Resynchronization of the left atrium may play an important role in cardiac resynchronization therapy

Introduction

Left atrial (LA) dyssynchrony is a predictor of response to cardiac resynchronization therapy (CRT). It is unknown, however, if LA resynchronization contributes to response to CRT. We hypothesize that there is a relationship between correction of LA dyssynchrony and response to CRT.

Purpose

To investigate the association between LA resynchronization and response to CRT.

Methods

In a prospective study of 171 heart failure patients with LBBB, myocardial strain was measured by speckle-tracking echocardiography, before and 6 months after CRT. As indicated by the white arrows in Figure 1, LA dyssynchrony was measured as the time delay between onset systolic stretch of the interatrial septum and the LA lateral wall. Response to CRT was defined as at least 15% reduction in left ventricular (LV) end systolic volume at 6 months follow up.

Results

119 (70%) patients responded to CRT. The panels in Figure 1 shows LA strain traces in a representative LBBB patient that did respond (upper panels), and a patient that did not respond (lower panels). The white arrows in the left panels indicate that both the responder and the non-responder had marked LA dyssynchrony before CRT (198 and 171 ms, respectively). However, after 6 months with CRT, there was recovery of LA synchrony only in the responder (time delay −40 ms), and still marked LA dyssynchrony of 191 ms in the non-responder (right panels).

Figure 2 confirms similar results for the whole study population: CRT response was associated with marked reduction of LA dyssynchrony (p=0.0001). In the CRT non-responders there was, however, only a modest, non-significant reduction of LA dyssynchrony.

Conclusions

Positive CRT response was associated with resynchronization of the left atrium. These findings suggest LA resynchronization as a potential additional target for CRT.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Institute for Surgical Research, Oslo University HospitalThe Intervention Centre, Oslo University Hospital

A deep learning approach for automatic echocardiographic right ventricular strain measurements using a limited dataset

Background

Speckle tracking echocardiography provides quantification of myocardial deformation and is useful in the assessment of myocardial function. Right ventricular (RV) strain has been suggested as a sensitive tool for diagnosing cardiomyopathies and assessing long term patient outcomes for patients with pulmonary hypertension, severe tricuspid regurgitation and COVID-19 infection. Recent advances in deep learning (DL) have made promising advances in automating the labour-intensive delineation of regions of interest (ROIs). However, compared to echocardiograms with left ventricular (LV) strain, RV strain data is scarce, making DL models difficult to train.

Purpose

To investigate whether annotated LV strain data could be beneficial in training a DL model for automatic RV strain when using a limited RV dataset.

Methods

The dataset consisted of anonymized still frames from 141 echocardiograms of the RV in the RV-focused 4 chamber view with corresponding cardiologist-defined ROI. Exams included healthy subjects and patients with heart failure, valvular disease, and conduction abnormalities. ROIs and still images were extracted at the mid-systole, and then quality assessed by an experienced cardiologist as high, medium, or low. The dataset was randomly split into 68%/17%/15% sets for training, validation, and testing. A convolutional neural network for image segmentation (U-Net) with a residual neural network (ResNet50) encoder was used, with a combination of binary cross entropy and Dice loss functions. Augmentation, predefined ImageNet weights and pre-training were also employed. For pre-training, 715 still images in the apical 4 chamber view with LV defined ROIs were used, both in their original and horizontally flipped view. Predicted ROIs were reintroduced into commercially available echocardiogram analysis software to automatically calculate longitudinal strain (LS) values.

Results

The model pre-trained with the flipped LV images achieved the highest performance with a mean absolute difference of 1.26 percentage points (95% confidence interval (CI): 0.62–1.89 percentage points) between manually measured and DL-assisted LS. Median absolute LS difference was 0.85 (95% CI: 0.28–1.57) percentage points. A Bland-Altman plot revealed two outliers and no obvious trends. In comparison, the mean and median absolute LS differences for the model without pre-training were 1.87 (95% CI: 0.73–3.00) and 1.09 (95% CI: 0.56–1.63) percentage points, respectively.

Conclusions

The current study demonstrates that DL-assisted, automated RV strain measurement is feasible even with a small dataset, and that performance can be increased by using images annotated for LV strain. While the majority of the predicted RV strain results were within the typical range of intra- and interobserver variability, a few outliers were observed. These outliers could possibly be avoided with the use of larger datasets.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Research council of Norway

Prognostic value of diastolic function assessment in patients undergoing cardiac resynchronization therapy

Funding Acknowledgements

Type of funding sources: None.

Objective

The best modality to assess diastolic function in CRT-candidates is an object of debate and the relationship between diastolic function, CRT-response and survival are not

clearly understood.

Purpose of the study: to assess diastolic patterns in patients undergoing CRT according to the 2016 recommendations of the American Society of Echocardiography/European Association of Cardiovascular Imaging and to evaluate the prognostic value of diastolic dysfunction (DD) in CRT candidates.

Methods

193 patients (age: 67 ± 11 years, QRS width: 167 ± 21 ms) were included in this multicentre prospective study. Patients were stratified according to DD grades (grade I

to III). CRT-response was defined as a reduction of left ventricular (LV) end-systolic volume >15% at 6-month follow-up (FU). The primary endpoint was defined as a

composite of heart transplantation, LV assisted device implantation or all-cause death during FU.

Results

During FU, 132 (68%) patients were CRT-responders. CRT delivery was associated with diastolic function degradation in non-responders. Grade I DD was able to predict

CRT-response with a sensitivity, specificity and accuracy of 70%, 65%, and 63%, respectively. After a median period of 35 months, the primary endpoint occurred in 29

(15%) patients. Grade I DD was associated with a better outcome [HR 0.26 95% CI: (0.10-0.66)], independently from ischemic cardiomyopathy, LV dyssynchrony and CRT-response (Table 1). Non-responders with grade II or grade III DD had the worse prognosis (HR 4.36, 95%CI: 2.10-9.06) Figure 1.

Conclusions

Grade I DD is associated with LV remodelling after CRT and is an independent predictor of prognosis in CRT candidates.

Abstract Figure.

Importance of systematic right ventricular assessment in cardiac resynchronization therapy candidates: a machine-learning approach

Funding Acknowledgements

Type of funding sources: None.

Background

Despite having all a systolic heart failure and broad QRS, patients proposed for cardiac resynchronization therapy (CRT) are highly heterogeneous and it remains extremely complicated to predict the impact of the device on left ventricular (LV) function and outcomes.

Objectives

We sought to evaluate the relative impact of clinical, electrocardiographic, and echocardiographic data on the left ventricular (LV) remodeling and prognosis of CRT-candidates by the application of machine learning (ML) approaches.

Methods

193 patients with systolic heart failure undergoing CRT according to current recommendations were prospectively included in this multicentre study. We used a combination of the Boruta algorithm and random forest methods to identify features predicting both CRT volumetric response and prognosis (Figure 1). The model performance was tested by the area under the receiver operating curve (AUC). We also applied the K-medoid method to identify clusters of phenotypically-similar patients.

Results

From 28 clinical, electrocardiographic, and echocardiographic-derived variables, 16 features were predictive of CRT-response; 11 features were predictive of prognosis.

Among the predictors of CRT-response, 7 variables (44%) pertained to right ventricular (RV) size or function. Tricuspid annular plane systolic excursion was the main feature associated with prognosis. The selected features were associated with a very good prediction of both CRT response (AUC 0.81, 95% CI: 0.74-0.87) and outcomes (AUC 0.84, 95% CI: 0.75-0.93) (Figure 1, Supervised Machine Learning Panel). An unsupervised ML approach allowed the identifications of two phenogroups of patients who differed significantly in clinical and parameters, biventricular size and RV function. The two phenogroups had significant different prognosis (HR 4.70, 95% CI: 2.1-10.0, p < 0.0001; log –rank p < 0.0001; Figure 1, Unsupervised Machine Learning Panel).

Conclusions

Machine learning can reliably identify clinical and echocardiographic features associated with CRT-response and prognosis. The evaluation of both RV-size and function parameters has pivotal importance for the risk stratification of CRT-candidates and should be systematically assessed in patients undergoing CRT.

Abstract Figure 1

Regional myocardial work as determinant of heart failure in left bundle branch block

Funding Acknowledgements

Type of funding sources: Public Institution(s). Main funding source(s): The Norwegian Health Association

Background

Left bundle branch block (LBBB) worsen prognosis in heart failure patients. LBBB may also cause heart failure in otherwise healthy individuals. The mechanical changes induced by LBBB are potential determinants of heart failure in these patients, but their relation to left ventricular (LV) systolic function is incompletely understood.

Purpose

This study investigates the contribution of regional contractile function to heart failure in patients with LBBB.

Methods

In 76 patients with LBBB and 11 healthy controls, myocardial strain was measured by speckle-tracking echocardiography and myocardial work by pressure-strain analysis. Patients with ischemic heart disease or myocardial scarring were excluded. LBBB patients were stratified by LV ejection fraction (EF) >50% (EFpreserved), 36-50% (EFmid), and ≤35% (EFlow). 62 LBBB patients subsequently underwent cardiac resynchronization therapy (CRT) implantation and was re-examined at 6 months.

Results

Septal work was significantly and successively reduced from controls, EFpreserved, EFmid, to EFlow (1977 ± 506, 1025 ± 342, 601 ± 494 and -41 ± 303 mmHg·%, respectively, all p < 0.01) (Figure 1). There was a strong correlation (R = 0.84, p < 0.01) between septal work and LVEF. In contrast, work in the LV lateral wall was preserved in both EFpreserved (2367 ± 459 mmHg·%) and EFmid (2252 ± 449 mmHg·%) vs controls (2062 ± 459 mmHg·%, all NS). In the EFlow group, however, LV lateral wall work was reduced (1473 ± 568 mmHg·%, p < 0.01 vs controls). Thus, lateral wall function was not correlated with LVEF in patients with LVEF >35% (NS). At six month CRT septal work was markedly increased (165 ± 485 vs 1288 ± 523 mmHg·%, p < 0.01) and LV lateral wall work reduced (1730 ± 620 vs 1264 ± 490 mmHg·%, p < 0.01). LVEF increased from 32 ± 8 to 47 ± 10 % (p < 0.01).

Conclusions

Heart failure in LBBB patients is determined by degree of septal dysfunction. LV lateral wall function, on the other hand, is preserved in the early phase of heart failure and was only reduced in patients with severe heart failure. Further clinical studies should investigate if measuring LV lateral wall function can increase precision in patient selection for CRT.

Abstract Figure.

Elevated septal wall stress - a driver of left ventricular dysfunction in left bundle branch block?

Funding Acknowledgements

Type of funding sources: Public Institution(s). Main funding source(s): The Norwegian Health Association

Background

Septal dysfunction is a main feature of left bundle branch block (LBBB), and increasing wall stress is a proposed mechanism of heart failure development in LBBB patients. To try to reveal the pathophysiologic pathway from dyssynchrony to heart failure, we investigated the relationship between septal and left ventricular (LV) lateral wall stress in patients with LBBB.

Hypothesis

Increased septal wall stress causes septal dysfunction in LBBB.

Methods

We included 24 LBBB-patients (65 ± 11 years, 11 males) with LV ejection fraction (EF) ranging from 18 to 67%, and 8 healthy controls (58 ± 10 years, 4 males). Wall stress was calculated at peak LV pressure (LVP) according to the law of La Place ([LVP x radius]/[wall thickness]). Wall thickness was measured using M-mode, and regional curvature was measured in mid-ventricular shortaxis from 2D echocardiographic images. We used a previously validated non-invasive method to estimate LVP from brachial blood pressure and adjusted for valvular events. Myocardial scar was ruled out by late gadolinium enhancement cardiac magnetic resonance imaging.

Results

Wall stress was significantly higher in septum than LV lateral wall at peak LVP (48 ± 12 vs 37 ± 11 kPa, p < 0.01) in LBBB patients, while no difference was seen in the controls (Figure A). In patients, septal wall thickening showed a strong correlation with LVEF (r = 0.77, p < 0.01) (Figure B). Similar correlation was not significant for the LV lateral wall (r = 0.13, NS). Attenuation of septal wall thickening in LBBB-patients correlated well with increasing septal wall stress (r=-0.60, p < 0.01). Wall thickening and stress did not correlate in the LV lateral wall (r=-0.14, NS).

Conclusion

Increased septal wall stress is associated with reduced systolic thickening in patients with LBBB. Septal wall thickening, in contrast to LV lateral wall thickening, was correlated to global LV function. These findings suggest that septal remodeling which could have normalized septal wall stress, was not achieved and heart failure may develop.

Abstract Figure.

Prognostic utility of the assessment of diastolic function in patients undergoing cardiac resynchronization therapy

Conflicting data exist about the relationship between cardiac resynchronization therapy (CRT) and diastolic function. Aims of the study are to assess diastolic patterns in patients undergoing CRT according to the 2016 recommendations of the American Society of Echocardiography/European Association of Cardiovascular Imaging and to evaluate the prognostic value of diastolic dysfunction (DD) in CRT candidates. METHODS AND RESULTS: One-hundred ninety-three patients (age: 67 ± 11 years, QRS width: 167 ± 21 ms) were included in this multicentre prospective study. Mitral filling pattern, mitral tissue Doppler velocity, tricuspid regurgitation velocity, and indexed left atrial volume were used to classify DD from grade I to III. CRT-response, defined as a reduction of left ventricular (LV) end-systolic volume > 15% at 6-month follow-up (FU), occurred in 132 (68%) patients. The primary endpoint was a composite of heart transplantation, LV assisted device implantation, or all-cause death during FU and occurred in 29 (15%) patients. CRT was associated with a degradation of DD in non-responders. At multivariable analysis corrected for clinical variables, QRS duration, mitral regurgitation, CRT-response and LV dyssynchrony, grade I DD was associated with a better outcome (HR 0.37, 95% CI: 0.14-0.96). Non-responders with grade II-III DD had the worse prognosis (HR 4.36, 95%CI: 2.10-9.06). CONCLUSIONS: The evaluation of DD in CRT candidates allows the prognostic stratification of patients, independently from CRT-response.

158 Myocardial work exposes afterload-dependent changes in strain

Funding Acknowledgements

Norwegian Health Association

Background

Global longitudinal strain (GLS) is used for detection of subclinical left ventricular (LV) dysfunction, for example when screening for chemotherapy-induced cardiotoxicity. A relative percentage reduction in GLS ≥8% is considered abnormal. However, as GLS is load-dependent, modest increases in afterload can potentially prove sufficient to cause clinical implication. In contrast, global myocardial work (GMW) which incorporates afterload, may be more accurate in detecting LV-dysfunction.

Purpose

We investigated the effect of increased afterload on GLS, and if GMW may be a more accurate parameter of myocardial function during increased afterload.

Methods

In 20 healthy individuals (age 49 ± 11 years (mean ± SD), 10 men), blood pressure was increased by a 3 minute arithmetic mental stress test. GLS was measured by speckle tracking echocardiography and LV ejection fraction (EF) by biplane Simpson. GMW was calculated from LV pressure-strain analysis using a non-invasive estimate for LV pressure (LVP).

Results

During the afterload-elevation, systolic blood pressure increased by 25 ± 16 mmHg (p < 0.01), and heart rate by 16 ± 13 bpm (p < 0.01). This was followed by a decrease in EF from 62 ± 5 to 59 ± 5% (p < 0.01) and GLS from 21.9 ± 2.2 to 20.8 ± 2.0% (p < 0.01). In contrast, GMW increased from 2052 ± 278 to 2382 ± 388 mmHg·% (p < 0.01). In 5 of 20 (25%) individuals, the relative percentage reduction of GLS was >8%, despite an increase in GMW. The figure shows an individual example during rest and afterload-elevation, where an increase in systolic blood pressure of 16 mmHg was associated with a 9% relative percentage reduction in GLS, but a small increase in GMW as illustrated by the loop areas.

Conclusions

This study demonstrated that a modest increase in afterload can result in significant reduction in GLS, that may lead to overdiagnosis of LV-dysfunction. GMW did not decrease, suggesting it has a better specificity in patients at risk for subclinical LV dysfunction. Future studies should investigate if GMW is more accurate than strain in detecting LV-dysfunction.

Abstract 158 Figure.

160 Echocardiographic assessment of CRT candidates. Does additional scar evaluation by MRI improve prediction of response?

Background

Myocardial scar presence and extent, has a considerable influence on response to cardiac resynchronization therapy (CRT). Apical rocking (ApRock) and septal flash (SF) are associated with favourable outcome after CRT. Little is known however to which extent visual assessment of mechanical dyssynchrony by ApRock, SF and scar predicts CRT response. We therefore investigated, if additional scar assessment by cardiac magnetic resonance imaging (MRI) adds to the predictive value of the visual evaluation of echocardiographic images in CRT candidates.

Methods

A total of 201 unselected patients referred for CRT, who fulfil the contemporary guidelines for CRT implantation, were enrolled in this prospective multicentre study. Two experienced observers visually assessed echocardiographic images before CRT implantation, focussing on the presence of ApRock, SF and location and extent of scar segments of the left ventricle (LV), resulting in a CRT response prediction (i.e. Integrative Prediction). A third observer provided a consensus reading in case of disagreement. All observers were blinded to all patient information other than the ischaemic aetiology of heart failure. Independent from that, segmental myocardial scar burden was objectified by late gadolinium enhancement (LGE) cardiac MRI (LGE > 50%). CRT response was defined as ≥15% reduction in LV end-systolic volume on echocardiography, one year after device implantation.

Results

Overall, 69 (34%) patients had an ischaemic aetiology of heart failure. Before CRT, ApRock and SF were present in 129 (64%) and 136 (68%) patients, respectively. ApRock and SF alone predicted CRT response with an area under the curve (AUC) of 0.85 (95% CI: 0.79-0.91) and 0.84 (95% CI: 0.77-0.91) (Figure A), while the echocardiographic Integrative Prediction had an AUC of 0.90 (95% CI: 0.84-0.95), with a sensitivity of 93% and a specificity of 87% for the prediction of CRT response (Figure B) (p < 0.05 vs. ApRock and SF alone). When combining information on ApRock, SF and the number of scarred segments on MRI in a statistical model, the AUC was comparable to the echocardiographic Integrative Prediction [0.90 (95% CI: 0.84-0.96)] as was sensitivity and specificity (91% and 83%, respectively, p = N.S. vs. Integrative Prediction) (Figure C).

Conclusions

An integrative visual assessment of LV function has an excellent predictive value for CRT response. Our data show, that the echocardiographic estimation of scar burden is sufficiently accurate and cannot be further improved by an additional MRI scar assessment.

Abstract 160 Figure.

561 Targeting septal work and viability identifies responders to cardiac resynchronization therapy

Funding Acknowledgements

The study was supported by Center for Cardiological Innovation.

Introduction

Septal dysfunction is the dominant mechanism of left ventricular (LV) failure in left bundle branch block (LBBB). We hypothesize that, provided septum is viable, septal function can recover and hence LV function improve after cardiac resynchronization therapy (CRT).

Purpose

To determine if combined assessment of septal function and viability identifies responders to CRT.

Methods

In a prospective multicenter study of 200 unselected patients referred for CRT, we measured myocardial strain by speckle-tracking echocardiography and regional work by pressure-strain analysis before and 7 ± 1 months after CRT. Viability was assessed by late gadolinium enhancement cardiac magnetic resonance imaging (n = 123). CRT response was defined as ≥15% reduction in LV end-systolic volume.

Results

Before CRT, septal work was 258 ± 463 and LV lateral wall work 1469 ± 674 mmHg·% (p < 0.0001). In CRT responders, septal work was restored to 1243 ± 495 mmHg·%, whereas non-responders showed less marked improvement (p < 0.0001). The figure illustrates a typical CRT responder with negative septal work and a large difference between work in the LV lateral wall and septum (panel A). There was no septal scar (panel B) and, after 6 months with CRT, septal work was recovered (panel C). Pressure-strain loops illustrate that CRT converted inefficient septal contractions with substantial negative (wasted) work to positive work throughout systole. For the entire study population, the difference between work in the LV lateral wall and septum predicted CRT response with area under the curve (AUC) 0.75 (95% CI: 0.68-0.83) and was feasible in 98% of patients. Furthermore, septal scar predicted non-response to CRT with AUC 0.76 (95% CI: 0.65-0.86). Combining work difference and septal viability improved AUC for CRT response to 0.85 (95% CI: 0.76-0.94) (figure panel D). The AUC was similar for QRS 120-150 and >150 ms.

Conclusions

The proposed combined approach with assessment of septal work and viability identified CRT responders with high precision.

Abstract 561 Figure.

553 Acute re-distribution of regional left ventricular work by cardiac resynchronization therapy determines long-term remodelling

Background

In patients with dilated cardiomyopathy and left bundle branch block (LBBB), different regions of the left ventricle (LV) have been shown to perform different amounts of work. In this study, we investigate the acute impact of cardiac resynchronization therapy (CRT) on regional LV work distribution and its relation to long-term reverse-remodelling.

Methods

We recruited 140 heart failure patients, referred for CRT. Regional myocardial work was calculated from non-invasive echocardiographic segmental stress-strain-loop-area before and immediately after CRT. The magnitude of volumetric reverse-remodelling was determined from the change in LV end-systolic volume (ESV), 11 ± 3 months after implantation. Characteristics of patients with the lowest and highest quartile of LV ESV reverse remodelling (LV ESV reduction of less than 10% and LV ESV reduction of more than -48%) were compared.

Results

Before CRT, myocardial work showed significant differences among the walls of the LV (Figure A). CRT caused an acute re-distribution of myocardial work, on average with most increase in the septum and most decrease laterally (all walls p < 0.05) and lead to a homogeneous work distribution (Figure B). The acute change in the difference between lateral and septal wall work (Δ Lateral-to-septal work) correlated significantly with LV ESV reverse-remodelling (r = 0.63, p < 0.0001). The smallest changes in work were seen in the patients with the least LV ESV reverse remodelling (Figure C, red markers), while patients with the most LV ESV reverse remodelling showed the largest changes in work (Figure C, green markers). In multivariate linear regression analysis, including conventional parameters such as pre-implant QRS duration, LV ejection fraction, LV end-diastolic volume and global longitudinal strain, the re-distribution of work across the septal and lateral walls appeared as the strongest determinant of volumetric reverse-remodelling after CRT (R²=0.393, p < 0.0001).

Conclusions

The acute re-distribution of regional myocardial work between the septal and lateral wall of the left ventricle is an important determinant of long term reverse-remodelling after CRT-implantation. Our data suggest that modification of regional loading is the mode of action of CRT treatment.

Abstract 553 Figure.

P602Septal function and viability determine response to cardiac resynchronization therapy

Introduction

Cardiac resynchronization therapy (CRT) has evolved as an important treatment in patients with symptomatic heart failure, reduced left ventricular (LV) ejection fraction and wide QRS. However, as one third of patients do not benefit from the therapy, there is need for better selection criteria. Previous studies have shown an association between recovery of septal function and response to CRT.

Purpose

To test the hypothesis that septal dysfunction in the absence of scar predicts response to CRT.

Methods

In 121 patients undergoing CRT implantation according to current European Society of Cardiology guidelines, we performed speckle-tracking echocardiography and estimated LV pressure non-invasively based on a method recently innovated in our lab. Pressure-strain analysis was used to calculate myocardial work. Septal dysfunction with asymmetric LV workload was calculated as the difference between LV lateral wall and septal work. Late gadolinium enhancement cardiac magnetic resonance imaging (LGE-CMR) was performed to assess septal scar. CRT response was defined as ≥15% reduction of LV end systolic volume by echocardiography at 6 months follow-up.

Results

Eighty-eight patients (73%) responded to CRT at 6 months follow-up. Multivariate logistic regression analysis including lateral-to-septal work difference, septal scar, QRS duration and QRS morphology found that only lateral-to-septal work difference and septal scar were significant predictors of CRT response (both p<0.005). Using logistic regression and receiver operating characteristic (ROC) curve analysis, we found that the combined approach of these two parameters identified CRT responders with a sensitivity of 86% and a specificity of 82%. The area under the curve (AUC) for CRT response prediction was 0.85 (95% CI: 0.76–0.94) (Figure). In comparison, the AUC value for QRS duration was 0.63 (95% CI: 0.52–0.75). Furthermore, for the subgroup of patients with QRS duration 120–150 ms (n=27), the AUC value for lateral-to-septal work difference in combination with septal scar was 0.90 (95% CI: 0.78–1.00).

Conclusions

A multimodality approach with strain echocardiography and LGE-CMR was able to detect CRT responders with high accuracy, also in the subset of patients with intermediate QRS duration. A dysfunctional but viable septum appears to be an ideal target for CRT.

Sphenopalatine Ganglion Stimulation to Augment Cerebral Blood Flow

Background and Purpose—

Many patients with acute ischemic stroke are not eligible for thrombolysis or mechanical reperfusion therapies due to contraindications, inaccessible vascular occlusions, late presentation, or large infarct core. Sphenopalatine ganglion (SPG) stimulation to enhance collateral flow and stabilize the blood-brain barrier offers an alternative, potentially more widely deliverable, therapy.

Methods—

In a randomized, sham-controlled, double-masked trial at 41 centers in 7 countries, patients with anterior circulation ischemic stroke not treated with reperfusion therapies within 24 hours of onset were randomly allocated to active SPG stimulation or sham control. The primary efficacy outcome was improvement beyond expectations on the modified Rankin Scale of global disability at 90 days (sliding dichotomy), assessed in the modified intention-to-treat population. The initial planned sample size was 660 patients, but the trial was stopped early when technical improvements in device placement occurred, so that analysis of accumulated experience could be conducted to inform a successor trial.

Results—

Among 303 enrolled patients, 253 received at least one active SPG or sham stimulation, constituting the modified intention-to-treat population (153 SPG stimulation and 100 sham control). Age was median 73 years (interquartile range, 64–79), 52.6% were female, deficit severity on the National Institutes of Health Stroke Scale was median 11 (interquartile range, 9–15), and time from last known well median 18.6 hours (interquartile range, 14.5–22.5). For the primary outcome, improved 3-month disability beyond expectations, rates in the SPG versus sham treatment groups were 49.7% versus 40.0%; odds ratio, 1.48 (95% CI, 0.89–2.47); P =0.13. A significant treatment interaction with stroke location (cortical versus noncortical) was noted, P =0.04. In the 87 patients with confirmed cortical involvement, rates of improvement beyond expectations were 50.0% versus 27.0%; odds ratio, 2.70 (95% CI, 1.08–6.73); P =0.03. Similar response patterns were observed for all prespecified secondary efficacy outcomes. No differences in mortality or serious adverse event safety end points were observed.

Conclusions—

SPG stimulation within 24 hours of onset is safe in acute ischemic stroke. SPG stimulation was not shown to statistically significantly improve 3-month disability above expectations, though favorable outcomes were nominally higher with SPG stimulation. Beneficial effects may distinctively be conferred in patients with confirmed cortical involvement. The results of this study need to be confirmed in a larger pivotal study.

Clinical Trial Registration—

URL: https://www.clinicaltrials.gov . Unique identifier: NCT03767192.