In healthcare, there is a constant need to increase and
improve a clinician’s ability to quickly and correctly
diagnose a patient’s condition. This need is paramount in
cardiac care where patients may have a critical condition
that needs to be assessed and treated immediately, such as
acute cardiac ischemia (ACI).
In support of your efforts to reduce discovery to treatment
time for STEMI patients, Philips has incorporated a
cardiac decision support tool into its HeartStart MRx
monitor/defibrillator: the Philips ACI-TIPI (Acute
Cardiac Ischemia Time-Insensitive Predictive
Instrument). This application note provides both
pre-hospital and emergency department (ED) clinicians
with an overview of the development and functionality of
this tool that aids the decision making process to
systematically improve patient outcomes.
Why Use the Philips ACI-TIPI
The Philips ACI-TIPI is a software tool that enhances the
computerized 12-lead ECG analysis capabilities of the
HeartStart MRx. It generates a predicted probability
score of ACI based on ECG features and patient
demographic information such as age, gender, and chest
determining when additional diagnostic testing was
The probability score supports EMS personnel’s efforts to
decrease time from the onset of ACI symptoms to
treatment with interventional cardiology. For example,
one pre-hospital study demonstrated ACI-TIPI
probabilities of ACI were comparable to those based on
physician ECG interpretations and may be useful in the
prehospital evaluation of chest pain.1
determining the probability of ischemia to influence
Studies have also shown the ACI-TIPI benefit to ED and
other acute care setting personnel by providing a second
opinion for decision making prior to cath lab activation
or patient release. Specifically, the ACI-TIPI was useful
not required in a low-risk ED population with
symptoms suggestive of acute coronary syndrome,2
identifying the likelihood of ischemia to help
determine the need for stress testing before hospital
the initial triage disposition of AMI patients to CCUs
or intermediate wards.4
It has been projected that if ACI-TIPI were used widely
throughout the United States, its potential incremental
impact may be more than 200,000 fewer unnecessary
hospitalizations and more than 100,000 fewer
unnecessary CCU admissions each year, for an overall
annual savings of $728 million.5
The ACI-TIPI: A Closer Look
The Difficulty of ACI Diagnosis
Diagnosing ACI depends heavily on obtaining an
accurate medical history and cardiac enzyme test results,
and on interpretation of the ECG tracings.
Unfortunately, patients with chest pain in the
pre-hospital environment and in the ED may not be
responsive to questioning and/or may not remember
previous chest pain history. Likewise, once an ECG is
taken, the early signs of ACI identified in the ECG can
be confusing, even to highly trained readers.
The inception of the ACI-TIPI resulted from researchers’
recognition of the need to increase the clinician’s ability
to quickly and correctly diagnose ACI. Reasoning that a
single numerical probability value might be easily
incorporated into clinical decision making, a “predictive
instrument” was created to compute a patient’s likelihood
of having ACI. The instrument generates a 0-100%
predicted probability score using an algorithm based on
weighted values for the patient’s age, gender, chest pain
status, and ECG waveform criteria.
Dr. Harry Selker and his colleagues conducted a
multicenter predictive instrument study in six New
England hospitals to develop and prospectively test a
method to reduce unnecessary CCU admissions.6 The
study was conducted in two phases, each lasting about
one year. The first phase developed a predictive
instrument from data on 2,188 patients who participated
in the study. The second phase was a prospective clinical
trial to test the usefulness of the predictive instrument in
improving CCU admitting practices. The study included
2,320 patients seen in the six hospitals' EDs.
The instrument was shown to be effective in helping
clinicians identify patients with ACI and in reducing false
positive identifications of ACI. Through the study, the
efficacy of the instrument was confirmed in hospital EDs
ranging from urban major teaching centers to rural
non-teaching hospitals and was widely considered to
represent an important new technology for the ED
diagnosis and treatment of patients with chest pain or
other symptoms suggestive of ACI.
Since the original study, Dr. Selker and his colleagues
have continued to improve on the predictive
performance of the instrument. Extensive work was done
by Philips with Dr. Selker to produce a version of his
ACI-TIPI for the Philips PageWriter cardiograph,
resulting in the Philips ACI-TIPI.
Understanding the ACI-TIPI
This section briefly describes the ECG features that are
used by the ACI-TIPI in computing the probability of
Abnormal Q Waves
The presence of abnormal Q waves generally indicates
myocardial infarction. In some cases, infarction can occur
without the generation of abnormal Q waves. Truly
pathological Q waves (see Figure 1) may be due to
previously unrecognized infarction. Conversely, a prior
infarction may mask new ischemia in the same area.
Pathological Q Wave
During development of the ACI-TIPI, seven predictors
of ACI were established - four clinical factors and three
The four clinical factors are:
Patient's age (yrs.)
The presence or absence of chest pain or pressure, or
left arm pain
Whether chest pain or equivalent symptom is the
patient's most important presenting symptom
The three ECG features are:
The presence or absence of pathological or significant
The presence and degree of ST segment elevation or
The presence and degree of T wave elevation or
While none of these features alone is diagnostic, together
they represent the most prominent indication of ACI. To
be considered significant, ECG features must be apparent
in at least two contiguous leads and must not be due to
any of the five exclusionary conditions that can skew
ST Segment Elevation
ST segment elevation (see Figure 2) is seen in over
two-thirds of patients admitted to the CCU who have
had an infarction. The percentage is even higher for
patients with both abnormal Q waves and ST segment
elevation. However, ST segment elevation can occur in
the absence of ischemia. For example, it can be due to
early repolarization, pericarditis, or left ventricular
right bundle branch blocks,
left bundle branch blocks,
“early repolarization variant,”
left ventricular hypertrophy, and
right ventricular hypertrophy.
ST Segment Depression
ST segment depression (see Figure 3) usually indicates
ischemia. Over half of patients admitted to CCUs with
ST segment depression have infarctions. ST segment
depression may also occur in normal individuals during
hyperventilation, in patients with hypokalemia or left
ventricular strain, and in those taking digitalis.
As indicated earlier, the Philips ACI-TIPI attempts to
differentiate primary from secondary ST and T changes.
This is done to identify non-ACI conditions often
associated with secondary ST and T chnages such as right
bundle branch blocks (RBBB), left bundle branch blocks
(LBBB), “early repolarization variant,” left ventricular
hypertrophy (LVH), and right ventricular hypertrophy
(RVH). These interpretations must be reviewed by a
physician for confirmation of primary vs. secondary ST
and T changes. The ACI-TIPI excludes the ST and/or T
factors (as appropriate) from its calculations when it
detects the following abnormalities:
LVH (can alter ST segment and T waves)
T Wave Inversion
Inverted T waves (see Figure 4) may also indicate ACI.
Isolated T wave inversion can be an indicator of AMI and
may also reflect prior myocardial damage, or
non-ischemic causes such as left ventricular strain.
T Wave Inversion
Secondary repolarization (ST and T) abnormalities
The presence of an artificial pacemaker
When it excludes such data from its calculations, the
ACI-TIPI issues the statement:
NOTE: Secondary Q wave and ST-T changes were not scored
due to presence of LBBB. IF ALL ST-T CHANGES WERE
SCORED AS ISCHEMIC, ACI-TIPI PROBABILITY
WOULD BE HIGHER.
Another exclusionary condition to keep in mind is that
the ACI-TIPI score is not calculated for patients whose
age is less than 18.
WARNING: If the tool issues an exclusion statement, the
clinician must consider that the actual probability may be
higher than what has been indicated.
The ACI-TIPI Algorithm
ST and T Measurements
The ACI-TIPI's probability score depends on the ST
segment and T wave measurements. However, ventricular
conduction abnormalities and ventricular hypertrophy
can cause secondary ST and T wave changes that, if
misinterpreted as primary, could cause erroneously high
ACI-TIPI scores. To avoid this, the ACI-TIPI sorts out
the primary ST and T changes from the secondary
changes when conduction abnormalities are present. In
certain cases, this distinction cannot reliably be made and
the ACI-TIPI issues a warning message. See the following
Exclusionary Conditions section for more information.
The Philips ACI-TIPI algorithm takes the coefficients
created by Dr. Selker’s logistic regression and inputs
information regarding the presence and/or level of each
clinical factor and ECG feature. The ACI-TIPI then
calculates the predicted probability value of ACI
(0-100%). Here are two calculation examples for
For a 63 year old female presenting with chest pain as
the primary complaint, 0.1 mV of ST elevation, -0.2
mV of T wave inversion, and abnormal Q waves, the
ACI-TIPI predicted probability is 87%.
For a 52 year old male presenting with chest pain as a
secondary complaint, 0.2 mV of ST depression, and
-0.5 mV of T wave inversion, the predicted
probability is 90%.
NOTE: Because the ACI-TIPI predicted probability was
designed to assist physicians making critical care
decisions, but not to replace them, a given range of
probability should not be taken to indicate specific
treatment decisions, such as “admit to CCU” or “send
the patient home”. Nonetheless, the issue of subranges
may arise as users apply the ACI-TIPI to clinical settings
and Dr. Selker’s article on the ACI-TIPI should be
consulted for this.7 Based on that article, a given
institution may want to make general recommendations
and/or monitor triage decisions for patients with
suspected ACI. Ultimately, it is the individual physician
who should choose how to apply the ACI-TIPI.
Using the Philips ACI-TIPI
This section describes input data and the resulting
printed report produced by the Philips ACI-TIPI.
concern. It contains 12-lead measurements, ACI-TIPI
interpretive output statements, and a predicted
probability of ACI.
Clinical Variable Input
As with standard 12-lead ECGs, the ACI-TIPI report can
be transmitted to the Philips 12-Lead Transfer Station
and the Philips TraceMaster Vue ECG Management
System, and stored on the TraceMaster Vue System.
The Philips ACI-TIPI is optional, but when used, it
requires entry of chest pain status prior to 12-lead ECG
Provided the chest pain is not due to non-cardiac trauma,
the chest pain status is entered as follows:
primary if chest pain or discomfort (or equivalent left
arm pain) is the main reason for the patient seeking
secondary if chest pain is present but is not the chief
none if the patient has no chest pain or discomfort.
The ACI-TIPI Report
There are no perfect tests or algorithms to exclude ACI,8
including the ACI-TIPI. For example, in one study
assessing its effect on the ED process of care outcomes,
ACI-TIPI did not appear to reduce resource utilization or
decrease length of stay.9 In another study, the addition of
the ACI-TIPI score did not improve diagnostic accuracy
or significantly change triage in rural hospitals.10 These
research results should be carefully weighted against a far
greater number of ACI-TIPI studies with positive
outcomes, as cited in this application note.
The ACI-TIPI report is meant to supplement Philips'
standard 12-lead interpretation report and be used in
clinical settings where ACI is a primary diagnostic
The diagnosis and management of ACI is a clear
challenge for emergency medical personnel. Strategies
must quickly and accurately identify all patients
requiring treatment, monitoring, and reperfusion
therapy to maximize outcomes without overdiagnosing.
The ACI-TIPI is one decision-support tool designed to
address this need.
It has proven to be an effective diagnostic test for
detecting ACI and cost-efficient at low to high rates of
ACI prevalence11 and shown to have excellent triage
accuracy for patients with ACI.
As a result, the ACI-TIPI should be considered to help in
the reduction of discovery to treatment times for STEMI
patients and used according to your organization’s policy
Aufderheide TP. Rowlandson I. Lawrence SW. Kuhn EM.
Selker HP. Test of the acute cardiac ischemia time-insensitive
predictive instrument (ACI-TIPI) for prehospital use. Annals
of Emergency Medicine. 27(2):193-8, 1996 Feb.
Mitchell AM. Garvey JL. Chandra A. Diercks D.
Pollack CV. Kline JA. Prospective multicenter study of
quantitative pretest probability assessment to exclude
acute coronary syndrome for patients evaluated in
emergency department chest pain units. Annals of
Emergency Medicine. 47(5):447, 2006 May
Chan GW, Sites FD, Shofer FS, Hollander JE. Impact of
stress testing on 30-day cariovascular outcomes for low-risk
patients with chest pain admitted to floor telemetry beds. Am J
Emerg Med. 21(4):282-7, 2003.
4 Arbelle JE, Porath A, Cohen E, Gilutz H, Garty M. Triage
disposition of patients with AMI--ACSIS 2000. Isr Med
Assoc J. 5(11):786-90, 2003.
5 Selker HP, Beshansky JR, Griffith JL, Aufderheide TP et al.
Use of the acute cardiac ischemia time-insensitive predictive
instrument (ACI-TIPI) to assist with triage of patients with
chest pain or other symptoms suggestive of acute cardiac
ischemia: A multicenter, controlled clinical trial. Annals of
Internal Medicine. 129(11)(pp 845-855), 1998.
Pozen MW, D'Agostino RB, Selker HP, Sytkowski
PA, and Hood WB. A predictive instrument to improve
coronary-care-unit admission practices in acute ischemic
heart disease. A prospective multicenter clinical trial. N
ENGL J Med 310: 1273-1278, 1984 May.
Selker HP. Griffith JL. D'Agostino RB. A tool for judging
coronary care unit admission appropriateness valid for both
real-time and retrospective use: A time-insensitive predictive
instrument (TIPI) for acute cardiac ischemia: a multicenter
study. Medical Care. 29:610-627, 1991.
Wilkinson K, Severance H. Identification of chest pain
patients appropriate for an emergency department observation
unit. Emerg Med Clin North Am 19(1):35-66, 2001.
Zalenski, R J; Shama, F; Waselewsky, D; Sherwin, R; Bock,
B; Kosnik, J Impact of ACI-TIPI on Resource Utilization in
Emergency Department Patients With Chest Pain. Annals of
Emergency Medicine. 34(4,2), 1999.
Westfall JM. Van Vorst RF. McGloin J. Selker HP. Triage
and diagnosis of chest pain in rural hospitals: implementation
of the ACI-TIPI in the High Plains Research Network. Annals
of Family Medicine. 4(2):153-8, 2006.
Milch C., Balk E., Salem D., Lau J. Diagnosing acute
cardiac ischemia in the emergency department: A
cost-effectiveness analysis. Journal of Applied Research.
Philips Healthcare is part
of Royal Philips Electronics
Koninklijke Philips Electronics N.V.
On the web
All rights are reserved.
Reproduction in whole or in part
is prohibited without the prior
written consent of the copyright
+31 40 27 64 887
By postal service
3000 Minuteman Road
Andover, MA 01810-1085
Tel: +852 2821 5888
Europe, Middle East, and Africa
Tel: +49 7031 463 2254
Tel: +55 11 2125 0744
Tel: +425 487 7000
1 800 285 5585 (USA only)
Philips Healthcare reserves the right
to make changes in specifications or
to discontinue any product at any
time without notice or obligation
and will not be liable for any
consequences resulting from the use
of this publication.
Published May 2009, Edition 1
Printed in the USA