Outcome of cardiac surgery in patients with low preoperative ejection fraction

Risk of conventional cardiac surgery among patients with severe left ventricular dysfunction in the era of mechanical circulatory support

BACKGROUND

Despite suggestions that severe left ventricle dysfunction may warrant selection of durable mechanical circulatory support over conventional surgery, comparative studies are lacking due to incomplete characterization of patients at highest risk after conventional surgery. We sought to define subsets of patients with severe left ventricle dysfunction who are at greatest mortality risk following conventional cardiac surgery.

METHODS

We studied 892 patients aged ≥ 18 years who underwent conventional coronary or valve surgery from 1993 to 2014, with preoperative ejection fraction ≤ 25%. Exclusions were transcatheter interventions, major concomitant procedures, active endocarditis, and prior/concurrent durable mechanical circulatory support use. Logistic and Cox regression identified determinants of early and late mortality.

RESULTS

Median age was 70 years (interquartile range, 62-76 years), 46% (n = 411) had New York Heart Association (NYHA) functional class IV symptoms, and 16% (n = 142) had undergone prior surgery. Operative mortality was 7.5%. NYHA functional class IV (odds ratio [OR], 1.88; P = .033), prior cardiac surgery (OR, 2.13; P = .017), peripheral vascular disease (OR, 2.55; P = .001), emergency status (OR, 2.68; P = .024), and intra-aortic balloon pump use (OR, 4.95; P < .001) independently predicted operative death. Risk imparted by presence of both NYHA functional class IV symptoms and prior surgery was additive, with a 4-fold increase in early mortality risk (OR, 3.95; P = .003). Prior surgery increased the hazard of late death by 60% (P < .001). In patients without prior surgery, late mortality was greatest in those aged ≥ 70 years (hazard ratio, 1.86; P < .001), especially if NYHA functional class IV symptoms were concurrently present (hazard ratio, 2.25; P < .001). Surgery type (coronary artery bypass graft surgery, aortic valve surgery, or mitral valve surgery) did not predict long-term outcome.

CONCLUSIONS

In patients referred for conventional surgery with an ejection fraction ≤ 25%, prior cardiac surgery, and/or NYHA functional class IV symptoms-particularly in those aged ≥ 70 years-confer significant and sustained survival disadvantages. Such high-risk subsets may benefit from durable mechanical circulatory support consideration.

Effect of Levosimendan on Renal Outcome in Cardiac Surgery Patients With Chronic Kidney Disease and Perioperative Cardiovascular Dysfunction: A Substudy of a Multicenter Randomized Trial

OBJECTIVE

Acute kidney injury (AKI) occurs frequently after cardiac surgery. Levosimendan might reduce the incidence of AKI in patients undergoing cardiac surgery. The authors investigated whether levosimendan administration could reduce AKI incidence in a high-risk cardiac surgical population.

DESIGN

Post hoc analysis of a multicenter randomized trial.

SETTING

Cardiac surgery operating rooms and intensive care units of 14 centers in 3 countries.

PARTICIPANTS

The study comprised 90 patients who underwent mitral valve surgery with an estimated glomerular filtration rate <60 mL/min/1.73 m² and perioperative myocardial dysfunction.

INTERVENTIONS

Patients were assigned randomly to receive levosimendan (0.025-0.2 μg/kg/min) or placebo in addition to standard inotropic treatment.

MEASUREMENTS AND MAIN RESULTS

Forty-six patients were assigned to receive levosimendan and 44 to receive placebo. Postoperative AKI occurred in 14 (30%) patients in the levosimendan group versus 23 (52%) in the placebo group (absolute difference -21.8; 95% confidence interval -41.7 to -1.97; p = 0.035). The incidence of major complications also was lower (18 [39%]) in the levosimendan group versus that in the placebo group (29 [66%]) (absolute difference -26.8 [-46.7 to -6.90]; p = 0.011). A trend toward lower serum creatinine at intensive care unit discharge was observed in the levosimendan group (1.18 [0.99-1.49] mg/dL) versus that in the placebo group (1.39 [1.05-1.76] mg/dL) (95% confidence interval -0.23 [-0.49 to 0.01]; p = 0.07).

CONCLUSIONS

Levosimendan may improve renal outcome in cardiac surgery patients with chronic kidney disease undergoing mitral valve surgery who develop perioperative myocardial dysfunction. Results of this exploratory analysis should be investigated in future properly designed randomized controlled trials.

Aortic valve replacement in patients with a left ventricular ejection fraction ≤35% performed via a minimally invasive right thoracotomy

BACKGROUND

We evaluated the outcomes of patients with aortic valve pathology in the setting of a left ventricular ejection fraction ≤35% who underwent minimally invasive aortic valve replacement (AVR), with or without concomitant mitral valve (MV) surgery.

METHODS

All minimally invasive AVR in patients with a left ventricular ejection fraction ≤35%, performed via a right thoracotomy for aortic stenosis or regurgitation between January 2009 and March 2013, were retrospectively evaluated. The operative characteristics, perioperative outcomes, and 30-day mortality were analyzed.

RESULTS

There were 75 patients identified: 51 who underwent isolated AVR, and 24 who had combined AVR plus MV surgery for moderate to severe mitral regurgitation. In patients undergoing MV surgery, there were 22 (91.7%) MV repairs [ring annuloplasty =7 (37.5%), transaortic edge-to-edge repair =15 (62.5%)], and 2 (8.3%) replacements. No patient required conversion to sternotomy for inadequate surgical field exposure. The median total mechanical ventilation time and intensive care unit length of stay were 14 (IQR, 8-20) and 42 hours (IQR, 26-93 hours) in the isolated AVR group, and 16.5 hours (IQR, 12-61.5 hours) and 95.5 hours (IQR, 43.5-159 hours) in the AVR plus MV surgery group, respectively. The most common post-operative complication was new-onset atrial fibrillation, which occurred in 15 (29.4%) isolated AVR and 4 (16.7%) AVR plus MV surgery patients. The median hospital length of stay and 30-day mortality was 7 days (IQR, 5-12 days) and 1 (2%) in the isolated AVR group, and 10.5 days (IQR, 5-21 days) and 1 (4.3%) for AVR plus MV surgery.

CONCLUSIONS

In patients with aortic valve pathology in the setting of a left ventricular ejection fraction ≤35%, minimally invasive AVR can be performed, with or without concomitant MV surgery, with a low morbidity and mortality.

Immediate increase of cardiac output after percutaneous mitral valve repair (PMVR) determined by echocardiographic and invasive parameters: Patzelt: Increase of cardiac output after PMVR

BACKGROUND

Successful percutaneous mitral valve repair (PMVR) in patients with severe mitral regurgitation (MR) causes changes in hemodynamics. Echocardiographic calculation of cardiac output (CO) has not been evaluated in the setting of PMVR, so far. Here we evaluated hemodynamics before and after PMVR with the MitraClip system using pulmonary artery catheterization, transthoracic (TTE) and transesophageal (TEE) echocardiography.

METHODS

101 patients with severe MR not eligible for conventional surgery underwent PMVR. Hemodynamic parameters were determined during and after the intervention. We evaluated changes in CO and pulmonary artery systolic pressure before and after PMVR. CO was determined with invasive parameters using the Fick method (COi) and by a combination of TTE and TEE (COe).

RESULTS

All patients had successful clip implantation, which was associated with increased COi (from 4.6±1.4l/min to 5.4±1.6l/min, p<0.001). Furthermore, pulmonary artery systolic pressure (PASP) showed a significant decrease after PMVR (47.6±16.1 before, 44.7±15.5mmHg after, p=0.01). In accordance with invasive measurements, COe increased significantly (COe from 4.3±1.7l/min to 4.8±1.7l/min, p=0.003). Comparing both methods to calculate CO, we observed good agreement between COi and COe using Bland Altman plots.

CONCLUSIONS

CO increased significantly after PMVR as determined by echocardiography based and invasive calculation of hemodynamics during PMVR. COe shows good agreement with COi before and after the intervention and, thus, represents a potential non-invasive method to determine CO in patients with MR not accessible by conventional surgery.