stenting (17), we controlled our results for GP IIIa genotypes. In our results we found no significant correlations between FVII and GP IIIa genotypes (p ⫽ 0.47). Arg353Gln polymorphism and FVIIc and FVIIa. To determine FVIIa and FVIIc in a representative sample of the patient population, the total of 666 patients was randomly mixed, and then every second patient, up to number 359, was sampled. Allelic and genotype frequencies of this sample did not differ from those of the total population (data not shown).
Table 2. Correlation Between FVII Genotypes, FVIIc and FVIIa
N FVIIc (%) FVIIa (mU/ml)
Arg353/Arg353
Arg353/Gln353
p Value
Gln353/Gln353
p Value
Arg353/Gln353 ⴙ Gln353/Gln353
p Value
264 135 (111, 201) 55 (25, 313)
90 107 (89, 144) 38 (17, 275)
⬍ 0.001 0.05
5 86 (66, 160) 28 (7, 152)
0.04 0.12
95 107 (86, 144) 37 (16, 260)
⬍ 0.001 0.03
Values are given as median (25th, 75th percentile). All comparisons refer to Arg353/Arg353.
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
1523
Table 3. 30-Day Composite End Point and FVII Genotypes Arg353/Gln353 ⴙ Gln353/Gln353
OR (95% CI)
p Value
470 (92.3%) 39 (7.7%)
153 (97.5%) 4 (2.5%)
0.32 (0.08–0.90)
0.013
199 (93.9%) 13 (6.1%)
66 (97.1%) 2 (2.9%)
0.46 (0.05–2.1)
0.24
73 (92.4%) 6 (7.6%)
25 (100.0%) 0 (0.0%)
198 (90.8%) 20 (9.2%)
62 (96.9%) 2 (3.1%)
353
Arg Total (n ⫽ 666) No complications End point PTCA (n ⫽ 280) No complications End point DCA (n ⫽ 104) No complications End point Stent (n ⫽ 282) No complications End point
/Arg
353
—
0.27
0.32 (0.04–1.38)
0.09
Composite end point: need for TVR, MI, and death.
As shown in Table 2, FVIIc levels were lower by 20.7% (p ⬍ 0.001) and 36.3% (p ⫽ 0.04), respectively, lower in heterozygous and homozygous carriers of Gln353, compared with Arg353/Arg353. Similarly, FVIIa was reduced by 30.9% (p ⫽ 0.05) and 49.0% (p ⫽ 0.12), respectively. In the group reaching the composite end point there were no significant differences in FVIIc and FVIIa levels between homozygous Arg353/Arg353 and carriers of Gln353. FVII Arg353Gln polymorphism and 30-day composite end point. As Table 3 shows, 280 patients underwent PTCA; 104, DCA; and 282, stenting. The composite end point occurred in 43 patients, of which 42 occurred within the first seven days in the hospital. Immediate repeat angiography was performed for these patients. The only case of death was sudden and out-of-hospital, after 14 days. Of these 43 patients, 4 were heterozygous Arg353/Gln353 and 39 were homozygous Arg353/Arg353. No Gln353/Gln353 homozygote was found in the group with complications; we therefore calculated the combined risk for both Arg353/ Gln353 and Gln353/Gln353. The incidence of the composite end point was 2.5% in carriers of the Gln353 allele and 7.7% in Arg353/Arg353 homozygotes (odds ratio [OR], 0.32; 95% CI, 0.08 – 0.90; p ⫽ 0.013). There was also a tendency to Gln353-allele-dependent risk reduction when patients were further subgrouped by device (Table 3). Moreover, comparison for each single end point between carriers of the Gln353 allele and Arg353/Arg353 homozygotes revealed, respectively, significant risk reduction for TVR: 3 (1.9%) versus 27 (5.3%), p ⫽ 0.05; trends for MI: 1 (0.6%) versus 11 (2.2%), and death: 0 (0%) versus 1 (0.2%). To control for potential confounding in the relationship between the Arg353 polymorphism and ischemic complications, we employed logistic regression in the next step, with the composite end point as dependent variable and the risk factors listed in Table 4 as independent variables. As shown in Table 4 with a relative risk of 0.28 (95% CI: 0.09 – 0.81, p ⫽ 0.02) for the Arg353 polymorphism, this logistic regression model confirmed the marked risk reduction established in univariate analysis (Table 3). Furthermore, lesion type C proved to be associated with higher risk when
compared with types A, B1, and B2, and von Willebrand factor demonstrated a slight association with excess risk (Table 4). FVII Arg353Gln polymorphism and acute coronary syndromes. In our intervention group the Arg353Gln polymorphism was not associated with protection against acute MI (OR: 1.33; 95% CI: 0.80 –2.24; p ⫽ 0.28) or unstable angina (OR: 0.89; 95% CI: 0.60 –1.32; p ⫽ 0.58). We have also examined the question of association in the overall population of the study, with 2,000 cases and controls (10). Here, as well, we determined no associations with the 578 patients with a history of MI (OR: 0.82; 95% CI: 0.61–1.12; p ⫽ 0.13), the 91 with acute MI (OR: 1.47; 95% CI: 0.90 –2.41; p ⫽ 0.14), or the 144 with unstable angina (OR: 1.06; 95% CI: 0.69 –1.63; p ⫽ 0.74). Table 4. Relative Risk for the Composite End Point in Logistic Regression Analysis Variables FVII genotype (Arg/Arg vs. Arg/Gln ⫹ Gln/Gln) Gender Age Diabetes Hypercholesterolemia Hypertension Smoking Angina (unstable) Infarction (acute) No. of diseased vessels (1 vs. 3) No. of diseased vessels (2 vs. 3) Lesion type (A vs. C) Lesion type (B1 vs. C) Lesion type (B2 vs. C) Recanalization Reference vessel diameter Fibrinogen CrP PAI-1 vWF
RR 0.28 0.69 0.99 0.89 1.68 1.61 1.00 0.97 2.05 0.81 0.63 0.11 0.40 0.41 0.53 0.72 1.10 1.00 0.99 1.0035
95% CI
p Value
0.09–0.81
0.02
0.31–1.50 0.95–1.03 0.40–2.01 0.83–3.37 0.81–3.19 0.50–2.09 0.42–2.43 0.33–12.85 0.35–1.87 0.26–1.49 0.01–0.91 0.13–1.27 0.19–0.88 0.18–1.54 0.40–1.27 0.81–1.50 0.99–1.01 0.96–1.03 1.0009–1.0061
0.35 0.62 0.78 0.15 0.18 0.96 0.99 0.44 0.63 0.29 0.04 0.12 0.02 0.24 0.25 0.52 0.93 0.71 0.01
CrP ⫽ C-reactive protein; PAI-1 ⫽ plasminogen activator inhibitor; vWF ⫽ von Willebrand factor.
1524
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
DISCUSSION Findings. The present study provides first evidence that the Gln353 allele of coagulation factor VII is associated with substantial risk reduction (in our findings, by two thirds) following coronary catheter interventions. It was a crucial finding of this study that the risk reduction was fairly consistent across the component end points as well as across the three different types of catheter interventions. In the large sample size of 666 patients, 43 patients suffered from major complications during the first 30 days and reached the composite end point. This small absolute number of complications may explain why the reduction in risk achieved significance only in the context of our total patient population. Because we had hypothesized that the Arg353Gln polymorphism associated with a reduction of FVIIa could reduce the vascular events such as thrombosis, we were slightly surprised to find no reduction of risk for acute coronary syndromes or for chronic MI. Conversely, these results of ours did concur with case-control studies with large populations (18) as well as with the results of the prospective Framingham Study (19). Potential limitations. The surprising results for the group as a whole (i.e., the substantial reduction of procedural risk among carriers of the Gln353 allele) must be interpreted with the necessary circumspection. Discussion of the possibility of coincidence in the finding data, or of selection bias in the patient group, is necessary with respect to Gln353 allelic frequency and the complication rate. The following, however, speaks against the possibility of selection bias: the fact that the Gln353 allelic frequency of 12.5% disclosed in our study lies within the reported range for Middle European Caucasians (7,9). Moreover, the prevalence of the composite end point of 6.5% as determined by us corresponds to comparable figures achieved by others with consecutive patients (20). From the standpoint of work being performed today, our DCA frequency of 16% appears to be very high. This level may be justified by considering that our interventions took place basically in 1996, at the beginning of the stent era, and that this percent figure is in line with data published in a comparable registry (20). Possible mechanisms. During catheter interventions the protective endothelial lining of the arterial wall is disrupted, and such catheter-induced endothelial injury may trigger thrombogenesis. It has been evidenced by our findings (Table 2) that the Gln353 mutation possesses functional importance, suggesting a gene-dose effect that consists in a FVIIc reduction of 21% in heterozygous and of 36% in homozygous Gln353 carriers. An analogous phenomenon could be observed in FVIIa levels, even though the 49% reduction in Gln353 homozygotes did not reach significance. These phenotypical findings accord well with the literature (6 –9) and may explain the observed protective effect of the Gln353 allele. Because a ⫺323 nt 0/10-bp polymorphism in the pro-
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
moter region has been described as functionally relevant in lowering FVII levels (16), we examined this aspect in our study group. The almost complete disequilibrium between this promoter polymorphism and the Arg353Gln polymorphism determined in our study accords with findings of other groups (21). By contrast, these “within-gene” allelic interactions do not in the clinical context of our study allow differentiation between the effects of each single polymorphism on decreased FVII levels. Instead, we can represent the decreased levels only as cumulative effects. Although many factors influence FVII concentration and activity, presence of the Gln353 allele is not necessarily associated with lower FVII levels, as demonstrated in the group of Gln353 allele carriers presenting with complications. By contrast to functional variations of FVIIc or FVIIa levels, the FVII genotype is a constant predictor of the lifelong tendency of individual FVII activity. Our observation (i.e., risk reduction for thrombotic events) is in line with findings by Iacoviello and colleagues (6), who described a protective effect against MI in carriers of the Gln353 allele. By contrast, several studies failed to demonstrate any protection (18,22). Clinical implications. Our results could well have two clinically important implications: first, that FVII polymorphism has an impact on the development of catheter complications and may serve as a risk predictor. Second, with the perspective of active site-blocked activated FVII (FVIIai) (23) as conjunctive medication during catheter interventions, our findings suggest that the FVII genotype should be taken into due consideration in use, dosage, and assessment of FVIIai effectiveness. Acknowledgment This work was supported in part by a grant of the German Federal Ministry of Education, Science, Research, and Technology (No. 01 GG 9845-5). Reprint requests and correspondence: Dr. Karl Stangl, Medizinische Klinik, Kardiologie, Angiologie, Pneumologie, Charite´, Campus Mitte, Humboldt Universita¨t zu Berlin, Schumannstrasse 20/21, D-10117 Berlin, Germany. E-mail: [email protected].
REFERENCES 1. Furie B, Furie BC. The molecular basis of blood coagulation. Cell 1988;53:505–18. 2. Rapaport SI, Rao LVM. Initiation and regulation of tissue factordependent blood coagulation. Arteriosler Thromb 1992;12:1111–21. 3. Nemerson Y. Tissue factor: then and now. Thromb Haemost 1995; 74:180 – 4. 4. Rao LV, Rapaport SI. Cells and the activation of factor VII. Haemostasis 1996;26 Suppl 1:1–5. 5. Muller YA, Ultsch MH, Kelley RF, de Vos AM. Structure of the extracellular domain of human tissue factor: location of the factor VIIa binding site. Biochemistry 1994;33:10864 –70. 6. Iacoviello L, Di Castelnuovo A, de Knijff P, et al. Polymorphisms in the coagulation factor VII gene and the risk of myocardial infarction. N Engl J Med 1998;338:79 – 85. 7. Green F, Kelleher C, Wilkes H, Temple A, Meade T, Humphries S. A common genetic polymorphism associated with lower coagulation
Mrozikiewicz et al. Factor VII Mutation and Coronary Interventions
JACC Vol. 36, No. 5, 2000 November 1, 2000:1520–5
8.
9.
10.
11.
12. 13. 14.
15.
factor VII levels in healthy individuals. Arterioscler Thromb 1991;11: 540 – 6. Ghaddar HM, Folsom AR, Aleksic N, et al. Correlation of factor VIIa values with factor VII gene polymorphism, fasting and postprandial triglyceride levels, and subclinical carotid atherosclerosis. Circulation 1998;98:2815–21. Hanault M, Arbini AA, Lopaciuk S, Carew JA, Bauer KA. The Arg353Gln polymorphism reduces the level of coagulation factor VII. In vivo and in vitro studies. Arterioscler Thromb Vasc Biol 1997;17: 2825–9. Laule M, Cascorbi I, Stangl V, et al. A1/A2 polymorphism of glycoprotein IIIa and association with excess procedural risk for coronary catheter interventions: a case-controlled study. Lancet 1999; 353:708 –12. Nomenclature and criteria for diagnosis of ischemic heart disease. Report of the Joint International Society and Federation of Cardiology/World Health Organization Task Force on Standardization of Clinical Nomenclature. Circulation 1979;59:607–9. Simonton CA, Leon MB, Baim DS, et al. ‘Optimal’ directional coronary atherectomy: final results of the Optimal Atherectomy Restenosis Study (OARS). Circulation 1998;97:332–9. Marchetti G, Patracchini P, Papacchini M, Ferrati M, Bernardi F. A polymorphism in the 5⬘ region of coagulation factor VII gene (F7) caused by an inserted decanucleotide. Hum Genet 1993;90:575– 6. Morrissey JH, Macik BG, Neuenschwander PF, Comp PC. Quantitation of activated factor VII levels in plasma using a tissue factor mutant selectively deficient in promoting factor VII activation. Blood 1993;81:734 – 44. Hellstern P, Beeck H, Fellhauer A, Fischer A, Faller-Stockl B. Measurement of factor VII and of activated factor VII in healthy
16. 17. 18. 19.
20. 21.
22.
23.
1525
individuals and in prothrombin complex concentrates. Thromb Res 1997;86:493–504. Pollak ES, Hung HL, Godin W, Overton GC, High KA. Functional characterisation of the human factor VII 5⬘ flanking region. J Biol Chem 1996;271:1738 – 47. Walter DH, Schachinger V, Elsner M, Dimmeler S, Zeiher AM. Platelet glycoprotein IIIa polymorphisms and risk of coronary stent thrombosis. Lancet 1997;350:1217–9. Lane DA, Grant PJ. Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease. Blood 2000;95:1517–32. Feng D, Tofler GH, Karson MG, et al. Factor VII gene polymorphism, factor VII levels, and prevalent cardiovascular disease: the Framingham Heart Study. Arterioscler Thromb Vasc Biol 2000;20: 593– 600. King SB III, Yeh W, Holubkov R, et al. Balloon angioplasty versus new device intervention: clinical outcomes. A comparison of the NHLBI and NACI registries. J Am Coll Cardiol 1998;31:5558 – 66. Di Castelnuovo A, D’Orazio A, Amore C, Falanga A, Donati MB, lacoviello L. The decanucleotide insertion/deletion polymorphism in the promoter region of the coagulation factor VII gene and the risk of myocardial infarction. Thromb Res 2000;98:9 –17. Lane A, Green F, Scarabin PY, et al. Factor VII Arg/Gln353 polymorphism determines factor VII coagulant activity in patients with myocardial infarction (MI) and control subjects in Belfast and in France but is not a strong indicator of MI risk in the ECTIM study. Atherosclerosis 1996;119:119 –27. Golino P, Ragni M, Cirillo P, et al. Antithrombotic effects of recombinant human, active site-blocked factor VIIa in a rabbit model of recurrent arterial thrombosis. Circ Res 1998;82:39 – 46.