| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Cardiology, Hôpital Robert Debré, Reims, France;2 Department of Biochemistry, American Memorial Hospital, Hôpital Robert Debré, Reims, France;3 Department of Cardiology, HEGP, Paris, France;4 Unité d’Aide Méthodologique à la Recherche Clinique, Hôpital Maison Blanche, Reims, France;
aaddress correspondence to this author at: Department of Cardiology, Hôpital Robert Debré, Centre Hospitalier Universitaire, 51092 Reims Cedex, France; fax 33-3-2678-4132, e-mail camille.brasselet{at}wanadoo.fr
Plasma homocysteine is a modifiable cardiovascular risk factor related to the extent of both coronary and carotid atherosclerosis (1)(2)(3)(4). Plasma homocysteine has been shown to predict the occurrence of cardiac events and mortality in patients with coronary atherosclerosis (5)(6)(7). The predictive value of homocysteine on restenosis after percutaneous coronary intervention (PCI) has been debated (8)(9)(10). Recent evidence, however, indicated that the pre-PCI homocysteine plasma concentration was an independent predictor of death, nonfatal myocardial infarction (MI), and target lesion revascularization (11). Cardiac troponins provide prognostic information in patients with acute coronary syndrome (ACS) (12). Several studies have demonstrated that PCI induces MI as assessed by increases in cardiac troponins, particularly in the case of ACS (13)(14)(15). Furthermore, increased cardiac troponin concentrations after PCI are associated with poor clinical outcome (16)(17)(18)(19). We therefore hypothesized that the pre-PCI plasma homocysteine concentration could be related to the occurrence of MI after PCI, as assessed by changes in plasma cardiac troponin I (cTnI) concentration.
Consecutive admissions for nonemergency PCI were studied prospectively. All patients had a stenosis >70% in 1 or more coronary arteries. Two groups were examined: patients with stable angina (SA) pain and those with ACS. The SA pain group had myocardial ischemia during exercise stress testing (n = 29). The ACS group included patients admitted with unstable angina without a subsequent increase in troponin concentrations (n = 28) and those with a definite MI with a documented cTnI increase and electrocardiogram changes that had occurred 7–14 days previously (n = 39). Patients with inflammatory diseases, as well as those being treated with corticosteroids or nonsteroidal antiinflammatory or immunosuppressive drugs, were excluded to minimize potential bias. This study complies with the Declaration of Helsinki, and the protocol was approved by the local institutional ethics committee. Informed written consent was obtained from each patient. Pre-PCI medications, including aspirin, heparin, nitrates, calcium channel–blocking agents, and ß-adrenergic–blocking drugs, were maintained throughout the study. None of the patients had received long-term vitamin B supplementation before enrollment. No vitamin B supplementation was planned during the study, and none of the patients were supplemented in case of hyperhomocysteinemia at the time of the study. Procedures were performed with standard angioplasty techniques. Almost all of the patients were treated with stent implantation (n = 85).
Immediately before and 24 h after PCI, venous blood was collected under standard conditions into glass tubes containing an anticoagulant (lithium heparin). Plasma cTnI was measured on an AxSYM system (first generation; Abbott Diagnostics; CV <8.0% at 10 µg/L; manufacturer-reported reference values, <0.4 µg/L; detection limit of the assay, 0.4 µg/L). cTnI values <0.4 µg/L were stated as 0. The variation between pre- and post-PCI cTnI was defined as 
cTnI [(post-PCI cTnI) – (pre-PCI cTnI) = cTnI]. For homocysteine measurements, samples were collected in plain glass tubes, transported on ice, and assayed within 4 h after venipuncture. Homocysteine was measured by ion-exchange HPLC (Hitachi 8800; Roche Diagnostics; CV <4.3% at 10 µmol/L; reference values, <12 µmol/L). Plasma creatine kinase and creatinine were measured with an Hitachi 911 analyzer (Roche Diagnostics). Post-PCI MI was defined by a post-PCI cTnI value >2 µg/L when pre-PCI cTnI was <0.4 µg/L, a post-PCI cTnI value >3 µg/L when pre-PCI cTnI was 0.4–2 µg/L, or an increase >50% of the initial value when pre-PCI cTnI was >2 µg/L. When MI was defined biologically, the difference between post-PCI non–Q-wave MI and post-PCI Q-wave MI was defined as the occurrence of a new Q-wave on post-PCI electrocardiograms. Acute renal failure was defined by a post-PCI increase in plasma creatinine concentration >2-fold higher than the initial value when the post-PCI plasma creatinine concentration was >120 µmol/L (1.4 mg/dL).
Statistical analyses were performed with Statview 5.0 software (SAS Institute). Continuous variables were expressed as the mean (SD) or median value and minimum–maximum values depending on distribution. Differences among groups were assessed by unpaired Student t-test or Mann–Whitney U-test, as appropriate. Because only pre-PCI homocysteine followed a gaussian distribution, correlations were assessed by Spearman correlation coefficient. After we checked the absence of interaction, a variance–covariance analysis was applied to explain first post-PCI cTnI and then cTnI variation. Variables proposed for the model showed a significance of 
0.20 at univariate analysis. All variables were included in the model, and a descending stepwise procedure was performed. P <0.05 was considered statistically significant.
Ninety-six consecutive patients were prospectively included in the study [males, n = 73; mean (SD) age, 60.1 (12.3) years]. The distribution of risk factors was as follows: diabetes, n = 20 (20.9%); smoker, n = 60 (62.3%); hypertension, n = 36 (37.5%); and hypercholesterolemia, n = 61 (63.5%). The procedural characteristics were as follows: scopy duration, 8.9 (6.7) min; procedure duration, 28.8 (18.9) min; mean treated lesions per patient, 1.3 (0.6); stents per patient, 1.1 (0.8); and area covered by the stent, 137.9 (96.8) mm2.
The biomarker findings are described in Fig. 1
. Baseline plasma homocysteine concentrations did not correlate with pre-PCI cTnI (r = 0.04; P = 0.65). In contrast, baseline plasma homocysteine concentrations correlated with both post-PCI cTnI and cTnI (r = 0.24; P = 0.018 and r = 0.33; P = 0.0008, respectively). Pre-PCI plasma homocysteine concentrations did not correlate with pre-PCI (r = –0.06; P = 0.56) and post-PCI (r = 0.10; P = 0.32) plasma creatine kinase concentrations. By multivariate analysis, the variables reaching statistical significance as independent predictors of post-PCI plasma cTnI concentrations were baseline homocysteine (P <0.0001) and pre-PCI cTnI (P <0.0001), with R2 = 0.76 (Table 1
). Moreover, when we considered cTnI after multivariate analysis, only baseline homocysteine (P <0.0001) and pre-PCI cTnI (P <0.0001) concentrations reached significance (R2 = 0.36). Post-PCI cTnI, cTnI, and the overall procedural data were not significantly different between the SA and ACS subgroups. For the whole population, the relationship between pre-PCI homocysteine and post-PCI cTnI or cTnI was particularly strong for the SA group (r = 0.83; P <0.0001 and r = 0.83; P <0.0001 for post-PCI cTnI and cTnI, respectively), whereas it was not strong in the ACS group. The baseline plasma homocysteine concentration was significantly higher in patients with post-PCI MI (n = 4), as defined previously, or acute renal failure after PCI (n = 8) than in patients without such a complication. Three patients had a new Q-wave MI after PCI [mean post-PCI cTnI, 107.0 (45.4) µg/L].
|
|
Our findings show that the baseline plasma homocysteine concentration strongly correlates with post-PCI markers of myocardial injury, whereas the baseline plasma cTnI concentration remains independent from the baseline plasma homocysteine concentration. This study was first designed to investigate whether pre-PCI homocysteine might influence the occurrence of MI after PCI, without previous clinical considerations. We found a correlation between homocysteine and post-PCI cTnI in the overall population, but this relationship was particularly strong in SA patients only because they are a homogeneous population. In contrast, the ACS group included patients with different evolving patterns of pre-PCI cTnI, which might have masked part of the occurring post-PCI MI. This finding suggests that homocysteine could be involved in PCI-induced MI, mainly in SA. Although the findings of this study must be considered limited in light of the small study size, we hypothesize that the pre-PCI plasma homocysteine concentration is able to predict the occurrence of MI in a population with lower risk because it is the only variable that remained statistically significant in multivariate analysis. In contrast, in ACS, other variables are involved and decrease the prothrombogenetic effect of homocysteine. In this study, only a few variables predicted an MI after PCI (i.e., baseline homocysteine and pre-PCI cTnI). None of the other variables correlated with post-PCI cTnI. Apparently, little is known with regard to risk markers identifying patients classically considered a "low-risk" population defined by pre-PCI troponin concentrations within the reference interval. Hamm and coworkers (20)(21) demonstrated the value of troponin measurements as strong and independent predictors of cardiac events in patients with acute chest pain, but in the setting of PCI, there is still a lack of markers to predict the occurrence of PCI-induced MI. Homocysteine, which has recently been described as an independent predictor of late outcome, might be an independent predictor of acute MI after coronary angioplasty (11).
Two different mechanisms may explain this finding. The first involves the prothrombogenetic effects of homocysteine via several pathways: promotion of endothelial dysfunction, formation of thromboxane A2, enhancement of platelet aggregation, reduction in the protective effect of nitric oxide, and the procoagulant effects (22)(23)(24)(25). These effects enhance thrombosis during PCI and promote the formation and migration of microthrombi, leading to MI. The second hypothesis takes into account the potential involvement of homocysteine in oxidative reactions. Indeed, it has been suggested that hyperhomocysteinemia may promote the production of hydroxyl radicals, known to be lipid peroxidation initiators, via homocysteine autooxidation and thiolactone formation (24)(26)(27). During transitory but complete coronary artery occlusion during PCI mimicking an ischemia–reperfusion model, the release of reactive oxygen species is important for the induction of cell death and the release of cTnI (28). Thus, hyperhomocysteinemia may enhance the generation of reactive oxygen species during PCI and presumably promote the death of myocardial cells in this model of ischemia–reperfusion. Further studies are needed to verify these 2 suspected mechanisms of damage.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
