Table 1. Twenty seven papers found from Medline and 7 papers found in the Cochrane Central Register of Controlled Trials.
| Author | Year | Study type | Population characteristics | Number of patients | Main outcome | Independent/dependant variables recorded & Follow-up period | Immediate Mx/ICU Mx & outcome (if available) |
|---|---|---|---|---|---|---|---|
| Licker. (65) | 2010 | Prospective cohort | High-risk candidates undergoing AVR were selected amongst a cohort of 108 patients with a predicted risk of mortality >9% (Bernstein-Parsonnet algorithm). 14 patients were excluded due to unobtainable Doppler and TDI measurements | 94 | • Difficulty weaning from CPB—38 paients (40%) out of the 94 patients required inotropic support +/- intraortic balloon pump for weaning from CPB • 3 independent predictors of post-op LV systolic dysfunction include: age (OR 1.1), aortic clamping time (OR 1.04), & Vp (OR 0.65). Amongst pre-op ECHO imaging, Vp was found to be superior in terms of prognostic value and reliability. Best Vp cut off was found to be 40 cm/s (sens 72%, spec 94%) •In line with previous studies, patients who experience post = op LV dysfunction had significantly higher in-hospital mortality (18.4% vs. 3.6%) and increased incidence of serious cardiac events (81% vs. 28.6%) |
Independent variables- PRE-OP—systolic function, LVEF, diastolic function (standard Doppler derived parameter, tissue Doppler imaging and transmitral flow propagation velocity) (Vp) - diastolic function was graded into four classes: normal (E/A>0.8, DT<200 ms, and E’/A’ >1 or S/D 1 to 1.5), impaired relaxation (E/A<0.8, DT>200 ms, IVRT≥100 ms and E’/A’<1 or S/D>1.5), pseudo-normalization (E/A=1 to 2, DT=150 to 200 ms, and E’/A’<1 or S/D<1.2), and restrictive pattern (E/A >2, DT<150 ms and E’/A’ <1 or S/D<0.8) Dependant variables-Post-CPB requirement of PIDs and difficult separation from cardiopulmonary bypass (DSB). | POST- OP - Requirement of PIDs ± mechanical circulatory support * Protocol for weaning off CPB - (I) target HR 70-100, MAP 70-100, LVEDD 2.2-2.8 cm/m2--> if not --> (II) if MAP> 70, assess ventricular function, and treat with vasodilator +/- vasodilators; (III) if MAP<70, assess ventricular function, and treat with vasopressors (PHE, NE) +/- inotropes, (IV) if not responsive, intra-aortic balloon pump or ventricular assist device + inotropes + vasopressors |
| Nakagawa et al. (75) | 2007 | Retrospective analysis | M:F=22:30. Isolated AS. Undergoing AVR | 52 | • No 30-day post-operative death observed in-group 1, 2, 3 cardiovascular 30day post-operative death observed in group 3 • Conclusion: systolic function is demonstrated in this study to be the only significant predictor of immediate post-operative complications & the presence of diastolic dysfunction did not affect the outcome |
Independent variables - patients were divided into 3 groups. Group 1 = normal LVEF normal PWP; Group 2 = normal LVEF + high PWP; Group 3 = low LVEF + normal PWP. Dependent variables- Follow-up period/ parameters measured - cardiovascular death within 30 days and 12 yr/systolic function measured by LVEF, diastolic function measured as PWP (pulmonary wedge pressure) | |
| Denault A et al. (76) | 2006 | Prospective cohort | M:F=112:57. CABG 106 (59%), AVR (30%) | 179 | • Difficult separation from CPB occurred in 19 patients (65.5%) with moderate or severe LVDD compared with 47patient (40.9%) with normal LV diastolic function or mild LVDD (P=0.0173) • Conclusion: moderate and severe LVDD and RVDD can be identified with very good reproducibility, and both degrees of diastolic dysfunction are associated with DSB |
• Independent variables—pre-operative TEE (E/A; pulmonary venous flow S/D; tricuspid annular velocity Et/At) • Dependant variables—post-CPB requirement of PIDs and difficult separation from cardiopulmonary bypass (DSB) • LVDD defined as—separated into mild, moderate, severe based on E/A, S/D, Et/At) |
Difficulty separating from CBP defined as—systolic BP <80 mmHg, diastolic pulmonary artery pressure or PCWP >15 mmHg, or hemodynamic instability resulting in the reintroduction of extracorporeal circuiting or the insertion of an intraortic balloon pump (IABP). These include- nor epinephrine, epinephrine, dobutamine, milrinon, intra-aortic balloon pump, or mechanical support |
| Bernard F et al. (80) | 2001 | Prospective cohort | 66 patients. 52 underwent CABG. 14 underwent CABG+AVR | 66 | • Patient with DD required longer PIDs post-CPB (P=0.003) up to 12 h after surgery • Total CPB time (P=0.004) and ischemic time (P=0.007) were longer in the DD group • Conclusions- Multivariate logistic regression identified female sex, DD, and total CPB time as predictive of difficult weaning and inotropic requirements up to 12 h after surgery • Note. 20 patients (30%) patient had DD- impaired relaxation 50%, pseudonormal 40%, & restrictive 10%. 15 required PIDs immediately post-CPB |
•Independent variables—sex, age, LV DD, ischemic time, total CPB time •Dependent variables—post-CPB requirement of PIDs and difficult separation from CPB |
• Difficulty separating from CBP defined as—systolic BP <80 mmHg, diastolic pulmonary artery pressure or PCWP >15 mmHg, or hemodynamic instability resulting in the reintroduction of extracorporeal circulation or the insertion of an intraortic balloon pump (IABP). These include- norepinephrine, pinephrine, dobutamine, milrinone, intra-aortic balloon pump, or mechanical support • The DD patterns were classified into three groups: (I) impaired relaxation (E/A <1.0, dt > 240 ms, and S > D); (II) pseudo normal (E/A< 1.0 –1.5, dt 160-200 ms, and S<D, with Ar>35 cm/s), and (III) restrictive (E/A1.5, dt<160 ms, S<D, and Ar >35 cm/s). DD is diagnosed using Doppler-derived parameters including: peak velocity of early diastolic filling wave (E) and late diastolic filling wave (A), the ratio of these two velocities (E/A), and deceleration time (dt). Peak systolic (S), diastolic (D), and atrial reversal (Ar) pulmonary vein flow velocities were measured in the left and the right upper pulmonary veins |
| Lund O, et al. (41) | 1997 | Prospective cohort study | • Patients undergoing AVR for AS were recruited. 32 patients underwent AVR + CABG • Mean age-61 years • Minimum postoperative observation period of 5-4 years • Aim: to identify preoperative and intraoperative variables predictive of early and late mortality |
91 | In Cox regression models on crude mortality and specific deaths, a subnormal ejection fraction and a fast filling fraction of < or =45% were the only independent risk factors. Patients with none of these risk factors had normal sex- and age-specific survival, those with any one factor had an early, and those with both factors a massive early and a late excess mortality, with 5-year crude survival of 92%, 77%, and 50%, respectively (P<0.0001) Further analyses indicated that impairment of left ventricular function occurred with increasing muscle mass over two phases: (I) diastolic dysfunction characterized by a pattern of severe relative concentric hypertrophy; (II) the addition of systolic dysfunction characterized by a more dilated, less concentric chamber geometry Conclusions: impaired systolic and diastolic left ventricular function, irrespective of afterload, were decisive independent pre-operative risk factors for early as well as late mortality after aortic valve replacement for aortic stenosis |
Independent variables—2D Doppler ECHO. LV end-diastolic & end-systolic diameter (EDD; ESD), end-diastolic & end-systolic posterior free wall thickness, LVEF, peak ejection rate, peak filling rate, time to peak filling, fast filling, late filling fraction. Dependent variables: 30 days & long term post operative cardiovascular complications/events & mortality | • Indices measures—the peak filling rate in the first half of diastole (1/ EDV.s), time (from end- systole) to peak filling (in the second half of diastole if the peak filling rate here was greater than in first half)-duration of diastole ratio, fast filling fraction (filling in the first half of diastole as a precent of total filling volume), and late filling fraction (filling volume during the PQ interval of the ECG, atrial contraction, as a precent of total filling volume) were used as diastolic function indices • Left ventricular diastolic dysfunction defined as a peak-filling rate of less than 2-86 end-diastolic volumes per second in patients younger than 49 years and of less than 200 in those older than 50 years |
| Hwang M, et al. (81) | 1989 | Prospective cohort study | AVR for AS. Aim: to identify preoperative and intraoperative variables predictive of LV dysfunction 6 months after AVR | 88 | • For the 88 patients with preoperative aortic stenosis, the most powerful predictor of postoperative left ventricular dysfunction in the final multivariate model was preoperative left ventricular ejection fraction (P=0.0001), followed by preoperative myocardial infarction (P=0.012), aortic valve gradient (P=0.020), and incomplete coronary revascularization (P=0.059). Abnormal preoperative left ventricular ejection fraction had a sensitivity of 72% and a specificity of 82% in identifying patients with postoperative left ventricular dysfunction | • LVDD was NOT assessed as a predictor • Pre-op factors assessed - LVEF, previous MI, AV gradient, • Post-op LV dysfunction definition - end-diastolic-volume index <101 mL/m2 or EF<0.5 |
LV dysfunction defined as LVEF <50%, or end-diastolic volume greater than or equal to 101 mL/m2 |
| Stassen P, et al. (77) | 2006 | Retrospective analysis | AVR for AS. Age 62.7±8.9; M:F=135:40; 45 required urgent operation; 103 mechanical 72 biological | 175 | 6.3% operative deaths. Predictors of early death - Logistic regression showed LVEF to be one of the significant predictors for early death. Other predictors of early deaths included: included pervious MI, and poor NYHA class. Predictors of unfavourable late outcome- Low LVEF, age, comorbidities, and operative cause | LVDD was NOT assessed as a predictor |
The first 4 studies listed examined LVDD as an independent predictor of immediate postoperative complications namely post-CPB requirement of PIDs and difficult separation from cardiopulmonary bypass (CPB) in patients undergoing AVR and CABG ± AVR; 3 out of these 4 studies concluded that LV DD was a significant independent predictor of immediate post-operative complication; The 5th study listed examined LVDD as an independent predictor of early (<30 day) and late mortality post- AVR ± CABG; The 6th & 7th study listed did not examine LVDD as an independent predictor in patients undergoing AVR for AS.