Clinical Outcomes after Mitral Valve Repair with the Physio II Annuloplasty Ring.

BACKGROUND: Mitral valve (MV) repair with annuloplasty is the standard of care in patients with primary degenerative mitral regurgitation (DMR). Newer generations of annuloplasty rings have been developed with the goals of closer reproduction of native annular geometry and easier implantation. This study investigates the short-term and 5-year clinical outcomes of MV repair with the Carpentier-Edwards (CE) Physio II annuloplasty ring. METHODS: This is an observational study including a total of 486 patients who underwent MV repair for DMR using the CE Physio II annuloplasty ring between 2011 and 2016. RESULTS: Mean age was 54.8 ± 12.1 years, 364 patients (74.9%) were males, and 84 patients (17.3%) presented with atrial fibrillation. Mean left ventricular ejection fraction was 62.3 ± 7.3%. Mean logistic EuroSCORE was 2.7 ± 2.4%. New York Heart Association functional class III-IV symptoms were present in 134 (27.6%) patients preoperatively. Isolated MV repair was performed via a right-sided mini-thoracotomy in 479 patients (98.6%). Concomitant procedures included ablation for atrial fibrillation in 83 patients (17.1%) and closure of atrial septum defect in 88 patients (18.1%). Median size of implanted annuloplasty rings was 34 mm (interquartile range: 34-38 mm). Mean cardiopulmonary bypass time was 116 ± 34 minutes and mean cross-clamp time was 74 ± 25 minutes. Thirty-day mortality was 0.4%. The Kaplan-Meier 4-year survival was 98.5%. Freedom from MV reoperation was 96.2 and 94.0% at 1 and 4 years. CONCLUSION: MV repair with the CE Physio II annuloplasty ring is associated with excellent midterm clinical outcome.

Complex cases of acquired pulmonary vein stenosis after radiofrequency ablation: is surgical repair an option?

AIMS: Results of catheter based interventional treatment for pulmonary vein stenosis (PVS) following radiofrequency ablation (RFA) for atrial fibrillation remain suboptimal. Surgical repair may represent an alternative therapy, though long-term results have not been thoroughly investigated. METHODS AND RESULTS: We retrospectively assessed all patients in our centre undergoing surgical repair for radiofrequency-induced PVS. Data regarding surgical technique, clinical outcome, and rate of pulmonary vein (PV) restenosis were collected and analysed. Between 2004 and 2016, the rate for PVS resulting from RFA for atrial fibrillation in our institution was 0.79% (76/9633). During this period, five male patients with multiple PVS (3 ± 1) underwent surgical repair of a total of 13 symptomatic PVS. Surgery was performed in a standard setting under cardiopulmonary bypass. Stenotic veins were incised longitudinally followed by a patch augmentation plasty using either bovine pericard (n = 7) or polytetrafluoroethylene (PTFE) patches (n = 5). Localization of incision was on the anterior side of the PV only (n = 8) or on both the anterior and posterior sides (n = 4). In one PVS lesion, mechanical dilatation was sufficient. Long-term follow-up after 60 ± 69 months revealed an average restenosis rate of 38%. Restenosis was defined as narrowing >70%. All patients reported clinical improvement of symptoms at follow-up. CONCLUSION: Even in the era of wide circumferential lesions, PVS still occurs. While surgical PV patch plasty represents a valuable treatment option, restenosis remains an issue during follow-up. Nevertheless, surgical repair achieves highly acceptable long-term results for RFA-acquired PVS. Hence, it should be routinely discussed as a therapeutic option in cases with multiple PVS.

Acute Effect of Mitral Valve Repair on Mitral Valve Geometry.

BACKGROUND: The aim of this study was to quantify acute mitral valve (MV) geometry dynamic changes throughout the cardiac cycle using three-dimensional transesophageal echocardiography (3D TEE) in patients undergoing surgical MV repair (MVR) with ring annuloplasty and optional neochord implantation. METHODS: Twenty-nine patients (63 ± 10 years) with severe primary mitral regurgitation underwent surgical MVR using ring annuloplasty with or without neochord implantation. We recorded 3D TEE data throughout the cardiac cycle before and after MVR. Dynamic changes (4D) in the MV annulus geometry and anatomical MV orifice area (AMVOA) were measured using a novel semiautomated software (Auto Valve, Siemens Healthcare). RESULTS: MVR significantly reduces the anteroposterior diameter by up to 38% at end-systole (36.8-22.7 mm; p < 0.001) and the lateromedial diameter by up to 31% (42.7-30.3 mm; p < 0.001). Moreover, the annular circumference was reduced by up to 31% at end-systole (129.6-87.6 mm, p < 0.001), and the annular area was significantly decreased by up to 52% (12.8-5.7 cm2; p < 0.001). Finally, the AMVOA experienced the largest change, decreasing from 1.1 to 0.2 cm2 during systole (at midsystole; p < 0.001) and from 4.1 to 3.2 cm2 (p < 0.001) during diastole. CONCLUSIONS: MVR reduces the annular dimension and the AMVOA, contributing to mitral competency, but the use of annuloplasty rings reduces annular contractility after the procedure. Surgeons can use 4D imaging technology to assess MV function dynamically, detecting the acute morphological changes of the mitral annulus and leaflets before and after the procedure.

Minimally Invasive Mitral Valve Surgery for Mitral Valve Infective Endocarditis.

BACKGROUND: The feasibility of minimally invasive mitral valve (MV) surgery in infective endocarditis (IE) has not been reported in detail. We assessed the safety, efficacy, and durability of the minimally invasive approach through a right anterolateral minithoracotomy for surgical treatment of MV IE. METHODS: A review of the Leipzig Heart Center database revealed 92 eligible patients operated on between 2002 and 2013. All patients had undergone minimally invasive surgery for IE. The indication for surgery was isolated IE of the MV in all patients. Baseline and intraoperative data, as well as clinical outcomes and short-term follow-up were analyzed retrospectively. RESULTS: The patients' mean age was 60.9 ± 15.3 years, the logistic EuroSCORE II was 19.6 ± 19.1%, and 64.1% (59) were male. MV repair was feasible in 23.9% (22/92) of patients. Repair techniques included annuloplasty ring implantation, anterior mitral leaflet resection, posterior mitral leaflet resection, and implantation of neochordae. MV replacement was performed in 69 patients (75%), a mitral annulus patch in 1 patient, and concomitant tricuspid valve surgery for tricuspid regurgitation in 5 patients. Bacteriological analysis showed staphylococcus infection in 45.5%, streptococcus in 36.4%, enterococcus in 13.6%, and others in 4.5%. The 30-day-mortality rate was 9.8% (9 patients). The 1-year follow-up showed a 1-year survival rate of 77.7 ± 4.4% and freedom from reoperation within 1 year due to reendocarditis of 93.3 ± 2.1%. CONCLUSIONS: The minimally invasive approach is suitable for the treatment of IE of the MV. It is a good technique in IE in selected patients.

Preoperative Predictors and Outcome of Triple Valve Surgery in 487 Consecutive Patients.

Background Triple valve surgery (TVS) is associated with an elevated risk for operative mortality and thus remains a surgical challenge. We report our experience and results of TVS procedures, especially with respect to identification of preoperative risk factors, to improve patient selection. Methods Between December 1994 and January 2013, 487 consecutive patients (240 male, 247 female) underwent TVS at the Heart Center Leipzig, University of Leipzig. The data were prospectively collected and retrospectively analyzed. Univariate and multivariable regression analyses were performed to identify risk factors. Results The 30-day mortality was 16.1% and the long-term survival at 1 year and 5 years was 71.8% and 54.6%, respectively. Multivariable logistic regression analysis identified previous myocardial infarction to be the only significant predictor for early mortality. Age, New York Heart Association functional class IV, previous myocardial infarction, dialysis, and liver dysfunction were identified as preoperative predictors for late mortality. Furthermore, an increase of operative risk, given for each year, was observed during the study period. In contrast, 30-day mortality decreased during the observation time. Conclusion TVS is associated with a high surgical risk. Long-term survival is decreased, but acceptable for these high-risk patients. The series demonstrates that increasing surgical risk, age, and comorbidities are the future challenges in TVS.

Learning minimally invasive mitral valve surgery: a cumulative sum sequential probability analysis of 3895 operations from a single high-volume center.

BACKGROUND: Learning curves are vigorously discussed and viewed as a negative aspect of adopting new procedures. However, very few publications have methodically examined learning curves in cardiac surgery, which could lead to a better understanding and a more meaningful discussion of their consequences. The purpose of this study was to assess the learning process involved in the performance of minimally invasive surgery of the mitral valve using data from a large, single-center experience. METHODS AND RESULTS: All mitral (including tricuspid, or atrial fibrillation ablation) operations performed over a 17-year period through a right lateral mini-thoracotomy with peripheral cannulation for cardiopulmonary bypass (n=3907) were analyzed. Data were obtained from a prospective database. Individual learning curves for operation time and complication rates (using sequential probability cumulative sum failure analysis) and average results were calculated. A total of 3895 operations by 17 surgeons performing their first minimally invasive surgery of the mitral valve operation at our institution could be evaluated. The typical number of operations to overcome the learning curve was between 75 and 125. Furthermore, >1 such operation per week was necessary to maintain good results. Individual learning curves varied markedly, proving the need for good monitoring or mentoring in the initial phase. CONCLUSIONS: A true learning curve exists for minimally invasive surgery of the mitral valve. Although the number of operations required to overcome the learning curve is substantial, marked variation exists between individual surgeons. Such information could be very helpful in structuring future training and maintenance of competence programs for this kind of surgery.

Gender differences in mitral valve surgery.

BACKGROUND: Knowledge regarding gender-specific mitral valve (MV) pathology and postoperative outcome is rare. We herein describe a single-center experience focusing on gender differences in MV surgery. MATERIALS AND METHODS: A total of 3,761 patients underwent minimal invasive MV surgery at our institution between 1999 and 2011. Demographic data, pre-, intra-, and postoperative characteristics have been collected, including details on MV pathology and surgical technique. Patient data have been analyzed with consideration of gender-specific differences. RESULTS: The cohort consisted of 2,124 male (56.5%; 58.8 ± 12.5 years) and 1,637 female (43.5%; 64.5 ± 13 years) patients. Mitral regurgitation was observed equally in women (91.3%) and men (92.4%). Additional MV stenosis has been diagnosed in 2.7% of men but in 13.9% of women (p < 0.001). Calcification of the posterior MV leaflet showed a similar trend: 20.1% in women compared with 6.5% in men. Prolapse of the posterior leaflet was present predominantly in men with 63.1 versus 35.7% in women (p < 0.001). Distinct MV repair differences were retrospectively detected between genders: posterior mitral leaflet resection was performed in 17.9% of men versus 10.1% of women; posterior mitral leaflet chordae replacement was performed in 39.3% of men compared with 20.4% of women. Prosthetic MV replacement was necessary in 26.8% of women compared with only 10.7% of men. Concomitant tricuspid valve surgery was mostly performed in women (14.4 versus 8.2%). Male patients showed a significant better postoperative long-term survival than females, with 96, 89, and 72% compared with 92, 82, and 58% after 1, 5, and 10 years, respectively (p < 0.0001). CONCLUSION: Substantial gender-specific differences regarding MV pathology, operative strategy, and long-term outcome are present that need to be addressed in clinical practice.

DynaCT-guided anatomical rotation of the SAPIEN XT valve during transapical aortic valve implantation: proof of concept.

BACKGROUND: Transcatheter aortic valve implantation (T-AVI) by using the Edwards SAPIEN (Edwards Lifesciences LLC, Irvine, California, United States) prosthesis is currently being performed by the implantation of valved stent into the aortic annulus without respecting the native commissural (rotational) orientation. Anatomical orientation may, however, be beneficial regarding optimal physiological valve performance, optimal coronary flow, and avoidance of the fully covered commissural stent part in front of the coronary ostia. With the recently introduced SAPIEN XT, transcatheter valve identification of the commissures during fluoroscopy is feasible. The aim of this study was to evaluate the concept of Dynamic-CT (DynaCT)-guided anatomical rotation of the SAPIEN XT valve during transapical-AVI (TA-AVI). METHODS: Intraoperatively, an automatically segmented DynaCT was performed using the Siemens Syngo Aortic ValveGuide (Siemens AG, Forchheim, Germany) software prototype. Commissures of SAPIEN XT could be identified with a high-quality fluoroscopic system. Before standard TA implantation, one radiopaque stent commissure of the crimped SAPIEN XT prosthesis was aligned with the native aortic valve commissure visualized by DynaCT. Resulting rotational orientation of the valve after implantation was assessed by transesophageal echocardiography. RESULTS: Feasibility of anatomical rotation was evaluated in 10 patients scheduled for TA-AVI by an interdisciplinary heart team. Mean logistic EuroSCOREs and Society of Thoracic Surgeons scores were 23.7 ± 4.9% and 8.6 ± 2.1%, mean aortic gradient improved from 46.0 ± 21.9 to 9.6 ± 3.1 mm Hg, and there was no death within 30 days. All valves were implanted successfully with none or trivial paravalvular regurgitation in seven patients, mild (1 + ) in two patients, and moderate (2 + ) in one patient. An optimal anatomical position could be achieved in six patients, minor rotational deviation (< 10 degrees) in three patients, and moderate deviation (10 to 20 degrees) in one patient only. CONCLUSIONS: DynaCT-guided anatomical rotation of the SAPIEN XT valve is feasible during TA-AVI, avoiding implantation of the fully covered commissural posts in front of the coronary ostia. This might reduce the risk of coronary obstruction. In addition, the technique provides the potential benefit of physiological valve position and performance.

A propensity score-adjusted retrospective comparison of early and mid-term results of mitral valve repair versus replacement in octogenarians.

AIMS: Feasibility and efficacy of mitral repair in the elderly remain controversial. This study aims to compare outcomes of mitral repair and replacement in octogenarians. METHODS AND RESULTS: We compared the outcomes of 322 consecutive octogenarian patients (mean age 82.6 ± 2.2 years) who underwent mitral repair (n = 227, 70%) or replacement (n = 95, 30%) at Mount Sinai Medical Center and Leipzig Herzzentrum between 1998 and 2008 using propensity score adjustment and univariate and multivariate analyses. Patients undergoing aortic valve replacement were excluded. Coronary bypass was performed in 47.5% (n = 153), and 31.1% (n = 100) required tricuspid repair. Propensity score adjustment yielded comparable groups. Thirty-day mortality in patients undergoing primary elective mitral repair for degenerative disease was 5.1% (2/39). Overall 90-day mortality was 18.9% (43/227) for repair compared with 31.6% (30/95) for replacement (P = 0.014). Pre-discharge echocardiography revealed less than moderate residual regurgitation in 99% of patients (231/232). Adjusted 1-, 3-, and 5-year survival for patients undergoing mitral repair was 71 ± 3, 61 ± 4, and 59 ± 4%, respectively, compared with 56 ± 5, 50 ± 6, and 45 ± 6% for patients undergoing mitral replacement (P = 0.046). Multivariate analysis demonstrated emergency surgery, previous myocardial infarction, concomitant coronary artery bypass surgery, and mitral replacement to be strong independent predictors of early mortality; mitral valve replacement was an independent predictor of reduced survival in degenerative patients. CONCLUSION: Elective mitral repair can be performed with low operative mortality and good long-term outcomes in selected octogenarians with degenerative mitral disease, and is associated with better long-term survival than mitral replacement. The survival benefit associated with surgery for non-degenerative disease is more questionable.

Innovations in minimally invasive mitral valve pair.

Mitral valve (MV) insufficiency is the second most common heart valve disease represented in cardiac surgery. The gold standard therapy is surgical repair of the valve. Today, most centers prefer a minimally invasive approach through a right-sided mini-thoracotomy. Despite the small access, there is still the need to use cardiopulmonary bypass (CPB), and the operation has to be performed on the arrested heart. New devices have been developed to optimize the results of surgical repair by implementing mechanisms for post-implantation adjustment on the beating heart or the avoidance of CPB. Early attempts with adjustable mitral annuloplasty rings go back to the early 1990s. Only a few devices are available on the market. Recently, a mitral valve adjustable annuloplasty ring was CE-marked and is under further clinical investigation. In addition, a sutureless annuloplasty band to be implanted on the beating heart is under preclinical and initial clinical investigation for transatrial and transfemoral transcatheter implantation. Furthermore, new neochord systems are being developed, which allow for functional length adjustment on the beating heart after implantation. Some devices were developed for percutaneous MV repair implanted into the coronary sinus to reshape the posterior MV annulus. Other percutaneous devices are directly fixed to the posterior annulus to alter its shape. Several disadvantages have been observed preventing a broad clinical use of some of these devices. There is a continuous effort to develop innovative techniques to optimize MV repair and to decrease invasiveness.

Double intrathoracic arterial cannulation plus peripheral cannulation for whole-body perfusion in an infant.

We report a technique for distal body perfusion in an infant with hypoplastic aortic arch and isthmus stenosis by ultrasound- guided cannulation of the femoral artery using an intra-arterial vascular sheath establishing whole-body perfusion by triple cannulation.

Triple-Arterial Cannulation Approach for Whole-Body Perfusion in Infant Hypoplastic Aortic Arch and Coarctation Repair.

Organ and end-organ protection in aortic arch surgery represents a substantial challenge, especially in infants. Selective antegrade cerebral perfusion has been reported to improve organ function during this procedure. Visceral perfusion can be optimized by cannulation of the descending aorta during infant aortic arch surgery, leading to a decrease in end organ damage. However, it is associated with extensive surgical manipulation and subsequent risk of major vessel and potential organ damage. In this report, we describe a technique for distal body perfusion in an infant with hypoplastic aortic arch and isthmus stenosis by ultrasound-guided cannulation of the femoral artery using an intra-arterial vascular sheath establishing whole-body perfusion by triple cannulation.

Embolic occlusion of the left main coronary artery following an isolated aortic valve replacement.

Coronary occlusion after aortic valve replacement due to embolization is a rare complication. We report the case of a patient who developed acute heart failure due to occlusion of the left main coronary artery following an aortic valve replacement. Successful treatment was achieved with emergent coronary bypass surgery.

The reinforced full-root technique for the Ross operation: surgical considerations and operative insights.

BACKGROUND: Surgical treatment of young and middle-aged patients suffering from aortic valve disease remains an unresolved issue due to the limited durability of bioprosthetic heart valve replacements and the valve-related morbidity of patients with mechanical valve substitutes. Theoretically, the "living valve" principle of the Ross operation may represent a potentially viable solution to this dilemma. In this paper, we report on the surgical techniques of the Ross procedure and present long-term post-operative outcomes using the reinforced full-root technique. METHODS: From 1995 to 2020, a total of 832 consecutive patients (mean age, 43.4±13.7 years; 617 males) underwent a Ross operation using the full-root technique. Patients were prospectively monitored with clinical and echocardiographic follow-up. Total follow-up was 9,046 patients-years and was 92% complete. Mean-follow-up was 10.9±6.9 years (range, 0-24.9 years). RESULTS: Survival at twenty years was 92% (95% CI: 90-94%). Freedom from autograft or right ventricle to pulmonary artery connection reoperation at twenty years was 79% (95% CI: 74-85%). Eighty-nine pulmonary autograft reoperations had to be performed in eighty patients; salvage of the pulmonary autograft could be performed in forty-six of them. Fifty-seven patients required sixty-three reoperations on the right ventricle to pulmonary artery connection. Major cerebral bleeding occurred in one patient and neurological events in seventeen patients, respectively. CONCLUSIONS: Over a follow-up interval of up to twenty-five years, the Ross operation with the reinforced full-root technique demonstrated excellent survival in young and middle-aged patients. The rate of pulmonary autograft and right ventricular outflow graft reoperations were low in this patient subset. Therefore, the Ross operation with the reinforced full-root technique represents an enduring and valid treatment option in young and middle-aged patients suffering from aortic valve disease.

The zone 2 concept and distal stent graft positioning in TH 2-3 are associated with high rates of secondary aortic interventions in frozen elephant trunk surgery.

OBJECTIVES: The goal of this study was to investigate the association between the localization of the distal anastomosis (zone 2/3), the stent graft length (100-160 mm), the position of the distal end of the hybrid prosthesis and the need for secondary aortic intervention (SAI) in acute and chronic thoracic aortic disease after the frozen elephant trunk procedure. METHODS: From 2009 through 2020, a total of 232 patients (137 men; mean age, 61.7 ± 13.8 years) were treated with the frozen elephant trunk procedure. The main indications were acute aortic dissection type A (n = 106, 46%), chronic aortic dissection type A (n = 52, 22%) and degenerative thoracic aortic aneurysm (n = 74, 32%). RESULTS: The rate of SAI was significantly higher when we performed a distal anastomosis in zone 2 rather than in zone 3, whereas the rate of SAI was less frequent if the distal positioning of the hybrid prosthesis was below TH 4-5. Combining the zone 2 concept and the short stent graft length (100 mm) was associated with a significantly higher rate of SAIs. Patients with a distal anastomosis in zone 2 were significantly less likely to have a recurrent laryngeal nerve injury (P < 0.001). However, no association between a specific arch zone of a distal anastomosis and the occurrence of spinal cord injury was observed. CONCLUSIONS: Rates of SAIs are highest in patients who were treated with a distal anastomosis in zone 2 and a short stent graft (100 mm) with the distal end of the hybrid prosthesis at vertebral level TH 2-3.

Loop neochord versus leaflet resection techniques for minimally invasive mitral valve repair: long-term results.

OBJECTIVES: Non-leaflet resection techniques including loop chordal replacement are being used with increasing frequency, but the long-term results of these techniques are still unknown. The aim of this study was to compare the long-term results of loop neochord replacement with leaflet resection techniques in patients undergoing minimally invasive mitral valve (MV) repair for MV prolapse. METHODS: Between 1999 and 2014, 2134 consecutive MV prolapse patients underwent minimally invasive MV repair with isolated loop (n = 1751; 82.1%) or resection techniques (n = 383, 17.9%) at our institution. Follow-up data were available for 86% of patients with a mean follow-up time of 6.1 ± 4.3 years. RESULTS: The 30-day mortality was 0.8% for all patients (loop: 0.7%, resection: 1.6%; P = 0.09). Leaflet resection was associated with more moderate or more mitral regurgitation on predischarge echocardiography (P = 0.003). The 1-, 5- and 10-year survival rates were 98 ± 1%, 95 ± 1% and 86 ± 2% for the loop technique versus 97 ± 1%, 92 ± 1% and 81 ± 2% for resection patients, respectively (P = 0.003). Significant predictors for late mortality were MV repair technique (P = 0.004), left ventricular ejection fraction (P < 0.001), age (P < 0.001) and myocardial infarction (P < 0.001). Freedom from MV reoperation at 1, 5 and 10 years was 98 ± 1%, 97 ± 1%, 97 ± 1% and 97 ± 1%, 97 ± 1%, 96 ± 1% for patients operated on with the loop technique and leaflet resection (P = 0.4). CONCLUSIONS: In our patient cohort, MV repair with loop chordal replacement is associated with less early recurrent mitral regurgitation and very good long-term results when compared to classical leaflet resection techniques for MV prolapse and is therefore an excellent option for such patients.

Minimally invasive isolated tricuspid valve surgery.

BACKGROUND AND AIM OF THE STUDY: Outcomes after minimally invasive isolated tricuspid valve (TV) surgery have not been well described. Hence, an assessment was made of the authors' results for minimally invasive, isolated TV surgery. METHODS: Between September 2000 and January 2008, at the authors' institution, a total of 35 patients (15 males, 20 females; mean age 59.2 +/- 14.9 years) underwent isolated TV surgery for TV regurgitation, using a right lateral mini-thoracotomy. The preoperative left ventricular ejection fraction was 57 +/- 11%. The TV pathology included annular dilatation (n = 22), recurrent regurgitation after previous repair (n = 4), ruptured chordae (n = 4), endocarditis (n = 2), intracardiac tumor (n = 2), and blunt chest trauma (n = 1). Twenty patients had previously undergone a total of 30 cardiac operations, eight of which involved the TV. RESULTS: A TV repair was performed in 27 patients (77%), and involved the implantation of an annuloplasty ring in all cases. A leaflet repair was performed in addition to an annuloplasty in two patients, and eight patients underwent TV replacement. The hospital mortality was 5.7%, with two deaths due to low cardiac output syndrome on days 1 and 9 after surgery. The latter patient underwent reoperation on day 7 for recurrent TV regurgitation and a ventricular septal defect. Early and mid-term echocardiographic follow up revealed no TV regurgitation in 19 patients, but trivial and mild regurgitation each in eight patients. The mean follow up time was 35 +/- 40 months, and was 100% complete. A Kaplan-Meier analysis revealed an estimated five-year survival of 90% (95% CI: 73-97). CONCLUSION: Isolated TV surgery can be performed through a minimally invasive approach, with good results. A high repair rate can be achieved, and the procedure has been particularly valuable in redo surgery.