ORIGINAL CONTRIBUTION
`
`Eugene Braunwald, MDPATIENTS PRESENTING WITH AN
`
`acute coronary syndrome with-
`out ST-segment elevation are di-
`agnosed as having unstable an-
`gina/non–ST elevation myocardial
`infarction (MI) (UA/NSTEMI). Given
`the heterogeneous nature of UA/
`NSTEMI, such patients have a wide
`spectrum of risk for death and cardiac
`ischemic events.1-5 Many attempts to es-
`timate a gradient of risk among pa-
`tients with UA/NSTEMI focus on a single
`variable, such as presence or absence of
`electrocardiographic (ECG) changes6-9
`or elevated serum cardiac markers.10-13
`Prognostication schemes have been
`developed that categorize patients quali-
`tatively into high, intermediate, or low
`risk, but they do not provide a quanti-
`tative statement about finer grada-
`tions of risk that exist clinically.2 Al-
`though univariate analyses are
`informative as an initial assessment of
`the importance of a potential prognos-
`
`See also p 876 and Patient Page.
`
`The TIMI Risk Score for
`Unstable Angina/Non–ST Elevation MI
`A Method for Prognostication
`and Therapeutic Decision Making
`Elliott M. Antman, MD
`Marc Cohen, MD
`Peter J. L. M. Bernink, MD
`Carolyn H. McCabe, BS
`Thomas Horacek, MD
`Gary Papuchis, MD
`Branco Mautner, MD
`Ramon Corbalan, MD
`David Radley, MS
`
`Context Patients with unstable angina/non–ST-segment elevation myocardial in-
`farction (MI) (UA/NSTEMI) present with a wide spectrum of risk for death and car-
`diac ischemic events.
`Objective To develop a simple risk score that has broad applicability, is easily cal-
`culated at patient presentation, does not require a computer, and identifies patients
`with different responses to treatments for UA/NSTEMI.
`Design, Setting, and Patients Two phase 3, international, randomized, double-
`blind trials (the Thrombolysis in Myocardial Infarction [TIMI] 11B trial [August 1996–
`March 1998] and the Efficacy and Safety of Subcutaneous Enoxaparin in Unstable An-
`gina and Non-Q-Wave MI trial [ESSENCE; October 1994–May 1996]). A total of 1957
`patients with UA/NSTEMI were assigned to receive unfractionated heparin (test co-
`hort) and 1953 to receive enoxaparin in TIMI 11B; 1564 and 1607 were assigned re-
`spectively in ESSENCE. The 3 validation cohorts were the unfractionated heparin group
`from ESSENCE and both enoxaparin groups.
`Main Outcome Measures The TIMI risk score was derived in the test cohort by
`selection of independent prognostic variables using multivariate logistic regression, as-
`signment of value of 1 when a factor was present and 0 when it was absent, and sum-
`ming the number of factors present to categorize patients into risk strata. Relative dif-
`ferences in response to therapeutic interventions were determined by comparing the
`slopes of the rates of events with increasing score in treatment groups and by testing
`for an interaction between risk score and treatment. Outcomes were TIMI risk score
`for developing at least 1 component of the primary end point (all-cause mortality, new
`or recurrent MI, or severe recurrent ischemia requiring urgent revascularization) through
`14 days after randomization.
`Results The 7 TIMI risk score predictor variables were age 65 years or older, at least
`3 risk factors for coronary artery disease, prior coronary stenosis of 50% or more, ST-
`segment deviation on electrocardiogram at presentation, at least 2 anginal events in
`prior 24 hours, use of aspirin in prior 7 days, and elevated serum cardiac markers. Event
`rates increased significantly as the TIMI risk score increased in the test cohort in TIMI
`11B: 4.7% for a score of 0/1; 8.3% for 2; 13.2% for 3; 19.9% for 4; 26.2% for 5;
`and 40.9% for 6/7 (P⬍.001 by ␹2 for trend). The pattern of increasing event rates
`with increasing TIMI risk score was confirmed in all 3 validation groups (P⬍.001). The
`slope of the increase in event rates with increasing numbers of risk factors was sig-
`nificantly lower in the enoxaparin groups in both TIMI 11B (P=.01) and ESSENCE (P=.03)
`and there was a significant interaction between TIMI risk score and treatment (P=.02).
`Conclusions In patients with UA/NSTEMI, the TIMI risk score is a simple prognos-
`tication scheme that categorizes a patient’s risk of death and ischemic events and pro-
`vides a basis for therapeutic decision making.
`JAMA. 2000;284:835-842
`
`www.jama.com
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, August 16, 2000—Vol 284, No. 7 835
`
`Author Affiliations and Financial Disclosures are listed
`at the end of this article.
`Corresponding Author and Reprints: Elliott M.
`
`Antman, MD, Cardiovascular Division, Brigham and
`Women’s Hospital, 75 Francis St, Boston, MA 02115
`(e-mail: eantman@rics.bwh.harvard.edu).
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`APPLE 1077
`Apple v. AliveCor
`IPR2021-00970
`
`1
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`tic variable, because of the complex pro-
`file of patients with an acute coronary
`syndrome, multivariate analyses that
`adjust for several prognostic variables
`simultaneously provide a more accu-
`rate tool for risk stratification.2,5,14
`Reports of the results of random-
`ized clinical trials of new therapeutic
`strategies for UA/NSTEMI typically pro-
`vide a statement of the overall effec-
`tiveness of a treatment in a population
`that is a mixture of patients at varying
`risks of the primary end point. Al-
`though univariate subgroup analyses
`are frequently presented in clinical trial
`reports, these provide only a partial pic-
`ture of the effect of the new treatment
`in a given subgroup unless adjust-
`ment is made for covariates. Given the
`spectrum of clinical presentations, it is
`plausible that the magnitude of the
`treatment effect of a therapy may vary
`depending on the profile of risk in any
`specific patient.15
`Prognostication of patient risk,
`therefore, is useful not only for allow-
`ing clinicians to triage patients to the
`optimum location for delivery of
`medical care (eg, intensive care unit vs
`hospital ward vs outpatient care)16,17
`but also for identification of patients
`who may be best served by potent but
`expensive—and sometimes risky—
`new therapies.5,18-20 To facilitate wide-
`spread use of a prognostic scoring sys-
`tem for patients with UA/NSTEMI, it
`must be readily applicable using stan-
`dard patient features that are part of
`the routine medical evaluation of such
`patients.
`The primary goal of this article is to
`report the development, testing, and
`clinical utility of a risk stratification tool
`for evaluation of patients with UA/
`NSTEMI. Previously, we reported that
`a risk stratification scheme based on age
`65 years or older, ST deviation on ECG,
`and positive serum cardiac markers seg-
`regated patients with UA/NSTEMI into
`low-, intermediate-, and high-risk
`groups, and the treatment effect of
`enoxaparin was greatest in the highest
`risk group.21 However, that risk strati-
`fication scheme used only a limited
`number of baseline characteristics. We
`
`developed a new, more comprehensive
`risk score for UA/NSTEMI using the da-
`tabase of the Thrombolysis in Myocar-
`dial Infarction (TIMI) 11B trial, a phase
`3 trial comparing low-molecular-
`weight heparin (enoxaparin) with un-
`fractionated heparin.22 Our purpose in
`designing a simple risk score was to pro-
`vide a tool that potentially could be ap-
`plied in clinical settings in which pa-
`tients with UA/NSTEMI present for
`evaluation.
`
`METHODS
`The design and results of the TIMI 11B
`and Efficacy and Safety of Subcutane-
`ous Enoxaparin in Unstable Angina and
`Non-Q-Wave MI (ESSENCE) trials have
`been reported previously.22,23 All pa-
`tients (n=3910 in TIMI 11B and n=3171
`in ESSENCE) presented within 24 hours
`of an episode of UA/NSTEMI at rest. Ad-
`ditional enrollment criteria included at
`least 1 of the following: ST-segment de-
`viation on the qualifying ECG (either
`transient ST elevation or persistent ST de-
`pression of ⱖ0.05 mV in TIMI 11B and
`ⱖ0.01 mV in ESSENCE), documented
`history of coronary artery disease, and
`elevated serum cardiac markers. (In TIMI
`11B, a history of coronary artery dis-
`ease was acceptable initially but was
`dropped later as the sole supportive cri-
`terion for UA/NSTEMI.) Major exclu-
`sion criteria were planned revascular-
`ization in 24 hours or less, a correctable
`cause of angina, and contraindications
`to anticoagulation.
`All patients received aspirin (100-
`325 mg/d) and, after providing writ-
`ten informed consent, were randomly
`assigned to 1 of 2 antithrombotic strat-
`egies. Both trials used a double dummy
`technique so that all patients received
`both an intravenous infusion (unfrac-
`tionated heparin or matched placebo)
`and subcutaneous injections (enoxa-
`parin or matched placebo). For the pur-
`poses of developing the TIMI risk score
`for UA/NSTEMI, the prespecified pri-
`mary efficacy end point from TIMI 11B
`was applied to both trials in a fashion
`similar to that reported for the TIMI
`11B–ESSENCE meta-analysis.24 This
`end point was a composite of all-cause
`
`mortality, new or recurrent MI, or se-
`vere recurrent ischemia prompting ur-
`gent revascularization. The analyses
`shown herein are based on rates for the
`primary end point through 14 days af-
`ter randomization.
`Initially, a multivariate model for
`prognostication of risk for experienc-
`ing at least 1 element of the primary end
`point was developed. The model incor-
`porated baseline characteristics that
`could be readily identified at presenta-
`tion and was restricted to the cohort of
`patients assigned to unfractionated hep-
`arin in TIMI 11B (test cohort). The ra-
`tionale for this approach was to focus on
`information that could be ascertained in
`a relatively short period after encoun-
`tering a patient and establishing a model
`that could be used for efficient triage for
`patient care without waiting for addi-
`tional tests or results of an initial pe-
`riod of medical observation over sev-
`eral days. Baseline characteristics that
`were evaluated include those previ-
`ously reported to be important vari-
`ables predicting outcomes in patients
`with UA/NSTEMI and are shown in
`TABLE 1.4,5,14,25,26
`A total of 12 baseline characteristics
`arranged in a dichotomous fashion were
`screened as candidate predictor vari-
`ables of risk of developing an end-
`point event (Table 1). A multivariate lo-
`gistic regression model was then used
`to assess the statistical significance of
`each candidate prognostic variable. Af-
`ter each factor was tested indepen-
`dently in a univariate logistic regres-
`sion model, those that achieved a
`significance level of P⬍.20 were se-
`lected for testing in a multivariate step-
`wise (backward elimination) logistic re-
`gression model. Variables associated
`with P⬍.05 were retained in the final
`model. Maximum likelihood estimates
`of the parameter coefficients were ob-
`tained using SAS PROC LOGISTIC (SAS
`Institute Inc, Cary, NC). The goodness
`of fit of the model to the observed event
`rates was evaluated by calculating the
`Hosmer-Lemeshow statistic.27 Low ␹2
`values and high corresponding P val-
`ues for the Hosmer-Lemeshow statistic
`indicate that the data can be ad-
`
`836 JAMA, August 16, 2000—Vol 284, No. 7 (Reprinted)
`
`©2000 American Medical Association. All rights reserved.
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`2
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`equately fit to a logistic function. The
`ability of the model to classify patients
`(ie, its predictive performance) was
`evaluated using the C statistic, a term
`equivalent to the area under a receiver
`operating characteristic curve for di-
`chotomous outcomes.28 Assessment of
`the impact of missing information for
`predictor variables was carried out by
`Monte-Carlo simulations that ran-
`domly set fixed proportions of the data
`to missing and then repeating the logis-
`tic regression analyses.
`After development of the multivar-
`iate model, the TIMI risk score for UA/
`NSTEMI was developed for the test co-
`hort using those variables that had been
`found to be statistically significant pre-
`dictors of events in the multivariate
`analysis. The score was then con-
`structed by a simple arithmetic sum of
`the number of variables present. Dif-
`ferences in the event rates for increas-
`ing TIMI risk score values were as-
`sessed using the ␹2 test for trend.
`The risk score was then validated in
`3 separate cohorts of patients: the enoxa-
`parin group from TIMI 11B (n=1953),
`the unfractionated heparin group from
`ESSENCE (n=1564), and the enoxapa-
`rin group from ESSENCE (n=1607). We
`tested for homogeneity of the unfrac-
`
`tionated heparin control groups in TIMI
`11B and ESSENCE by comparing the
`slope of the increase in the rate of events
`with increasing TIMI risk score using
`least squares linear regression analysis.
`Differences between the unfraction-
`ated heparin and enoxaparin groups in
`both TIMI 11B and ESSENCE were also
`assessed by comparing the slope of the
`increase in rate of events with increas-
`ing TIMI risk score using least squares
`linear regression analysis. In addition,
`using a merged database of the TIMI 11B
`and ESSENCE studies, testing for a het-
`erogeneous treatment effect stratified by
`risk score was carried out by examin-
`ing the statistical significance of the in-
`teraction term in a multivariate logistic
`regression model of the following form:
`outcome=constant + risk score + treat-
`ment (eg, unfractionated heparin vs
`enoxaparin) + risk score ⴱ treatment.
`The asterisk in the model desig-
`nates an interaction between the ad-
`joining terms. To explore whether the
`interaction of risk score ⴱ treatment was
`affected by the trial in which the pa-
`tient was enrolled, we tested for statis-
`tical significance of terms for trial (TIMI
`11B vs ESSENCE) and interaction of
`trial with risk score and treatment when
`added to the model.
`
`As a secondary goal, we examined the
`ability of the TIMI risk score to pre-
`dict development of each of the indi-
`vidual components of the composite
`primary end point as well as the com-
`posite end point of all-cause mortality
`or nonfatal MI.
`
`RESULTS
`The test cohort for development of the
`TIMI risk score consisted of the 1957
`patients assigned to receive unfraction-
`ated heparin in TIMI 11B.22 The pri-
`mary end point (all-cause mortality, MI,
`or urgent revascularization) occurred
`by 14 days in 16.7% of patients in the
`test cohort. Of the 12 original candi-
`date variables, 7 remained statistically
`significant in the multivariate analysis
`and formed the final set of predictor
`variables (Table 1). The Hosmer-
`Lemeshow statistic was 3.56 d f 8,
`(P=.89). The C statistic for the model
`in the test cohort was 0.65.
`Since the parameter estimates for
`each of the 7 predictor variables were
`of a similar magnitude (Table 1), the
`risk score was calculated by assigning
`a value of 1 when a variable was pres-
`ent and then categorizing patients in the
`test cohort by the number of risk fac-
`tors present, as shown in FIGURE 1. The
`
`Table 1. Baseline Characteristics Analyzed for Development of TIMI Risk Score for UA/NSTEMI*
`Univariate Analysis
`
`Multivariate Analysis
`
`␤ Coefficient
`0.4681
`0.3717
`0.5473
`
`P Value
`⬍.001
`.009
`⬍.001
`
`OR (95% CI)
`1.60 (1.25-2.04)
`1.45 (1.10-1.91)
`1.73 (1.34-2.23)
`
`␤ Coefficient
`0.5575
`0.4336
`0.5284
`
`P Value
`⬍.001
`.003
`⬍.001
`
`OR (95% CI)
`1.75 (1.35-2.25)
`1.54 (1.16-2.06)
`1.70 (1.30-2.21)
`
`Characteristics†
`Age, ⱖ65 y
`At least 3 risk factors for CAD‡
`Significant coronary stenosis
`(eg, prior coronary stenosis ⱖ50%)
`Prior MI
`Prior CABG
`Prior PTCA
`ST deviation
`Severe anginal symptoms
`(eg, ⱖ2 anginal events in last 24 h)
`Use of aspirin in last 7 days
`Use of IV unfractionated heparin
`within 24 hours of enrollment
`1.42 (1.12-1.80)
`.004
`0.3486
`Elevated serum cardiac markers§
`0.90 (0.53-1.53)
`.70
`−0.1058
`Prior history of CHF
`*UA/NSTEMI indicates unstable angina/non−ST elevation myocardial infarction; OR, odds ratio; CI, confidence interval; CAD, coronary artery disease; MI, myocardial infarction;
`CABG, coronary artery bypass graft surgery; PTCA, percutaneous transluminal coronary angioplasty; IV, intravenous; and CHF, congestive heart failure.
`†Bold indicates variables that remained statistically significant in the multivariate analysis and were used as the final set of predictor variables.
`‡Risk factors included family history of CAD, hypertension, hypercholesterolemia, diabetes, or being a current smoker.
`§Creatine kinase MB fraction and/or cardiac-specific troponin level.
`
`0.2386
`0.3004
`0.4828
`0.3356
`0.4521
`
`0.6179
`0.1665
`
`.06
`.07
`.004
`.02
`<.001
`
`.002
`.19
`
`1.27 (0.99-1.63)
`1.35 (0.97-1.88)
`1.62 (1.16-2.26)
`1.40 (1.06-1.85)
`1.57 (1.24-2.00)
`
`1.86 (1.26-2.73)
`1.18 (0.92-1.51)
`
`0.4125
`0.4279
`
`0.5534
`
`.005
`.001
`
`.006
`
`1.51 (1.13-2.02)
`1.53 (1.20-1.96)
`
`1.74 (1.17-2.59)
`
`0.4420
`
`⬍.001
`
`1.56 (1.21-1.99)
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, August 16, 2000—Vol 284, No. 7 837
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`3
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`sults of a prior cardiac catheterization.
`Construction of the TIMI risk score us-
`ing the TIMI 11B database was accom-
`plished from the case report form data
`for each patient and, therefore, com-
`plete information for the predictor of
`prior coronary stenosis of 50% or more
`was available for all patients; a value of
`0 was assigned if no cardiac catheteriza-
`tion had been previously performed or
`if a prior cardiac catheterization re-
`vealed no coronary stenoses of 50% or
`more; a value of 1 was assigned if a prior
`cardiac catheterization revealed at least
`1 coronary stenosis of 50% or more.
`Since the results of a prior cardiac
`catheterization might not be immedi-
`ately available to a clinician attempt-
`ing to use the TIMI risk score when a
`patient with UA/NSTEMI presents for
`evaluation, we investigated the effect of
`missing values on the prior coronary
`stenosis of 50% or more variable. Us-
`ing Monte-Carlo simulation, a fixed
`proportion of data on prior coronary
`stenosis of 50% or more was ran-
`domly set as missing. The model was
`reevaluated assuming 0 for missing pa-
`tients and then reevaluated once again
`excluding the missing patients. When
`10%, 30%, or 50% of the prior coro-
`nary stenosis data were randomly set
`as missing and a 0 was assumed for the
`missing patients, the variable for prior
`coronary stenosis of 50% or more re-
`mained a significant predictor of the
`composite outcome at 14 days: for 10%
`missing, odds ratio (OR)=1.44 (95%
`confidence interval [CI], 1.18-1.75;
`P⬍.001); for 30% missing, OR = 1.35
`(95% CI, 1.09-1.68; P=.007); and for
`50% missing, OR=1.58 (95% CI, 1.25-
`2.01; P⬍.001). For the same assump-
`tions of data randomly set as missing
`but excluding missing patients, the vari-
`able for prior coronary stenosis of 50%
`or more also remained a significant pre-
`dictor, with ORs of 1.53 (95% CI, 1.25-
`1.88), 1.50 (95% CI, 1.19-1.90), and
`1.63 (95% CI, 1.25-2.13), respectively
`(P⬍.001 for all). Therefore, under a va-
`riety of assumptions about missing val-
`ues, prior coronary stenosis of 50% or
`more remained a significant predictor
`of outcome.
`
`Validation of Risk Score
`Validation of the TIMI risk score is
`shown in FIGURE 2. The unfraction-
`ated heparin control groups in TIMI
`11B and ESSENCE showed a homogen-
`eous pattern when patients were strati-
`fied by risk score since the slope of the
`increase in event rates with increasing
`number of risk factors was not statis-
`tically different (P=.13) in the 2 un-
`fractionated heparin groups (Figure 2).
`For all 3 validation cohorts (the enoxa-
`parin group from TIMI 11B, the un-
`fractionated heparin group from
`ESSENCE, and the enoxaparin group
`from ESSENCE) there was a signifi-
`cant increase in the rate of events as the
`TIMI risk score increased (P⬍.001).
`
`Application of TIMI Risk Score
`As shown in Figure 2, the relative rate
`of increase in events among patients
`with higher TIMI risk scores was dif-
`ferent for the unfractionated heparin
`and enoxaparin groups. For both TIMI
`11B and ESSENCE, the slope of the in-
`crease in event rates with increasing
`numbers of risk factors was signifi-
`cantly lower in the enoxaparin groups
`(3.92 vs 6.41; P=.01 in TIMI 11B; 2.18
`vs 4.36; P=.03 in ESSENCE). A gener-
`ally consistent pattern of increasing ab-
`solute risk difference and correspond-
`ing decrease in the number of patients
`requiring treatment to prevent 1 end
`point event by 14 days after random-
`ization favoring enoxaparin was seen
`in both trials as the TIMI risk score in-
`creased.
`Using a merged database from the
`TIMI 11B and ESSENCE trials
`(N=7081), multivariate logistic regres-
`sion analysis revealed that the TIMI risk
`score and treatment (unfractionated hep-
`arin vs enoxaparin) were significant pre-
`dictors (P⬍.001 for both terms) of all-
`cause mortality, MI, or urgent
`revascularization by 14 days after
`randomization (C statistic = 0.63).
`An interaction term for TIMI risk
`score ⴱ treatment was also a significant
`predictor of the composite outcome at
`day 14 (P=.02). However, the follow-
`ing terms were not significant pre-
`
`Figure 1. TIMI Risk Score
`
`40.9
`
`26.2
`
`19.9
`
`13.2
`
`3
`4
`5
`No. of Risk Factors
`627
`573
`267
`(32.0)
`(29.3)
`(13.6)
`
`6/7
`
`66
`(3.4)
`
`8.3
`
`4.7
`
`0/1
`
`2
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`15
`
`10
`
`5 0
`
`Rate of Composite End Point, %
`
`Test Cohort
`No.
`(%)
`
`
`85
`(4.3)
`
`339
`(17.3)
`
`Rates of all-cause mortality, myocardial infarction, and
`severe recurrent ischemia prompting urgent revascu-
`larization through 14 days after randomization were
`calculated for various patient subgroups based on the
`number of risk factors present in the test cohort (the
`unfractionated heparin group in the Thrombolysis in
`Myocardial Infarction [TIMI] 11B trial; n=1957) (see
`Table 1). Event rates increased significantly as the TIMI
`risk score increased (P⬍.001 by ␹2 for trend).
`
`pattern of the number of risk factors was
`normally distributed. Because of the
`small number of patients with ex-
`treme risk scores, patients with 0 or 1
`risk factor(s) and 6 or 7 risk factors were
`combined. There was a progressive, sig-
`nificant pattern of increasing event rates
`as the TIMI risk score increased in the
`test cohort (P⬍.001 by ␹2 for trend).
`In the final model, an age cutoff of
`65 years was used because this value
`was close to the median age for the un-
`fractionated heparin group (66 years)
`and was the median age for the enoxa-
`parin group. Use of different age cut-
`offs showed very little effect on perfor-
`mance of the model: the C statistic
`ranged between 0.63 and 0.66 for vary-
`ing age cutoffs in 5-year increments
`from 50 to 80 years. Furthermore, treat-
`ing age as a continuous variable (prob-
`lematic for the development of a simple
`risk score) also had little effect on model
`performance: the C statistic was 0.66
`in a model using age as a continuous
`variable.
`One of the 7 predictor variables shown
`in Table 1, prior coronary stenosis of 50%
`or more, requires knowledge of the re-
`
`838 JAMA, August 16, 2000—Vol 284, No. 7 (Reprinted)
`
`©2000 American Medical Association. All rights reserved.
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`4
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`dictors of the composite outcome at day
`14: trial (P = .18), trial ⴱ treatment
`(P=.51), trial ⴱ TIMI risk score (P=.13),
`and trial ⴱ treatment ⴱ TIMI risk score
`(P=.84); inclusion of these in the lo-
`gistic regression analysis had no effect
`on overall model performance (C
`statistic=0.63).
`The ability of the TIMI risk score to
`predict outcomes other than all-cause
`mortality, MI, or urgent revasculariza-
`tion was assessed in TIMI 11B. In the
`entire trial population, there were pro-
`gressive, significant (P⬍.001) in-
`creases in the rates of all-cause mortal-
`ity, MI, urgent revascularization, and
`the composite of all-cause mortality or
`nonfatal MI as the TIMI risk score in-
`creased (FIGURE 3). The event rates
`stratified by risk score for the unfrac-
`tionated heparin and enoxaparin groups
`in TIMI 11B are shown in TABLE 2. For
`both treatment groups, there was a con-
`sistent, significant increase in the rate
`of events for each outcome with in-
`creasing risk score. Also, for each out-
`
`come, the slope of the increase in events
`with increasing risk score was lower in
`the enoxaparin group: 68% lower for
`all-cause mortality (P=.02), 25% lower
`for MI (P=.41), 38% lower for urgent
`revascularization (P = .05), and 39%
`lower for all-cause mortality or nonfa-
`tal MI (P=.15).
`
`COMMENT
`Our results indicate that standard clini-
`cal characteristics routinely obtained dur-
`ing the initial medical evaluation of pa-
`tients with UA/NSTEMI can be used to
`construct a simple classification system
`that is predictive of risk for death and car-
`diac ischemic events. The TIMI risk score
`includes variables that can be easily as-
`certained when a patient with UA/
`NSTEMI presents to the medical care sys-
`tem. The variables used to construct the
`score were based on observations from
`prior studies of risk stratification and in-
`corporate demographic and historical
`features of the patient, measures of the
`tempo and acuity of the presenting ill-
`
`ness, and indicators of the extent of myo-
`cardial ischemia and necrosis.2,9,17,29-31 The
`predictor variables were derived from a
`logistic regression model that con-
`firmed their independent predictive
`power after multivariate adjustment in
`the TIMI 11B and ESSENCE data sets.
`The simple arithmetic sum of the
`number of variables present that con-
`stitutes the risk score can be calcu-
`lated without the aid of a computer.
`This distinguishes the TIMI risk score
`from other scoring systems that are
`more complex computationally since
`they require weighting terms for the
`predictor variables and cannot be imple-
`mented easily without computer assis-
`tance.25 The approach taken in devel-
`oping the TIMI risk score is similar to
`that taken by Centor et al,32 who intro-
`duced a scoring system for assessment
`of the likelihood of streptococcal phar-
`yngitis based on clinical findings as-
`certained in the emergency depart-
`ment, and Croft et al,33 who developed
`a simple clinical prediction rule for
`
`Figure 2. Validation of TIMI Risk Score and Assessment of Treatment Effect According to Score
`
`TIMI 11B
`(n = 3910)
`
`ESSENCE
`(n = 3171)
`
`Unfractionated Heparin (n = 1957)
`Enoxaparin (n = 1953)
`
`40.9
`
`Unfractionated Heparin (n = 1564)
`Enoxaparin (n = 1607)
`
`26.2
`
`28.8
`
`19.9
`
`20
`
`13.2 14.1
`
`14.9
`
`8.3 8.6
`
`4.7 3.5
`
`15.8
`
`16.8
`
`12
`
`12.4
`
`11.6
`
`9.5
`
`7.2 7.3
`
`38.1
`
`31
`
`18.3
`
`20
`
`0/1
`
`2
`
`4
`3
`No. of Risk Factors
`
`5
`
`6/7
`
`0/1
`
`2
`
`3
`4
`No. of Risk Factors
`
`5
`
`6/7
`
`85 (4.3)
`
`339 (17.3)
`
`627 (32.0)
`
`573 (29.3)
`
`267 (13.6)
`
`66 (3.4)
`
`265 (16.9)
`
`438 (28.0)
`
`476 (30.4)
`
`280 (17.9)
`
`84 (5.4)
`
`21 (1.3)
`
`86 (4.4)
`
`362 (18.5)
`
`631 (32.3)
`
`536 (27.4)
`
`265 (13.6)
`
`73 (3.7)
`
`261 (16.2)
`
`465 (28.9)
`
`515 (32.0)
`
`258 (16.1)
`
`93 (5.8)
`
`15 (0.9)
`
`1.2
`83
`
`–0.3
`–333
`
`–0.9
`–111
`
`5
`20
`
`6.2
`16
`
`12.1
`8
`
`– 0.1
`–910
`
`2.1
`46
`
`3.8
`27
`
`4.4
`23
`
`12.7
`8
`
`18.1
`6
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`15
`
`10
`
`5 0
`
`Rate of Composite End Point, %
`
`No. (%) of Patients
`Unfractionated
`Heparin Group
`Enoxaparin
`Group
`ARD
`NNT
`
`Rates of all-cause mortality, myocardial infarction, and severe recurrent ischemia prompting urgent revascularization through 14 days after randomization were cal-
`culated for the enoxaparin and unfractionated heparin groups in the Thrombolysis in Myocardial Infarction (TIMI) 11B trial and the Efficacy and Safety of Subcutaneous
`Enoxaparin in Unstable Angina and Non-Q-Wave MI trial (ESSENCE), based on the TIMI risk score. The pattern of increasing event rates with increasing TIMI risk score
`was confirmed in all 3 validation cohorts (P⬍.001 by ␹2 for trend). C statistics were 0.65 for the unfractionated heparin group and 0.61 for the enoxaparin group in
`TIMI 11B; and 0.65 for the unfractionated heparin group and 0.59 for the enoxaparin group in ESSENCE. The rate of increase in events as more risk factors were
`present was significantly lower in the enoxaparin group in both studies (for TIMI 11B, P=.01; for ESSENCE, P=.03). Positive values for absolute risk difference (ARD)
`and number needed to treat to prevent 1 event (NNT) indicate calculations favoring enoxaparin, while negative values indicate calculations favoring unfractionated
`heparin.
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, August 16, 2000—Vol 284, No. 7 839
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`5
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`identifying nerve function impair-
`ment in patients with leprosy.
`The TIMI risk score appears statis-
`tically robust in that it was validated in-
`
`ternally within TIMI 11B as well as in
`2 separate cohorts of patients from the
`ESSENCE trial. The model is easy to re-
`call and apply clinically since a simple
`
`Figure 3. Outcomes for Individual Components of the Composite Primary End Point
`Stratified by TIMI Risk Score
`
`Myocardial Infarction
`
`15.8
`
`8.5
`
`5.0
`
`3.7
`
`2.3
`
`2.1
`
`0/1
`
`2
`
`3
`
`4
`
`5
`
`6/7
`
`All-Cause Mortality or Nonfatal MI
`
`19.4
`
`11.5
`
`6.7
`
`4.7
`
`2.9
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`25
`
`20
`
`15
`
`10
`
`5
`
`Rate, %
`
`Rate, %
`
`All-Cause Mortality
`
`1.2
`
`0/1
`
`1.0
`
`2
`
`1.7
`
`3
`
`6.5
`
`5.6
`
`2.5
`
`4
`
`5
`
`6/7
`
`Urgent Revascularization
`
`20.9
`
`14.3
`
`12.2
`
`9.5
`
`6.0
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`25
`
`20
`
`15
`
`10
`
`5
`
`Rate, %
`
`Rate, %
`
`age cutoff of 65 years provided similar
`predictive ability to a more complex
`model using age as a continuous vari-
`able. Also, variables such as knowl-
`edge of whether the patient had a pre-
`viously documented coronary stenosis
`of 50% or more appeared relatively in-
`sensitive to missing information and re-
`mained a significant predictor of events.
`The TIMI risk score offers several
`promising applications for clinical use.
`It categorizes patients with UA/
`NSTEMI into groups that span a wide
`range of risk for clinical events—about
`a 5- to 10-fold range of risk. A contribu-
`tion of the TIMI risk score that has not
`been emphasized in other risk stratifi-
`cation studies is the actual testing of its
`use for identifying patients who would
`be expected to show particular benefit
`from new antithrombotic regimens such
`as enoxaparin.4 As evidenced by the
`lower slope of the increase in event rates
`with increasing risk score in Figure 2 and
`the statistical significance of the inter-
`action term between risk score and treat-
`ment, the benefit of enoxaparin was
`greatest in those patients with higher
`TIMI risk scores. That the logistic re-
`gression modeling did not indicate that
`the trial in which the patient was en-
`rolled was a predictor of outcome and
`that the interactions between trial and
`risk score were not significant are con-
`sistent with an independent effect of
`enoxaparin across the 2 trials and illus-
`trates the use of the TIMI risk score for
`therapeutic decision making. The abso-
`
`0
`
`1.2
`
`0/1
`
`2
`
`3
`
`4
`
`5
`
`2.9
`
`0
`
`6/7
`0/1
`No. of Risk Factors
`
`2
`
`3
`
`4
`
`5
`
`6/7
`
`Rates of all-cause mortality, myocardial infarction (MI), urgent revascularization, and all-cause mortality or
`nonfatal MI through 14 days after randomization were calculated for the entire population in the Thromboly-
`sis in Myocardial Infarction (TIMI) 11B trial based on the TIMI risk score. There was a progressive, significant
`increase in the rate of events for each end point as the TIMI risk score increased (P⬍.001 by ␹2 for trend for
`all). C statistics for the 4 end points shown were 0.74 (all-cause mortality), 0.66 (MI), 0.68 (urgent revascu-
`larization), and 0.63 (all-cause mortality/MI), respectively.
`
`Table 2. Event Rates in TIMI 11B Stratified by TIMI Risk Score*
`TIMI Risk Score, Rate of Events, %
`
`0/1
`
`2
`
`3
`
`0
`2.3
`
`3.5
`1.2
`
`1.2
`1.2
`
`0.9
`1.1
`
`1.8
`2.5
`
`6.2
`5.8
`
`All-cause mortality
`Unfractionated heparin
`Enoxaparin
`Myocardial infarction
`Unfractionated heparin
`Enoxaparin
`Urgent revascularization
`Unfractionated heparin
`Enoxaparin
`All-cause mortality or nonfatal
`myocardial infarction
`4.8
`2.4
`3.5
`Unfractionated heparin
`4.6
`3.3
`2.3
`Enoxaparin
`*TIMI 11B indicates the Thrombolysis in Myocardial Infarction 11B trial.
`
`1.8
`1.7
`
`4.0
`3.3
`
`9.1
`10.0
`
`4
`
`2.6
`2.4
`
`5.9
`4.1
`
`14.0
`10.3
`
`7.5
`5.8
`
`5
`
`7.1
`4.2
`
`9.7
`7.2
`
`15.0
`13.6
`
`13.5
`9.4
`
`6/7
`
`10.6
`2.7
`
`16.7
`15.1
`
`27.3
`15.1
`
`22.7
`16.4
`
`P Value by ␹2
`for Trend
`
`C Statistic
`
`Slope
`
`P Value for
`Comparison
`of Slopes
`
`⬍.001
`.05
`
`⬍.001
`⬍.001
`
`⬍.001
`⬍.001
`
`⬍.001
`⬍.001
`
`0.78
`0.72
`
`0.68
`0.65
`
`0.64
`0.62
`
`0.70
`0.65
`
`1.88
`0.61
`
`2.47
`1.85
`
`3.93
`2.43
`
`3.53
`2.15
`
`.02
`
`.41
`
`.05
`
`.15
`
`840 JAMA, August 16, 2000—Vol 284, No. 7 (Reprinted)
`
`©2000 American Medical Association. All rights reserved.
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`6
`
`

`

`THE TIMI RISK SCORE FOR UNSTABLE ANGINA
`
`lute difference in event rates increased
`and the corresponding number of pa-
`tients needed to treat for prevention of
`1 event with enoxaparin decreased as the
`risk score increased (Figure 2). As shown
`in Figure 3, the risk score also appears
`useful for stratification of patients at risk
`for the individual components of com-
`posite end points used in many contem-
`porary trials of therapi