Septal Flash as a Predictor of Cardiac Resynchronization Therapy Response: A Systematic Review and Meta-Analysis.

Cardiac resynchronization therapy (CRT) in heart failure patients has been shown to improve patient outcomes in some but not all patients. A few studies have identified that septal flash on imaging is associated with response to CRT, but there has yet to be systematic review to evaluate the consistency of the finding across the literature. A search of MEDLINE and EMBASE was conducted to identify studies, which evaluate septal flash and its association with CRT response. Studies that met the inclusion criteria were statistically pooled with random-effects meta-analysis and heterogeneity was assessed using the I2 statistic. A total of nine studies were included with 2307 participants (mean age 76 years, 67% male). Septal flash on imaging before CRT implantation was seen in 53% of patients and the proportion of CRT responders from the included studies varied from 52% to 77%. In patients who were CRT responders, septal flash was seen in 40% of patients compared to 10% in those deemed to be CRT nonresponders. Meta-analysis of eight of the nine included studies suggests that the presence of septal flash at preimplant was associated with an increased likelihood of CRT response (relative risk 2.55 95% confidence interval 2.04-3.19, P < 0.001, I2 = 51%). Septal flash was also reported to be associated with left ventricular reverse remodeling, but the association with survival and symptomatic improvement was less clear. Septal flash is a well-defined and distinctive contraction pattern that is easily recognizable on cardiac imaging. Septal flash may be associated with CRT response and should be evaluated in the patients that are considered for CRT devices. Copyright:

INTRODUCTION

Heart failure (HF) affects approximately 1%–2% of the adult population, rising to over 10% in those aged 70 years and above.[1] Despite significant advances in drugs and therapeutics, HF continues to be associated with poor quality of life,[2] and patient outcomes.[3] In patients with HF and cardiac dyssynchrony, cardiac resynchronization therapy (CRT) is designed to restore synchronous electromechanical contraction, resulting in improved cardiac output and favorable left ventricular (LV) remodeling.[456] Data from large-scale randomized studies has consistently demonstrated that, in patients with a broad left bundle branch block (LBBB), conduction pattern and severe LV systolic dysfunction, CRT improves hard outcomes, including mortality and HF hospitalization.[56] Furthermore, CRT is associated with improved exercise capacity, symptoms, and quality of life.[7] However, it is also recognized that some patients fail to respond to CRT (nonresponders) whilst others exhibit negative-response to CRT.[8] The a priori identification of patients who are likely to be responders would enable superior patient selection criteria. However, to date, this remains an area of primary research. It has been proposed that echocardiographic assessment of ventricular/mechanical dyssynchrony before CRT could improve patient selection and identify responders to CRT irrespective of LV ejection fraction (LVEF) or QRS duration.[91011]

Septal flash, as first described by Dillon et al. in 1974,[12] is a sign of intra-ventricular dyssynchrony, easily recognizable on cardiac imaging including transthoracic echocardiography (TTE) and, to a lesser extent, cardiac magnetic resonance (CMR) imaging. It is defined as an abnormal and rapid leftward motion of the interventricular septum during the isovolumetric ventricular contraction phase [Figure 1]. Several studies have identified the presence of septal flash which, when present before CRT implantation, can predict both reverse LV remodeling,[13] improved in symptomatic status,[14] and improved survival rates.[15]

Figure 1

Echocardiography M-mode image displaying septal flash. M-mode image of the parasternal short-axis view displaying septal flash (*)

To the best of our knowledge, there was no previous systematic review to determine if the association between septal flash and CRT response is consistent across the literature and the strength of the association. Therefore, we conducted a systematic review and meta-analysis of septal flash and response after CRT implantation.

METHODS

This systematic review and meta-analysis were conducted in accordance with the PRISMA guidelines.

Eligibility criteria

We selected studies that evaluated septal flash and the response rate to CRT. There were no restrictions on how septal flash was defined with all imaging modalities being included. There was no restriction based on patient age, presence of congenital heart disease, study design, or language. In addition, the studies had to include an end-point that was an indicator of response to CRT which included changes in LV volume or LV reverse remodeling or an evaluation of symptoms or hospitalizations. We excluded studies identified as conference abstracts, editorials, and articles without original data including reviews, case studies, and case series. Animal studies were also excluded.

Search strategy

We searched MEDLINE and EMBASE using OVID with no date or language restriction on October 26, 2020. The exact search terms were: “cardiac dyssynchrony,” “septal flash” and “septal dyssynchrony.” We reviewed the bibliography of relevant studies and reviews for additional studies that met the inclusion criteria.

Study selection and data extraction

Two reviewers (SB and JT) independently screened all titles and abstracts retrieved from the search for studies. The studies that potentially met the inclusion criteria were reviewed and the final decision to include or exclude studies was made by consensus. The data extraction was carried out by SB and checked by CSK and JT independently. Data were collected on study design, country of study origin, year, sample size, mean age, percentage of male participants, and inclusion criteria. The imaging modality used, definition and assessment of septal flash, and definition of CRT response were also extracted. Follow-up duration, pertinent results from each study along with an overview of the usefulness of septal flash in predicting CRT response was collated also.

Risk of bias assessment

Methodological quality assessment of the included studies was based on the risk of bias assessment with consideration of the following: (i) prospective study, (ii) reliable ascertainment of septal flash, (iii) reliable outcome ascertainment, (iv) low missing data/loss to follow up, (v) adjustments for confounders in the estimate for septal flash and response in CRT, and (vi) generalizability of the cohort. This assessment was done by one reviewer (SB) and checked independently by two other reviewers (CSK and JT).

Data analysis

Data were extracted into predesigned tables which collected data on study design, the country where it took place, the year when the study took place, sample size, mean age, percentage of the cohort that was male, percentage of the cohort that had ischemic cardiomyopathy and inclusion criteria. We also collected data on the definition for septal flash for each study and the results including CRT response and its definitions. Using the inverse variance method, random-effects meta-analysis of septal flash and response in patients with CRT was undertaken on Review Manager (Version 5.3, The Cochrane Collaboration, 2014). Where available the most adjusted risk or odds estimates were collected and pooled in meta-analysis. For studies which only reported crude results, we calculated the risk ratio using the Mantel–Haenszel method in a random-effects model. Statistical heterogeneity was considered with the I statistic where 30%–60% represents a moderate level of heterogeneity[16] and leave-one-out sensitivity analysis was performed to evaluate the source of statistical heterogeneity. Publication bias was assessed by asymmetry testing with funnel plots if the dataset contains more than 10 studies and there is no evidence of significant heterogeneity.[17]

RESULTS

Study selection and description of included studies

Our search yielded 543 studies and after detailed screening and review, a total of nine studies were included [Figure 2].[131415181920212223]

Figure 2

Flow diagram to illustrate the study selection process and the reason for study exclusion

The description of the included studies is shown in Table 1. There were four retrospective studies and four prospective studies while one study was of unclear design. Overall, there were 2307 participants, and the average age across the eight studies which reported mean age was 76 years. The proportion of male patients in the included studies as 67% and 42% of all patients had ischemic HF. Five studies reported percentage of LBBB among participants, the average was 77%. The mean QRS duration across all studies ranged from 155 to 170 ms.

Table 1

Study design, patient characteristics, and inclusion criteria

Study ID	Study design; Country; year	Sample size	Age	Male (%)	Ischemic etiology (%)	Left bundle branch block at baseline (%)	QRS duration (ms)	Patient inclusion criteria
Bernard 2017	Retrospective cohort study; France; 2016	275	67	70	65	73	161±26	Patients with HF requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms)
Brunet- Bernard 2014	Retrospective cohort study; France; 2010-2012	207	66	70	35	69	161±25	Patients with HF requiring a CRT device (LVEF ≤35%, QRS duration ≥120msecs), with adequate ultrasound acoustic windows
Doltra 2014	Retrospective cohort study; Spain; 2005-2009	200	67.3	77	40	70	169±30	Patients with HF requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms), who were not awaiting heart transplantation or without significant comorbidity that would shorten life expectancy
Gabrielli 2014	Prospective cohort study; Spain; 2013	94	69	79	28	85	166±35	Patients with HF and long-standing atrial fibrillation (≥3 months) who required a CRT device according to international guidelines (LVEF≤35%, QRS duration ≥120 ms)
Gasior 2016	Prospective cohort study; Poland; 2009-2012	133	63	77	51	Not stated	165±25	Patients enrolled in the ViaCRT study and who met standard indications for a CRT device (LVEF ≤35%, QRS duration ≥120 ms)
Mada 2016	Retrospective cohort study; Belgium and Germany; unclear	125	63	71	54	Not stated	170±26	Patients with HF requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms) with good echocardiographic windows
Parsai 2009	Prospective cohort study; France and Netherlands; 2008	161	66	80	51	Not stated	156±21	Patients with HF requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms)
Sohal 2014	Prospective cohort study; UK; 2014	52	65	88	48	Not stated	155±24	Patients with HF requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms)
Stankovic 2016	Retrospective cohort study; Europe; 1999-2012	1060	64	76	43	88	170±29	HF patients enrolled in predict-CRT study requiring a CRT device according to international guidelines (LVEF ≤35%, QRS duration ≥120 ms)

HF=Heart failure, CRT=Cardiac resynchronization therapy, LV: Left ventricular, LVEF=LV ejection fraction

The definitions for septal flash used in each study are presented in Table 2. All studies used either two-dimensional (2D or 3D) TTE to assess for the presence of septal flash.[1314151819202122] One study used a combination of 2D TTE and CMR.[23] Details results extracted from each of the included studies is shown in Table 3.

Table 2

Definition of septal flash and definition of cardiac resynchronization therapy response

Study ID	Imaging modality used	Definition of SF	Assessment of SF	Definition of CRT response
Bernard 2017	2D echocardiography	Early septal thickening/thinning within the isovolumetric contraction period	Visual and m-mode assessment	Reduction of >15% in LVESV at 6-months follow-up and/or occurrence of major cardiovascular event
Brunet- Bernard 2014	2D echocardiography	Leftward displacement of the interventricular septum during preejection, followed by the rightward (paradoxical) motion	Visual and m-mode assessment	Not clear
Doltra 2014	2D echocardiography	Fast contraction and relaxation (inward/outward motion) of the septum occurring during the isovolumetric contraction period (within the QRS width)	Visual and m-mode assessment	Reduction of ≥15% in LVESV at follow-up on echocardiogram. Clinically defined as: alive, without heart transplantation, and with the improvement of ≥20% in the 6-min walk test or of at least 1 NYHA functional class
Gabrielli 2014	2D echocardiography	Early fast inward/outward motion of the interventricular septum within the isovolumetric contraction period and the QRS complex duration. The presence of SF was confirmed with an excursion >1 mm and corrected SF at 12-months follow-up was defined by the disappearance of SF or a reduction of the excursion ≥3 mm	Visual or color tissue Doppler m-mode assessment	Reduction of ≥15% in LVESV at follow-up
Gasior 2016	2D echocardiography	Not stated	Not stated	Not clear
Mada 2016	2D and 3D echocardiography	Not stated	Visual assessment	Reduction of ≥15% in LVESV
Parsai 2009	2D echocardiography	The presence of an early septal thickening/thinning within the isovolumetric contraction period	Visual or color tissue Doppler m-mode assessment	Reduction of >10% in LVESV. Reduction in NYHA class by ≥1 class
Sohal 2014	MRI and 2D echocardiography	MRI definition: inward septal motion within the first 15% of the cycle measured by the software platform Echo definition: Early inward and outward motion within the isovolumetric contraction period	MRI: Time volume curves and visual assessment. Echocardiography: visual and m-mode assessment	Changes in NYHA, 6-min walk distance, and minnesota living questionnaire
Stankovic 2016	2D echocardiography	Short inward motion of the septum in early systole (within the QRS width)	Visual assessment	Reduction of ≥15% in LVESV at follow-up

2D=2-dimensional, 3D=3-dimensional, CRT=Cardiac resynchronization therapy, LVESV=Left ventricular end-systolic volumes, MRI=Magnetic resonance imaging, NYHA=New York Heart Association, SF: Septal Flash

Table 3

Results of included studies

Study ID	Follow-up	Results	Septal flash a useful parameter in CRT response?
Bernard 2017	Median 720 days	CRT response rate: 172/275 (62.5%) Responders versus nonresponders  Septal flash: 140/172 (84%) versus 43/97 (46%), P<0.001  LVEF: 31±10 versus 28±8, P=0.005.  Major cardiovascular events: 21/275 (8%)  Cardiovascular mortality: 16/275 (6%)	Yes, septal flash is associated with CRT response
Brunet-Bernard 2014	6 months	Responders versus nonresponders  Derivation cohort: 76/98 (78%) versus 28/64 (44%), P<0.0001  Validation cohort: 24/32 (77%) versus 4/13 (31%)  Combined: 100/130 (77%) versus 32/77 (42%)  Multivariate predictor of good response with septal flash: OR 5.16 95% CI 2.24-11.86, P=0.0001. LVEF: Derivation cohort 27±6 versus 27±6, P=0.67  Validation cohort: 27±3 versus 24±3, P=0.003	Yes, septal flash is associated with CRT response
Doltra 2014	12 months	Septal flash: 106/200 (53%) Echo response at 12 months: Septal flash 85/106 (80%) The proportion with septal flash  Death or no clinical response and LVESV reduction: <6.8%: 5/37 (13.5%)  Clinical response and LVESV reduction: <6.8%: 12/40 (30.0%)  LVESV reduction of 6.8%-26.68%: 37/62 (59.7%)  LVESV reduction >26.68%: 52/61 (85.2%)  Septal-to-lateral delay and echocardiographic response: Univariate OR 1.00 95% CI 0.99-1.01, P=0.583  Septal-to-lateral delay and cardiovascular mortality: Univariate OR 1.00 95% CI 0.99-1.01, P=0.563	Yes, septal flash is associated with CRT response
Gabrielli 2014	12 months	CRT responders: 56/94 (59%) Responders versus nonresponders  Septal flash: 74.1% (41/56) versus 25.9% (10/38), P<0.01 significant reverse remodeling and response with septal flash: Multivariate OR 5.24 95% CI 1.95-14.11, P<0.01  LVEF: 24±6 versus 29±10, P=0.04	Yes, septal flash was associated with CRT response
Gasior 2016	12 months	Septal flash: 24/87 Septal flash (n=24) versus no septal flash (n=63) ΔLVEDV: −46±10 versus −32±7, P<0.05 ΔLVESV: −63±10 versus −36±6, P<0.05 ΔLVEF: 14±2 versus 8±1, P<0.05 ΔNYHA: −1.1 versus −0.8, P=0.06	No, unable to comment on septal flash and its associated with CRT response
Mada 2016	6 months	CRT responders: 77/125 (61.6%) Septal flash: 83/125 (66.4%) Responders versus nonresponders  Septal flash: 67/77 (87.0%) versus 16/48 (33.3%), P<0.001  LVEF at baseline: 24±6 versus 27±7, P=0.07	Yes, septal flash was associated with CRT response
Parsai 2009	1,3 and 6 months	CRT responders: 121/161 (75%) Septal flash: 84/161 (54%) Responders versus nonresponders  Septal flash: 77/121 (64%) versus 10/40 (25%)  Septal flash in clinical responders 77/121, volume responders 73/92, and nonresponders 22/40  Septal flash sensitivity 64% 95% CI 54-72%, specificity 55% 95% CI 39-70%, PPV 81% 95% CI 71-88%, NPV 33% 95% CI 23-46%	Yes, septal flash was associated with CRT response
Sohal 2014	6 months	CRT responders: 27/52 (51.9%) Septal flash: 24/52 (46.1%) Responders versus nonresponders  Echo septal flash: 21/27 (78%) versus 5/20 (20%), P=0.001  CMR septal flash: 21/27 versus 3/25, P=0.001  Degree of CMR septal flash: 18.5±7.7 versus 8.9±2.7, P=0.003  LVEF with septal flash versus no flash: 33.9±8.9 versus 30.0±10.2, P=0.18  Septal flash and reverse remodeling: Sensitivity 78% 95% CI 67-89%, specificity 88% 95% CI 79-97%  No scar and septal flash and reverse remodeling: Sensitivity 52% 95% CI 38-66%, specificity 96% 95% CI 90-100%  Scar and septal flash and reverse remodeling: Sensitivity 24% 95% CI 12-36%, specificity 86% 95% CI 76-96%  Septal flash and as a predictor of CRT response: Multivariate prevalence ratio 3.25 95% CI 1.42-7.42	Yes, septal flash was associated with CRT response
Stankovic 2016	Median 48 months	CRT response: 58% Septal flash: 438/677 (65%) Responders versus nonresponders  Septal flash: 329/396 (83%) versus 109/279 (39%), P<0.001  Septal flash and all-cause mortality: multivariable HR 0.456 95%CI 0.342-0.608, P<0.0001	Yes, septal flash was associated with CRT response

CRT=Cardiac resynchronization therapy, PPV=Positive predictive value, NPV=Negative predictive value, LBBB=Left bundle branch block, RVP=Right ventricular pacing, CMR=Cardiac magnetic resonance, LVEF=Left ventricular ejection fraction, LVEDV=Left ventricular end-diastolic volume, LVESV=Left ventricular end-systolic volume, NYHA=New York Heart Association, CI=Confidence interval, OR=Odds ratio

Study quality assessment

Quality assessment of the included studies is shown in Supplementary Table 1. Ascertain of septal was deemed reliable in seven of the included studies.[13141518192223] There was no definition for the assessment of septal flash in two studies.[2021] Outcomes measures included TTE parameters (improvement in LVEF and a reduction in LV volumes) and clinical characteristics (improvement in NYHA classification and survival rates at follow-up). Outcomes were assessed with the evaluation of TTE, CMR or events from medical records. With respect to missing data, eight studies had no significant (<10%) missing data or loss to follow-up.[1314151819212223] The percentage of missing data/lost to follow-up was not clear in the remaining included study.[20] Only three studies used some form of adjustment in the analysis which evaluated the association of septal flash in predicting CRT response.[131923] Only one study included HF patients with atrial fibrillation.[14] all remaining studies included HF patients with sinus rhythm[13151819212223]

Supplementary Table 1

Study quality assessment

Study ID	Prospective design	Reliable septal flash ascertainment	Reliable outcome ascertainment	Low missing data/lost to follow up	Adjustments for confounders	Generalizable cohort
Bernard 2017	No, retrospective	Yes, assessment by echocardiography	Yes, echocardiographic parameters and MACE	Yes, 6 (2.8%) lost to follow-up	No, unadjusted	Yes, HF patients requiring a CRT device
Brunet- Bernard 2014	Not clear	Yes, assessment by echocardiography	Yes, echocardiographic parameters and clinical characteristics	Yes, 5 (2.4%) excluded for missing data	Yes, adjusted for LBBB, nonischemia HF, >70 years, LVEDD <40 mm/msq	Yes, HF patients requiring a CRT device
Doltra 2014	No, retrospective	Yes, assessment by echocardiography	Yes, echocardiographic parameters and clinical characteristics	Yes, no loss to follow up/missing data described	No, unadjusted	Yes, HF patients requiring a CRT device
Gabrielli 2014	Yes, prospective	Yes, assessment by echocardiography	Yes, echocardiographic parameters and clinical characteristics	Yes, 6 (6.3%) excluded for missing data	Yes, adjusted but unclear what variables	No, patients with atrial fibrillation and HF requiring a CRT device
Gasior 2016	Yes, prospective	No, assessment by echocardiography however no definition of septal flash given	Yes, echocardiographic parameters and clinical characteristics	Unclear, subanalysis of a larger study	No, unadjusted	Yes, HF patients requiring a CRT device
Mada 2016	No, retrospective	No, assessment by echocardiography however no definition of septal flash given	Yes, echocardiographic parameters and clinical characteristics	Yes, no loss to follow up/missing data described	No, unadjusted	Yes, HF patients requiring a CRT device
Parsai 2009	Yes, prospective	Yes, assessment by echocardiography	Yes, echocardiographic parameters and clinical characteristics	Yes, no loss to follow up/missing data described	No, unadjusted	Yes, HF patients requiring a CRT device
Sohal 2014	Yes, prospective	Yes, assessment by MRI and echocardiography	Yes, MRI and echocardiographic parameters and clinical characteristics	Yes, no loss to follow up/missing data described	Yes, adjusted prevalence ratios for myocardial scar, QRS duration, echo derived SDI, and presence of septal flash	Yes, HF patients requiring a CRT device
Stankovic 2016	No, retrospective	Yes, assessment by MRI and echocardiography	Yes, MRI and echocardiographic parameters and clinical characteristics	Yes, 2 (0.2%) excluded for missing data	No, unadjusted	Yes, HF patients requiring a CRT device

MACE=Major adverse cardiac events, MRI=Magnetic resonance imaging, LBBB=Left bundle branch block, HF=Heart failure, LVEDD=Left ventricular end-diastolic diameter, LV=Left ventricle, LVESD=LV end-systolic diameter, SDI=Systolic dyssynchrony index, CRT=Cardiac resynchronization therapy

Results for septal flash in predicting overall cardiac resynchronization therapy response

Six of the nine studies had clearly defined criteria for the assessment of CRT response. This was characterized as either a10%[22] or o15%[13151821] reduction in LV end-systolic volumes. The percentage of CRT responders from the included studies varied from 52% to 77%. All studies reported that septal flash was present in a greater proportion for patients who were reported as CRT responders compared to those who were deemed CRT nonresponders. In patients who were deemed CRT responders, septal flash was seen in approximately 40% (n = 865/2174) of patients compared to only 10% (n = 232/2174) who were classed as CRT nonresponders. Only one study by Gabrielli et al.,[14] was not included in the pooled meta-analysis to quantify the association between septal flash and CRT response [Figure 3]. The presence of septal flash at baseline was associated with an increase in CRT response (relative risk 2.55 95% confidence interval [CI] 2.04–3.19, P < 0.001, I = 51%). The moderate statistical heterogeneity was explored and sequential evaluation of I after exclusion of each study did not reduce the statistical heterogeneity lower than 42%.

Figure 3

Meta-analysis of septal flash and response in cardiac resynchronization therapy. Pooled results from 8 studies suggest a 2.5-fold increase in response in patients with cardiac resynchronization therapy who have septal flash compared to no septal flash

Results for septal flash in predicting left ventricular reverse remodeling

Doltra et al. reported that patients with septal flash at baseline had a TTE response rate of 80.2%. This included a reduction in LV end-systolic volumes and LVEF.[13] Doltra et al. reported that CRT corrected the septal flash abnormality in 93% of patients.[13] Gabrielli et al. found that septal flash was the only independent predictor for significant LV reverse remodeling (odds ratios 5.24 95% CI 1.95–14.1, P < 0.01).[14] Gąsior et al. found that septal flash at baseline was associated with significant improvements in LVEF, LV systolic, and diastolic volumes.[20] Of the 87 patients who were identified as having septal flash at baseline by Parsai et al., 88% of these showed resolution of septal flash following CRT implantation, and improvements in LV volumes were also noted, though no significant improvement in overall LVEF was observed.[22] Sohal et al. reported that septal flash without scar was able to predict reverse remodeling with a sensitivity 78% 95% CI 67%–89% and specificity 88% 95% CI 79%–97%.[23]

DISCUSSION

Our review and meta-analysis have several key findings. First, septal flash is common and present in more than half of HF patients undergoing CRT therapy. Second, when septal flash is present there is more than two-fold increase in the response to CRT, with a further suggestion that septal flash is associated with positive LV remodeling after CRT implantation. These findings suggest that in the workup for potential CRT implantation patients should be assessed for the presence of septal flash, regardless of the imaging modality used, as the presence of septal flash is associated with a positive response to CRT therapy.

Our review suggests that amongst patients with CRT device, an LVEF ≤35% and a QRS duration ≥120msec, 53% have septal flash on baseline imaging. Although there have been previous articles providing comprehensive overviews on septal flash,[11] CRT non-response causes[24] and opinion pieces on selection criteria for CRT,[25] this is the first systematic review and meta-analysis which has been undertaken assessing the significance of septal flash in predicting CRT response in patients with HF. Although all studies examined used the inclusion criterion of QRS duration >120 ms, overall the mean duration was 163 ms. This is important as, it highlights that, beyond the assessment of LVEF, assessment of septal flash retains its relevance in a contemporary population as part of the pre-implant workup of patients for CRT.

The benefits of CRT in patients with HF are well known and include positive LV remodeling, improved symptomatic status, and reduced mortality rates.[26] However, approximately one-third of CRT patients do not show a clinical improvement.[27] Thus, there is a potentially large cohort of patients receiving expensive therapies which require long-term follow-up with no clear clinical benefit associated with it. Although there are well-defined international guidelines for the use of CRT in HF patients which include LBBB morphology and QRS duration of >130 ms,[7] there are several studies which indicate that prolonged QRS duration is not a precise and specific marker of true ventricular dyssynchrony.[2829] This was seen in Gabrielle et al. also where they concluded that QRS duration had no independent prognostic value in assessing CRT response.[14]

Various echocardiographic dyssynchrony parameters have been investigated previously which were thought to better reflect mechanical dyssynchrony. These included septal to posterior wall motion delay, longitudinal velocities, and differences in LV to right ventricular ejection timing intervals.[30] These parameters have been used in large multicenter trials including PROSPECT to assess for CRT response, however, these measurements were hindered with poor reproducibility, as such they were thought to be unreliable to assist in the selective criteria of patients for CRT in mainstream clinical practice.[3031] This is not unexpected as many of the parameters involve complex and time-consuming calculations. The assessment of septal flash overcomes the limitations due to its distinctive and recognizable pattern of motion which, as seen in this review can easily be assessed by the routine use of 2D TTE. The evaluation of septal flash is advantageous as most patients have 2D TTE to assess the LVEF when considering whether patients met CRT implantation guideline criteria. Furthermore, in two studies the inter-observer agreement was shown to be high.[1321] This is an important consideration which cardiovascular imaging reporters should be aware of particularly when assessing patients who may be eligible for CRT.

The limitations of this review include the small number of studies investigating the use of septal flash of which five were retrospective in nature. Our study was also limited because we did not apply any restriction on how septal flash was quantified which could be by any imaging modality including echocardiography and CMR imaging. More research on the quality of life measurements, cardiac event rates such as recurrent hospitalization with HF and mortality are required to assess the association of septal flash and its potential impact on mortality rates as data in this area is lacking. Patients in atrial fibrillation are increasingly offered CRT, they represent a particularly challenging group for management, outcome measurement, and suitability for this therapy. The presence or absence of septal flash in patients with atrial fibrillation also requires more investigation.

The underlying mechanisms of action of septal flash remain relatively unknown as such, a better understanding of the pathophysiology would enable a more targeted approach of patients when being considered for CRT. Furthermore, this may lead to improved understanding in how LBBB, septal flash, and LVEF are interrelated. All of the included studies used CRT indications of LBBB duration of >120 ms and an LVEF LVEF%, however, the latest international guidelines now require an LBBB duration of >130 ms. Due to the date of the included studies of this review, these new guidelines were not implemented. Although it must be noted that in the included studies, the overall averaged QRS duration is still in keeping with the new guidelines of >130 ms.

CONCLUSIONS

Septal flash is a well-defined and distinctive contraction pattern that is easily recognizable on cardiac imaging. Septal flash may be associated with CRT response and should be evaluated in the patients that are considered for CRT devices.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.