Neurovascular Outcomes in Relation With Carotid Artery Stenosis in Patients Undergoing Transcatheter Aortic Valve Implantation

Abstract

Introduction. Data regarding the prevalence of carotid artery stenosis (CAS) in patients undergoing transcatheter aortic valve implantation (TAVI) are scarce. Whether CAS, especially severe or bilateral, is a predictor of worse prognosis after TAVI is unknown. We aimed to address these questions. Methods. We included all patients who underwent TAVI between 2018 and 2021. Using pre-TAVI carotid Doppler ultrasound, atherosclerosis of the right and left carotid internal arteries was assessed. CAS was defined as moderate (50%-69% stenosis, peak systolic velocity of 125-230 cm/sec) or severe (≥70% stenosis, peak systolic velocity of >230 cm/sec). When both carotid arteries presented with ≥50% stenosis, CAS was defined as bilateral. Endpoints included the 30-day incidence of stroke or transient ischemic attack (TIA), 30-day all-cause mortality, and periprocedural complications. Results. Among 448 patients, 56 (12.5%) had CAS, of which 15 had bilateral and 15 had severe CAS. Patients with CAS were more often men and had higher rates of peripheral artery disease, coronary artery disease, and previous percutaneous coronary intervention. There was no association between CAS and 30-day stroke or TIA (adjusted hazard ratio [aHR], 2.55; 95% confidence interval [CI], 0.73-8.91; P=.14), even when considering severe CAS only. However, a significant association was found between bilateral CAS and 30-day stroke or TIA (aHR, 8.399; 95% CI, 1.603-44.000; P=.01). No association between CAS and 30-day mortality or periprocedural complications was found. Conclusions. CAS is common among TAVI patients. While CAS as a whole was not a predictor of neurovascular complications, the subgroup of bilateral CAS was associated with an increased risk of stroke.

J INVASIVE CARDIOL 2023;35(3):E136-E142. Epub 2023 January 26.

Key words: aortic valve stenosis, carotid artery stenosis, stroke, transcatheter aortic valve implantation, transcatheter aortic valve replacement

During the past 20 years, transcatheter aortic valve implantation (TAVI) has emerged as the first-line procedure for the treatment of symptomatic severe aortic stenosis in patients aged ≥75 or those younger and at high surgical risk.¹ Hence, the volume of TAVI interventions has dramatically increased, and this trend is expected to continue in the coming years.^2,3 Periprocedural ischemic stroke and transient ischemic attack (TIA) are dreaded complications, with an incidence of 2%-4%, and are associated with reduced survival and quality of life.^4,5 However, the pathophysiological mechanisms behind this increased risk are still poorly understood. Cerebral debris embolization is thought to play a central role, especially since up to 80% of patients who undergo TAVI have been found to have new silent bilateral ischemic embolic lesions.⁶ However, in clinical practice, despite the use of cerebral embolic protection devices, there are residual strokes; in the SENTINEL trial, which studied the use of the Sentinel embolic protection device (Claret Medical) in TAVI, strokes at 30 days still occurred in 5.6% of patients randomized to the device group,⁷ suggesting other mechanisms than cerebral debris embolization may be involved.

Carotid artery stenosis (CAS) has been shown to be associated with an increased risk of periprocedural cerebrovascular complications after coronary artery bypass graft surgery and surgical aortic valve replacement.^8,9 Whether this holds true for TAVI is currently unknown, in particular in the setting of bilateral CAS, even though a few previous studies have been published^10-13 and recent guidelines do not recommend to systematically screen for CAS before TAVI.¹ From a pathophysiological standpoint, CAS could in theory be a risk factor for ischemic neurovascular complications in patients undergoing TAVI, because the latter may induce transient hypotension and peripheral hypoperfusion via rapid ventricular pacing.¹⁴ Using prospectively collected data from our TAVI registry, we performed the present study to: (1) assess the rates of CAS (including moderate, severe, or bilateral), neurovascular complications, and mortality in patients undergoing TAVI; and (2) analyze whether there is an association between CAS and the 30-day risk of stroke or TIA, as well as with 30-day all-cause mortality risk.

Methods

Patient population. We included all patients who underwent TAVI from January 1, 2018 to December 31, 2021 at Lausanne University Hospital (Centre hospitalier universitaire vaudois). During that timeframe, a screening of CAS using carotid Doppler ultrasound was routinely performed in all candidates for TAVI. The diagnosis of severe aortic stenosis was based on the clinical, echocardiographic, and hemodynamic criteria of the European Society of Cardiology 2017 guidelines.¹⁵ For all cases, suitability for TAVI and the choice of vascular access (transfemoral as first choice, with transcarotid considered if transfemoral was not possible) were assessed by a heart team consisting of at least a senior interventional cardiologist, an echocardiographer, a cardiac surgeon, and an anesthesiologist.

Ethical statement. All patients belonged to the SWISS-TAVI Registry and provided written informed consent for the use of their data for research purposes. Our study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was given by the local ethics committee (Commission cantonale d’éthique de la recherche sur l’être humain, decision CER-VD 211/13, dated May 10, 2013).

Carotid artery stenosis. Carotid Doppler ultrasound was usually performed during the month preceding TAVI intervention. Atherosclerosis of both the right and left internal carotid arteries was assessed by an experienced angiologist. We defined CAS as any carotid lesions exceeding 50%, keeping in line with previous studies.^13,16 CAS was further defined as moderate (stenosis between 50% and 69% and peak systolic velocity [PSV] between 125 cm/sec and 230 cm/sec) or severe (≥70% stenosis and PSV >230 cm/sec).¹⁷ It was considered bilateral when both carotid arteries presented at least 50% stenosis; in those cases, severity was defined as the most important stenosis found either on the right or left carotid artery.

Endpoints. The primary endpoint was the incidence and timing of ischemic neurovascular complications (stroke or TIA) within the first 30 days after TAVI. Ischemic stroke was defined as a new neurological deficit associated with an acute ischemic lesion evidenced on neuroimaging. TIA was diagnosed when the patient presented a transient neurological deficit, with no lesion found on neuroimaging. All diagnoses of stroke or TIA were independently made by neurologists. Secondary endpoints included 30-day mortality and periprocedural complications such as the need for permanent pacemaker implantation, postoperative acute kidney injury, major vascular complications, and life-threatening bleeding. All endpoints were reported according to the Valve Academic Research Consortium (VARC)-2 criteria¹⁸ and not the more recent VARC-3 criteria,¹⁹ as data collection in our registry preceded the publication of the latter.

Statistical analyses. Categorical variables were expressed as frequencies and percentages and compared using Pearson’s Chi-squared test. Continuous variables were expressed as medians with 25th and 75th percentiles and were compared using the Mann-Whitney test. The incidence of neurovascular complications at 30 days and 30-day mortality were modeled using the Kaplan-Meier method and were analyzed using a log-rank test. The Cox proportional hazards model was used to identify baseline variables and periprocedural complications associated with the primary endpoint. Variables and complications considered clinically relevant or having P<.10 at the univariate level were included in the multivariate model. Results were expressed as hazard ratios (HRs) and 95% confidence intervals (CIs). All statistical analyses were performed using the SPSS 28.0 software (SPSS, Inc) and P<.05 was considered statistically significant.

Results

Baseline characteristics. Between January 1, 2018 and December 31, 2021, a total of 448 patients underwent TAVI at our institution, of which 56 (12.5%) had CAS. Among those, 41 (9.2%) had unilateral CAS and 15 (3.3%) had bilateral CAS. Likewise, 41 patients (9.2%) had moderate CAS while 15 patients (3.3%) had severe CAS. Patients’ baseline characteristics are presented in Table 1. Patients with CAS were more often men and had a higher prevalence of lower-extremity artery disease (LEAD), coronary artery disease, and previous percutaneous coronary intervention. Regarding technical aspects, the transcarotid vascular access was more frequently used in patients with CAS, and there was a trend toward more use of self-expandable transcatheter heart valves (THVs) in patients without CAS (P=.05). Supplemental Table S1 and Supplemental Table S2 show patient characteristics in the subgroups of CAS; no significant difference was found between patients with unilateral CAS and those with bilateral CAS, while those with severe CAS had significantly more LEAD (P=.04) compared with those with moderate CAS. One patient with bilateral CAS underwent prophylactic carotid revascularization (right carotid artery stenting) prior to TAVI and was considered as having only left CAS.

30-day and periprocedural outcomes. Rates of 30-day stroke or TIA, 30-day all-cause mortality, and periprocedural outcomes are presented in Table 2. Kaplan-Meier curves for 30-day stroke and mortality are presented in Figure 1. The overall incidence of ischemic neurovascular complications in our cohort was 2.9%. The incidence of stroke or TIA at 30 days was higher in patients with CAS, as compared with those without CAS (7.1% vs 2.3%; P=.04). In contrast, there was no significant difference with respect to all-cause 30-day mortality. Likewise, no significant difference was found regarding the other secondary endpoints (namely, the perioperative complications).

Adjusted primary endpoint. The univariate analyses of predictors of stroke or TIA at 30 days are shown in Table 3. Variables with a P-value <.10 were CAS (HR, 3.15; 95% CI, 0.97-10.23; P=.056), LEAD (HR, 2.98; 95% CI, 0.92-9.67; P=.07), postoperative acute kidney injury (HR, 7.65; 95% CI, 1.69-34.48; P<.01), and major vascular complications (HR, 6.68; 95% CI, 1.83-24.22; P<.01). No significant association was found between the other variables and the risk of neurovascular complications.

After adjustment for LEAD, postoperative acute kidney injury, and major vascular complications, CAS was no longer independently associated with the primary outcome (adjusted HR, 2.55; 95% CI, 0.73-8.91; P=.14). However, a significant association was found with bilateral CAS and the primary outcome (adjusted HR, 8.40; 95% CI, 1.60-44.00; P=.01) (Table 4). An independent subgroup analysis did not find severe CAS to be a predictor of stroke or TIA at 30 days (adjusted HR, 2.30; 95% CI, 0.25-21.39; P=.46).

Causality between CAS and stroke lesions. The medical records of the 4 patients with CAS who had stroke or TIA after TAVI were reviewed. The side of CAS and the locations of the ischemic neurovascular lesions are described in Supplemental Table S3. The stroke lesion was attributable to CAS in only 1 case (in that case, severe bilateral CAS); in the 3 other cases, an embolic cause was considered the most likely, due to the number and the distribution of the lesions.

Discussion

Our results can be summarized as follows: (1) 12.5% of patients undergoing TAVI had CAS and the majority of CAS was unilateral; (2) patients with CAS were found to have a higher prevalence of LEAD, coronary artery disease, and previous percutaneous coronary intervention; (3) after adjustment for the predictors of stroke or TIA at 30 days on univariate analyses (LEAD, postoperative acute kidney injury, and major vascular complications), bilateral CAS (but not CAS in general or severe CAS) was associated with a higher risk of stroke or TIA at 30 days; and (4) CAS was not associated with all-cause 30-day mortality and perioperative complications.

The prevalence of CAS in patients undergoing TAVI in our cohort (12.5%) lies in the lower range of the prevalence values reported in literature, which vary from 6.9% to 33.1%.^10-13 This important variation in the reported prevalence of CAS may be partly due to the way data were collected (for example, data based on administrative claims or discharge letters may be less reliable); differences in ethnicity may also play a role, as the prevalence of carotid disease has been shown to vary significantly by ethnicity.²⁰ In contrast, the incidence of stroke or TIA at 30 days in our cohort (2.9%) lines up with previous data; in the PARTNER trial (n = 2621), 3.8% of patients experienced a stroke or TIA within 30 days after TAVI,²¹ whereas the international CENTER Collaboration Registry (n = 10,982) found a 30-day stroke rate of 2.4%.⁵

A few previous studies have explored the subject of the prognostic impact of CAS on mortality and neurovascular outcomes after TAVI; they suggested that CAS (defined in most studies as carotid stenosis >50%) is not a predictor of post-TAVI mortality or stroke.^10-13 However, data for the specific setting of bilateral CAS are conflicting or lacking. In a United States (US) registry-based analysis, bilateral CAS appeared as a predictor of perioperative ischemic strokes following TAVI (adjusted odds ratio, 4.46; 95% CI, 2.03-9.82; P<.001),²² but that was not confirmed in another US observational multisite registry; no difference was found in the 1-year composite endpoint of stroke/mortality between patients with bilateral severe CAS (>80% stenosis) when compared with non-CAS patients.¹³ Although these studies included significant numbers of patients, they also had some important limitations, such as the lack of a single approach and protocol to analyze the carotid arteries, the lack of rigorous physician adjudication of strokes (the diagnoses were based on administrative data), and the possibility that some unmeasured confounding factors were not taken into account during comparisons. Our study adds up new up-to-date data to the previous ones; it confirms the absence of association between CAS in general (including severe CAS) and the risk of neurovascular complications, except for bilateral CAS, which may indeed result in a higher risk of stroke.

To further advance in the reasoning, a thorough understanding of the pathophysiological mechanisms of ischemic neurovascular complications in TAVI is important. One explanation is the dislodgment and embolization of crushed calcified native valve and aortic debris by the use of catheters and delivery systems, balloon valvuloplasty, and implantation of the prosthetic valve.⁵ This hypothesis is supported by histopathologic analyses of the types of debris frequently found in cerebral protection filters used during TAVI, which mainly originated from the native wall or the arterial wall, or were calcifications.^5,23,24 However, in a US national observational study including 123,186 patients, the use of cerebral embolic protection devices was only associated with a modest decrease in the risk of in-hospital stroke, with an adjusted odds ratio of 0.82 (95% CI, 0.69-0.97).²⁵ In our cohort, there was no association found between the use of cerebral embolic protection devices (Sentinel) and the incidence of neurovascular complications. All this suggests other pathophysiological mechanisms may be at stake. One possible explanation is the transient hypotension and peripheral hypoperfusion induced by rapid ventricular pacing. Briefly, the latter is necessary to implant balloon-expandable THVs to reduce cardiac output and achieve cardiac standstill, thus allowing optimal positioning of the THVs.¹⁴ Rapid ventricular pacing may also be used with self-expandable THVs when pre- or postdilation of the aortic valve is needed. Patients with bilateral CAS at baseline may be more prone to having neurovascular complications following rapid ventricular pacing. Furthermore, a systemic proinflammatory state during the periprocedural period may lead to an increased risk of carotid plaque rupture.¹³ The same mechanisms have been used to explain, among others, the pathogenesis of acute mesenteric ischemia and acute kidney injury after TAVI.^14,26 Our analysis of the mechanisms behind the 4 cases of stroke in patients with CAS suggests that while unilateral CAS is possibly merely a marker of a high comorbidity (in particular vascular disease) burden, with no cerebral lesions directly attributable to it, severe bilateral CAS may indeed be a direct cause of stroke.

Finally, vascular access for TAVI (transfemoral or transcarotid) was not a predictor of neurovascular complications. This is interesting as there has been some debate as to whether the transcarotid access could result in more strokes, because of the direct manipulation of the carotid artery and transient reduction in blood flow during surgery.^27,28

Our results have practical implications; they suggest that screening of CAS before TAVI may be beneficial. In case of bilateral severe CAS, preoperative carotid revascularization may be considered, although the authors acknowledge the lack of data regarding the benefits of such a strategy.

Study limitations. Our study is subject to several limitations. First, it is based on observational data and there is a possibility of unmeasured confounding factors not taken into account during comparisons. Second, the number of patients with CAS, and in particular bilateral CAS, was relatively small, making it difficult to make any definitive conclusion regarding this subgroup of patients. Finally, the results presented here are those of a single Swiss tertiary center and may not be transposable to other populations and hospitals.

Affiliations and Disclosures

From the ¹Service of Cardiology, ²Service of Internal Medicine, ³Service of Cardiovascular Surgery, ⁴Service of Angiology, and ⁵Adult Intensive Care Unit, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Fournier reports consulting fees from Bayer and Cathworks. Dr Eeckhout reports grants from Edwards Lifesciences. Dr Muller reports grants from Abbott and Edwards Lifesciences. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript accepted November 8, 2022.

Address for Correspondence: Henri Lu, MD, Service of Cardiology, Lausanne University Hospital (CHUV), Rue du Bugnon 46, 1010 Lausanne, CH. Email: henri.lu@chuv.ch

Conclusion

The prevalence of CAS is high among patients undergoing TAVI. Our data show that while CAS as a whole was not a predictor of neurovascular complications or mortality at 30 days (being in line with results from previous studies), the subgroup of bilateral CAS may be associated with an increased risk of stroke at 30 days. Further prospective research is warranted regarding the utility of systematic screening of CAS in patients undergoing TAVI and the prognostic significance of bilateral CAS on neurovascular complications.

References

1. Vahanian A, Beyersdorf F, Praz F, et al. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J. 2022;43(7):561-632. doi:10.1093/eurheartj/ehab395

2. Mauri V, Abdel-Wahab M, Bleiziffer S, et al. Temporal trends of TAVI treatment characteristics in high volume centers in Germany 2013–2020. Clin Res Cardiol. 2022;111(8):881-888. doi:10.1007/s00392-021-01963-3

3. Lu H, Muller O, Eeckhout E, et al. TAVI: une revue de la littérature des voies alternatives à l’accès trans-fémoral. Presse Médicale Form. 2020;1:249‑56. doi:10.1016/j.lpmfor.2020.04.016

4. Armijo G, Nombela-Franco L, Tirado-Conte G. Cerebrovascular events after transcatheter aortic valve implantation. Front Cardiovasc Med. 2018;5:104. eCollection 2018. doi:10.3389/fcvm

5. Vlastra W, Jimenez-Quevedo P, Tchétché D, et al. Predictors, incidence, and outcomes of patients undergoing transfemoral transcatheter aortic valve implantation complicated by stroke. Circ Cardiovasc Interv. 2019;12(3):e007546. doi:10.1161/CIRCINTERVENTIONS.118.007546

6. Teitelbaum M, Kotronias RA, Sposato LA, Bagur R. Cerebral embolic protection in TAVI: friend or foe. Interv Cardiol. 2019;14(1):22-25. doi:10.15420/icr.2018.32.2

7. Kapadia SR, Kodali S, Makkar R, et al. Protection against cerebral embolism during transcatheter aortic valve replacement. J Am Coll Cardiol. 2017;69(4):367-377. doi:10.1016/j.jacc.2016.10.023

8. Brott TG, Halperin JL, Abbara S, et al. 2011 ASA/ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/SAIP/SCAI/SIR/SNIS/SVM/SVS Guideline on the management of patients with extracranial carotid and vertebral artery disease: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American Stroke Association, American Association of Neuroscience Nurses, American Association of Neurological Surgeons, American College of Radiology, American Society of Neuroradiology, Congress of Neurological Surgeons, Society of Atherosclerosis Imaging and Prevention, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of NeuroInterventional Surgery, Society for Vascular Medicine, and Society for Vascular Surgery. Developed in collaboration with the American Academy of Neurology and Society of Cardiovascular Computed Tomography. Catheter Cardiovasc Interv. 2013;81(1):E76-123. doi:10.1002/ccd.22983

9. Likosky DS, Marrin CA, Caplan LR, et al. Determination of etiologic mechanisms of strokes secondary to coronary artery bypass graft surgery. Stroke. 2003;34(12):2830-2834. doi:10.1161/01.STR.0000098650.12386.B3

10. Chakraborty S, Faisaluddin M, Ashish K, et al. In-hospital clinical outcomes of transcatheter aortic valve replacement in patients with concomitant carotid artery stenosis: insights from the national inpatient sample. Int J Cardiol Heart Vasc. 2020;31:100621. doi:10.1016/j.ijcha.2020.100621

11. Ben-Shoshan J, Zahler D, Steinvil A, et al. Extracranial carotid artery stenosis and outcomes of patients undergoing transcatheter aortic valve replacement. Int J Cardiol. 2017;227:278-283. Epub 2016 Nov 9. doi:10.1016/j.ijcard.2016.11.107

12. Lepidi S, Squizzato F, Fovino LN, et al. Prevalence and prognostic impact of carotid artery disease in patients undergoing transcatheter aortic valve implantation. Ann Vasc Surg. 2022;S0890509622001479. doi:10.1016/j.avsg.2022.03.018

13. Kochar A, Li Z, Harrison JK, et al. Stroke and cardiovascular outcomes in patients with carotid disease undergoing transcatheter aortic valve replacement. Circ Cardiovasc Interv. 2018;11(6):e006322. doi:10.1161/CIRCINTERVENTIONS

14. Lu H, Rotzinger D, Monney P, et al. Prevalence and prognostic value of mesenteric artery stenosis in patients undergoing transcatheter aortic valve implantation. Front Cardiovasc Med. 2022;9:750634. doi:10.3389/fcvm.2022.750634

15. Baumgartner H, Falk V, Bax JJ, et al. 2017 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J. 2017;38(36):2739-2791. doi:10.1093/eurheartj/ehx391

16. Steinvil A, Leshem-Rubinow E, Abramowitz Y, et al. Prevalence and predictors of carotid artery stenosis in patients with severe aortic stenosis undergoing transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2014;84(6):1007-1012. doi:10.1002/ccd.25585

17. Grant EG, Benson CB, Moneta GL, et al. Carotid artery stenosis: gray-scale and Doppler US diagnosis—Society of Radiologists in Ultrasound Consensus Conference. Radiology. 2003;229(2):340-346. doi:10.1148/radiol.2292030516

18. Kappetein AP, Head SJ, Généreux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. Eur Heart J. 2012;33(19):2403-2418. doi:10.1093/eurheartj/ehs255

19. VARC-3 writing committee, Généreux P, Piazza N, et al. J Am Coll Cardiol. 2021;77(21):2717-2746. doi:10.1016/j.jacc.2021.02.038

20. Rockman CB, Hoang H, Guo Y, et al. The prevalence of carotid artery stenosis varies significantly by race. J Vasc Surg. 2013;57(2):327-337. doi:10.1016/j.jvs.2012.08.118

21. Kapadia S, Agarwal S, Miller DC, et al. Insights into timing, risk factors, and outcomes of stroke and transient ischemic attack after transcatheter aortic valve replacement in the PARTNER trial (Placement of Aortic Transcatheter Valves). Circ Cardiovasc Interv. 2016;9(9):e002981. doi:10.1161/CIRCINTERVENTIONS

22. Thirumala PD, Muluk S, Udesh R, et al. Carotid artery disease and periprocedural stroke risk after transcatheter aortic valve implantation. Ann Card Anaesth. 2017;20(2):145-151. doi:10.4103/aca.ACA_13_17

23. Schmidt T, Schlüter M, Alessandrini H, et al. Histology of debris captured by a cerebral protection system during transcatheter valve-in-valve implantation. Heart. 2016;102(19):1573-1580. doi:10.1136/heartjnl-2016-309597

24. Van Mieghem NM, Schipper ME, Ladich E, et al. Histopathology of embolic debris captured during transcatheter aortic valve replacement. Circulation. 2013;127(22):2194-2201. doi:10.1161/CIRCULATIONAHA.112.001091

25. Butala NM, Makkar R, Secemsky EA, et al. Cerebral embolic protection and outcomes of transcatheter aortic valve replacement: results from the Transcatheter Valve Therapy Registry. Circulation. 2021;143(23):2229-2240. doi:10.1161/CIRCULATIONAHA.120.052874

26. Cheungpasitporn W, Thongprayoon C, Kashani K. Transcatheter aortic valve replacement: a kidney’s perspective. J Renal Inj Prev. 2016;5(1):1-7. doi:10.15171/jrip.2016.01

27. Lu H, Monney P, Fournier S, et al. Transcervical approach versus transfemoral approach for transcatheter aortic valve replacement. Int J Cardiol. 2021;327:58-62. Epub 2020 Nov 24. doi:10.1016/j.ijcard.2020.11.026

28. Lu H, Monney P, Hullin R, et al. Transcarotid access versus transfemoral access for transcatheter aortic valve replacement: a systematic review and meta-analysis. Front Cardiovasc Med. 2021;8:687168. doi:10.3389/fcvm.2021.687168