Summary
Background
Compensatory mechanisms activated after myocardial infarction include an increase in systolic wall stress (SWS) and activation of the neurohormonal system. Nevertheless, left ventricular ejection fraction (LVEF) and infarct size are the established primary predictors of outcome after ST-segment elevation myocardial infarction.
Aims
To assess the relative impact of various cardiac magnetic resonance (CMR) imaging variables, such as infarct size, LVEF and SWS, on pre- and post-discharge heart failure (HF).
Methods
CMR was performed in a prospective study involving 169 patients with first ST-segment elevation myocardial infarction. Common CMR findings, such as SWS, were computed.
Results
Mean SWS was 16.3 ± 5.1 × 10 3 N·m −2 , and was systematically higher in patients exhibiting either pre- or post-discharge HF (18.9 ± 5.7 and 21.3 ± 7.6 × 10 3 N·m −2 , respectively). SWS was moderately related to initial infarct size ( r = 0.405; P < 0.001). In total, 28 patients presented with HF during the hospitalization phase and 14 during follow-up, with a median time of event of 93 days (25th–75th percentiles, 29–139.25 days). The univariate predictors of HF were age, LVEF, infarct size, SWS, microvascular obstruction, anterior infarction and heart rate at admission. Multivariable analysis revealed infarct size and age to be the predictors of predischarge HF, while SWS and heart rate at admission predicted post-discharge HF. The greatest SWS quartile provided a negative predictive value of 95.9%.
Conclusion
Regardless of LVEF and infarct size, SWS was shown to be an independent predictor of post-discharge HF after ST-segment elevation myocardial infarction.
Résumé
Contexte
La survenue d’un infarctus du myocarde est responsable d’une augmentation du stress pariétal systolique (SWS) et de l’activation du système neurohormonal. Cependant, la fraction d’éjection du ventricule gauche et la taille d’infarctus restent les deux principaux déterminants d’événements cliniques au décours.
Objectifs
Nous avons cherché à établir l’impact relatif de différents paramètres issus de l’IRM cardiaque tels que la taille d’infarctus, la fraction d’éjection et le SWS, et ceci à la fois sur la survenue d’épisode d’insuffisance cardiaque au cours de la phase hospitalière, mais aussi au cours du suivi ultérieur.
Méthodes
Une analyse par IRM cardiaque a été réalisée à la phase hospitalière et répétée à 3 mois chez 169 patients consécutifs présentant un infarctus du myocarde inaugural avec élévation du segment ST.
Résultats
Vingt-huit patients ont présenté un épisode d’insuffisance cardiaque lors de l’hospitalisation initiale et 14 pendant leur suivi avec une médiane de survenue de 93 jours (25 e –75 e percentiles, 29–139,25 jours). La valeur moyenne du SWS était de 16,3 ± 5,1 × 10 3 N·m −2 et était retrouvée systématiquement plus élevée chez les patients qui présentaient un épisode d’insuffisance cardiaque, que ce soit hospitalier ou lors du suivi (respectivement 18,9 ± 5,7 et 21,3 ± 7,6). Le SWS était faiblement corrélé à la taille d’infarctus initiale ( r = 0,405 ; p < 0,001). L’analyse multivariée montrait que la taille d’infarctus et l’âge étaient les paramètres prédictifs de la survenue d’insuffisance cardiaque lors de la phase hospitalière alors que le SWS et la fréquence cardiaque à l’admission prédisaient la survenue d’une insuffisance cardiaque au cours du suivi ultérieur. Le SWS le plus élevé (4 e quartile) présentait une valeur prédictive négative de 95,9 %.
Conclusion
Quelles que soient les fractions d’éjection et taille d’infarctus, le stress pariétal systolique est un paramètre prédictif indépendant de la survenue d’insuffisance cardiaque au cours du suivi des patients ayant présenté un infarctus du myocarde avec le sus-décalage du segment ST.
Background
The current modalities used to manage acute myocardial infarction with ST-segment elevation have improved patient survival rates . Nevertheless, the loss of contractile performance is still accompanied by an increase in neurohormonal activity and wall stress; these are exerted on both infarcted and healthy remote myocardial areas, and induce both metabolic and histological changes . An excessive increase in these compensatory mechanisms induces left ventricular remodelling, with left ventricular enlargement being a precursor of heart failure (HF) and death. HF is, in fact, the most common event during post-myocardial infarction patient management, and remains a matter of grave public health concern.
Over the last few decades, cardiac imaging has been specifically dedicated to assessing individual prognosis issues , with a large range of variables described. Some are functional, such as left ventricular ejection fraction (LVEF) and left ventricular volumes , others are directly related to the ischaemic injury in question, such as infarct size , a factor that was even considered a surrogate marker for therapeutic efficacy, while others are related to reperfusion success, such as microvascular obstruction assessment . Alongside these variables, and given the fact that left ventricular remodelling (LVR) implicates changes in left ventricular shape that may be investigated through geometrical factors, we wished to share our interest in evaluating end-systolic wall stress (SWS). This may, in fact, be considered a load-dependent functional variable, and was found to be independently related to cardiovascular prognosis and, more precisely, HF events.
In this study, we evaluated acute myocardial infarction patients with the aim of determining which cardiac magnetic resonance (CMR) variables out of left ventricular SWS, infarct size and MVO, assessed during the initial phase are independent predictors of HF at both pre- and post-discharge levels.
Methods
Study population
A total of 169 patients with acute myocardial infarction, referred to our catheterization laboratory for emergency primary transcutaneous coronary angioplasty, were evaluated prospectively. Patients were enrolled if they met the following criteria: ST-segment elevation myocardial infarction within 12 hours of chest pain onset; successful angioplasty with stent implantation; written informed consent provided. The exclusion criteria were as follows: history of myocardial infarction or coronary bypass grafting; age < 18 years; clinical signs of cardiogenic shock; major co-morbidities limiting life expectancy; contraindications for CMR (pacemaker, metallic devices, claustrophobia or chronic renal insufficiency). This study conformed to the principles outlined in the declaration of Helsinki. All patients provided written informed consent for completion of the CMR, and the study protocol was approved by the hospital’s ethics committee (CHU Angers).
Imaging technique
CMR was performed using a 1.5 Tesla imager (Avanto, Siemens, Erlangen, Germany) with the application of an eight-element phased-array cardiac receiver coil. All imaging was obtained at the end of expiration in order to reduce the ventricular volume changes induced by Valsalva manoeuver in forced inspiration. Left ventricular function was analysed using the steady-state free precession sequence performed on contiguous short-axis slices covering the entire left ventricle. The typical in-plane resolution applied was 1.6 × 1.9 mm, with a 7 mm section thickness (repetition time/echo time, 2.6 ms/1.30 ms; flip angle, 80°; matrix, 256 × 208; temporal resolution, 35–45 ms).
Late gadolinium enhancement (LGE) sequences were performed 12–15 minutes after the injection, at a dose of 0.2 mmol/kg, by means of a two-dimensional segmented inversion recovery gradient-echo pulse sequence. An initial inversion time scout sequence was conducted to determine the optimal inversion time. Contiguous short-axis slices covered the entire ventricle. The typical in-plane resolution used was 1.68 × 1.68 mm, with a 7 mm section thickness (echo time, 4.66 ms; flip angle, 30°; imaging triggered to every other heartbeat; matrix, 256 × 208). Steady-state free precession pulse sequences and LGE sequences were acquired in breathhold state, each with identical section positioning.
Image analysis
The CMR images were transferred to a workstation for analysis and calculation (QMass 7.1, Medis, Leiden, The Netherlands).
Left ventricular function
On all short-axis cine slices, the endocardial and epicardial borders were outlined manually on end-diastolic and end-systolic images, excluding the trabeculae and papillary muscles. Left ventricular end-diastolic and end-systolic volumes, such as left ventricular mass, were determined.
Infarct size measurement
Infarct size was quantified on LGE images by means of the ‘full width at half maximum’ method , corresponding to the sum of the LGE extent measured on all sections, given in grams.
Microvascular assessment
If present, central hypoenhancement was manually delineated, and its extent was systematically added to the hyperenhanced area. The variability assessment for left ventricular volumes, infarct size and MVO extent produced good results, published elsewhere .
Wall stress measurement
Global wall stress was calculated using dedicated software, specially built by our laboratory, using a three-dimensional model analysis . In brief, a median border between endo- and epicardial borders was generated on each slice. The barycentre of the section was then defined as the mass centre of the median border. Each short-axis was centred on the barycentre. The radius of curvature and wall thickness were calculated on end-systole in a series of contiguous short-axis slices (5–12 sections, depending on heart size) in order to compute the SWS . All apical slices absent of ventricular cavity and basal slices presenting open borders were excluded from the analysis. The SWS was calculated on each slice, with the SWS of the whole heart (global wall stress), defined as the average value of all slices, then used for the statistical analyses.
Wall stress was calculated using the modified Grossman formula : 0.133 × SBP × (R/[2Wt × (1 + Wt/2R)]), where SBP is systolic blood pressure, R is radius of curvature and Wt is wall thickness. Three measurements were taken from the SBP cuff during the acquisition of cine magnetic resonance imaging.
Outcome measures
Outcomes were defined by a consensus achieved between two physicians blinded to the CMR results in accordance with the ESC guidelines . We distinguished between deaths from cardiovascular causes and other causes. Death and hospitalization for HF were tabulated per subject. Patients who died were judged irrelevant for the HF-dedicated analysis and thus excluded. We reported all cases of HF during hospitalization and follow-up.
Statistical analysis
Data are presented as means ± standard deviations or medians (25th–75th percentiles) in cases of non-normal distribution, with categorical data expressed as frequencies and percentages. Comparisons of variables were performed using analysis of variance, the unpaired Student’s t -test or the Chi 2 test, where appropriate. We compared the baseline with 3-month quantitative CMR variables by means of the Wilcoxon test. Initial and follow-up CMR findings were compared with initial SWS and infarct size using the Pearson’s correlation coefficient. A stepwise binomial logistic regression analysis was performed to identify the determinants of SWS. LVEF and left ventricular volumes were not included in the model because of a high level of collinearity.
For multivariable analysis assessing HF during hospitalization, clinical and CMR data were tested by means of a stepwise binomial logistic regression analysis, including variables with P values < 0.05. The Hosmer-Lemeshow statistic was used to assess the goodness-of-fit of the applied models. A similar analysis was performed to identify predictors of any level of HF.
For multivariable analysis assessing HF during follow-up, clinical and CMR data were tested by means of a stepwise Cox analysis, including variables with P values < 0.05. The proportional-hazard assumptions were tested by analysing the Schoenfeld residuals.
The analyses were performed using SPSS version 15.0 for Windows (SPPS Inc., Chicago, IL, USA). A P value < 0.05 was considered statistically significant.
Results
Patient characteristics in terms of heart failure
This study included 169 patients presenting with a first episode of acute myocardial infarction with persistent ST-segment elevation, referred between January 2008 and May 2012. The patient characteristics are summarized in Table 1 . We observed that the population with pre- and post-discharge HF exhibited more anterior infarctions with higher creatine kinase peaks (75.0% with 4839 ± 2643 IU/L and 85.7% with 5709 ± 2935 IU/L, respectively). Discharge treatments systematically combined dual anti-platelet aggregation, angiotensin-converting enzyme inhibitors and statins. Two patients were not administered beta-blocker treatment because of a history of asthma; 57 patients (33.7%) received an aldosterone blocker.
| Total | No predischarge HF | Predischarge HF ( n = 28) | P a | No post-discharge HF | Post-discharge HF ( n = 14) | P a | |
|---|---|---|---|---|---|---|---|
| Age (years) | 58.0 ± 11.6 | 56.8 ± 10.9 | 61.3 ± 13.0 | 0.001 | 57.0 ± 11.3 | 62.0 ± 14.2 | 0.18 |
| BMI (kg/m 2 ) | 27.2 ± 4.1 | 26.8 ± 4.0 | 28.9 ± 4.3 | 0.011 | 27.2 ± 4.1 | 26.7 ± 4.5 | 0.68 |
| Risk factors | |||||||
| Male | 140 (85.9) | 117 (86.7) | 23 (82.1) | 0.36 | 128 (85.9) | 12 (85.7) | 0.62 |
| Hypertension | 58 (35.6) | 44 (32.6) | 14 (50.0) | 0.07 | 50 (33.6) | 8 (57.1) | 0.07 |
| Diabetes mellitus | 25 (15.3) | 21 (15.6) | 4 (14.3) | 0.56 | 22 (14.8) | 3 (21.4) | 0.37 |
| Hypercholesterolaemia | 86 (52.8) | 70 (51.9) | 16 (57.1) | 0.38 | 79 (53.0) | 7 (50.0) | 0.52 |
| Tobacco use | 74 (45.4) | 62 (45.9) | 12 (42.9) | 0.89 | 69 (46.3) | 5 (35.7) | 0.41 |
| Characteristics at admission | |||||||
| SBP (mmHg) | 141.2 ± 26.9 | 141.0 ± 25.6 | 142.1 ± 32.8 | 0.84 | 141.4 ± 27.3 | 138.8 ± 21.7 | 0.73 |
| Heart rate (beats/min) | 75.5 ± 18.5 | 75.5 ± 17.9 | 75.5 ± 21.4 | 0.98 | 74.4 ± 17.7 | 87.1 ± 22.6 | 0.012 |
| LAD culprit lesion | 91 (55.8) | 70 (51.9) | 21 (75.0) | 0.019 | 79 (53.0) | 12 (85.7) | 0.016 |
| Multivessel disease | 73 (44.8) | 61 (45.1) | 12 (42.8) | 0.38 | 68 (45.6) | 5 (35.7) | 0.75 |
| Creatinine (μmol/L) | 84.1 ± 27.1 | 82.6 ± 27.2 | 91.5 ± 26.3 | 0.11 | 82.3 ± 20.6 | 104.2 ± 62.4 | 0.003 |
| Blood glucose (mmol/L) | 9.0 ± 6.3 | 8.9 ± 6.7 | 9.4 ± 3.8 | 0.67 | 8.9 ± 6.5 | 9.3 ± 3.0 | 0.83 |
| Creatine kinase peak (IU/L) | 3085 ± 2196 | 2718 ± 1908 | 4839 ± 2643 | < 0.001 | 2836 ± 1950 | 5709 ± 2935 | < 0.001 |
| HbA1c (%) | 6.2 ± 1.3 | 6.2 ± 1.3 | 6.2 ± 1.4 | 0.76 | 6.2 ± 1.3 | 6.0 ± 0.8 | 0.59 |
| Haemoglobin (g/dL) | 15.0 ± 1.6 | 15.0 ± 1.5 | 15.0 ± 2.0 | 0.86 | 14.9 ± 1.5 | 15.5 ± 2.4 | 0.26 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
