4
Differences in TIMI Frame Count Following Successful Reperfusion With Stenting or Percutaneous Transluminal Coronary Angioplasty for Acute Myocardial Infarction Martin E. Edep, MD, Erminia M. Guarneri, MD, Paul S. Teirstein, MD, Paul S. Phillips, MD, and David L. Brown, MD The Thrombolysis In Myocardial Infarction (TIMI) flow grade achieved in the infarct-related artery (IRA) during reperfusion therapy for acute myocardial infarction (AMI) is directly related to myocardial salvage. Recently, several series have demonstrated the safety of stenting in AMI and documented a larger postprocedure luminal diameter than that found at angioplasty, although no study has compared the effect of PTCA and stenting in AMI on flow characteristics of the IRA. The residual stenosis and the number of frames required to opacify standardized angiographic landmarks normalized for vessel length (corrected TIMI frame count) or compared with flow in a corresponding normal coronary artery (TIMI frame count index) were determined for the IRA of 39 patients who underwent angioplasty or stenting for AMI. Baseline characteristics were similar for the 20 patients who underwent stenting and the 19 patients who underwent percutaneous transluminal coronary an- gioplasty. After intervention, the luminal diameter was greater (3.24 vs 2.09 mm, p <0.0001) and the residual stenosis was less (29.4% vs. 26.7%, p <0.0001) after stenting than after percutaneous transluminal coronary angioplasty. These changes in vessel geometry were associated with a lower corrected TIMI frame count (16.1 vs 30.7, p <0.002) and a lower TIMI frame count index (0.68 vs 1.3, p <0.002). Thus, stenting in AMI is associated with a greater postprocedure luminal diam- eter and improvement in coronary blood flow as mea- sured by the TIMI frame count method. Q1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;83:1326 –1329) M yocardial salvage during acute myocardial in- farction (AMI) is directly related to the quality of coronary flow restored in the infarct-related artery (IRA) by reperfusion therapy. 1 The optimal method of reperfusion remains an area of active investigation. 2–5 Until recently, stenting in AMI was thought to be relatively contraindicated because of the potential risk for enhanced thrombus formation. Several small series have now reported on the safety of stenting in AMI and have documented a larger postprocedure luminal diameter than that found after angioplasty. 6 –10 How- ever, the added expense of coronary stenting will be difficult to justify if the cosmetic improvement in the vessel lumen does not translate into patient benefit. 11 To our knowledge, no study has compared the effect of percutaneous transluminal coronary angioplasty (PTCA) and stenting in AMI on flow in the IRA. In the present study, we tested the hypothesis that the greater luminal diameter associated with stenting would translate into improvements in coronary blood flow (as measured by the Thrombolysis In Myocardial Infarction [TIMI] trial frame count method) compared with PTCA. METHODS Patient selection: One hundred forty-one consecu- tive patients with AMI at the University of California San Diego Medical Center, Mercy Medical Center, or Green Hospital of Scripps Clinic and Research Foun- dation who underwent emergency mechanical revas- cularization were identified by review of the respec- tive interventional databases. From this group, a sub- set of 39 patients with successful reperfusion defined as TIMI 3 flow in the IRA at the end of the procedure and whose pre- and postintervention angiograms were suitable for analysis by the TIMI frame count method (see below) were identified. AMI was defined as an episode of ischemic chest discomfort associated with ST-segment elevation of .1.0 mm (0.1 mV) in $2 contiguous electrocardiographic leads or elevation in creatinine kinase .2 times the upper limit of normal and an MB fraction of .2.5%. Patients were separated into 2 groups on the basis of whether they had undergone PTCA alone or PTCA with stenting. In all cases only the IRA was treated. All patients received aspirin and nitroglycerin before the procedure. Systemic anticoagulation was achieved by an initial bolus dose of 10,000 U of intravenous heparin, with additional heparin given to maintain an activated clotting time of .300 seconds. No patient in From the University of California San Diego, San Diego; Green Hos- pital of the Scripps Clinic and Research Foundation, La Jolla; and Mercy Medical Center, San Diego, California. Manuscript received September 10, 1998; revised manuscript received and accepted January 5, 1999. Address for reprints: David L. Brown, MD, Division of Cardiovas- cular Medicine, Albert Einstein College of Medicine, 1825 Eastches- ter Road, Bronx, New York 10461. E-mail: dbrown@montefiore.org. 1326 ©1999 by Excerpta Medica, Inc. 0002-9149/99/$–see front matter All rights reserved. PII S0002-9149(99)00094-6

Differences in TIMI frame count following successful reperfusion with stenting or percutaneous transluminal coronary angioplasty for Acute Myocardial Infarction

Embed Size (px)

Citation preview

Differences in TIMI Frame CountFollowing Successful Reperfusion WithStenting or Percutaneous Transluminal

Coronary Angioplasty for AcuteMyocardial Infarction

Martin E. Edep, MD, Erminia M. Guarneri, MD, Paul S. Teirstein, MD,Paul S. Phillips, MD, and David L. Brown, MD

The Thrombolysis In Myocardial Infarction (TIMI) flowgrade achieved in the infarct-related artery (IRA) duringreperfusion therapy for acute myocardial infarction(AMI) is directly related to myocardial salvage. Recently,several series have demonstrated the safety of stentingin AMI and documented a larger postprocedure luminaldiameter than that found at angioplasty, although nostudy has compared the effect of PTCA and stenting inAMI on flow characteristics of the IRA. The residualstenosis and the number of frames required to opacifystandardized angiographic landmarks normalized forvessel length (corrected TIMI frame count) or comparedwith flow in a corresponding normal coronary artery(TIMI frame count index) were determined for the IRA of39 patients who underwent angioplasty or stenting forAMI. Baseline characteristics were similar for the 20

patients who underwent stenting and the 19 patientswho underwent percutaneous transluminal coronary an-gioplasty. After intervention, the luminal diameter wasgreater (3.24 vs 2.09 mm, p <0.0001) and the residualstenosis was less (29.4% vs. 26.7%, p <0.0001) afterstenting than after percutaneous transluminal coronaryangioplasty. These changes in vessel geometry wereassociated with a lower corrected TIMI frame count(16.1 vs 30.7, p <0.002) and a lower TIMI frame countindex (0.68 vs 1.3, p <0.002). Thus, stenting in AMI isassociated with a greater postprocedure luminal diam-eter and improvement in coronary blood flow as mea-sured by the TIMI frame count method. Q1999 byExcerpta Medica, Inc.

(Am J Cardiol 1999;83:1326–1329)

Myocardial salvage during acute myocardial in-farction (AMI) is directly related to the quality

of coronary flow restored in the infarct-related artery(IRA) by reperfusion therapy.1 The optimal method ofreperfusion remains an area of active investigation.2–5

Until recently, stenting in AMI was thought to berelatively contraindicated because of the potential riskfor enhanced thrombus formation. Several small serieshave now reported on the safety of stenting in AMIand have documented a larger postprocedure luminaldiameter than that found after angioplasty.6–10 How-ever, the added expense of coronary stenting will bedifficult to justify if the cosmetic improvement in thevessel lumen does not translate into patient benefit.11

To our knowledge, no study has compared the effectof percutaneous transluminal coronary angioplasty(PTCA) and stenting in AMI on flow in the IRA. Inthe present study, we tested the hypothesis that thegreater luminal diameter associated with stentingwould translate into improvements in coronary bloodflow (as measured by the Thrombolysis In Myocardial

Infarction [TIMI] trial frame count method) comparedwith PTCA.

METHODSPatient selection: One hundred forty-one consecu-

tive patients with AMI at the University of CaliforniaSan Diego Medical Center, Mercy Medical Center, orGreen Hospital of Scripps Clinic and Research Foun-dation who underwent emergency mechanical revas-cularization were identified by review of the respec-tive interventional databases. From this group, a sub-set of 39 patients with successful reperfusion definedas TIMI 3 flow in the IRA at the end of the procedureand whose pre- and postintervention angiograms weresuitable for analysis by the TIMI frame count method(see below) were identified. AMI was defined as anepisode of ischemic chest discomfort associated withST-segment elevation of.1.0 mm (0.1 mV) in$2contiguous electrocardiographic leads or elevation increatinine kinase.2 times the upper limit of normaland an MB fraction of.2.5%.

Patients were separated into 2 groups on the basisof whether they had undergone PTCA alone or PTCAwith stenting. In all cases only the IRA was treated.All patients received aspirin and nitroglycerin beforethe procedure. Systemic anticoagulation was achievedby an initial bolus dose of 10,000 U of intravenousheparin, with additional heparin given to maintain anactivated clotting time of.300 seconds. No patient in

From the University of California San Diego, San Diego; Green Hos-pital of the Scripps Clinic and Research Foundation, La Jolla; andMercy Medical Center, San Diego, California. Manuscript receivedSeptember 10, 1998; revised manuscript received and acceptedJanuary 5, 1999.

Address for reprints: David L. Brown, MD, Division of Cardiovas-cular Medicine, Albert Einstein College of Medicine, 1825 Eastches-ter Road, Bronx, New York 10461. E-mail: [email protected].

1326 ©1999 by Excerpta Medica, Inc. 0002-9149/99/$–see front matterAll rights reserved. PII S0002-9149(99)00094-6

the PTCA group received abciximab, whereas 1 pa-tient in the stent group was administered abciximab.PTCA was performed in the standard fashion, with thedilating balloon-to-reference diameter ratio not ex-ceeding 1.1:1. The final angiographic result was doc-umented in at least 2 separate views of the dilatedvessel. Stenting of the IRA was accomplished with theuse of Palmaz-Schatz coronary stents in most cases. Acoronary stent slightly larger than the reference diam-eter of the vessel to be stented was selected anddeployed using high-pressure balloon inflation. Theanticoagulation/antiplatelet regimen for patients re-ceiving stents included aspirin (100%), ticlopidine(93%), and/or warfarin (27%).

Angiographic analysis: Quantitative measurementswith the use of hand-held calipers were obtained in theoptimal, nonforeshortened, single-plane projectionthat identified the stenosis in its greatest severity.TIMI flow was determined using published criteria.12

Thrombolysis In Myocardial Infarction trial framecount : The TIMI frame was obtained for the IRAusing the technique described by Gibson et al.13 Inbrief, the TIMI frame count is the number of framesrequired for dye to first opacify a standard distallandmark. To be considered suitable for TIMI framecount analysis, a pre- and postintervention angiogramwas needed to demonstrate the initial entry of contrastinto the coronary artery and follow the contrast until itreached the prespecified distal landmark within thecoronary bed.

Most patients excluded from the study had angio-grams that failed to document entry of contrast intothe artery and arrival of contrast at the distal land-mark. The TIMI frame count of 64 angiographicallynormal coronary arteries from patients not having anischemic syndrome were obtained as controls. To cor-rect for the longer length of the left anterior descend-ing coronary artery compared with the right or leftcircumflex coronary arteries, when the IRA was theleft anterior descending coronary artery, the TIMIframe count of the left anterior descending coronaryartery was divided by the unadjusted TIMI framecount of the normal left anterior descending coronaryartery divided by the mean TIMI frame count of theright and left circumflex coronary arteries, yielding acorrected TIMI frame count. To express flow in theIRA relative to a corresponding normal coronary ar-tery, a TIMI frame count index, defined as the cor-rected TIMI frame count of the IRA divided by themean corrected TIMI frame count for the correspond-ing coronary artery in the control group was calcu-lated. A TIMI frame count index.1 would thereforesignify abnormally slow flow, and a TIMI frame countindex ,1 would indicate faster flow than normal(hyperemia).

Statistical analysis: Data are presented as themean6 SD. All analyses were performed using Excelversion 5.0 (Microsoft, Redmond, Washington). Com-parisons of continuous variables and categorical vari-ables between the 2 groups were calculated by theStudent’st test and chi-square test, respectively. A pvalue#0.05 was considered significant.

RESULTSPatient characteristics: Patient characteristics are

listed in Table I. Patients in the PTCA and stentgroups were similar in age and gender. Mechanicaltreatment was the primary modality in 95% of thestent group and in 83% of patients in PTCA group(p 5 NS). Mechanical treatment followed unsuccess-ful thrombolysis in 5% of stent patients and in 17% ofPTCA patients, respectively (p5 NS). There was nosignificant difference with respect to culprit vesseldistribution between the 2 groups. Because commonlyused angiographic views most often capture the originand distal aspects of the right coronary artery withoutpanning, the right coronary artery was the most com-mon culprit vessel in both groups.

Angiographic characteristics: The angiographic fea-tures of the 2 groups are listed in Table II. The meanreference vessel diameter (3.02 vs 3.45 mm; p5 NS),mean minimal luminal diameter (0.29 vs 0.23 mm;p 5 NS), and percent diameter stenosis (92% vs 93%;p 5 NS) of the culprit vessels were similar in the stentand PTCA patients before intervention. There was nodifference in TIMI flow distribution at baseline be-

TABLE I Clinical Characteristics

Stent(n 5 20)

Angioplasty(n 5 19)

Age (yr) 60.3 6 11 60.3 6 14Men 15 (75%) 12 (63%)Infarct-related artery

Left anterior descending 5 (25%) 4 (21%)Right 13 (65%) 13 (68%)Left circumflex 2 (10%) 2 (11%)

Chest pain time (min) 160 6 209 200 6 278Primary PTCA 95% 83%Rescue PTCA 5% 17%

TABLE II Angiographic Characteristics

Stent(n 5 20)

Angioplasty(n 5 19) p Value

Reference vesseldiameter (mm)

3.02 6 0.79 3.45 6 0.66 NS

PreprocedureMLD (mm)

0.29 6 0.55 0.23 6 0.38 NS

Preprocedurestenosis

92% 93% NS

PreprocedureTIMI flow

0 68% 63% NS1 0% 0% NS2 5.3% 5.3% NS3 26.3% 31.5% NS

PostprocedureMLD (mm)

3.24 6 0.82 2.09 6 0.68 ,0.0001

Postprocedurestenosis

29.4% 26.7% ,0.0001

PostprocedureTIMI 3 flow

100% 100% NS

Corrected TIMIframe count

16.1 6 6.7 30.7 6 16.8 0.0018

TIMI frame countindex

0.68 6 0.28 1.3 6 0.71 0.0017

MLD 5 minimal lumen diameter.

CORONARY ARTERY DISEASE/CORONARY BLOOD FLOW IN MYOCARDIAL INFARCTION 1327

tween the 2 groups. Sixty-eight percent of patientsundergoing stenting had occluded vessels at presenta-tion, whereas 63% in the PTCA group had TIMI grade0 flow (p5 NS). Twenty-six percent of patients in thestent group presented with TIMI grade 3 flow com-pared with 31.5% in the PTCA group (p5 NS).

By definition all patients in each group had TIMIgrade 3 flow restored in the culprit vessel after inter-vention. All patients in the stent group achieved a finalresidual stenosis of,50%, whereas 79% of patientsdid so in the PTCA group. Both groups experienced asignificant increase in minimal luminal diameter and asignificant decrease in residual stenosis. However, thepostprocedure improvement noted in the stent groupwas significantly greater than that in the PTCA group.The minimal luminal diameter after stent implantationwas 3.23 mm, whereas the minimal luminal diameterafter PTCA was 2.09 mm (p,0.0001). The percentresidual stenosis of29.4% observed in the stent groupwas significantly less than the 26.7% residual stenosisin the PTCA group (p,0.0001).

The baseline corrected TIMI frame counts of thenormal coronary arteries in control patients were24.96 in the left anterior descending artery, 22.52 inthe right coronary artery, and 27.4 in the left crcum-flex coronary artery. There were significant differ-ences in IRA flow between the 2 groups after inter-vention as measured by both the corrected TIMI framecount and the TIMI frame count index (Table II). Inthe stent group, the mean corrected TIMI frame countfor all IRAs was 16.1 compared with a mean correctedTIMI frame count of 30.7 in the PTCA group (p,0.002 ), indicating a more rapid flow rate (Table II).The TIMI frame count index demonstrated similarfindings with a TIMI frame count index of 0.68 in thestent group compared with a TIMI frame count indexof 1.3 in the PTCA group (p,0.002 ) (Table II).

DISCUSSIONIt is well established that flow is a key determinant

of myocardial salvage and survival in AMI, with re-duction in flow (i.e., TIMI grade#2) being associatedwith decreased regional and global left ventricularfunction, higher enzyme peaks, and increased morbid-ity and mortality.1,14–17Although the factors that gov-ern flow in the IRA after angioplasty are complex andinclude epicardial coronary artery luminal diameter,vasospasm, anterograde perfusion pressure, and distalrunoff, the use of stents offers several advantages overangioplasty alone. Stenting produces changes in ves-sel wall geometry which maximize acute gain andreduce immediate recoil resulting in decreased resid-ual stenosis.5 Additionally, stents tend to normalizecalculated resistance to flow and pressure gradientsacross stenoses via their smoothing effects.5 Whereasthe increased postprocedure luminal diameterachieved with stenting contributes to the decreasedrate of restenosis,3 these changes may additionallyimprove flow characteristics acutely. Since flow isproportional to the fourth power of the radius, the

larger lumen size obtained with stenting comparedwith angioplasty, coupled with decreased vascular re-sistance, may translate into improvements in coronaryflow.

In the present study the improvement in meanluminal diameter and residual stenosis achieved bystent placement appear to translate into improvementsin coronary blood flow, with the corrected TIMI framecount and TIMI frame count index in the stent groupbeing reduced (i.e., faster flow) compared with thePTCA group. In fact, as indicated by a TIMI framecount index of 0.68, stenting was associated withcoronary blood flow that was brisker than flow in acorresponding normal vessel. Whereas coronary bloodflow was also improved by primary PTCA, the flowrates achieved were still slower than normal.

In addition to changes in luminal diameter, stentingmay have a beneficial effect on distal microvascula-ture and coronary flow reserve in the IRA.6,7 Thelarger lumen associated with stenting may providegreater flow to the distal bed, resulting in less micro-vascular plugging, enhanced delivery of vasodilatorysubstances, and improved endothelial cell function.The combined effect of improvements in vessel ge-ometry at the site of the stenosis and enhancement ofdistal flow may account for the hyperemia observedwith stenting in the present study.

In the present study the time frame count methodwas used to quantitate flow in the IRA. This techniquehas been shown to be a more sensitive measurement ofcoronary blood flow than the more widely used TIMIflow grading system. The TIMI grading method islimited by several factors which include interobservervariability, the unimodal distribution of flow, and useof a potentially inaccurate standard to compare flowrates.13 The current study highlights one of thestrengths of the TIMI frame count method by docu-menting statistically significant differences in coro-nary blood flow among patients classified as havingequivalent TIMI flow.

Study limitations: There are several potential limi-tations of the current study that may influence theobserved results. First, this study is a nonrandomizedretrospective comparison of 2 cohorts of AMI pa-tients. Most PTCA patients had their procedure per-formed before 1994 when stenting became widelyavailable, whereas the stented group was treated morerecently. The technique of PTCA and the managementof patients with AMI may have improved during thestudy period and may bias the data so that the resultsof the PTCA group appear inferior to that of the stentgroup. An additional limitation is the potential biasingeffects of the angiographic requirements for framecounting. Last, the TIMI frame count is only a surro-gate for direct measurement of coronary blood flowand therefore subject to potential confounding vari-ables. Nevertheless, our results suggest that improve-ments in vessel wall geometry obtained by stent place-ment translate into improvements in coronary bloodflow as measured by the TIMI frame count method.

1328 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 83 MAY 1, 1999

1. The Gusto Angiographic Investigators. The effects of tissue plasminogenactivator, streptokinase or both on coronary artery patency, ventricular functionand survival after acute myocardial infarctionN Engl J Med1993;329:1615–1622.2. Every NR, Parsons LS, Hlatky M, Martin JS, Weaver WD. A comparison ofthrombolytic therapy with primary angioplasty for acute myocardial infarction.N Engl J Med1996;335:1253–1260.3. Lange RA, Hillis LD. Should thrombolysis or primary angioplasty be thetreatment of choice for acute myocardial infarction?N Engl J Med1996;335:1311–1312.4. Grines CL. Primary angioplasty—the strategy of choice.N Engl J Med1996;335:1313–1314.5. Garcia-Cantu E, Spaulding C, Thierry C, Khaldoun BH, Roussel L, FavereauX, Guerin Y, Chalet Y, Soucffrant G, Guerin F. Stent implantation in acutemyocardial infarction.Am J Cardiol1996;77:451–454.6. Rodriguez AE, Fernandez M, Santaera O, Larribau M, Bernardi V, Castano H,Palacios IF. Coronary stenting in patients undergoing percutaneous transluminalcoronary angioplasty during acute myocardial infarction.Am J Cardiol1996;77:685–689.7. Saito S, Hosokawa G, Kim K,Tanaka S, Miyake S. Primary stent implantationwithout coumadin in acute myocardial infarction.J Am Coll Cardiol1996;28:74–81.8. Le May MR, Labinaz M, Beanlands RS, Laramee LA, O’Brien ER, MarquisJ-F, Williams WL, Al-Sadoon K, Davies RF, Kearns SA, Johansen HL, Higgin-son LA. Usefulness of intracoronary stenting in acute myocardial infarction.Am JCardiol 1996;78:148–152.9. Antoniucci D, VaIenti R, Buonamici P, Santoro GM, Leoncini M, BologneseL, Fazzini PF. Direct angioplasty and stenting in the infarct related vessel in acutemyocardial infarction.Am J Cardiol1996;78:568–571.10. Schomig A, Neumann FJ, Walter H, Schuhlen H, Hadamitzky M, Zitzmann-

Roth EM, Dirschinger J, Hausleiter J, Blasini R, Schmitt C, Alt E, Kastrati A.Coronary stent placement in patients with acute myocardial infarction: a com-parison of clinical and angiographic outcome after randomization to antiplateletor anticoagulant therapy.J Am Coll Cardiol1997;29:28–34.11. Topol EJ, Nissen SE. Our preoccupation with coronary luminology. Thedisassociation between clinical and angiographic findings in ischemic heartdisease.Circulation 1995;92:2333–2342.12. The TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI)trial: phase I findings.N Engl J Med1984;33:523–530.13. Gibson CM, Cannon CP, Daley WL, Dodge JT, Alexander B, Marble SJ,McCabe BS, Raymond L, Fortin T, Poole WK, Braumwald E, for the TIMI 4Study Group. TIMI Frame Count: a quantitative method of assessing coronaryartery flow.Circulation 1996;93:879–888.14. Karagounis LA, Sorensen SG, Menlove RL, Moreno F, Anderson JL. DoesThrombolysis in Myocardial Infarction (TIMI) perfusion grade 2 represent amostly patient artery or a mostly occluded artery? Enzymatic and electrocardio-graphic evidence from the TEAM-2 study. J Am Coll Cardiol1992;19:1–10.15. Anderson JL, Karagounis LA, Becker LC, Sorensen SG, Menlove RL, for theTEAM-3 investigators. TIMI perfusion grade 3 but not grade 2 results inimproved outcome after thrombolysis for myocardial infarction. Ventriculo-graphic, enzymatic and electrocardiographic evidence from the TEAM-3 study.Circulation 1993;87:1829–1839.16. Vogt A, von Essen R, Tebbe U, Feuerer W, Appel KF, Neuhaus KL. Impactof early perfusion status of the infarct-related artery on short-term mortality afterthrombolysis for acute myocardial infarction: retrospective analysis of fourmulticenter studies.J Am Coll Cardiol1993;21:1391–1395.17. Anderson JL, Sorensen SG, Moreno FL, Hackworthy RA, Browne KF, DaleHT, Leya F, Dangoisse V, Eckerson HW, Marder VJ, and the TEAM-2 StudyInvestigators. Multicenter patency trial of intravenous anistreplase compared withstreptokinase in acute myocardial infarction.Circulation 1991;83:126–140.

CORONARY ARTERY DISEASE/CORONARY BLOOD FLOW IN MYOCARDIAL INFARCTION 1329