A commonly encountered scenario in the clinical practice of an interventional cardiologist involves a patient who has a stent and requires non-cardiac surgery: what should be done with the antiplatelet regimen? Defining the perioperative risk for stent thrombosis in patients with coronary artery disease and prior stenting is a frequent dilemma. The challenge of selecting individuals who can safely interrupt dual antiplatelet therapy (DAPT) to minimize surgical bleeding exposes these patients to an angst-provoking mechanism of post-operative myocardial infarction, namely, increased platelet reactivity.1 Recently, the Surgery After Stenting 2 score (SAS 2) was published and a mobile app version (“Stent & Surgery” by Araneum Group srl) created to assist with decision-making strategies for guiding perioperative DAPT management (interrupting, bridging, resuming) in patients with coronary stents undergoing non-cardiac surgery.2 Emphasis is placed on balancing the thrombotic-to-hemorrhagic risk profiles through a multidisciplinary collaboration between cardiologist and surgeon. Our first case highlights the limitations of commonly utilized risk predictors for perioperative stent thrombosis, while our second case features the potential utility of optical coherence tomography (OCT) as a risk-sharpening tool to predict safe DAPT interruption in high-risk patient subsets.
Case 1: Limitations of Commonly Utilized Risk Predictors for Perioperative Stent Thrombosis
A 75-year-old male with history of ischemic heart disease and on long-term DAPT presented with intractable trigeminal neuralgia refractory to medical therapy. Neurosurgical recommendations included trigeminal nerve decompression off DAPT to minimize intracranial hemorrhagic risk. Cardiology was thereafter consulted to define his perioperative cardiac ischemic-thrombotic risk. Percutaneous coronary intervention (PCI) was first performed more than 10 years ago in the setting of acute coronary syndrome, and again 3 years prior in the setting of stable ischemic symptoms. Stent types and implant guiding strategies were unknown. He was asymptomatic with mild ischemic cardiomyopathy at baseline (left ventricular ejection fraction [LVEF] of 48%). Preoperative testing included echocardiography, which showed reduction of LVEF to 30%, along with inferolateral wall hypokinesis. Transradial coronary angiography showed patent stents in the left anterior descending (LAD) artery (Figure 1A) and right coronary artery (RCA) (Figure 1B), and otherwise no obstructive disease. Clinical features for ischemic risk were thus limited to LVEF <35%, while the risk of stent thrombosis was estimated to be low based on the time from last stenting to non-cardiac surgery being >1 year. The patient had been maintained on guideline-directed medical therapy for chronic ischemic heart disease which, in addition to DAPT, included a beta-blocker and high-intensity statin.
DAPT was held for several days prior to surgery without a bridging strategy, based on high neurosurgical bleeding risk. Successful vascular decompression of the trigeminal nerve was achieved through a retro-mastoid craniotomy in <2.5 hours with <4 hours total anesthesia time. Hematocrit remained normal with minimal blood loss. The patient made an early, uneventful recovery, but developed severe chest pain, nausea, and vomiting 11 hours later, near midnight. An electrocardiogram (EKG) demonstrated findings consistent with an inferior wall ST-elevation myocardial infarction (STEMI) (Figure 1C). The 24-hour in-hospital interventional cardiologist was alerted3, and recommendations for emergent cardiac catheterization with intended primary PCI and antithrombotic pharmacotherapy (cangrelor and unfractionated heparin) were recognized as clinically necessary by the neurosurgeon.
Angiography of the suspected non-culprit left coronary system showed an unexpected acute non-occlusive thrombus in the mid-LAD stent with TIMI-3 flow (Figure 1D), while the culprit mid-right coronary artery (RCA) was confirmed to demonstrate findings of occlusive stent thrombosis. Aspiration thrombectomy (Figure 1E) followed by angioplasty restored TIMI-3 flow. Intravascular ultrasound (IVUS) demonstrated a well-apposed stent that was considered undersized/under-expanded, based on reference measurements distal and proximal to the stent (Figure 1F-H). Re-expansion with a balloon sized 1:1 to the distal reference beyond nominal pressure failed to maintain appropriate vessel patency that was ultimately achieved by deploying a 4.0 x 33 mm drug-eluting stent. The procedural course was complicated by incessant ventricular fibrillation and cardiac arrest. Rapid sequence intubation, advanced cardiac life support, and placement of an intra-aortic balloon pump resulted in electrical-mechanical stabilization. Daily head computed tomography (CT) scans fortunately showed no evolving intracranial hemorrhage while the patient was maintained on IV cangrelor and resolving postoperative changes. An magnetic resonance imaging (MRI) scan of the brain on post-PCI day 6 showed multifocal punctate acute ischemic changes in a watershed distribution. The patient steadily recovered from consequent mild neurocognitive deficits and remained without recurrent facial pain off all prior pain medications.
Case 2: OCT as a Risk-Sharpening Tool to Predict Safe DAPT Interruption
A 72-year-old established female patient with history of diabetes and ischemic heart disease on >1-year DAPT duration following management of subacute stent thrombosis presented for perioperative risk assessment prior to elective endoscopy-guided esophageal dilatation for treatment of dysphagia. The gastroenterologist requested interruption of DAPT for several days prior to and during the procedure. The patient’s initial encounter was in the setting of stent thrombosis (Figure 2A) that occurred 2 days following an index angio-guided PCI of the mid LAD at an outside hospital. The mechanism of stent thrombosis, defined by IVUS, included geographic miss (Figure 2B) of the stent to the intended lesion in addition to stent under-expansion and malapposition. Treatment included stent re-expansion with a noncompliant balloon and placement of an additional drug-eluting stent to bridge the missed residual lesion, each sized according to IVUS guidance. As a VerifyNow assay (Accriva) indicated high on clopidogrel platelet reactivity (298 PRU), ticagrelor was selected for P2Y12 inhibition. The patient remained clinically stable with no recurrent angina, though subsequent echocardiograms demonstrated apical wall akinesia consistent with an infarct. Her LVEF stabilized at 40%. An alternative strategy was selected to define the patient’s individualized risk for stent thrombosis off DAPT with the use of transradial coronary angiography and OCT imaging to confirm stent strut neointimal coverage prior to safely interrupting DAPT. Angiography confirmed widely patent stents in the mid LAD (Figure 2C), while OCT imaging demonstrated normal stent apposition (Figure 2D) with a diffusely homogenous neointimal pattern (Figure 2E). Ticagrelor was thus interrupted for 5 days and aspirin for 3 days prior to the procedure, which was performed under monitored anesthesia care with no perioperative complications.
PCI has become the most commonly performed procedure in medicine.4 More than 600,000 patients per year receive a coronary stent in United States, while 23% require some type of non-cardiac surgery within 2 years of stent implantation.5 Stent thrombosis is a potentially catastrophic complication that commonly presents as STEMI or sudden cardiac death. Although newer-generation stents are very efficacious in preventing restenosis, stent thrombosis continues to remain a concern, albeit at a lower rate.6 Compliance with DAPT reduces stent thrombosis event rates; however, seeking to avoid potential for major bleeding is a frequent cause of interruption when non-cardiac surgery is necessary.
Studies from the pre drug-eluting stent (DES) era suggested a potential for major adverse cardiovascular outcomes as high as 20% in patients undergoing surgery within 6 weeks of stent placement. Subsequent studies showed a persistent elevated thrombotic risk beyond 6 weeks following bare metal stent placement and 1 year after DES placement.7 Second- and third-generation DES have lowered the risk of stent thrombosis such that current American College of Cardiology/American Heart Association guidelines recommend delaying non-cardiac surgery until 30 days after bare metal stent placement, and ideally, 6 months after DES placement, unless clinical judgment indicates that the benefits exceed the risks for earlier (3-6 months after DES placement) surgery.8
The recently published SAS 2 score is derived from best efforts of multidisciplinary societies in Italy to create a systematic approach for guiding antithrombotic management in patients with coronary stents undergoing various non-cardiac surgeries.2 The main predictors of stent thrombosis are largely based on combined clinical and angiographic ischemic risk characteristics, although the greatest determinant is essentially the time from surgery to PCI. Using OCT, Adriaenssens et al identified uncovered stent struts, malposition, and stent dismantling as key morphological features of stent thrombosis.9 Uncovered struts are expected findings on OCT following stent implant; however, in-vivo case control studies have shown that up to 80% of patients with late stent thrombosis exhibit still as-yet uncovered struts.10 Finn et al reported >30% uncovered struts as a significant predictor of late stent thrombosis.11 In addition, stent malapposition and underexpansion, neointimal hyperplasia, and neoatherosclerosis have also been associated with late (<1 year) and very late (>1 year) stent thrombosis.9,12-14
Our first case presentation highlights the importance of weighing the perioperative thrombotic-to-bleeding risk according to individualized patient risk profiles, from coronary artery disease burden to stent type and implant guiding strategies, if known (i.e., angiography vs OCT or IVUS guidance). It also, however, highlights the limitations of the SAS 2 score as a risk-defining tool, as well as the inability of angiography to predict stent thrombosis. The greatest predictor for stent thrombosis utilizing the SAS 2 score includes the time of stent implant to surgery, with the lowest risk if stents were placed >1 year out from surgery. Preoperative diagnostic coronary angiography is often considered in patients with coronary stents when clinically appropriate, as in our patients. Clinical uncertainty remains as to whether imaging beyond luminography (OCT or IVUS) need be considered, especially if an angio-only approach was used to guide the index stent implant. Intravascular imaging may be useful to confirm appropriate neointimal stent coverage and confirm appropriately sized, apposed, and expanded stents, and thus may further enhance a thrombotic risk profile. Finally, utilizing a standardized bridging protocol as described by Rossini et al2 can be considered when hemorrhagic risk is redefined as acceptable or low by the surgical team.
Disclosures: The authors report no conflicts of interest regarding the content herein.
The authors can be contacted via Dr. Louie Kostopoulos at email@example.com.
- Landesberg G, Beattie WS, Mosseri M, et al. Perioperative myocardial infarction. Circulation. 2009 Jun;119(22): 2936-2944.
- Rossini R, Tarantini G, Musumeci G, et al. A multidisciplinary approach on the perioperative antithrombotic management of patients with coronary stents undergoing surgery: surgery after stenting 2. JACC Cardiovasc Interv. 2018 Mar;11(5):417-434.
- Allaqaband S, Jan MF, Banday WY, et al. Impact of 24-hr in-hospital interventional cardiology team on timeliness of reperfusion for ST-segment elevation myocardial infarction. Catheter Cardiovasc Interv. 2010 Jun; 75(7): 1015-1023.
- Nabel EG, Braunwald E. A tale of coronary artery disease and myocardial infarction. N Engl J Med. 2012 Jan; 366(1): 54-63. Erratum in: N Engl J Med. 2012 Mar; 366(10): 970.
- Hawn MT, Graham LA, Richman JS, et al. Risk of major adverse cardiac events following noncardiac surgery in patients with coronary stents. JAMA. 2013 Oct; 310(14): 1462-1472. Erratum in: JAMA. 2014 Feb; 311(5): 528.
- James SK, Stenestrand U, Lindbäck J, et al. Long-term safety and efficacy of drug-eluting versus bare-metal stents in Sweden. N Engl J Med. 2009 May; 360(19): 1933-1945.
- van Werkum JW, Heestermans AA, Zomer AC, et al. Predictors of coronary stent thrombosis: the Dutch Stent Thrombosis Registry. J Am Coll Cardiol. 2009 Apr; 53(16): 1399-1409.
- Levine GN, Bates ER, Bittl JA, et al. 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Thorac Cardiovasc Surg. 2016 Nov; 152(5): 1243-1275.
- Adriaenssens T, Joner M, Godschalk TC, et al. Optical coherence tomography findings in patients with coronary stent thrombosis: a report of the PRESTIGE consortium (Prevention of Late Stent Thrombosis by an Interdisciplinary Global European Effort). Circulation. 2017 Sep; 136(11): 1007-1021.
- Guagliumi G, Sirbu V, Musumeci G, et al. Examination of the in vivo mechanisms of late drug-eluting stent thrombosis: findings from optical coherence tomography and intravascular ultrasound imaging. JACC Cardiovasc Interv. 2012 Jan; 5(1): 12-20.
- Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of late drug-eluting stent thrombosis: strut coverage as a marker of endothelialization. Circulation. 2007 May; 115(18): 2435-2441.
- Souteyrand G, Amabile N, Mangin L, et al. Mechanisms of stent thrombosis analysed by optical coherence tomography: insights from the national PESTO French registry. Eur Heart J. 2016 Apr; 37(15): 1208-1216.
- Taniwaki M, Radu MD, Zaugg S, et al. Mechanisms of very late drug-eluting stent thrombosis assessed by optical coherence tomography. Circulation. 2016 Feb; 133(7): 650-660.
- Prati F, Kodama T, Romagnoli E, et al. Suboptimal stent deployment is associated with subacute stent thrombosis: optical coherence tomography insights from a multicenter matched study. From the CLI Foundation investigators: the CLI-THRO study. Am Heart J. 2015 Feb; 169(2): 249-256.