PET evaluation of myocardial perfusion function after percutaneous coronary intervention in patients with chronic total occlusion: a systematic review and meta-analysis

Abstract

Objective. The benefit of percutaneous coronary intervention (PCI) in chronic complete coronary artery occlusion (CTO) remains controversial. PCI is currently indicated only for symptom and myocardial ischemia abolition, but large chronically occluded vessels with extensive afferent myocardial territories may benefit most from this procedure. The noninvasive evaluation of myocardial perfusion is critical before and after revascularization, and positron emission tomography (PET) can determine absolute myocardial perfusion. Here, we aimed to explore and compare myocardial perfusion in CTO territories and their remote associated areas before and after PCI. Design. We searched for relevant articles published before November 28, 2022, in the Cochrane Library and PubMed. We calculated 95% confidence intervals (CIs) and standardized mean differences (SMDs) for parameters related to myocardial perfusion in CTO territories and remote areas in CTO patients before and after PCI. Results. We included five studies published between 2017 and 2022, with a total of 592 patients. Stress myocardial blood flow (MBF) was increased in CTO territories after PCI when compared to pre-PCI (mean difference [MD]: 1.70, 95% confidence interval [CI] 1.33-2.08, p < 0.001). Coronary flow reserve (CFR) in CTO regions was also higher after PCI (MD 1.37,95% [CI]1.13-1.61, p < 0.001). Stress MBF in remote regions was also increased after PCI (MD 0.27,95% [CI]0.99 ∼ 0.45, p = 0.004), as was CFR in remote regions (MD 0.32,95% [CI] 0.14-0.5, p = 0.001). Conclusions. According to our pooled analysis of current literature, there was an increase in stress MBF and CFR in both CTOs and remote regions after PCI, suggesting that patients with CTO have widespread recovery of blood perfusion after the procedure. These results provide evidence that patients with CTO arteries and high ischemic burdens would indeed benefit from CTO-PCI. Future research on the correlation of ischemia burden reduction with hard clinical endpoints would contribute to a clearer demarcation of the role of CTO PCI with prognostic potential.

Keywords:

• Coronary artery occlusion
• positron emission tomography
• myocardial perfusion function
• myocardial blood flow

Introduction

Chronic complete coronary occlusion (CTO) refers to 100% stenosis of the coronary arteries and myocardial infarction thrombolysis (TIMI) of 0 flow based on angiography or symptoms for more than three months [1]. CTO is found in up to 30% of patients with coronary artery disease (CAD) [2,3]. There are several symptoms associated with CTO, including angina, and it is also associated with poor disease prognosis [4]. At present, CTO surgery remains difficult, with high complications and limited success rates, and some clinical trials have obtained negative results in terms of major cardiovascular events (MACE) [5–7]. Moreover, the prolonged nature of the CTO procedure and the high degree of contrast ingestion both increase patients’ risks for contrast nephropathy [8]. Many physicians also contend that patients with CTO might not benefit from percutaneous coronary intervention (PCI) because the remaining collateral circulation is considered sufficient to prevent ischemia. Additionally, the non-viable myocardium supplied by the CTO artery may not show functional recovery after CTO recanalization [9,10]. Guan et al. [11] found that optimal access benefits CTO patients. High-quality CTO openings may lead to better clinical prognoses, but non-high-quality CTO openings may increase perioperative complications. As a result, non-invasive tests of ischemia and viability are needed to guide CTO revascularization [12].

Amongst non-invasive nuclear medicine techniques, [15O] H₂O positron emission tomography (PET) has the greatest flow linearity for assessing myocardial perfusion and can determine the absolute level of myocardial perfusion [13]. The recovery of quantitative myocardial perfusion after PCI is crucial for its benefit in CTO patients. However, because fewer CTO patients have had their myocardial perfusion monitored by PET before receiving PCI, there is limited evidence that PCI reduces ischemic burdens, reduces symptoms, or improves myocardial perfusion in this patient population. There is no current gold standard evidence relating to myocardial perfusion in CTO patients after PCI.

Thus, the present study aims to explore and compare myocardial perfusion in CTO patients before and after PCI. A meta-analysis was performed on studies focusing on myocardial reperfusion in the CTO and distant regions. Patients with CTO may benefit from our findings.

Methods

Information sources and search

We searched Pubmed and Cochrane library for studies published in the English language from inception until 28 November 2022. We used several search terms, including ‘coronary flow reserve, CFR, Positron Emission Tomography, pet, blood flow, MBF, myocardial perfusion’ and ‘Chronic Total Occlusion, CTO’. To supplement the first search, we manually reviewed the manuscript reference list. Z.Y.A and J.F.T independently retrieved these studies. X.T.S resolved any inconsistencies. The protocol for this systematic review was registered with PROSPERO (CRD42022383483).

Selection criteria

Selection criteria for studies were as follows: (1) Studies that included patients with CTO (2) Studies where ischemia detection was performed using [O]H ₂ O PET perfusion imaging, followed by successful CTO PCI. (3) Studies where successful revascularization was performed at least three months after surgery, and patients had follow up [15O]H2O PET scans. (4) Studies that excluded Rubidium-82 or [13N] Ammonia radiotracers for radiographic applications of PET. (5) Studies that were RCTs or observational in nature.

Duplicates were removed, title reviews were conducted, and abstract screening was performed. Studies were excluded for the following reasons: (1) There was no clear outcome; (2) No PCI or PET examinations were performed; (3) Studies were reviews, meta-analyses, commentaries, study protocols, abstracts, case reports, or letters. Figure 1 shows the search and screen protocols.

Figure 1. Search for and selection of eligible studies, culminating in the inclusion of five studies.

Data extraction and outcomes

Data were independently extracted by two authors (A.Z.Y and J.F.T) and verified by a third author (X.T.S). The extracted data included the type of study, population, age, gender, Body Mass Index (BMI), CAD risk factors, and CTO artery. The PET myocardial perfusion data extraction included myocardial blood flow (MBF) during rest and hyperemia (Stress MBF), as well as coronary flow reserve (CFR) in CTO and remote areas (e.g. remote myocardial areas without coronary artery disease.).

Statistical analysis

Statistical analyses were performed using Stata 11.0 (StataCorp, College Station, Tex). First, we performed a meta-analysis based on rest MBF, Stress MBF, and CFR before and after PCI in the CTO area. Second, we conducted a meta-analysis of rest MBF, Stress MBF, and CFR before and after PCI in remote, non-CTO cardiac areas. A standardized mean difference (SMD) of 95% confidence intervals (CIs) was calculated. The I² test was used to assess heterogeneity between studies (low heterogeneity: I²≤25%; Moderate heterogeneity: I² 25-50%; Severe heterogeneity: I² > 50%). Statistical analysis was performed using M-H fixed effects models if the I² was equal or less than 50%, and using M-H random effects models if the I² was greater than 50%. P values ≤0.05 were considered to be statistically significant.

Quality assessment and sensitivity analysis

When heterogeneity was high, all endpoints of each study were excluded for sensitivity analysis. Recalculating the set estimate involved random effects. Publication bias was evaluated using Egger’s linear regression tests and funnel plots. The Newcastle-Ottawa Scale (NOS) was used to assess the retrospective study’s quality.

Results

Database searches and analyses yielded 102 references, while other sources yielded 23 references. 104 cases were excluded from the screening process based on their titles and abstracts (Figure 1). 14 studies were included in the final analysis, but nine were excluded for the following reasons: (1) Missing primary endpoint (n = 2); (2) Did not include PCI (n = 3); and (3) Different methods of measurement (n = 4). A total of five references were included in the final meta-analysis.

Baseline characteristics

As shown in Table 1, a total of 592 patients were included in the five selected studies. Four of these studies were prospective [14–17], and one was observational [18]. There were five studies that included myocardial blood perfusion in the primary CTO region and two studies that included myocardial blood perfusion in remote non-CTO regions. The average age range was 62 - 63 years old. The majority of patients were male (80-86%), 16-27% had diabetes, 49-64% had high blood pressure, and 29-62% were current smokers.

Table 1. Basic characteristics of included studies.

	Study	Wijnand(14), 2017	Stefan(16), 2019	Zhang (18) ,2021	Ruben (17), 2022	Schumacher(15), 2018
Basic Information	design	prospectively study	prospectively study	observational study	prospectively study	prospectively study
	population	CTO	CTO	CTO	CTO	CTO
	patients	69	193	74	164	92
	age	63 ± 10	63 ± 11	62 ± 7.5	62 ± 11	62 ± 10
	male	59(86)	163 (84)	64(86)	137(84)	74(80)
	BMI	28 ± 3.9	28 ± 4	/	28 ± 4	28 ± 4
CAD risk factors	Diabetes	11 (16)	49 (25)	16(21)	41(25)	25(27)
	hypertension	34 (49)	103 (53)	47(64)	169(54)	56(61)
	current smoking	20(29)	59 (31)	46(62)	53(32)	28(30)
CTO artery	RCA	51(74)	130 (67)	/	115(79)	64(70)
	LAD	14(20)	40 (21)	/	32(20)	16(17)

Myocardial perfusion before and after PCI in CTO areas

Table 2 details the mean values and standard deviations of MBF, Stress MBF and CFR in the CTO areas before and after PCI in each article.

Table 2. Mean values and standard deviations of MBF, Stress MBF, and CFR in CTO and Remote areas before and after PCI in each article.

PET-releated
CTO area
Study	MBF	MBF-after	Stress MBF	Stress MBF-after	CFR	CFR-after
Wijnand, 2017	0.82 ± 0.24	0.82 ± 0.27	1.22 ± 0.36	2.40 ± 0.90	1.62 ± 0.78	3.13 ± 1.42
Stefan, 2019	0.82 ± 0.23	0.86 ± 0.26	1.24 ± 0.40	2.34 ± 0.84	1.59 ± 0.55	2.88 ± 1.14
Zhang, 2021	0.76 ± 0.17	0.94 ± 0.19	1.32 ± 0.39	2.27 ± 0.52	1.72 ± 0.47	2.73 ± 0.73
Ruben, 2022	0.87 ± 0.24	0.87 ± 0.24	1.71 ± 0.53	2.46 ± 0.81	2.05 ± 0.68	2.91 ± 1.02
Schumacher, 2018	0.86 ± 0.25	0.86 ± 0.22	1.3 ± 0.37	2.48 ± 0.73	1.66 ± 0.75	3.03 ± 1.05
Remote area
Study	MBF	MBF-after	Stress MBF	Stress MBF-after	CFR	CFR-after
Wijnand, 2017	0.99 ± 0.27	0.95 ± 0.27	2.58 ± 0.68	2.77 ± 0.7	2.75 ± 0.93	3.05 ± 0.99
Ruben, 2022	0.95 ± 0.26	0.94 ± 0.25	2.29 ± 0.67	2.48 ± 0.75	2.48 ± 0.76	2.74 ± 0.85

Five studies were included, and meta-analysis techniques were used to summarize and compare the rest of MBF, stress MBF, and CFR values in the CTO areas of patients before and after PCI. We found that the rest of MBF in the CTO region was slightly higher after PCI (mean difference (MD 0.22,95% [CI]-0.09 ∼ 0.53, p = 0.164, Figure 2). Stress MBF in the CTO region was also increased after PCI (MD 1.70,95% [CI]1.33 ∼ 2.08, p < 0.001, Figure 2). In addition, CFR in the CTO region was higher after PCI (MD 1.37,95% [CI] 1.13-1.61, p < 0.001, Figure 2).

Figure 2. Forest plots reveal the mean differences between MBF, stress MBF, and CFR in the CTO regions. The MBF in the CTO region was slightly higher after PCI (MD 0.22, 95% [CI]-0.09 ∼ 0.53, p = 0.164). The hyperemic Stress MBF in the CTO region was increased after PCI (MD 1.70, 95% [CI]1.33 ∼ 2.08, p < 0.001). The CFR in the CTO region was higher after PCI (MD 1.37, 95% [CI] 1.13–1.61, p < 0.001).

Myocardial perfusion before and after PCI in remote areas

Non-CTO remote areas are defined as remote areas of the myocardium without significant coronary artery disease. Table 2 details the mean values and standard deviations for rest MBF, Stress MBF, and CFR in the remote areas before and after PCI in each article.

Two studies were included and meta-analysis was used to summarize and compare the values of rest MBF, stress MBF, and CFR in remote areas after PCI. Mean rest MBF in remote areas decreased slightly after PCI compared with pre-PCI (MD −0.07,95% [CI] − 0.25-0.11, p = 0.441, Figure 3). On the contrary, mean stress MBF in the remote regions increased after PCI (MD 0.27,95% [CI]0.99 ∼ 0.45, p = 0.004, Figure 3). In addition, CFR in the remote regions was higher after PCI (MD 0.32,95% [CI] 0.14-0.5, p = 0.001, Figure 3).

Figure 3. Forest plots reveal the mean differences between MBF, stress MBF, and CFR in the remote regions. MBF in the remote area was decreased slightly after PCIMD -0.07,95% [CI] -0.25-0.11, p = 0.441). Stress MBF in the remote areas was increased after PCI (MD 0.27,95% [CI]0.99 ∼ 0.45, p = 0.004). CFR in the remote areas was higher after PCI (MD 0.32,95% [CI] 0.14-0.5, p = 0.001).

Quality of studies, publication bias, and sensitivity analysis

Heterogeneity was greater than 50% in the above meta-analysis for studies focusing on CTO region reperfusion after PCI. We conducted a sensitivity analysis on the forest map of the CTO region. In the sensitivity analysis, the conclusions drawn from forest plots of mean differences in rest MBF, stress MBF, and CFR in the CTO region were largely consistent with the preliminary analysis when each trial was conducted separately (supplementary Figures 1-3).

Supplementary Table 1 illustrates the quality assessment of the five included studies. Four of the studies scored seven points on the NOS scale, and the remaining study scored six points. We used R studio (R Core Team (2022)) to visualize the quality assessment results. The high bias risks included “Demonstration that the outcome of interest was not present at the start of study” and “Assessment of outcome” (Figure 4). The independent visualization bias risks of each study are shown in Supplementary Figure 4. Additionally, a funnel plot (Supplementary Figures 5-7) indicated publication bias.

Figure 4. Risk of bias graph reveals the high bias risks are “demonstration that outcome of interest was not present at the start of study” and “assessment of outcome”.

Discussion

We conducted a meta-analysis of relevant articles comparing myocardial perfusion in CTO areas and remote, non-CTO myocardial areas before and after CTO-PCI. The main results were as follows: (1) There was no change in the resting MBF of CTO patients after PCI in either CTO or remote areas. (2) There was an increase in the stress MBF and CFR of CTO patients after PCI in both CTO and remote areas. The results suggested some clinical benefits of PCI for CTO patients in terms of PET-evaluated ischemia.

Ischemic burden in patients with CTO

Our study population was patients with CTO, primarily because treatment for this population remains highly controversial, with inconsistent data and questions about the utility of revascularization in chronic coronary syndromes in general [19,20]. The extent of improvements in MACE events in patients after CTO-PCI remains unclear. No improvement in MACE was found in the DECISION-CTO trial (NCT 01078051) [6], but the newer EURO-CTO (NCT 01760083)trial, which was just published in 2023 [21], showed that successful revascularization PCI in patients with CTO was effective in reducing the occurrence of MACE events. Moreover, the OPEN-CTO [22] (Outcomes, Patient Health Status, and Efficiency in Chronic Total Occlusion Hybrid Procedures) registry revealed that CTO patients with refractory angina had larger improvements in SAQ Angina Frequency and SAQ Summary Scores after successful CTO PCI. These results may be derived from improvement in ischemia within the CTO territory. The mechanics of coronary blood flow in patients with CTO are complicated because absolute perfusion of the myocardium in this condition may be dependent on the function of collateral circulation [23–25]. However, well-developed collaterals do not always translate into diminished ischemic burdens. Previous studies have shown that, even with well-developed collaterals, blood supply remains insufficient in situations with increased demand, resulting in myocardial ischemia in the CTO area in more than 90% of patients [10, 26]. According to a study by Schumacher et al. there were supply and demand mismatches in the CTO-supported ischemic myocardium, and collateral perfusion was recruited to this region, suggesting that even in the presence of collaterals there was reduced distal myocardial perfusion [27]. Previous studies by our group [28,29] have shown that there is no significant correlation between the severity and degree of perfusion defects, myocardial viability, or grade of collateral circulation. Khan et al. [30] used thermodilution to demonstrate improved blood flow in CTO patients after PCI. Therefore, early revascularization and early opening of CTO vessels is critical. The present study also suggested that myocardial perfusion in CTO areas was increased after PCI, indicating that successful CTO PCI could improve ischemia in these areas.

Myocardial perfusion in non-CTO remote areas after PCI

Previous studies have shown further regression of collateral circulatory function after revascularization in CTO [31,32]. In addition, in terms of blood flow reserve, previous studies have demonstrated the influence of myocardial tissue perfusion on FFR values, and have suggested that chronic complete occlusion of collaterals may be associated with reductions in FFR values [33]. Other studies have demonstrated that collateral donor artery FFR increases and mean peak flow velocity decreases in CTO patients after PCI. These findings suggest that there is a collateral flow to the CTO myocardium even in the absence of obstructive CAD in the donor artery [34]. Our analysis indicated that myocardial blood perfusion in remote myocardial areas in CTO patients was also increased after PCI, which is consistent with the above studies. Mechanistically, revascularization of CTO vessels may improve overall myocardial congestion and perfusion conditions, including in the distal myocardium, and/or blood flow may return to collateral circulation in the CTO region at the time of revascularization, thus increasing blood perfusion within the collaterals and resulting in enhanced distal myocardial perfusion [35]. Thus, after PCI revascularization in CTO patients, increases in absolute myocardial perfusion in CTO areas are collinear with improvements in distal myocardial perfusion.

Higher perioperative risks in CTO-PCI

Because of the particularities of CTO vessels, CTO-PCI has higher perioperative risks than conventional PCI. CTO surgery is prone to donor vessel injury, which can lead to extensive ischemia and hemodynamic decompensation [36]. Coronary perforation is another of the most feared complications of CTO-PCI. In a study by Azzalini [37] et al. 5.5% of CTO-PCI cases had perforations, with half of these cases requiring management and 20% resulting in cerebral infarction. Collateral occlusion may also occur in patients with CTO after PCI, especially when the subintimal dissection/reentry strategy is used. This approach is also associated with a high risk of myocardial infarction after PCI [38]. Radiation exposure for patients undergoing PCI for CTO is higher than that of normal PCI due to the increased procedure times [39]. Ge [40] et al. showed that the use of a low frame rate protocol for CTO-PCI during the procedure made it safer. As for how to perform complex CTO-PCI surgeries more safely, Achim et al. [41] showed that the use of distal radial access for complex CTO intervention was safe and feasible, and could reduce radiation doses and make dual radial access easier to achieve, with no increased perioperative risk or long-term adverse outcomes. In addition, a study on CTO operators conducted by Young et al. [42] showed that there was an experiential learning curve for procedural success among new CTO PCI surgeons. Thus, performing CTO-PCI under the guidance of more experienced surgeons may be a good solution.

These studies suggest that CTO PCI procedures should try to minimize risks and that carefully selected patients, particularly those with LAD or dominant LCX/RCA problems, can help improve outcomes. Our pooled analysis indicated that myocardial perfusion in both the CTO region and distant cardiac regions is improved after early revascularization in CTO patients. In addition, many studies [43] have shown that CTO recanalization reduces the burden of ischemia, is conducive to reverse remodeling, and can improve quality of life. Therefore, we would recommend CTO-PCI for patients with high ischemic burdens.

Limitation

Our meta-analysis had several limitations. First, the studies we included were observational and prospective studies, with no RCT-type studies, suggesting that further research is needed to confirm and refine our findings. Second, our research analysis has great heterogeneity, which was related to the small number of studies and the small sample sizes within the studies. Third, there are relatively few articles on myocardial blood perfusion evaluated by PET in distant non-CTO myocardial regions. Although we obtained consistent results, further studies are needed to confirm them. Finally, this is a PET CT-only paper. Other cohorts of CTO PCI patients have been assessed with other imaging modalities, such as CMR [43,44], which can also assess viability, scars, and stress tests [45], but we did not include those studies here.

Conclusion

In conclusion, we reviewed studies evaluating MBF PET in CTO regions before and after PCI. Myocardial perfusion status after PCI in CTO patients was improved both in CTO regions and remote areas. Patients with CTO arteries, and especially those with ischemia symptoms or evidence of ischemia, could potentially benefit from CTO-PCI. More studies on the correlation between ischemic burden reduction and clinical endpoints are needed to further define the role of PCI in patients with CTO.

Consent for publication

All authors have reviewed the final version of this manuscript and have approved it for publication.

Author contributions

Xiantao Song and Liying Chen helped conceive of the idea. Ziyu An and Jinfan Tian wrote the manuscript. Xueyao Yang, Libo Liu, and Lijun Zhang helped structure the text. Mingduo Zhang and Xin Zhao helped revise the manuscript.

Abbreviations
CTO	=	coronary complete occlusion
CAD	=	coronary artery disease
MACE	=	major cardiovascular events
PET	=	positron emission tomography
PCI	=	Percutaneous coronary intervention
MBF	=	myocardial blood flow
CFR	=	coronary flow reserve

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (82100486 and 82270341), the Capital Health Development Research Project (No. 2022-1-2061), the Beijing Municipal Science and Technology Project (Z161100000516139), the ‘Qingmiao’ talent plan (QML20210603), ‘Deng Feng’ Talent Training Program (DFL20220603) and High-level Public health Technical Talents Construction Project (Leading Talents).