Study population

Between 1st January 2010 and 1st January 2012, we retrospectively selected 790 patients who had been referred to our institution, and for whom the primary or associated PMSI (Programme Médicalisé des Systèmes d’Information) diagnosis was DCM. In order to isolate idiopathic DCM, this database was crossed with hospitalization records, clinical and echocardiographic data. Exclusion criteria were defined to precisely distinguish idiopathic DCM from all other causes of LV dysfunction or dilatation, and required a systematic review of coronary angiograms. Ischaemic substrate was defined as any stenosis ≥ 50% in either of the main coronary arteries or in one of their main side branches. Data from patients with insufficient echocardiography or examination performed in an unstable clinical state were not retained for analysis ( n = 92). When the initial cause of hospitalization was acute HF, the echocardiography assessment at discharge was retained. Data from 136 patients presenting with idiopathic DCM were available for analysis.

Follow-up

Follow-up was censored on 1st January 2014, and consisted of a telephone interview with the patient’s general physician. For all patients, we recorded data on the presence and date of occurrence of all major adverse cardiac events (MACE), defined as a composite criterion, including death from cardiac cause (sudden cardiac death, ventricular tachycardia, acute HF), heart transplantation and cardiac-related hospitalizations.

The delay from the first hospitalization between January 2010 and January 2012 to the first event was retained for survival analysis.

Echocardiographic assessment

Transthoracic echocardiograms were recorded on various generations of Vivid systems (GE Vingmed Ultrasound, Horten, Norway), and analyses were performed off-line on an EchoPAC workstation (GE Healthcare, Milwaukee, MI, USA).

Measurements were made according to guidelines . LV ejection fraction (LVEF) was measured according to Simpson’s method. DCM was defined as LVEF ≤ 45% and left ventricular end-diastolic diameter ≥ 55 mm. Global peak systolic longitudinal strain was assessed by two-dimensional speckle imaging in the apical four-chamber view. Diastolic function analysis was based on mitral-pulsed Doppler inflow and tissue-Doppler imaging at the lateral mitral annulus. A restrictive pattern was defined as E-wave deceleration time (DT) < 145 ms – the median value and in concordance with previous studies . Left and right atrial areas were measured from the apical four-chamber view.

TAPSE was measured by M-mode, after two-dimensional echocardiography guidance at the lateral tricuspid annulus, as the maximal systolic excursion. Tissue Doppler imaging at the tricuspid annular free wall allowed the assessment of S-wave velocity. Fractional area change (FAC) in the right ventricle was obtained according to image quality. Systolic pulmonary artery pressure was calculated from tricuspid regurgitation (TR) flow added to right atrial pressure estimated from the inferior vena cava. Right and left atrial areas were measured at end-ventricular systole in a four-chamber view.

Cardiac magnetic resonance imaging assessment

Thirty-seven patients (27.2%) underwent cardiac magnetic resonance imaging (MRI), performed using a 1.5-Tesla MRI scanner (Signa HDxt; GE Healthcare, Milwaukee, WI, USA) connected to an eight-element cardiac phased-array surface coil. Cine images were obtained with a steady-state free precession sequence in base-to-apex contiguous short-axis slices for volume and mass quantification.

Statistical analysis

All continuous variables are described as means ± standard deviations; all categorical variables are described with absolute and relative frequencies.

Intra- and interobserver variabilities of measure were assessed using a Bland-Altman diagram for TAPSE. A receiver operating characteristic curve was built to define an appropriate threshold for TAPSE, based on gold-standard cardiac MRI-derived RV ejection fraction (RVEF). The area under the curve of the model was 0.63, and the most pertinent value was 15.4 mm (Se 0.86; 1–Sp 0.44). RV dysfunction was defined by TAPSE ≤ 15 mm, and patients were divided into two groups according to the presence (group 1) or absence (group 2) of RV dysfunction. Comparisons according to the presence or absence of RV dysfunction and the occurrence of MACE were realized with Student’s t test for continuous variables and the Chi ² test for discrete variables.

Bivariate and multivariable logistic regressions were used to identify determinants of RV dysfunction. In the multivariable model, LVEF, LV end-diastolic diameter and age were forced, and a stepwise variable selection method, with a P value < 0.1 to enter the stepwise selection and a P value < 0.05 to remain in the final model, was applied. Results are expressed as odds ratios (ORs) with 95% confidence intervals (CIs).

To assess the prognostic effect of RV dysfunction on the occurrence of MACE, several analyses were used. Survival curves according to the presence or absence of RV dysfunction were generated by the Kaplan-Meier method, and were compared using log-rank tests. Next, bivariate and multivariable Cox regression analyses were used to assess predictors of MACE, particularly including RV dysfunction as a dependent variable.

Finally, to take into account the likely imbalance in the baseline characteristics of patients with or without RV dysfunction and, more specifically, to address the effect of the extent of LV dysfunction on RV dysfunction, we used a propensity analysis . Propensity scores of RV dysfunction were calculated for each patient, including variables predictive of RV dysfunction obtained by multivariable logistic regression and variables supposed to influence both the presence of RV dysfunction and patient outcome (LVEF, LV end-diastolic diameter and age). A Cox multivariable regression stratified by quintiles of propensity scores was performed to predict the occurrence of MACE. Results are expressed as hazard ratios (HRs) with 95% CIs.

For all tests, P < 0.05 was considered significant. All statistical analyses were performed using SAS software, version 9.3 (SAS Institute, Cary, NC, USA). Variables with more than half non-missing values and with correlation with other co-variables < 0.6 were selected for all bivariate analyses. Significant variables in the bivariate analyses were candidates for all multivariable analyses.

Baseline characteristics

Patient characteristics are listed in Table 1 . The cause of initial hospitalization was acute HF for 39 patients (28.7%); the others (71.3%) were considered to be in a stable haemodynamic state at inclusion, and most were hospitalized for device implantation, coronary angiography or cardiac rehabilitation. The median duration of disease at inclusion was 2.1 years, with no significant intergroup difference (1.5 years vs 2.6 years for groups 1 and 2, respectively; U test P = 0.66).

Table 1

Patient characteristics according to right ventricular dysfunction.

	All patients	Group 1 (TAPSE ≤ 15 mm)	Group 2 (TAPSE > 15 mm)	P
Baseline features
Age (years)	59.01 ± 13.24	58.50 ± 14.40	59.18 ± 12.89	0.80
Men	100 (73.5)	29 (85.3)	71 (69.6)	0.07
Prior acute HF	82 (60.2)	28 (82.4)	54 (54.5)	0.0004 a
NYHA functional status				0.20
I–II	78 (57.3)	7 (47.1)	33 (60.8)
III	48 (35.3)	13 (38.2)	35 (34.3)
IV	10 (7.4)	5 (14.7)	5 (4.9)
Systolic blood pressure (mmHg)	118.81 ± 21.55	111.83 ± 16.80	121.53 ± 22.66	0.036 a
Sinus rhythm	115 (84.6)	24 (70.6)	91 (89.2)	0.009 a
BNP (pg/mL)	751.18 ± 912.60	1020.93 ± 999.55	627.36 ± 849.92	0.06
Peak VO 2 (L/min/m 2 )	21.38 ± 6.49	23.34 ± 6.20	20.69 ± 6.5	0.14
VE/VCO 2 (%)	35.52 ± 7.01	34.72 ± 5.45	35.84 ± 7.57	0.57
Therapy
ACEi	120 (88.2)	29 (85.3)	91 (89.2)	0.54
> 50% maximal dose	101 (74.3)	21 (72.4)	80 (87.9)	0.046 a
Beta-blockers	123 (90.4)	32 (94.1)	91 (89.2)	0.40
> 50% maximal dose	68 (50.0)	12 (38.7)	56 (63.6)	0.016 a
Aldosterone antagonists	50 (36.8)	16 (47.1)	34 (33.3)	0.15
Diuretics	108 (79.4)	31 (91.2)	77 (75.5)	0.05 a
> 80 mg/day furosemide	46 (33.8)	17 (54.8)	29 (38.2)	0.11
ICD	58 (42.6)	11 (32.4)	47 (46.1)	0.16
CRT	42 (30.9)	8 (23.5)	34 (33.3)	0.28
Echocardiography: LV
LVEDD (mm)	65.83 ± 9.21	64.68 ± 8.46	66.22 ± 9.46	0.40
LVEF (%)	27.51 ± 8.71	24.24 ± 7.93	28.60 ± 8.73	0.014 a
Four-chamber LVEF (%)	26.42 ± 9.29	22.96 ± 8.43	27.58 ± 9.31	0.012 a
Four-chamber LPS (%)	–8.00 ± 3.98	–5.62 ± 2.54	–8.73 ± 4.07	< 0.001 a
LVOT TVI (cm/s)	16.02 ± 5.64	12.28 ± 4.30	17.31 ± 5.48	< 0.001 a
E-wave DT (ms)	154.94 ± 59.70	119.89 ± 34.42	166.15 ± 61.81	< 0.001 a
LA surface (cm 2 )	25.32 ± 6.95	28.60 ± 7.82	24.24 ± 6.32	0.002 a
MR > grade 2	45 (33.0)	15 (44.1)	30 (29.4)	0.12
Echocardiography: RV
TAPSE (mm)	18.6 ± 5.4	11.9 ± 2.1	20.8 ± 4.2
RA surface (cm 2 )	20.84 ± 9.71	26.85 ± 13.64	18.82 ± 6.98	< 0.001 a
TR > grade 2	17 (12.5)	8 (23.5)	9 (8.9)	0.13
sPAP (mmHg)	36.94 ± 12.21	38.83 ± 10.64	36.29 ± 12.70	0.34
S-wave (cm/s)	9.71 ± 2.87	7.76 ± 1.73	10.39 ± 2.89	< 0.001 a
FAC	0.40 ± 0.14	0.30 ± 0.12	0.43 ± 0.13	0.001 a
Cardiac MRI
LVEF (%)	28.34 ± 11.20	24.82 ± 13.24	29.59 ± 10.34	0.23
RVEF (%)	42.99 ± 17.52	32.55 ± 17.67	47.41 ± 15.77	0.016 a

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

Share this:

• X
• Facebook

Related

Related posts:

1. Long-term prognosis of left-sided native-valve Staphylococcus aureusendocarditis
2. Prediction of new onset of resting pulmonary arterial hypertension in systemic sclerosis
3. Novel materials and devices in the transcatheter management of congenital heart diseases – the future comes slowly (part 1)
4. A French patient cohort with low prevalence of diabetes mellitus and Takotsubo syndrome

Stay updated, free articles. Join our Telegram channel

Join