Prognostic value of exercise stress echocardiography in patients with known coronary artery disease

Funding Acknowledgements

Type of funding sources: None.

Background

Exercise stress echocardiography (EE) is useful for diagnosis and risk stratification in patients (pts) with suspected coronary artery disease (CAD). Pts with known CAD carry a high risk of events. Our aim was to assess the utility of EE in predicting outcomes in this population.

Methods

Single center retrospective study of consecutive pts with known CAD who performed an EE between 2018 and 2019. The primary endpoint was a composite of admission for acute coronary syndrome (ACS), coronary revascularization and cardiovascular death during the follow-up.

Results

A total of 76 pts were included (mean age 59±9 years; 87% male). Fifty-nine (78%) pts had history of ACS, 51 (67%) of percutaneous coronary intervention (PCI) and 14 (18%) of coronary artery bypass graft. The majority of pts had 2 or more vessel disease (42 pts; 55%).

The main reason for performing EE was new onset of chest pain (38 pts; 50%) followed by functional assessment of coronary stenoses after incomplete revascularization (29 pts; 38%). The majority of pts had a preserved left ventricular ejection fraction (67; 88%). The exam was performed under beta-blocker effect in 35 (46%) pts.

The results of EE were positive for myocardial ischemia in 7 (9%) pts, negative in 37 (49%) pts and inconclusive in 32 (42%) pts. Mean exercise time was 8±3 minutes and mean METs achieved was 9.4±2.6. ST-segment depression fulfilling electrocardiographic criteria for ischemia occurred in 17 (22%) pts. Eleven (14%) pts complained of chest pain during the exam.

During a median follow up of 22 months (IQR 15-26), the primary endpoint occurred in 9 pts (admission for ACS in 5 pts; revascularization for chronic coronary syndrome in 4 pts). No cardiovascular death occurred. The positive predictive value of EE for the primary endpoint was 57.1% and the negative predictive value (NPV) 97.3%.

In a survival analysis, the predictors of the primary endpoint were lower exercise time, lower METs achieved, untreated significant coronary lesions and a positive EE (figure).

After adjustment in a multivariate analysis, a positive EE was an independent predictor of the primary endpoint (HR 4.6, 95%CI: 1.1-16.7, p=.044).

Conclusion

In pts with known CAD, EE had a high NPV and a positive result was an independent predictor of future cardiovascular events.

A left systolic dysfunction study in a group of breast cancer patients and its clinical impact

Funding Acknowledgements

Type of funding sources: None.

Background

 Cancer therapy-related cardiac dysfunction (CTRCD) is one of the major complications being reported in patients with breast cancer (BC) under chemotherapy, in particular with anthracyclines (A) or trastuzumab (T). CTRCD is defined as a reduction of left ventricular ejection fraction (LVEF) >10% to a value <50% or as a relative reduction of global longitudinal strain (GLS) >15%.

Purpose

We aim to evaluate the frequency of CTRCD and its clinical impact on BC patients.

Methods

 Retrospective study of patients with BC treated with A and/or T between Jan 2017 and Dec 2018 who underwent a  transthoracic echocardiography (TTE) before, during, and after chemotherapy. We analysed their baseline characteristics and outcomes based on the type of treatment received, and focusing specially on parameters of ventricular function.

Results

 We included 128 females with median age of 54 ± 11 years-old, treated with A (78; 60.9%), T (14; 10.9%) or A followed by T (36; 28.1%).

At the end of A therapy, there was a significant decrease in LVEF (64,2 ± 5,4% vs 62.6 ± 5,3%, p < 0.05) and in GLS (-20,7 ± 1.9% vs -18.8 ± 2.2%, p < 0.001) compared to baseline. No change in tricuspid annular systolic velocity (S") (p = 0,842) was observed. At 2 years of follow-up, this group of patients had no significant difference in LVEF (64,58 ± 5,58 vs 62,63 ± 6,48, p = 0,053) but maintained a significant lower GLS (-20,84 ± 2,13 vs -18,51± 2,52, p < 0,001) compared to baseline. A significant decrease in S’ (14,36 ± 2,55 vs 13,25 ± 2,26, p < 0,05) was observed.

During T therapy, there was a significant decrease in LVEF (65,04 ± 5,41 vs 59,30 ± 6,21, p < 0,001), in GLS (-21,21 ± 2,75 vs -17,89 ± 2,77, p < 0,001) and in S’ (14,39 ± 3,05 vs 12,19 ± 1,62, p < 0,001) compared to baseline. At the end of T therapy (with or without A), this group of patients maintained a significant decrease in LVEF (65,11 ± 5,65 vs 61,29 ± 6,39, P < 0,001) and in GLS (-21.41%±2,86 vs -19.54%±3,50, p < 0,01). S’ returned to normal values (p = 0,10). At 2 years of follow-up, this group of patients maintained a significant decrease in LVEF (65,00 ± 5,99 vs 61,18 ± 6,30, p < 0,05) but GLS (p = 0,235) returned to normal values. No change in S’ was observed (p =0,379).

During a mean follow-up of 38 months, 35 (27%) patients developed CTRCD with a higher prevalence in patients who took A followed by T (18; 14%), 9 (7%) presented with clinical heart failure, 1 (1%) needed hospitalization, 5 (4%) needed to temporarily suspend treatment and 3 (2%) needed to definitely stop treatment.

Conclusions

 Treatment with A was associated with a decrease in LV systolic function at the end of therapy. A decrease in right ventricular (RV) systolic function was observed at 2 years of follow-up. Treatment with T was associated with a decrease in RV and LV systolic function during therapy. CTRCD occurred in 27%, mainly on patients with both A and T therapy, a frequency within literature data. CTRCD had clinical impact leading to HF and suspension of chemotherapy.

Cancer therapeutics-related cardiac dysfunction: what is the role of cardioprotective medication?

Funding Acknowledgements

Type of funding sources: None.

Introduction

Cancer therapy-related cardiac dysfunction (CTRCD) is one of the major complications being reported in patients with breast cancer (BC) under chemotherapy, in particular with anthracyclines (A) or trastuzumab (T). Guidelines recommend regular left ventricular ejection fraction (LVEF) assessments and CTRCD management with cardioprotective medication (CPM). However, while secondary prevention has already entered clinical practice, primary prevention is still in the research domain.

Purpose

Our aim was to evaluate the role of CPM and the risk of CTRCD in BC patients.

Methods

Retrospective study of BC patients treated with A and/or T between Jan 2017 and Dec 2018 who underwent a transthoracic echocardiography (TTE) before, during, and after chemotherapy. Patients with baseline LVEF <50% were excluded. CTRCD is defined as reduction of EF >10% to a value <50% or as relative reduction of global longitudinal strain (GLS) >15%. As CPM we considered angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blocker (ARB) and beta-blockers (BB).

Results

We included 128 patients, all female with median age of 54 ± 11 years-old, treated with A (78; 60.9%), T (14; 10.9%) or A followed by T (36; 28.1%).

At baseline, mean LVEF was 64,7 ± 5,7% and mean GLS was -20,8 ± 2,3.

During a mean follow-up of 38 months, 35 (27%) patients developed CTRCD with a higher prevalence in patients who took A followed by T (18; 14%), 9 (7%) presented with clinical heart failure, 1 (1%) needed hospitalization, 5 (4%) needed to temporarily suspend treatment and 3 (2%) needed to definitely stop treatment. ACEI/ARB therapy was initiated in 12 (34%) and BB therapy was initiated in 10 (29%). Twenty-one (60%) of CTRCD patients recovered. CPM initiation after CTRCD was not associated with a significant higher rate of cardiac function recovery (p = 0,682).

When comparing patients already medicated with CPM before cancer treatment (39; 30,5%) to those naïve of CPM, the first group presented a lower incidence of CTRCD (7% vs 20%) but it was not statistically significant (p = 0,473).

In patients treated with T or A + T, the group of patients treated with CPM before cancer treatment did not present a significantly lower incidence of CTRCD (p = 0,449) compared to patients CPM naïve. Nonetheless, there was a significant higher LVEF in the TTE during T therapy, after chemotherapy and at 2 years of follow-up (P <0,05) in patients treated with CPM before cancer treatment compared to patients CPM naïve.

In patients treated with A (without T) there was no statistically significant difference between the two groups.

Conclusion

 In our study, pre-treatment with CPM was associated with a significant higher LVEF in patients treated with T but no significant association was found with respect to the occurrence of CTRCD. CPM initiation after CTRCD was not associated with a statistically significant cardiac function recovery.

Impact of an inconclusive exercise stress echocardiogram on cardiovascular outcomes

Funding Acknowledgements

Type of funding sources: None.

Background

Exercise stress echocardiography (ESE) is used for the assessment of suspected or known coronary artery disease (CAD); however, a certain percentage of ESE studies are inconclusive. We aim to evaluate the prognostic impact of an inconclusive ESE on cardiovascular outcomes.

Methods

Single-center retrospective study of consecutive patients (pts) who performed an ESE between 2018 and 2019 for diagnosis or stratification of CAD. All pts performed a symptom-limited Bruce protocol. ESE was considered inconclusive when 85% of age-predicted maximum HR was not reached. Primary endpoint was a composite of admission for acute coronary syndrome (ACS), coronary revascularization and cardiovascular death during the follow-up.

Results

A total of 141 pts were included (76% male; mean age 60 ± 9 years). ESE was inconclusive in 51 (36%) pts, positive in 11 (8%) and negative in 79 (56%).

The mean exercise time of pts with inconclusive ESE was 7 ± 2 minutes and 76% had normal functional capacity. Fatigue (25 pts; 49%) and lower extremities pain (11 pts; 22%) were the main reasons for ESE stopping. Five pts (10%) complained of chest pain during exam.

Pts with an inconclusive ESE were more diabetic (inconclusive 43%, positive 9%, negative 13%; p=.001), had more chronic obstructive pulmonary disease (COPD) (14%, 0%, 2.5%; p=.026), performed more frequently the exam under beta blocker (BB) therapy (59%, 27%, 18%; p=.041) and had less ST-segment depression fulfilling electrocardiographic criteria for ischemia (10%, 64%, 27%, p<.001). Their functional capacity was worse than negative ESE pts (p<.001).

During a median follow up of 22 months (IQR 15-27), 13 (9.2%) pts had the primary endpoint, including 5 pts (3.5%) with ACS.

Pts with an inconclusive ESE had a lower incidence of the primary outcome than pts with a positive ESE and a higher incidence of events than pts with negative results (figure).

In multivariate analysis, after adjusting for functional capacity and electrocardiographic criteria for ischemia, an inconclusive ESE was an independent predictor of the primary endpoint (HR 9.7, IC95% 1.1-87,6 p= .042)

Conclusions

The frequency of inconclusive ESE is not negligible (36%). These pts had more diabetes and COPD and performed the exam under BB therapy more frequently, highlighting the importance of proper patient selection. An inconclusive ESE was associated to a higher risk of cardiovascular events compared to negative ESE. Abstract Figure.

Prevalence and clinical impact of latent obstruction in hypertrophic cardiomyopathy

Funding Acknowledgements

Type of funding sources: None.

BACKGROUND

Hypertrophic cardiomyopathy (HCM) is characterized by varying degrees of left ventricular outflow tract obstruction (LVOTobs). We aim to define the prevalence, clinical profile and impact of LVOTobs under physiological exercise in HCM patients (pts).

METHODS

Single center retrospective study of consecutive HCM pts without LVOTobs at rest (resting gradient <30mmHg), referred for exercise stress echocardiogram (ESE) between 2015 and 2019. Significative latent obstruction was defined as a LVOT gradient ≥50 mmHg during exercise or at early recuperation.

RESULTS

A total of 56 pts were included (64% men, mean age 57 ± 11 years, 61% septal HCM). The majority of pts (47; 84%) were in NYHA I functional class, 20 (36%) had history of syncope or pre-syncope (S/pS) and 7 (13%) had an implantable cardioverter defibrillator (ICD). Twelve (21%) pts had systolic anterior motion (SAM) of the mitral valve at rest.

Thirty-five (63%) pts performed ESE under beta-blocker (BB) therapy. Mean exercise time was 8 ± 3 min. During ESE, 2 (4%) pts developed a LVOT gradient between 30 to 50 mmHg and 17 pts (30%) developed a significative LVOTobs gradient (mean 85 ± 18 mmHg).

Pts with significative latent LVOTobs had more previous complaints of S/pS (59% vs 26%, p=.017), a tendency for a higher NYHA functional class (p=.082) and were more frequently on BB therapy (82% vs 18%, p=.043). Mean septum thickness was similar between groups (17 ± 0.7 vs 16 ± 0.6 mm, p=.536). The presence of SAM at rest was more frequent in the significative latent LVOTobs pts (p<.001). No differences were noted in exercise tolerance (p=.526).

During a median follow-up of 43 (IQR 15-53) months, 7 (41%) pts with significative latent LVOTobs had a pre-syncope, 3 (18%) were diagnosed with atrial fibrillation and 2 (12%) had a cardiovascular admission . There was up-titration/initiation of BB therapy in 5 (29%) pts, referral for septal myectomy in 3 (18%) and ICD implantation for primary prevention in 3 (18%) pts. No proper ICD shocks, sustained ventricular arrythmias or deaths occurred.

Comparing significative latent to non LVOTobs pts, the first ones had more ICD implantation (log rank p =.04) and performed a surgical myectomy more frequently (log rank p= .018) during the follow-up.

CONCLUSION

In our study, significative latent LVOTobs was observed in 30% of pts. Its presence can have clinical implications in HCM pts approach and should be suspected in more symptomatic pts and when SAM is present at rest.

Characterization of temporal trends of systolic and diastolic indexes over time in breast cancer patients treated with trastuzumab

Funding Acknowledgements

Type of funding sources: None.

Background

Trastuzumab therapy increases survival in patients (pts) with HER2 positive breast cancer, however, it is associated with a risk of cardiotoxicity (CT). Our aim was to: 1) assess the temporal evolution of systolic and diastolic left and right ventricular indexes during and after T therapy and 2) study the incidence of CT in this group.

Methods

Retrospective study of breast cancer pts treated with T in a single center, during 2017 and 2018, who underwent a comprehensive echocardiographic examination before, quarterly, and after conclusion of T therapy. Pts with a baseline left ventricular ejection fraction (EF) <50% were excluded. CT was defined as a reduction of EF >10% to a value <50% or as a relative reduction of global longitudinal strain (GLS) of more than 15%.

Results

We included 50 women with mean age of 56 ± 10 years and mean body mass index (BMI) of 27 ± 5 kg/m2. Thirty-six (72%) patients underwent anthracycline chemotherapy prior to T and 44 (88%) had performed radiotherapy.

During treatment with T there was an increase of indexed left ventricle end-diastolic volume (38.6 ± 7.8 to 46.5 ± 10.3 ml/m2, p<.001) and a reduction of LVEF (65.1 ± 5.4 to 59.3 ± 6.2%, p < 0.001), GLS (-21.2 ± 2.7 to -17.9 ± 2.8%, p<.001) and right ventricle S’ (14.4 ± 13.1 to 12.2 ± 1.6 cm/s, p<.001) compared to baseline. There was no change in diastolic function parameters.

CT occurred in 23 (46%) pts, the majority based on GLS criteria (18; 78%). Heart failure symptoms were reported in 4 (8%) of these pts (1 in the GLS group and 3 in the EF group) and 3 (6%) pts had to permanently suspend therapy with T due to CT.

Patients with CT had a lower BMI (25 ± 4 vs 29 ± 4 kg/m2), but were otherwise similar to pts without CT. No association was found between pre-treatment with anthracyclines or radiotherapy with the risk of CT.

An echocardiographic revaluation at 2 years from treatment with T revealed a tendency for LVEF improvement (58.7 ± 7.1 to 61.2 ± 6.3%, p=.059) and a significative recovery of GLS (-17.4 ± 3.5 to -19.1 ± 3.3%, p=.026). No significative variations were found on right ventricular systolic function indexes or on diastolic parameters.

During a follow-up of 43 (IQR 32–47) months, 7 (14%) pts had tumor relapse and 1 (2%) died. CT was not associated with the composite endpoint of tumor relapse or death (p = 0.585).

Conclusion

T therapy is associated with a significative reduction on left and right ventricle systolic indexes during treatment. Although left ventricle systolic function tends to normalize over time, this does not happen with the right ventricle. CT induced by T had clinical impact leading to overt heart failure in 8% of pts and treatment discontinuation in 6% of pts.  A lower BMI was associated with CT, which may be taken into account when defining the treatment strategy.

Family screening in brugada syndrome patients

Background

Brugada syndrome (BS) is a rare inherited channelopathy associated with sudden cardiac death (SCD) and family screening (FS) of index patients (pts) is recommended. However, data about pts identified through FS is lacking.

Aim

To compare index pts to non-index pts identified through systematic FS.

Methods

Single-center retrospective study of BS pts followed by the Arrhythmology Department. FS was offered to 1st degree relatives of all index pts through primary care services and a once-weekly voluntary open appointment. Genetic counselling was performed when indicated. Index and non-index pts were compared regarding baseline characteristics and events during the follow-up (syncope of probable arrhythmic origin, ventricular tachycardia/ventricular fibrillation (VT/VF) and SCD).

Results

We included 165 pts (61% males, mean age 47±15 years) and 72 (44%) were identified through FS. Non-index pts were diagnosed at a younger age (42±14 vs 51±14 years, p <.001), were more often female (57% vs 25%, p<.001), were diagnosed predominantly through provocative test with ajmaline/flecainide (88% vs 47%, p<.001) and had less documented spontaneous type 1 ECG pattern (17% vs 59%, p<.001). A type 2 pattern was identified in 18 (25%) non-index pts.

Genetic testing was performed in 38 (53%) non-index pts: 6 had a pathogenic SCN5A mutation, 18 a likely pathogenic SCN5A mutation and 12 a mutation of uncertain significance.

At diagnosis, 24 (33%) non-index pts had history of syncope, 3 (4%) had nocturnal agonal respiration and 11 (15%) had palpitations with no differences between both groups (p=.119). Non-index pts were less likely to implant a cardioverterdefibrillator (14% vs 38%, p=.001).

During a median follow-up of 28 (IQR 16–41) months, 10 (6%) pts had an event - 2 (3%) in the non-index group (2 syncope) and 8 (9%) in the index group (1 syncope; 7 VT/VF) - with no significative differences between groups (p=.432). There were nocardiovascular deaths.

Conclusions

FS identified a considerable proportion of BS pts. Non-index pts were younger at the time of the diagnosis and had less spontaneous type 1 pattern. No differences were found in events between index and non-index pts, however, the event rate was low. Systematic FS can identify individuals at risk of SCD earlier, allowing close monitoring and, when indicated, appropriate treatment.

Funding Acknowledgement

Type of funding sources: None.

Electrophysiologic study for risk stratification in Brugada Syndrome: does it still matters?

Background

The value of eletrophisiologic study (EPS) with programmed ventricular stimulation for risk stratification in patients with Brugada Syndrome (BS) remains controversial.

Purpose

The aim of this study is to determine the clinical and electrocardiographic predictors of positive EPS and to evaluate whether the induction of malignant ventricular dysrhythmias in the EPS is a predictor of events

Methods

A retrospective study was carried out of patients with spontaneous type 1 Brugada pattern followed up in Arrhythmology consultation at our Hospital. From this population, patients who underwent EPS for stratification of dysrhythmic risk were selected. Clinical and electrocardiographic variables were analyzed. Hospital records and monitoring data from cardiac devices were consulted. Statistical analysis was performed using SPSS 20.0.

Results

Fourty nine patients were included, with a mean age at the beginning of follow-up of 45±14 years, 40 (82%) of whom were male. In 16 individuals (33%) the EPS was positive with induction of malignant ventricular dysrhythmias. All patients with positive EPS implanted an implantable cardioverter-defibrillator and of the 32 patients with negative EPS, 10 implanted an implantable event recorder. The group of patients with positive EPS had a higher proportion of male patients (88% vs 78%). The median follow-up time was 31 months.

The family history of sudden death, family history of BS, or identification of a genetic variant classified as pathogenic or probably pathogenic did not present any relationship with EPS positivity. Of the analyzed electrocardiographic markers, PR interval (178±29 vs 171±27) and QRS duration (119±24 vs 113±15) tended to be longer in patients with positive EPS. Additionally, it was found that 74% of patients with a QRS of less than 120 ms had a negative EPS.

In the analysis of the value of EPS in the stratification of dysrhythmic risk, it was found that of all the patients who suffered events (4), 75% had positive EPS. Of the patients with negative EPS, only 3% (1) presented events in the follow-up. However, there was no significant association between these variables.

Conclusion

In this population, the analysed clinical elements did not correlate with the EPS result. Although there was no statistical significance, there was a tendency for patients with narrower QRS to be more likely to have negative EPS. Accordingly with published data, it was found that the EPS result was not a predictor of events during the follow-up period, which highlights the difficulty of risk stratification in patients with BS.

Funding Acknowledgement

Type of funding sources: None.

Gender related differences in brugada syndrome

Background

Brugada syndrome (BS) is a heritable channelopathy with male predominance. Males (M) seem to have a higher risk of arrhythmic events, although, there is limited data regarding gender differences in BS.

Aim

To compare the differences between genders in a Portuguese sample of BS patients (pts).

Methods

Single-center retrospective study of BS pts followed by the Arrhythmology Department. Pts were divided according to gender and compared regarding baseline characteristics and electrocardiographic (ECG) parameters that possibly predict the arrhythmic risk (significative S wave in DI, R wave sign, Tpeak-Tend interval and fragmented QRS). The events during follow-up were syncope of probable arrhythmic origin, ventricular tachycardia/ventricular fibrillation (VT/VF) and sudden cardiac death (SCD).

Results

A total of 165 pts were included; 64 (39%) were female (F) and the mean age at diagnosis was 47±15 years. The diagnosis was made by family screening (FS) in 72 (44%) pts. Sixty-seven pts (41%) had spontaneous type 1 pattern, 59 (36%) had history of syncope and 5 (3%) had history of aborted SCD. A positive genetic test was identified in 41 (25%) pts. Sixty-three (38%) pts were referred for an electrophysiological study (EEF) which was positive in 17 (27%) pts. A cardioverter-defibrillator was implanted in 45 (27%) pts. Females were more often diagnosed by FS (64% vs 31%, p <.001), had less type 1 spontaneous pattern (22% vs 53%, p<.001) and had no atrial fibrillation (0% vs 7%, p=.043). They performed EEF less frequently (22% vs 49%, p<.001) and had less spontaneous pattern during treadmill stress test (8% vs 33%, p=.004).

Regarding ECG parameters, females had shorter QRS interval (104±12 vs 115±18 ms, p<0.001), less frequent aVR sign (3% vs 27%, p<0.001), less significative S wave in DI (31% vs 55%, p=0.004), and a tendency for a shorter Tp-Te interval (80 vs 100 ms, p=0.051). Corrected QT interval was longer in females (396 vs 392ms, p=0.044). During a median follow-up of 28 (IQR 16–41) months, 7 pts had VT/VF (2F, 5M) and 3 had syncope (3M), with no differences between gender (p=0.287). There were no cardiovascular deaths.

Conclusion

In BS, female pts are more often diagnosed by FS and have less spontaneous type 1 pattern. Gender appears to affect basal ECG characteristics in BS, namely in parameters that may predict arrhythmic risk. Further studies are important to clarify the role of gender in prognosis and risk stratification of BS pts.

Funding Acknowledgement

Type of funding sources: None.

Device therapy guided by implantable loop recorders: predictors of bradyarrhythmic events

Background

Implantable loop recorders (ILR) improved diagnostic yield in unexplained syncope patients (pts). Data on possible predictive factors for bradycardia requiring device implantation in these pts is limited. We aim to identify clinical predictors for device implantation due to a significant bradyarrhythmia in patients who underwent ILR insertion for the study of syncopal/presyncopal (S/pS) events.

Methods

Single-center retrospective study of patients who implanted an ILR for the study of unexplained S/pS episodes between 2013 and 2018. The primary endpoint was a documented bradyarrhythmia requiring device implantation during the follow-up.

Results

A total of 251 pts were included. Mean age was 68±15 years and 53% were female. The majority of pts (220; 88%) had normal ejection fraction. Fifty-two (21%) pts had atrial fibrillation (AF). Regarding basal electrocardiographic characteristics, 34 (14%) pts had 1st degree auriculo-ventricular block (AVB), 31 (12%) pts had left bundle branch block (L-BBB), 22 (9%) pts had R-BBB and 13 (5%) pts had bifascicular block.

During a median follow-up of 20 (IQR 9–34) months, 74 (29%) pts required device implantation because of a significant bradyarrhythmia (92% pacemaker, 4% CRT, 4% ICD). Median duration from ILR until device implantation was 5 (IQR 1–10) months. The indications were sick-sinus-syndrome in 47 (64%) pts, advanced AV block in 23 (31%) pts and symptomatic slow AF in 4 (5%) pts.

Patients who required device implantation were older (73±12 vs 66±15 years, p <0.001), had more hypertension (73% vs 59% p=0.048), a higher prevalence of AF (34% vs 15% p=0.001) and a lower glomerular filtration rate (GFR<60 ml/min: 32% vs 21%, p=0.047). They also had more 1st AVB (22% vs 11%, p=0.026) and intraventricular conduction disturbances (38% vs 25%, p=0.025). There was a trend for a higher device implantation in pts with concomitant 1st AVB and left anterior fascicular block (7% vs 2%, p=0.063)

In a logistic regression model, age >75 years (HR: 1.7; 95% CI: 1.1–2.8) and the presence of AF (HR: 1.8; 95% CI: 1.1–3.0) were independent predictors for device implantation.

Conclusion

An older age and the presence of AF were independent predictors for device therapy in pts who implanted an ILR for the study of S/pS. These factors may identify a higher risk group and should be considered in the initial workup of these pts.

Funding Acknowledgement

Type of funding source: None

P1456 Pacemaker and percutaneous aortic valve endocarditis - a complex decision

CASE REPORT

We describe the case of a 64-year-old man who was referred because of a fever. He had history of hypertension, dyslipidemia, atrial fibrillation, chronic obstructive pulmonary disease and a right lung pneumectomy due to epidermoid lung cancer diagnosed twenty-five years ago. He implanted dual-chamber definite pacemaker ten years ago and had a transcatheter aortic valve implantation (TAVI) in the previous year due to severe aortic stenosis and a high surgical risk.

He was hospitalized 2 months ago because of a respiratory infection. Blood cultures were positive for an Enterococcus faecalis. A transthoracic echocardiogram was performed and had no evidence of vegetations.

After one month, he had recurrence of fever and was again admitted. On physical examination he was hemodynamically stable, with signs of peripheral congestion and no heart murmur on cardiac auscultation. Blood cultures were again positive for Enterococcus faecalis. The transesophageal echocardiogram (TEE) showed a biologic aortic valve with thickened leaflets and small nodular structures suggestive of vegetations. There was a small periprosthetic leak and no obstruction or suspected perivalvular abscesses. There was also a small vegetation with 0,45*0,3 cm in the auricular pacemaker lead. A thoraco-abdomino-pelvic computed tomography scan showed no embolic complications. A diagnosis of pacemaker and TAVI endocarditis was made. The patient started directed antibiotic therapy (ceftriaxone and ampicillin).

The case was discussed in Heart Team and because of high surgical risk, medical treatment was decided. The patient completed antibiotic treatment with negative blood cultures and apyrexia. A repeated TEE showed persistence of vegetations, without development of local complications.

It was decided to discharge the patient under palliative suppressive antibiotic treatment with levofloxacin after discussion with the infectious disease doctor. In a follow-up evaluation, he remained clinically stable, without recurrence of fever.

DISCUSSION

As TAVI procedures are performed more frequently, a higher number of late complications are expected. Prosthetic valve endocarditis after TAVI is a complex situation, whose treatment strategy is not well-defined, particularly because these patients are usually of high surgical risk. This case describes a complex clinical picture and highlights the difficulty in decision-making in these situations. Also, it pretends to reinforce the need to discuss in a Heart Team the best treatment options.

P199 Mitral valve annular disjunction prevalence and significance amongst patients with mitral valve prolapse

Background

Mitral valve prolapse (MVP) is a common and usually benign entity. Occasionally, patients with MVP present with mitral annular disjunction (MAD), whose prevalence and clinical significance are still not clearly established.

Purpose

This study aimed to define the prevalence of MAD in a population of patients with MVP and study its echocardiographic and clinical implication.

Methods

A total of 31 patients with MVP who underwent echocardiographic evaluation in our laboratory were retrospectively evaluated. Echocardiographic, demographic and clinical variables were assessed. Disjunction amplitude (DA) was measured in parasternal long axis view (PLAX) and MAD was considered present if a separation of ≥ 5mm was verified. Annular diameter was measured in PLAX and apical four chamber view (A4C) both in systole and diastole.

Results

MAD was identified in 9 patients (29%), having a mean DA of 12.3 ± 3.2 mm. The group of patients with MAD was significantly younger than the group without MAD (mean age of 54 ± 18 vs 67 ± 15 years; p = 0.001), but there were no significant differences regarding gender, height, weight or cardiovascular risk factors. DA was inversely correlated with body surface area (r=-0.8, p = 0.009). Systolic annular diameters in PLAX and A4C views were increased in patients with MAD (4.2 ± 0.6 vs 3.6 ± 0.7 mm, p= 0.04 and 4.8 ± 0.7 vs 4.1 ± 0.7 mm, p = 0.025, respectively). The difference between systolic and diastolic diameters in PLAX was also greater in MAD (0.5 ± 0.2 vs 0.1 ± 0.3 mm, p = 0.007). Posterior wall thickness (8.5 ± 1.1 vs 9.7 ± 1.4 mm, p = 0.035), indexed left ventricular mass (89.0 ± 15.8 vs 110.6 ± 40,2 g/m2, p = 0.04) and ascending aortic dimensions (28.7 ± 6.7 vs 37.4 ± 3.6 mm, p = 0.018) were notably inferior in MAD patients, as was left ventricle ejection fraction (LVEF) (57.5 ± 5.8 vs 62.6 ± 4.9 %, p = 0.0023). No differences were found in chamber volumes. Similarly, eletrocardiographic parameters were identical in both groups. Five patients (56%) in the group with MAD had documentation of events (palpitations, dizziness, syncope or sudden cardiac death), in contrast with just 3 (21%) in the group without MAD, although statistical significance was not achieved (p = 0.078).

Conclusion

This study revealed that MAD is common among patients with MVP and is associated with altered annulus dynamics during the cardiac cycle. Its association with younger individuals, lower LVEF and, apparently, more clinical events highlight the importance of this entity.

P4408Cardiovascular response in diabetic patients submitted to treadmill electrocardiographic exercise stress test

Introduction

Diabetes mellitus (DM) is an entity commonly associated with neuropathy, a factor that may have repercussions on the cardiovascular system, specifically in its capacity to respond to stimuli.

Objectives

To evaluate the cardiovascular response to exercise in diabetic patients, comparing it with a control group.

Methods

Patients who underwent treadmill eletrocardiographic exercise stress test (EST) between January 2016 and November 2018 without the influence of negative chronotropic medication were included.

Results

A total of 187 patients were identified, having a mean age of 57±13 years, of whom 120 (64%) were of the male gender. Eighty-four (45%) were diabetic. The group of diabetic patients had a higher prevalence of arterial hypertension, however the number of classes of antihypertensive drugs was not significantly different between diabetic and non-diabetic patients. There were no significant differences in the remaining demographic variables. Patients with DM presented lower maximal heart rates (HR) (141±14 vs 148±19 beats/minute, p=0.015), lower HT reserve (59±16 vs 67±21 beats/minute, p=0.005), as well as a lower rate of HR fall in the recovery period (13±5 vs 16±5 beats/minute2, p<0.001). Total EST time was also lower in diabetic patients (median 7.0 IQR=3 vs 7.3 IQR=3.5 minutes, p=0.044). Additionally, the number of years since the time of diagnosis of DM was inversely correlated with the degree of increase in systolic blood pressure (SBP) with exercise (r=−0.22, p=0.045), and the pre-test SBP (median 135 IQR=24 vs 130 IQR=20 mmHg, p=0.048) and post-test SBP (median 150 IQR=20 vs 140 IQR=25 mmHg, p=0.007) were higher in patients with DM.

Discussion

In this study it was found that patients with DM present an impaired chronotropic response, both in exercise and recovery periods, and it was observed that the capacity to increase SBP is inversely correlated to the duration of DM. These data may reflect the neuropathic involvement in DM and its influence in the cardiovascular response to exercise.

[Diet and risk of myocardial infarction. A case-control community-based study]

OBJECTIVE

To evaluate diet as a risk factor for myocardial infarction.

DESIGN

Community based case-control study.

SETTING

University Hospital, Oporto.

PARTICIPANTS

First time consecutive cases of acute myocardial infarction (n = 100) and 198 community controls, older than 39 years and living in Oporto, were compared.

METHODS

Data were collected by trained interviewers using a structured questionnaire designed to obtain information on socio-demographic, medical and behavioural aspects, emphasising the description of diet and food habits (using a semi-quantitative food frequency questionnaire). Controls were selected by random digit dialing with a participation rate of 70%. Odds ratios and 95% confidence intervals (CI) according to quartiles of nutrient ingestion were calculated using unconditional logistic regression.

RESULTS

Female controls presented significantly higher mean intakes of protein, omega-3 fatty acids, fiber, cholesterol and vitamin C. Male controls had a significantly higher mean daily intake of fiber, vitamin C, vitamin E, and carotenes. After adjusting for age, sex, education, body mass index, ethanol, smoking and total energy intake, there was a protective effect of vitamin C (OR = 0.2, 95% CI: 0.1-0.6, for the 4th quartile), vitamin E (OR = 0.3, 95% CI: 0.1-0.9 for the 4th quartile) and total fiber (OR = 0.3, 95% CI: 0.1-0.9) for the 4th quartile). No significant effect was found for trans-fatty acids, but there was a higher risk with increased energy intake.

CONCLUSIONS

This study showed that diet has an important independent effect on myocardial infarction, a protective independent role for anti-oxidant vitamin C and E was verified.