Author + information
- Received June 11, 2018
- Accepted June 22, 2018
- Published online September 10, 2018.
- Mohit K. Turagam, MDa,
- Venkat Vuddanda, MDb,
- Niels Verberkmoes, MDc,
- Toshiya Ohtsuka, MDd,
- Ferdi Akca, MDc,
- Donita Atkins, RNe,
- Sudharani Bommana, MSce,
- Maximilian Y. Emmert, MDf,
- Rakesh Gopinathannair, MD, MAg,
- Gansevoort Dunnington, MDh,
- Abdi Rasekh, MDi,
- Jie Cheng, MDj,
- Sacha Salzberg, MDk,
- Andrea Natale, MDl@andreanatalemd,
- James Cox, MDm and
- Dhanunjaya R. Lakkireddy, MDe,∗ (, )@DJ_Lakkireddy@OPRMC
- aDepartment of Cardiology, Icahn School of Medicine at Mount Sinai, New York, New York
- bHarvard Medical School, Boston, Massachusetts
- cHeart Center, Catharina Hospital, Eindhoven, the Netherlands
- dDepartment of Cardiovascular Surgery, Tokyo Metropolitan Tama Medical Center, Tokyo, Japan
- eKansas City Heart Rhythm Institute & Research Foundation, Overland Park, Kansas
- fClinic for Cardiovascular Surgery, University Hospital, Zurich, Switzerland
- gDivision of Cardiovascular Medicine, University of Louisville, Louisville, Kentucky
- hSt. Helena Hospital, St. Helena, California
- iBaylor College of Medicine, Houston, Texas
- jTexas Heart Institute, Houston, Texas
- kHeartClinic, Hirslanden Hospital, Zurich, Switzerland
- lTexas Cardiac Arrhythmia Institute, Austin, Texas
- mFeinberg School of Medicine, Northwestern University, Chicago, Illinois
- ↵∗Address for correspondence:
Dr. Dhanunjaya Lakkireddy, The Kansas City Heart Rhythm Institute (KCHRI), HCA MidWest, 12200 West 106th street, Overland Park Regional Medical Center, Overland Park, Kansas 66215.
Background Percutaneous left atrial appendage exclusion (LAAE) has evolved as an alternative strategy for stroke prevention in atrial fibrillation (AF). Recent observational data have suggested that epicardial LAAE can have substantial impact on arrhythmia burden and hemodynamic profile.
Objectives The authors aimed to study the impact of epicardial versus endocardial LAAE on systemic blood pressure in hypertensive AF patients.
Methods This was a prospective, nonrandomized study comparing 247 patients who underwent epicardial LAAE with 124 patients with endocardial exclusion. Clinical outcomes were measured at 3 months and 1 year. Primary outcome was improvement in systolic blood pressure (SBP) between both groups compared with baseline. Secondary outcome included changes in diastolic pressures (DBP), serum electrolytes, and creatinine.
Results There was no significant difference in baseline SBP between epicardial and endocardial groups. SBP was significantly lower in the epicardial group both at 3 months (122 ± 11.8 mm Hg vs. 129.7 ± 8.2 mm Hg; p < 0.001) and 1 year (123 ± 11.6 mm Hg vs. 132.2 ± 8.8 mm Hg; p < 0.001) compared with the endocardial group. An adjusted multivariate linear mixed effects model demonstrated that epicardial LAAE significantly decreased SBP by 7.4 mm Hg at 3 months and by 8.9 mm Hg at 1 year (p < 0.0001). There was a trend toward lower DBP with epicardial LAAE at 3 months by 1.3 mm Hg (p = 0.2) and at 1 year by 1.8 mm Hg (p = 0.09). There was no significant difference in serum electrolytes and creatinine between both groups.
Conclusions In hypertensive AF patients, epicardial LAAE significantly decreases SBP both at 3 and 12 months compared with endocardial exclusion.
Left atrial appendage (LAA) exclusion (LAAE) using an endocardial or epicardial approach has recently emerged as an alternative strategy for stroke prevention in atrial fibrillation (AF) in patients who are considered poor candidates for long-term oral anticoagulation (1–4). The advent of such exclusion strategies has stimulated interest in the broader function of the LAA and its role in cardiovascular physiology and systemic homeostasis. There are additional observational data that epicardial LAAE can significantly decrease AF burden (5) and can be associated with early and persistent reduction in systemic blood pressure on short-term follow-up (6,7). Thus far, no such adjunctive effects were seen with endocardial LAAE devices. Although, the exact mechanism for this reduction in systemic blood pressure remains unclear, persistent down-regulation of the renin-angiotensin-aldosterone system (RAAS) and its interaction with the sympathetic nervous system and natriuretic peptides with epicardial LAAE remains the most plausible explanation (7). Currently, there is a lack of data whether such favorable effects on systemic blood pressure with epicardial LAAE is temporary or persists on long-term follow-up. We hypothesized that in AF patients with a history of hypertension, epicardial LAAE with Lariat (SentreHEART, Redwood City, California) or AtriClip (AtriCure, West Chester, Ohio) device can significantly decrease systemic blood pressure at 3-month and 1-year follow-up when compared with endocardial LAAE with the Watchman device (Boston Scientific, Marlborough, Massachusetts).
This was a prospective observational study including 371 patients with AF and history of hypertension, considered poor candidates for long-term oral anticoagulation, and underwent either a successful epicardial or endocardial LAAE for stroke prevention at 4 centers. Epicardial LAAE was performed with AtriClip (n = 37) and the Lariat system (n = 210), whereas endocardial exclusion was performed with a Watchman device (n = 124). A successful LAAE procedure was defined as Lariat suture, Watchman, or AtriClip deployment with ≤2-mm leak by intraprocedural transesophageal echocardiography (TEE). Major adverse cardiac complications included either death, cardiac perforation (requiring intervention or surgery), or stroke at 30 days.
Inclusion and exclusion criteria
Inclusion criteria were as follows: 1) age ≥18 years; 2) history of nonvalvular AF; 3) CHA2DS2-VASc score ≥2; 4) poor candidate or unwilling to take long-term oral anticoagulation therapy; 5) a life expectancy of ≥1 year; 6) history of hypertension on ≥2 antihypertensive medications in the past 3 months; and 7) an estimated glomerular filtration rate of ≥45 ml/min/1.73 m2 by the Modification of Diet and Renal Disease formula.
Exclusion criteria included: 1) nonvalvular AF with no prior history of hypertension; 2) history of hypertension with <2 antihypertensive medications; 3) history of secondary hypertension; 4) prior renal artery intervention; 5) history of cardiothoracic surgery; 6) pectus excavatum; 7) myocardial infarction within the last 3 months; 8) embolic event within the last 30 days; 9) New York Heart Association functional class III to IV heart failure; 10) left ventricular ejection fraction (LVEF) <30%; 11) history of thoracic radiation; and 12) patients who required hospital admission for hypertension emergency within the past 1 year.
Patients meeting the above initial criteria for the study enrollment underwent a screening contrast cardiac computed tomographic scan or TEE (Online Figure 1). Additional exclusion criteria based on LAA anatomy included: 1) an LAA width >40 mm (for Lariat); 2) a superiorly oriented LAA with the LAA apex directed behind the pulmonary trunk (for Lariat); 3) bilobed LAA or multilobed LAA in which lobes were oriented in different planes exceeding 40 mm (Lariat); 4) a posteriorly rotated heart (Lariat); 5) shallow LAA (LAA ostium >31 mm or length <17 mm) (Watchman); and 6) LAA thrombus. Once the patient met the specified study inclusion and not the exclusion criteria, the type of procedure performed was at the discretion of the clinician. The institutional review boards of the respective institutions approved the study protocol. Informed consent was obtained from all the patients.
LAAE was performed using the AtriClip device in 37 patients (4) and the Lariat system in 210 patients, as described previously (5,8). AtriClip implantation was a thoracoscopic-guided, minimally invasive standalone procedure in all patients.
LAAE was performed using the Watchman device in 124 patients, as described previously (1,2).
Primary outcome was improvement in systolic blood pressure (SBP) at 3 months and 1 year. Secondary outcome included changes in diastolic pressure (DBP) at 3 months and 1 year, and change in serum electrolytes and creatinine at 3 months follow-up.
Baseline and procedural characteristics were obtained. SBP, DBP, and antihypertensive medications (type and dose) were assessed at 3 weeks before the procedure in the clinic (baseline), on the day of the procedure in the electrophysiology/cardiac catheterization lab (pre-procedural), and again at 3 and 12 months in the clinic. All blood pressure measurements were obtained twice in the supine position by a registered nurse using a blood pressure cuff with an automated blood pressure machine. The average of the 2 measurements was recorded.
Electrolytes and renal function were also checked at 3 weeks before the procedure in the clinic (baseline) and on clinic follow-up at 3 months post-procedure after overnight fasting. All patients were instructed to maintain their routine diet and were advised against consuming excess salt. Diuretic agents were held on the day of the procedure in all patients. The rest of the blood pressure medications were routinely continued. Acute renal insufficiency was defined as an increase in serum creatinine of ≥0.5 mg/dl from baseline. All patients were seen in clinic at 3, 6, 9, and 12 months. All patients were continued on their baseline antihypertensive regimen and underwent changes only when clinically necessary.
Categorical variables were presented as counts (percentage), and continuous variables as mean ± SD or median (interquartile range [IQR]), as appropriate. Descriptive statistics of patient demographics, comorbidities, CHA2DS2-VASc and HAS-BLED scores, medication use, and pre-procedural echocardiographic variables, that is, LVEF and LA size were displayed in the cohorts stratified based on the LAAE approach (epicardial vs. endocardial). Differences in the baseline characteristics were examined using Welch independent 2-sample t-test or Wilcoxon rank sum test as appropriate for continuous variables and chi-square test or Fisher exact test for categorical variables. Linear mixed models were used to compare outcomes between the 2 LAAE approaches, assuming linear change of blood pressure over time. The models included participant-specific and institution specific random effects to account for within-subject and within-institution correlations as a result of repeated assessments. Fixed effects in the model included LAAE approach, follow-up time, and the interaction between LAAE approach and follow-up time. All statistical analyses were performed with the use of R statistical computing environment (R Foundation for Statistical Computing, Vienna, Austria). All the hypothesis tests were 2-sided, and p values of <0.05 were considered to indicate statistical significance.
A total of 491 patients were initially enrolled. After undergoing screening TEE/cardiac computed tomography, 107 were excluded because they did not meet the additional pre-specified criteria. Final study enrollment included 384 patients in whom LAAE was attempted. The procedures were unable to be completed in 10 patients in the epicardial group (Lariat = 9, AtriClip = 1) and 3 in the endocardial group. A total of 371 patients successfully underwent the procedure (Online Figure 1). Successful epicardial and endocardial LAAE occurred in 95.9% and 97.5%, and major adverse complications occurred in 1.5% and 1%, respectively. There were no deaths.
Table 1 demonstrates the baseline characteristics of patients who underwent epicardial and endocardial LAAE. There was no significant difference between age (67.8 ± 10.8 years vs. 69.5 ± 7 years; p = 0.11), male sex (63.5% vs. 67.8%; p = 0.50), HAS-BLED score (3 [IQR: 2.5 to 4.0] vs. 3 [IQR: 2 to 4]; p = 0.50) or LVEF (55% [IQR: 55% to 60%] vs. 60% [IQR: 50% to 60%]; p = 0.20) between epicardial and endocardial groups. There was no significant difference between baseline SBP (138.9 ± 10.5 mm Hg vs. 139.3 ± 2.3 mm Hg; p = 0.72), but DBP (76.6 ± 8.1 mm Hg vs. 79.4 ± 8.1 mm Hg; p = 0.002) was significantly lower in the epicardial compared with the endocardial cohort.
There was no significant difference in median number of antihypertensive medications (3 [IQR: 2 to 5] vs. 3.5 [IQR: 2 to 5]; p = 0.50) at baseline between both groups. The proportion of patients on angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, aldosterone antagonists, beta-blockers, or amiodarone was not significantly different between both groups. The number of patients receiving calcium channel blockers was significantly higher in the endocardial group compared with epicardial (48% vs. 35%; p = 0.02). There was no significant difference in baseline serum sodium, potassium, magnesium, and creatinine levels between both groups.
Online Table 1 demonstrates the individual baseline characteristics of patients who underwent AtriClip (n = 37), Lariat (n = 210), and Watchman (n = 124) procedures. There was no significant difference in age, sex, diabetes, coronary disease, stroke, or heart failure between all 3 groups. LA size was significantly different between AtriClip, Lariat, and Watchman groups (35 [IQR: 30 to 40] mm vs. 30 [IQR: 27 to 37] mm vs. 41 [IQR: 38.7 to 44] mm; p < 0.001). There was a significant difference in baseline SBP (131 ± 13.6 mm Hg vs. 140.4 ± 9.1 mm Hg vs. 139.3 ± 2.2 mm Hg; p < 0.001) and DBP (79.8 ± 7.6 mm Hg vs. 76 ± 8.1 mm Hg vs. 79.4 ± 8 mm Hg; p < 0.001) between the 3 groups.
Blood pressure outcomes
All patients who underwent the procedure were followed up to 1 year. There were no patients lost to follow-up. Table 2 shows SBP, DBP, serum sodium, potassium, magnesium, and creatinine levels at 3 and 12 months in epicardial (n = 247) and endocardial (n = 124) groups. SBP was significantly lower in the epicardial cohort compared with endocardial at both 3 months (122 ± 11.9 mm Hg vs. 129.7 ± 8.2 mm Hg; p < 0.001) and 12 months (123 ± 11.7 mm Hg vs. 132.1 ± 8.8 mm Hg; p < 0.001), respectively. Similarly, DBP was also significantly lower in the epicardial cohort compared with endocardial at 3 months (71.7 ± 8.3 mm Hg vs. 75.8 ± 6.3 mm Hg; p < 0.001) and 12 months (71.1 ± 1.0 mm Hg vs. 75.8 ± 6.4 mm Hg; p < 0.001) post-procedure. There were significant differences in SBP and DBP at 3 months and 12 months in the individual groups between the AtriClip, Lariat, and Watchman groups (Online Table 2). The Central Illustration demonstrates the differences in SBP and DBP at baseline, and 3 and 12 months with epicardial and endocardial LAAE.
Number of antihypertensive drugs was significantly lower in the epicardial cohort compared with endocardial (median: 2 [IQR: 1 to 3] vs. 3 [IQR: 2 to 5]; p < 0.001) at 12 months of follow-up. An adjusted multivariate linear mixed effect model after adjusting for all differences in both groups demonstrated that the epicardial approach significantly decreased both SBP and DBP at 3 and 12 months. Epicardial approach significantly reduced SBP by 7.4 mm Hg at 3 months and 8.9 mm Hg at 12 months (p < 0.0001). There was a trend toward lower DBP at 3 (−1.3 mm Hg; p = 0.20) and 12 (−1.8 mm Hg; p = 0.09) months that did not reach statistical significance with the epicardial approach (Online Table 3).
Figure 1 shows the between-group differences in the change in SBP from baseline to 12 months in various pre-specified subgroups. There were significant differences between the epicardial and endocardial groups at 12 months compared with baseline across all subgroups. The highest reduction in SBP with epicardial exclusion was observed in patients with a history of heart failure (−11.66 mm Hg; 95% confidence interval [CI]: −5.85 to −17.5; p < 0.001), followed by diabetes (−10.65 mm Hg; 95% CI: −4.62 to −16.68; p < 0.001), and then age >65 years (−10.54 mm Hg; 95% CI: −7.5 to −13.6; p < 0.001).
Electrolyte and renal function
Online Table 2 demonstrates the electrolyte and creatinine changes at the predefined endpoints. Serum sodium, potassium, magnesium, and creatinine levels were not significantly different between the epicardial and endocardial groups and also among the individual types of LAA techniques.
First, epicardial LAAE significantly decreased SBP at 3 and 12 months in patients with a history of AF and hypertension. Second, the epicardial LAAE cohort had a significantly lower number of antihypertensive medications than the endocardial exclusion cohort at 1 year. Third, there was no significant difference in serum electrolytes (sodium, potassium, magnesium) and creatinine at 3 months between both groups. This substantial decrease in SBP on both short-term and long-term follow-up is due to the impact of epicardial exclusion on 2 separate pathways—the natriuretic peptide and neural pathways—both ultimately leading to inhibition of the RAAS. The impact on systemic blood pressure seems to be that of a group effect with epicardial exclusion compared with endocardial occlusion.
Prior studies have reported that although the right atrial appendage is the major source of cardiac atrial natriuretic peptides (ANP), the LAA stores about 30% of ANP, which is released into the blood circulation with atrial stretch or distension (9). ANP is an important neurohormonal regulator that plays a vital role in cardiovascular homeostasis by regulating natriuresis and diuresis in response to volume expansion to maintain adequate salt and water balance (10). Earlier studies in patients who underwent a maze procedure with bilateral atrial appendectomies demonstrated reduced ANP in the early post-operative phase (11,12) that remained low at 2 years (12). Despite the decrease in ANP levels, there were no reported data on its impact on systemic blood pressure until the recent observations from our group performing epicardial LAAE in patients with AF (6,7). In a retrospective study of 76 patients with AF who underwent epicardial LAAE with the epicardial exclusion (6), there was a significant reduction in SBP at 24 h post-procedure that was sustained at 3-month follow-up. Further insights regarding these changes were reported by another study from our group that showed that significant down-regulation of adrenaline, noradrenaline, renin, and aldosterone at 3 months occurred with epicardial LAAE, but no such changes were observed with endocardial exclusion (6,7,13). The reduction in short-term blood pressure seen with epicardial LAAE was attributed to inhibition of the natriuresis pathway from ANP degranulation from LAA necrosis. This hypothesis is further supported by our findings in the current study because >75% of patients developed post-procedural hypotension within 24 to 48 h that resolved subsequently after discontinuing or changing the antihypertensive drug they were previously taking. It is, however, interesting to note that there was a persistent decline in clinic blood pressure recordings at 3 months and 12 months follow up. These findings are consistent with our prior observations of approximately 15% decline in SBP at 3 months that also correlated with an inhibition of the neural pathway by epicardial LAAE (6,7).
The LAA is richly innervated by both sympathetic and parasympathetic nerves that may participate in reflex response to stretch (14,15). It is possible that epicardial LAAE can impact the afferent neural connections and the peri-ganglionated plexi, subsequently modifying the neural reflexes and autonomic feedback mechanisms that may be interconnected with RAAS and natriuretic peptides, thus regulating systemic hemodynamics. This association has not been demonstrated in animal models or clinical studies and remains an area of further investigation.
Although this study was a prospective, nonrandomized study, the significant decline in 1-year SBP with epicardial LAA ligation, after adjusting all the other clinical variables, is much higher than previously observed with any interventional therapy targeting systemic blood pressure (16–19). Furthermore, the reduction in SBP remained consistent across all subgroups, including age >65 years, sex—both males and females, body mass index >30, history of diabetes, history of heart failure, and use of aldosterone antagonist. Our results suggest that epicardial LAAE can be another way of targeting the autonomic innervations and the RAAS system in hypertensive AF patients to improve overall clinical outcomes. Earlier studies on pharmacological therapy in hypertension have shown a benefit in cardiovascular morbidity with improvement in SBP of 2 to 5 mm Hg (20).
As we are aware, RAAS up-regulation is also known to be associated with atrial fibrosis, myocyte hypertrophy, and AF (21). Theoretically, improvement in blood pressure can impact these anatomic and hemodynamic changes by decreasing atrial wall stress, and decreasing left ventricular diastolic and pulmonary capillary wedge pressure, improving clinical outcomes in these patients.
This was a prospective, nonrandomized study that has its inherent limitations. First, we recorded baseline blood pressure 3 weeks before the procedure in the clinic setting, which we believe is a reasonable surrogate of long-term blood pressure control. However, we cannot rule out the possibility of minor variations in blood pressure due to stress, nutritional status, medication compliance, and so on. Second, medication and dietary adherence cannot be reliably confirmed between both groups. Third, changes in antihypertensive medications during a follow-up visit could have impacted our results. Fourth, although blood pressure was not recorded using an ambulatory 24-h recording, care was taken to schedule appointments and measure blood pressure in the clinic at the same time during the 3 visits to minimize discrepancy. Fifth, it is possible that the impact of epicardial LAAE on hypertension may diminish over time and longer follow-up. Despite, the limitations, these study results are provocative and open a window of knowledge on how a primordial embryological remnant like the LAA can play a major role in systemic hemodynamics. This study and prior data suggest the possibility of utilizing epicardial LAAE as a potential intervention for multidrug-resistant hypertension in patients with AF. A multicenter randomized control trial is required to further confirm our findings.
Epicardial LAAE in AF patients with a history of hypertension significantly decreases systolic blood pressure both on short-term and long-term follow-up compared with endocardial exclusion. Further studies are required to examine the underlying mechanism for these physiological changes.
COMPETENCY IN MEDICAL KNOWLEDGE: LAA exclusion has evolved as an alternative strategy for prevention of thromboembolism in patients with AF. In patients with AF and hypertension, epicardial LAA exclusion decreases systolic blood pressure more than endocardial exclusion both acutely and during long–term follow-up.
TRANSLATIONAL OUTLOOK: Further studies are required to elucidate the neurohormonal mechanisms responsible for the differential effects of epicardial and endocardial LAA exclusion.
Dr. Verberkmoes has been a consultant for AtriCure. Dr. Emmert has received speaker fees from AtriCure. Dr. Gopinathannair has received research grants and/or honoraria from Pfizer, Bristol-Myers Squibb, Zoll Medical, Abbott Medical, HealthTrust PG, Boston Scientific, and Abiomed. Dr. Dunnington has been a consultant for AtriCure. Dr. Natale has been a consultant for Biosense Webster, Stereotaxis, Abbott, and Boston Scientific; and has received speaker/travel honoraria from Medtronic, AtriCure, EPiEP Biotronik, and Janssen. Dr. Cox has been a consultant for AtriCure and SentreHEART; holds stock in Adagio Medical, AtriCure, and SentreHEART; and is a cofounder of and has served on the board of directors of Adagio Medical. Dr. Lakkireddy is a principal investigator for the aMAZE study sponsored by SentreHEART and the AMULET investigational device exemption trial sponsored by St. Jude Medical/Abbott. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- atrial natriuretic peptide
- diastolic blood pressure
- left atrial appendage
- left atrial appendage exclusion
- left ventricular ejection fraction
- renin-angiotensin-aldosterone system
- systolic blood pressure
- transesophageal echocardiography
- Received June 11, 2018.
- Accepted June 22, 2018.
- 2018 American College of Cardiology Foundation
- Holmes D.R. Jr..,
- Kar S.,
- Price M.J.,
- et al.
- Bartus K.,
- Han F.T.,
- Bednarek J.,
- et al.
- Smith N.E.,
- Joseph J.,
- Morgan J.,
- Masroor S.
- Lakkireddy D.,
- Sridhar Mahankali A.,
- Kanmanthareddy A.,
- et al.
- Lakkireddy D.,
- Turagam M.,
- Afzal M.R.,
- et al.
- Price M.J.,
- Gibson D.N.,
- Yakubov S.J.,
- et al.
- Stewart J.M.,
- Dean R.,
- Brown M.,
- et al.
- Bartus K.,
- Podolec J.,
- Lee R.J.,
- et al.
- Li Z.,
- Zhao Q.Y.,
- Huang H.,
- Yang B.,
- Jiang H.,
- Huang C.X.
- Pokushalov E.,
- Romanov A.,
- Corbucci G.,
- et al.
- Esler M.D.,
- Krum H.,
- Schlaich M.,
- Schmieder R.E.,
- Bohm M.,
- Sobotka P.A.