Exercise testing scores as an example of better decisions through science

Automated interpretation of stress echocardiography reports using natural language processing

Aims

Stress echocardiography (SE) findings and interpretations are commonly documented in free-text reports. Reusing SE results requires laborious manual reviews. This study aimed to develop and validate an automated method for abstracting SE reports in a large cohort.

Methods and results

This study included adult patients who had SE within 30 days of their emergency department visit for suspected acute coronary syndrome in a large integrated healthcare system. An automated natural language processing (NLP) algorithm was developed to abstract SE reports and classify overall SE results into normal, non-diagnostic, infarction, and ischaemia categories. Randomly selected reports (n = 140) were double-blindly reviewed by cardiologists to perform criterion validity of the NLP algorithm. Construct validity was tested on the entire cohort using abstracted SE data and additional clinical variables. The NLP algorithm abstracted 6346 consecutive SE reports. Cardiologists had good agreements on the overall SE results on the 140 reports: Kappa (0.83) and intraclass correlation coefficient (0.89). The NLP algorithm achieved 98.6% specificity and negative predictive value, 95.7% sensitivity, positive predictive value, and F-score on the overall SE results and near-perfect scores on ischaemia findings. The 30-day acute myocardial infarction or death outcomes were highest among patients with ischaemia (5.0%), followed by infarction (1.4%), non-diagnostic (0.8%), and normal (0.3%) results. We found substantial variations in the format and quality of SE reports, even within the same institution.

Conclusions

Natural language processing is an accurate and efficient method for abstracting unstructured SE reports. This approach creates new opportunities for research, public health measures, and care improvement.

French Society of Cardiology guidelines on exercise tests (part 1): Methods and interpretation

The exercise test is still a key examination in cardiology, used for the diagnosis of myocardial ischemia, as well as for the clinical evaluation of other heart diseases. The cardiopulmonary exercise test can further define functional capacity and prognosis for any given cardiac pathology. These new guidelines focus on methods, interpretation and indications for an exercise test or cardiopulmonary exercise test, as summarized below. The safety rules associated with the exercise test must be strictly observed. Interpretation of exercise tests and cardiopulmonary exercise tests must be multivariable. Functional capacity is a strong predictor of all-cause mortality and cardiovascular events. Chest pain, ST-segment changes and an abnormal ST/heart rate index constitute the first findings in favor of myocardial ischemia, mostly related to significant coronary artery disease. Chronotropic incompetence, abnormal heart rate recovery, QRS changes (such as enlargement or axial deviations) and the use of scores (based on the presence of various risk factors) must also be considered in exercise test interpretation for a coronary artery disease diagnosis. Arrhythmias or conduction disorders arising during the exercise test must be considered in the assessment of prognosis, in addition to a decrease or low increase in blood pressure during the exercise phase. When performing a cardiopulmonary exercise test, peak oxygen uptake and the volume of expired gas/carbon dioxide output slope are the two main variables used to evaluate prognosis.

Prognostic value of electrocardiogram exercise testing for risk stratification in asymptomatic coronary artery disease

BACKGROUND

Several variables of electrocardiogram exercise testing (EET) predict cardiovascular events in the general population and in patients with coronary artery disease (CAD). However, most of the studies have not included patients with asymptomatic CAD. The aim of this study was to evaluate the prognostic value of EET in asymptomatic CAD patients.

PATIENTS AND METHODS

We carried out a retrospective single-center analysis including all patients with asymptomatic CAD documented by angiography who underwent EET from January 2010 to December 2013. A number of EET variables and three exercise scores [Duke Treadmill Score (DTS), Morise score, and FIT score] were analyzed. The primary endpoint was the combined incidence of myocardial infarction (MI), myocardial revascularization, and death from any cause during follow-up.

RESULTS

A total of 306 patients were included (mean age was 65±10 years, 61% had previous MI, and the median exercise capacity was 9.4±2.7 metabolic equivalent of task). The primary endpoint occurred in 15.7% of patients during 3.3 years of follow-up. The DTS and FIT were independent predictors of the primary endpoint unlike the Morise score (DTS: hazard ratio=0.91, 95% confidence interval: 0.85-0.99, P=0.018; FIT score: 0.99, 0.98-0.996, P=0.001; Morise score: 0.97, 0.93-1.02, P=0.20). The DTS was independent predictor of MI or revascularization, whereas FIT predicted death from any cause. Excluding patients with early revascularization, DTS had no predictive power at the composite endpoint.

CONCLUSION

In our population with asymptomatic CAD, FIT and DTS had significant value for risk prediction and consequently the EET can be a valid tool in the clinical follow-up of this population.

The use of pre-test and post-test probability values as criteria before selecting patients to undergo coronary angiography in patients who have ischemic findings on myocardial perfusion scintigraphy

Objective:

Myocardial perfusion scintigraphy (MPS) is a diagnostic test which is frequently used in the diagnosis of coronary heart disease (CHD). MPS is generally interpreted as ischemia present or absent; however, it has a power in predicting the disease, similar to other diagnostic tests. In this study, we aimed to assist in directing the high-risk patients to undergo coronary angiography (CA) primarily by evaluating patients without prior CHD history with pre-test and post-test probabilities.

Methods:

The study was designed as a retrospective study. Between January 2008 and July 2011, 139 patients with positive MPS results and followed by CA recently (<6 months) were evaluated from patient files. Patients’ pre-test probabilities based on the Diamond and Forrester method and the likelihood ratios that were obtained from the literature were used to calculate the patients’ post-exercise and post-MPS probabilities. Patients were evaluated in risk groups as low, intermediate, and high, and an ROC curve analysis was performed for the post-MPS probabilities.

Results:

Coronary artery stenosis (CAS) was determined in 59 patients (42.4%). A significant difference was determined between the risk groups according to CAS, both for the pre-test and post-test probabilities (p<0.001, p=0.024). The ROC analysis provided a cut-off value of 80.4% for post-MPS probability in predicting CAS with 67.9% sensitivity and 77.8% specificity.

Conclusion:

When the post-MPS probability is ≥80% in patients who have reversible perfusion defects on MPS, we suggest interpreting the MPS as “high probability positive” to improve the selection of true-positive patients to undergo CA, and these patients should be primarily recommended CA.

Development of a composite model derived from cardiopulmonary exercise tests to predict mortality risk in patients with mild-to-moderate heart failure

OBJECTIVE

Cardiopulmonary exercise testing (CPET) is used to predict outcome in patients with mild-to-moderate heart failure (HF). Single CPET-derived variables are often used, but we wanted to see if a composite score achieved better predictive power.

METHODS

Retrospective analysis of patient records at the department of cardiology, Castle Hill Hospital, Kingston-upon-Hull. 387 patients (median (25th-75th percentile)) (age 65 (56-72) years; 79% men; LVEF 34 (31-37) %) were included. Patients underwent a symptom-limited, maximal CPET on a treadmill. During a median follow-up of 8.6±2.1 years in survivors, 107 patients died. Survival models were built and validated using a hybrid approach between the bootstrap and Cox regression. Nine CPET-derived variables were included. Z-score defined each variable's predictive strength. Model coefficients were converted to a risk score.

RESULTS

Four CPET-related variables were independent predictors of all-cause mortality in the survival model: the presence of exertional oscillatory ventilation (EOV), increasing slope of the relation between ventilation and carbon dioxide production (VE/VCO2 slope), decreasing oxygen uptake efficiency slope (OUES), and an increase in the lowest ventilatory equivalent for carbon dioxide (VEqCO2 nadir). Individual predictors of mortality ranged from 0.60 to 0.71 using Harrell's C-statistic, but the optimal combination of EOV+VE/VCO2 slope+OUES+VEqCO2 nadir reached 0.75. The Hull CPET risk score had a significantly higher area under the curve (0.78) when compared to the HF Survival Score (AUC=0.70; p<0.001).

CONCLUSIONS

A composite risk score using variables from CPET out-performs the traditional single variable approach in predicting outcome in patients with mild-to-moderate HF.

Validation of a cardiopulmonary exercise test score in heart failure

BACKGROUND

Cardiopulmonary exercise test (CPX) responses are strong predictors of outcomes in patients with heart failure. We recently developed a CPX score that integrated the additive prognostic information from CPX. The purpose of this study was to validate the score in a larger, independent sample of patients.

METHODS AND RESULTS

A total of 2625 patients with heart failure underwent CPX and were followed for cardiovascular (CV) mortality and major CV events (death, transplantation, left ventricular assist device implantation). Net reclassification improvement (NRI) for the score and each of its components were determined at 3 years. The VE/VCO2 slope was the strongest predictor of risk and was attributed a relative weight of 7, with weighted scores for abnormal heart rate recovery, oxygen uptake efficiency slope, end-tidal CO2 pressure, and peak VO2 having scores of 5, 3, 3, and 2, respectively. A summed score of >15 was associated with an annual mortality rate of 12.2% and a relative risk >9 for total events, whereas a score of <5 was associated with an annual mortality rate of 1.2%. The composite score was the most accurate predictor of CV events among all CPX responses considered (C indexes, 0.70 for CV mortality and 0.72 for the composite outcome). Each component of the score provided significant NRI compared with peak VO2 (category-free NRI, 0.61-0.77), and the score provided significant NRI above clinical risk factors for both CV events and mortality (NRI, 0.63 and 0.65 for CPX score compared with clinical variables alone).

CONCLUSIONS

These results validate the application of a simple, integrated multivariable score based on readily available CPX responses.

The Clinical Utility of Cardiopulmonary Exercise Testing in Suspected or Confirmed Myocardial Ischemia

Heart disease is a major cause of morbidity and mortality in the United States, with coronary artery disease (CAD) representing more than half of all cardiovascular events. Stable patients presenting with symptoms suggestive of CAD are likely to undergo an exercise electrocardiogram (ECG) and/ or imaging study as a first-line diagnostic assessment. A cardiopulmonary exercise test (CPX) is an ECG stress test plus ventilatory gas analysis. Recently, CPX has been used to detect exercise-induced myocardial ischemia (EIMI) suggestive of underlying CAD. Two CPX variables, oxygen pulse (VO2/HR) and the slope of oxygen consumption versus work rate (Δ VO2/ Δ WR), have been identified to be especially indicative of EIMI. Currently, there are a number of diagnostic tests available for the identification of CAD, with the most widely used being stress ECG, stress myocardial perfusion imaging (MPI) and echocardiography, and cardiac catheterization. Exercise ECG, although inexpensive, has a number of well-recognized limitations, including low sensitivity resulting in false-negative results. Stress (exercise or pharmaceutically induced) MPI and catheterization are more accurate but also more invasive and expensive. It appears that CPX may improve the diagnostic accuracy of exercise ECG. This review will address the potential utility of CPX in patients with suspected or confirmed myocardial ischemia.

The lowest VE/VCO2 ratio during exercise as a predictor of outcomes in patients with heart failure

BACKGROUND

The lowest minute ventilation (VE) and carbon dioxide production (VCO(2)) ratio during exercise has been suggested to be the most stable and reproducible marker of ventilatory efficiency in patients with heart failure (HF). However, the prognostic power of this index is unknown.

METHODS AND RESULTS

A total of 847 HF patients underwent cardiopulmonary exercise testing (CPX) and were followed for 3 years. The associations between the lowest VE/VCO(2) ratio, maximal oxygen uptake (peak VO(2)), the VE/VCO(2) slope, and major events (death or transplantation) were evaluated using proportional hazards analysis; adequacy of the predictive models was assessed using Akaike information criterion (AIC) weights. There were 147 major adverse events. In multivariate analysis, the lowest VE/VCO(2) ratio (higher ratio associated with greater risk) was similar to the VE/VCO(2) slope in predicting risk (hazard ratios [HR] per unit increment 2.0, 95% CI 1.1-3.4, and 2.2, 95% CI 1.3-3.7, respectively; P < .01), followed by peak VO(2) (HR 1.6, 95% CI 1.1-2.4, P=.01). Patients exhibiting abnormalities for all 3 responses had an 11.6-fold higher risk. The AIC weight for the 3 variables combined (0.94) was higher than any single response or any combination of 2. The model including all 3 responses remained the most powerful after adjustment for beta-blocker use, type of HF, and after applying different cut points for high risk.

CONCLUSIONS

The lowest VE/VCO(2) ratio adds to the prognostic power of conventional CPX responses in HF.

A cardiopulmonary exercise testing score for predicting outcomes in patients with heart failure

OBJECTIVE

The aim of this study is to evaluate the predictive accuracy of a cardiopulmonary exercise test (CPX) score.

BACKGROUND

Cardiopulmonary exercise test responses, including peak VO(2), markers of ventilatory inefficiency (eg, the VE/VCO(2) slope and oxygen uptake efficiency slope [OUES]), and hemodynamic responses, such as heart rate recovery (HRR) and chronotropic incompetence (CRI) are strong predictors of outcomes in patients with heart failure (HF). However, there is a need for simplified approaches that integrate the additive prognostic information from CPX.

METHODS

At 4 institutions, 710 patients with HF (568 male/142 female, mean age 56 +/- 13 years, resting left ventricular ejection fraction 33 +/- 14%) underwent CPX and were followed for cardiac-related mortality and separately for major cardiac events (death, hospitalization for HF, transplantation, left ventricular assist device implantation) for a mean of 29 +/- 25 months. The age-adjusted prognostic power of peak VO(2), VE/VCO(2) slope, OUES (VO(2) = a log(10)VE + b), resting end-tidal carbon dioxide pressure (PetCO(2)), HRR, and CRI were determined using Cox proportional hazards analysis, optimal cutpoints were determined, the variables were weighted, and a multivariate score was derived.

RESULTS

There were 175 composite outcomes. The VE/VCO(2) slope (> or =34) was the strongest predictor of risk and was attributed a relative weight of 7, with weighted scores for abnormal HRR (< or =6 beats at 1 minute), OUES (>1.4), PetCO(2) (<33 mm Hg), and peak VO(2) (< or =14 mL kg(-1) min(-1)) having scores of 5, 3, 3, and 2, respectively. Chronotropic incompetence was not a significant predictor and was excluded from the score. A summed score >15 was associated with an annual mortality rate of 27% and a relative risk of 7.6, whereas a score <5 was associated with a mortality rate of 0.4%. The composite score was the most accurate predictor of cardiovascular events among all CPX responses considered (concordance indexes 0.77 for mortality and 0.75 for composite outcome composed of mortality, transplantation, left ventricular assist device implantation, and HF-related hospitalization). The summed score remained significantly associated with increased risk after adjusting for age, gender, body mass index, ejection fraction, and cardiomyopathy type.

CONCLUSION

A multivariable score based on readily available CPX responses provides a simple and integrated method that powerfully predicts outcomes in patients with HF.

Quantifying physical activity in first-through fourth-grade physical education via pedometry

The aim of this study was to determine pedometry steps per minute (SPM) cutscores that accurately quantify physical activity (PA) time in first- through fourth-grade physical education. A total of 257 participants were grouped in two data pools, first- and second-grade (n = 126), and third- and fourth-grade (n = 131). Systematic observation was the PA criterion instrument and pedometry was the predictor instrument. Correlations between physical activity measures were strong (r = .82-.89, p < .01). Ten min of PA and 33.33% of the lesson time engaged in PA within a 30-min class can be quantified by 61-63 SPM for first- and second-grade, and 58-61 SPM for third- and fourth-grade. IN CONCOLUSION: (a) SPM values were a valid indicator of students achieving or not achieving PA criteria, and (b) pedometry is a valid and practical tool for physical activity surveillance within physical education.