|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2006; 48:1326-1335, doi:10.1016/j.jacc.2006.03.064
(Published online 12 September 2006). © 2006 by the American College of Cardiology Foundation |
* Pitié-Salpêtrière University Hospital, Paris, France
University of Ottawa Heart Institute, Ottawa, Ontario, Canada
University Hospital of Leicester, Leicester, United Kingdom.
Manuscript received December 13, 2005; revised manuscript received February 27, 2006, accepted March 16, 2006.
* Reprint requests and correspondence: Dr. Gilles Montalescot, Institut de Cardiologie, Bureau 2-236, Centre Hospitalier Universitaire Pitié-Salpêtrière, 47 Boulevard de l’Hôpital, 75013 Paris, France. (Email: Gilles.montalescot{at}psl.aphp.fr).
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
24 h) PCI versus delayed or ischemia-guided PCI; 3) fibrinolysis-facilitated PCI versus primary PCI alone. BACKGROUND: The impact of PCI strategies after fibrinolysis on mortality or reinfarction remains to be established.
METHODS: The meta-analysis was performed using the odds ratio (OR) as the parameter of efficacy with a random effect model. Fifteen randomized trials (5,253 patients) were selected. The primary end point was mortality or the combined end point of death or reinfarction.
RESULTS: Rescue PCI for failed fibrinolysis reduced mortality (6.9% vs. 10.7%) (OR, 0.63; 95% confidence interval [CI], 0.39 to 0.99; p = 0.055) and the rate of death or reinfarction (10.8% vs. 16.8%) (OR, 0.60; 95% CI, 0.41 to 0.89; p = 0.012) compared with a conservative approach. Systematic and early PCI performed during the "stent era" led to a nonsignificant reduction in mortality compared with delayed or ischemia-guided PCI (3.8% vs. 6.7%) (OR, 0.56; 95% CI, 0.29 to 1.05; p = 0.07) and to a 2-fold reduction in the rate of death or reinfarction (7.5% vs. 13.2%) (OR, 0.53; 95% CI, 0.33 to 0.83; p = 0.0067). This benefit contrasted with a nonsignificant increase in the rate of both mortality (5.5% vs. 3.9%, p = 0.33) or death or reinfarction (9.6% vs. 5.7%, p = 0.06) observed in the "balloon era." Fibrinolysis-facilitated PCI was associated with more reinfarction as compared with primary PCI alone (5.0% vs. 3.0%) (OR, 1.68; 95% CI, 1.12 to 2.51; p = 0.013) without significant impact on mortality (OR, 1.30; 95% CI, 0.92 to 1.83; p = 0.13).
CONCLUSIONS: Our findings support rescue PCI and systematic and early PCI after fibrinolysis. However, the current data do not support fibrinolysis-facilitated PCI in lieu of primary PCI alone.
| ||||||||
Primary percutaneous coronary intervention (PCI) is now the reperfusion strategy of choice in STEMI because it is superior to fibrinolysis in reducing the rate of death regardless of time from symptom onset (2). Even if transfer to a PCI center is necessary, primary PCI performed in a timely fashion remains superior to fibrinolysis (3). However, access to a catheterization laboratory within an acceptable time is the main limitation of this strategy (4–6). Therefore, the combination of fibrinolysis and mechanical reperfusion appears logical. Obviously, there is a need to assess the role of PCI done at different time intervals after fibrinolysis and whether the approach should be selective or systematic.
Currently, there are 3 different time-related PCI strategies that can be applied after initiation of fibrinolysis. Rescue PCI is attempted when there is failure of fibrinolysis, usually documented by ongoing chest pain and/or persistent ST-segment elevation at 60 to 90 min after initiation of fibrinolytic therapy. The second strategy consists of a systematic and early (24 h) PCI approach after administration of fibrinolysis irrespective of the latter’s success rather than the traditional conservative approach of delayed and/or ischemia-guided PCI. The third strategy of fibrinolysis-facilitated PCI involves the administration of fibrinolytic therapy to improve flow in the infarct-related artery before and/or during the transfer for PCI.
We have performed 3 separate meta-analyses with all studies available to assess mortality and reinfarction benefits of each strategy.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Study objective and design. Our objectives were to assess the relative benefits of the following different strategies in patients with STEMI: 1) for failed fibrinolysis: rescue PCI versus a conservative treatment; 2) after fibrinolysis regardless of success: systematic and early (within 24 h) PCI versus delayed and/or ischemia-guided PCI; and 3) fibrinolysis-facilitated PCI versus primary PCI alone.
Definitions of the characteristics of the trials selected for the meta-analysis, including definitions of STEMI, fibrinolysis failure, and PCI delays, were those of the trials concerned. It was recognized that there would be heterogeneity in the studies such as symptom onset to fibrinolysis, and fibrinolysis to PCI, as well as adjunctive pharmacologic agents and devices used. To minimize this issue, we selected only studies that included patients with STEMI 12 h of symptom onset and studies that provided data on death and reinfarction. In addition, "balloon era" studies and "stent era" studies were analyzed separately to assess whether the improvement of PCI techniques may have improved the outcomes of the post-fibrinolysis PCI strategies.
For rescue PCI, we restricted our analysis to studies that attempted to perform PCI <12 h after failed fibrinolytic therapy. There were no exclusion criteria regarding the type of fibrinolytic agent used and the type of PCI. Failed fibrinolysis was defined: 1) clinically as failure of ST-segment resolution with persistent chest pain; or 2) angiographically as Thrombolysis In Myocardial Infarction (TIMI) flow grade 0 to 1 within the infarct-related artery. In that respect, the RESCUE (Randomized Evaluation of Salvage Angioplasty with Combined Utilization of Endpoints) II study was excluded because only patients with TIMI flow grade 2 were considered for randomization (8).
Systematic and early intervention after fibrinolytic treatment was defined as catheterization performed 24 h of fibrinolytic therapy irrespective of its success. We selected trials with fibrinolysis being the initial mode of reperfusion in all patients with further randomization to early and systematic catheterization <24 h or delayed ischemic-driven revascularization. We excluded studies with a systematic invasive strategy delayed by >24 h after fibrinolysis (6,9–13). Studies that compared delayed PCI to no PCI after fibrinolysis were also excluded, being considered as obsolete strategies (14–18). There were no exclusion criteria regarding the type of fibrinolytic agent and the type of PCI.
When evaluating fibrinolysis-facilitated PCI, we selected studies with PCI being the main mode of reperfusion in all patients, with a randomization process evaluating fibrinolysis as facilitating agent during transfer to the catheterization laboratory. Studies using combined-treatments facilitation with reduced dose of lytics + full dose of glycoprotein (GP) IIb/IIIa inhibitors were not considered. The delay between initiation of fibrinolysis and PCI had to be <6 h. Studies that did not report death and reinfarction rates were excluded (19).
End points and definitions. The primary end point was mortality. The combined end point of death or reinfarction was also examined. When complete data were not available, the principal investigators of the studies were contacted. End points were evaluated at the longest available follow-up. The definition of reinfarction was different in each trial, and thus we decided to use the trial-specific definition of reinfarction. Major bleeding, primarily a safety end point, was defined differently among trials. Again, the trial-specific definition was retained.
Data management. Two authors (J-P.C., M.B.) independently reviewed abstracts and published manuscripts. A total of 15 nonoverlapping studies met the inclusion criteria. Independent data extraction was performed by the 2 reviewers and confirmed by consensus.
Data analysis. The results from each trial were those obtained on an intention-to-treat basis. The meta-analysis was performed using the odds ratio (OR) as the parameter of efficacy with a random effect model (Mantel-Haenszel), with appropriate tests for association and heterogeneity. The p value for significance of association and heterogeneity tests was set at 0.1 as suggested (20). Meta-analysis calculation (association test) and heterogeneity were performed with the EasyMa software (INSERM, University Claude Bernard, Lyon, France) (21).
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
12 h of symptom onset in all trials. Stents and GP IIb/IIIa receptor antagonists were used only in the 2 most recent trials (22–24). Patients who were readministered fibrinolytic therapy after failed fibrinolysis were not considered in the present meta-analysis because our primary attempt was to compare rescue PCI with a conservative strategy including no reperfusion therapies of any kind.
|
Within the 30 days of follow-up, there was a nonsignificant reduction in mortality in the rescue PCI group as compared with the group treated conservatively in either the "balloon era" studies and in the "stent era" studies (Fig. 2A). Combining all the trials, there was a 37% reduction in mortality in the rescue PCI group as compared with the group treated conservatively (p = 0.055) (Fig. 2A). At longest-term follow-up (up to 1 year), there was a nonsignificant reduction in the rate of death in the rescue group as compared with the conservative group (41 of 462 vs. 55 of 458) (OR, 0.69; 95% confidence interval [CI], 0.41 to 1.57; p = 0.16).
|
Rescue PCI was associated with major bleeding in 11.9% (44 of 368) of patients as compared with 1.3% (5 of 373) in the group treated conservatively (OR, 9.05; 95% CI, 3.71 to 22.06; p < 0.001). In patients treated with rescue PCI, major bleeding was associated with the femoral sheath used for catheterization in 82% (22 of 27), and none were fatal. There was 1 retroperitoneal bleed (transfused) and 4 gastrointestinal bleeds (2 transfused) without sequelae. Of interest, 69% of these rescue PCI patients had received abciximab compared with 43% overall.
All the analyses for heterogeneity were nonsignificant for all the end points in the preceding text.
Systematic and early angioplasty after fibrinolysis. Six randomized trials were identified for the comparison of a systematic and early catheterization versus delayed and/or ischemia-guided catheterization after fibrinolysis (n = 1,508 patients) (Table 2). Three of these trials were performed in the early days of mechanical intervention (13,28,29).
|
12 h. After initiation of fibrinolysis, systematic and early PCI was performed 6 h in 5 studies (13,28–31). In the stent era studies, PCI took place at least 90 min after fibrinolysis initiation in all trials and 
3 h in the 2 trials (Table 2). The pharmacologic approaches were different between trials of the "balloon era" (13,28,29) and those of the "stent era" (30–32). During the "balloon era," aspirin was not given immediately to all patients, and thienopyridines were not available. In the stent era, aspirin and thienopyridines were given to >80% of patients treated with early PCI, and GP IIb/IIIa receptor antagonists were used in 10% to 30% of these patients.
Mortality and reinfarction data were available within the first 6 weeks in all studies, at 6 months in 2 studies (30,31), and at 1 year in 1 study (32). The composite end point of death or reinfarction was consistent throughout the studies with few differences between trials. Data on major bleeding were available only in contemporary trials. At late follow-up, the death or reinfarction rate in the control arm (conservative approach) did not differ between trials of the "balloon era" (7.8%) and trials of the "stent era" (10.1%).
The systematic and early PCI strategy in the "stent era" (n = 832) was associated with a 44% reduction of death (p = 0.07); however, it was not significant (Fig. 3A). This benefit was obtained without significant excess of major bleeding (OR, 1.18; 95% CI, 0.60 to 2.30; p = 0.64). The rate of death or reinfarction in the contemporary trials was 2-fold lower in the systematic and early PCI group than in the delayed or ischemia-guided PCI group (7.5% vs. 13.2%, respectively, p = 0.0067). In contrast, the meta-analysis of trials done in the "balloon era" (n = 675) showed a nonsignificant 44% increase of mortality in the systematic and early PCI group as compared with the delayed or ischemia-guided PCI group (Fig. 3A). Similarly, there was a 76% increase in death or reinfarction in the systematic and early PCI group as compared with the delayed or ischemia-guided PCI group (9.6% vs. 5.7%, p = 0.064). When combining balloon and stent trials all together, there was a nonsignificant trend for a benefit for a systematic and early PCI as compared with a more conservative strategy.
|
Facilitated PCI with fibrinolytic therapy. Four trials were identified, representing a total of 2,679 patients (Table 3) (6,33–35). The end points of death and reinfarction were evaluated separately because the combined end point was not available in any of the selected studies. The overall death rate was 5.4%. Fibrinolysis-facilitated PCI was associated with a nonsignificant increase in the rate of death at 90 days (Fig. 4A). In addition, there was a 68% increase in the rate of reinfarction in the fibrinolysis-facilitated PCI group (5.0%) as compared with the primary PCI group (3.0%, p = 0.013) (Fig. 4B). There was no significant increase in the rate of major bleeding in the fibrinolysis-facilitated PCI (6.50% [80 of 1,230]) as compared with primary PCI (4.98% [62 of 1,243]) (OR, 1.23; 95% CI, 0.74 to 2.05; p = 0.42).
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The identification and management of failed fibrinolysis has been the subject of much debate (36,37). A conservative approach has often been considered a pragmatic option, particularly in hospitals without interventional facilities (38). Rescue PCI has also been shown to be feasible and safe more than 10 years ago, but the benefit has been questioned (39). A pooled analysis of randomized trials comparing rescue PCI to medical therapy during the "balloon era" (25–27) suggested a reduction in severe heart failure but remained inconclusive for death or reinfarction (8) and did not provide clear guidance on whether rescue PCI should be systematically performed in patients with failed fibrinolysis. Two recent studies (22–24) have readdressed this issue given the potential benefits of coronary stenting and new pharmacologic adjuncts. However, none of these studies were powered to assess the impact of rescue PCI on mortality.
The first major finding of our meta-analysis is that "rescue angioplasty" after failed fibrinolysis improves survival as compared with a conservative approach. The mortality benefit is mainly driven by the contemporary studies. These studies have chosen more precise inclusion criteria such as failure of ST-segment resolution, which can identify easily most failures with low false positives for patent arteries (40) and with a good correlation to short-term outcome (41). In addition, stent placement and concomitant administration of GP IIb/IIIa receptor antagonists may have greatly improved the results of rescue PCI. Indeed, most of the benefit on mortality was driven by REACT (Rescue Angioplasty Versus Conservative Therapy or Repeat Thrombolysis Trial), in which 69% of patients received stent and 43% GP IIb/IIIa receptor antagonists. In the MERLIN (Middlesbrough Early Revascularisation to Limit Infarction) trial, PCI slightly increased survival but clearly improved event-free survival, mainly subsequent revascularization. Only 50% of patients underwent stent placement, and 3% were given IIb/IIIa inhibitors in this trial. As a consequence, there was no reduction in reinfarction as opposed to the REACT trial.
Systematic and early (24 h) PCI approach after administration of fibrinolysis irrespective of the latter’s success rather than the traditional conservative approach of delayed and/or ischemia-guided PCI remains a controversial issue. Randomized trials performed during the "balloon era" showed no benefit of a systematic mechanical strategy on hard clinical end points (13,28,29,39). This strategy was progressively abandoned because of disappointing results and unacceptable bleeding. Contemporary studies have recently reassessed this hypothesis but were too small to examine mortality (30–32).
The present meta-analysis shows that early stent-PCI after fibrinolysis is associated with a better survival and a significant reduction of death or reinfarction of STEMI patients as compared with a delayed or ischemia-guided PCI strategy. These new findings draw the attention to the need for an appropriate network organization with adequate ambulance systems, pre-hospital management, and well-organized catheterization laboratories as previously outlined for transfer for primary PCI (3,42). Another important point is that PCI in the stent era trials was attempted with a delay of several hours after fibrinolysis initiation (Table 2). Up to 60% of patients randomized in the early PCI strategy presented with TIMI flow grade 3 irrespective of the time delay from fibrinolysis to angiography. Our data support the idea that time delay required for transfer to the catheterization laboratory after fibrinolysis may not be as critical as for primary PCI alone. Therefore, systematic catheterization within 24 h after fibrinolysis may be feasible in the majority of patients.
The benefits of these 2 PCI strategies after fibrinolysis were mainly observed in contemporary studies. It underscores the benefit of stenting in STEMI but also the critical impact of adjunctive pharmacology as stents have never been shown to reduce mortality in the setting of primary PCI. Although GP IIb/IIIa inhibitors in combination with stenting in primary PCI may save lives, their use was highly variable and overall low in the fibrinolytic studies chosen for this meta-analysis (43–45). This reflects the lack of safety data on the use of GP IIb/IIIa inhibitors for post-fibrinolysis PCI (43). However, the expected increase in major bleeding observed with rescue PCI was offset by the mortality benefit. The vast majority of bleeding events was located at the arterial access site and could be avoided with a larger use of the radial approach (46). As opposed to rescue PCI, there was no excess of major bleeding in the early systematic PCI strategy after lysis. This may be due to the longer delay between lysis and PCI as compared with rescue PCI. Given the impressive benefit of rescue PCI and the difficult task of identifying failed fibrinolysis, immediate and systematic routine angiography in patients with large myocardial infarctions appears to be a reasonable recommendation.
Facilitated PCI is defined as a pharmacologic reperfusion treatment administered before primary PCI to bridge the delay between first medical contact and mechanical reperfusion (47–49). Only 3 randomized studies have evaluated this strategy against primary PCI alone (33–35), and the efficacy of this concept remains unproven. Our analysis did not find any benefit of this strategy with a significant increase of early infarction. Full-dose fibrinolysis was used for facilitation except for PACT (Plasminogen Activator-Angioplasty Compatibility Trial) (34). In the GRACIA-2 (Primary Optimal Percutaneous Coronary Intervention versus Facilitated Intervention in STEMI patients) trial, there was also an important difference in the time delay between randomization and catheterization between the 2 groups of patients (32). The recent premature interruption of the ASSENT-4 (Assessment of the Safety and Efficacy of a New Thrombolytic Regimen-4) trial (35), opposing tenecteplase (TNK)-facilitated PCI to primary PCI, goes, however, in the same direction. The significant excess of reinfarction related to early stent thrombosis in the TNK-facilitated group suggests a deleterious prothrombotic effect of fibrinolysis given immediately before PCI. This further supports that fibrinolysis is not the ideal facilitating agent for primary PCI. Glycoprotein IIb/IIIa receptor antagonists reduce ischemic events including mortality when used with primary PCI and in particular when given as facilitating agents during transfer to the catheterization laboratory (43–45,50–52). The ongoing FINESSE (Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events) trial will bring more information on this specific issue.
Study limitations. Definitions of end points and the time at which these were recorded may vary in the studies included for the meta-analysis. However, short of large randomized trials adequately powered to look at end points such as mortality, meta-analyses such as this one provides an alternative to examine these end points. Overlap of time intervals may exist between studies subgrouped into early PCI and facilitated PCI, which represents another potential limitation. However, categorization of studies according to the type of strategy rather than time window may be more clinically relevant. Finally, the different length of follow-up among studies is another limitation that cannot be addressed in the present meta-analysis.
In conclusion, our results support the concepts of rescue PCI for failed fibrinolysis and early (<24 h) PCI regardless of the success achieved after fibrinolysis. The safety of such approaches and the benefit of GP IIb/IIIa inhibitors in this context warrant additional studies. Alternatively, fibrinolysis does not appear to facilitate PCI for STEMI (53,54).
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. G. Goodman, V. Menon, C. P. Cannon, G. Steg, E. M. Ohman, and R. A. Harrington Acute ST-Segment Elevation Myocardial Infarction: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 708S - 775S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. McDonald, Y. Fu, U. Zeymer, G. Wagner, S. G. Goodman, A. Ross, C. B. Granger, F. Van de Werf, P. W. Armstrong, and for the ASSENT-4 PCI Investigators Adverse outcomes in fibrinolytic-based facilitated percutaneous coronary intervention: insights from the ASSENT-4 PCI electrocardiographic substudy Eur. Heart J., April 1, 2008; 29(7): 871 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 American College of Cardiology/American Heart Asso, Developed in Collaboration With the Canadian Cardi, Endorsed by the American Academy of Family Physici, 2007 Writing Group to Review New Evidence and Upda, E. M. Antman, M. Hand, P. W. Armstrong, E. R. Bates, L. A. Green, L. K. Halasyamani, et al. 2007 Focused Update of the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction J. Am. Coll. Cardiol., January 15, 2008; 51(2): 210 - 247. [Full Text] [PDF]
|
|
|
|
|
|
American College of Cardiology/American Heart Asso, 2007 Writing Group to Review New Evidence and Upda, S. B. King III, S. C. Smith Jr, J. W. Hirshfeld Jr, A. K. Jacobs, D. A. Morrison, and D. O. Williams 2007 Focused Update of the ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention J. Am. Coll. Cardiol., January 15, 2008; 51(2): 172 - 209. [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Antman, M. Hand, P. W. Armstrong, E. R. Bates, L. A. Green, L. K. Halasyamani, J. S. Hochman, H. M. Krumholz, G. A. Lamas, C. J. Mullany, et al. 2007 Focused Update of the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines: Developed in Collaboration With the Canadian Cardiovascular Society Endorsed by the American Academy of Family Physicians: 2007 Writing Group to Review New Evidence and Update the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction, Writing on Behalf of the 2004 Writing Committee Circulation, January 15, 2008; 117(2): 296 - 329. [Full Text] [PDF]
|
|
|
|
|
|
S. B. King III, S. C. Smith Jr, J. W. Hirshfeld Jr, A. K. Jacobs, D. A. Morrison, D. O. Williams, 2005 WRITING COMMITTEE MEMBERS, S. C. Smith Jr, T. E. Feldman, J. W. Hirshfeld Jr, et al. 2007 Focused Update of the ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines: 2007 Writing Group to Review New Evidence and Update the ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention, Writing on Behalf of the 2005 Writing Committee Circulation, January 15, 2008; 117(2): 261 - 295. [Full Text] [PDF]
|
|
|
|
 |
|
 J B Jones and A Docherty Non-invasive treatment of ST elevation myocardial infarction Postgrad. Med. J., December 1, 2007; 83(986): 725 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Dixon, C. L. Grines, and W. W. O'Neill The Year in Interventional Cardiology J. Am. Coll. Cardiol., July 17, 2007; 50(3): 270 - 285. [Full Text] [PDF]
|
|
|
|
|
|
T. J. Kiernan, H. H. Ting, and B. J. Gersh Facilitated percutaneous coronary intervention: current concepts, promises, and pitfalls Eur. Heart J., July 1, 2007; 28(13): 1545 - 1553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Rihal, A. S. Jaffe, D. R. Holmes Jr, H. H. Ting, B. J. Gersh, and M. R. Bell Percutaneous Coronary Intervention vs Thrombolysis for ST-Elevation Myocardial Infarction JAMA, March 28, 2007; 297(12): 1313 - 1313. [Full Text] [PDF]
|
|
|
|
|
|
U. Stenestrand, J. Lindback, and L. Wallentin Percutaneous Coronary Intervention vs Thrombolysis for ST-Elevation Myocardial Infarction--Reply JAMA, March 28, 2007; 297(12): 1314 - 1315. [Full Text] [PDF]
|
|
|
|
|
|
J. E. Tcheng and K. G. Kinney After Fibrinolysis: Is Cardiac Catheterization the Answer? J. Am. Coll. Cardiol., October 3, 2006; 48(7): 1336 - 1338. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
