Author + information
- Sandeep Joshi, MD,
- Jonathan S. Steinberg, MD, FACC⁎ (, )
- Robert C. Ashton Jr., MD,
- Sandhya Balaram, MD, PhD,
- Avi Fischer, MD, FACC and
- Joseph J. DeRose Jr., MD
- ↵⁎Division of Cardiology, St. Luke’s–Roosevelt Hospital Center, Division of Cardiology, 1111 Amsterdam Avenue, New York, NY 10025
To the Editor:Cardiac resynchronization therapy (CRT) via coronary sinus (CS) lead placement has been shown to improve heart failure (HF) symptoms (1–3). However, a 5% to 10% implant failure rate has resulted in alternative approaches for optimal CRT. Robotically assisted left ventricular (LV) epicardial lead placement via the posterior approach was developed by our group as a minimally invasive rescue procedure for patients with failed CS leads (4). Our goal was to prospectively assess the clinical, electrophysiological, and echocardiographic effects of this approach at a median 15 months postoperatively.
Forty-two patients were referred for robotic LV lead implantation (Table 1).Demographic, operative, and post-operative data were collected and analyzed per institutional review board guidelines. All patients had been on maximal medical therapy for at least three months before CRT. Pre-operative evaluation consisted of an electrocardiogram for verification of QRS >130 ms. Tissue Doppler imaging (TDI) was used to localize the latest site of mechanical activation in our last 10 patients with intraoperative TDI confirming resynchronization.
The technique used for robotically assisted LV lead placement has been described previously (4). General anesthesia with selective right lung ventilation was used with patients positioned in the full posterolateral thoracotomy position. The da Vinci robotic surgical system was used for all portions of the surgery. A camera port was placed in the seventh intercostal space (ICS) in the posterior axillary line. The left and right arms were positioned in the ninth and fifth ICS, and a working port was inserted posterior to the camera port, used for the introduction of leads and sutures.
Two leads were placed along the posterolateral LV surface using the robotic arms. Pericardial closure with leads tunneled to the left subpectoral pocket occurred next. One lead, the active lead, was connected to the device and the second lead was left in the pocket as a back-up lead. The remainder of the CRT system was implanted using standard transvenous techniques.
Patient follow-up occurred at a median 15 months (range, 3 to 34 months) (3-month intervals for device interrogation and within 6 months for echocardiograms). Response was defined as ≥1 New York Heart Association functional class improvement.
Paired ttesting and analysis of variance compared pre-operative and post-operative changes between groups. Cox regression analysis determined predictors of response.
Eighty-four epicardial leads were placed. All patients had successful lead placement within the targeted area. Intraoperative lead threshold and R-wave were 0.8 ± 0.5 V and 16 ± 9 mV. Operative time was 51 ± 32 min. Left ventricular injury and pleural adhesions necessitated conversion to mini-thoracotomy in two patients. When the learning curve of the first five cases was removed from analysis (108 ± 55 min), operative time decreased to 45 ± 13 min for the next 37 cases.
Patients were divided into two groups (first cardiac surgery [primary]) (n = 21) and more than one prior cardiac surgery (re-op) (n = 21). Although the re-op group manifested higher incidences of ischemic cardiomyopathy, diabetes mellitus, and age, no significant differences (operative/follow-up) were found between the groups.
Discharge occurred at a mean of one postoperative day with no perioperative mortality. In-hospital and 30-day complications included pneumonia (n = 1), medically treated ischemic colitis (n = 1), renal insufficiency (n = 1), failure (n = 1), urinary tract infection (n = 1), intercostal neuropathy (n = 2), and an infected device requiring explant (n = 1).
The three-month clinical response rate was 81% (34 of 42) and 71% (30 of 42) at maximum follow-up (Fig. 1).Five deaths occurred, four patients experienced worsening HF, and three patients experienced no change. One patient underwent cardiac transplantation.
Echocardiography at 12 ± 7 months showed significant decreases in LV internal dimension index (LVIDI) systolic (s) [−10%] and diastolic (d) [−7%]. The LV ejection fraction improved by 7%.
Multivariate Cox regression analysis of various clinical and echocardiographic parameters showed pulmonary hypertension (right-sided ventricular pressure ≥35 mm Hg) was associated with a worse outcome (p < 0.001).
Three patients experienced lead failures at 1, 9, and 14 months secondary to failure of lead capture. Successful revision via the second lead was performed in these patients.
Successful CRT relies on accurate LV lead placement. Numerous investigators have shown that posterolaterally placed LV leads result in greater degrees of intraventricular resynchronization (5), and with TDI, the site for optimal lead placement may now be determined pre-operatively, therefore lead insertion techniques should aim to place leads in the most optimal zone for resynchronization, a goal for alternative approaches for CRT.
Variations in CS venous tributaries remain the major limitation for optimal CRT via percutaneous LV leads. A minimally invasive approach is particularly desirable in this setting. Limited thoracotomy carries a morbidity of a rib spreading incision and limits lead placement options. Surgeons have championed thoracoscopic approaches to the LV for full surgical access to the entire epicardial surface. The technical demands required for re-operative surgery and that with massive cardiomegaly have resulted in significant conversion rates based on prior studies. In comparison, robotic technology affords complete LV access, and the three-dimensional vision, scaled motion, and wrist-like mechanism allow for delicate pericardial dissection during re-operative surgery and fine dissection in the limited working space afforded by these large hearts.
In all of the patients in this series, the desired site for LV pacing was achieved. The 81% three-month response rate achieved in this series compares favorably to the 60% to 70% response rate reported with percutaneous CS LV lead placement (1–3), which may be attributed to 100% posterolateral lead placement in our patients. However, a non-responder rate did exist, and it is postulated that because all patients presumably had maximal resynchronization, other factors, including severe LV dysfunction, can impact outcome.
Left ventricular reverse remodeling has been reported with percutaneous CRT and seems to be dependent on successful resynchronization confirmed by TDI. This study is the first to confirm reverse remodeling in patients undergoing robotic surgical CRT. However, our data on TDI predicting clinical response is too limited to make any conclusions.
In summary, robotically assisted LV epicardial lead implantation is a safe and effective technique for CRT in all patients. Clinical response rates and LV remodeling data compare favorably with those achieved with CS leads. It is an invaluable technique allowing for minimally invasive rescue therapy after failed CS leads. Its role in primary implantation awaits further study.
Please note: Dr. DeRose receives consultancies from Intuitive Surgical Inc. and research funding from Medtronic Inc. Dr. Steinberg receives research funding from Medtronic Inc. and is a consultant and receives research support from Guidant Corp., St. Jude Medical, and Biotronik, Inc.
- American College of Cardiology Foundation
- Alonso C.,
- Leclercq C.,
- d’Allones F.R.,
- et al.
- Butter C.,
- Auricchio A.,
- Stellbrink C.,
- et al.
- Butter C.,
- Auricchio A.,
- Stellbrink C.,
- et al.