Enlargement of the Small Aortic Root During Aortic Valve Replacement: Is There a Benefit?

CARDIOVASCULAR

Enlargement of the Small Aortic Root During Aortic Valve Replacement: Is There a Benefit? Alexander Kulik, MD, Manal Al-Saigh, MD, Vincent Chan, MD, Roy G. Masters, MD, Pierre Bédard, MD, B-Khanh Lam, MD, MPH, Fraser D. Rubens, MD, Paul J. Hendry, MD, Thierry G. Mesana, MD, PhD, and Marc Ruel, MD, MPH Division of Cardiac Surgery, University of Ottawa Heart Institute, and Department of Epidemiology, University of Ottawa, Ottawa, Ontario, Canada

Background. Aortic root enlargement (ARE) at the time of aortic valve replacement (AVR) is an often proposed but still unproven technique to prevent prosthesis-patient mismatch. To evaluate the risks and benefits of ARE, we examined the outcomes of patients with small aortic roots who underwent AVR with or without the use of ARE. Methods. Patients (n ⴝ 712) with small aortic roots who underwent AVR were prospectively followed (followup, 3,730 patient-years; mean, 5.2 ⴞ 4.1 years). All patients had a small aortic annulus that would have led to the insertion of an aortic prosthesis of 21 or less in size. Multivariate techniques were used to compare outcomes between patients who underwent AVR alone (n ⴝ 540) versus AVR plus ARE (n ⴝ 172). Results. Aortic cross-clamp times were 9.9 minutes longer in the AVRⴙARE group (p ⴝ 0.0002). There were no differences in reopening or stroke rates or perioperative mortality (all p ⴝ not significant). All patients in the AVR-alone group received size 19 to 21 prostheses,

whereas 51% of the AVRⴙARE patients received size 23 prostheses. Postoperative gradients were reduced (p < 0.01) and indexed effective orifice areas were larger (p < 0.0001) in the AVRⴙARE group. While the incidence of postoperative prosthesis-patient mismatch (indexed effective orifice area < 0.85 cm2/m2) was lower in the AVRⴙARE group (p < 0.0001), the presence of mismatch did not significantly impact long-term outcomes after surgery. The ARE was associated with a trend toward better freedom from late congestive heart failure (p ⴝ 0.19), but not an improvement in long-term survival (p ⴝ 0.81). Conclusions. For patients with small aortic roots, ARE at the time of AVR is a safe procedure that reduces postoperative gradients and the incidence of prosthesispatient mismatch. However, ARE does not appreciably improve long-term clinical outcomes.

T

atic improvement [4 – 6], less left ventricular mass regression [4, 6], and poor early and late survival after surgery [6 –9]. Patients with small aortic roots, especially those with a large body surface area, are at high risk for having PPM after AVR. To minimize postoperative gradients and left ventricular outflow tract obstruction, several strategies have been developed to enlarge the small aortic root at the time of surgery, including the Nicks [10], the Konno [11], and the Manouguian [12] procedures. A number of cardiac surgery centers have reported their experience with aortic root enlargement (ARE) procedures, suggesting that these operations are both safe and feasible [12–15]. However, others have cautioned against the routine use of ARE because of its technical difficulty and its possible association with adverse outcomes [16, 17]. To our knowledge, data evaluating the effect of ARE on the incidence of postoperative PPM and late clinical outcomes are lacking. Therefore, the purpose of this study was to assess the risks and benefits of ARE by comparing the hemodynamic and clinical outcomes of patients with small aortic roots who underwent AVR with or without the use of ARE.

he goals of aortic valve replacement (AVR) are to reduce pressure and volume overload on the left ventricle, relieve symptoms, and improve survival in patients with advanced aortic valve disease [1]. Ideally, postoperative transprosthetic pressure gradients should be minimal. However, high-pressure gradients may be encountered after surgery despite normally functioning prostheses, especially in patients who received small-sized prosthetic valves, and may result in postoperative left ventricular outflow tract obstruction and prosthesis-patient mismatch (PPM). First described by Rahimtoola in 1978, PPM is a condition in which the “effective prosthetic valve area, after insertion into the patient, is less than that of a normal valve” [2]. Subsequently, Pibarot and Dumesnil [3] defined PPM as a prosthetic valve effective orifice area (EOA) indexed to body surface area of 0.85 cm2/m2 or less. We and others have documented worse postoperative outcomes in AVR patients with PPM, with less symptomAccepted for publication July 23, 2007. Address correspondence to Dr Ruel, University of Ottawa Heart Institute, 40 Ruskin St, Suite 3403, Ottawa, Ontario, K1Y 4W7, Canada; e-mail: [email protected].

© 2008 by The Society of Thoracic Surgeons Published by Elsevier Inc

(Ann Thorac Surg 2008;85:94 –101) © 2008 by The Society of Thoracic Surgeons

0003-4975/08/$34.00 doi:10.1016/j.athoracsur.2007.07.058

Patients and Methods Patient Population and Follow-Up This study focused on patients with small aortic roots. Patients (n ⫽ 712) with small aortic roots who underwent AVR were prospectively followed between 1989 and 2006 at the University of Ottawa Heart Institute. All patients were 18 years or older and had a small aortic annulus identified before or at surgery that would have led to the insertion of an aortic prosthesis of size 21 or smaller. Patients were excluded from the study population if they received stentless prostheses or if they received prostheses that are no longer commercially available. Patients were also excluded from the study population if they did not survive the perioperative period, except for the purposes of reporting procedural data and perioperative mortality rates. Of the 712 patients in this cohort, 540 patients underwent AVR and 172 underwent AVR plus ARE (AVR⫹ARE). After AVR, patients were assessed 6 months postoperatively and thereafter on an annual basis by a physician in a dedicated valve clinic. At each visit, patients underwent a medical history focused on the determination of functional status and the occurrence of valve-related complications, a physical examination, electrocardiogram, chest radiograph, complete blood count, serum chemistries, and international normalized ratio determinations when applicable. The methods of the valve clinic follow-up have been reviewed and approved by the University of Ottawa Heart Institute Human Research Ethics Board. Before 2004, a waiver of consent was granted by the Board. Since 2004, all valve clinic patients have provided explicit, fully informed consent for enrollment. All patients were followed for at least one outpatient visit. Surgeries were performed between January 24, 1989, and August 4, 2006, and the closing interval for vital status determination was April 1 to May 5, 2007. The total follow-up for the entire cohort was 3,730 patient-years (mean, 5.2 ⫾ 4.1 years; maximum, 17.5). The follow-up period for the patients who had AVR alone was 2,848 patient-years (mean, 5.3 ⫾ 4.3 years; maximum, 17.5), and for the patients who had AVR⫹ARE, the follow-up was 883 patient-years (mean, 5.1 ⫾ 3.8 years; maximum, 15.0).

Prostheses and Root Enlargement Procedure The decision whether to perform an ARE was taken by the operating surgeon, depending on the patient’s age and comorbid conditions, the anatomy of the aortic root, and the surgeon’s judgment and comfort level. The ARE was performed based upon the techniques of Nicks and associates [10] and Manouguian and Seybold-Epting [12] using either autologous pericardium or Hemashield (Boston Scientific, Natick, Massachusetts) patches and 4-0 Prolene (Ethicon, Somerville, New Jersey). The ARE technique enabled the insertion of a prosthetic valve at least one size larger than the original annulus could accommodate. Prosthesis type and size were documented in all patients. Valve sizing and selection were performed with the sizers provided by each respective prosthetic valve manufacturer. Data were collected from

KULIK ET AL AORTIC ROOT ENLARGEMENT

95

operative records to audit technical details regarding the root enlargement procedures.

Echocardiograms All patients underwent a complete M-mode, twodimensional and Doppler transthoracic echocardiogram before and after AVR, and thereafter as clinically indicated. Measurements were documented from the Mmode recordings as per the recommendations of the American Society of Echocardiography [18]. The EOA for each prosthesis type and size was obtained from the literature of patients with normally functioning prostheses [5, 19, 20], and averaged if more than one published value was available. This was supplemented with phase I regulatory data provided by the valve manufacturer if published data were insufficient with respect to a specific prosthesis size. The indexed EOA was obtained by dividing the EOA by the patient’s body surface area at the time of operation. PPM was defined a priori as an indexed EOA of 0.85 cm2/m2 or less, as this constitutes the most common definition of PPM in the literature [5, 19, 21]. To account for the differences between the sizes of the different prostheses used in this study, the actual internal and external diameters for each prosthesis type and size were determined from the literature [20].

Statistical Analyses CLINICAL OUTCOMES. Data were analyzed in Intercooled Stata 9.2 (Stata, College Station, Texas). Prosthesis hemodynamics and clinical outcomes were compared between small aortic root patients treated with AVR versus those treated with AVR and ARE. Continuous data are presented as a mean ⫾ SD and were compared between groups using unpaired two-sided Student’s t tests for parametric data and Wilcoxon rank sum tests for non-

Table 1. Preoperative Patient Characteristics AVR Group AVR ⫹ ARE p (n ⫽ 540) Group (n ⫽ 172) Value Age at operation (years) 69.1 ⫾ 11.8 Body mass index (kg/m2) 27.3 ⫾ 6.5 Female 404 (74.8%) Atrial fibrillation 34 (6.5%) Cigarette smoking 56 (10.4%) Preoperative NYHA 221 (40.9%) class 3–4 Ejection fraction ⬍ 50% 66 (12.2%) Preoperative mean 48.4 ⫾ 25.4 transvalvular pressure gradient (mm Hg) Preoperative peak 68.8 ⫾ 28.9 transvalvular pressure gradient (mm Hg) Preoperative aortic 0.72 ⫾ 0.34 valve area (cm2)

66.8 ⫾ 12.3 28.2 ⫾ 6.4 119 (69.2%) 6 (3.5%) 12 (12.8%) 66 (38.4%)

0.03 0.20 0.16 0.19 0.40 0.59

24 (14.0%) 39.1 ⫾ 18.0

0.60 0.21

67.7 ⫾ 25.1

0.84

0.70 ⫾ 0.17

0.57

ARE ⫽ aortic root enlargement; AVR ⫽ aortic valve replacement; NYHA ⫽ New York Heart Association.

CARDIOVASCULAR

Ann Thorac Surg 2008;85:94 –101

CARDIOVASCULAR

96

KULIK ET AL AORTIC ROOT ENLARGEMENT

Ann Thorac Surg 2008;85:94 –101

Table 2. Operative and Prosthetic Valve Characteristics

Cardiopulmonary bypass duration (minutes) Aortic cross-clamp duration (minutes) Aortic root enlargement Manouguian Nicks Concomitant CABG Concomitant mitral procedurea Concomitant tricuspid procedureb Mechanical valves Medtronic Hall St. Jude Carbomedics MCRI On-x Bioprosthetic valves Medtronic Hancock Medtronic Mosaic Edwards Pericardial Prosthesis size Size 19 Size 20 Size 21 Size 23 a

Denotes either mitral valve repair or replacement.

ARE ⫽ aortic root enlargement;

b

AVR Group (n ⫽ 540)

AVR ⫹ ARE Group (n ⫽ 172)

p Value

129.9 ⫾ 44.4 86.5 ⫾ 27.9

137.5 ⫾ 45.6 94.1 ⫾ 26.7

0.05 0.002

— — 214 (39.6%) 73 (13.5%) 29 (5.4%) 217 (40.2%) 57 (10.6%) 86 (15.9%) 28 (5.2%) 46 (8.5%) 323 (59.8%) 214 (39.6%) 22 (4.0%) 87 (16.1%)

123 (71.5%) 49 (28.5%) 75 (43.6%) 12 (7.0%) 1 (0.6%) 74 (43.0%) 8 (4.7%) 33 (19.2%) 20 (11.6%) 13 (7.6%) 98 (57.0%) 58 (33.7%) 3 (1.7%) 37 (21.5%)

—

59 (10.9%) 18 (3.3%) 463 (85.7%) 0 (0%)

0 (0%) 3 (1.7%) 82 (47.7%) 87 (50.6%)

0.37 0.02 0.004 0.53

0.53

⬍ 0.0001

Denotes either tricuspid valve repair or replacement.

AVR ⫽ aortic valve replacement;

parametric data. Categorical data are presented as proportions and were compared between groups using a Fisher’s exact test. Statistical significance was set at a p value of less than 0.05. A composite congestive heart failure (CHF) endpoint was defined a priori as (1) New York Heart Association (NYHA) functional class III or IV for more than 4 consecutive weeks or (2) death for which the primary or main contributing diagnosis was CHF [5]. Clinical impressions were corroborated with physical examination, chest radiograph, electrocardiogram, and echocardiographic findings. Deaths from an unknown cause were not considered to result from CHF and were treated as a censored event if the patient had not previously experienced NYHA class III or IV symptoms. Nonparametric estimates of freedom from all-cause death and freedom from the composite CHF endpoint were determined using the Kaplan-Meier method and are reported as mean ⫾ SE. MULTIVARIATE ANALYSES. Potential univariate predictors of outcomes were individually tested for equality with a log-rank test. To account for positive or negative confounding, multivariate Cox proportional hazards models were developed by incorporating all variables that had a p value of 0.20 or less on log-rank testing. Stepwise forward selection and backward elimination techniques were employed with p ⫽ 0.20 for entry and removal criteria. Risk factors (left ventricular dysfunction, age, atrial fibrillation, preoperative heart failure functional class, coronary artery disease, smoking, insulin-

CABG ⫽ coronary artery bypass grafting.

dependent diabetes mellitus, postoperative hypertension) for decreased survival and freedom from CHF after AVR that have been previously identified [5] were considered in each model. Effect modification between factors was tested with the use of interaction terms in the models. Scaled Schoenfeld residuals were employed to assess the Cox proportional hazard assumption for each variable. Stratified Cox analysis was applied if the proportional hazard assumption was not met for a particular factor. Hazard ratios (HR) are reported along with standard errors or 95% confidence intervals (CI).

Results Patient Characteristics The preoperative characteristics of the AVR and AVR⫹ARE groups are presented in Table 1. Preoperative characteristics were equivalent between the groups, except patients in the AVR⫹ARE group were significantly younger (p ⫽ 0.03). Operative characteristics differed between the two groups, as demonstrated in Table 2. The ARE procedure was performed based on either the techniques of Nicks [10] (28.5%) or Manouguian [12] (71.5%). Patients in the AVR⫹ARE group were less likely to undergo intervention on the mitral or tricuspid valve (all p ⬍ 0.05). Baseline and operative factors that differed between the two groups with a p value of 0.20 or less were included in the multivariate analysis models examining long-term outcomes.

KULIK ET AL AORTIC ROOT ENLARGEMENT

97

Table 3. Perioperative Outcomes

Stroke Mortality Reopening Length of hospital stay (days)

AVR Group (n ⫽ 540)

AVR ⫹ ARE Group (n ⫽ 172)

p Value

4 (0.7%) 35 (6.5%) 33 (6.1%) 10.6 ⫾ 7.8

1 (0.6%) 12 (7.0%) 9 (5.2%) 13.8 ⫾ 24.2

1.00 0.74 1.00 0.42

ARE ⫽ aortic root enlargement;

AVR ⫽ aortic valve replacement.

Perioperative Outcomes The addition of an ARE significantly increased the duration of surgery, including total aortic cross-clamp times and cardiopulmonary bypass times (all p ⬍ 0.05; see Table 2). Among patients undergoing isolated aortic valve surgery (no CABG, mitral, or tricuspid surgery; n ⫽ 346), AVR⫹ARE patients had significantly longer aortic cross-clamp times by 9.9 ⫾ 2.6 minutes (82.1 ⫾ 21.3 versus 72.2 ⫾ 20.9 minutes, p ⫽ 0.0002, AVR⫹ARE versus AVR), and significantly longer cardiopulmonary bypass times by 12.2 ⫾ 4.3 minutes (119.3 ⫾ 44.5 versus 106.9 ⫾ 30.8 minutes, p ⫽ 0.004, AVR⫹ARE versus AVR). Reopening rates in the immediate postoperative period were similar between the two groups (5.2% versus 6.1%, p ⫽ 1.00, AVR⫹ARE versus AVR). Perioperative mortality during the time period of this study was similar between the two groups (7.0% versus 6.5%, p ⫽ 0.74, AVR⫹ARE versus AVR). Perioperative outcomes are summarized in Table 3.

Prosthesis Characteristics Patients in the AVR and AVR⫹ARE groups received mechanical and bioprosthetic valves in a similar propor-

Fig 2. (A) Survival and (B) freedom from New York Heart Association class III or IV congestive heart failure (CHF) or CHF death among patients with small aortic roots who underwent aortic valve replacement (AVR [dotted line]) or AVR plus aortic root enlargement (ARE [solid line]). After adjusting for potential confounders in the multivariate analysis, an ARE was associated with a trend toward better long-term freedom from CHF (p ⫽ 0.19) but not long-term survival (p ⫽ 0.81).

Fig 1. Postoperative transprosthesis gradients in patients with small aortic roots who underwent aortic valve replacement (AVR [shaded bars]) or AVR plus aortic root enlargement (ARE [open bars]). Both mean and peak transprosthetic gradients were significantly lower in the AVR plus ARE group.

tion (p ⫽ 0.53). However, the AVR group received size 19 to 21 aortic prostheses, whereas 50.6% of AVR⫹ARE patients received size 23 aortic prostheses (Table 2). The mean diameters of the implanted aortic prostheses were significantly larger in the AVR⫹ARE group (internal diameters, 19.4 ⫾ 1.6 mm versus 18.0 ⫾ 1.3 mm, p ⬍ 0.0001; external diameters, 27.7 ⫾ 2.3 mm versus 25.8 ⫾ 2.4 mm, p ⬍ 0.0001; AVR⫹ARE versus AVR, respectively). Postoperative transprosthesis gradients were significantly lower in the AVR⫹ARE group compared with the AVR group, as illustrated in Figure 1 (peak gradient, 28.3 ⫾ 14.0 mm Hg versus 34.2 ⫾ 15.1 mm Hg, p ⫽ 0.0001; mean gradient, 15.4 ⫾ 7.8 mm Hg versus 18.3 ⫾ 8.8 mm Hg, p ⫽ 0.001; AVR⫹ARE versus AVR, respectively). The

CARDIOVASCULAR

Ann Thorac Surg 2008;85:94 –101

CARDIOVASCULAR

98

KULIK ET AL AORTIC ROOT ENLARGEMENT

indexed EOAs after surgery were significantly greater in the AVR⫹ARE group (0.89 ⫾ 0.18 cm2/m2 versus 0.79 ⫾ 0.16 cm2/m2, p ⬍ 0.00001, AVR⫹ARE versus AVR). The incidence of postoperative PPM was significantly reduced in the AVR⫹ARE group (indexed EOA ⱕ 0.85 cm2/m2, 42.6% versus 69.4%, p ⬍ 0.0001).

Long-Term Survival Among patients who survived the perioperative period, survival at 1, 5, and 10 years in the AVR⫹ARE group was 96.8% ⫾ 1.4%, 86.2% ⫾ 3.2%, and 69.7% ⫾ 5.4%, respectively. Survival at 1, 5, and 10 years in the AVR group was 97.0% ⫾ 0.8%, 85.2% ⫾ 1.9%, and 65.0% ⫾ 3.2%, respectively. Long-term survival was not significantly better among the AVR⫹ARE patients, either on univariate analysis (HR: 0.8; 95% CI: 0.6 to 1.3; p ⫽ 0.41; Fig 2A), or after adjustment for covariates in the multivariate model (HR: 1.1; 95% CI: 0.7 to 1.6; p ⫽ 0.81). Among AVR⫹ARE patients, 10-year survival was 84.6% ⫾ 9.7% in those with postoperative PPM and 84.5% ⫾ 6.1% in those without postoperative PPM (p ⫽ 0.83). Among AVR patients, 10-year survival was 77.8% ⫾ 4.2% in those with postoperative PPM and 77.3% ⫾ 5.4% in those without postoperative PPM (p ⫽ 0.94).

Freedom From Congestive Heart Failure Freedom from the composite CHF outcome for AVR⫹ARE patients at 1, 5, and 10 years was 100%, 96.5% ⫾ 1.7%, and 86.4% ⫾ 4.8%, respectively. Freedom from the composite CHF outcome for AVR patients at 1, 5, and 10 years was 99.4% ⫾ 0.3%, 95.5% ⫾ 1.1%, and 83.3% ⫾ 2.7%. On univariate analysis, freedom from CHF was not significantly better for AVR⫹ARE patients (HR: 1.0; 95% CI: 0.5 to 1.8; p ⫽ 0.94; Fig 2B). After adjustment for covariates in the multivariate model, however, an ARE was associated with a trend toward better freedom from CHF (HR: 0.6; 95% CI: 0.3 to 1.3; p ⫽ 0.19). Among AVR⫹ARE patients, 10-year freedom from CHF was 97.3% ⫾ 2.7% for those with postoperative PPM and 88.8% ⫾ 5.6% for those without postoperative PPM (p ⫽ 0.61). Among AVR patients, 10-year freedom from CHF was 89.4% ⫾ 3.2% for those with postoperative PPM and 90.8% ⫾ 3.7% for those without postoperative PPM (p ⫽ 0.54).

Comment The surgical management of the small aortic root at the time of AVR has been discussed in the cardiac surgery literature for more than 30 years. After initial attempts at supra-annular enlargement, the first report of aortic annular enlargement with a posterior incision was made by Nicks and colleagues [10] in 1970. Subsequently, in 1975, Konno and associates [11] demonstrated their surgical technique involving anterior annular enlargement for congenital aortic stenosis. In the late 1970s, Manouguian and Seybold-Epting [12] reported a novel enlargement technique through the commissure between the left and noncoronary sinuses of Valsalva [12]. These surgical strategies were developed with the objectives of implant-

Ann Thorac Surg 2008;85:94 –101

ing prostheses as large as possible to minimize postoperative gradients. Recent investigations documenting the adverse effects of postoperative PPM after AVR have renewed interest in the ARE operation [4 – 8, 19]. However, there remains a general perception among cardiac surgeons that ARE procedures increase the risk and technical difficulty of aortic valve surgery [16, 17]. Thus, the purpose of this study was to evaluate whether the addition of an ARE to an AVR in a patient with a small aortic root increases the perioperative risk and to determine whether it is associated with any clinical benefit. In this cohort of patients with small aortic roots followed annually after AVR, we observed that (1) the addition of an ARE to an AVR increased the aortic cross-clamp time, on average, by 9.9 minutes; (2) there was no significant increase in perioperative morbidity or mortality associated with the addition of an ARE; (3) patients with small aortic roots who underwent root enlargement procedures received larger prostheses, with lower postoperative gradients and larger EOAs; (4) patients who underwent ARE⫹AVR had a lower incidence of postoperative PPM, but the presence of PPM did not independently impact long-term outcomes; and (5) the addition of an ARE did not significantly improve longterm clinical outcomes among the entire cohort. Thus, patients with small aortic roots treated with ARE⫹AVR have better hemodynamic outcomes after surgery, without suffering an increased risk of perioperative morbidity or mortality. However, the ARE procedure did not appreciably improve long-term clinical outcomes after AVR. Early reports of the ARE experience in the 1980s suggested that these operations were both safe and feasible, albeit at the expense of longer aortic crossclamp times [12, 22, 23]. In 1997, however, Sommers and David [16] published a negative experience with the ARE procedure in a retrospective review of 530 patients who underwent AVR with Hancock II prostheses. Of this cohort, 98 patients underwent ARE in an effort to avoid postoperative PPM. The addition of an ARE increased the aortic cross-clamp time (by 11 minutes), increased the rate of perioperative reopening for bleeding (10.2% versus 6.7%, p ⫽ 0.23, AVR⫹ARE versus AVR), and increased the operative mortality rate (7.1% versus 3.5%, p ⫽ 0.10, AVR⫹ARE versus AVR) [16]. With similar long-term outcomes between the two groups, and yet increased early morbidity and mortality, there appeared to be no clinical advantage to the routine use of ARE at the time of AVR. Recent experience with the ARE procedure has been more favorable, with reported operative mortality rates less than 4% in case series [13, 14]. Even in centers initially reporting high operative mortality rates, increasing experience has led to declining mortality rates (7.2% down to 2.9%) [16, 24]. The operative mortality during the era of the current study is higher than expected (6% to 7%), but this may be related to the surgical cohort (patients with small aortic roots), a high proportion of female patients, and a high proportion of patients requiring coronary revascularization. Both female sex and

small prosthesis size have been demonstrated to increase AVR perioperative mortality [9]. Nevertheless, it appears that similar perioperative outcomes can be achieved in patients with small aortic roots undergoing AVR⫹ARE compared with AVR alone. The results of the current study demonstrate that, in experienced hands, there is no additional risk of perioperative bleeding or mortality with the addition of an ARE procedure at the time of AVR. The ARE procedure adds approximately 9.9 minutes of aortic cross-clamp time and 12.2 minutes of cardiopulmonary bypass time to the length of the operation. Our study has documented the hemodynamic benefit of the ARE in patients with small aortic roots undergoing AVR. The ARE procedure enabled the insertion of larger prosthetic valves (size 23 in more than half the patients), reduced postoperative gradients, and reduced the incidence of postoperative PPM. The concept of PPM was first introduced by Rahimtoola [2] in 1978 to describe the condition in which the prosthetic valve orifice area is less than that of the native human valve. Subsequent studies examining the physiologic sequalae of PPM have fostered the recommendation that the indexed EOA of an aortic prosthesis should ideally be greater than 0.85 cm2/m2 to minimize postoperative gradients and improve clinical results [19]. There continues to be controversy in the literature as to the relevance of PPM. In the current study, we found that ARE reduced the incidence of postoperative PPM, but PPM did not significantly impact long-term outcomes after AVR, regardless of whether patients underwent ARE. Earlier work from our group has demonstrated that PPM is an independent predictor of adverse outcomes after AVR specifically in patients with preoperative left ventricular dysfunction or low-gradient aortic stenosis [4, 6]. However, we did not have enough statistical power in the present study to focus on the outcomes of these patients. Previous ARE studies have produced conflicting results when comparing outcomes in patients who underwent AVR⫹ARE to those who underwent AVR alone [13, 16]. In 1996, Kitamura and colleagues [13] reported a series of 45 patients with small aortic annuli (19 mm) who underwent AVR with or without ARE procedures. Although early mortality was similar between the two groups (3.6% versus 5.9%, p ⫽ not significant, AVR⫹ARE versus AVR), 10-year survival (85.7% versus 62.7%, p ⬍ 0.10, AVR⫹ARE versus AVR) and 10-year freedom from valve-related events (81.0% versus 58.8%, p ⬍ 0.05, AVR⫹ARE versus AVR) were improved in the AVR⫹ARE group, suggesting that long-term outcomes may be superior after AVR with ARE [13]. In the current study, we found that the addition of ARE for patients with small aortic roots was associated with a trend toward better long-term freedom from CHF (HR: 0.6; 95% CI: 0.3 to 1.3; p ⫽ 0.19), but not long-term survival (HR: 1.1; 95% CI: 0.7 to 1.6; p ⫽ 0.81) after AVR. Whether ARE procedures improve the long-term outcomes in patients with small aortic roots can not definitively be determined in the absence of prospective data. However, we believe that a randomized controlled trial

KULIK ET AL AORTIC ROOT ENLARGEMENT

99

prospectively evaluating the ARE operation is unlikely ever to be conducted. Complicating the issue, enlargement of the aortic root is generally applied to patients with small aortic roots who are deemed to be good operative candidates with prospects for excellent longterm survival [16]. In the current study, the AVR⫹ARE patients were significantly younger compared with those who underwent AVR without root enlargement. In contrast, high-risk AVR patients with lower probability of long-term survival, as well as those with long crossclamp times due to concomitant procedures (namely, mitral surgery), may be less likely to undergo aortic root enlargement. Although multivariate analyses were used in this study, confounding by indication, a form of selection bias, can not be fully accounted for by multivariate analyses [7].

Limitations Other strategies have been described for the management of the small aortic root at the time of AVR, including the use of stentless valves [25]. However, the ARE procedure is the favored approach at our institution, thus limiting our ability to draw comparisons to other techniques. Group differences and known confounders were controlled for in this observational study by using multivariate analysis. Despite the sample size and statistical adjustments applied, however, unmeasured or unknown confounders may have influenced the results. Moreover, the Cox proportional hazards regression method requires an assumption of independent censoring that may not always be met. In this regard, it is possible that patients lost to follow-up after a number of visits may have had subsequent outcomes that were not accounted for in the analyses. Finally, like that of other observational cohorts, the results of these analyses may not be generalizable to all patients with small aortic roots who have undergone AVR at other centers because this study represents a single institution’s experience. In conclusion, for patients with small aortic roots, the addition of an ARE procedure at the time of AVR slightly increases the length of the operation, but does not increase perioperative morbidity or mortality. The ARE procedure improves postoperative transprosthesis gradients and reduces the incidence of PPM after surgery. However, ARE does not appreciably improve long-term clinical outcomes for patients with small aortic roots undergoing AVR.

References 1. Bonow RO, Carabello BA, Kanu C, et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/ American Heart Association task force on practice guidelines (writing committee to revise the 1998 guidelines for the management of patients with valvular heart disease): developed in collaboration with the Society of Cardiovascular Anesthesiologists: endorsed by the Society for Cardiovascular Angiography and Interventions and The Society of Thoracic Surgeons. Circulation 2006;114:e84 –231.

CARDIOVASCULAR

Ann Thorac Surg 2008;85:94 –101

CARDIOVASCULAR

100

KULIK ET AL AORTIC ROOT ENLARGEMENT

2. Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation 1978;58:20 – 4. 3. Pibarot P, Dumesnil JG, Lemieux M, Cartier P, Metras J, Durand LG. Impact of prosthesis-patient mismatch on hemodynamic and symptomatic status, morbidity and mortality after aortic valve replacement with a bioprosthetic heart valve. J Heart Valve Dis 1998;7:211– 8. 4. Kulik A, Burwash IG, Kapila V, Mesana TG, Ruel M. Longterm outcomes after valve replacement for low-gradient aortic stenosis: impact of prosthesis-patient mismatch. Circulation 2006;114(1 Suppl):I553– 8. 5. Ruel M, Rubens FD, Masters RG, et al. Late incidence and predictors of persistent or recurrent heart failure in patients with aortic prosthetic valves. J Thorac Cardiovasc Surg 2004;127:149 –59. 6. Ruel M, Al-Faleh H, Kulik A, Chan KL, Mesana TG, Burwash IG. Prosthesis-patient mismatch after aortic valve replacement predominantly affects patients with preexisting left ventricular dysfunction: effect on survival, freedom from heart failure, and left ventricular mass regression. J Thorac Cardiovasc Surg 2006;131:1036 – 44. 7. Rao V, Jamieson WR, Ivanov J, Armstrong S, David TE. Prosthesis-patient mismatch affects survival after aortic valve replacement. Circulation 2000;102(19 Suppl 3):III5–9. 8. Blais C, Dumesnil JG, Baillot R, Simard S, Doyle D, Pibarot P. Impact of valve prosthesis-patient mismatch on short-term mortality after aortic valve replacement. Circulation 2003; 108:983– 8. 9. Morris JJ, Schaff HV, Mullany CJ, et al. Determinants of survival and recovery of left ventricular function after aortic valve replacement. Ann Thorac Surg 1993;56:22–30. 10. Nicks R, Cartmill T, Bernstein L. Hypoplasia of the aortic root. The problem of aortic valve replacement. Thorax 1970; 25:339 – 46. 11. Konno S, Imai Y, Iida Y, Nakajima M, Tatsuno K. A new method for prosthetic valve replacement in congenital aortic stenosis associated with hypoplasia of the aortic valve ring. J Thorac Cardiovasc Surg 1975;70:909 –17. 12. Manouguian S, Seybold-Epting W. Patch enlargement of the aortic valve ring by extending the aortic incision into the anterior mitral leaflet. New operative technique. J Thorac Cardiovasc Surg 1979;78:402–12. 13. Kitamura M, Satoh M, Hachida M, Endo M, Hashimoto A, Koyanagi H. Aortic valve replacement in small aortic annu-

Ann Thorac Surg 2008;85:94 –101

14. 15. 16. 17. 18.

19. 20. 21.

22.

23.

24.

25.

lus with or without annular enlargement. J Heart Valve Dis 1996;5(Suppl 3):289 –93. Castro LJ, Arcidi JM Jr, Fisher AL, Gaudiani VA. Routine enlargement of the small aortic root: a preventive strategy to minimize mismatch. Ann Thorac Surg 2002;74:31– 6. Feindel CM. Aortic root enlargement in the adult. Op Tech Thorac Cardiovasc Surg 2006;11:2–15. Sommers KE, David TE. Aortic valve replacement with patch enlargement of the aortic annulus. Ann Thorac Surg 1997;63: 1608 –12. Carrier M, Pellerin M, Perrault LP, et al. Experience with the 19-mm Carpentier-Edwards pericardial bioprosthesis in the elderly. Ann Thorac Surg 2001;71(Suppl):249 –52. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072– 83. Pibarot P, Dumesnil JG. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol 2000;36:1131– 41. Christakis GT, Buth KJ, Goldman BS, et al. Inaccurate and misleading valve sizing: a proposed standard for valve size nomenclature. Ann Thorac Surg 1998;66:1198 –203. Dumesnil JG, Honos GN, Lemieux M, Beauchemin J. Validation and applications of indexed aortic prosthetic valve areas calculated by Doppler echocardiography. J Am Coll Cardiol 1990;16:637– 43. Mori T, Kawashima Y, Kitamura S, Nakano S, Kawachi K, Nakata T. Results of aortic valve replacement in patients with a narrow aortic annulus: effects of enlargement of the aortic annulus. Ann Thorac Surg 1981;31:111– 6. Pugliese P, Bernabei M, Santi C, Pasque A, Eufrate S. Posterior enlargement of the small annulus during aortic valve replacement versus implantation of a small prosthesis. Ann Thorac Surg 1984;38:31– 6. Peterson MD, Borger MA, Feindel CM, David TE. Aortic annular enlargement during aortic valve replacement: improving results with time. Ann Thorac Surg 2007;83: 2044 –9. Hvass U, Palatianos GM, Frassani R, Puricelli C, O’Brien M. Multicenter study of stentless valve replacement in the small aortic root. J Thorac Cardiovasc Surg 1999;117: 267–72.

INVITED COMMENTARY Patient prosthesis mismatch is undesirable and continues to be a controversial topic since the beginning of aortic valve replacement. The major issue with patient prosthesis mismatch is the effective orifice area: all valve prostheses except stentless bioprostheses had a significantly smaller effective orifice area than the normal native valve. Therefore, in many patients with prosthetic heart valves, left ventricular outflow obstruction improved from severe to moderate [1], but a pathologic transprosthetic pressure gradient remains. In the case of a small aortic root, the remaining pressure gradient can be dramatical. Despite the variety of artificial heart valves, no ideal prosthesis for the small aortic root is currently available. Small mechanical valves have superior hemodynamics compared with tissue valves but are often contraindicated in older patients. Conventional stented valves are hemodynamically disadvantageous due to the higher transprosthetic gradients. © 2008 by The Society of Thoracic Surgeons Published by Elsevier Inc

Until now, aortic root enlargement, followed by implantation of an aortic mechanical valve or stentless bioprosthesis, seems to be the best choice to establish a nonobstructive left ventricular outflow and is a good match with the native annulus and body surface area. The aim of the study by Kulik and colleagues [2] was to describe the patient benefit after enlargement of the small aortic root during aortic valve replacement. In contrast to previous reports, the enlargement was completed in a short period of time without increased mortality compared with the aortic valve replacement group. Nevertheless, the combination of a calcified root, potential bleeding risks, an insignificant longterm benefit, and an increased cross-clamp time of 10 minutes may yet prove problematic for the recovery of elderly patients. The combined procedure (aortic root enlargement plus aortic valve replacement) resulted in a lower transprosthetic gradient compared with a simple aortic valve 0003-4975/08/$34.00 doi:10.1016/j.athoracsur.2007.09.014