Nicorandil in ST-segment elevation myocardial infarction patients undergoing percutaneous coronary intervention: A systematic review and meta-analysis

Abstract

BACKGROUND

Primary percutaneous coronary intervention (PCI) is the standard treatment for ST-elevation myocardial infarction (STEMI). However, its effectiveness may be limited by reperfusion injury and the no-reflow phenomenon.

AIM

To improve myocardial microvascular function and reduce no-reflow.

METHODS

A systematic search was conducted on MEDLINE, Scopus, EMBASE, and Cochrane CENTRAL from inception to January 25, 2025. Randomized controlled trials comparing nicorandil to control in STEMI patients undergoing PCI were included. Primary outcomes included ST-segment resolution, Thrombolysis in Myocardial Infarction grades, Thrombolysis in Myocardial Infarction Myocardial Perfusion Grade, left ventricular (LV) functions, and infarct size. R version 4.4.0 was used to conduct statistical analyses. This review has been registered with PROSPERO (CRD42024614029).

RESULTS

A total of 26 randomized controlled trials involving 3258 participants were included in this review. Individuals receiving nicorandil had a higher rate of complete ST segment resolution than the control group. It significantly reduced infarct size and the LV end-systolic volume index at 6 months. Furthermore, LV ejection fraction showed significant improvement with nicorandil therapy at 1 week, 1 month, 3 months, and 6 months. Patients receiving nicorandil also had a notably higher incidence of normal epicardial flow (Thrombolysis in Myocardial Infarction grade 3) and normal tissue-level perfusion (Thrombolysis in Myocardial Perfusion grade 3) compared to controls. Additionally, nicorandil significantly lowered the risk of major adverse cardiovascular events relative to the control group.

CONCLUSION

Nicorandil improves myocardial perfusion and LV function following PCI in STEMI patients while maintaining a favorable safety profile. Further high-quality trials are needed to confirm its long-term efficacy and clinical benefits.

Key Words: Nicorandil; Percutaneous coronary intervention; ST-segment elevation myocardial infarction; Myocardial perfusion; Myocardial infarction

Core Tip: Nicorandil, a nitrate-like agent that opens adenosine triphosphate-sensitive potassium channels, has been studied as an adjunct during primary percutaneous coronary intervention for ST-elevation myocardial infarction. In this systematic review and meta-analysis of randomized trials, nicorandil improved ST-segment resolution, Thrombolysis in Myocardial Infarction grade 3 flow, and Thrombolysis in Myocardial Perfusion grade 3. It also enhanced left ventricular function and lowered major adverse cardiovascular events. These findings support its value in reducing microvascular injury and improving post-percutaneous coronary intervention outcomes.

INTRODUCTION

Percutaneous coronary intervention (PCI) is the mainstay of treatment for ST-elevation myocardial infarction (STEMI) in patients that present within 12 hours of symptom onset. Timely reperfusion therapy improves outcomes by reducing infarct size and improving ventricular function[1]. However, the clinical efficacy of PCI is limited in the case where reperfusion injury occurs and leads to no-reflow (NR) phenomenon, as it is associated with an increase in the infarct size and lead to poor clinical outcomes[2]. Several mechanisms have been proposed that might contribute to NR including vasoconstriction, embolization of atherothrombotic material, platelet aggregation and generation of toxic free radicals.

Nicorandil is a hybrid agent with both nitrate-like and adenosine triphosphate-sensitive potassium channel-opening properties and is indicated in chronic stable angina wherein it reduces symptoms by improving microvascular dysfunction and coronary vasospasm[3]. Due to its effects on coronary microvasculature and the mimicking of ischemic preconditioning, several trials have been performed to demonstrate the efficacy and safety of nicorandil during primary PCI in patients for STEMI in case of NR. Recent trials have shown that this drug can improve ischemic reperfusion injury and NR phenomenon by improving the myocardial microvascular function[4,5]. However, some other trials show no beneficial effects of nicorandil in this setting[6].

A meta-analysis done by Geng et al[7] in 2021 showed that nicorandil was very effective in reducing NR and was associated with a decreased incidence of major adverse cardiovascular events (MACEs). Since then, several randomized controlled trials (RCTs), including the CHANGE trial, have been conducted aiming at better understanding the real efficacy of this drug[5]. Furthermore, we performed subgroup analyses based on modes of administration to assess efficacy accordingly. Each outcome was further stratified based on the duration of follow-up. Due to the conflicting evidence in recent trials for the use of nicorandil, we performed an updated meta-analysis and systematic review to better quantify the clinical outcomes of this drug in patients undergoing primary PCI.

MATERIALS AND METHODS

We followed guidelines provided by the Preferred Reporting Items for Systematic Review and Meta-Analyses to conduct our meta-analysis[8]. We registered our review protocol with the International Prospective Register of Systematic Reviews (PROSPERO: CRD42024614029).

Data sources and search strategy

An electronic search was done on the following databases: MEDLINE (via PubMed), Scopus, EMBASE (via OVID), and the Cochrane CENTRAL Library from their inception until January 25, 2025. Relevant trials were also searched through ClinicalTrials.gov. Snowballing was conducted, during which we identified review articles in the final studies and searched for any potentially eligible studies to include in our review. Detailed search strategy for our review is demonstrated in Supplementary Table 1.

Study selection

Two reviewers (Akram U and Ahmad E) assessed all relevant studies individually and in case of any discrepancies a third reviewer (Ahmed S) was consulted. Studies inclusion criteria involved: (1) Study participants had STEMI and underwent PCI; (2) Nicorandil was used as an intervention; (3) Placebo involved no agent or any other positive drug; (4) Primary outcomes involving at least one of the following: ST-segment resolution, Thrombolysis in Myocardial Infarction grades (TIMI), TIMI Perfusion Grade (TMPG), left ventricular (LV) function, infarct size; (5) Published in English language; and (6) Were RCTs. Studies were excluded from our review: Conference abstracts, narrative reviews, case reports, and case series.

Data collection and data items

After the complete screening, following data was extracted from the final included studies: First author’s last name, publication year, study population in each arm, mean age of participants. Extraction for all the reported primary outcomes (ST-segment resolution, TIMI, TMPG, LV function, and infarct size) was done along with secondary outcomes that involved MACEs. For outcomes where median and interquartile ranges were provided, methods by Luo et al[9] for mean and Wan et al[10] for standard deviations were utilized.

Risk of bias assessment

The Cochrane Risk of Bias tool for randomized trials (RoB 2.0) was used to assess the risk of bias in the included RCTs[11]. This tool evaluates a study based on five domains: (1) The randomization process; (2) Deviations from intended interventions; (3) Missing outcome data; (4) Measurement of the outcome; and (5) Selection of the reported result. Two independent reviewers conducted this assessment and discrepancies were sorted by consulting a third reviewer.

Statistical analysis

Statistical analysis was conducted on R version 4.4.0 using the packages “meta” and “metasens”. For continuous outcomes, mean difference (MD) along with their 95% confidence intervals (CIs) for continuous outcomes were pooled utilizing the inverse variance method, while risk ratio (RR) and their 95%CIs were calculated for dichotomous outcomes using the Mantel-Haenszel method within a random effects model[12]. The derSimonian-Laird estimator was applied for tau² and the Jackson method for the CI of tau² and tau. Forest plots were formed to evaluate the pooled outcome. Subgroup analyses were conducted based on the mode of administration for nicorandil. Heterogeneity was assessed using values as recommended by the Cochrane Handbook of Systematic Reviews of Interventions for the Higgins I² statistic, keeping in view the results of the χ² test - 0% to 40%: Low heterogeneity; 30% to 60%: Moderate heterogeneity; 50% to 90%: Substantial heterogeneity; and 75% to 100%: Considerable heterogeneity[13]. Funnel plots were generated along side Egger’s P value to report publication bias[14]. However, Doi plots and the Luis-Furuya Kanamori index (LFK index) were generated for outcomes with less than ten studies since they possess more sensitivity and statistical power in such cases[15]. When there were more than two studies that had a high heterogeneity for an outcome, sensitivity analysis was conducted using the leave-one-out method. A two-tailed P value < 0.05 was considered significant for all instances.

Certainty of evidence

We used the Grading of Recommendations Development, Assessment and Evaluation approach to determine the certainty of evidence.

RESULTS

The initial search of the electronic database yielded 1957 potentially relevant studies. Following duplicate removal and title and abstract screening, 42 records were included for further review. Finally, a total of 26 clinical trials[4-6,16-38] involving 3258 patients (nicorandil, n = 1655; control, n = 1603) were included in this systematic review and meta-analysis (Figure 1). The majority of studies (n = 14) originated from Japan, followed by China (n = 8), Korea (n = 2), with one study each from Iran and Pakistan. The mean age of the participants in the intervention group was 60.86 years, whereas that of the controls was 60.53 years. Table 1 shows the baseline characteristics of the included studies.

Figure 1 PRISMA flow diagram. RCT: Randomized controlled trial.

Table 1 Baseline characteristics of studies included in systematic review and meta-analysis.

Ref.	Country	N	C	Participants (n)		Age (years)		Male (%)		DM (%)		HTN (%)		Dyslipidemia (%)		Smokers (%)
				N	C	N	C	N	C	N	C	N	C	N	C	N	C
Ito et al[16], 1999	Japan	4 mg bolus IV; 6 mg/hour continuous infusion for 24 hours; 15 mg/day oral (for a mean of 28 days)	No agent	40	41	60	60	80	78	25	29	53	51	27	34	60	51
Fukuzawa et al[17], 2000	Japan	4 mg bolus IV; 6 mg/hour infusion for 24 hours	No agent	31	31	61.5	61.3	74.2	71	32.3	29	58.1	67.7	45.2	48.4	61.3	61.3
Ikeda et al[18], 2004	Japan	Infusion 6 mg/hour IV for 72 hours; 2 mg IC	Isosorbide dinitrate 6 mg/hour IV for 72 hours, 2 mg IC	30	30	60	63	76.7	83.3	16.7	20	NA	NA	NA	NA	NA	NA
Nameki et al[19], 2004	Japan	4 mg IV + 4 mg IC before reperfusion; 4 mg/hour IV for 24 hours	Control 1: Magnesium: 10 mmol IV before reperfusion, 0.4 mmol/hour for 24 hours; control 2: No agent	13	27	64	62	85	85	31	26	46	44	23	44	77	78
Ono et al[20], 2004	Japan	4 mg bolus IV, 8 mg/hour IV infusion for 24 hours after PCI	No agent	33	25	64	66	67	64	33	32	55	56	42	36	45	36
Ishii et al[21], 2005	Japan	12 mg dissolved in normal saline 100 mL IV	Normal saline	185	183	63	64	77.8	84.1	33	31.7	28.1	31.1	25.9	28.4	35.7	42
Kasama et al[22], 2005	Japan	2 mg bolus IV, continuous infusion 4 mg/hour, oral 15 mg/day	Isosorbide dinitrate 2 mg IV bolus, continuous infusion 4 mg/hour, oral 40 mg/day	25	25	62	63	72	76	24	32	48	44	32	40	52	60
Akagi et al[23], 2006	Japan	Intervention 1: Continuous infusion IV 4 mg/hour, 2 mg IC; intervention 2: Continuous infusion IV 4 mg/hour, 2 mg IC, oral 15 mg/day	No agent	20	10	65.5	61	50	90	NA	NA	NA	NA	NA	NA	NA	NA
Miyazawa et al[24], 2006	Japan	2 mg IC; 2 mg/hour infusion for 24 hours IV; 15 mg/day oral	No agent	35	35	64	60	88	74	28	40	48	57	31	37	54	57
Ota et al[25], 2006	Japan	Intervention 1: 1-2 mg IC (1-2 times); intervention 2: 96 mg dissolved in 100 mL saline, 4 mg IV, infusion 6 mg/hour, IC same dose as in intervention 1 group	No agent	63	27	62.2	64.2	82.5	74.1	29	33	44	41	60	44	54	52
Toyama et al[26], 2006	Japan	4 mg IV bolus, infusion 4 mg/hour over 24-hour period, 2 mg IC	No agent	33	35	65	63	68.5	57	24	34.5	35.5	37	35.5	22.5	69.5	77.5
Fujiwara et al[27], 2007	Japan	4 mg IV bolus, infusion 8 mg/hour for 24 hours	No agent	31	31	62	62	80.6	80.6	38.7	38.7	58.1	51.6	45.2	35.5	70.9	80.6
Kitakaze et al[28], 2007	Japan	0.067 mg/kg bolus IV, 1∙67 μg/kg per minute infusion IV for 24 hours	Normal saline	276	269	61.1	63.7	89.1	81.8	39.5	32.9	48.5	53.9	46.7	46.2	68.7	66.1
Lee et al[29], 2008	Korea	2 mg IC before CAG, 2 mg IC before stenting	No agent	37	36	56.4	60.2	83.8	83.3	27	36.1	51.4	66.7	NA	NA	86.5	69.3
Yamada et al[30], 2016	Japan	0.2 mg/kg IC before the initial and final angiograms, 2.0 mg/hour infusion IV for 4 days	Nitroglycerin 0.2 mg/kg IC before the initial and final angiograms; 2.0 mg/hour infusion IV for 4 days	28	24	67	65	82	91	21	41	32	33	14	37	32	37
Chen et al[31], 2015	China	Intervention 1: 2 mg IC, anisodamine 2 mg IC; intervention 2: 2 mg IC	Control 1: Anisodamine 2 mg IC; control 2: Only PCI	52	52	58.1	59.8	75	73.1	26.9	30.8	40.4	46.2	21.2	23.1	67.3	65.4
Qi et al[32], 2018	China	2 mg IC	Control 1: Nitroprusside: 200 μg IC; control 2: Saline only	40	80	56	60	72.5	71.3	42.5	33.8	55	53.8	40	35	44	46.3
Pi et al[33], 2019	China	Intervention 1: 4 mg IC, 4 mg/hour IV infusion for 24 hours; intervention 2: 4 mg IC, normal saline 4 mL/hour IV for 24 hours	Normal saline	95	45	68.5	68.7	72.6	72.1	63.2	62.2	62.1	66.7	61.1	57.8	71.6	71.1
Chen et al[34], 2020	China	0.06 mg/kg IC; 2 mg/hour IV infusion for 36 hours, tirofiban 10 μg/kg IC; 0.1 μg/kg/minute IV infusion for 36 hours	Tirofiban 10 μg/kg IC, 0.1 μg/kg/minute IV infusion for 36 hours	39	39	67	66.2	92.3	89.7	30.7	25.6	51.3	48.7	NA	NA	74.4	76.9
Feng et al[35], 2019	China	2-6 mg IC; thrombectomy; tirofiban 10 mg/kg IC	Saline 2-6 mL IC; thrombectomy; tirofiban 10 mg/kg IC	84	86	69.2	68.5	70	72	42	36	56	59	NA	NA	45	49
Wang et al[36], 2019	China	5 mg oral, 5 mg three times/day for 6 months	5 mg oral nicorandil	60	57	58.4	55.8	54	47	20	17.5	46.7	38.6	NA	NA	58.3	47.4
Akbari et al[37], 2020	Iran	40 mg oral	No agent	116	124	58.9	57.9	78.4	83.1	21.6	18.5	39.7	41.1	14.7	16.1	47.4	54
Wang et al[38], 2021	China	IV infusion	5 mg oral nicorandil before PCI, then no agent given	59	60	54	55	79.7	78.3	10.2	8.3	42.4	43.3	NA	NA	67.8	68.3
Qian et al[5], 2022	China	6 mg IV bolus, IV infusion 6 mg/hour	No agent	120	118	60	58	85	87.3	25.8	18.6	50	56.8	18.3	21.2	58.3	61
Choe et al[6], 2023	Korea	4 mg IV bolus, infusion IV 6 mg/hour for 24 hours (two 2 mg IC doses to reduce adverse events)	No agent	40	43	59.5	59.8	82.5	86	30	44.2	47.5	37.2	NA	NA	50	58.1
Ilyas et al[4], 2024	Pakistan	Nicorandil along with conventional treatment	Conventional treatment (tablet nitroglycerin 0.5 mg sublingually stat, tablet aspirin 300 mg orally stat, tablet clopidogrel 300 mg orally stat, injection heparin 5000 units IV)	70	70	35.1	25.7	82.1	92.9	NA	NA	NA	NA	NA	NA	NA	NA

N: Nicorandil; C: Control; DM: Diabetes mellitus; HTN: Hypertension; IV: Intravenous; IC: Intracoronary; CAG: Coronary angiogram; PCI: Percutaneous coronary intervention.

Risk of bias

Most of the included studies had a high risk of bias while only five studies[5,6,21,7,38] had low risk of bias when assessed using the RoB-2 tool (Supplementary Figures 1 and 2). Bias in the studies predominantly arose from challenges with randomization and deviations from the intended interventions.

ST-segment resolution

Our meta-analysis of 9 trials[4,5,21,25,31,32,34,37,38] involving 1461 patients reported a significantly higher rate of completely settled ST segments in patients receiving nicorandil compared to the control group (RR = 1.29, 95%CI: 1.12-1.49, P < 0.01, I² = 67%). There was no significant difference between the subgroups when stratified by the route of intervention (P = 0.68) (Figure 2). For partially settled ST segments the pooled analysis showed no statistically significant difference between the nicorandil group and the control group (RR = 0.41, 95%CI: 0.16-1.07, P = 0.07, I² = 65%) (Supplementary Figure 3), while the risk of non-settled ST segments was significantly lower in the nicorandil group compared to the control group (RR = 0.14, 95%CI: 0.03-0.59, P < 0.01, I² = 0%) (Supplementary Figure 3).

Figure 2 Forest plot for ST-segment resolution following percutaneous coronary intervention. The overall pooled analysis demonstrated a significantly higher rate of complete ST-segment resolution with nicoranadil compared with control. Subgroup analysis stratified by route of administration showed no statistically significant difference. MH: Mantel-Haenszel; CI: Confidence interval; IV: Intravenous; IC: Intracoronary; PO: Per orem.

Corrected TIMI frame count

Our meta-analysis of 7 trials[5,20,21,25,32-34] involving 1052 patients reported a non-significant reduction in corrected TIMI frame count (CTFC) by nicorandil therapy following PCI (MD: -0.56, 95%CI: -5.15 to 4.03, P = 0.81, I² = 91%). There was no significant difference between the subgroups when stratified by the route of intervention (P = 0.34) (Figure 3).

Figure 3 Forest plot for corrected Thrombolysis in Myocardial Infarction frame count. The pooled and subgroup analysis showed no significant difference between nicoranadil and the control group. Substantial heterogeneity was noted in the overall and subgroup analyses. CI: Confidence interval; IV: Intravenous; IC: Intracoronary.

TIMI grades

Patients receiving nicorandil had a higher incidence of TIMI grade 3 after PCI when compared with the control group (RR = 1.08; 95%CI: 1.02-1.45, P < 0.01, I² = 50%). The incidence of TIMI grade 2 was significantly lower (RR = 0.64; 95%CI: 0.44-0.94, P = 0.02, I² = 0%). However, no significant differences were observed in the incidence of TIMI grade 1 (RR = 0.64; 95%CI: 0.27-1.53, P = 0.32, I² = 0%) or TIMI grade 0 (RR = 0.57; 95%CI: 0.22-1.46, P = 0.24, I² = 0%) (Figure 4).

Figure 4 Forest plot for Thrombolysis in Myocardial Infarction grades (0, 1, 2, and 3). A: Thrombolysis in Myocardial Infarction (TIMI) grades 0; B: TIMI grades 1; C: TIMI grades 2; D: TIMI grades 3. Nicorandil was associated with a higher incidence of TIMI grade 3 and a lower incidence of TIMI grade 2, while no statistically significant differences were observed for TIMI grades 0 or 1. Subgroup analysis for TIMI grade 3 demonstrated significant differences by route of administration. MH: Mantel-Haenszel; CI: Confidence interval; IV: Intravenous; IC: Intracoronary; PO: Per orem.

There was a significant difference (P = 0.02) between the subgroups for TIMI grade 3 when stratified by the route of intervention: RR = 1.09 (95%CI: 1.00-1.20, P = 0.05, I² = 57%) for intravenous, RR = 1.15 (95%CI: 1.05-1.26, P < 0.01, I² = 0%) for intra-coronary (IC), and RR = 1.11 (95%CI: 0.96-1.28, P = 0.15, I² = 17%; ) for intravenous + IC (Figure 4).

Thrombolysis in Myocardial Perfusion grades

Patients receiving nicorandil had a higher incidence of Thrombolysis in Myocardial Perfusion (TMP) grade 3 after PCI when compared with the control group (RR = 1.12; 95%CI: 1.01-1.24, P = 0.03, I² = 50%) (Figure 5). However, no significant differences were observed in the incidence of TMP grade 2 (RR = 0.67; 95%CI: 0.39-1.18, P = 0.17, I² = 0%), TMP grade 1 (RR = 0.58; 95%CI: 0.26-1.28, P = 0.17, I² = 0%) or TMP grade 0 (RR = 0.64; 95%CI: 0.25-1.63, P = 0.35, I² = 0%). There was a significant difference (P = 0.02) between the subgroups for TMP grade 3 when stratified by the route of intervention: RR = 1.17 (95%CI: 1.06-1.29, P < 0.01, I² = 0%) for IC (Figure 5).

Figure 5 Forest plot for Thrombolysis in Myocardial Infarction Myocardial Perfusion Grades (0, 1, 2, and 3). A: Thrombolysis in Myocardial Infarction Myocardial Perfusion Grades (TMPG) grades 0; B: TMPG grades 1; C: TMPG grades 2; D: TMPG grades 3. Nicorandil significantly increased the incidence of TMPG 3 following percutaneous coronary intervention, while no such differences were observed for TMPG 0,1 and 2. Subgroup analysis showed a statistically significant difference for TMPG 3 when stratified by route of administration. MH: Mantel-Haenszel; CI: Confidence interval; IV: Intravenous; IC: Intracoronary; PO: Per orem.

Left ventricle function

LV ejection fraction: Following PCI, STEMI patients receiving nicorandil therapy demonstrated significant improvement in LV ejection fraction (LVEF) at 1 week (MD: 3.72, 95%CI: 0.30-7.14, P = 0.03, I² = 93%), 1 month (MD: 2.36, 95%CI: 0.87-3.86, P < 0.01, I² = 0%), 3 months (MD: 2.60, 95%CI: 0.37-4.83, P = 0.02, I² = 0%), and 6 months (MD: 3.31, 95%CI: 1.85-4.78, P < 0.01, I² = 82% ) (Figure 6).

Figure 6 Forest plot for left ventricular ejection fraction following percutaneous coronary intervention in ST-segment elevation myocardial infarction patients receiving nicorandil after 1 week, 1 month, 3 months, and 6 months. A: Left ventricular ejection fraction (LVEF) at 1 week; B: LVEF at 1 month; C: LVEF at 3 months; D: LVEF at 6 months. Nicorandil was associated with significant improvement in LVEF at all follow-up intervals. No statistically significant differences were observed between subgroups stratified by route of administration at any follow-up interval. CI: Confidence interval; IV: Intravenous; IC: Intracoronary.

However, no significant change was observed immediately after PCI (MD: -0.04; 95%CI: -1.02 to 0.94, P = 0.94, I² = 0%; Supplementary Figure 4). There was no significant difference between the subgroups when stratified by the route of intervention immediately after PCI (P = 0.69) at week 1 (P = 0.89), month 1 (P = 0.82), month 3 (P = 0.58) and month 6 (P = 0.69) (Figure 6).

Other LV function measures: Nicorandil significantly reduced the LV end-systolic volume index at 6 months (MD: -10.70, 95%CI: -12.69 to -8.71, P < 0.01, I² = 63%). However, the reduction in LV end-diastolic diameter at 3 months (MD: -1.66, 95%CI: -7.90 to 4.58, P = 0.60, I² = 84%), LV end-diastolic volume (LVEDV) at 6 months (MD: -13.65, 95%CI: -29.85 to 2.54, P = 0.10, I² = 68%), LVEDV index immediately after PCI (MD: -0.34, 95%CI: -4.31 to 3.63, P = 0.87, I² = 87%), and LVEDV index at 6 months (MD: -8.55, 95%CI: -17.45 to 0.35, P = 0.06, I² = 83%) was not significant compared to the control group (Figure 7).

Figure 7 Forest plot for left ventricular functions (left ventricular end-diastolic diameter at 3 months, left ventricular end-diastolic volume at 6 months, left ventricular end-diastolic volume index immediately after percutaneous coronary intervention, left ventricular end-diastolic volume index at 6 months, and left ventricular end-systolic volume index at 6 months). Nicorandil was associated with a significant reduction in left ventricular end-systolic volume index at 6 months, while no statistically significant differences were observed for left ventricular end-diastolic diameter, left ventricular end-diastolic volume, or left ventricular end-diastolic volume index. CI: Confidence interval; IV: Intravenous; LVEDD: Left ventricular end-diastolic diameter; LVEDV: Left ventricular end-diastolic volume; LVESV: Left ventricular end-systolic volume.

Infarct size

Nicorandil did not significantly reduce infarct size on cardiac magnetic resonance imaging at 5 days to 7 days (MD: -2.51, 95%CI: -7.88 to 2.85, P = 0.36, I² = 79%) or at 6 months (MD: -2.94, 95%CI: -8.35 to 2.47, P = 0.29, I² = 83%; Supplementary Figure 5). The reduction in LV infarct size by nicorandil was also not significant at 5 days to 7 days (MD: -3.00, 95%CI: -6.17 to 0.16, P = 0.06, I² = 65%). However, at 6 months, there was significant reduction in infarct size (MD: -2.96, 95%CI: -4.51 to -1.41, P < 0.01, I² = 0%) compared to the control group (Supplementary Figure 5).

MACE

Nicorandil significantly reduced the incidence of MACEs when compared with the control group (RR = 0.54; 95%CI: 0.41-0.70, P < 0.01, I² = 52%). There was no significant difference between the subgroups when stratified by the route of intervention (P = 0.59) (Supplementary Figure 6).

When stratified by individual MACEs, nicorandil significantly reduced the incidence of ventricular tachycardia/fibrillation (RR = 0.51; 95%CI: 0.39-0.66, P < 0.01, I² = 0%), congestive heart failure (RR = 0.36; 95%CI: 0.20-0.66, P < 0.01, I² = 0%), and angina (RR = 0.25; 95%CI: 0.11-0.59, P = 0.14, I² = 0%); however, it did not significantly reduce target vessel revascularization (RR = 0.60; 95%CI: 0.10-3.63, P = 0.58, I² = 25%), death (RR = 0.62; 95%CI: 0.33-1.15, P = 0.13, I² = 0%), myocardial infarction (RR = 0.79; 95%CI: 0.28-2.23, P = 0.66, I² = 0%), acute stent thrombosis (RR = 0.65; 95%CI: 0.08-5.12, P = 0.68, I² = 0%), stroke (RR = 0.67; 95%CI: 0.08-5.35, P = 0.70, I² = 0%), heart failure (RR = 0.55; 95%CI: 0.25-1.22, P = 0.14, I² = 0%), and postoperative hypotension (RR = 0.52; 95%CI: 0.15-1.81, P = 0.31, I² = 0%) (Supplementary Figure 7; Table 2).

Table 2 Analysis of individual major adverse cardiovascular events demonstrating the safety of the nicorandil group.

Outcome	Number of studies	Effect sizes		Heterogeneity
		RR (95%CI)	P value	Tau2	I2 (%)
VT/VF	15	0.51 (0.39-0.66)	< 0.01	0.01	0
Death	13	0.62 (0.33-1.15)	0.13	0	0
Myocardial infarction	6	0.79 (0.28-2.23)	0.66	0	0
TVR	5	0.60 (0.10-3.63)	0.58	0.87	25
Congestive heart failure	4	0.36 (0.20-0.66)	< 0.01	< 0.0001	0
Pericardial effusion	3	0.45 (0.17-1.21)	0.12	0.27	34
Chest pain	3	0.64 (0.31-1.33)	0.24	0.23	59
Acute stent thrombosis	2	0.65 (0.08-5.12)	0.69	0	0
Stroke	2	0.67 (0.08-5.35)	0.7	0	0
Heart failure	2	0.55 (0.25-1.22)	0.14	0	0
Angina	2	0.25 (0.11-0.59)	< 0.01	0	0
Postoperative hypotension	2	0.52 (0.15-1.81)	0.31	0	0
Cardiogenic shock	1	1.00 (0.06-15.43)	-	-	-
Intraoperative hypotension	1	0.13 (0.03-0.55)	-	-	-

RR: Risk ratio; CI: Confidence interval; VT: Ventricular tachycardia; VF: Ventricular fibrillation; TVR: Target vessel revascularization.

Sensitivity analysis

Sensitivity analysis using the leave-one-out approach did not impact the results of clinical outcomes and adverse events much (Supplementary Figures 8-20). Omitting Akbari et al[37] reduced heterogeneity to 0% from pooled analysis for TMP grade 3 (Supplementary Figure 21). Omitting Ilyas et al[4] reduced heterogeneity from pooled analysis for a completely settled ST segment (I² = 22%) (Supplementary Figure 22).

Publication bias assessments

Publication bias was detected in LVEF at 1 month (LFK index = -2.45), LVEF at 3 months (LFK index = -6.57), LVEF immediately after PCI (LFK index = -5.31), and TIMI grade 0 (LFK index = -2.22). TMP grade 0 (LFK index = -1.22), TIMI grade 2 (LFK index = -1.01), and TMP grade 1 (LFK index = -2) revealed minor asymmetry. No publication bias was observed for CTFC (LFK index = 0.18), LVEF at 1 week (LFK index = -0.65), TIMI grade 1 (LFK index = 0.8), and TMP grade 2 (LFK index = -0.04). Funnel plots showed no publication bias for LVEF at 6 months (Egger’s P value = 0.33), while there was publication bias for the non-settled or partially settled ST segment and TIMI grade 3 (Egger’s P value = 0.04) (Supplementary Figures 23-36).

DISCUSSION

The results of our systematic review and meta-analysis indicate that the use of nicorandil in patients with STEMI who underwent primary PCI was associated to the following positive effects: A higher rate of complete ST segment resolution and final TIMI 3 flow, a reduced infarct size at 6 months, and an improvement in LVEF. Importantly, this drug was associated with a reduced risk of MACE during follow up (Figure 8).

Figure 8 Graphical abstract. STEMI: ST-elevation myocardial infarction; PCI: Percutaneous coronary intervention; LVEF: Left ventricular ejection fraction; RR: Risk ratio; CI: Confidence interval; MACE: Major adverse cardiovascular event; MD: Mean difference.

Our findings are in line with the results of a previous study from Geng et al[7]. However, it is interesting to note that these results were not found significant for oral or combined intravenous and intracoronary routes and were only significant for intravenous and intracoronary routes upon subgroup analysis. The decreased efficacy of nicorandil when administered orally could be attributed to longer absorption times in comparison to intravenous or intracoronary routes of administration, especially in the setting of primary PCI[37]. Interestingly, in our meta-analysis, the occurrence of MACE significantly decreased even with peroral administration. A previous study concluded a non-significant difference in functional outcomes between oral nicorandil therapy vs control, further adding to the controversy regarding the efficacy of oral nicorandil therapy in patients undergoing PCI[39]. Furthermore, a combination of intravenous and intracoronary administration did not result in a significant increase in the occurrence of completely resolved ST segment elevation, but this can also be attributed to the lack of studies with this combination.

Our results revealed a non-significant difference in CTFC following nicorandil therapy in PCI patients in comparison to control. Geng et al[7] combined various parameters including TMPG, TIMI grade, CTFC, and myocardial contrast echocardiography to assess NR phenomenon and concluded a significant benefit of nicorandil therapy. Another meta-analysis which studied CTFC separately also revealed significantly reduced CTFC with nicorandil in comparison to placebo[40]. Our results contrast these findings. However, CTFC has been found to not accurately assess the degree of microvascular injury after PCI in patients with their first anterior wall acute myocardial infarction[41]. It has also found to not be a reliable predictor of MACE a month after PCI[42]. Even though CTFC is considered to provide a more objective assessment of reperfusion in comparison to TIMI grades[7], it is essential to consider these results in combination with other parameters of coronary reperfusion. One of these parameters include TIMI flow grade, which assesses epicardial flow and is stratified into four grades, with grade 0 indicating lack of anterograde flow and grade 3 indicating complete perfusion[43]. Our results indicate a significantly increased number of patients with TIMI grade 3 after with nicorandil therapy. A higher TIMI grade has been found to be positively associated with early and late survival[44]. Myocardial perfusion can be assessed with TMP, which has been similarly stratified into four grades[43]. Normal perfusion, as indicated by TMP grade 3, has been associated with a low risk of mortality[45]. A significantly higher number of patients were observed to have TMP grade 3 with nicorandil when compared to control in our meta-analysis.

We would also like to emphasize how the results of our systematic review and meta-analysis revealed an improvement in LVEF with nicorandil, and persisting for up to 6 months. Nicorandil also significantly reduced LV end-systolic volume index at 6 months in comparison to control. Ejection fraction has been found to be primarily dependent on end-systolic volume[46], which may be why changes in LV end-diastolic diameter and LVEDV index were non-significant despite the significant change in LVEF. Patients with a reduced LVEF beyond 3 months have been observed to have a higher risk for heart failure hospitalization and all-cause mortality, explaining the prognostic importance of this parameter[47]. A significant improvement with nicorandil supports its efficacy in patients undergoing primary PCI.

Clinically, the findings of this meta-analysis suggest that nicorandil may serve as a useful adjunct to primary PCI in selected patients with STEMI, particularly those at higher risk of microvascular dysfunction and the NR phenomenon. Improvements in ST-segment resolution, TIMI 3 flow, TMP grade 3, and LV systolic function indicate that nicorandil primarily exerts benefit at the level of myocardial perfusion rather than epicardial patency alone. In contemporary PCI practice, nicorandil could be considered as an adjunctive pharmacologic strategy in patients with large infarct burden, delayed presentation, or suboptimal reperfusion following initial balloon inflation or stent deployment. Notably, the observed benefits were more consistent with intravenous or intracoronary administration, supporting its benefit over delayed oral therapy. While these findings do not warrant routine incorporation into standard PCI protocols at present, they highlight a potential role for nicorandil in tailored, high-risk STEMI management pending further large, multicenter randomized trials.

Although our novel findings on the effects of nicorandil are of some importance, there are several limitations that must be recognised. First, all of the included studies were conducted in Asia, with the exception of Akbari et al[37] which was conducted in the Middle East. This limits the applicability of the results on other ethnicities. Second, the duration of therapy varied across studies, which might have been a source of heterogeneity. Third, the variation in the control group could also be a cause of heterogeneity in our results. To address this, we conducted subgroup analyses based on the mode of administration and performed sensitivity analyses. Last, we employed the DerSimonian and Laird method, and CIs were not adjusted with Knapp-Hartung adjustment, which may have resulted in narrow CIs[48].

CONCLUSION

The results of our systematic review and meta-analysis revealed significantly improved LV function, higher rate of completely resolved ST segment elevation, and higher incidences of normal epicardial and myocardial perfusion with nicorandil therapy in patients undergoing PCI following ST elevated myocardial infarction. Future research should focus on expanding the inclusion of various ethnicities as well as studying the impact of various routes of administration, especially peroral, on post-PCI clinical outcomes.