Independent Effect of Low Flow on Outcomes in Patients Undergoing Aortic Valve Replacement for Severe Aortic Stenosis

Transcription

1 Circ J 2018; 82: doi: /circj.CJ ORIGINAL ARTICLE Valvular Heart Disease Independent Effect of Low Flow on Outcomes in Patients Undergoing Aortic Valve Replacement for Severe Aortic Stenosis Suguru Miyazaki, MD; Kenji Kuwaki, MD, PhD; Kan Kajimoto, MD, PhD; Satoshi Matsushita, MD, PhD; Shizuyuki Dohi, MD, PhD; Taira Yamamoto, MD, PhD; Hiroaki Hata, MD, PhD; Atsushi Amano, MD, PhD Background: Low flow (LF; i.e., reduced left ventricular stroke volume index <35 ml/m 2 ) can occur with severe aortic stenosis (AS). However, few studies have investigated the effects of LF on early and late outcomes after aortic valve replacement (AVR) for severe AS. Methods and Results: In all, 285 severe AS patients undergoing isolated AVR at Juntendo University Hospital between August 2002 and August 2015 were enrolled in the study. In this cohort, 52 patients (18%) had LF. Compared with patients with normal flow (NF) severe AS, early postoperative mortality (9.6% vs. 1.2%; P=0.006), gastrointestinal complications (5.7% vs. 0.8%; P=0.04), and the duration of the intensive care unit (ICU) stay (81.7 vs h; P=0.02) were increased in LF patients with severe AS. LF was an independent predictor of early mortality (Model A, odds ratio [OR] 6.81, P=0.01; Model B, OR 6.69, P=0.01) and composite complications (Model A, OR 2.44, P=0.02). In propensity score-matched comparisons, early mortality (12.8% vs. 0%; P=0.02), composite complications (28.2% vs. 10.2%; P=0.04), and duration of ICU stay (97.4 vs h; P=0.006) were significantly increased in LF than NF patients. Conclusions: LF, as an important independent risk factor for postoperative mortality and morbidity, should be included in risk stratification and assessment in severe AS patients. Key Words: Aortic stenosis; Aortic valve replacement; Low flow Low flow (LF), defined as a reduced left ventricular stroke volume index (SVI) <35 ml/m 2, has been shown to be an important factor in the evaluation of severe aortic stenosis (AS; aortic valve area [AVA] 1.0 cm 2 ). 1 3 LF may occur in patients with either reduced or preserved left ventricular ejection fraction (EF), known as classical and paradoxical LF, respectively. LF is often associated with a low gradient (LG; mean transvalvular gradient <40 mmhg) of the aortic valve despite severe AS. 4 Patients with classical LF have a poor prognosis with medical treatment but a higher operative mortality than patients with normal flow (NF) AS. 1 Patients with paradoxical LF also have a worse prognosis with medical therapy than with surgical aortic valve replacement (AVR), despite higher operative mortality. 1,2,5 LF, but not low EF or LG, has been shown to be an independent predictor of early and late mortality following transcatheter AVR (TAVR) in high-risk severe AS patients. 6,7 However, the independent effect of LF on outcomes after surgical AVR is not well known. 8,9 Therefore, the aim of the present study was to examine the effects of LF, LG, and low EF on early and late outcomes following AVR for severe AS. Methods In all, 285 consecutive patients who underwent AVR for severe AS at Juntendo University Hospital between August 2002 and August 2015 were included in the present cohort study. Patients were excluded from the study if they underwent a concomitant valve procedure (mitral and/or tricuspid) or coronary artery bypass grafting (CABG). Patients with moderate AS (AVA >1.0 cm 2 ) and concomitant aortic and/or mitral valve regurgitation were also excluded from the study. Baseline characteristics of the study population are given in Table 1. This study was approved by the Medical Ethics Committee of Juntendo University. Early mortality and complications after AVR were defined as those within 30 days of surgery or as those occurring at any time before discharge from hospital. Completeness of follow-up in this study was 100%, and the Received August 17, 2017; revised manuscript received April 1, 2018; accepted April 11, 2018; released online May 25, 2018 Time for primary review: 15 days Department of Cardiovascular Surgery, Juntendo University, Tokyo, Japan Mailing address: Kenji Kuwaki, MD, Department of Cardiovascular Surgery, Juntendo University, Hongo, Bunkyo-ku, Tokyo , Japan. kuwakikj@yahoo.co.jp ISSN All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 2200 MIYAZAKI S et al. Table 1. Patient Characteristics According to the SVI Before and After Propensity Score Matching Before propensity score matching After propensity score matching Variable LF (SVI <35 ml/m 2 ; NF (SVI 35 ml/m 2 ; LF (SVI <35 ml/m ; NF (SVI 35 ml/m 2 ; n=52) n=233) Age (years) 72.2± ± ± ± Age 80 years 12 (23.1) 47 (20.1) (17.9) 14 (35.8) 0.08 Male 27 (51.9) 111 (47.6) (58.9) 15 (38.4) 0.07 LVEF <50% 12 (23.1) 18 (7.7) (25.6) 11 (28.2) 0.79 Mean AV gradient <40 mmhg 4 (7.6) 39 (16.7) (7.6) 8 (20.5) 0.11 Aortic valve area (cm 2 ) 0.66± ± ± ± NYHA Class III IV 11 (21.1) 34 (14.5) (23.1) 13 (33.3) 0.31 Chronic dialysis 9 (17.3) 18 (7.7) (23.1) 4 (10.2) 0.12 Hypertension 33 (63.4) 160 (68.6) (64.1) 26 (66.6) 0.81 Diabetes mellitus 8 (15.3) 49 (21.0) (15.3) 7 (17.9) 0.76 AF 4 (7.6) 5 (2.1) (5.1) 1 (2.5) 0.51 PVD 3 (5.7) 8 (3.4) (5.1) 2 (5.1) 0.69 History of stroke 3 (5.7) 17 (7.2) (5.1) 1 (2.5) 0.51 COPD 4 (7.6) 23 (9.8) (10.2) 4 (10.2) 0.64 LVDd (mm) 43.1± ± ± ± LVDs (mm) 30.2± ± ± ± Emergency/urgent surgery 1 (1.9) 2 (0.8) (2.5) 1 (2.5) 0.75 EuroSCORE II 2.5± ± ± ± Data are given as the mean ± SD or as n (%). AF, atrial fibrillation; AV, aortic valve; COPD, chronic obstructive lung disease; LF, low flow; LVDd, left ventricular end-diastolic dimension; LVDs, left ventricular end-systolic dimension; LVEF, left ventricular ejection fraction; NF, normal flow; NYHA, New York Heart Association; PVD, peripheral vascular disease; SVI, stroke volume index. mean follow-up period was 5.3 years. All patients underwent comprehensive 2-dimensional and Doppler echocardiographic examinations. Left ventricular EF was calculated using the Simpson s biplane method. AVA was calculated using the continuity equation. Stroke volume was measured by a volumetric method based on the Teichholz formula. Low EF was defined as left ventricular EF <50%. LF was defined as left ventricular SVI <35 ml/m 2. LG was defined as a mean transvalvular gradient <40 mmhg. All surgical procedures were performed through a full or partial sternotomy using cardiopulmonary bypass with systemic normothermia or mild systemic hypothermia. Myocardial protection was achieved with combined antegrade and retrograde cold blood cardioplegia. The aortic valve prostheses used in this study included 225 biological valves (Carpentier-Edwards, n=152; Trifecta, n=47; Mosaic, n=18; Mitroflow, n=8) and 60 mechanical valves (St. Jude Medical, n=24; OnX, n=23; ATS, n=11; CarboMedics, n=2). Anticoagulation with warfarin was commenced 1 2 days after the operation for patients who received a mechanical prosthesis. Statistical Analysis Categorical variables are given as percentages and were compared between groups using a Chi-squared test or Fisher s exact test. Continuous variables are reported as the mean ± SD and were compared between groups using Student s t-test. Kaplan-Meier curves and the log-rank test were used for long-term survival analysis. P<0.05 was considered statistically significant. Multivariable analysis using multiple logistic regression was performed to identify independent predictors of early mortality. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs) and s. A Cox proportional hazards multivariable model was used to identify predictors of long-term mortality, with results presented as hazard ratios (HRs) with 95% CIs and s. To reduce differences in preoperative baseline characteristics, propensity score analysis was performed using a 1:1 nearest neighbor matching algorithm with a ±0.05 caliper and no replacement. All variables listed in Table 1 were included in the analysis. Propensity score matching produced 39 matched pairs with similar patient characteristics. The significance of differences between the 2 matched groups was determined using the Chi-squared test or Fisher s exact test for categorical variables and the t-test or Mann-Whitney U-test for continuous variables. Statistical analyses were performed using SPSS version 18.0 (IBM, Armonk, NY, USA). Results Of the 285 patients included in this study, 52 (18%) were in the LF group and 233 (82%) were in the NF group. The baseline and clinical data in this study population divided into LF and NF patients and before and after propensity matching are given in Table 1. Age and gender were similar between the LF and NF patients. LF patients had a higher incidence of low EF, a higher prevalence of chronic dialysis, a higher surgical risk of AVR estimated by the EuroSCORE II, and a smaller left ventricular end-diastolic dimension than NF patients (Table 1). After performing propensity score matching in the overall patient population, 39 LF patients were matched to the NF patients (. In this matched cohort, the differences between LF and NF patients in the preoperative baseline characteristics were no longer significant (Table 1). There were 8 early deaths, giving an early mortality rate of 2.8% (Table 2). The causes of early deaths included

3 Low-Flow Severe AS and Surgery 2201 Table 2. Early Mortality and Morbidity After AV Replacement According to the SVI Before and After Propensity Score Matching Before propensity score matching After propensity score matching LF (SVI <35 ml/m 2 ; n=52) NF (SVI 35 ml/m 2 ; n=233) LF (SVI <35 ml/m 2 ; NF (SVI 35 ml/m 2 ; Early death 5 (9.6) 3 (1.2) (12.8) 0 (0) 0.02 Stroke 1 (1.9) 6 (2.5) (2.5) 0 (0) 0.51 New renal failure (CHDF/HD) 4 (7.6) 10 (4.2) (7.6) 2 (5.1) 0.51 Respiratory failure 6 (11.5) 10 (4.2) (17.9) 1 (2.5) 0.03 GI complications 3 (5.7) 2 (0.8) (7.6) 0 (0) 0.12 Systemic infection 1 (1.9) 2 (0.8) (5.1) 0 (0) 0.24 Composite complications 10 (19.2) 25 (10.7) (28.2) 4 (10.2) 0.04 EOAI (cm 2 /m 2 ) 1.15± ± ± ± EOAI 0.85 cm 2 /m 2 3 (5.7) 22 (9.4) (5.1) 6 (15.3) 0.26 ICU stay (h) 81.7± ± ± ± POLOS (days) 14.0± ± ± ± Data are given as the mean ± SD or as n (%). CHDF, continuous hemodiafiltration; EOAI, effective orifice area index; GI, gastrointestinal; HD, hemodialysis; ICU, intensive care unit; POLOS, postoperative length of stay. Other abbreviations as in Table 1. Table 3. Univariate and Multivariate Predictors of Early Mortality Variables Univariate analysis Multivariate analysis: Model A Multivariate analysis: Model B OR (95% CI) OR (95% CI) OR (95% CI) LF (SVI <35 ml/m 2 ) 8.15 ( ) ( ) ( ) 0.01 Chronic dialysis 6.32 ( ) ( ) 0.06 EuroSCORE II 3% 5.02 ( ) ( ) 0.08 CI, confidence interval; LF, low flow; OR, odds ratio; SVI, stroke volume index. EuroSCORE II was not included in the multivariate analysis in Model A, but was entered into the analysis in Model B. Figure 1. Early mortality according to flow and EuroSCORE II. Patients with low flow and a EuroSCORE II 3% showed very high early mortality compared with other patients. cardiac failure in 2 patients, gastrointestinal complications in 2 patients, cerebral infarction in 2 patients, fatal arrhythmia in 1 patient, and mediastinitis in 1 patient. In univariate analysis, variables associated with increased early mortality were LF (P=0.006), chronic dialysis (P=0.03), and EuroSCORE II 3% (P=0.03; Table 3). On multivariate analysis, LF remained an independent predictor of early mortality (OR 6.81, 95% CI , P=0.01; Table 3, Model A). EuroSCORE II was not included in the multivariate analysis in Model A. When EuroSCORE II was entered into the multivariate analysis (instead of its individual predictors), LF (OR 6.69, 95% CI , P=0.01) was found to be an independent predictor of early mortality and EuroSCORE II 3% was identified as a marginal risk factor (OR 3.72, 95% CI , P=0.08; Table 3, Model B). Figure 1 shows the early mortality according to flow and EuroSCORE II. Patients with LF and a EuroSCORE II 3% had the highest early mortality of 13.3%, whereas those with NF and a EuroSCORE II <3% had the lowest mortality of 0.5%. In

4 2202 MIYAZAKI S et al. Table 4. Univariate and Multivariate Predictors of Long-Term Mortality Variables Univariate analysis Multivariate analysis: Model A Multivariate analysis: Model B HR (95% CI) HR (95% CI) HR (95% CI) Chronic dialysis 7.26 ( ) ( ) ( ) Low EF <50% 3.13 ( ) ( ) ( ) 0.44 EuroSCORE II 3% 1.92 ( ) ( ) 0.03 CI, confidence interval; EF, ejection fraction; HR, hazard ratio. Table 5. Univariate and Multivariate Predictors of Postoperative Composite Complications Variables Univariate analysis Multivariate analysis: Model A Multivariate analysis: Model B OR (95% CI) OR (95% CI) OR (95% CI) LF (SVI <35 ml/m 2 ) 2.49 ( ) ( ) ( ) 0.07 Age 80 years 2.74 ( ) ( ) 0.01 PVD 4.17 ( ) ( ) 0.08 EuroSCORE II 3% 4.14 ( ) ( ) Abbreviations as in Tables 1,3. Figure 2. Incidence of composite complications following aortic valve replacement according to flow and EuroSCORE II. Patients with a low flow and a EuroSCORE II 3% showed a higher incidence of composite complications compared with other patients. the propensity score-matched comparison between LF and NF patients, early mortality (P=0.02) and composite complication rate (P=0.04) were significantly higher in LF than NF patients (Table 2). The causes of early deaths in the propensity-matched LF group included cardiac failure in 1 patient, gastrointestinal complication in 2 patients, and cerebral infarction in 2 patients. We compared the length of intensive care unit (ICU) stay (in hours) and postoperative length of hospital stay (POLOS; in days) between patients with LF and NF. The ICU stay was significantly longer in LF than NF patients both before (P=0.02) and after (P=0.006) propensity score matching (Table 2). Although not statistically significant, there was a trend towards a longer POLOS after AVR in the LF than NF group in the propensity score-matched comparison (P=0.08; Table 2). Overall, there were 28 deaths, including 20 late deaths (3 LF patients, 17 NF patients) during a mean follow-up period of 5.3±3.2 years. The causes of late deaths in the 3 LF patients included malignancy in 2 and pneumonia in 1. The causes of death among the 17 NF patients were unknown in 10 patients, malignancy in 2 patients, sudden death in 1patient, pneumonia in 1patient, systemic infection in 1patient, stroke in 1patient, and amyotrophic lateral sclerosis in 1patient. Although the follow-up data regarding the reasons for late death are incomplete because there were 10 unknown late deaths in the NF group, non-cardiac causes of late death were more common than cardiac causes, particularly in 3 LF patients, in whom all deaths were related to non-cardiac diseases. The 10-year survival rates were similar between the LF and NF patients (80% vs. 83%, respectively; P=0.20). In univariate analysis, significant preoperative factors of late mortality were chronic dialysis (P=0.001), low EF (P=0.004), and EuroSCORE II 3% (P=0.001; Table 4). In multivariate analysis, the predictor independently associated with increased risk of late mortality was chronic dialysis (HR 6.05, 95% CI , P=0.001; Table 4, Model A). When EuroSCORE II was entered into the analysis, EuroSCORE II 3% (HR 1.58, 95% CI , P=0.03) was identified as an independent risk factor of long-term mortality (Table 4, Model B). Postoperative major complications were defined as the presence of the following: stroke (n=7), new renal failure (n=14), respiratory failure (n=17), gastrointestinal complications (n=5), and systemic infection (n=4). Thirty-seven (14%) patients had at least 1 major postoperative complication. There were no differences in any of the individual complications (stroke, new renal failure, respiratory failure, and systemic infection) between LF and NF patients, but the rate of gastrointestinal complications was significantly higher in patients with LF than those with NF (5.7% vs. 0.8%, respectively; P=0.04; Table 2). Univariate analysis of composite complications showed that LF (P=0.01), patient age 80 years (P=0.003), peripheral vascular disease (P=0.03), and EuroSCORE II 3% (P=0.001) were risk factors (Table 5). Multivariate analysis revealed that LF (OR 2.44, 95% CI , P=0.02) and patient age 80 years (OR 2.54,

5 Low-Flow Severe AS and Surgery % CI , P=0.01) were associated with increased postoperative composite complications (Table 5, Model A). In Model B, which included EuroSCORE II, a EuroSCORE II 3% (OR 3.77, 95% CI , P=0.001) was found to be an independent predictor of postoperative composite complications, and LF was identified as a marginal risk (OR 2.09, 95% CI , P=0.07; Table 5, Model B). Evaluation of the combination of these 2 risk factors (LF and EuroSCORE II) revealed that LF severe AS patients with EuroSCORE II 3% had a high risk of composite complications compared with other patients (40% vs. 11%, respectively; P=0.001; Figure 2). In the propensity scorematched comparison between LF and NF patients, the composite complication rate was significantly higher in LF than NF patients (28% vs. 10%, respectively; P=0.04; Table 2). Major postoperative complications significantly increased the likelihood of non-home discharge and overall death after surgery. Patients who developed postoperative complications were 9-fold more likely to experience non-home discharge (43.2% vs. 4.4%; P=0.0001; Figure 3) and were significantly less likely to survive at 10 years (56% vs. 94%; P=0.0001) than those without postoperative complications. Discussion There are several major findings of the present study: (1) LF is relatively common in patients with severe AS undergoing AVR, occurring in 18% (52/285 patients); (2) LF is an independent predictor of early mortality and postoperative major complications following AVR in multivariate analysis; and (3) in the propensity scorematched comparison, early mortality (12.8% vs. 0%; P=0.02), composite complications (28.2% vs. 10.2%; P=0.04), and duration of ICU stay (97.4 vs h; P=0.006) were significantly increased in LF than NF patients. LF has recently been identified as an independent predictor of early and late mortality in high-risk patients with severe AS undergoing TAVR. 6,10 The present study demonstrated that LF was significantly associated with early mortality after AVR in univariate analysis, and this association remained significant after adjustment for several strong risk factors, including dialysis and EuroSCORE II 3%. It is well recognized that low EF is an important risk factor for increased mortality after AVR in AS patients, and EF is therefore included and evaluated in every risk score system for cardiac operations However, the present study identified no significant association between low EF and higher early mortality in univariate analysis. Similar results have been reported in other TAVR studies, 6,10 where LF, but not low EF or LG, was found to be an independent predictor of mortality following TAVR in high-risk patients with severe AS. A possible explanation for these results is that a direct measure of left ventricular pump function (stroke volume) is a more important risk factor of outcome than EF, which may underestimate the extent of left ventricular systolic dysfunction in severe AS patients who often show concentric left ventricular hypertrophy. In severe AS patients, LF is considered to be due to several factors, including ventricular concentric remodeling with decreased left ventricular cavity size, higher global left ventricular afterload, a reduction in left ventricular compliance and filling with preserved EF, or severely Figure 3. Non-home discharge according to the presence of postoperative composite complications. Patients who developed a postoperative complication had significantly higher rates of non-home discharge than those without complications. depressed EF with left ventricular dilatation due primarily to associated ischemic heart disease. 10,18 One important finding of the present study is that LF was found to be an independent predictor of early mortality even after adjustment for EuroSCORE II (Table 3). LF is not included in the EuroSCORE II, and therefore consideration of LF in addition to the EuroSCORE II may be useful for improving surgical risk stratification in AVR for severe AS (Figure 1). Some contemporary studies regarding AS reported that a more advanced stage of myocardial dysfunction in the presence of severe concentric myocardial hypertrophy, regardless of left ventricular EF, would be a possible reason for increased operative mortality in LF severe AS patients. 1,2,19 Herrmann et al 19 revealed more extensive myocardial fibrosis in patients with LF than those with NF. Longitudinal myocardial shortening is reduced to a larger extent in LF patients due to more advanced fibrosis in the subendocardial layer, where fibers are oriented longitudinally. Hence, LF severe AS would be associated with more advanced impairment of myocardial function. Of note, 5 of the 8 patients who died early after AVR were LF patients (Table 2). Two patients from the LF group died of low cardiac output and non-occlusive mesenteric ischemia (gastrointestinal complication), which may be related to the LF characterized by reduced stroke volume and advanced myocardial dysfunction. Neither of these patients received intra-aortic balloon pumping (IABP) or percutaneous cardiopulmonary support (PCPS) postoperatively. One of these 2 patients, a 78-year-old male, had a preoperative SVI of 27 ml/m 2 and underwent AVR with a 23-mm Magna valve without any problem during surgery. However, postoperatively, the patient had low cardiac output without myocardial ischemia or left ventricular systolic dysfunction, and died of low output syndrome on Postoperative Day 30. The other patient was a 78-year-old male with severe AS and a preoperative SVI of 34 ml/m 2. AVR with a 23-mm Mosaic valve was performed smoothly. However, the patient s general condition deteriorated on Postoperative Day 6 and he was diagnosed with nonocclusive mesenteric ischemia based on clinical appearance, hyperlacticacidemia, and contrast-enhanced abdominal

6 2204 MIYAZAKI S et al. computed tomography, and died on Postoperative Day 8. Non-occlusive mesenteric ischemia is a serious complication after cardiac surgery, and special care must be taken to prevent this complication, particularly in an LF state. The intestine is very likely to be prone to ischemia and infarction in the face of hypovolemia early after cardiac surgery, particularly in patients with decreased left ventricular cavity size. Therefore, postoperative management of excess fluid balance is critical, and rapid correction of excess water should be avoided because it could lead to poor blood circulation, causing mesenteric ischemia, particularly in patients with LF severe AS. Postoperative ICU stay and hospital stay, as measures of postoperative recovery, are important issues when examining outcomes after AVR. The results of the present study showed that LF was associated with a longer ICU stay and hospital stay (Table 2) than was NF. Slow recovery in LF patients would be anticipated, because LF was a risk factor for early mortality and morbidity in our series of patients. Causes of late death provide important information. Although our follow-up data regarding the reasons for late deaths are incomplete, the predominant causes of late death were non-cardiac diseases in LF patients. However, Eleid et al 20 reported different results in their LF severe AS patients. In that study, Eleid et al 20 examined long-term survival and causes of late death after AVR for LF patients with a mean follow-up duration of 2.2 years, concluding that LF was associated with higher cardiac mortality after AVR and that congestive heart failure was the predominant cause of cardiac mortality in LF patients. Eleid et al 20 speculated that persistent myocardial dysfunction even after successful AVR in their population could be the reason for their findings. The discrepancy between the results of Eleid et al 20 and those of the present study regarding the causes of late death could be explained by differences in patient characteristics. Further detailed follow-up studies are needed in this regard. One important limitation of the present study is the small number of operative deaths, which may be the reason why potential predictors of operative mortality, such as low EF, New York Heart Association class, and chronic dialysis, did not reach statistical significance. Therefore, the results of the present study may not be generalizable to other institutions, and a larger validation study is needed to confirm our findings. Another limitation of the present study is the potential selection bias of patients, particularly in high-risk LF severe AS patients, referred for surgery. In addition, we do not have information as to how many patients initially referred for surgery were ultimately denied surgery. So, it is not possible to quantify the number of patients considered inoperable who continued with medical management or were referred for TAVI. Another limitation of the study is the possible measurement errors of stroke volume by echocardiography, which may lead to misclassification of patients into LF or NF. Echocardiography is widely used to assess cardiac function and can measure the stroke volume by several methods. 21 One common approach uses a volumetric method based on the Teichholz formula, which was used in the present study. Pitfalls of the Teichholz method include off-axis left ventricular measurement, concomitant mitral valve regurgitation or aortic valve regurgitation, and discordance between regional and global left ventricular systolic function, which can often occur in patients with severe coronary artery disease. Another method uses Doppler imaging to measure stroke volume based on blood flow. This approach may also be subject to measurement errors, particularly in the presence of an elliptic shape of the left ventricular outflow tract, bulky calcification of the aortic annulus extending into the left ventricular outflow tract, sigmoid septum, and atrial fibrillation. 22 It is possible that the Teichholz or Doppler methods could lead to different values and conclusions. In our series, only patients with isolated severe AS were examined and those with concomitant mitral or aortic regurgitation, and coronary artery disease, were excluded. However, patients with regional left ventricular wall motion abnormalities were not completely excluded from the study, which is another limitation, because measurement of stroke volume based on the Teichholz formula is valid in patients without regional left ventricular wall motion abnormalities. Conclusions LF is an independent predictor of early mortality and morbidity in severe AS patients undergoing AVR. LF should be incorporated into the risk stratification and assessment process for these patients. None declared. Conflict of Interest References 1. Pibarot P, Dumesnil JG. Low-flow, low-gradient aortic stenosis with normal and depressed left ventricular ejection fraction. J Am Coll Cardiol 2012; 60: Clavel MA, Dumesnil JG, Capoulade R, Mathieu P, Sénéchal M, Pibarot P. Outcome of patients with aortic stenosis, small valve area, and low-flow, low-gradient despite preserved left ventricular ejection fraction. J Am Coll Cardiol 2012; 60: Dumesnil JG, Pibarot P. Low gradient severe aortic stenosis with preserved ejection fraction: Don t forget the flow! Rev Esp Cardiol 2013; 66: Dayan V, Vignolo G, Magne J, Clavel MA, Mohty D, Pibarot P. Outcome and impact of aortic valve replacement in patients with preserved LVEF and low-gradient aortic stenosis. J Am Coll Cardiol 2015; 66: Hachicha Z, Dumesnil JG, Bogaty P, Pibarot P. Paradoxical low-flow, low-gradient severe aortic stenosis despite preserved ejection fraction is associated with higher afterload and reduced survival. Circulation 2007; 115: Le Ven F, Freeman M, Webb J, Clavel MA, Wheeler M, Dumont É, et al. Impact of low flow on the outcome of high-risk patients undergoing transcatheter aortic valve replacement. J Am Coll Cardiol 2013; 62: Le Ven F, Thébault C, Dahou A, Ribeiro HB, Capoulade R, Mahjoub H, et al. Evolution and prognostic impact of low flow after transcatheter aortic valve replacement. Heart 2015; 101: Clavel MA, Berthelot-Richer M, Le Ven F, Capoulade R, Dahou A, Dumesnil JG, et al. Impact of classic and paradoxical low flow on survival after aortic valve replacement for severe aortic stenosis. J Am Coll Cardiol 2015; 65: Parikh R, Goodman AL, Barr T, Sabik JF, Svensson LG, Rodriguez LL, et al. Outcomes of surgical aortic valve replacement for severe aortic stenosis: Incorporation of left ventricular systolic function and stroke volume index. J Thorac Cardiovasc Surg 2015; 149: Herrmann HC, Pibarot P, Hueter I, Gertz ZM, Stewart WJ, Kapadia S, et al. Predictors of mortality and outcomes of therapy in low-flow severe aortic stenosis: A Placement of Aortic Transcatheter Valves (PARTNER) trial analysis. Circulation 2013; 127: Nashef SA, Roques F, Sharples LD, Nilsson J, Smith C, Goldstone AR, et al. EuroSCORE II. Eur J Cardiothoracic Surg 2012; 41:

7 Low-Flow Severe AS and Surgery O Brien SM, Shahian DM, Filardo G, Ferraris VA, Haan CK, Rich JB, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: Part 2 isolated valve surgery. Ann Thorac Surg 2009; 88(Suppl): S23 S Ambler G, Omar RZ, Royston P, Kinsman R, Keogh BE, Taylor KM. Generic, simple risk stratification model for heart valve surgery. Circulation 2005; 112: Motomura N, Miyata H, Tsukihara H, Takamoto S; Japan Cardiovascular Surgery Database Organization. Risk model of valve surgery in Japan using the Japan Adult Cardiovascular Surgery Database. J Heart Valve Dis 2010; 19: Yamaoka H, Kuwaki K, Inaba H, Yamamoto T, Kato TS, Dohi S, et al. Comparison of modern risk scores in predicting operative mortality for patients undergoing aortic valve replacement for aortic stenosis. J Cardiol 2016; 68: Kuwaki K, Inaba H, Yamamoto T, Dohi S, Matsumura T, Morita T, et al. Performance of the EuroSCORE II and the Society of Thoracic Surgeons Score in patients undergoing aortic valve replacement for aortic stenosis. J Cardiovasc Surg (Torino) 2015; 56: Kuwaki K, Amano A, Inaba H, Yamamoto T, Dohi S, Matsumura T, et al. Predictors of early and mid-term results in contemporary aortic valve replacement for aortic stenosis. J Card Surg 2012; 27: Weidemann F, Herrmann S, Störk S, Niemann M, Frantz S, Lange V, et al. Impact of myocardial fibrosis in patients with symptomatic severe aortic stenosis. Circulation 2009; 120: Herrmann S, Störk S, Niemann M, Lange V, Strotmann JM, Frantz S, et al. Low-gradient aortic valve stenosis: Myocardial fibrosis and its influence on function and outcome. J Am Coll Cardiol 2011; 58: Eleid MF, Micheiena HI, Nkomo VT, Nishimura RA, Malouf JF, Scott CG, et al. Causes of death and predictors of survival after aortic valve replacement in low flow vs. normal flow severe aortic stenosis with preserved ejection fraction. Eur Heart J 2015; 16: Dele-Michael AO, Fujikura K, Devereux RB, Islam F, Hriljac I, Wilson SR, et al. Left ventricular stroke volume quantification by contrast echocardiography comparison of linear and flowbased methods to cardiac magnetic resonance. Echocardiography 2013; 30: Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP, et al. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. Eur J Echocardiogr 2009; 10: 1 25.