Hybrid intervention approach to coronary artery and valvular heart disease

Abstract

Coronary angiography is used to assess the burden of coronary artery disease prior to surgical valve repair/replacement and often leads to concomitant bypass and valve surgery. We sought to evaluate outcomes of an alternative, hybrid approach involving percutaneous coronary intervention (PCI) and valve surgery, assessing the rate of stent thrombosis as a primary outcome. We reviewed charts of consecutive patients who underwent planned PCI prior to surgical valve repair/replacement by a single surgeon from January 2008 to December 2016. We calculated rates of surgical complication, duration of dual antiplatelet therapy (DAPT) prior to surgery, and rates of stent thrombosis and in-stent restenosis. Twenty-four patients were included in this study. Surgery was performed a median of 52.5 days following PCI. DAPT was withheld an average of 8 days before and resumed an average of 4 days after surgery. Ninety-two percent of surgeries were minimally invasive. There were no bleeding complications, stent thromboses, or restenosis events. All patients survived the 1-year follow-up. For patients with mixed coronary and valvular heart disease, a heart team approach involving preoperative PCI followed by staged minimally invasive valvular surgery appears to be safe and warrants further exploration.

Patients in need of surgical valve repair/replacement routinely undergo preoperative assessments of coronary artery disease burden via left heart catheterization to determine the appropriateness of coronary artery bypass grafting (CABG) at the time of valve surgery. ¹ However, vein graft patency has been as low as 70% 1 year following CABG, and improvements in surgical techniques afford less invasive approaches to valve replacement and repairs, at times forgoing sternotomy altogether. ² Doing so may lead to earlier ambulation, decreased intensive care unit stay, decreased risk of perioperative myocardial infarction and neurologic complications, and increased patient satisfaction. ³ In addition, improvements in drug-eluting stents and dual antiplatelet therapy (DAPT) have contributed to a decreased risk of stent thrombosis within 1 year of percutaneous coronary intervention (PCI). ^4–6 While the concept of hybrid PCI to bypass is not new, work has been limited to feasibility evaluations for patients who required bypass grafting with an internal mammary artery to the left anterior descending artery and PCI to other vessels. ⁷ As such, we sought to evaluate the hybrid approach of PCI to valve surgery, focusing on the rate of stent thrombosis.

METHODS

This study involved a retrospective review of consecutive cases at Baylor University Medical Center who underwent staged procedures of revascularization via PCI and subsequent surgical valve replacement or repair from 2008 to 2016. Specifically, we first identified nonemergent surgical valve replacement or repair cases that also had preoperative left heart catheterization performed at Baylor University Medical Center. We used the American College of Cardiology CathPCI registry to subsequently identify those who received PCI at the time of preoperative left heart catheterization, limiting the timeframe between PCI and surgery to 1 year. Chart review was performed to capture perioperative events including cerebrovascular accident, major and minor bleeding, sepsis, and return to the operating room for surgical revision. All patients were followed for 1 year following surgery, per approval from the Baylor University Medical Center’s institutional review board with a waiver of informed consent.

RESULTS

Twenty-four patients were included in this study ( Table 1) . All patients received a single stent and had an average stent length of 14 mm (range 8–38 mm). One patient (4%) received an everolimus-eluting bioabsorbable scaffold, and all others received drug-eluting stents. Of the drug-eluting stents, 18 (78%) were zotarolimus eluting, 4 (17%) were everolimus eluting, and 1 (4%) was paclitaxel eluting.

Table 1.

Patient characteristics and procedure details

Case	Age	Sex	Days from PCI to valve surgery	Valve replacement		Valve repair			Bioprosthetic valve	PCI to coronary artery				Cross clamp time (min)	P2Y12 inhibitor held prior to surgery (days)	Medical history
				Aortic	Mitral	Aortic	Mitral	Tricuspid		LM	LAD	LCx	Right			Diabetes mellitus	Hypertension	Dyslipidemia	CVA	Systolic heart failure
1	74	M	248	+	–	–	–	–	+	–	+	–	–	45	12	–	–	–	–	+
2	73	M	39	+	–	–	–	–	+	–	–	–	+	61	8	+	+	+	–	–
3	70	M	36	+	–	–	–	–	+	–	–	–	+	61	7	+	+	+	–	–
4	69	M	35	+	–	–	–	–	+	–	–	+	–	90	7	–	+	+	–	–
5	74	M	29	+	–	–	–	–	+	–	+	–	–	73	7	–	+	+	–	+
6	63	M	27	+	–	–	–	–	+	–	–	–	+	139	9	–	+	+	–	+
7	71	F	97	+	–	–	–	–	+	–	+	–	–	60	6	–	+	+	–	+
8	72	M	81	+	–	–	–	–	+	–	–	+	–	68	7	+	+	+	–	–
9	80	M	70	+	–	–	–	–	+	–	–	+	–	55	39	–	+	+	–	–
10	72	M	53	+	–	–	–	–	+	–	–	–	+	75	5	+	+	+	–	–
11	66	M	52	+	–	–	–	–	+	–	+	–	–	61	3	–	+	+	–	+
12	71	M	52	+	–	–	–	–	+	–	–	+	–	57	22	+	+	–	+	+
13	71	M	49	+	–	–	–	–	+	–	–	+	–	62	3	–	+	+	–	+
14	70	M	49	+	–	–	–	–	+	–	–	–	+	63	8	–	+	+	–
15	65	M	42	+	–	–	–	–	+	–	+	–	–	63	12	–	–	+	–	–
16	80	M	158	+	+	–	–	–	+/+	–	–	–	+	117	–	+	+	+	–	+
17	67	F	234	–	+	–	–	–	+	–	–	+	–	76	–	–	+	+	–
18	51	M	69	–	+	–	–	–	+	–	+	–	–	83	7	+	–	+	–	+
19	62	F	144	–	+	–	–	–	+	+	–	–	–		8	–	+	+	–	+
20	64	M	92	–	–	–	+	–	–	–	–	–	+	86	8	–	+	–	–	–
21	77	F	81	–	–	–	+	–	–	–	+	–	–	127	14	+	+	+	–	+
22	77	M	50	–	–	–	+	–	–	–	–	+	–	116	6	–	+	+	+	+
23	74	M	31	–	–	–	+	–	–	–	–	+	–	136	8	–	+	+	–	+
24	63	M	267	–	–	–	–	+	–	–	–	–	+		10	–	+	+	–	+

LAD indicates left anterior descending artery; LCx, left circumflex artery; LM, left main coronary artery. Diabetes mellitus is defined as hemoglobin A1c >6.5% or use of oral glucose-lowering therapy or insulin; hypertension, systolic blood pressure >140 mm Hg or diastolic >90 mm Hg or use of American Heart Association/American College of Cardiology antihypertensive therapies; dyslipidemia, low-density lipoprotein >160 mg/dL or triglycerides > 150 mg/dL not on lipid-lowering therapy or use of lipid-lowering therapy; CVA, cerebral vascular accident: history of radiologic evidence of ischemic or hemorrhagic disruption of blood supply to brain; systolic heart failure, ejection fraction <50% by echocardiography or magnetic resonance imaging.

Surgery was performed a median of 52.5 days following PCI. Importantly, 22 patients (92%) were able to undergo a minimally invasive surgical approach, avoiding a full sternotomy. Two patients (8%) were not on DAPT prior to surgery, as their referring cardiologists had deemed prior DAPT duration to be sufficient. Clopidogrel was the predominant P2Y12 inhibiting agent, used in 23 patients (96%), and was discontinued an average of 9 days prior to surgery. One patient (4%) took ticagrelor, which was withheld 4 days prior to surgery. DAPT was resumed in 21 of 22 patients (95%) and was done so an average of 4 days (range 1–15) after surgery. Timing of P2Y12 inhibition cessation before valve surgery and reinitiation after was at the discretion of the surgeon. Four patients (17%) were on anticoagulation therapy prior to surgery for atrial fibrillation (three on warfarin and one on rivaroxaban).

All patients survived the 1-year follow-up. Additionally, there were no stent thromboses. Only 1 patient (4%) had repeat left-sided heart catheterization within 1 year, and there was no evidence of target vessel failure. There were no instances of major or minor bleeding. Complications occurred in three patients (13%); one (4%) had a cerebrovascular accident and two (8%) developed sepsis (during the surgical hospitalization). Two patients (8%) developed postoperative atrial fibrillation; they had been on DAPT for 234 and 158 days, respectively, and were transitioned to aspirin and apixaban.

DISCUSSION

In this study of 24 patients who underwent staged PCI to valve surgery, only two (8%) required a full sternotomy. Additionally, there were no instances of confirmed or suspected early thrombosis or in-stent restenosis, despite short-term perioperative cessation of DAPT. Further, only one patient (4%) was required to repeat angiography within 1 year of surgery, and there was no evidence of target vessel failure for that patient.

While DAPT is recommended following PCI to decrease the risk of stent thrombosis, there is an increased risk of bleeding, transfusion, and reexploration for patients on DAPT during cardiac surgery compared to those on aspirin monotherapy. ^8–10 As such, both the Society of Thoracic Surgeons (STS) and the European Association for Cardiothoracic Surgery assign a class I and IIa recommendation, respectively, for the cessation of P2Y12 therapy prior to surgery. ^{11 , 12} Hence, this raises concern for cases, such as those in our study, in which surgery is planned within a relatively short period following stent placement. Yet, our finding of no thrombotic events in this cohort suggests that the hybrid approach of PCI to valve surgery may be safe. There are no consensus guidelines regarding the timing of DAPT resumption following surgery beyond a risk-benefit assessment in regards to bleeding risk; patients in this study resumed DAPT an average of 4 days following valve surgery.

This hybrid approach of PCI to valve surgery decreases surgical risk. In fact, the STS score for estimating perioperative risk assigns a higher risk value to patients undergoing combined valvular and CABG surgery compared to isolated procedures. ^{13 , 14} Further, the STS 2018 update on outcomes and quality revealed that 83% of combination mitral valve replacement and CABG surgeries required blood product, compared to approximately 66% and 32% of isolated mitral valve replacement and repair cases, respectively. ¹⁴ Other studies have reiterated the reduction in blood product usage and perioperative complication rates in minimally invasive aortic valve repair or replacement, mitral valve repair or replacement, and hybrid PCI and CABG. ^15–17 In the same STS report, the rates of permanent stroke following combination surgeries (aortic valve replacement and CABG, mitral valve replacement and CABG, and mitral valve repair and CABG) ranged from 2.2% to 3.1%; in our study, one patient (4%) experienced a cerebrovascular accident. ¹⁴ Additionally, the STS reported in-hospital mortality outcomes ranging from 2.7% to 8.3%; all patients in our study survived to discharge after undergoing this hybrid approach, indicating that the hybrid approach may have a comparable or improved safety profile. Although national rates of 30-day (or longer) mortality are not available via STS, all patients in our study survived the full 1-year follow-up.

We are not the first to investigate hybrid PCI and valve surgery. In 2012, Santana and colleagues compared results from a hybrid PCI to valve surgery (an average of 24 days prior) approach compared to a matched cohort of combination CABG and valve repair. They found the incidence of postoperative complications occurred significantly less frequently with the hybrid approach than with the traditional approach ¹⁸ ; however, there was no mention of the incidence of stent thrombosis. In 2005, Byrne and colleagues published confirmatory results with hybrid PCI and valve surgery in high-risk patients during the same hospitalization. ¹⁹ Similarly, George and colleagues in 2015 demonstrated safety in hybrid stenting followed by immediate valve surgery in high-risk patients. ²⁰ With the improving safety profile of drug-eluting stents and suboptimal long-term outcomes of venous grafting, the question of a hybrid approach deserves further exploration in larger clinical trials. ^{2 , 7}

This work has all of the limitations associated with a small, retrospective, observational study, including a lack of standardized wait times between PCI and surgery. Accordingly, our findings may not be generalizable to all centers. Further, in-stent restenosis and stent thrombosis are rare events given the improvements in stent design, antiproliferative coating, and DAPT, so our limited sample size may have simply been too small to observe such events. Additionally, DAPT duration is related to surgical timing after PCI, which is unique in each case due to a multitude of factors, including patient preference, surgeon availability, and operating room availability.

In conclusion, our observations suggest that, for patients with mixed coronary and valvular heart disease, a heart team approach involving interventional cardiology and cardiothoracic surgery to perform preoperative PCI followed by staged valvular surgery may be safe without substantial increased risk for stent thrombosis or early restenosis, despite holding DAPT prior to surgery. This allows the employment of minimally invasive surgical techniques, leading to faster patient recovery and reduced risk of complications. Further investigation with larger randomized controlled trials and longer follow-up is warranted.

DISCLOSURE STATEMENT

Dr. Stoler is a proctor and advisory board member for Boston Scientific and Medtronic, as well as a proctor for Edwards. The remaining authors have nothing to disclose.

Funding

This study was partially funded by the Baylor Health Care System Foundation.