Assessing the efficacy of VerifyNow platelet-function testing in predicting postoperative thromboembolic complications of neuroendovascular surgery: A systematic review and meta-analysis (part 1)

Abstract

Background

Despite the heavily debated use of routine platelet-function testing, the VerifyNow Platelet Reactivity Unit (PRU) assay has been increasingly adopted as standard of care for assessing risk of postoperative thromboembolic complications of neuroendovascular surgery.

Objective

We conducted a systematic review and meta-analysis to examine the relationship between platelet response and risk of ischemic events from neuroendovascular surgery, assess the efficacy of point-of-care platelet-function testing in predicting thromboembolic outcomes, and assess whether a clinically useful threshold for platelet response can be defined in order to standardize guidelines.

Methods

PubMed, Embase, and Scopus were searched. Following deduplication, articles were first screened for relevance by title and abstract, followed by full text.

Results

Of 735 resultant articles, 22 studies consisting of 3266 patients undergoing neuroendovascular intervention were included. Diagnoses included both intracranial and extracranial pathologies, of which 45.8% were treated with flow diversion, 16.4% with stent-assisted coil embolization, 15.8% with intracranial stenting, 12.0% with simple coil embolization, 3.4% with balloon-assisted coil embolization, 3.6% with extracranial stenting, and 3.0% with an alternate method. 54.5% (12/22) of studies determined platelet hyporesponse to be an independent predictor of postoperative thromboembolic complications, with 27.3% (6/22) of studies reporting a similar, but non-statistically significant trend. 18.2% (4/22) of studies found no relationship between platelet response and postoperative thromboembolic complications. The estimated clinical threshold for PRU to prevent thromboembolic complications varied greatly across studies (Range: > 144–295 PRU). Meta-analysis found platelet hyporesponse to have a 2.23-fold increased risk of thromboembolic complications compared to normoresponders (RR = 2.23, P = 0.03).

Conclusion

While PRU demonstrates a significant predictive value for postoperative thromboembolic complications of neuroendovascular surgery, the target therapeutic threshold for minimizing ischemic events remains unclear. Further studies, such as large multicenter cohorts of the existing data, are needed to standardize guidelines.

Introduction

Neuroendovascular surgery represents a burgeoning field of clinical research and innovation, offering less invasive alternatives to traditional, open surgical techniques.^1,2 While these minimally invasive procedures have revolutionized the treatment of intracranial and extracranial pathologies alike, thromboembolic events, including ischemic strokes and acute vessel occlusions, represent major concerns in the perioperative period. Suboptimal response to antiplatelet therapy or insufficient platelet inhibition are critical factors that may predispose patients to thromboembolic events.^3,4

Given the significant variability in Clopidogrel resistance across the population, identifying patients at high risk of thromboembolic events through effective platelet-function testing can aid in tailoring antiplatelet therapy and optimizing patient outcomes. ⁵ Clopidogrel resistance can be due to a variety of pharmacokinetic and pharmacodynamic factors, including poor intestinal absorption due to genetic variation in ABCB1, variability in hepatic CYP P450 enzymes needed to convert Clopidogrel to its active form, increased BMI or diabetes causing increased sensitivity to ADP, and genetic polymorphisms of the GPIIb/IIIa receptor or platelet membrane receptors, to name a few. ⁶

The VerifyNow platelet-function test (PFT) is a point-of-care assay that measures platelet reactivity units (PRU) in response to antiplatelet agents, thus allowing clinicians to assess the efficacy of antiplatelet therapy and gauge individual patient response. Point-of-care platelet-function testing offers advantages such as rapid results, individualized treatment, risk stratification, and ongoing monitoring of antiplatelet therapy. ⁷ However, challenges related to standardization, limited evidence, interpretation complexity, resource requirements, and uncertain impact on clinical outcomes renders the utility of routine testing in this setting unclear.^7,8

While the role of point-of-care platelet-function-testing in predicting postoperative complications of cardiovascular surgery has been thoroughly evaluated in the existing literature,^3,9 no such critical evaluation within the neuroendovascular field has been conducted. A recent review and meta-analysis ¹⁰ evaluated the role of PFT on hemorrhagic and thrombotic complications, but the study population was limited to patients with cerebral aneurysms treated with pipeline embolization devices (PEDs). We performed a systematic review and meta-analysis to critically evaluate the available evidence on the efficacy of VerifyNow platelet-function testing in predicting postoperative thromboembolic complications following neuroendovascular surgery. These findings may inform clinical decision-making, refine antiplatelet therapy strategies, and contribute to improved patient outcomes via appropriate risk stratification.

Methods

Search strategy

A systematic review and meta-analysis was conducted using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) ¹¹ guidelines to assess the efficacy of the VerifyNow platelet-function assay in predicting postoperative thromboembolic complications of neuroendovascular surgery. Three databases were searched in October 2022: PubMed MEDLINE (National Library of Medicine), Embase (Elsevier), and Scopus (Elsevier). Search strategies for Embase and Scopus were adapted from the PubMed search strategy shown in Supplementary Table 1. Date, study type, and language limits were not applied. Searches were conducted using controlled vocabulary (MeSH), keywords, and keyword synonyms for ‘P2Y12 assay’, ‘VerifyNow’, ‘neuroendovascular’, ‘neurointervention’, ‘carotid artery’, ‘vertebral artery’, ‘stenosis’, ‘stent’, ‘hemorrhagic’, ‘thromboembolic’, and ‘ischemic’.

Screening

Duplicates were removed after search completion using the automated deduplication feature in Endnote ×9 (Clarivate, London, United Kingdom). The remaining eligible studies were screened by title and abstract for relevance. Those selected for full text review were screened using predetermined inclusion and exclusion criteria. Two authors, DLM and LSM, completed both stages of screening. Conflicts were reconciled by discussion to achieve consensus.

Inclusion and exclusion criteria

Inclusion criteria were publications in or translated into English, with full text and abstract available, studying patients undergoing neuroendovascular surgery, and assessing both platelet sensitivity and postoperative thromboembolic complications. Studies including data on patients who did not undergo neuroendovascular intervention were included if definite outcomes for patients who did undergo neuroendovascular intervention could be identified. Exclusion criteria were as follows: abstracts only, secondary literature, not assessing platelet response in terms of PRU or percent inhibition, and not reporting incidence of thromboembolic outcomes.

Outcome variables

The primary outcome measure was the incidence of postoperative thromboembolic complications, including both symptomatic (stroke, acute vessel occlusion) and asymptomatic (ischemic lesions on imaging) events. No secondary outcomes were evaluated.

Data extraction

Articles selected for final inclusion were assessed for bibliographic data, study design, participants, intervention where applicable, and outcome data.

Quality assessment

The income status for countries of study origin was determined using the World Bank designation. ¹² Quality and risk of bias were assessed using the GRADE framework ¹³ and Cochrane ROBINS-I (Risk of Bias in Non-Randomized Studies of Interventions) ¹⁴ tool for each of the included studies. Overall risk of bias for this systematic review was inferred based on risk of bias of each included study.

Statistical analysis

Descriptive statistics were completed to summarize the findings of the studies included in the systematic review. For the meta-analysis, the proportion of low responders and standard error for each proportion was calculated by study. Studies without the reporting data necessary to compute incidence were excluded. A pooled incidence proportion was calculated using a random-effects model with R (R Foundation for Statistical Computing, Vienna, Austria, Version 4.3.1). Relative risks (RR) were computed for each study by comparing the risk of thromboembolic complications between hyporesponders and normoresponders. Studies not containing data compatible with this analysis were excluded. A random-effects model, specifically the inverse variance method, was used to account for both within-study and between-study variations. This model provided a pooled RR, representing the combined effect size across all studies. Egger's regression was calculated to evaluate for potential publication bias.

Results

Search results

The search identified 735 articles, of which 22 were included (Figure 1, Table 1).^15–36 Study designs included randomized controlled trials (1, 4.5%), prospective cohort (6, 27.3%), retrospective cohort (13, 59.1%), and both prospective and retrospective cohort (2, 9.1%). The quality of most studies was high (12, 54.5%). Most studies had a low risk of bias (12, 54.5%), resulting in this review having a low risk of bias as well. All but one of the included studies originated from high-income countries, including the United States (12, 54.5%), Japan (5, 22.7%), Korea (2, 9.1%), Germany (1, 4.5%), and Spain (1, 4.5%). Only one study originated from a middle-income country.

Figure 1.

PRISMA flowchart outlining the search and review process used to identify and select articles for inclusion in the systematic review.

Table 1.

Studies included in the review.

Reference	Country	Quality	Bias	Study Design	Patients, N	Mean Age, yrs	Mean PRU (Range)	LR Threshold (≥)	LRs, N (%)	Diagnosis	Intervention (N)	Total TECs, N (%)	LR TECs, N (%)	Key Findings
Asai et al., 2015 15	Japan	High	Low	Retrospective cohort	189	59.9	189.9 (5–370)	230	66 (34.9)	ICA	Simple coiling (129), SACE (60)	8 (4.2)	2 (3.0)	Low response to clopidogrel, current smoker, posterior location, and longer neck size were independent risk factors for postop new large ischemic lesions on MRI-DWI.
Batur et al., 2021 16	Turkey	Moderate	Moderate	Retrospective cohort	295	51.1	68 (0–240)	240	130 (44.1)	ICA	SACE, flow diverter (not specified)	20 (6.8)	4 (3.1)	The TEC rate was significantly lower in the hyporesponder group compared to non-hyporesponders, though ROC analysis could not identify the risk threshold for TECs.
Bender et al., 2017 17	United States	Moderate	Moderate	Retrospective cohort	52	57	Not defined	200	52 (100.0)	ICA	PED (52)	3 (5.8)	3 (5.8)	P2Y12 hyporesponse was not associated with increased periprocedural complications.
Corliss et al., 2021 18	United States	High	Low	Retrospective + prospective cohorts	165	61.3	142	Not defined	N/A	ICA, CAS	Extracranial stenting (41), intracranial stenting (1), SACE (32), PED (100)	16 (9.7)	N/A	PRU had a significant dose-dependent effect on the rates of thrombosis, with ROC analysis yielding a therapeutic threshold value of 144 PRU to avoid TECs.
Delgado Almandoz et al., 2013 19	United States	Moderate	Moderate	Retrospective cohort	44	59.2	130.2 (0–292)	240	2 (4.5)	ICA	PED (48)	6 (13.6)	2 (100.0)	A last-recorded PRU value of >240 was the only independent predictor of all TECs up to 6 months post-procedure.
Delgado Almandoz et al., 2013 20	United States	Moderate	Moderate	Retrospective cohort	44	59.2	111.7 (5–275)	240	3 (6.8)	ICA	PED (48)	4 (9.1)	2 (66.7)	Pre-procedure PRU >240 and a technically difficult procedure were independent predictors of all perioperative TECs.
Delgado Almandoz et al., 2013 21	United States	High	Low	Retrospective cohort	100	57.3	136.5	240	15 (15.0)	ICA	Simple/balloon coiling (40), SACE (18), PED (43)	5 (5.0)	3 (20.0)	Clopidogrel hyporesponse was associated with a significantly increased risk of TECs.
Drazin et al., 2011 22	United States	High	Low	Retrospective cohort	52	62.6	279.5, 30.7%	20%	19 (36.5)	ICA, ICS	Intracranial stenting (52)	1 (1.9)	1 (5.3)	Although 37% of patients had a suboptimal clopidogrel response, only one patient had a TEC in this group. No patients with > 20% inhibition had such events.
Fifi et al., 2013 23	United States	High	Low	Prospective cohort	96	64	205	240	36 (37.5)	CAS, VAS, CAD, ICA, ICS	Intracranial stenting (95), extracranial stenting (1)	7 (7.3)	6 (16.7)	Clopidogrel resistance, higher diastolic BP, and lack of statin use were significantly associated with periprocedural TECs.
Flechtenmacher et al., 2015 24	Germany	Moderate	Moderate	Prospective cohort	97	67.2	Not defined	236	52 (53.6)	ECS, ICS, ICA	Intracranial stenting (16), extracranial stenting (77), SACE (10)	9 (9.3)	6 (11.5)	Periprocedural TECs occurred more frequently in patients determined by all three methods (LTA, VerifyNow, Multiplate) to be clopidogrel resistant. Sensitivity and specificity rates of clopidogrel resistance in relation to embolic complications were 67% and 51% for VerifyNow.
Fujita et al., 2022 25	Japan	High	Low	Retrospective cohort	197	68	148 (82–203)	Not defined	N/A	ICA, ICS/ECS, other	Simple coiling (70), flow diverter (36), angioplasty (85), other (6)	8 (4.1)	N/A	Higher PRU was significantly associated with symptomatic TECs. ROC analysis showed that PRU greater than or equal to 212 was the optimal threshold to identify patients with TECs.
Gonzalez et al., 2014 26	Spain	High	Low	Randomized- controlled trial	214	67.1	216	230	99 (46.3)	CAS	Carotid artery stenting (214)	8 (3.7)	4 (4.0)	The incidence of TIA, stroke or death up to 30 days after CAS in the hyporesponder patients was slightly but not statistically significantly higher than in clopidogrel responders.
Higashiguchi et al., 2021 27	Japan	High	Low	Retrospective + prospective cohorts	217	60.9	159.5 (CPG group), 270.7 (PSG group)	240	27 (12.4)	ICA	Simple coiling (66), SACE (124), flow-diverter (24)	19 (8.8)	1 (3.7)	ROC curve analysis assessing TECs confirmed a PRU value of 175.5. Tailored antiplatelet medications using PRU as an index reduced the TECs compared with standard DAPT using aspirin and clopidogrel in a non-tailored cohort, without increasing hemorrhagic complications.
Hosoo et al., 2022 28	Japan	High	Low	Retrospective cohort	162	61.2	198.7	240	47 (29.0)	ICA	PED (162)	10 (6.2)	1 (2.1)	Significant risk factors independently associated with periprocedural TECs were PRU values of 190 or higher and neck sizes of 8 mm or larger.
Jiang et al., 2018 29	United States	High	Low	Retrospective cohort	742	56.3	148.6	200	N/A	ICA	PED (742)	37 (5.0)	N/A	Acute in situ thrombosis occurred in 37 cases despite no significant difference in postprocedural day 0 or day 1 PRU values.
Kang et al., 2010 30	Korea	High	Low	Prospective cohort	186	58.3	276.9	332	47 (25.3)	ICA	Simple, balloon, or SACE (186)	14 (7.5)	8 (17.0)	ROC analysis showed PRU levels significantly discriminate between patients with and without procedure-related TECs, with a PRU of 295 identified as the optimal cutoff point to predict the events.
Kashiwazaki et al., 2014 31	Japan	High	Low	Prospective cohort	66	63.1	47%	20%	19 (28.8)	CAS, ICA	Carotid stenting (35), simple coiling (13), SACE (18)	13 (19.7)	9 (47.4)	Low percent inhibition on clopidogrel was an independent predictor of 30-day TECs. Optimal cutoff for P2Y12 percent inhibition rate to prevent complications is 26%.
Kass-Hout et al., 2014 32	United States	Moderate	Moderate	Retrospective cohort	37	55.3	Not defined	237	11 (29.7)	ICA	Intracranial stenting (37)	3 (8.1)	2 (18.2)	Even though the combined complication rate was similar between the two groups, the hyporesponder group had a trend towards a higher chance of postoperative stroke (18% vs 0%).
Kayan et al., 2016 33	United States	Moderate	Moderate	Prospective cohort	103	57	118 (0–315)	240	3 (2.9)	ICA	Simple or balloon-assisted coiling (56), SACE (21), flow diverter (26)	7 (6.8)	0 (0.0)	There was no association between out-of-range PRU and TECs.
Nordeen et al., 2013 34	United States	Moderate	Moderate	Retrospective cohort	81	65	Not defined	20%	17 (21.0)	ICS, ICA, stroke, TIA	Angioplasty (32), SACE (29), CEA (1) intracranial stenting (12), simple coiling (2), DC (5)	5 (6.2)	0 (0.0)	There was an increased rate of death when a complication occurred in the resistant group (<20% inhibition) by 30 days postop and 90 days postop.
Ryu et al., 2010 35	Korea	Low	High	Prospective cohort	53	60.3	39%	40%	N/A	ICS, CAS, CAD, ICA	Simple coiling (11), SACE (17), intracranial stenting (25)	5 (9.4)	N/A	Ten patients demonstrated resistance to both drugs, 5 of which suffered TECs after neurointervention (mean platelet inhibition < 23%).
Tan et al., 2015 36	United States	Moderate	Moderate	Retrospective cohort	74	60	210 (2–450)	230	N/A	ICA	PED (74)	5 (6.8)	N/A	A preprocedural PRU > 208 had an OR of 11.32 for symptomatic TECs, but the result was not statistically significant.

Legend: N refers to number of study participants. Low response threshold is defined in terms of PRU or percentage platelet inhibition. Abbreviations: PRU, platelet reactivity units; OR, odds ratio; postop, postoperatively; TEC, thromboembolic complication; ICA, intracranial aneurysm; CAS, carotid artery stenosis; ICS, intracranial stenosis; VAS, vertebral artery stenosis; CAD, carotid artery dissection; ECS, extracranial stenosis; TIA, transient ischemic attack; PED, pipeline embolization device; CPG, clopidogrel; PSG, prasugrel; SACE, stent-assisted coil embolization; CEA, carotid endarterectomy; DC, decompressive craniectomy; LR, low responder.

Patient demographics

The 22 included studies consisted of 3266 participants, of which 2151 were female (Table 1).^15–36 The mean age of study participants at time of assessment ranged from 51.1 to 68.0 years, with a weighted mean of 59.6 years. Diagnoses included both intracranial and extracranial pathologies, of which 45.8% were treated with flow diversion, 16.4% with stent-assisted coil embolization, 15.8% with intracranial stenting, 12.0% with simple coil embolization, 3.4% with balloon-assisted coil embolization, 3.6% with extracranial stenting, and 3.0% with an alternate method.

All studies utilized an antiplatelet regimen consisting of dual-antiplatelet therapy (DAPT) with Aspirin and either Clopidogrel or Prasugrel. Platelet sensitivity was measured using the VerifyNow platelet reactivity assay and mean values were reported in terms of PRU (Range: 68.0–279.5 PRU) or percentage platelet inhibition (Range: 30.7–47.0%). Four studies did not report a mean platelet response for the study population. Eight studies defined a clinical threshold for high platelet response, of which 62.5% utilized ≤ 60 PRU to classify hyperresponders. Twenty studies defined a clinical threshold for low platelet response, of which 59.1% utilized ≥230–240 PRU to classify hyporesponders.

Impact of platelet function testing on perioperative management

As demonstrated in Supplementary Table 2, perioperative management based on the results of platelet function testing was widely varied. Six studies^{17,25,29–31,35} did not perform any preoperative intervention or alter the DAPT regimen for patients determined to be hyporesponders. Fourteen studies^{15,16,18–24,26–28,32–34,36} had a protocol in place to adjust the DAPT regimen for hyporesponders, but these varied in the PRU cutoff used to define resistance, the alternative medications patients were switched to, and the dosing. One study ¹⁸ stated that DAPT adjustments were left up to the surgeon's discretion, while another ¹⁵ stated that the regimen was altered depending on risk of ischemic events or allergies, but did not elaborate as to how they defined risk.

Predictive value of platelet sensitivity for thromboembolic complications

In total, 214 postoperative thromboembolic complications were reported across the included studies. Results are summarized in Table 2 and Figure 2. 54.5% (12/22) of studies determined platelet hyporesponse to be an independent predictor of postoperative thromboembolic complications, with 27.3% (6/22) of studies reporting a similar, but non-statistically significant trend. 18.2% (4/22) of studies found no relationship between platelet response and postoperative thromboembolic complications. The estimated clinical threshold for PRU to prevent thromboembolic complications varied greatly across studies (Range: >144–295 PRU).

Table 2.

Study findings regarding the efficacy of verifyNow testing in predicting postoperative TECs of neuroendovascular intervention with calculated threshold values for prevention.

Reference	Predictive Value			Threshold (PRU or % Inhibition)
	Statistically Significant	Non-Significant Association	No Association
Asai et al., 2015 15	X			Not defined
Batur et al., 2021 16	X			Not defined
Bender et al., 2017 17			X	N/A
Corliss et al., 2021 18	X			>144 PRU
Delgado Almandoz et al., 2013 19	X			>240 PRU
Delgado Almandoz et al., 2013 20	X			>240 PRU
Delgado Almandoz et al., 2013 21	X			>240 PRU
Drazin et al., 2011 22			X	N/A
Fifi et al., 2013 23	X			Not defined
Flechtenmacher et al., 2015 24		X		Not defined
Fujita et al., 2022 25	X			>212 PRU
Gonzalez et al., 2014 26		X		Not defined
Higashiguchi et al., 2021 27	X			>175.5 PRU
Hosoo et al., 2022 28	X			>190 PRU
Jiang et al., 2018 29			X	N/A
Kang et al., 2010 30	X			>295 PRU
Kashiwazaki et al., 2014 31	X			<26%
Kass-Hout et al., 2014 32		X		Not defined
Kayan et al., 2016 33			X	N/A
Nordeen et al., 2013 34		X		Not defined
Ryu et al., 2010 35		X		<23%
Tan et al., 2015 36		X		>208 PRU
TOTAL (N)	12	6	4

Table Legend: N refers to the total number of studies with the relevant finding. Abbreviations: PRU, platelet reactivity units; TEC, thromboembolic complications.

Figure 2.

Percentage of hyporesponders and postoperative thromboembolic complications in each included study's patient population. Abbreviations: TECs, thromboembolic complications.

Formal meta-analysis was performed to evaluate the relative risk of thromboembolic complications for hyporesponders compared to normoresponders. Results are summarized in Figures 3–4. Pooled incidence proportions of hyporesponders were 26% across studies (Pooled Incidence = 0.26, Standard Error = 0.04). Hyporesponders were observed to have a 2.23-fold increased risk of thromboembolic complications compared to normoresponders (RR = 2.23, P = 0.03). Additionally, analysis of Egger's regression demonstrated that no significant publication bias impacted the findings of this meta-analysis (Bias coefficient = −0.53, Intercept = 1.24, P = 0.60).

Figure 3.

Meta-analyses examining the pooled incidence proportion of thromboembolic complications after neuroendovascular intervention in hyporesponders across studies. The number of hyporesponders in each study population is represented by proportion, with higher values indicating a greater proportion of study hyporesponders and proportional study weight estimated by the size of each box. Error bars represent the 95% confidence interval (CI), with arrows representing an upper limit of the scale.

Figure 4.

Meta-analyses examining the relative risk (RR) of thromboembolic complications after neuroendovascular intervention based on preoperative platelet response grouping (normoresponder vs hyporesponder) in terms of PRU. The likelihood of postoperative TEC for hyporesponders is represented by RR, with values > 1 indicating increased risk and proportional study weight estimated by the size of each box. Error bars represent the 95% confidence interval (CI), with arrows representing an upper limit off the scale. The solid line represents RR = 1.

Discussion

We present a systematic review and meta-analysis of the efficacy of VerifyNow point-of-care platelet-function testing in predicting postoperative thromboembolic complications of neuroendovascular interventions. To the best of our knowledge, this is the first systematic review and meta-analysis on the topic. We emphasize three primary findings: 1) Platelet hyporesponse is largely predictive of postoperative thromboembolic complications of neuroendovascular interventions, solidifying a role for VerifyNow platelet function testing in the perioperative management of neuroendovascular procedures; 2) However, the optimal clinical threshold for PRU to minimize risk of postoperative complications is very poorly defined, as are the definitions of platelet hyporesponse vs hyperresponse in terms of PRU; and 3) There is a large cohort of patients across studies, primarily normoresponders, who conferred no significant benefit from point of care VerifyNow testing, demonstrating the need for additional screening tools to assess cost vs. benefit to patients at the individual level.

Evidence-based benefits of VerifyNow testing

The majority of thromboembolic complications are catastrophic by nature, often resulting in significant disability, poor functional outcomes, and death. As such, the ability to identify high risk patients by this metric provides physicians a critical opportunity to correct any dangerous variability in platelet sensitivity prior to neurointervention, thereby improving the likelihood of a good functional outcome. The results of this review and meta–analysis largely demonstrate that patients defined as platelet hypo-responders on preoperative testing confer a significant mortality benefit from being identified and medically managed prior to neurointervention.^15–36 While four of the included studies did not identify a significant association between platelet hyporesponse and risk of thromboembolic complications, we hypothesize that small study size^17,22 or lack of a comparison or control group^29,33 likely contributed to the lack of significant findings.

Furthermore, platelet function testing with VerifyNow allows for rapid results and ongoing management, meaning patients can be tested continuously throughout treatment to ensure they reach and maintain PRU values in a therapeutic range. This is especially relevant given the significant variability across individuals in terms of baseline platelet reactivity, changes in reactivity over time, and response to treatment, with some patients requiring more intensive treatment or being resistant to common antiplatelet regimens.^37,38 As additional, large scale studies are conducted to finetune the therapeutic thresholds of platelet response in terms of PRU, platelet function testing with VerifyNow can be used more judiciously in routine management protocols.

Shortcomings and socioeconomic considerations

A major point of contention with respect to the use of point-of-care platelet function testing is the test's consistency with light transmission aggregometry (LTA), the historical gold standard for measuring platelet function. While some large-scale studies, such as the VERITAS trial, ³⁹ found VerifyNow to be a reliable, fast, and sensitive metric for platelet inhibition, others ⁴⁰ have demonstrated variable agreement between the two methods. This is problematic as inaccurate measurements could result in inappropriate adjustments to medical management and an inadvertent but increased risk of complications.

Additionally, the added cost of routine PFT with VerifyNow to the patient and the hospital must be considered. While use of VerifyNow to guide long-term DAPT has been demonstrated to be a cost-effective strategy, ⁴¹ this model cannot be applied to its use in the acute, perioperative care setting. Instead, these patients confer a one-time additional cost associated with their procedural hospital stay, which can be increasingly expensive depending on the number of repeat tests and additional antiplatelet therapy needed. Furthermore, cheaper test alternatives to VerifyNow are continually in development, as are studies investigating whether simply changing the antiplatelet regimen can eliminate the need for confirmatory testing.^42,43 As such, the predictive benefits of VerifyNow testing summarized in this review must be routinely considered against more cost-effective alternatives.

Opportunities for intervention

Pending explicit determination of the sensitivity and specificity of the VerifyNow assay for predicting ischemic complications, the creation of a standardized risk tool for physicians could be extremely valuable. Once additional studies have examined risk factors associated with clopidogrel hyporesponse and related ischemic complications, points could be assigned to individual risk factors based on their significance, with the summative score pointing to whether point of care platelet function testing is indicated. For example, risk factors for platelet hyporesponse defined in the literature include diabetes mellitus, age, body weight, BMI and Japanese race, to name a few.^22,44,45 Additionally, several genetic polymorphisms, particularly the CYP2C19*2 allele, have been associated with clopidogrel hyporesponsiveness, with significant differences in allelic prevalence across racial groups. ⁴⁶ According to a recent review, approximately 50% of Asians and 25% of Caucasians are thought to harbor the CYP2C19*2 allele. ⁴⁷ As such, genetic sequencing could be a valuable future tool for assessing cumulative risk of platelet hyporesponse.

Similar tools, such as the Well's criteria, have been utilized for diagnosing deep vein thrombosis (DVT), with marked success and a resultant increase in high value care. ⁴⁸ Such a screening tool for VerifyNow testing would likely increase the proportion of patients benefiting from its use. As such, we propose that the use of VerifyNow PFT should be utilized preoperatively for patients undergoing neurointervention who demonstrate relevant risk factors but should otherwise be used selectively to limit unnecessary testing, associated costs, and over-management of those unlikely to confer significant benefit from its use.

Another consideration is the use of alternative pre-procedural antiplatelet medications, such as ticagrelor, and whether these medications confer less resistance and are a better first-line option for patients undergoing neurointervention. Recent studies have shown that DAPT combining ticagrelor and aspirin seems to be an effective treatment for preventing thromboembolic complications in patients with intracranial aneurysms, without any subsequent increase in hemorrhagic complications.^49,50 However, another study found that, in order to mitigate the risk of thromboembolic complications, the safe PRU range for ticagrelor needs to be shifted to 0–100, which is lower than clopidogrel. ⁵¹ Further validation of the optimal PRU range for ticagrelor and other alternatives is needed.

Limitations

As only published full-text studies were included, results are at risk for publication bias. Studies not written or translated into English were excluded from this review, potentially resulting in missed findings. Only one of the included studies were randomized controlled trials, which are necessary to provide the highest-quality evidence on associated outcome data. Data from low- or middle-income countries were lacking. The variability in RR estimates and calculated standard error of pooled incidence suggests potential heterogeneity, which may be attributed to differences in study populations, methodologies, or other factors. Some studies had few events, leading to wide confidence intervals and potentially less reliable estimates. Despite these limitations, strict PRISMA guidelines were followed to systematically assess and provide a comprehensive analysis of the published literature.

Conclusions

The results of this review and meta-analysis illustrate a critical role for the VerifyNow assay in the perioperative management of neuroendovascular patients, with most relevant studies demonstrating a significant, predictive value for the assay in minimizing the risk of thromboembolic complications. Nonetheless, a widespread lack of standardization in DAPT regimen, thresholds for defining platelet hyper- and hyporesponse, treatment modifications, and perioperative management demonstrate the need for large-scale randomized controlled trials to determine best practice. Furthermore, testing may not be indicated for every patient, risk factors depending, and the routine use of VerifyNow as part of a standardized protocol may confer unnecessary costs to the patient and be wasteful of hospital resources.