Home SVCC Area: English - Español - Português
Prospective Validation of a Clinical Method
for the Diagnosis of the Syndrome
of Unstable Angina
Payares, Agustín; Patete, Alexis; Chavez, Parmy;
Millán, Miguel; Romero, Rodolfo; Márquez, Cira;
García, Merbin; VanGrieken, León; Gutierrez,
Sergio; Ortiz, Freddy; Lugo, Fanny.
Experimental National University "Francisco
de Miranda". Universitary Hospital
"Dr. Alfredo Vangrieken". Coro Falcón State Venezuela.
Introduction: The unstable angina (Ua) is a multifactorial and heterogeneous illness, whose syntomatologic and prognosis is very variable, and his diagnosis (Dg) is clinical and subjective.
Objectives: To design a method of Dg clinical measurable and objective, based on a chart of criteria (Cr) epidemiologist, semiologist and of laboratory of Ua in presence of precordial pain.
Design: Blind, randomized, cross sectional study of Ua. And risk factors of cardiovascular disease (RF), included the Independent Cr of the diabetes mellitus, to test the diagnostic power of a chart of Cr. We had classified the Dg of VERY PROBABLE Ua, PROBABLE Ua and UNLIKELY Ua, for the presence of one or more mayor or minor criterias, in which the RF was classified. These results were compared with our Gold Standard constituted by the simultaneous Dg of three clinical cardiologists that ignored the purpose of our study. It was considered positive if two or more was according with the Dg.
Statistical Analysis: Through the table 2x2, it was calculated: sensivity, specificity, positive predictive value, negative predictive value, accuracy, prevalence, Odds ratio, chi square and p value.
The Table of Dg of Ua is useful, highly predictive, of low cost, applicable in a big mass of populations , easy handling, effective, exactly and high sensitive to identify and classify the patients with UA that consult for angina pectoris associated to RF.
The angor pectoris (AP) classically is defined as a precordial pain, oppressive or constrictive character, of ischemic origin that increases with the exercise and diminishes with the rest, and constitutes the cardinal symptom of the acute coronary syndromes, they embrace a wide range clinical manifestations that: its go from the silent ischemia, the angina induced by the exercise, the unstable angina (UA) until the myocardial infarct (MI). The UA occupies the center of this spectrum, causing a high morbidities and a very high risk of MI and death (1).
The UA is a very heterogeneous illness whose origin are a coronary spasm, or a coronary obstruction due to the rupture and thrombosis of the atherosclerotic plaque, causing from a significant dynamic estenosis until a severe illness of three coronary arteries. Therefore it is not surprising that their sintoms and prognosis are so variable (2,20,21,22).
The coronaries lesions can be
of two types: The fissure or erosion of the endotelia that causes the formation
of the trombus on the plaque, and the rupture or laceration of the covered,
with a plaque rich in lipids which allows that over there enters blood into
the lipidic nucleus forming the trombus. Therefore; the rupture of a vulnerable
plaque contributes to the increase the obstruction of the lumen and alter the
geometry of the coronary estenosis, precipitating the UA(20,21,22) and its will
show clinically with one of this following variables:
· Isquemics symptoms in rest of sudden appearance, or precipitated by the physical effort, in a patient without a previous history of coronary arteries disease (CAD)
· Augmenting or change of the pattern of the isquemics symptoms in a patient with history of CAD
· Recurrence of the ischemic symptoms immediately or until four weeks after a MI (1).
It is of making notice that that the diagnosis of an UA is carried out based on clinical data, the objective demonstration of the signs of myocardial ischemia increases the accuracy of the diagnosis: the most frequently sign looked are the changes into the electrocardiogram (ECG) which can be documented in 70% of the cases (1), the previous history of the illness, the presence of risk factors and the advanced age increases the specificity of the diagnosis (2). The troponin T and I are more sensitive and more specific markers of ischemia with minimum myocardial necrosis, and the high levels are associate with a bad prognosis (3,4). Also has been documented that the systemic markers of inflammation like the Reactive C Protein , can provide an independent prognostic information (5 )
This analysis indicates that the diagnosis of an UA constitutes a challenge to the cardiologist, for the heterogeneous and multifactorial of the syntomatologic that in some cases are silent. For this reasons we decide to built and evaluate a clinical test, based on the presence of signs, symptoms and the presence of risk cardiovascular factors, to carry out a diagnostic test that was more objective and measurable for the diagnosis of UA.
MATERIAL AND METHODS
From January 1º, 1999 until January 7, 2000 we included 460 patients of different ages and sexes, whose consultation reason was a pain in the chest, with more than 10 minutes of duration, without taking into account other associate factors, and that it gathered one of this following characteristic:
1. Suspicion of UA: for the appearance of complex symptoms of a increasing ischemic process, or a rest angor with absence of changes in the ST segment of the ECG of 12 derivations
2. Definitive UA: complex symptoms of angina, increcendo or a "novo" pattern with or without angor in rest in presence of changes in the ST segment of the ECG compatible with ischemia.
3. Secondary angina: symptoms of complex angina in rest that was precipitated by a condition extra cardiac or an arrhythmia (6).
We settled down as condition "sine qua non" that the symptom has induced to the patient to consult at our Hospital, and independent to the diagnosis we applied the test based on the chart of diagnoses of UA.
Criterion diagnoses of UA: As the diagnosis of the UA is eminently clinical (2,7), based on three approaches that include: Epidemiology criterion (antecedents and risk factors), Semiologic criterion (interrogation and physical exam) and the Paraclinic criterion (laboratory), we build a diagnostic chart () basing of these three items like it continues:
The Epidemiology criterion: we subclassify into mayors and minors. In mayors we included: Masculine sex, age >40 years in the man and >45 years in the woman, menopause, more than two labor, tobacco habits , hiperlipidemia, arterial hypertension (AH), and personal antecedent of CAD; and in the minors we include : family antecedents of CAD, stress, sedentary and the obesity.
The Semiologic criterion: We classifies them in mayors for the presence of a precordial pain of more than 10 minutes that it increases with the exercise and diminish with the rest, AH 3C (8), Acute Pulmonary Edema, Heart Rate more than 140 per minute or minor than 55, or a murmur mitral insufficiencies "de novo" or a changing pattern; and in the minors criterion we include: the precordial pain of atypical character, the presence of 4th or 3th sound, and a murmur of mitral insufficiencies .
The paraclinic criterion: includes as mayors, the presence of myocardial ischemia, in the ECG or in the Holter of analysis of the ST segment, the presence of Troponins T or I. and signs of lung congestion into the X ray, and the minor criterion include the presence of alterations of the myocardial motilities seen through the Bidimensional Echocardiography .
Taking into account the high prevalence of alterations of the endothelial, function and the hiperlipidemia in patients with diabetes mellitus (DM) (9,10), we classifies this illness like mayor and independent criterion , settling down in our study that: All precordial pain that justify a medical consultation, in a patient with DM, is an UA, until to demonstrate the opposite.
We classify the diagnosis of VERY PROBABLE UA if the precordial pain was presented of typical character, associated to 1 or more mayor criterion, or under the presentation of a pain typical precordial associated to two or more minor criterion or as a atypical precordial pain, jointly to two or more mayor criterion of different category. We classified the PROBABLE UA when it was presented as a atypical precordial pain associated to 2 or more minor criterion, or under the form of a pain typical precordial in absence of risk factors; and the UNPROBABLE UA when the pain precordial was presented in atypical character with 1 minor criterion or in absence of factors of risk ().
We compare this classification with our "gold standard" (11,19) which conformed to with the simultaneous diagnostic opinion of three cardiologists that ignored the purpose of our study, and we take as definitive diagnosis of VERY PROBABLE or PROBABLE UA if two or three cardiologists respectively was agreed with the diagnosis of UA, and we excluded the diagnosis of UA when two cardiologists or the three didn't agree. We excluded from our study the patients that entered to the emergency with the diagnosis of MI.
Statistical analysis: We calculated the statistical descriptive to the population included in the study, and the data obtained from the diagnostic chart was analyzed, through a 2x2 table (11,19) to compare it with our gold standard. The following parameters were calculated: Sensibility, Specificity, Predictive Positive Value (PPV), Predictive Negative Value (PNV), Accuracy, Prevalence, Odds Ratio, Chi squared (X2) and p value (12)
The data obtained were represented in form of charts.
Of the total of studied patients (n=460) they corresponded 51% to male and 49% to female, with an Average of 61,2±15,3 year for the men and 60,2±13,3 years for the women.
We obtained from the analysis of the 2x2 chart that the 81%(n=372) corresponded to the true positive, and the 4%(n=22) corresponded to the false negative, coming off that the 85%(n=394) carried the illness. With regard to the false positive they represented the 11%(n=50) and the true negatives the 3%(n=16), therefore we deduce that the 14%(n=66) didn't have the illness.
In concerning with the analysis of the stable properties of the 2x2 table, we find that the diagnostic chart of UA possesses a Sensibility =94% and a Specificity =24%; and with regard to dependent properties of the frequency we obtained PPV=88%, PNV=42%, Accuracy=84% and Prevalence=86%. Odds Ratio= 5,41(2,47-11,56), with a value of X2 (corrected by Yates)=23,57 and a p value=<0,0000012. ()
It was obtained that the classification of VERY PROBABLE UA have a 95% of probabilities that the patient has the syndrome of UA, the classification of PROBABLE UA have a 58% of probabilities that the patient possesses an UA and the UNLIKELY UA have a 5% of probabilities of UA. ()
The UA is not a specific illness, is a clinical syndrome (13,14) and it is usually classified by the use of simple describers, such as the presence or absence of precordial pain, changes in the ECG and the presence of biochemical markers of lesion in the myocites (14), and in this form of clinical presentation intervene many factors. Their cardinal symptom is the precordial pain, however; although the AP has been one of the syndromes more studied in the medicine, continuous being a challenge the precise diagnosis of the pathology that originates it.
In the outlined lapse of our investigation was found the frequency of consultation for AP was of 1,26 patients/day, and the frequency of hospitalization was 1,17 patients per day, observing a proportion with respect the sex of 0,96:1, in opposite to reported for other investigators (15). The ages oscillated around the 60 years in both sexes; associated to a high prevalence of cardiovascular risk factors in our series. It got us the attention the similar incidence in both sexes, attributing it to the similarity of the risk of CAD in the menopausal woman. We appreciated that the chart was able to identify correctly in 94% to the patient with UA, and with regard to the percentage of patients identified positive that really had the illness we detect that it was of 88%. We found a Prevalence of 86%, nevertheless its capacity to detect the individuals healthy payees of a precordial pain was very low: 24%, with a capacity to identify those individuals with a negative result that they didn't really have the illness in 42%. In consequence we being able to appreciate that the test is really effective to identify and to diagnose the really sick individuals, being able to detected an Accuracy of the chart of 84%; being completed one of the objectives of our study. With regard to the little capacity to detect the healthy individuals, we believe that it is an intrinsic property of the chart, since it was devised to carry out the diagnosis of the individuals with the illness, on the other hand, it is of making notice that the drop specificity and NPV of our Chart it is shared for some methods of clinical and paraclinic diagnoses used nowadays (16,17,18).
The patients that we classify
as VERY PROBABLE AI had a probability of having the illness of 95%, the
patients in the class PROBABLE UA reached a probability of having the
illness of 58%, and the patients contained as UNPROBABLE UA they presented
a probability of having the illness of 5% (
1. CLASS 1 (VERY PROBABLE UA): with 95% of probabilities of guessing right the diagnosis, in other words: is a patient of high risk, in consequence the patient should enter to a Unit of Coronary Cares and to begin the treatment immediately.
2. CLASS 2 (PROBABLE UA): patient of moderate risk; with 58% of probabilities of guessing right the diagnosis, therefore it can be studied in an area of cares intermissions.
3. CLASS 3 (UNPROBABLE UA) it possesses 5% of probabilities of guessing right the diagnosis, so that is a patient of low risk, therefore it can be studied it in an ambulatory way through the external consultation.
We notice in the measures of association calculated that those patients with a precordial pain identified by the chart like VERY PROBABLE AI or PROBABLE AI possess a risk of 5,41 times more to have an unstable angina , and this diagnosis was statistically significant in connection with our gold standard (X2=23,57, p=<0.0000012).
The most important of this method diagnosis rests in their simplicity and use easiness, to the point that can be used by a doctor without experienced in the cardiologic diagnosis, to decide a quick and appropriate treatment to the case, and you can apply on the base of objective approaches in all the patients with suspicion of unstable angina that possess cardiovascular risk factors, that it is where it is really effective the chart. In connection with the patient with a precordial pain and they don't have the illness, it becomes obligatory to create an exam paraclinic that is highly sensitive and specific, of easy access, applicable to big number of population, of low cost and that it really excludes the illness.
- The diagnostic chart of Unstable Angina is useful, highly predictive, of very low cost, of easy handling, effective, exact and sensitive to identify and to classify to the patient with UA that consult for precordial pain, associated to the presence of cardiovascular risk factors.
- You can apply in every scenario to big number of population, for any professional of the medicine, with high probabilities of making an exact diagnosis.
- Is low specific to identify the patients without coronary illness that consult for precordial pain.
- Permit to classify the patients with angor according to the risk of presenting the illness, settling down that the patients included in the CLASS 1 and CLASS 2, are considered of risk; therefore they should be hospitalized, to be evaluated and to begin the treatment immediately.
- The prevalence of UA in our country is high
Your questions, contributions and commentaries
answered by the authors in the Ischemic Heart Disease list.
Please fill in the form (in Spanish, Portuguese or English) and press the "Send" button.
2nd Virtual Congress of Cardiology
Dr. Florencio Garófalo
Dr. Raúl Bretal
Dr. Armando Pacher
Technical Committee - CETIFAC
Copyright© 1999-2001 Argentine Federation of Cardiology
All rights reserved
This company contributed to the Congress: