Whether transesophageal echocardiography (TEE) should be routinely performed before hospital discharge after Bentall surgery remains unclear. The investigators took advantage of this practice at their institution to evaluate its benefit.
All patients who had undergone the Bentall procedure at Bichat Hospital from January 2010 to March 2014 were included. For each patient, transthoracic echocardiographic and transesophageal echocardiographic data and clinical events were retrospectively collected from the various reports.
One hundred ninety-eight patients underwent the Bentall procedure during the study period. Postoperative TEE was performed in 117 patients (59.1%), including nine with abnormalities observed on transthoracic echocardiography (a vibrating element on the new prosthetic valve, suspicion of peritubular complications in two patients, and aortic regurgitation in six patients). In 108 patients, routine TEE was performed (i.e., without clinical indication beyond baseline postoperative imaging). Patients with and those without routine TEE were identical, except for more frequent endocarditis as an indication for surgery in patients with routine TEE. Routine TEE did not reveal any new findings that prior transthoracic echocardiography had not shown. The most frequent finding on transthoracic echocardiography or TEE was periaortic hematoma, which sometimes led to the performance of computed tomography. This imaging did not change the care of the patients in this population.
This study does not support the performance of TEE after Bentall surgery during the in-hospital course in the absence of a specific indication. Baseline postoperative imaging using TEE or computed tomography should preferably be recommended beyond the early postoperative period after periaortic hematoma has resolved.
The Bentall procedure has become the reference technique for treating the proximal ascending aorta (aneurysm, dissection) when the aortic valve cannot be spared (because of associated aortic valvular disease or intraoperative impossibility to keep the valve). The Bentall procedure involves the surgical implantation of a composite graft to replace the aortic valve, aortic root, and a portion of the ascending aorta, with reimplantation of the coronary arteries into the graft. In addition to proximal and distal anastomoses (usually at the annulus and in the tubular ascending aorta), the proximal coronary arteries are also anastomosed into the graft. Furthermore, in contrast to isolated aortic valve replacement, replacement of the ascending aorta is usually responsible for periaortic hematoma.
Numerous studies have described the main complications as well as long-term results and prognosis after this surgery. They concluded that transesophageal echocardiography (TEE) was useful to detect complications of this surgery, and some programs of systematic follow-up with TEE have been developed. However, few studies have focused on the early postoperative period. Those small studies were performed 20 years ago and included limited numbers of patients. As a result, there is no clear evidence supporting recommendations. The European Society of Cardiology guidelines published in 2014 recommend imaging 1 month after surgery of the ascending aorta, without specifying an imaging modality, and the American College of Cardiology and American Heart Association guidelines published in 2010 recommend imaging 1 month after ascending aortic surgery using computed tomography (CT) scan or magnetic resonance imaging. Both the European Society of Cardiology (2012) and the American Heart Association (2014) guidelines on valvular diseases recommend the performance of baseline postoperative transthoracic echocardiography (TTE) 6 to 12 weeks after aortic valve replacement, but they do not individualize with respect to Bentall intervention. The French Society of Cardiology recommends the performance of baseline postoperative TTE 2 to 3 months after aortic valve replacement and the performance of TEE if surgery of the ascending aorta is associated (Bentall). Last, the European Association of Cardiovascular Imaging and the European Association of Echocardiography note the importance of postoperative TEE in aortic diseases, without further precision.
As a result, TTE is routinely performed after Bentall surgery before discharge from the hospital, but the indication for TEE during this early postoperative period remains unclear. It was nevertheless a frequent practice at Bichat Hospital in recent years to routinely perform TEE before discharge in patients after Bentall surgery. Bichat Hospital is a cardiologic medical-surgical center including a department of cardiac surgery in which >1,000 heart lung bypass procedures are performed per year, a department of medical cardiology, a department of vascular surgery, and the national reference center for Marfan syndrome and related diseases.
We took advantage of this practice at our institution to evaluate the potential benefit associated with the routine performance (without clinical indication beyond baseline postoperative imaging) of TEE before hospital discharge after Bentall surgery.
All patients who had undergone the Bentall procedure at Bichat Hospital from January 2010 to March 2014 were included.
For each patient, we retrospectively collected data from the hospital files, as follows:
demographic characteristics (age and sex);
medical history (previous cardiac surgery, diabetes, hypertension, obesity, chronic renal or respiratory failure, indication for surgery and its timing [emergency or not], and the presence or absence of Marfan syndrome);
surgical data (presence or absence of a bicuspid aortic valve, names of the surgeon and the responsible senior medical doctor, and type of prosthetic valve implanted [biological or mechanical]);
medical events during hospitalization (atrial fibrillation, pericardial effusion requiring drainage, pacemaker implantation, stroke, fever, pneumonia, heparin-induced thrombocytopenia);
duration of hospitalization (after surgery, after ultrasound evaluation);
transthoracic and transesophageal echocardiographic evaluations before hospital discharge (date, left ventricular ejection fraction, mean and maximal aortic transvalvular gradient, presence or not and quantification of intra- or paraprosthetic regurgitation, maximal thickness of the periaortic hematoma and its location [if the hematoma was qualified as small and no measure had been made, we considered that the thickness was 5 mm], other remarkable features, and modifications of care after TEE); and
electrocardiographically gated cardiac and aortic CT, including computed tomographic angiography using an iodinated contrast agent (performance or not and its clinical consequences on care).
Only planned TTE and TEE performed in the echocardiography laboratory, when patients were out of the intensive care unit, were considered (TEE in the intensive care unit was always performed because it was required by the clinical course of the patient).
This study was an observational retrospective study. The conduct of the study did not interfere with the usual care of patients. TEE, TTE, and all examinations were performed in the context of care. In accordance with French law, no institutional review board approval was required.
Data are expressed as mean ± SD (quantitative variables) or as numbers and percentages (qualitative variables). Comparisons were made using unpaired t tests (quantitative variables) or χ 2 tests (qualitative variables). The program JMP version 7.0.1 was used (SAS Institute, Cary, NC).
Bentall procedures were performed in 198 patients between January 2010 and March 2014 in the Department of Cardiac Surgery at Bichat Hospital.
Indications for surgery are presented in Table 1 . Clinical characteristics of the patients are summarized in Table 2 . Most of the patients were men (83%), with a mean age of 55.6 years. Marfan syndrome was diagnosed in 24 patients (12%). Bicuspid aortic valves were observed in 79 patients (40%).
|Indication for surgery||n (%)|
|Aneurysm and AR||98 (50)|
|Aneurysm and AS||35 (18)|
|Aortic dissection||13 (6.6)|
|Aneurysm, AR, and AS||8 (4.0)|
|False aneurysm after previous Bentall procedure||1 (0.5)|
|Mean age (y)||55.6 ± 0.98|
|Previous cardiac surgery||37 (19%)|
|Chronic renal failure||4 (2.0%)|
|Chronic respiratory failure||2 (1.0%)|
|Marfan syndrome||24 (12%)|
|Bicuspid aortic valve||79 (40%)|
Clinical events occurring before discharge were new-onset postoperative atrial fibrillation in 73 patients (36.9%), pneumonia in 26 patients (13.1%), persistent atrioventricular block requiring pacemaker implantation in 24 patients (12.1%), pericardial effusion requiring surgical drainage in 24 patients (12.1%), stroke in 13 patients (6.6%; 12 ischemic strokes with one hemorrhagic transformation and one transient ischemic attack), and heparin-induced thrombocytopenia in six patients (3.0%). In-hospital mortality was 3.5%.
Among the 198 patients, seven underwent neither TTE nor TEE in the echocardiography laboratory, because of early death. One hundred ninety-one patients underwent TTE at least once in the echocardiography laboratory ( Figure 1 ).
Postoperative TEE had been performed in 117 of the 198 patients (59%). In nine patients, TEE was performed because of abnormalities observed on TTE. In the remaining 108 patients, routine TEE was performed (without clinical indication beyond baseline postoperative imaging).
Patients who underwent both TEE and TTE were indistinguishable from patients who underwent only TTE for most variables ( Table 3 ). The two only variables that differed significantly between the two groups were indication for Bentall surgery and the average size of the peritubular hematoma.With respect to indication for Bentall surgery, surgery for endocarditis led to more frequent TEE (91%) compared with surgery indicated either for aneurysm (60%) or for dissection (42%) ( P = .049), but the prevalence rates of fever without documented infection (negative blood cultures, no pneumonia or urinary, skin, or other apparent infection), elevated white blood cell count, and heparin-induced thrombocytopenia were similar in the two groups. The average size of the peritubular hematoma measured with TTE was smaller in patients in whom TEE was performed (4.8 vs 6.4 mm, P = .048). The clinical meaning of this observation is uncertain. When the measurement of the size of the hematoma obtained by TTE could be compared with the measurement obtained by TEE, which was possible in 93 patients, the size of the hematoma was considered larger on TEE (9.5 vs 4.8 mm, P = .001).
|Variable||TTE alone ( n = 74)||TTE and TEE ( n = 117)||P|
|Men||63 (85%)||97 (83%)||.70|
|Age (y)||56.7 ± 1.6||55.1 ± 1.3||.40|
|Hypertension||37 (50%)||47 (40%)||.20|
|Diabetes||5 (6.7%)||8 (6.8%)||1.00|
|Previous cardiac surgery||13 (18%)||19 (16 %)||.80|
|Obesity||12 (16%)||19 (16%)||1.00|
|Chronic renal failure||1 (1.4%)||2 (1.7%)||.80|
|Chronic respiratory failure||0 (0%)||2 (1.7%)||.30|
|Urgent surgery||6 (8.1%)||12 (10%)||.60|
|Occult fever||6 (8.1%)||16 (14%)||.20|
|Elevated white blood cell count||5 (6.7%)||7 (6.0%)||.90|
|Stroke||5 (6.7%)||7 (6.0%)||.90|
|AF||29 (39%)||44 (38%)||.20|
|HIT||1 (1.4%)||5 (4.3%)||.70|
|Hematoma thickness on TTE (mm)||6.4 ± 0.6||4.8 ± 0.5||.048|
|Mechanical valve||46 (63%)||83 (71%)||.20|
|Bicuspid aortic valve||29 (40%)||47 (40%)||.90|
|Marfan syndrome||6 (8.1%)||17 (15%)||.20|
|LVEF (%)||52 ± 1.2||54 ± 0.9||.40|
|Mean transaortic gradient (mm Hg)||10.3 ± 0.5||10.8 ± 0.4||.40|
TEE Performed after Abnormalities on TTE
TEE was performed in nine patients because of the following findings, recognized during TTE:
A vibrating element was observed on the new prosthetic valve that was not found on TEE in one patient.
Suspicion of peritubular complications was present in two patients. In one patient, TTE revealed color flow between the neo-Valsalva and the tube, with an aspect of kinking at the level of the anastomosis. TEE did not demonstrate any abnormal flow. CT confirmed the absence of communication between the lumen and the hematoma. In the other patient, an anterior hematoma 18 mm in size led to the performance of TEE and CT, which did not find any evidence of active bleeding within the hematoma.
Suspicion of nonphysiologic intra- or paraprosthetic regurgitations in six patients. Two patients had mild intraprosthetic aortic regurgitation. TEE demonstrated one physiologic regurgitation and one mild regurgitation, which did not alter the care of the patient. Four patients had paraprosthetic aortic regurgitation seen on TTE (two mechanical valves, one bioprosthesis, one homologous graft), which was also found on TEE in only one patient (mechanical valve). The leak was mild on TEE and on TTE in this last patient. The reality of these paraprosthetic regurgitations is questionable when the surgical anatomy is considered.
All other transesophageal echocardiographic procedures ( n = 108) were routine (i.e., without any clinical indication beyond baseline postoperative imaging). No new findings were obtained with TEE, notably no aortic prosthetic thrombosis, no pathologic prosthetic or paraprosthetic regurgitation, and no echocardiographic suggestion of endocarditis or aortic ring abscess. No complications related to TEE were reported, and postoperative hospitalization length was similar whether TEE was performed or not (16.3 days if TEE was performed vs 17.3 days if only TTE was performed, P = .60).
Other In-Hospital Imaging: Indications and Results of CT
CT was performed in 18 patients (9%). CT was performed for better visualization of periaortic hematomas in six patients after both TTE and TEE. In two patients, TEE was performed because of suspicion of active bleeding within hematomas on TTE (not confirmed by CT). In four patients, routine TEE was performed and showed inconclusive images: one active bleed within a false aneurysm at the base of the tube, which was not confirmed by CT; one hematoma with an unusual image including a hypoechogenic zone without Doppler color flow, which was diagnosed as an embedded air bubble within the hematoma on CT; and two large hematomas (30 and 18 mm). Similar findings were seen on CT, which are common in the early postoperative period and not pathologic ( Figure 2 ).