Prognostic value of aerobic capacity and exercise oxygen pulse in postaortic dissection patients

Although recommendations encourage daily moderate activities in post aortic dissection, very little data exists regarding cardiopulmonary exercise testing (CPET) to personalize those patient's physical rehabilitation and assess their cardiovascular prognosis.

that is, 97 (83-113) % of its predicted value, suggesting cardiac exercise limitation in a population under beta blockers (92% of the population). During a follow-up of 39(20-51) months from CPET, 42 aortic event recurrences and 22 MCE not related to aorta occurred. Low peak oxygen pulse (<85% of predicted value) was independently predictive of aortic event recurrence, while low peak oxygen uptake (<70% of predicted value) was an independent predictor of MCE occurrence.
Conclusion: CPET is safe in postaortic dissection patients should be used to not only to personalize exercise rehabilitation, but also to identify those patients with the highest risk for new aortic events and MCE not directly related to aorta. Post discharge management is based on optimal blood pressure control and aortic imaging follow-up. 1 Regular physical activity is a key point of blood pressure management and cardiovascular fitness in both primary and secondary care of cardiovascular diseases. [2][3] Although recommendations encourage daily moderate activities in post aortic dissection, the safety and benefit of physical activity in this specific context has been poorly explored. [4][5] While moderate physical activity could be beneficial by lowering blood pressure and thus preventing aortic disease progression, intensive exercise is feared to aggravate aortic expansion. Importantly since promoting physical activity in post aortic dissection patients might be difficult because of their very low aerobic capacities, 6 cardiopulmonary exercise testing (CPET) represents an affordable tool to evaluate individual fitness and thus personalize patient advices to their daily life activities. 7 Moreover, CPET might provide prognostic insights in post aortic dissection as it does in several severe medical conditions (e.g. end-stage heart failure, chronic respiratory failure, end-stage liver disease). [8][9] In line we aimed at testing the prognostic significance of CPET regarding aortic and cardiovascular events by exploring a cohort of patients followed-up after acute aortic dissection.

| Clinical data and blood pressure recording
Clinical characteristics and medication were collected at inclusion. 24-h blood pressure monitoring was performed and allowed to ensure proper control if blood pressure: blood pressure measure was performed every 15 min with a Spacelabs Medical 90 207 ambulatory blood pressure monitor (Spacelabs Medical Ltd. Issaquah, Wa).
In case of aortic stenting, the cuff was placed on the right arm. Medical treatment was recorded at the time of blood pressure monitoring.  10 Left ventricular filling pressure was considered elevated if E/A mitral inflow ratio was>2 or E/é ratio > 14. 11 An angio-CT scan confirmed an aortic diameter below the surgical indication. CT angiograms were realized with an acquisition from the apex of the lungs to the femoral heads. The maximal diameter of the descending aorta behind the left pulmonary artery was used according the Kato's technic. 12 2.4 | Cardiopulmonary exercise testing CPET was performed under antihypertensive drugs including betablocker. CPET was performed in standard manner on a cycle ergometer and firstly aimed at personalizing patient rehabilitation. Patients were encouraged to exercise to exhaustion (symptom-limited maximal test). Blood pressure was measured every 2 min. A 12-lead electrocardiogram was monitored during exercise test. Incremental exercise tests were performed on a calibrated electromagnetically braked cycle ergometer (ER-900; Jaeger, Hochberg, Germany), using a 1-min step protocol at 10 W min − 1 until exhaustion. Maximal incremental exercise was performed with the subjects maintaining a pedaling frequency of 60 ± 5 revolutions per min. Peak exercise was defined as the highest work level reached during incremental exercise test. During testing, continuous noninvasive monitoring of heart rate and blood pressure was done (Marquette electrocardiogram system, Marquette Electronics, WI, USA). CPET was performed with a metabolic cart (Oxycon Pro, Viasys, France). Continuous measurement of inspired and expired oxygen and carbon dioxide output were done with inline sensors. Peak O2 pulse was calculated by dividing peak oxygen uptake (VO2) by heart rate at the time of peak VO2. The quality of effort was assessed by the respiratory exchange ratio, that is, maximal exercise was defined when peak respiratory exchange ratio > 1.05.

| Morphological data
Hemoglobin level was checked at the time of exercise testing realization.

| Clinical follow-up
Events were recorded after CPET. Clinical follow-up was obtained by regular phone calls of patients, as well as interview with their cardiologist and general practitioner, and review of medical files. A new aortic event was defined as a new dissection, aortic rupture, or new scheduled surgical or endovascular procedure, and was the primary endpoint of the study. A scheduled aortic intervention was performed in case of rapid aortic expansion (>5 mm per year) or if aortic diameter reached 55 mm during follow-up. A secondary endpoint was defined as a composite criterion of major cardiovascular events (MCE) not directly related to aortic event, that is, cardiovascular death, stroke, acute coronary syndrome, hospitalization for cardiac failure and coronary or peripheral artery revascularization.

| RESULTS
Among the 213 patients referred to our hospital for post aortic dissection between September 2012 and October 2017, 165 patients underwent CPET. Median delay between the acute aortic event and CPET was 6 4-20 months. No adverse event was observed during exercise. Flow-chart of the study is presented in Figure 1 and patient characteristics at inclusion in Table 1 Per-exercise hemoglobin level was normal in the entire population (12.9 ± 1.8 g/dl). Resting heart rate and blood pressure were within the expected ranges in a population under beta-blockers and optimized anti-hypertensive drugs. Details of exercise testing data are provided in Table 2. Peak respiratory exchange ratio was 1.2 ± 0.13 reflecting maximal exercise. Aerobic capacities were globally altered with a mean peak oxygen uptake of 18.8 ± 5.1 ml/Kg/min, corresponding to 75(63-87) % of predicted values and 5.5(4. 5-6.7) METS. Heart rate at peak exercise was missing for three patients, so peak exercise oxygen pulse data were available for 162 patients. Peak  Table 1. Optimal threshold of event predictors was determined for numerical parameters reaching a p value <0.20. A threshold value to best predict event free-probabilities was defined using the algorithm of maximization of Hazard ratio. The optimal cut-off to predict aortic event recurrence were respectively 85% for the predictive oxygen pulse value, 50 mm for the left ventricular end-diastolic diameter, 70% for the predictive peak oxygen uptake and 40 mm for the descending aorta diameter.
Peak oxygen pulse <85% of predicted value, left ventricular enddiastolic diameter > 50 mm and thoracic descending aorta diameter > 40 mm were identified by multivariable analysis as independent predictors of aortic event recurrence during follow-up ( Table 3).
The association between the risk of aortic events during follow-up and of peak oxygen pulse is illustrated in Figure 2.
MCE not related to aortic events occurred in 22 patients during follow-up: three cardiovascular deaths, eight strokes, one acute coronary syndrome, three hospitalizations for cardiac failure and seven scheduled arterial revascularizations. Predicted peak oxygen uptake and 24-h systolic blood pressure were associated with MCE onset, with thresholds of 135 mmHg for the 24-h systolic blood pressure and 70% of predicted peak oxygen uptake identified as the best cutoff values to predict events using the algorithm of maximization of the Hazard ratio. Interestingly, peak oxygen uptake <70% of predicted value and 24-h systolic blood pressure > 135 mmHg were the sole independent prognostic factor associated with MCE occurrence (Table 3). The association between the risk of MCE during follow-up and peak oxygen uptake is illustrated in Figure 3.

| DISCUSSION
Exploring clinical significance of CPET in postaortic dissection syndrome, we confirmed that CPET is safe in patients with well-controlled blood pressure and can thus be used to tailor and personalize exercise rehabilitation. We also showed that peak exercise oxygen pulse and oxygen uptake are independent predictors respectively of the risk of new aortic events and MCE not directly related to aortic events.

Management of patients surviving aortic dissection is mainly
focused on blood pressure control thanks to intensive antihypertensive drugs. Guidelines encourage daily moderate activities for these patients to improve global cardiovascular health and blood pressure control 1,5 : exercise at an intensity of 3 to 5 METs for at least 30 min on most days of the week, for a total of 150 min/week or more. In our study, most of patients had very limited aerobic capacity with maximum workload very close or even lower than 5 METS. The recommended daily practice of moderate exercise at 3-5 METS should thus be adapted and personalized to each patient thanks to CPET. 6 Aggravation of aortic expansion is feared after intense exercise. We performed CPET in >150 patients under beta-blockers F I G U R E 1 Flow chart of the study population without any complication, confirming the safety of CPET in those patients with proper control of blood pressure. 13,14 Mechanisms influencing aortic dilatation and cardiac events in postaortic dissection patients are poorly explored. Aortic evolution is understood as being dependent of aorta properties and systemic factors such as blood pressure control. 15 In line, we showed that the descending aorta size was associated with a higher risk of aortic event during follow-up. We also highlighted that oxygen pulse is related to aortic events. Oxygen pulse is the product of stroke volume and arteriovenous oxygen difference during exercise, and thus an indicator of cardiac performance in the absence of anemia. 16 Thus, oxygen pulse increases linearly with increasing workload until it reaches its highest value, that is, a kinetic similar to stroke volume raise. 17 We interpret abnormal peak oxygen pulse in our study as an index of impaired cardiovascular performance secondary to elevated aortic-related afterload. 18 Indeed, abnormal pulsatile aortic loading during exercise can occur in postdissection aorta independent of hypertension because of thrombosed false lumens, abnormal aortic compliance and pulse wave reflection. Ventriculo-arterial uncoupling despite wellcontrolled blood pressure would develop with stress and reflect arterial tree burden. This hypothesis should be tested further thanks to noninvasive evaluation of ventriculo-arterial coupling during exercise. [19][20] Oxygen pulse is also linked to arteriovenous difference during exercise. This parameter was not directly evaluated in our observational study since its assessment requires invasive measurements.
Although peripheral oxygen extraction at peak exercise is influenced by the level of physical activity, this phenomenon is of limited impact  Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; GFR, glomerular filtration rate, LDL, low density lipoprotein, RAAS, renin-angiotensinaldosterone system, SBP, systolic blood pressure. Values are expressed as mean (standard deviation) unless otherwise as indicated. Note: Heart rate at peak was missing for three patients, peak exercise oxygen pulse data were thus available for 162 patients. a p < 0.05 vs. patients with O2 pulse≥85% predicted value.
in nonathlete individuals. 21 Finally we can also hypothesized that a lower oxygen pulse reflected higher sympathetic activity. A higher sympathetic activity increases heart rate at rest and during exercise. Different situations associated with high sympathetic activity are known to influence aortic root size, aortic aneurysm expansion and also influence aortic expansion following in post aortic dissection. Obesity is associated with aortic size at different level and influence aortic enlargement. Sleep apnea syndrome, which was not specifically explored in our study, has been shown to be very prevalent in the population of patients suffering from aortic dissection and is associated with an oversympathetic activity.
Our findings have potential clinical implications. If clinical benefits of cardiac rehabilitation post aortic dissection has not been evaluated well so far, cardiac rehabilitation activities seems to be beneficial in this context. In a small sample size study, peak oxygen consumption was significantly increased after a cardiac rehabilitation program. 22 Enhancement of peak oxygen uptake has already shown its benefit on cardiovascular secondary prevention in atherosclerotic populations. These results reflect the benefit of moderate physical activity. Physical activities recommendations must be advised with caution, intensive physical activity have probably deleterious effect. 23 By increasing peak oxygen uptake and decreasing maximal heart rate at exercise, regular physical activities are expected to improve peak oxygen pulse.
We can thus hypothesize that regular activity could reduce residual aortic and cardiovascular risk. Moreover, regular physical activity is an integrant part of lowering blood pressure and weight loss.
Aerobic capacity was evaluated for the first time in our study in postaortic dissection to stratify the risk of cardiovascular events. Peak oxygen uptake predicted nonaortic cardiac events. CPET is thus shown to be of prognostic value, along the same line as in several other chronic medical conditions. 24 Aerobic capacity evaluation might be implemented in post aortic dissection management to only to guide cardiac rehabilitation but also to stratify cardiovascular risk. 25,26 This work must be read with its limits, as the single center observational design. The exercise testing was performed once during follow-up, and data on patient's physical activities before and after acute aortic dissection were not collected. The population is quite heterogeneous with different specific management in the aortic dissection acute phase. Exploration of sleep apnea syndrome and sympathetic hyperactivity would have brought interesting data regarding the interplay between neurovegetative system, blood pressure and outcomes. These explorations were not performed and would need further research." In conclusion, the risk of recurrent aortic event remains high after aortic dissection. On top of personalization of exercise rehabilitation, CPET helps stratifying the risks of new aortic events and MCE not directly related to aortic events in postaortic dissection patients. The influence of aerobic capacity enhancement on residual risk following an aortic dissection needs to be further investigated. Cardiac rehabilitation program appears to be essential following aortic dissection.

CONFLICT OF INTEREST
The authors declare no conflicts of interest regarding this work.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.