Lifetime exercise dose and ventricular arrhythmias in patients with mitral valve prolapse

Abstract

Aims

Patients with mitral valve prolapse (MVP) have high risk of life-threatening ventricular arrhythmias (VAs). Data on the impact of exercise on arrhythmic risk in these patients are lacking. We explored whether lifetime exercise dose was associated with severe VA and with established risk factors in patients with MVP. Furthermore, we explored the circumstances at the VA event.

Methods and results

In this retrospective cohort study, we included patients with MVP and assessed lifetime exercise dose as metabolic equivalents of task (MET) hours/week. Severe VA was defined as sustained ventricular tachycardia or fibrillation, aborted cardiac arrest, and appropriate shock by a primary preventive implantable cardioverter defibrillator. We included 136 MVP patients (48 years [interquartile range (IQR) 35–59], 61% female), and 17 (13%) had previous severe VA. The lifetime exercise dose did not differ in patients with and without severe VA (17 MET h/week [IQR 9–27] vs. 14 MET h/week [IQR 6–31], P = 0.34). Lifetime exercise dose > 9.6 MET h/week was a borderline significant marker for severe VA (OR 3.38, 95% CI 0.92–12.40, P = 0.07), while not when adjusted for age (OR 2.63, 95% CI 0.66–10.56, P = 0.17). Ventricular arrhythmia events occurred most frequently during wakeful rest (53%), followed by exercise (29%) and sleep (12%).

Conclusion

We found no clear association between moderate lifetime exercise dose and severe VA in patients with MVP. We cannot exclude an upper threshold for safe levels of exercise. Further studies are needed to explore exercise and risk of severe VA.

Graphical Abstract

Open in new tabDownload slide

Introduction

Mitral valve prolapse (MVP) is a condition characterized by systolic displacement of one or both mitral leaflets above the mitral annulus.¹ It affects ∼2–3% of the general population,² and most patients have excellent prognosis. However, a subgroup of MVP patients is at risk of life-threatening arrhythmias and sudden cardiac death (SCD).¹

Exercise guidelines state that patients with MVP and with arrhythmic risk factors should refrain from high intensity exercise.³ Other studies also propose exercise restriction in MVP patients guided by the presence of varying risk factors.⁴ The evidence behind these recommendations is, however, not well documented. A recent study showed that athletes with MVP have similar incidence of clinical events as the general MVP population,⁵ indicating unnecessary exercise restrictions in MVP. The influence of exercise on short- and long-term arrhythmic risk is unknown.

We aimed to explore the association between lifetime exercise dose and severe ventricular arrhythmia (VA), and the association to established arrhythmia risk factors in MVP patients. We hypothesized that exercise increases arrhythmic risk in MVP patients. Furthermore, we explored the acute effect of exercise on arrhythmic risk by assessing the circumstances and level of physical activity at time of severe VA.

Methods

Study population and recruitment

Patients with MVP and mitral annular disjunction (MAD) were recruited from two centres (Oslo University Hospital and Drammen Hospital) from 2014 to 2021 as previously reported.⁶ Inclusion criteria were presence of MVP and/or MAD with severe VAs, or premature ventricular complexes (PVCs) with left sided origin detected by Holter or exercise electrocardiogram (ECG), or arrhythmic symptoms, or PVCs with other configurations and no other obvious cause of arrhythmias. At inclusion, patients were evaluated with clinical examination, 12-lead ECG, 24-h ECG, exercise stress ECG, transthoracic echocardiography, and cardiac magnetic resonance (CMR) imaging.⁶ We excluded other known aetiologies of arrhythmias. In patients with severe VA, we additionally performed genetic testing for channelopathies and cardiomyopathies.

We included patients from the previous cohort who responded to a questionnaire on lifetime physical activity (from August 2021 to April 2022) (Figure 1).

The study complied with the Declaration of Helsinki and was approved by the Regional Committee for Medical Research Ethics (2015/596/REK Nord). All study participants provided written informed consent.

Exercise questionnaire

Exercise history was assessed by the standardized and validated lifetime physical activity questionnaire,⁷ and patients responded either by written response or by a structured telephone interview. We calculated the total lifetime exercise dose, expressed as metabolic equivalents of task (MET), from age 6 until a severe VA or April 2022, whichever came first. To standardize for age, we divided the total exercise dose by number of years of performed exercise excluding the first 6 years of life (total lifetime exercise dose in METs/[52 weeks × age − 6]) giving us the average total lifetime exercise dose in MET hours/week. Additionally, we calculated the lifetime high intensity exercise dose including only activities > 6 METs as defined by consensus as the cut-off for high intensity exercise⁸ and standardized for age as described above. We calculated the total time spent in high intensity exercise (>6 METs). We performed analyses on patients who exercised >6 METs for ≥4 h/week for at least 5 consecutive years.

Severe ventricular arrhythmia

We defined severe VA as aborted cardiac arrest, ventricular fibrillation, sustained ventricular tachycardia (>100 bpm lasting >30 s or terminated earlier due to haemodynamic instability), or appropriate shock from a primary preventive implantable cardioverter defibrillator.

Circumstances at the time of the severe ventricular arrhythmia

Patients with a severe VA were interviewed to assess their activity state at the time of severe VA. The activities were categorized as exercise (defined as physical activity > 6 MET), wakeful rest, sleep or missing.

Echocardiography and cardiac magnetic resonance imaging

We assessed cardiac volumes and function.^9,10 Imaging data were analysed offline [echocardiographic data by EchoPAC v203 (GE Healthcare, Horten, Norway) and CMR data by Sectra Workstation IDS7 v18.1 (Sectra AB, Linköping, Sweden)]. We defined MVP as superior displacement ≥ 2 mm of any part of the mitral leaflet beyond the mitral annulus on echocardiography in the parasternal long-axis view.¹ We defined MAD as the separation of left atrial wall at the junction of the posterior mitral leaflet and the myocardium in the lateral view on echocardiography.⁶ We graded mitral regurgitation (MR).¹¹ Left ventricular (LV) ejection fraction was assessed by Simpson biplane method. Late gadolinium enhancement (LGE) was reported if present on CMR.⁶

Risk factors advising against high intensity exercise

We analysed each patient for factors of increased risk of SCD according to current guidelines on sports cardiology.³ These risk factors included ECG T-wave inversion in inferior leads, long QT-interval, bileaflet MVP, family history of SCD, documented VA, basal-inferolateral wall fibrosis on LGE CMR, severe MR, and severe LV dysfunction/dilatation. Patients with no risk factors were compared to patients with ≥1 risk factors.

Statistical analysis

Values were expressed as mean ± standard deviation (SD), frequencies (%), or median with interquartile range (IQR). Groups were compared with independent Student’s t-test, Mann–Whitney U test, χ², Fisher exact tests, one-way ANOVA, and Kruskal–Wallis test as appropriate. We tested lifetime exercise dose as a risk marker of severe VA using logistic regression models. The parameters’ ability to detect risk of severe VA was assessed by receiver operating characteristic curves, and we calculated the area under the curve (AUC). The Akaike information criterion (AIC) was evaluated for the logistic regression models to assess the precision/complexity relationships. We used single threshold regression analysis to investigate for threshold of exercise doses associated with increased risk of severe VA. Cumulative incidence plot showed age-related incidence of a severe VA above and below this threshold, compared by log-rank test. We performed univariate and multivariate logistic regression analyses with the parameters exercise dose below and above the threshold, and age. We used log base 10 transformation of the exercise dose to meet model linearity assumptions. Two-sided P-values < 0.05 were considered significant (Stata/SE v17.0, StataCorp LLC, TX, USA).

Results

Study population

Of the 160 patients who received the exercise questionnaire, 136 (85%) responded and were included in the study (age 48 [IQR 35–59], 61% female) (Table 1). Of these, 104 (76%) had previously been reported.⁶ Of the 136 included patients, 87 (64%) fulfilled the EHRA criteria for AMVP.¹ Of the remaining 49, 17 (13%) had PVCs with right bundle branch block configuration. The remaining 32 (24%) had symptoms of arrhythmia with various configuration of PVCs and with no other obvious aetiology.

Median total lifetime exercise dose was 14 MET h/week (IQR 6–30). A severe VA had occurred in 17 (13%) patients at a median age of 33 years (IQR 24–40) (Table 1, Figure 1). Patients with a severe VA were younger at time of inclusion (P < 0.01) and had higher PVC burden (Table 1).

Relationship between lifetime exercise dose and a severe ventricular arrhythmia

Neither the total lifetime exercise dose, nor the lifetime high intensity exercise dose, differed between patients with and without a severe VA (Table 1). Single threshold regression analysis of the total lifetime exercise dose identified 9.6 MET h/week as optimal threshold for increased arrhythmic risk (OR 1.18 per 1 MET h/week increase above threshold, 95% CI 1.10–1.27, P < 0.01), while there was no association between total lifetime exercise dose and a severe VA below this threshold (P = 0.21). Patients above 9.6 MET h/week had higher cumulative incidence of a severe VA compared to patients below 9.6 MET h/week (log-rank P = 0.046) (Figure 2A). By univariate logistic regression, patients with exercise dose above 9.6 MET h/week had a borderline significant association to severe VA (OR 3.38, 95% CI 0.92–12.40, P = 0.07). However, when adjusting for age, this effect was no longer evident (OR 2.63, 95% CI 0.66–10.56, P = 0.17) (Figure 2B). Patients with exercise dose above 9.6 MET h/week were significantly younger at time of inclusion (P = 0.01) (Table 2). However, age at arrhythmic event did not differ between those with exercise dose above or below 9.6 MET h/week (P = 0.53) (Table 2). Total lifetime exercise dose was not associated to age, neither in those with nor in those without severe VA (P = 0.58 and P = 0.07, respectively) (Figure 3). In patients that had performed the most exercise (n = 36, exercising >6 MET for ≥4 h/week for >5 years), the incidence of VAs was 7/36 (26%) compared to 10/100 (10%) in the remaining (P = 0.14).

Exercise dose and risk factors restricting high intensity exercise

In all, 106 (78%) patients were classified as having ≥1 risk factors and should have been advised against participating in high intensity exercise according to guidelines.³ As expected, VA was more frequent in those with ≥1 risk factors (17 patients [16%] vs. 0 patients [0%], P = 0.02). Number of risk factors correlated with severe VA (OR 2.11 (95% CI 1.33–3.37) P = 0.01). However, the regression model did not improve when adding total lifetime exercise dose (OR 1.52 (95% CI 0.69–3.37) P = 0.30) and the AUC did not differ between the two models (P = 0.11) (Figure 4).

Total lifetime exercise dose did not differ between those with >1 risk factor compared those without risk factors (14 MET h/week [IQR, 6–31] vs. 14 MET h/week [IQR, 7–29], P = 0.96). Furthermore, there was no difference in lifetime high intensity exercise dose between those with ≥1 risk factor compared those without risk factors (9 MET h/week [IQR, 3–26] vs. 11 MET h/week [IQR, 1–20], P = 0.50).

Circumstances around the severe ventricular arrhythmia

Among the 17 patients with a severe VA, 9 (53%) had their event at wakeful rest, 5 (29%) during exercise, and 2 (12%) during sleep. One patient had amnesia after the cardiac arrest and was classified as missing.

In the five patients with a severe VA during exercise, total lifetime exercise dose was 23 (IQR 12–45) MET h/week (P = 0.31 compared to total cohort).

Patients with severe VA during exercise had spent median 1560 (IQR 800–4240) h in high intensity exercise (>6 METs), patients with VA during rest had spent median 2016 (IQR 572–4832) h, and patients without severe VA had spent median 2235 (432–5410) h in high intensity exercise. Adjusted for age, there was no differences between groups (P = 0.61).

Discussion

Our results showed no clear association between exercise habits and severe VA in patients with MVP. There may exist an association between severe VA and a total lifetime exercise dose above 9.6 MET h/week and between severe VA and the highest levels of exercise, but these associations were biased by age and by limited power of analyses. Our findings suggested that MVP patients may be encouraged to be physically active at a moderate intensity to derive the common benefits from exercise, while precaution should be taken towards high intensity exercise.

Relationship between lifetime exercise dose and a severe ventricular arrhythmia and risk factors for a ventricular arrhythmia

In our study, the median total lifetime exercise dose was 14 MET h/week. This dose is comparable with previous reports with exercise doses ranging from 9 to 18 MET h/week.^12,13 Patients with severe VA did not have higher total lifetime exercise doses, nor higher lifetime high intensity exercise dose than patients without severe VA.

A threshold of 9.6 MET h/week was identified as optimal threshold for increased arrhythmic risk by single threshold regression analyses. This method has inherent limitations for predicting a clinically valid exercise threshold and should be interpreted with care. Importantly, we found no association between total lifetime exercise dose below 9.6 MET h/week and long-term arrhythmic risk, reassuring exercise performance at this level. Recommended level of exercise to gain health benefits have been reported to be around 7.5 MET h/week.¹⁴

We cannot exclude that a lifetime exercise dose above 9.6 MET h/week may impose increased long-term arrhythmic risk. There was a weak association between severe VA and exercise above 9.6 MET h/week (P = 0.07). However, this association was outperformed by age, possibly reflecting the age clustering of arrhythmic events at younger age in MVP that has been previously reported.^6,15 Importantly, patients with higher levels of exercise dose did not have events at younger age compared to those with lower exercise dose. Patients with high intensity exercise over time (>6 METs for >4 h/week for 5 years) did not have significantly higher incidence of severe VA, however, there was a numerical trend (26% vs.10%, P = 0.14).

A previous paper by Caselli et al.⁵ showed low risk for severe VA in athletes with MVP, and they reported no episodes of SCD. Our cohort was not comparable to the study by Caselli et al. regarding arrhythmic risk, by our inclusion of MVP patients with previous arrhythmias and with 13% of patients having a history of severe VA. Our patient cohort was therefore not representative for the general MVP population.

In all, our findings question if exercise dose and exercise intensity play a role in arrhythmogenesis in MVP. Larger studies are warranted to clarify the effect of exercise on arrhythmic outcome in MVP.

High-risk patients and exercise

We found no clear association between lifetime exercise dose and frequency of risk factors for severe VA (Table 2). None of the established risk factors for severe VA (syncope, non-sustained ventricular tachycardia,¹⁶ reduced left ventricular ejection fraction (LVEF), ECG T-wave inversions, MAD, and presence of myocardial fibrosis)^17–19 were more frequent in patients with higher lifetime exercise doses, indicating that exercise did not seem to aggravate these parameters.

The majority of our patients had one or more factors of increased risk for VA, and there was a high prevalence of MAD and bileaflet MVP. The high incidence of MAD can be explained by using patients from our previous study using MAD as an inclusion criterion.⁶ Only LVEF differed between patients with and without VA. However, patients with more risk factors had more frequently severe VA (Figure 4). This finding supports the validity of the established risk factors and supports the incremental value of aggregation of multiple risk factors. Adding lifetime exercise dose to the risk model did not improve the prediction for severe VA (Figure 4), indicating that lifetime exercise dose, as calculated in this study, was not the major player in arrhythmic risk.

Circumstances around the severe ventricular arrhythmia

Approximately 30% of patients experienced their severe VA during exercise, while ∼70% of events occurred at wakeful rest or sleep. These results support a previous study by Basso et al.,²⁰ including 43 patients with MVP, where SCD occurred during physical or emotional stress in 19%, during rest in 70%, and during sleep in 12%. Together, these two studies indicate that VA events most often occurred during wakeful rest, and 20–30% during exercise. Considering that time spent exercising normally is considerably lower than 30% per day, these results may indicate a proportionally higher risk of VA during exercise than in the rested state. Exercise as an acute trigger for severe VA should be further studied.

Clinical implications

Exercise is important for an individual’s quality of life and for maintaining cardiovascular health.^21–23 In other cardiac diseases, reports have suggested that too restrictive exercise advise may lead to harmful long-term effects.^24,25 Our results show that higher lifetime exercise doses not clearly increase risk of severe VA beyond established risk factors and thereby do not challenge guidelines for exercise in MVP patients (sports guidelines) recommending moderate doses of exercise. We suggest maintaining the recommended caution regarding high intensity exercise for an extended period, similar to recommendations in other cardiovascular diseases.

Limitations

This study has a retrospective observational design with inherent limitations, including recruitment, referral, recall, and survival bias. The sample size was limited, and underpowered. Physical activity was self-reported and therefore susceptible for recall bias. The older population reported less participation in organized sporting activities of today, which may have underestimated the physical activity in this population. Our study did not include high performing professional athletes, limiting the interpretation and advice for this category.

A subset of patients did not fulfil EHRA AMVP diagnosis¹ due to inclusion before that document was published. However, patients not fulfilling EHRA AMVP diagnosis were carefully considered, and other causes for arrhythmias were excluded.

Conclusions

This study showed no clear association between lifetime exercise dose and severe VA in patients with MVP. Exercise dose was not clearly associated with occurrence of severe VA, nor with the presence of established arrhythmic risk factors in MVP. However, there might exist an upper threshold for safe levels of exercise. In all, our findings indicate that MVP patients should follow the general ESC sports guidelines and, as a precaution, limit high intensity exercise for an extended period. Further studies are needed to explore exercise as an acute trigger of VA.

Acknowledgements

We thank the patients for participating in this study.

Funding

This work was supported by the Research council of Norway, Precision Health Center for optimized cardiac care (ProCardio) (grant number 309762) to C.K.F., K.H., N.E.H., K.V., and L.T.A., Research council of Norway, Genepositiv (grant number 288438) to A.I.C. and E.W.A., and European Research Area Network on Cardiovascular Diseases (ERA-CVD, EMPATHY project) (grant number 298736) to C.R.-N.

Data availability

The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.

References

Author notes