2025 年 7 巻 11 号 p. 1029-1036
Background: Several antiarrhythmic agents, including amiodarone (AMD) and nifekalant (NIF), are used in the management of life-threatening arrhythmias such as ventricular fibrillation (VF) and ventricular tachycardia (VT). Although a few studies have compared the antiarrhythmic effect of NIF and AMD, most have focused on termination or defibrillation of VF/VT. In this meta-analysis, we aimed to assess the current evidence on the preventive effects of NIF and AMD on recurrent VF and electrical storm (ES).
Methods and Results: The protocol was registered in PROSPERO (CRD42024578486). PubMed (MEDLINE and PMC), Cochrane Central Register of Controlled Trials (CENTRAL), Science Citation Index Expanded, and Igaku Chuo Zasshi were searched for observational studies and randomized controlled trials evaluating patients with cardiac arrest due to recurrent VF and ES treated with AMD or NIF, published before 28 August 2024. The primary endpoint was the recurrence of VF/VT. Secondary outcomes included short-term death and incidence of torsades de pointes (TdP). 5 eligible studies comprising 222 participants were included. Evidence regarding differences between NIF and AMD in the recurrence of VF/VT, short-term death, and incidence of TdP was of very low certainty and remains highly uncertain.
Conclusions: NIF and AMD may have little to no difference in preventing the recurrence of VF/VT, short-term death, or incidence of TdP; however, the certainty of the evidence is very low.

Several drugs, including amiodarone (AMD) and nifekalant (NIF), are effective for managing fatal arrhythmias such as ventricular fibrillation (VF) and ventricular tachycardia (VT).1–6 Although AMD is recommended as the first-line agent for VF/VT according to the American Heart Association guidelines, NIF has been widely used for VF/VT for over 2 decades, especially in Japan.7,8 NIF exerts a rapid effect through blockade of potassium channels, particularly the rapid delayed rectifier potassium current (IKr), whereas AMD acts on potassium, sodium and calcium channels as well as β-receptors, exhibiting potent antiarrhythmic activity.9,10 Given their distinct pharmacological profiles, the efficacy and adverse events associated with these agents may be different. Although several studies have examined the effect of NIF and AMD, most have focused on their roles in termination and defibrillation of VF/VT.11–13 In this meta-analysis, we evaluated the preventive effects of NIF and AMD on recurrent VF and electrical storm (ES).
The Japan Resuscitation Council (JRC) Arrhythmia Task Force for the JRC guidelines 2025 was formed by the Japanese Circulation Society and Japanese Society of Internal Medicine. The JRC Arrhythmia Task Force developed clinically relevant questions. Following discussions with the Guidelines Editorial Committee, the Population, Intervention, Comparator, Outcome, Study design, and Time frame (PICOST) approach was adopted to guide the systematic review:
P (Population): Patients (≥18 years) with recurrent VF and ES, including those with implantable cardioverter defibrillators (ICD)
I (Intervention): NIF
C (Comparator): AMD
O (Outcomes): Recurrence of VF or VT, short-term death, TdP
S (Study design): Observational studies and randomized trials; case series involving ≥5 patients were included; experimental studies were excluded
T (Time frame): Literature published up to 28 August 2024
This systematic review and meta-analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, registered in the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42024578486), and carried out according to PRISMA methodology.14,15
Search StrategiesSearch strategies were developed using medical subject headings and keywords related to cardiac arrest with recurrent VF and ES. A systematic search was performed across PubMed (MEDLINE and PMC), Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Library, Science Citation Index Expanded (SCIE) via the Web of Science, and Japana Centra Revuo Medicina (Igaku Chuo Zasshi) via the Ichushi-Web to retrieve relevant articles published from inception to 28 August 2024 (Supplementary Appendix). Only full-text human studies published before 28 August 2024 were included.
Study Selection and Inclusion CriteriaEligible studies included adult patients with cardiac arrest, recurrent VF, or ES. Studies were not restricted by country. However, only publications in English were included. Observational studies and randomized trials were eligible, as were case series involving ≥5 patients. Experimental studies were excluded.
Data Extraction and ManagementThe following data were extracted: authors, article title, journal name, year of publication, website URL, and abstract. After removing duplicates, 2 independent reviewers (K.S. and K.I.) screened titles and abstracts, followed by full-text assessments. Discrepancies were resolved through discussion until consensus was reached. Full-text articles selected for inclusion were independently reviewed by both reviewers. Disagreements were resolved by a third independent reviewer (T.F.).
Assessment of the Risk of BiasThe 2 reviewers (K.S. and K.I.) independently assessed the risk of bias in all included studies using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool.16 Studies were evaluated across 7 domains and judged as having a low, moderate, serious, or critical risk of bias, or classified as lacking sufficient information. Disagreements regarding the risk of bias were resolved by discussion. Studies determined to have a critical risk of bias were excluded from the meta-analysis.
Statistical AnalysisResults were synthesized using a random-effects model to pool treatment effect estimates. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for dichotomous outcomes. Heterogeneity across studies was evaluated using the I2 statistic. Inconsistency was considered significant if the I2 value was ≥50% and no clear explanation for heterogeneity was identified.17
Funnel plots were constructed to assess potential publication bias. All available and published data were included. Analyses were conducted using Review Manager software (RevMan v.5.4).
Rating Certainty of EvidenceThe Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework was used to assess the certainty of the evidence regarding the effect of NIF and AMD on recurrent VF and ES.18–20 Certainty was rated as high, moderate, low, or very low based on risk of bias, inconsistency, indirectness, imprecision, and publication bias.
Figure 1 presents the study selection flow diagram, adapted from the PRISMA 2020 statement.21

PRISMA flow diagram depicts the identification, screening, eligibility, and inclusion process for studies in this systematic review.
The database search across PubMed (MEDLINE and PMC), CENTRAL, SCIE, and Igaku Chuo Zasshi identified 3,090 records. After removing 688 duplicates and excluding 2,397 studies that did not meet the inclusion criteria, 5 articles were retained for final analysis.22–26
Study CharacteristicsTable 1 summarizes the characteristics of the included studies. All studies were prospective or retrospective observational investigations conducted between 2010 and 2016. The arrhythmias assessed were ES in the studies by Mera and Miyauchi; VF/VT in those by Sasaki and Suzuki; and refractory VF/VT in the study by Yamazaki. The number of patients in the NIF groups ranged from 13 to 33, and in the AMD groups from 8 to 41, totaling 122 patients treated with NIF and 100 with AMD. Three studies reported the etiology of heart disease, patient age, and sex. Among these, 50–70% of patients had ischemic heart disease, the mean age was approximately 70 years, and the majority were men.
Summary of Included Studies
| Author | Year | Study design |
Types of arrhythmias |
Sample size (NIF/AMD) |
IHD (%) (NIF/AMD) |
Age (years) (NIF/AMD) |
Men (%) (NIF/AMD) |
Recurrence of VF/VT |
Short-term death* |
Torsades de pointes |
|---|---|---|---|---|---|---|---|---|---|---|
| Mera et al.24 |
2012 | Retrospective cohort |
ES | 33/41 | –/– | –/– | –/– | ○ | ○ | ○ |
| Yamazaki et al.22 |
2010 | Retrospective cohort |
Refractory VF/VT |
29/11 | 73/64 | 72/71 | 66/55 | ○ | ○ | – |
| Sasaki et al.25 |
2010 | Retrospective cohort |
VF/VT | 23/19 | 52/48 | 66/69 | 70/63 | ○ | – | ○ |
| Miyauchi et al.23 |
2010 | Retrospective cohort |
ES | 13/8 | –/– | –/– | –/– | ○ | – | – |
| Suzuki et al.26 |
2016 | Prospective cohort |
VF/VT | 24/21 | 58/71 | 69/67 | 88/71 | ○ | ○ | – |
*Definition of short-term death: Mera et al., Yamazaki et al. and Suzuki et al. evaluated acute death, 1-month death and in-hospital death as a short-term death, respectively. AMD, amiodarone; ES, electrical storm; IHD, ischemic heart disease; NIF, nifekalant; VA, ventricular arrhythmia; VF, ventricular fibrillation; VT, ventricular tachycardia.
The primary outcome of the recurrence of VF/VT was assessed in all 5 studies. In addition, Mera et al.,24 Yamazaki et al.,22 and Suzuki et al.26 reported short-term death outcomes, defined respectively as acute death, 1-month death, and in-hospital death. Only the studies by Mera et al.24 and Sasaki et al.25 provided data on the incidence of TdP.
Risk of BiasThe risk of bias for each study is summarized in Figure 2. All studies were judged to have a “serious” risk of bias due to confounding, while the risk related to participant selection was “low.” In the studies by Miyauchi,23 Sasaki et al.,25 and Yamazaki et al.,22 classification of interventions was rated as “serious” due to treatment decisions being based on patient condition. The risk of bias due to deviations from intended intervention was “moderate” in Suzuki’s study and “serious” in the other 4. Risks related to missing data, outcome measurement, and selective reporting ranged from “low” to “moderate.” The overall risk of bias was judged as “serious” in all 5 studies. Visual inspection of funnel plots did not suggest publication bias (Figure 3).

Risks of bias summary. The overall judgment of the risk of bias in five studies were all “serious”.

Visual inspection of the funnel plot suggested no existence of publication bias.
Analyses of Outcomes
The primary outcome, the recurrence of VF/VT, was reported in all 5studies (Figure 4A). The recurrence of VF/VT occurred in 37 of 122 patients (30.3%) in the NIF group, and in 21 of 100 patients (21.0%) in the AMD group. The meta-analysis yielded an OR of 1.33 (95% CI, 0.69–2.59) for NIF compared to AMD. No statistical heterogeneity was observed among the included studies (I2=0%). Short-term death was reported in 3 studies and the pooled OR was 1.63 (95% CI, 0.58–4.54) in favor of AMD (Figure 4B). There was low statistical heterogeneity across the studies (I2=27%). Two studies reported the incidence of TdP and the pooled analysis produced an OR of 5.25 (95% CI, 0.58–47.69) for NIF compared to AMD (Figure 4C). There was no statistical heterogeneity was detected between the 2 analyzed studies (I2=0%).

Forest plots of (A) recurrence of VF/VT, (B) short-term death, and (C) torsades de pointes. CI, confidence interval.
Certainty of Evidence
The certainty of evidence for each outcome is presented in Table 2. For all outcomes (i.e., the recurrence of VF/VT, short-term death, and incidence of TdP) the certainty was rated as “very low.”
Evidence Profile
| No. of studies | Certainty assessment | No. of patients | Effect | Certainty | Importance | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Study design | Risk of bias | Inconsistency | Indirectness | Imprecision | Other considerations |
NIF | AMD | Relative (95% CI) |
Absolute (95% CI) |
|||
| Recurrence of VF/VT | ||||||||||||
| 5 | Non-randomized studies |
Very serious | Serious | Not serious | Serious | None | 37/122 (30.3%) |
21/100 (21.0%) |
OR 1.33 (0.69 to 2.59) |
93 more per 1,000 (from 21 fewer to 207 more) |
⊕○○○ Very low |
CRITICAL |
| Short-term death | ||||||||||||
| 3 | Non-randomized studies |
Very serious | Serious | Not serious | Serious | None | 24/86 (27.9%) |
12/73 (16.4%) |
OR 1.63 (0.58 to 4.54) |
115 more per 1,000 (from 13 fewer to 242 more) |
⊕○○○ Very low |
CRITICAL |
| Torsades de pointes | ||||||||||||
| 2 | Non-randomized studies |
Very serious | Not serious | Not serious | Serious | None | 4/56 (7.1%) |
0/60 (0.0%) |
OR 5.25 (0.58 to 47.69) |
71 more per 1,000 (from 3 fewer to 144 more) |
⊕○○○ Very low |
IMPORTANT |
CI, confidence interval; OR, odds ratio; NIF, nifekalant. Other abbreviations as in Table 1.
To our knowledge, this is the first systematic review comparing the preventive efficacy and adverse effects of NIF and AMD in patients with VF/VT. In this meta-analysis, we found little to no difference between NIF and AMD in terms of preventing the recurrence of VF/VT, short-term death, and incidence of TdP, although the certainty of the evidence remains very low.
Suppression of the Recurrence of VF/VTAll 5 included studies reported on the suppressive effect on VF/VT recurrence. Among them, Sasaki et al.25 and Mera et al.24 reported a favorable trend for NIF, while the other 3 studies favored AMD. Overall, no significant difference was observed in the preventive effect of NIF vs. AMD on the recurrence of VF/VT. Although previous studies have compared NIF and AMD in terms of pharmacologic defibrillation success or facilitation of electrical defibrillation, only 1 prior systematic review published in 2014 assessed their prophylactic effects on resuscitation outcomes.27–29 In our systematic review, the number of patients analyzed increased to 220, and we added 1 article published after the 2014 systematic review.
Impact on Short-Term DeathShort-term death was analyzed in 3 studies. Although each showed a trend toward lower mortality rates with AMD, the meta-analysis indicated that NIF and AMD may result in little to no difference in short-term deaths, with substantial uncertainty. Definitions of short-term death varied: death within 1 month (Yamazaki et al.22), in-hospital death (Suzuki et al.26), and death in the coronary care unit (Mera et al.24). Despite differences in terminology, these outcomes all reflect acute death following hospitalization and were deemed appropriate for pooled analysis.
Appearance of TdPBoth NIF and AMD prolong the QT interval by inhibiting the rapid delayed IKr, which may lead to TdP. Although AMD has a lower reported risk of TdP than other pure IKr blockers such as NIF or dofetilide, the pathogenesis of TdP is multifactorial. Early afterdepolarizations (EADs) and increased transmural dispersion of repolarization (TDR) are considered central to its development. AMD’s activity on multiple ion channels (Na, Ca) and its β-adrenergic blocking properties may contribute to reduced EADs and TDR, thereby lowering TdP risk.30 In contrast, NIF, being a selective IKr blocker, may cause marked QT prolongation and, in some cases, increased TDR, thereby elevating TdP risk.31 However, only 2 studies in this review reported TdP outcomes, and no meta-analysis on QT duration was feasible. The complex interplay between QT prolongation and TdP, influenced by multiple patient-specific and drug-specific factors, limits definitive conclusions.
Clinical Use of NIF and AMDGiven the very low certainty of evidence, no definitive conclusion can be drawn regarding the superiority of AMD or NIF in suppressing VF/VT. This uncertainty does not preclude the use of either drug as recommended in current guidelines.2 NIF is associated with QT prolongation, whereas AMD may induce bradycardia or hypotension due to its broader pharmacologic effects. Therefore, individual risk profiles must be considered when selecting therapy depending on the patient’s clinical presentation, such as in cases of hereditary arrhythmias or in patients with ICD.32–37 AMD dosing is well established and can generally be managed through monitoring of heart rate and blood pressure. NIF, used in continuous infusion for VF/VT suppression, requires careful QT interval monitoring, with recommended dosages ranging from 0.1 to 0.4 mg/kg/h. Evidence from a prior study suggests that lower doses of NIF (e.g., 0.2 mg/kg/h) may be both effective and safe in patients with acute coronary syndromes.38 Effective and safe use of either agent requires an understanding of their distinct pharmacologic properties and patient-specific considerations.
Study LimitationsThis systematic review is limited by the retrospective design of all included studies and the overall “serious” risk of bias. Most data were derived from Japanese cohorts, in which NIF is more commonly used, limiting the generalizability to other populations. Additionally, all included studies were published over 10 years ago. Advances in critical care, such as the introduction of mechanical circulatory support (e.g., IMPELLA), may influence contemporary outcomes.39 Updated clinical data under current intensive care practices are needed.
This systematic review suggests that NIF and AMD may have minimal differences in their effects on the recurrence of VF/VT, short-term death, and TdP risk, although the evidence is highly uncertain. Robust, adequately powered randomized controlled trials are urgently needed to provide higher-quality evidence to guide clinical decision-making.
The authors thank Mr. Shunya Suzuki and Ms. Tomoko Nagaoka, librarians at Dokkyo Medical University, Tochigi, Japan, for their assistance in searching for articles.
Not applicable.
T.M. is a member of Editorial Team for Circulation Reports’. The other authors declare no conflicts of interest with regard to this article.
This work was supported by the Japan Resuscitation Council, Japan Circulation Society, and JSPS KAKENHI (Grant Number JP23K08454).
All data generated or analyzed during this study are included in this published article.
Please find supplementary file(s);
https://doi.org/10.1253/circrep.CR-25-0138