The superiority of high‐power short‐duration radiofrequency catheter ablation strategy for atrial fibrillation treatment: A systematic review and meta‐analysis study

Abstract Background Radiofrequency catheter ablation (RFCA) using the high‐power short duration (HPSD) results in better ablation lesion formation in the swine model. This systematic review and meta‐analysis purposed to investigate the safety and efficacy profile between HPSD and low‐power long‐duration (LPLD) ablation strategies to treat atrial fibrillation (AF) patients. Methods We completed the literature review after identifying the relevant articles comparing HPSD and LPLD ablation methods for AF recorded in ClinicalTrials.com, CENTRAL, PubMed, and ScienceDirect until February 2021. The overall effects were calculated using pooled risk ratio (RR) and mean difference (MD) for categorical and continuous data, respectively. We also estimated the 95% confidence interval (CI). Results The HPSD strategy took shorter procedure time (MD = −33.75 min; 95% CI = −44.54 to −22.97; P < .01), fluoroscopy time (MD = −5.73 min; 95% CI = −8.77 to −2.70; P < .001), and ablation time (MD = −17.71; 95% CI = −21.02 to −14.41) than LPLD strategy. The HPSD RFCA was correlated with lower risk of esophageal thermal injury (RR = 0.75; 95% CI = 0.59 to 0.94; P = .02). The HPSD method resulted in higher first‐pass pulmonary vein isolation (PVI) (RR = 1.36; 95% CI = 1.13 to 1.64; P < .01), lower PV reconnection (RR = 0.47; 95% CI = 0.34 to 0.64; P < .01), and lower recurrent AF (RR = 0.72; 95% CI = 0.54 to 0.96; P = .02) than LPLD strategy. Conclusion HPSD RFCA was superior to the conventional LPLD RFCA in terms of safety and efficacy in treating AF patients.


| INTRODUC TI ON
Compared with the optimal medical treatment (OMT), catheter ablation results in better atrial fibrillation (AF) outcomes. 1,2 Catheter ablation for pulmonary vein isolation (PVI) is recommended by the current guideline to restore the sinus rhythm in paroxysmal AF or persistent AF. 3 The sinus rhythm is successfully maintained in 60.8%-71% of AF patients following the catheter ablation procedure. 4 The complete PVI can be achieved through permanent, continuous, and transmural tissue damage using radiofrequency catheter ablation (RFCA). 5 However, several complications, such as pericardial effusion/tamponade, esophageal injury, vascular access complication, or pulmonary vein (PV) stenosis, can occur during RFCA procedure. 6,7 Inappropriate energy delivery might be the possible cause of the procedural complications and failure in sinus rhythm preservation.
RFCA induces thermal injury through resistive and conductive heating. The equilibrium between power and duration of radiofrequency (RF) delivery during resistive and conductive heating is a critical determinant for lesion generation. The resistive heating directly leads to permanent myocardial tissue damage with necrosis, whereas conductive heating spreads to the deeper tissue layers, leading to reversible damage in myocardial tissue. 5,[8][9][10][11] In daily clinical practice, the low-power long-duration (LPLD) ablation strategy is more commonly used. 12 That conventional method is correlated with longer RF application time, longer conduction heating, and deeper tissue heating. 5,[8][9][10][11] So, the risk of complications is predicted to be higher. The new approach called the high-power short-duration (HPSD) ablation strategy might be used to overcome those limitations. 5,9,13 In silico and animal studies demonstrated that catheter ablation using the HPSD approach resulted in shorter ablation time, better linear continuity, better lesion uniformity, and better lesion transmurality. 5,9 However, the safety and efficacy profile of HPSD and LPLD ablation strategies in humans is still unclear. Therefore, we conducted a systematic review and meta-analysis to investigate the safety and efficacy profile between HPSD and LPLD ablation strategies for AF treatment.

| ME THODS
This systematic review and meta-analysis were conducted based on preferred reporting items for systematic reviews and meta-analyses (PRISMA). 14

| Literature search
We searched for and identified the relevant studies comparing HPSD and LPLD ablation strategies for AF patients from the electronic scientific databases such as ClinicalTrials.com, CENTRAL, PubMed, and ScienceDirect. We applied the following keywords during the literature searching process: ("catheter ablation" OR "radiofrequency ablation" OR "RF ablation" OR "RFA" OR "radiofrequency catheter ablation" OR "RFCA" OR "ablation") AND ("high-power short-duration" OR "HPSD") AND ("low-power long-duration" OR "LPLD") AND ("atrial fibrillation" OR "AFib" OR "AF"). We completed the literature searching process in February 2021. Three investigators conducted the literature search.

| Eligibility criteria
We included the studies with the following criteria: (i) original research articles comparing HPSD and LPLD RFCA strategies for AF, (ii) the aim of RFCA was for rhythm control, (iii) article written in English, (iv) availability of the data about power and duration during RF delivery, and (v) availability of the detailed information about the treatment, procedural aspects, safety outcomes, and efficacy outcomes. We also excluded articles with the following criteria: (i) duplications, (ii) the full-text manuscript unavailability, (iii) the article used the data from similar studies, (iv) incomparable treatment group and control group, (v) ablation index (AI) guided catheter ablation, and (vi) outcomes of interest were not reported. The study selection process was performed by three investigators.

| Exposure and outcomes
The exposure was the RFCA method. Patients were classified into the "HPSD group" and "LPLD group." HPSD was defined as the catheter ablation performed using the highest Power ≥40 W and duration ≤10 seconds in any ablation or less than duration in the LPLD group. In comparison, LPLD was defined as the catheter ablation performed using the highest power <40 W and duration ≥10 seconds in any ablation or longer than duration in the HPSD group. The outcomes measured included: procedural aspects (procedure time,

| Study quality assessment and data extraction
All eligible randomized controlled trials (RCTs) and cohort studies comparing HPSD and LPLD ablation strategies for AF patients were involved in this study. The quality assessment of RCTs was performed using the modified Jadad scale, which ranged from 0 to 8. 15 A good-quality RCT is defined as an RCT with a modified Jadad score ranged from 4 to 8. 16 For cohort studies, study quality assessment was completed using the Newcastle-Ottawa scale (NOS). According to the NOS, a good quality cohort study was defined as a study with 3-4 stars in the selection area, 1-2 stars in the comparability area, and 2-3 stars in the outcome area. 17 To minimize the risk of bias in this systematic review and meta-analysis, we only involved high-quality studies. Two investigators conducted the study quality assessment.
The disagreement between both investigators was resolved through discussion and the second opinion of the third investigator.
The essential data about: (i) the name of the first author; (ii) publication date; (iii) design of the study; (iv) center involved; (v) number of patients; (vi) AF type; (vii) ablation strategy; (viii) HPSD ablation criteria; (ix) LPLD ablation criteria; (x) length follow-up period; (xi) arrhythmia detection method; (xii) demographic data (sex and age); (xiii) were obtained from each article. Three investigators performed the data extraction process. We reported the categorical data and continuous data using number (percentage) and mean ± standard deviation (SD), respectively. For continuous data, we also quantified mean ± SD from the median and interquartile range (IQR). 18-20

| Statistical analysis
The statistical analysis was completed based on the standard guideline. 21 Assessment of heterogeneity and potential publication bias was conducted before the conclusion determination. The Q-test was used to assess the heterogeneity. We used a cut-off point of 0.1 for P for heterogeneity. We used the random-effect analysis model in the presence of heterogeneity (P < .1). On the other hand, in the absence of heterogeneity (P ≥ .1), we used the fixed-effect analysis model. 22 We applied the combination of Begg's and Egger's tests to assess the presence of publication bias. The P-value of Begg's test and/or Egger's test <.05 indicated the presence of publication bias. 23 The pooled risk ratio (RR) and 95% CI for categorical data were calculated using the Mantel-Haenszel statistical method. The pooled mean difference (MD) and 95% CI for continuous data were determined using the inverse variance statistical method. A P-value of <.05 was considered significant statistically. 24 Both Review Manager

| Study quality and publication bias
Based on the assessment using the modified Jadad scale for RCTs (Table S1) and NOS for cohort studies (Table S2), we only included good-quality studies in our analysis. It was our effort to minimize the risk of bias. Moreover, we did not find any publication bias because no P-values of <.05 were obtained from Begg's and Egger's tests (Tables 3 and 4).

| Outcomes
The HPSD ablation strategy took shorter the procedure time (MD = We divided the efficacy outcomes into short-term and long-term efficacy outcomes. Short-term efficacy outcomes included first-pass PV isolation and PV reconnection. The HPSD ablation strategy was correlated with higher first-pass PV isolation (RR = 1.36; 95% CI = 1.13 to 1.64; P < .01) and lower PV reconnection (RR = 0.47; 95% CI = 0.34 to 0.64; P < .01) than LPLD ablation strategy ( Figure 4). The recurrent AF and recurrent AFL or AT were the long-term efficacy outcomes. Conducting RFCA using HPSD approach significantly reduced the risk of recurrent AF (RR = 0.72; 95% CI = 0.54 to 0.96; P = .02). However, the risk of recurrent of AFL or AT was not significantly different in both groups (RR = 1.14; 95% CI = 0.89 to 1.47; P = .30) ( Figure 5).

| D ISCUSS I ON
Several essential findings were obtained from this systematic review and meta-analysis study. First, conducting AF catheter ablation using the HPSD approach was more efficient than the LPLD approach due to shorter procedure time, fluoroscopy time, and ablation time. Second, compared with LPLD RFCA, the HPSD approach reduced the risk of ETI. Third, HPSD was associated with greater first-pass PVI. Fourth, the HPSD ablation method successfully reduced the risk of PV reconnection and recurrent AF following a single RFCA procedure.

| The role of ablation power and duration
An effective and efficient PVI can be achieved by: (1)    .87

<.01
Abbreviations: CI, confidence interval; HPSD, high-power short duration; LPLD, low-power long duration; MD, mean difference. on the balance between power and duration of RF application. 39 During resistive heating, a resistive component located close to the tip of the catheter causes energy dissipation and local heating. 38 Resistive heating achieves the maximum value within few seconds following RF energy delivery. During RFCA using the conventional LPLD method (power 25-30 W), the temperature rises above 50°C.

TA B L E 4 Summary of safety and efficacy outcomes
However, tissue necrosis occurs within a radius of 1-1.5 mm from the tip of the ablation catheter. 40 The higher power application can Their study revealed that the HPSD method generated more extensive lesions with similar depth and greater lesion-to-lesion uniformity. In HPSD RFCA, the heat generated during the resistive phase can affect the tissue until the depth of 3.5 to 4 mm. 5 It is suitable for atrial tissue because the maximum left atrial wall thickness is about 3.5 mm. 42
The radiofrequency application duration for each point in the LPLD group was <30 seconds. However, in the LPLD group, the maximum power used ranged from 30 to 40 W. The power applied in the posterior wall was also lower than the anterior wall. The radiofrequency application duration for each point in the HPSD group ranged from 3 to 42 seconds (Table 1). [25][26][27][28][29][30][31][32][33][34][35][36][37] To simplify the data analysis process, HPSD was defined as the RFCA performed using the highest power was supported the results from prior meta-analysis studies. 12,[46][47][48] From the safety aspects, our meta-analysis revealed that performing RFCA for AF using the HPSD approach could reduce the incidence of ETI. It was not similar to the results of the previous meta-analysis studies. 46,47 We added the results of the study from Wielandts et al. 35 and analyzed the results of the study from Castrejón-Castrejón et al. 27 using a different approach from the prior meta-analysis from Chen et al. 46 and Li et al. 47  to prove the net benefit of ETI risk reduction using HPSD RFCA approach ( Figure 3A). It was suggested that the advantage of the ETI risk reduction of the HPSD approach was more significant using the higher power. In the study from Castrejón-Castrejón et al., 27 the HPSD RFCA using the power of 60 W took shorter radiofrequency application time than HPSD RFCA using the power of 50 W (17 ± 5 min vs 24 ± 8 min; P < .01). 27 It seemed that the risk of ETI was more associated with the duration of RFCA, not the power.
Both groups did not show a significant difference for PNP, pericardial effusion, and cardiac tamponade. Those data were not reported in Our systematic review and meta-analysis revealed that the HPSD approach was a safe procedure. Several studies in animals and humans had confirmed the lower complication rate of the HPSD approach. 5,9,25,49 For the short-term efficacy outcomes, our study revealed that HPSD RFCA had better first-pass PVI than LPLD RFCA. This result was consistent with the findings from the previous metaanalysis. [46][47][48] Moreover, the HPSD approach was also associated with a lower risk of PV reconnection. The data about the risk of PV reconnection were provided by the meta-analysis study from Ravi et al. 48  The ability of the HPSD method in increasing first-pass PVI and reducing PV reconnection and recurrent AF was due to larger area, more uniform, and more consistent lesion generated by the HPSD method.
The stability of catheter-tissue contact is the essential part contributing to lesion formation. The instability of the ablation catheter in the beating heart can disrupt the RF energy delivery from the catheter to the tissue. 32 The reduction of ablation time can minimize catheter instability and perhaps improve lesion generation by increasing the possibility of maintaining catheter stability during the RF energy application. 50 The catheter instability was a problem in the LPLD approach that leads to various lesions, tissue edema, wider tissue damage, lower rate first-pass PVI, and higher rate of PV reconnection. The perfect PVI with transmural scar formation is an essential factor in ensuring freedom from recurrent AF. 51,52 Therefore, the HPSD RFCA can provide permanent lesions with better continuity and transmurality.
Theoretically, the relatively shorter RF application duration and consequently smaller RF energy delivery in the LPLD RFCA can reduce the first-pass PVI, increase the PV reconnection, and increased AF recurrence. However, to the best of our knowledge, no specific study directly compared long duration versus short duration of RF energy application using similar low power (<40 W) in both groups. Moreover, the research that compares high power versus low power RFCA using equal duration is still not available. Our study also showed that the HPSD method could not reduce the risk of recurrent AFL or AT. This result could be caused by several factors: (i) different arrhythmogenic mechanisms, (ii) some patients received not only PVI, and (iii) AFL or AT could be caused by scar formation due to RFCA. 51,53

| Study limitations
We recognized several limitations in this systematic review and metaanalysis. First, most studies involved in this study were cohort studies, causing unwanted selection bias and referral bias. However, we overcame this situation by including only high-quality studies. Second, the publication bias possibility could not be avoided. To minimize this situation, we performed a double publication bias evaluation using Begg's and Egger's tests. We did not find any publication bias in this metaanalysis. Third, the settings of power and duration in the involved studies were not uniform. Fourth, the variation in the follow-up period duration and arrhythmia detection methods could be confounding factors. Fifth, the inability to get data at individual patient-level limited our effort to determine the real effects at the patient level.

| CON CLUS ION
Our systematic review and meta-analysis study revealed that the HPSD RFCA was a safe, effective, and efficient procedure to treat AF. The superiority offered by the HPSD method over the conventional LPLD strategy method: (i) shorter procedure time, fluoroscopy time, and ablation time; (ii) lower risk of ETI; (iii) higher first-pass PVI; and (iv) lower risk of PV reconnection and recurrent AF. However, the universal definition of HPSD RFCA was not available. Our findings suggested that the RCT with a large number of patients, better design, more clear HPSD definition, longer follow-up duration, and more appropriate arrhythmia detection methods should be conducted.

ACK N OWLED G EM ENTS
We wish to thank Faculty of Medicine, Universitas Brawijaya, Malang, Indonesia.