Advertisement

Open hemiarch versus clamped ascending aorta replacement for aortopathy during initial bicuspid aortic valve replacement

Open ArchivePublished:September 25, 2019DOI:https://doi.org/10.1016/j.jtcvs.2019.09.028

      Abstract

      Background

      There is controversy regarding the extent of aortic resection necessary in patients with aortopathy related to bicuspid aortic valve disease. To address this issue, we reviewed our experience in patients undergoing ascending aorta replacement during bicuspid aortic valve replacement.

      Methods

      We reviewed 702 patients who underwent ascending aorta replacement at the time of initial nonemergent native bicuspid aortic valve replacement at our institution between January 2000 and June 2017. Treatment cohorts included an open hemiarch replacement group (n = 225; 32%) and a clamped ascending aorta replacement group (n = 477; 68%).

      Results

      Median patient age was 60 years (interquartile range [IQR], 51-67 years), female sex was present in 113 patients (16%), ejection fraction was 62% (IQR, 56%-66%), and aortic arch diameter was 33 mm (IQR, 29-36 mm). Cardiopulmonary bypass time was longer in the hemiarch replacement group (188 minutes vs 97 minutes; P < .001). Procedure-related complications (36%) and mortality (<1%) were similar in the 2 groups; however, the hemiarch group had an increased odds of blood transfusion (odds ratio, 1.62; 95% confidence interval [CI], 1.15-2.28; P = .006). The median duration of follow-up was 6.0 years (95% CI, 5.3-6.8 years). Overall survival was 94 ± 1% at 5 years and 80 ± 2% at 10 years. Multivariable analysis demonstrated similar survival in the 2 groups (hazard ratio, 0.83; 95% CI, 0.51-1.33; P = .439). No repeat aortic arch operations were done for aortopathy over the duration of clinical follow-up.

      Conclusions

      Compared with patients in the clamped ascending aorta replacement group, patients in the hemi-arch replacement group had longer cardiopulmonary bypass and aortic cross-clamp times, along with an increased risk of blood transfusion, but similar freedom from repeat aortic arch operation and survival. We identified no advantage of performing hemiarch replacement in the absence of aortic arch dilation.

      Graphical abstract

      Key Words

      Abbreviations and Acronyms:

      CI (confidence interval), HR (hazard ratio), OR (odds ratio)
      Figure thumbnail fx2
      Kaplan-Meier estimates of survival in the ascending and hemiarch groups.
      We identify no advantage to hemiarch replacement in comparison to ascending aorta replacement with respect to follow-up repeat arch operation or survival in the absence of aortic arch dilation.
      Hemiarch and ascending aorta replacement can be done with low morbidity and mortality during bicuspid aortic valve replacement. Hemiarch replacement had longer bypass and cross-clamp times, higher risk of blood transfusion, but similar follow-up freedom from repeat aortic arch operation and survival. We identify no advantage of hemiarch replacement in the absence of aortic arch dilation.
      See Commentaries on pages 21 and 23.
      Ascending aortopathy is common in patients with a bicuspid aortic valve. The reported prevalence of aneurysm ranges from 20% to 40% in patients with a bicuspid aortic valve.
      • Michelena H.I.
      • Khanna A.D.
      • Mahoney D.
      • Margaryan E.
      • Topilsky Y.
      • Suri R.M.
      • et al.
      Incidence of aortic complications in patients with bicuspid aortic valves.
      ,
      • El-Hamamsy I.
      • Yacoub M.H.
      A measured approach to managing the aortic root in patients with bicuspid aortic valve disease.
      Current United States and European guidelines are in agreement that concomitant ascending aorta replacement should be performed in selected patients at the time of bicuspid aortic valve replacement
      • Hiratzka L.F.
      • Creager M.A.
      • Isselbacher E.M.
      • Svensson L.G.
      • Nishimura R.A.
      • Bonow R.O.
      • et al.
      Surgery for aortic dilatation in patients with bicuspid aortic valves. A statement of clarification from the American College of Cardiology/American Heart Association task force on clinical practice guidelines.
      • Vahanian A.
      • Alfieri O.
      • Andreotti F.
      • Antunes M.J.
      • Barón-Esquivias G.
      • Baumgartner H.
      • et al.
      Guidelines on the management of valvular heart disease (version 2012). Joint task force on the management of valvular heart disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS).
      • Borger M.A.
      • Fedak P.W.M.
      • Stephens E.H.
      • Gleason T.G.
      • Girdauskas E.
      • Ikonomidis J.S.
      • et al.
      The American Association for Thoracic Surgery consensus guidelines on bicuspid aortic valve-related aortopathy: executive summary.
      ; however, how best to replace the ascending aorta—hemiarch replacement or clamped ascending aorta replacement—remains a matter of debate.
      • Bilkhu R.
      • Youssefi P.
      • Soppa G.
      • Theodoropoulus P.
      • Phillips S.
      • Liban B.
      • et al.
      Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      • Singh R.
      • Yamanaka K.
      • Reece T.B.
      Hemiarch: the real operation for ascending aortic aneurysm.
      There are limited published studies comparing the outcomes of hemiarch replacement versus simple clamped ascending aorta replacement during bicuspid aortic valve replacement. No study to date has demonstrated a clear benefit for hemiarch replacement in patients with bicuspid aortic valve.
      • Bilkhu R.
      • Youssefi P.
      • Soppa G.
      • Theodoropoulus P.
      • Phillips S.
      • Liban B.
      • et al.
      Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
      ,
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      In a study of 168 patients, a St George's Hospital group concluded that prophylactic arch replacement was not supported in patients with bicuspid aortic valve.
      • Bilkhu R.
      • Youssefi P.
      • Soppa G.
      • Theodoropoulus P.
      • Phillips S.
      • Liban B.
      • et al.
      Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
      A University of Pittsburg group found that hemiarch replacement did not increase operative risk and thus concluded that its use should not be limited.
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      Despite the lack of any demonstrable evidence-based clinical benefit to hemiarch replacement, controversy persists. The Inova Fairfax Hospital and University of Colorado groups entitled a recent review article “Hemiarch: The real operation for ascending aortic aneurysm.”
      • Singh R.
      • Yamanaka K.
      • Reece T.B.
      Hemiarch: the real operation for ascending aortic aneurysm.
      To address this controversy, we reviewed our experience in patients with aortopathy and bicuspid aortic valve disease. The focus of the study was on the outcomes of freedom from repeat aortic arch operation and survival in patients who underwent either hemiarch replacement or simple clamped ascending aorta replacement during initial bicuspid aortic valve replacement.

      Materials and Methods

      This study was approved by our Institutional Review Board (approved September 29, 2017, approval 17-007553). We retrospectively reviewed the records of 723 consecutive patients who underwent ascending aorta replacement at the time of initial nonemergent bicuspid aortic valve replacement between January 2000 and June 2017. The initial cohort query excluded patients with aortic dissection. We then excluded 21 patients (3%), including 20 with active infective endocarditis and 1 with a connective tissue disorder. Thus left a total of 702 patients (97%) eligible for the study.
      We assessed differences in outcomes between patients who underwent clamped ascending aorta replacement (ascending group; n = 477, 68%) and those who underwent open hemiarch and ascending aorta replacement during circulatory arrest (hemiarch group; n = 225, 32%). The choice of procedure was at the discretion of the operating surgeon. This included the decision to perform hemiarch replacement versus clamped ascending aorta replacement, as well as any other additional operative procedure (eg, valve conduit vs separate valve and ascending aorta replacement). Yearly case volumes are reported in Figure E1.
      Patient baseline, operation, and outcome data were recorded using criteria defined in the Society of Thoracic Surgeons adult cardiac surgery database. Coronary artery disease was a compilation of the variables of previous myocardial infarction, coronary artery bypass graft operation, or coronary artery stenosis of ≥50%.
      Aorta measurements were recorded for the mid-ascending aorta at the level of the right main pulmonary artery and the aortic arch at the level just proximal to the left common carotid artery. Data were first obtained from direct measurement of a computed tomography or magnetic resonance imaging scan, then from the echocardiography report, and finally from the surgeon's notes. The source of the measurement is reported in Table E1 for mid-ascending aorta diameter and Table E2 for aortic arch diameter.
      The endpoints of the study were in-hospital procedure-related complications, repeat aortic arch operation after discharge, and survival. The last dates of clinical and vital status follow-up were determined through a review of the electronic medical record and Department of Cardiovascular Surgery patient surveys sent out to patients at 1, 3, 5, 10, 15, and 20 postoperative years. A yearly vital status review was obtained through Accurint (LexisNexis, New York, NY).
      Categorical data are reported as count (percent), and continuous data are reported as median (interquartile range [IQR]). Categorical data were analyzed with Fisher's exact test or the χ2 test, as appropriate. Continuous data were analyzed with the Wilcoxon rank-sum test. Logistic regression models were used to assess treatment effect of hemiarch replacement and concomitant valve conduit replacement on procedure-related blood transfusion and complication endpoints.
      The median duration of follow-up was calculated using the reverse Kaplan-Meier estimator, whereas survival estimates were calculated using the Kaplan-Meier estimator. The cumulative incidence of a repeat aortic arch operation was estimated accounting for the competing risk of death. Cause-specific Cox proportional hazard models were used to compare variables with time-dependent distributions. Multivariable model covariates were selected a priori. The α level was set at 0.05. All statistical analyses were done with R version 3.4.2 (R Project for Statistical Computing, Vienna, Austria).

      Results

      The median patient age was 60 years (IQR, 51-67 years), 113 patients were female (16%), the median ejection fraction was 62% (IQR, 56%-66%), aortic valve moderate/severe stenosis was present in 462 patients (66%), and aortic valve moderate/severe regurgitation was present in 354 patients (50%). Baseline patient characteristic data were similar in the hemiarch and ascending groups except for a greater prevalence of hypertension in the hemiarch group (64% vs 55%; P < .015) (Table 1).
      Table 1Patient baseline data in the ascending and hemiarch groups
      VariableAscending group

      (N = 477; 68%)
      Hemiarch group

      (N = 225; 32%)
      P value
      Age, y, median (IQR)60 (51-67)61 (53-67).770
      Body mass index, kg/m2 median (IQR)28 (26-32)29 (26-33).084
      Creatinine, μg/dL, median (IQR)1.0 (0.9-1.2)1.0 (0.9-1.2).551
      Ejection fraction, %, median (IQR)63 (57-67)62 (55-66).115
      Aortic arch diameter, mm, median (IQR)33 (30-36)34 (30-37).127
      Ascending aorta diameter, mm, median (IQR)48 (45-51)50 (46-53)<.001
      Female sex, n (%)83 (17)30 (13).171
      Diabetes, n (%)40 (8)24 (11).327
      Hypertension, n (%)261 (55)145 (64).015
      Chronic lung disease, severe, n (%)5 (1)4 (2).478
      Peripheral vascular disease, n (%)23 (5)14 (6).438
      Coronary artery disease, n (%)132 (28)63 (28).928
      Atrial fibrillation, n (%)57 (12)27 (12).985
      Aortic valve stenosis, moderate/severe, n (%)320 (67)142 (63).300
      Aortic valve regurgitation, moderate/severe, n (%)239 (50)115 (51).804
      Urgent status, n (%)28 (6)12 (5).775
      IQR, Interquartile range.
      The mid-ascending aorta diameter measurement data were obtained from echocardiography in 511 patients (73%), radiologic scan in 185 (26%), and surgeon notes in 6 (1%). The median diameter of the mid-ascending aorta was 48 mm (IQR, 45-51 mm) in the ascending group and 50 mm (IQR, 46-53 mm) in the hemiarch group (P < .001). Importantly, the range of diameters of the ascending group (27-70 mm) encompassed that of the hemiarch group (34-66 mm) (Figure 1). The aortic arch measurement data were obtained from echocardiography in 332 patients (47%) and radiologic scan in 187 (27%), and were missing in 183 (26%). The median aortic arch diameter was 33 mm (IQR, 30-36 mm) in the ascending group and 34 mm (IQR, 30-37 mm) in the hemiarch group (P = .127) (Figure 2).
      Figure thumbnail gr1
      Figure 1Mid-ascending aorta diameter median, interquartile range, 10% and 90% levels, and range in the ascending and hemiarch groups.
      Figure thumbnail gr2
      Figure 2Aortic arch diameter median, interquartile range, 10% and 90% levels, and range in the ascending and hemiarch groups.
      Valve composite root replacement was performed less commonly in the ascending group compared with the hemiarch group (36% [n = 173] vs 63% [n = 142]; P < .001), but other operative procedures were distributed similarly in the 2 groups (Table 2). Procedure times were shorter in the ascending group for both cardiopulmonary bypass (97 minutes vs 188 minutes; P < .001) and aortic cross-clamp time (78 minutes vs 136 minutes; P < .001), even when stratified by the presence of concomitant valve conduit replacement (all P < .001) (Table 2). The median circulatory arrest time was 18 minutes (IQR, 15-21 minutes), and median temperature was 18°C (IQR, 18°C-18°C). Adjunctive cerebral perfusion was done in 64 patients (28%), including antegrade cerebral perfusion in 5 patients and retrograde perfusion in the other 59.
      Table 2Procedure-related data in the ascending and hemiarch groups
      VariableAscending group

      (N = 477)
      Hemiarch group

      (N = 225)
      P value
      Valve conduit, n (%)173 (36)142 (63)<.001
      Concomitant other cardiac operation, n (%)146 (31)73 (32).624
      Coronary artery bypass graft, n (%)81 (17)39 (17).908
      Mitral valve operation, n (%)17 (4)9 (4).775
      Tricuspid valve operation, n (%)2 (<1)2 (1).597
      Other cardiac operation, n (%)76 (16)37 (16).863
      Valve type, n (%).167
       Mechanical264 (55)121 (54)
       Biological202 (42)103 (46)
       Homograft11 (2)1 (<1)
      Aortic cross-clamp time, min, median (IQR)78 (58-98)136 (95-166)<.001
       Valve conduit operation90 (75-121)155 (126-176)<.001
       Separate aortic valve and ascending aorta replacement69 (50-90)100 (75-127)<.001
      Cardiopulmonary bypass time, min, median (IQR)97 (74-126)188 (158-210)<.001
       Valve conduit operation113 (93-148)197 (175-218)<.001
       Separate aortic valve and ascending aorta replacement87 (64-114)173 (140-196)<.001
      IQR, Interquartile range.
      Procedure-related morbidity rates were low in both groups with respect to new-onset dialysis (ascending group, <1%; hemiarch group, 0; P = 1.000), sepsis (1% each group; P = .658), stroke (2% vs 1%; P = 1.000), and mortality (1% vs 0; P = .555) (Table 3). Overall complication rates were also similar in the 2 groups (35% vs 38%; P = .440), but the repeat operation for bleeding rate was higher in the ascending group (5% vs 1%; P = .017). After adjusting for concomitant valve conduit replacement, hemiarch replacement was associated with an increased odds of the need for blood transfusion (odds ratio [OR], 1.62; 95% confidence interval [CI], 1.15-2.28; P = .006) but not the occurrence of any complication (OR, 1.10; 95% CI, 0.78-1.55; P = .582).
      Table 3Procedure-related morbidity and mortality in the ascending and hemiarch groups
      VariableAscending group (N = 477)Hemiarch group (N = 225)P value
      Mortality, n (%)3 (1)0 (0).555
      Any complication, n (%)168 (35)86 (38).440
      Atrial fibrillation, n/N (%)
      New-onset atrial fibrillation, pacemaker insertion, or dialysis, which excludes patients who had the characteristic(s) before the operation.
      128/420 (30)69/198 (35).276
      Pacemaker, n/N (%)
      New-onset atrial fibrillation, pacemaker insertion, or dialysis, which excludes patients who had the characteristic(s) before the operation.
      25/467 (5)14/222 (6).613
      Stroke, n (%)7 (2)3 (1)1.000
      Sepsis, n (%)3 (1)2 (1).658
      Dialysis, n/N (%)
      New-onset atrial fibrillation, pacemaker insertion, or dialysis, which excludes patients who had the characteristic(s) before the operation.
      1/476 (<1)01.000
      Blood transfusion, n (%)268 (56)145 (64).038
      Repeat operation for bleeding, n (%)24 (5)3 (1).017
      New-onset atrial fibrillation, pacemaker insertion, or dialysis, which excludes patients who had the characteristic(s) before the operation.
      Clinical follow-up was complete at a median duration of 5.4 years (IQR, 2.0-10.0 years) for all patients, 6.5 years (IQR, 2.7-10.1 years) for the ascending group, and 4.3 years (IQR, 1.7-7.0 years) for the hemiarch group (P < .001). An aortic arch operation under circulatory arrest was done after discharge in 9 patients: 6 in the ascending group (3 with structural valve deterioration, 2 with endocarditis, and 1 with valve thrombosis) and 3 in the hemiarch group (all for endocarditis). No operations were done for arch aortopathy or aneurysm. The cumulative incidence of repeat aortic arch operation after discharge (while accounting for the competing risk of death) was 1% at 5 years and 2% at 10 years. The rate of repeat aortic arch operation after discharge appears to be higher in the hemiarch cohort at earlier time points, although there was no statistically significant difference (HR, 1.66; 95% CI, 0.41-6.68; P = .480) (Figure 3).
      Figure thumbnail gr3
      Figure 3Cumulative incidence of repeat aortic arch operation after discharge in the ascending and hemiarch groups. Hemiarch group: hazard ratio, 1.66; 95% confidence interval, 0.41–6.68; P = .480.
      Vital status follow-up was complete in all patients at 6.0 years (95% CI, 5.3-6.8 years) for all patients, 5.0 years (95% CI, 3.7-5.1 years) for the hemiarch group, and 7.6 years (95% CI, 6.7-8.7 years) for the ascending group (P < .001). Survival was 94 ± 1% at 5 years, 80 ± 2% at 10 years, 57 ± 5% at 15 years. The median time to death using the Kaplan-Meier estimator was 16 years (IQR, 12 years to not available); in total, there were 97 deaths. A multivariable Cox proportional hazard model was created for survival with 9 clinically relevant variables listed in Table 4. Hemiarch replacement was not statistically associated with reduced mortality (HR, 0.83; 95% CI, 0.51-1.33; P = .439) (Figure 4). The predictors of mortality were increasing age (years; HR, 1.05; 95% CI, 1.03-1.08; P < .001), diabetes (HR, 2.69; 95% CI, 1.59-4.54; P < .001), severe chronic lung disease (HR, 3.69; 95% CI, 1.28-10.64; P = .015), and coronary artery disease (HR, 1.65; 95% CI, 1.04-2.59; P = .032). Year of surgery was not related to survival in univariate or multivariable models.
      Table 4Multivariable Cox proportional hazard model of survival
      VariableUnivariate analysisMultivariable analysis
      OR (95% CI)P valueOR (95% CI)P value
      Age1.08 (1.05-1.10)<.0011.05 (1.03-1.08)<.001
      Hypertension2.37 (1.50-3.74)<.0011.33 (0.81-2.18).255
      Diabetes3.01 (1.81-4.99)<.0012.69 (1.59-4.54)<.001
      Chronic lung disease, severe6.06 (2.21-16.62)<.0013.69 (1.28-10.64).015
      Peripheral vascular disease1.64 (0.72-3.75).2421.18 (0.50-2.78).702
      Coronary artery disease0.33 (0.22-0.49)<.0010.61 (0.39-0.96).032
      Hemiarch operation1.06 (0.67-1.68).8000.83 (0.51-1.33).439
      Valve conduit operation0.58 (0.37-0.89).0120.92 (0.57-1.47).720
      Valve type mechanical1 (Reference)1 (Reference)
      Valve type biological2.52 (1.65-3.83)<.0011.25 (0.77-2.06).369
      Valve type homograft0.93 (0.29-3.05).9080.77 (0.22-2.70).689
      OR, Odds ratio; CI, confidence interval.
      Figure thumbnail gr4
      Figure 4Kaplan–Meier estimates of survival in the ascending and hemiarch groups.

      Discussion

      This study compared the outcomes of hemiarch and ascending aorta replacement during concomitant bicuspid aortic valve replacement in 702 consecutive patients without aortic arch aneurysm (Figure 5). We found that hemiarch replacement required longer cardiopulmonary bypass time (188 minutes vs 97 minutes; P < .001) and aortic cross-clamp time (136 minutes vs 78 minutes; P < .001). Procedure-related complications (36%) and mortality (<1%) were similar in the 2 groups; however, the risk of receiving blood transfusion was higher in the hemiarch replacement group (OR, 1.62; 95% CI, 1.15-2.28; P = .006). Finally, the 2 groups had a similar cumulative incidence of repeat aortic arch operation after discharge (HR, 1.66; 95% CI, 0.41-6.68; P = .480) and survival (HR, 0.83; 95% CI, 0.51-1.33; P = .439).
      Figure thumbnail gr5
      Figure 5Study reporting important operative outcomes, need for repeat aortic arch operation, and survival. OR, Odds ratio; CI, confidence interval; HR, hazard ratio.
      Our treatment groups had similar baseline demographic and comorbidity characteristics with the exception of hypertension, which was more common in the hemiarch group (64% vs 55%; P = .015). Another important difference was the higher percentage of concomitant valve conduit replacement procedures in the hemiarch group (63% vs 36%; P < .001). We addressed the differences by including hypertension in the adjusted model for survival; furthermore, we included concomitant valve conduit replacement in the adjusted models for complication, blood transfusion, and survival. Given the number of patients in the study (n = 702), we feel that the multivariable analysis was appropriate to address potential confounding of our prespecified variables.
      The United States and European guidelines are in general agreement that ascending aorta replacement should be performed in selected patients with aortopathy at the time of bicuspid valve replacement.
      • Hiratzka L.F.
      • Creager M.A.
      • Isselbacher E.M.
      • Svensson L.G.
      • Nishimura R.A.
      • Bonow R.O.
      • et al.
      Surgery for aortic dilatation in patients with bicuspid aortic valves. A statement of clarification from the American College of Cardiology/American Heart Association task force on clinical practice guidelines.
      • Vahanian A.
      • Alfieri O.
      • Andreotti F.
      • Antunes M.J.
      • Barón-Esquivias G.
      • Baumgartner H.
      • et al.
      Guidelines on the management of valvular heart disease (version 2012). Joint task force on the management of valvular heart disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS).
      • Borger M.A.
      • Fedak P.W.M.
      • Stephens E.H.
      • Gleason T.G.
      • Girdauskas E.
      • Ikonomidis J.S.
      • et al.
      The American Association for Thoracic Surgery consensus guidelines on bicuspid aortic valve-related aortopathy: executive summary.
      The general recommendations are to replace the ascending aorta in selected asymptomatic patients (diameter ≥5.5 cm), asymptomatic patients with additional risk factors (diameter ≥5.0 cm), and patients undergoing aortic valve replacement (diameter ≥4.5 cm). Compliance with the guidelines appears safe.
      • Rinewalt D.
      • McCarthy P.M.
      • Malaisrie S.C.
      • Fedak P.W.
      • Andrei A.C.
      • Puthumana J.J.
      • et al.
      Effect of aortic aneurysm replacement on outcomes after bicuspid aortic valve surgery: validation of contemporary guidelines.
      The controversy surrounds how much of the aorta to replace.
      • Bilkhu R.
      • Youssefi P.
      • Soppa G.
      • Theodoropoulus P.
      • Phillips S.
      • Liban B.
      • et al.
      Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      • Singh R.
      • Yamanaka K.
      • Reece T.B.
      Hemiarch: the real operation for ascending aortic aneurysm.
      We found that hemiarch replacement required more cardiopulmonary bypass and aortic cross-clamp time to perform than ascending aorta replacement even when stratified by concomitant valve conduit replacement (differences of 91 and 58 minutes, respectively; both P < .001). The increased times are consistent with those reported by Malaisrie and colleagues
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      in their propensity-matched aortic root replacement study (differences of 65 and 36 minutes, respectively; both P < .001) and by Sultan and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      in their propensity-matched all types of cardiac surgery study (differences of 39 and 16 minutes; both P < .050). The findings are important because the duration of cardiopulmonary bypass time has been identified as an independent risk factor for procedure-related complications to include death.
      • Salis S.
      • Mazzanti V.V.
      • Merli G.
      • Salvi L.
      • Tedesco C.C.
      • Veglia F.
      • et al.
      Cardiopulmonary bypass duration is an independent predictor of morbidity and mortality after cardiac surgery.
      We believe that the additional time needed to perform hemiarch replacement resulted in greater need of blood transfusion in the hemiarch group compared with the ascending aorta group (64% vs 56%; P = .038). This is consistent with the findings of Salis and colleagues,
      • Salis S.
      • Mazzanti V.V.
      • Merli G.
      • Salvi L.
      • Tedesco C.C.
      • Veglia F.
      • et al.
      Cardiopulmonary bypass duration is an independent predictor of morbidity and mortality after cardiac surgery.
      who reported that increased cardiopulmonary bypass time was an independent risk factor for blood transfusion. Malaisrie and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      also found a higher rate of blood transfusion in their hemiarch group compared with their aorta replacement group (67% vs 51%; P = .009). Sultan and colleagues,
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      in contrast, reported markedly lower blood transfusion rates in the hemiarch and ascending groups (28% versus 24%). Although their hemiarch group had a higher transfusion rate, the difference was not statistically significant (P = .455).
      The open anastomosis of hemiarch replacement has been postulated to be technically easier or safer than ascending aorta replacement.
      • Singh R.
      • Yamanaka K.
      • Reece T.B.
      Hemiarch: the real operation for ascending aortic aneurysm.
      This may be perceived as being supported in our experience based on fewer repeat operations for bleeding. Malaisrie and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      reported no difference in repeat operation for bleeding (P = .84). Sultan and colleagues reported a significantly lower rate of repeat operations following hemiarch operation (1% vs 16%; P < .001); however, this was all-cause return to the operating room and not specifically identified as for bleeding.
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      It is possible that our ascending aorta group was more aggressively returned to the operating room for exploration whereas the hemiarch group was observed, resulting in a higher rate of blood transfusion.
      We found low and similar procedure-related morbidity event rates of dialysis (all patients, <1%; between-group difference, P = 1.000), stroke (1%; P = 1.000), and mortality (<1%; P = .555) in both treatment groups. Similar low event rates were also noted by Malaisrie and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      in their propensity-matched groups with respect to dialysis (2%; P = .31), stroke (3%; P = .31), and mortality (2%; P = .41). Sultan and colleagues
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      likewise reported comparable low event rates in their propensity-matched cohorts for dialysis (6%; P = .775), stroke (3%; P = .408), and mortality (3%; P = .408). Our event rates are in line with those reported following even isolated bicuspid aortic valve replacement.
      • Girdauskas E.
      • Disha K.
      • Borger M.A.
      • Kuntze T.
      Long-term prognosis of ascending aortic aneurysm after aortic valve replacement for bicuspid versus tricuspid aortic valve stenosis.
      The low percentages may be deceptively reassuring, because they are counter to the reported independent relationship between the duration of cardiopulmonary bypass and procedure-related complications, as noted by Salis and colleagues.
      • Salis S.
      • Mazzanti V.V.
      • Merli G.
      • Salvi L.
      • Tedesco C.C.
      • Veglia F.
      • et al.
      Cardiopulmonary bypass duration is an independent predictor of morbidity and mortality after cardiac surgery.
      We found a 2% incidence of repeat aortic arch operation at 10 years after discharge; importantly, there were no cases of repeat aortic arch operation for arch aortopathy or aneurysm; furthermore, we found that hemiarch replacement was not associated with improved freedom from any repeat arch operation. Bilkhu and colleagues
      • Bilkhu R.
      • Youssefi P.
      • Soppa G.
      • Theodoropoulus P.
      • Phillips S.
      • Liban B.
      • et al.
      Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
      reported no repeat operation on the arch or the remaining aorta at a median follow-up of 5.9 years after ascending aorta replacement. Malaisrie and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      similarly noted no repeat operation on the arch at a mean follow-up of 3.8 years after ascending aorta replacement. From our own institution in 2011, Park and colleagues
      • Park C.B.
      • Greason K.L.
      • Suri R.M.
      • Michelena H.I.
      • Schaff H.V.
      • Sundt III, T.M.
      Should the proximal arch be routinely replaced in patients with bicuspid aortic valve disease and ascending aortic aneurysm?.
      reported that in patients with paired echocardiography scans, the diameter of the aortic arch remained unchanged over a median follow-up of 4.2 years.
      Our median patient follow-up was 6.0 years (95% CI, 5.3-6.8 years) with Kaplan-Meier estimated survival of 94 ± 1% at 5 years and 80 ± 2% at 10 years. Multivariable analysis demonstrated that hemiarch replacement was slightly protective of mortality (HR, 0.83; 95% CI, 0.51-1.33), but the difference was not statistically significant (P = .439). Malaisrie and colleagues
      • Malaisrie S.C.
      • Duncan B.F.
      • Mehta C.K.
      • Badiwala M.V.
      • Rinewalt D.
      • Kruse J.
      • et al.
      The addition of hemiarch replacement to aortic root surgery does not affect safety.
      reported reduced survival at 5 years in the hemiarch group (88%) compared with the ascending aorta group (91%), but again the difference was not statistically significant (P = .24). Sultan and colleagues
      • Sultan I.
      • Bianco V.
      • Yazji I.
      • Kilic A.
      • Dufendach K.
      • Cardounel A.
      • et al.
      Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
      noted better survival at 5 years in the hemiarch group (86% vs 81%); however, the difference was also not statistically significant (P = .420). Mortality is a hard endpoint. We extended the Kaplan-Meier estimate of survival out to 10 years in the present series. In that regard, there appears to be equipoise between the 2 groups.
      This study included only patients without aortic arch aneurysm and is limited by its retrospective nature. This was an as-treated group of patients with incomplete data and a limited duration of follow-up, especially with respect to baseline and follow-up aortic arch diameter and modality of measurement. We cannot discern why the surgeon chose hemiarch replacement over ascending aorta replacement, which is a potential for both selection and treatment bias; furthermore, the low rates of procedure-related complications put the study at risk of both type I and II statistical errors. Finally, we lack information on other factors important to the development of aortopathy, such as quality of the aortic tissue, management of hypertension after operation, and others.

      Conclusions

      Hemiarch and ascending aorta replacement can be done with low procedure-related morbidity and mortality during initial bicuspid aortic valve replacement. Hemiarch replacement requires longer cardiopulmonary bypass and aortic cross-clamp times and is associated with a greater risk of blood transfusion. We found that repeat aortic arch operation and survival were similar in the 2 treatment groups. We identified no specific advantage to hemiarch replacement in the absence of aortic arch dilation. Our current practice is to remove the abnormal and aneurysmal aorta. Surgical judgement guides the use of hemiarch replacement.

      Conflict of Interest Statement

      Authors have nothing to disclose with regard to commercial support.

      Appendix E1

      Figure thumbnail fx3
      Figure E1Yearly case volumes in the ascending and hemiarch groups.
      Table E1Preoperative mid-ascending aorta measurement source in the ascending and hemiarch groups (P < .001)
      There were no missing data.
      GroupEchocardiographyRadiologic scanSurgeon's noteTotal
      Ascending group, n (%)369 (77)104 (22)4 (1)477
      Hemiarch group, n (%)142 (63)81 (36)2 (1)225
      Total, n (%)511 (73)185 (26)6 (1)702
      There were no missing data.
      Table E2Preoperative aortic arch measurement source in the ascending and hemiarch replacement groups (P < .001)
      GroupEchocardiographyRadiologic scanMissing dataTotal
      Ascending group, n (%)223 (47)106 (22)148 (31)477
      Hemiarch group, n (%)109 (48)81 (36)35 (16)225
      Total, n (%)332 (47)187 (27)183 (26)702

      References

        • Michelena H.I.
        • Khanna A.D.
        • Mahoney D.
        • Margaryan E.
        • Topilsky Y.
        • Suri R.M.
        • et al.
        Incidence of aortic complications in patients with bicuspid aortic valves.
        JAMA. 2011; 306: 1104-1112
        • El-Hamamsy I.
        • Yacoub M.H.
        A measured approach to managing the aortic root in patients with bicuspid aortic valve disease.
        Curr Cardiol Rep. 2009; 11: 94-100
        • Hiratzka L.F.
        • Creager M.A.
        • Isselbacher E.M.
        • Svensson L.G.
        • Nishimura R.A.
        • Bonow R.O.
        • et al.
        Surgery for aortic dilatation in patients with bicuspid aortic valves. A statement of clarification from the American College of Cardiology/American Heart Association task force on clinical practice guidelines.
        J Thorac Cardiovasc Surg. 2016; 151: 959-966
        • Vahanian A.
        • Alfieri O.
        • Andreotti F.
        • Antunes M.J.
        • Barón-Esquivias G.
        • Baumgartner H.
        • et al.
        Guidelines on the management of valvular heart disease (version 2012). Joint task force on the management of valvular heart disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS).
        Eur J Cardiothorac Surg. 2012; 33: 2451-2496
        • Borger M.A.
        • Fedak P.W.M.
        • Stephens E.H.
        • Gleason T.G.
        • Girdauskas E.
        • Ikonomidis J.S.
        • et al.
        The American Association for Thoracic Surgery consensus guidelines on bicuspid aortic valve-related aortopathy: executive summary.
        J Thorac Cardiovasc Surg. 2018; 156: 473-480
        • Bilkhu R.
        • Youssefi P.
        • Soppa G.
        • Theodoropoulus P.
        • Phillips S.
        • Liban B.
        • et al.
        Fate of the aortic arch following surgery on the aortic root and ascending aorta in bicuspid aortic valve.
        Ann Thorac Surg. 2018; 106: 771-776
        • Sultan I.
        • Bianco V.
        • Yazji I.
        • Kilic A.
        • Dufendach K.
        • Cardounel A.
        • et al.
        Hemiarch reconstruction versus clamped aortic anastomosis for concomitant ascending aortic aneurysm.
        Ann Thorac Surg. 2018; 106: 750-756
        • Singh R.
        • Yamanaka K.
        • Reece T.B.
        Hemiarch: the real operation for ascending aortic aneurysm.
        Semin Cardiothorac Vasc Anesth. 2016; 20: 303-306
        • Rinewalt D.
        • McCarthy P.M.
        • Malaisrie S.C.
        • Fedak P.W.
        • Andrei A.C.
        • Puthumana J.J.
        • et al.
        Effect of aortic aneurysm replacement on outcomes after bicuspid aortic valve surgery: validation of contemporary guidelines.
        J Thorac Cardiovasc Surg. 2014; 148: 2060-2069
        • Malaisrie S.C.
        • Duncan B.F.
        • Mehta C.K.
        • Badiwala M.V.
        • Rinewalt D.
        • Kruse J.
        • et al.
        The addition of hemiarch replacement to aortic root surgery does not affect safety.
        J Thorac Cardiovasc Surg. 2015; 150: 118-124.e2
        • Salis S.
        • Mazzanti V.V.
        • Merli G.
        • Salvi L.
        • Tedesco C.C.
        • Veglia F.
        • et al.
        Cardiopulmonary bypass duration is an independent predictor of morbidity and mortality after cardiac surgery.
        J Cardiothorac Vasc Anesth. 2008; 22: 814-822
        • Girdauskas E.
        • Disha K.
        • Borger M.A.
        • Kuntze T.
        Long-term prognosis of ascending aortic aneurysm after aortic valve replacement for bicuspid versus tricuspid aortic valve stenosis.
        J Thorac Cardiovasc Surg. 2014; 147: 276-282
        • Park C.B.
        • Greason K.L.
        • Suri R.M.
        • Michelena H.I.
        • Schaff H.V.
        • Sundt III, T.M.
        Should the proximal arch be routinely replaced in patients with bicuspid aortic valve disease and ascending aortic aneurysm?.
        J Thorac Cardiovasc Surg. 2011; 142: 602-607

      Linked Article

      • Commentary: Open hemi-arch replacement in bicuspid aortic valve aortopathy without arch dilatation? If it's not broken, no need to fix it!
        The Journal of Thoracic and Cardiovascular SurgeryVol. 161Issue 1
        • Preview
          In the current issue of the Journal, Greason and colleagues1 report an interesting study comparing an open hemi-arch replacement (OHAR) with a clamped ascending aortic replacement (CAAR) as a distal aortic repair strategy in patients with bicuspid aortic valve aortopathy. The authors reviewed 702 consecutive patients operated for a nonemergent bicuspid aortic valve replacement between 2000 and 2017 receiving either a CAAR (n = 477) or an OHAR (n = 225). The OHAR was conducted under deep hypothermia (18°C) with adjunctive cerebral perfusion in 28% of cases.
        • Full-Text
        • PDF
        Open Archive
      • Commentary: Aortic replacement for bicuspid aortic valve disease—How much is too much (or too little)?
        The Journal of Thoracic and Cardiovascular SurgeryVol. 161Issue 1
        • Preview
          Bicuspid aortic valve disease is one of the most common congenital cardiac malformations encountered. Although it is classified as a single pathology, recent findings suggest that bicuspid aortic valve disease may represent a heterogeneous range of phenotypes yielding a similar fusion of aortic valve leaflets.1 Nevertheless, bicuspid valvular disease is frequently associated with ascending aortopathy,2,3 and current European and American guidelines advocate for selective ascending aortic replacement, depending on the extent of aneurysmal degeneration at the time of valve repair.
        • Full-Text
        • PDF
        Open Archive