Study

This study complied with the Helsinki Declaration and was approved by the ethics committee of our institution. Informed consent was not sought from the patients, as this was an observational study and did not involve any changes to diagnostic tests or therapeutic interventions.

Patient selection

Data were collected prospectively for all patients who underwent elective LVAD or biventricular VAD (BiVAD) implantation and preoperative echocardiography between November 2010 and August 2011, at the Clinic for Thoracic and Cardiovascular Surgery in Bad Oeynhausen, Germany. The devices implanted were the HeartMate II (Thoratec, Pleasanton, CA, USA), the HeartWare HVAD (HeartWare, Oakville, CA, USA) and the Thoratec Paracorporeal BiVAD (Thoratec, Pleasanton, CA, USA). Patients receiving a total artificial heart were excluded from the analysis because, at our institution, the choice to implant a total artificial heart was often based on the presence of mechanical prostheses, active endocarditis, severe pulmonary insufficiency, extensive LV apical thrombus or hypertrophic cardiomyopathy. Patients were also excluded if image quality was deemed insufficient for the analysis of RV function.

Clinical data

Baseline clinical, demographic, haemodynamic and laboratory data were recorded prospectively in the electronic record. An Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profile and Michigan RV risk score were calculated for each patient. The Michigan score assigns points based on four variables, with vasopressor use adding 4 points, creatinine > 2.3 mg/dL adding 3 points, bilirubin > 2 mg/dL adding 2.5 points and aspartate aminotransferase > 80 IU/dL adding 2 points. Higher scores (especially if ≥ 5.5) are associated with a greater risk of RVF.

Patients with a glomerular filtration rate < 60 mL/min/1.73 m ² for 3 months were classified as having chronic renal insufficiency.

Deaths during hospitalization and the causes of death were recorded. Univariate risk factors for death were analysed.

Echocardiographic assessment

Preoperative transthoracic echocardiography results were reviewed and analysed by a reader blinded to clinical outcome. Standard echocardiographic measurements of the RV were made in accordance with current guidelines , including basal RV end-diastolic diameter (RVEDD), end-systolic and end-diastolic RV areas, fractional area change, maximal systolic excursion of the tricuspid annulus (TAPSE), tissue Doppler systolic and diastolic velocities of the RV lateral wall (S _RV and E _RV ), pulsed Doppler tricuspid E wave (Et), systolic pulmonary arterial pressure and maximal and minimal diameters of the inferior vena cava. Longitudinal strain of the RV free wall was measured on the stored DICOM loops with standard commercially available software (QLAB CMQ, Philips, Amsterdam, Netherlands). Images were searched for evidence of severe mitral, aortic, tricuspid and pulmonary regurgitation. The systolic and diastolic functions of the left ventricle (LV) were also assessed: LV end-diastolic diameter and volume, LV end-systolic diameter and volume, ejection fraction (Biplan Simpson), aortic time-velocity integration, tissue Doppler imaging, systolic and diastolic lateral (S _LAT and E _LAT ) and septal (S _SEPT and E _SEPT ) velocities, pulsed Doppler transmitral E wave (Em) and mitral deceleration time.

Several markers were calculated: cardiac index, right-to-left ventricular end-diastolic diameter, Et/E _RV ratio, Em/E _SEPT ratio, Em/E _LAT ratio, Em/S _SEPT ratio and Em/S _LAT ratio.

Data collection

The data were collected in the 24 hours before LVAD implantation. The median period between admission and VAD implantation was 5.0 days (interquartile range, 2.0–7.0).

Outcomes

Patients were divided into three groups: LVAD patients without RVF, LVAD patients with RVF and patients for whom BiVAD implantation was planned. RVF was defined as the unplanned insertion of a right VAD (RVAD) or the use of an intravenous inotrope for 14 days after surgery .

Statistical analysis

Preoperative variables were compared between the three groups with GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA). Continuous variables were compared in unpaired t tests for normally distributed variables and Wilcoxon rank-sum tests for non-normally distributed variables. Chi ² or Fisher’s exact tests were used for categorical variables. Analysis of variance was used to compare continuous variables between groups. A P value < 0.05 was considered significant. Relative risks (RRs) are presented with 95% confidence intervals (CIs).

Multivariable analyses were based on stepwise multiple logistic regression analysis and were used to assess predictors of RVF. BiVAD patients were not included in multivariable analyses, because they were generally in a more critical state and the definition of postoperative RVF failure is more complex in this specific context. The following variables were identified as significant predictors ( P < 0.05) in univariate analyses between the two LVAD groups and were, therefore, included in multivariable analyses: INTERMACS profile, basal RVEDD, minimal inferior vena cava diameter and Em/S _LAT ratio.

Bootstrap estimation with resampling from 1000 simulations (570 simulations for the RVF LVAD group and 430 simulations for the LVAD group without RVF) was used. Univariate and multivariable analysis were performed on bootstrap samples.

An ARVADE (assessment of right ventricular dysfunction predictors before the implantation of a left ventricular assist device in end-stage heart failure patients using echocardiographic measures) score was devised by rounding the exponentiated regression model coefficients of independent predictors of RVF to the nearest 0.5. A receiver operating characteristic curve was plotted for ARVADE score, and the area under the curve (AUC) was calculated. AUCs were also calculated for Michigan score and for the independent predictors of RVF.

Population characteristics

During the study period, 67 patients received mechanical circulatory support. We excluded 18 patients because they had received a total artificial heart (11 patients) or because complete echocardiographic evaluations with strain rate imaging were not available (7 patients; all undergoing emergency implantation); thus, 49 patients were included in the study. Twenty-seven patients received the HeartWare HVAD (HeartWare), 15 patients received the HeartMate II (Thoratec) and seven patients received a Thoratec Paracorporeal BiVAD. Detailed results are shown in Table 1 .

Table 1

Preoperative patient characteristics according to postoperative occurrence of RVF.

Characteristic	Study patients, ( n = 49)	LVAD patients without RVF, ( n = 18)	LVAD patients with RVF, ( n = 24)	BiVAD patients, ( n = 7)
Age (years)	56 [46–67]	60 [48–69]	52 [43–61]	51 [48–57]
Men	44 (90)	18 (100)	21 (88)	5 (71)
BMI (kg/m 2 )	25.3 [22.4–27.2]	24.2 [22.7–26.9]	26.0 [22.3–37.0]	26.4 [21.6–28.7]
Ischaemic aetiology	23 (47)	10 (56)	9 (38)	2 (29)
History of COPD	5 (10)	3 (17)	2 (8)	0
Chronic renal insufficiency	24 (49)	11 (61)	9 (38)	2 (29)
Previous cardiac surgery	18 (37)	7 (39)	11 (46)	0
Resynchronization therapy	20 (41)	10 (56)	6 (25)	4 (57)
Bridge to transplantation	33 (67)	11 (61)	15 (63)	7 (100)
Preoperative mechanical ventilation	12 (25)	1 (6)	9 (38) a	2 (29)
Preoperative haemodialysis	9 (18)	3 (17)	3 (13)	3 (43)
Preoperative IABP	21 (43)	5 (28) b	11 (46)	6 (86)
Preoperative ECMO	4 (8)	1 (6)	3 (13)	0
Preoperative inotrope	47 (96)	16 (89)	24 (100)	7 (100)
SOFA score	4 [3–7]	3 [2–7]	4 [3–5]	7 [7–10]
SAPS II	30 [23–40]	24 [20–37] b	29 [23–38]	35 [33–42]
Michigan score	3.0 [0–5.5]	2.3 [0.0–4.1] b	2.8 [0.5–4.4] c	8.5 [7.5–9.5]
INTERMACS level	1 [1–3]	3 [1–3] b	1 [1–2] a	1 [1]
Serum sodium concentration (mmol/L)	134 [129–136]	135 [128–139] b	135 [131–137] c	126 [125–132]
Blood urea nitrogen concentration (mg/dL)	76 [54–107]	80.0 [57.5–111.5]	76.0 [54.8–102.8]	111 [63.8–195.5]
Creatinine concentration (mg/dL)	1.5 [1.2–2.2]	1.8 [1.3–2.3]	1.3 [1.0–1.8]	2.3 [2.2–3.7]
CRP concentration (mg/L)	2.3 [1.0–7.8]	1.7 [0.3–6.2]	2.2 [1.1–6.9]	5.1 [4.4–7.2]
AST concentration (IU/L)	31 [25–62]	27 [22–42]	37 [31–49]	31.0 [21.3–185.5]
Bilirubin concentration (mg/dL)	1.7 [1.0–2.5]	1.4 [0.6–2.1] b	1.6 [0.9–3.5] c	3.6 [2.6–4.5]
INR	1.5 [1.2–2.3]	2.0 [1.3–2.5]	1.4 [1.1–2.3]	1.5 [1.3–2.0]
Haemoglobin concentration (g/dL)	11.9 [10.6–13.5]	12.6 [10.6–13.2]	11.6 [10.4–14.1]	11.8 [10.3–13.5]
Platelets (10 9 /L)	194 [160–256]	183 [166–233]	232 [166–284]	156 [95–224]
Lactate concentration (mmol/L)	1.6 [1.3–2.5]	1.3 [1.0–1.7] b	2.1 [1.6–3.1]	2.7 [1.4–3.0]
Heart rate (beats/min)	84 [70–102]	76 [64–83] b	85 [69–105] c	102 [98–120]
Mean systemic arterial pressure (mmHg)	70 [64–78]	65 [61–75]	65 [58–75]	72 [69–78]
Central venous pressure (mmHg)	16 [12–21]	19 [11–21]	15 [12–22]	16 [15–19]
sPAP (mmHg)	50 [42–62]	50 [47–58]	51 [36–64]	34 [33–54]
mPAP (mmHg)	35 [26–44]	35 [28–38]	38 [27–46]	27 [26–42]
PCWP (mmHg)	28 [19–30]	28 [27–31] b	29 [16–38] c	23 [19–27]
Cardiac index (L/min/m 2 )	1.7 [1.4–1.9]	1.7 [1.4–2.0]	1.8 [1.5–2.3] c	1.4 [1.1–1.6]
PVR (Dynes/s/cm 5 /m 2 )	183 [117–311]	127 [101–223]	193 [134–255]	131 [86–257]
RV stroke work index (mmHg/m 2 /L)	2.9 [2.0–4.2]	8.7 [6.7–12.5] b	5.6 [4.3–7.4] c	2.7 [2.1–3.5]

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. End-systolic wall stress predicts post-discharge heart failure after acute myocardial infarction
2. Erratum to “High-sensitivity cardiac troponin assays: Answers to frequently asked questions” [Arch Cardiovasc Dis. 108 (2015) 132–149]
3. Sequential management of post-myocardial infarction ventricular septal defects
4. Erratum to “Management of non-traumatic chest pain by the French Emergency Medical System: Insights from the DOLORES registry” [Arch Cardiovasc Dis. 108 (2015) 181–188]

Stay updated, free articles. Join our Telegram channel

Join