Abstracts DGTHG

Patient related outcome measurements for postoperative pain and mobility for transapical and transaortal TAVI

(Abstract ID: 128)

P. Wolf¹

¹Paracelsus Klinik, München

Background:

Transcatheter aortic valve implantation (TAVI) is an effective therapeutic method for well evaluated patients with symptomatic aortic valve stenosis. Beside the transfemoral (TF) TAVI procedure, typical approaches for surgical TAVI are the transapical (TA) and transaortal (TAO) access. Interestingly, patient related outcome measurements are rare concerning postoperative pain and mobility in surgical TAVI patients.

The aim of this prospective study was to compare patient related outcome concerning postoperative pain and mobility for transapical and transaortal TAVI.

Materials and methods:

Between September 2013 and June 2016 85 surgical TAVI patients were included in this study - 62 patients with TAO approach and 23 patients with TA approach. Postoperative chest pain was assessed at rest and after coughing using the numeric rating scale on day 1, 3, 7, 30, and 90 after TAVI procedure. The grade of mobility was evaluated at the same timepoints.

Results:

All 85 surgical TAVI procedures were successful. Mean age in the TA group was 79.34 ± 4.66 years, while it was 81.37 ± 7.87 years for the TAO group (p = 0,147). Demographic data demonstrated significant more female patients in TAO group compared to TA group (p= 0.001). The TA group showed an increased incidence of preoperative significant coronary heart disease compared to the TAO group (p= 0.008). Logistic EuroSCORE II tended to be higher in TA TAVI recipients (p= 0.01). 30-day mortality demonstrated no significant differences between the two groups investigated (p= 0.75). There were statistically significant differences in pain intensity when comparing TA and TAO TAVI approach at day 1 and day 30 after coughing (day 1: TA vs. TAO, p= 0.018 and day 30: TA vs. TAO, p= 0.039), as well as on day 30 at rest (day 30: TA vs. TAO, p= 0.01). For all other timepoints of pain measurements no significant differences could be observed in both groups investigated. Throughout the whole collective, the intensity of pain is greater with deep inspiration and cough compared to pain intensity at rest. The pain intensity in the entire patient population decreased during the survey period. In the multivariate regression models type of procedure (TA vs. TAO) was only a predictor of postoperative pain on day 30 (p< 0,001), with less pain intensity in TAO group.

Overall, both access routes showed a rapid re-mobilization despite the surgical approach. Seven days after TAVI procedure the patients almost matched the preoperative status. There were no significant differences between TA and TAO approach concerning grade of mobility at all timepoints investigated.

Conclusion:

Patient related outcome measurements demonstrated very good results for transapical and transaortal TAVI approaches for postoperative pain and mobility. All patients experienced both modest pain intensity and quick recover in mobility after TA and TAO access with slightly less pain in the TAO group.

The impact of pericardial versus porcine aortic valve replacement on clinical outcome of isolated aortic valve replacement

(Abstract ID: 135)

S. Tugtekin¹, M. Loose¹, M. Wilbring¹, K. Matschke¹, K. Alexiou¹

¹Herzzentrum Dresden

Background:

The choice of the most suitable aortic valve prosthesis (AVP) for isolated aortic valve replacement (AVR) is still a matter of debate and is influenced by various parameters, not least by the preference of the patient himself. A common decision is the one between bovine or porcine xenograft, since biological AVPs are now the most frequently implanted models in Germany.The aim of this study is the intra- and posthospital comparison of a bovine and a porcine xenograft after isolated AVR. The hemodynamic outcome of two selected valve prostheses postoperatively and further clinical follow up was investigated.

Materials and methods:

A retrospective single center study was conducted including 200 patients who received an elective isolated AVR between January 2000 and May 2012. Patients with reoperation, older than 75 years and endocarditis were excluded. The reference group was a group of patients who received a bovine AVP from Edwards Modell Perimount. These were compared with patients who received a porcine prosthesis from Medtronic model Hancock II or Ultra according to age, sex and EuroSCORE. Prostheses with a diameter of more than 25 mm or less than 19 mm were excluded. For the follow-up, the most recent echocardiographic findings of the patient at that time were determined by telephone interviews. Kaplan-Meier estimators were used to compare the survival rates of the patient collectives.

Results:

A total of 200 patients received either a bovine (n = 100) or porcine (n = 100) AVP. Both patient collectives showed a comparable preoperative risk with an additive EuroSCORE of 4.3 ± 1.3 % for the bovine AVR and 4.5 ± 1.2 % for the porcine AVR (P-value = 0.26519). The evaluation of the postoperative echocardiographic data showed a significant difference regarding the pressure gradients with a maximum and mean pressure gradient (ΔPmax, ΔPmean) of 31.42 ± 11.92 mmHG and 17.71 ± 7.11 mmHG respectively in the porcine group and 25.66 ± 8.06 mmHG and 14.53 ± 4.93 mmHG in the bovine group (P-value = < 0.01). Early mortality showed no significant difference and was 1 % in both groups. The average follow-up time was 5.35 ± 2.51 years for 77% (n = 154) of the patients. Also, posthospitally there were significant differences in hemodynamic outcome with a ΔPmax and ΔPmean in the porcine group of 32.48 ± 13.51 mmHG and 19.44 ± 8.98 mmHG respectively vs. 24.98 ± 10.35 mmHG and 4.39 ± 6.90 mmHG respectively in the bovine collective (P-value = < 0.01). However, no significant difference was observed for late and all-cause mortality. Mortality was 14 % and 16 % for the bovone group. The cumulative 5-and 10 year survival rates were 86 % and 74 % for the bovine group vs . 84 % and 72 %,respectively, for the porcine patient population and thus no significant difference (p=0.86333).

Conclusion:

Bovine aortic valve replacement demonstrated hemodynamic superiority, but no advantage in terms of early, late and all cause mortality compared to porcine valve replacement

Minimally-Invasive Approach to the Mitral Valve Does Not Impair Repairability

(Abstract ID: 267)

T. Doenst¹, G. Färber¹, N. Zeynalov¹, H. Kirov¹, S. Tkebuchava¹, M. Diab¹, C. Sponholz¹

¹Universitätsklinikum Jena

Background:

Mitral valve repair (MVR) is the gold standard for structural mitral valve regurgitation and high repair rates have been described by expert centers. Many surgeons have adopted minimally-invasive techniques approaching the mitral valve for various reasons. However, the concern has been raised that minimally-invasive access surgery compromises the quality of work "on site". We reviewed our experience with minimally invasive mitral valve cases scheduled for repair.

Materials and methods:

We selected all patients who had been scheduled for minimally invasive mitral valve repair from September 2010 to December 2016. We retrospectively analyzed the preoperative profiles, perioperative courses and surgical outcomes.

Results:

There were 420 patients with planned mitral valve repair through a right sided mini-thoracotomy approach (85% of all repair patients). Patient population wasd characterized as follows: mean age 65±12.4 years, male: 256 (61%), atrial fibrillation 193 (46%), treated diabetes mellitus 74 (18%), COPD 66 (16%), active endocarditis 15 (3.5%), previous neurological events 33 (7.9%), dialysis dependence 16 (3.8%), liver cirrhosis 21 (5%), previous cardiac surgery 26 (6.2%), EuroSCORE II of 3.7±4.9 (0.5 to 37.2). Mitral regurgitation was functional in 160 patients (38.1%) and structural in 252 patients (60%). Additional procedures were: tricuspid valve surgery 168 (40%), PFO closure 219 (52%), VSD closure 1 (0.2%), cryo ablation 164 (39%), LAA closure 135 (32%). Mean procedural duration was 177±42min and cross clamp time was 60±24min. The repair rate for mitral valve regurgitation was 99%, with 4 patients (1%) requiring mitral valve replacement due to failed repair. Two patients (0.5%) were converted to sternotomy due to severe lung adhesions in one and intraoperative Type A dissection in another patient. Postoperatively, 24 patients (5.7%), required re-exploration due to bleeding. Wound infection occurred in 7 (1.7%) patients, stroke occurred in 14 (3.3%). The 30-day mortality rate was 1.9%.

Conclusion:

Minimally-invasive access to the mitral valve, including a wide range of pathologies, does not seem to limit repair quality.

Individualized surgical strategy for asymmetric aortic root dilation – technique and outcome analysis of 65 patients with replacement of the tubular ascending aorta and the non-coronary sinus

(Abstract ID: 707)

C. Etz¹, J. Haunschild¹, M. Misfeld¹, P. Davierwala¹, M. Borger¹

¹Herzzentrum Leipzig

Background:

Asymmetric aortic root dilation, frequently localized in the non-coronary sinus in patients with a stenotic—often bicuspid/alized aortic valve—determines a surgical dilemma: isolated ascending replacement leaves structurally weakened (locally thinned-out) tissue behind with the patient possibly prone to long-term complications — alternatively, a Bentall procedure requires coronary reimplantation inevitably increasing operative risk. Non-coronary sinus repair (NCSR) cures the localized root aneurysm without no need to re-implant the coronaries. The aim of the study was to analyze surgical short and long-term outcomes for AVR and tubular ascending replacement with concomitant NCSR.

Materials and methods:

Our institutional database was retrospectively reviewed over a 19-year period (Jan 2000 to Aug 2018). After exclusion of reoperations, emergencies, and concomitant multiple valve or bypass procedures, a total of 644 elective primary aortic root replacements included 579 Bentall procedures (institutional standard as previously described with direct reimplantation of both coronary buttons into a prosthetic conduit) and 65 patients with NCSR. Repair of the non-coronary sinus (NCSR) involves resection of the entire non-coronary sinus to eliminate all structurally weakened aortic wall tissue from the annulus to the sinotubular junction between the left-to-non and right-to-non coronary commissures. After resection of the calcified aortic valve and implantation of the aortic valve prosthesis, a patch (Dacron or pericardium) is tailored to a ‘sinus-shape’. The patch is anchored at the nadir of the non-coronary sinus directly to the prosthetic ring of the valve implant running the suture continuously towards both adjacent commissures to achieve a durable and hemostatic repair. Thereafter, aortic repair is completed by connecting the non-coronary neosinus and the remaining 2/3 of the aortic root at the level of the sinotubular junction to the ascending graft; in a limited number of patients, the proximal part of the ascending prothesis can be tailored to form a ‘non-coronary lip’ and sutured in continuously for a seamless replacement.

Chi2, unpaired t-test, logrank test, logistic regression for in-house mortality, inverse Kaplan-Meier and Cox regression for time-to-death were utilized.

Results:

No significant difference in age (60.4±11 vs. 59.9±12 years), LVEF (56±11 vs. 58±11), arterial hypertension (82 vs. 82%), hyperlipidemia (49 vs. 42%), smoking (38 vs. 35%) and BMI (29.2±6 vs. 28.2±5), however NCSR patients had more often diabetes (18 vs. 12%, p=0.003) and pulmonary hypertension (23 vs. 13%, p=0.04). Nominal aortic diameters (CT-angio) were larger in the Bentall cohort at the root ( 3.1mm vs. NCSR; p=0.0012) and sinotubular junction level ( 5.3 mm vs. NCSR; p=0.0001 ), however equivalent at tubular ascending level (p=0.7).

The aortic cross-clamp time was in median 18 minutes shorter in the NCSR cohort (69 [53; 82] min), and the NCSR group required shorter ventilation times and had a shorter ICU stay. There was no significant difference in postoperative complications, i.e. bleeding, pericardial effusion.

Conclusion:

Non-coronary sinus repair is a safe alternative to the Bentall procedure offering low reoperation rates and equivalent longevity for selected patients with structural wall pathologies delimited to the non-coronary sinus in asymmetric aortic root dilatation.

Picture:

 

Applying the principles of deliberate practice to train CABG surgery: Development of the CABG module of the Early Exposure and Assessment (EASE) Training Program

(Abstract ID: 842)

F. Kirchhoff¹, C. Merz¹, W. Korte¹, E. Beckmann¹, F. Fleissner¹, M. Arar¹, T. Goecke¹, A. Haverich¹, M.-L. Shrestha¹, A. Martens¹

¹Medizinische Hochschule Hannover

Background:

Coronary artery bypass grafting (CABG) is highly standardized and requires a professionally trained team. Surgeons have to perform a high number of CABG procedures to achieve ideal results. As the learning curve is very steep in the beginning, the operation should be trained and simulated in advance. The amount and intensity of "deliberate practice" is a predictor of expert level performance in non-surgical and surgical domains. Deliberate practice is characterized as intense and focused training with the specific aim to improve performance. Herein we describe the development of a CABG simulation curriculum based on the principles of "deliberate practice".

Materials and methods:

In 2017 we initiated CABG simulation training using a biological low fidelity setup on chicken legs. In a first step needle control and instruments handling (level 1) were trained. In level 2 participants performed stitch simulations on a chicken legs (level 2a) followed by an anastomosis (level 2b). Surgical performance of participants increased significantly. However, translation into the OR needed a more realistic simulation setup. In 2018 additional levels were added. In level 3a and 3b, the procedure gains complexity by training on a porcine heart. In level 4a and 4b the procedure is trained with surgical loupes (2.5–3.0x magnification). Performance of each level is filmed before and after an intense training period was analyzed in subgroups of student and young residents. Experienced residents and fellows that did not engage in specific training tasks served as controls.

Results:

A curriculum was devised that defined component tasks with increasing complexity. Video tutorials and a 30-page training workbook were created. A training box for self-organized training sessions on porcine hearts was developed, including a LED ring light and a smartphone/webcam holder for videotaping. A scoring system was devised to assess surgical performance (handling, motion, tissue, product, time) in video recordings. Training tasks were timed. Based on the "deliberate practice" principles performance targets were defined for self-organized training sessions. Regular training workshops with a senior instructor were conducted for feedback. An entry and exit assessment were filmed in each level. Residents and students showed significant improvements in both technical accuracy and completion times. Training basic surgical techniques (e.g. stich angles, instrument handling) remains a major component of the curriculum in novices. Residents reached an overall performance level comparable to fellows. Students reached similar accuracy of surgical skills with longer completion times (Level 3B Exit Score/Time: Fellows: 18 [17–21] 1/min; Residents 17 [16-18] 1/min, P=0.13; Students 12 [10-15] 1/min, P=0.014).

Conclusion:

The principles of deliberate practice can be effectively applied to surgical training programs: Detailed print and video material are provided, component tasks are defined and performance targets set. Feedback is given through comparison of videotaped training performances with video tutorials and during regular instructed workshops. A target simulation performance should be defined before trainees perform real cases in patients. Dedicated simulation training should be a mandatory part of cardiothoracic residency. Protected training times should by scheduled on a regular basis. We hypothesize that deliberate practice simulation training of CABG procedures will improve learning curves and increase patient safety.

3D Printing of Patient-Specific Mitral Valve Replicas from Echocardiography for Preoperative Surgical Planning

(Abstract ID: 305)

M. Kruttschnitt¹, K. Tiemann¹, N. Hitschrich², T. C. Lueth¹, R. Sodian³

¹Technical University of Munich

²Tomtec Imaging Systems GmbH, Unterschleissheim

³Ludwig-Maximilians-University Munich

Background:

Mitral valve regurgitation is one the most common heart valve diseases and the favored therapy is mitral valve repair. However, the planning of this surgery is challenging. It is difficult to properly understand mitral valve anatomy and pathology spatially since preoperatively echocardiography images must be interpreted and intraoperatively the mitral valves collapse, making their shape and disease not clearly visible. Furthermore, the planning of mitral valve repair must be done purely mentally. To cope with these problems, 3D printed patient-specific mitral valve replicas were developed. They should simplify understanding of the patient-specific anatomy and enable preoperative planning of mitral valve repair by simulating the surgery on the replicas.

Materials and methods:

3D echocardiography of patients’ hearts was used to segment the mitral valve leaflets and the subvalvular apparatuses in systolic shape. Based on the segmentations, models of the leaflets and the subvalvular apparatuses were designed and manufactured using 3D printing. The mitral valve leaflets were not 3D printed directly, but cast from silicone with 3D printed casting molds, giving them a more realistic feel. The subvalvular apparatuses were 3D printed directly from soft material. Joining the mitral valve leaflets and the subvalvular apparatuses results in the patient-specific mitral valve replicas.

Results:

Patient-specific mitral valve replicas were manufactured for several patients and assessed by an expert heart surgeon. He found that the mitral valve replicas look and feel realistic and enhance anatomical understanding, since it is easier to inspect the physical, systolic mitral valve replica than to interpret echocardiography images or to inspect a collapsed mitral valve. Further it was possible for him to realistically simulate all common repair techniques, resection, implantation of artificial chordae tendineae and annuloplasty, on the replicas. He concluded that the replicas are a helpful tool for planning complex cases and can help to improve the outcome of the surgery.

Conclusion:

3D printed patient-specific mitral valve replicas were developed based on 3D echocardiography. The replicas are optically and haptically comparable to real mitral valves and allow simulation of all common mitral valve repair techniques. Inspection of the replicas can enhance anatomical understanding and the simulation of mitral valve repair on the replicas can be a valuable preparation for the operation of the patient.

Picture:

Patient-Specific Mitral Valve Replica for Preoperative Surgical Planning

Lost in Circulation

(Abstract ID: 268)

H. Kirov¹, S. Tkebuchava¹, G. Färber¹, M. Diab¹, T. Sandhaus¹, T. Doenst¹

¹Universitätsklinikum Jena

Background:

Developments in interventional cardiology have resulted in the ability to address more and more complex and heavily calcified coronary lesions. This may lead to a rise in complications such as broken, entrapped or left in place guide-wires or rotablator devices. In such cases surgical removal, potentially combined with coronary bypass grafting, is the most common therapy. Although several single case reports exist, an evaluation of surgical therapy for such complications in a larger case series is practically not existent.

Materials and methods:

We prospectively followed all cases of retained guide-wires and rotablator devices in 2015 and 2016, referred for surgical removal from five different cardiology departments. Additionally, we searched our database for such cases between 01/2010 and 12/2014.

Results:

While from 2015 on, seven cases with lost parts of cardiac interventions were referred, there was not a single case in the 5 years before. From the seven patients (5 male, 2 female), five were operated emergently. In one case, a broken guide-wire was an accidental finding during an LVAD implantation (14.3%). The patients were 60±5.7 years old, overweight (BMI 29.97±4.06) and except in one case (LVAD) with preserved ejection fraction (mean EF 53.5±17.07; EuroSCORE II was 4.76±3.21 for the whole group.

The broken devices were mostly located in the RCA (42.9%) followed by RCX (42.9%) and RD2 (14.3%). In two cases a vessel perforation was observed. The devices were mostly remnants of guide-wires (n=4); balloon catheters (n=2) and in 1 case a rotablator. The surgical approach was either through complete sternotomy (n=5) or left sided mini-thoracotomy (n=2). Out of the seven cases, four were operated On-Pump (57.1%) and three Off-Pump (42.9%). Complete extraction of the foreign bodies was possible in all patients. The patient with LVAD implantation died after 13 days in multi-organ-failure unrelated to the foreign body. All other patients survived and had uneventful postoperative courses.

Conclusion:

Retained foreign bodies from cardiac interventions such as broken guide-wires or rotablator devices can safely be removed surgically using individualized approaches. There appears to be a trend towards a rising incidence of such interventional complications possibly reflective of the increased complexity of interventions performed in many cath labs nowadays.

Interdisciplinary hybrid approach for combined mitral valve and coronary artery pathologies - a feasibility study of minimally invasive mitral valve surgery with PCI

(Abstract ID: 521)

M. Moscoso Ludueña¹, T. Ghazy¹, M. Hiresh¹, T. Andrasi², M. Irqsusi², A. Abugameh¹, M. Vondran², H. Nef³, A. Rastan²

¹Herz-Kreislauf-Zentrum Rotenburg an der Fulda

²Universitätsklinikum Gießen und Marburg, Standort Marburg

³Universitätsklinikum Gießen und Marburg, Standort Gießen

Background:

We conducted this study to evaluate the feasibility and safety of our hybrid concept of combined minimally invasive mitral surgery (MIMVS) and percutaneous coronary intervention (PCI) for patients with mitral valve (MV) and coronary artery pathologies.

Materials and methods:

Between 01/2012 and 04/2018, 534 patients underwent isolated or combined MIMVS at our institution. Eleven patients suffered from combined MV pathology and single-vessel coronary artery disease (CAD), who underwent a hybrid approach comprising a MIMVS with PCI of the CAD. The procedures were performed in a standard manner. The perioperative data of this highly selective cohort were retrospectively collected and analyzed. Continuous variables are reported as median with interquartile range.

Results:

The patients age, left ventricular ejection fraction and EuroSCORE II were 75 (64 - 80) years, 59 (37 - 60) % and 8.1% (2.5 - 8.6) respectively. Three patients had previous cardiac surgery. The time between MIMVS and PCI was 21 (7 - 162) days, where 9 patients received the PCI before and 2 patients after MIMVS. The right coronary artery was stented in 4 patients, the left anterior descending artery in 3 patients, the diagonal branch in one patient and the circumflex artery in 3 patients respectively. Ten patients were on dual antiplatelet therapy before surgery. The surgical duration was 230 (200–255) minutes. The bypass time was 137 (124 -152) minutes. Mitral valve repair could be performed in all patients. There was no conversion to sternotomy but 2 patients had to be operated on beating heart without aortic cross-clamping due to extensive adhesions. The ventilation time was 20 (15-34) hours and the length of stay on ICU was 3 (1-5) days. No procedural or in-hospital death occurred. One Patient needed postoperative hemofiltration. There was no re-exploration for bleeding, myocardial infarction or disabling stroke during the hospital stay. There was no significant difference in postoperative bleeding between the patients who underwent MIMVS first and those who underwent the PCI first.

Conclusion:

Our hybrid concept in patients with MV pathology combined with single-vessel CAD is safe and feasible. The MV pathology should not play a major role in the decision of revascularization strategy. In patients with CAD amenable to PCI therapy, the decision should not change in case of concomitant mitral valve surgery if the expertise for MIMVS is present.

Impact of aortic pathology on TEVAR-induced post-implantation syndrome

(Abstract ID: 650)

L. Volevskiy¹, I. Talipov¹, M. Vondran¹, M. Irqsusi¹, T. Andrasi¹, A. Rastan¹

¹Universitätsklinikum Marburg

Background:

Thoracic endovascular aortic repair (TEVAR) is the minimal-invasive treatment option of choice for almost all pathologies of the descending thoracic aorta. However, post-implantation syndrome (PIS) after TEVAR was shown to increase the risk of major adverse events, need for reintervention and all-cause mortality. The impact of aortic pathology on PIS has not yet been determined. The aim of this study was to investigate the biomarker profile and the clinical impact of PIS after TEVAR for different thoracic aortic pathologies.

Materials and methods:

50 patients undergoing TEVAR for aortic dissection (AD, 21), aortic aneurysm (AA, 19), and aortic rupture (AR, 10) were included in this retrospective study. The clinical endpoints were persistent inflammation at hospital discharge and in-hospital mortality.

Results:

AR patients were slightly younger (66.6±11.4) than AD (70.21±8) and AA patients (61.8 ±15.1 years), whereas AA patients had slightly damaged kidney function (33.33% vs. 26% and 10%, P=0.43) and suffered more form arterial hypertension (95.24% vs. 68.4% and 60.0%, P < 0.05) when compared to AD and AR patients, respectively. Whereas WBC remained comparable in all groups before (9.9±3.4, 8.6±2.7 and 9.6±2.7) and after the intervention (9.96±4.6, 9.94±2.7 and 8.6±2.6 in AD, AA and AR, respectively), CRP values increased in all groups after TEVAR (156.6± 94.5, P< 0.001; 108.1±57.7, P< 0.01 and 117.8±70.4, P< 0.05) compared to baseline levels (58.1±77.5, 31.94±52.1 and 31.9±52.1, in AD, AA and AR respectively), and this increase was more accentuated in the AD patients (P< 0.05 vs. AA). Hospital stay was similar in AD (12.6±7.3 days) and AA (12.7±7.2 days) patients and slightly prolonged in AR patients (15.7±11.2 days, P=0.192 vs. AD). PIS was diagnosed in 3.06% of all patients. Cohort-mortality was 8% and did neither correlate with PIS nor with the preoperative parameters.

Conclusion:

The presence of aortic dissection was associated with more inflammation after TEVAR than aortic aneurysms or aortic rupture. Inflammation did neither prolong hospital stay nor was a predictor for mortality after TEVAR. Thus, TEVAR remains safe for the treatment of both

Pseudocoarctation after frozen elephant trunk due to sharp kinking of a gothic arch in a patient with acute non-A-non-B aortic dissection

(Abstract ID: 812)

D. Gray¹, T. Schmitz-Rixen¹, P. Risteski¹

¹Universitätsklinikum Frankfurt am Main

Background:

Although the natural history of acute non-A non-B dissection is not well known, high early mortality with conservative treatment has been repetitively reported. Early single stage treatment using the frozen elephant trunk technique may close the entry, treat the involvement of the proximal aorta and promote early positive remodeling of the descending aorta.

Materials and methods:

A 63-year old male presented with acute non-A non-B aortic dissection with persistent pain and hypertension as well as true lumen collapse in the descending thoracoabdominal aorta. A gothic arch configuration with sharp angulation in zone 3 was additional finding. The patient underwent urgent minimal access aortic arch repair with the E-vita Open Plus® hybrid stent graft with distal anastomosis in zone 2 and debranching of the left subclavian artery. Intraoperative and early postoperative course was uneventful. Routine CT angiography showed a pseudocoarctation of only 5.5 mm at the level of the sharp angulation in zone 3, in addition to persistent near-true-lumen-collapse and false lumen perfusion. Resting ankle-brachial index was normal but the patient stated having fatique in his lower extremities during mobilization.

Results:

A stage II endovascular treatment was performed with relining of the hybrid prosthesis and distal extention with a Medtronic Navion stent graft via a transfemoral access. Secure true lumen cannulation was acquired via a right transbrachial thru-and-thru wire. For more stiffness and to avoid recoil of the stenosis of the flexible stentgraft, a stainless steal ballon expandable bare metal stent (Andramed XXL) was additionally placed at the level of the pseudocoarcation. Postoperative CT angiography showed a successful treatment of the pseudocoarctation with a lumen of 18 mm without true lumen collapse and reduced false lumen perfusion in the descending aorta. He was discharged symptom-free a week after the stage II intervention.

Conclusion:

Severe kinking of the aortic arch can cause pseudocoarctation of the hybrid prosthesis. Relining with a flexible stentgraft is possible even in a very steep angle. A higher stiffness ballon expandable bare metal stent is effective tool to reopen the pseudocoarctation and avoid re-kinking.

Picture:

 

Coronary artery surgery combined with carotid endarterectomy in patients with asymptomatic carotid artery stenosis: A multicenter randomized controlled trial (CABACS)

(Abstract ID: 1020)

S. Knipp¹, T. Holst², K. Bilbilis³, H. Jakob¹, A. Ruhparwar⁴, C. Weimar⁴

¹Westdeutsches Herz- und Gefäßzentrum, Universitätsklinikum Essen

²Herzzentrum Essen-Huttrop GmbH, Essen

³Institut für Medizinische Informatik, Biometrie und Epidemiologie, Essen

⁴Universitätsklinikum Essen

Background:

The debate on optimal management of patients with asymptomatic carotid artery stenosis undergoing coronary artery bypass graft (CABG) surgery has been ongoing for decades. We hypothesized that the perioperative event rates of carotid endarterectomy (CEA) combined with CABG versus isolated CABG were not equal.

Materials and methods:

This was an investigator-initiated, randomized, controlled, open, multi-center, group sequential study in 17 tertiary care cardiovascular centers in Germany and the Czech Republic. Patients scheduled for elective CABG were eligible if they had high-grade asymptomatic carotid artery stenosis assessed by Doppler/Duplexsonography. Patients were randomized using a web-based stratified blocked randomization (1:1, stratified by center, age < 60 versus >= 60 years, sex, and modified Rankin Score 0-1 versus 2-3) to receive either synchronous CEA and CABG or CABG without CEA. The primary efficacy composite endpoint was stroke or death within 30 days after operation. Events were adjudicated by blinded observers, and the primary analysis was performed on the intention-to-treat (ITT) population.

Results:

A total of 129 patients were enrolled in the trial. Two patients of the isolated CABG arm withdrew consent before operation and were excluded. Enrolment was terminated early due to withdrawal of funding following insufficient recruitment. Among 127 patients in the ITT population, the 30-day composite rate of non-fatal stroke or death was 12/65 (18.5%) patients in the synchronous CEA and CABG arm and 6/62 (9.7%) patients in the isolated CABG arm (absolute risk reduction 8.8%, 95% confidence interval:-3.2% to 20.8%; pWALD =0.12). There was also no evidence for a treatment group effect for all secondary endpoints at 30 days even though patients who had isolated CABG tended to have better outcomes. The trial was registered with Current Controlled Trials (ISRCTN13486906) and funded by the German Research Council (DFG, WE2585-3).

Conclusion:

Although our results cannot rule out a treatment group effect, a superiority of the synchronous combined CEA+CABG approach seems unlikely. Five-year follow-up of patients is still ongoing.

The impact of cardioplegia type on bypass flow measurements after coronary revascularizations for three-vessel disease

(Abstract ID: 637)

I. Vasiloi¹, L. Volevskiy¹, M. Moscoso Luduena¹, A. Abugameh¹, A. Rastan¹, T. Andrasi¹

¹Universitätsklinikum Marburg

Background:

The predictors and clinical effects of coronary bypass flows areinsuffic iently explored. This study thought to investigate pre-andintraoperative factors influenc ing bypass blood flow after sequential fullyarterialrevasc ularization (bilateral internal mammary artery as T-graft) of allmyocardial regions for coronary three-vessel disease.

Materials and methods:

Data of 102 patients were evaluated. Preoperative, intraoperative andpostoperative medical support was identic al in all patients to achieve standardhemodynamic and laboratory parameters. The effects of patient- and diseasecharacteristic s and of the surgical features on the intraoperatively measured totalbypass flow, as well as the association of bypass flows with postoperativeoutcome were analyzed.

Results:

Multivariate regression analyzes revealed that age (68.6±9.9), BMI (28.4±4.5), diabetes (41.2%) preop LV-EF (56.7±9.6), presenc e of left main disease(29.4%), the number of distal anastomoses (3.7±0.7), completeness ofrevasc ularization (64.7%) and revasc ularization of right coronary vessels(92.2%) did not affect intraoperative total bypass flows (85±48ml/min).Moreover, total bypass flow neither influenc ed postoperative CK-MB increase(max. 42±36 U/L), nor inversely c orrelated with operation time (252±61min),CPB time (94±24min), cardiac arrest (71±19min), intubation time (12±4h) andhospital stay (13±5d). Nonetheless, the use of warm blood cardioplegia (56.86%)was associated with higher bypass flows (92±48 vs.74±45ml/min, p=0.01) inspite of similar completeness of revasc ularization (67.2% vs. 61.3%),smaller number of distal anastomoses performed (3.6±0.6 vs. 3.9±0.8,p=0.04) and longer CPB time (98±27 vs. 88±23min, p=0.02) whencompared with patients receiving cold crystalloid c ardioplegia. The use ofblood cardioplegia was also associated with higher values of the lowestintraoperative Hemoglobin (9.1±1.6 vs. 8.0±1.2 g/dL, p=0.01) with similar max.lactate (1.6±0.3 vs. 1.8±0.5 mmol/L) and min. temperature (33.5±1.4 vs. 34.2±0.7°C); and lower postoperative increase of creatinine compared to preop.values (1.4±1.1 vs. 1.2±0.9, n.s. in blood- and 1.1±0.3 vs.1.0±0.3, p=0.03 incrystalloid cardioplegia).

Conclusion:

Total coronary bypass flow of sequential bypasses measured after CPB seem to beaffected more by the type of cardioplegia and its subsequent state of myocardialand systemic homeostasis than by heart function, number of distal anastomoses,CPB time or completeness of revasc ularization.

The value of transit-time flow measurement in different types of coronary revascularization techniques

(Abstract ID: 645)

M. Mantov¹, T. Andrasi¹, I. Talipov¹, M. Irqsusi¹, A. Rastan¹

¹Universitätsklinikum Marburg

Background:

The sequential bypass grafting (SQBG) is widely used in the coronary revascularization (CABG); however its superiority over multiple single bypass grafting (MSBG) remains unclear. The aim of the present study was to investigate the flows of sequential and multiple single coronary grafts using transit-time flow measurement.

Materials and methods:

From January 2017 to January 2019, 142 patients underwent isolated CABG for coronary disease (73 SQBG and 69 MSBG). Preoperative characteristics, intraoperative flow measurements and postoperative monitoring of cardiac enzymes were analyzed.

Results:

SQBG technique was not associated with significantly higher total bypass flow than the MSBG technique (181.6 ±78.4 vs. 163.3 ±70.1 ml/min respectively, p=0.14). A higher number of distal anastomosis has been also performed with SQBG than with MSBG (3.97 ±0.8 vs. 2.72 ±0.6 respectively, p<0.000), that was liable for the lower flow per anastomosis found in the SQBG group than in the MSBG (46.3 ±18.7 vs. 61.03 ±25.7 ml/min respectively, p<0.0001).

Similar maximal increase in postoperative CK-MB (50.7 ±61 and 54.3 ±78.8 U/l, p=0.76) and Troponin I (9722 ±15114 and 8229 ±14352 U/l, p=0.55) was assessed in both SQBG and MSBG, respectively.

Whereas total bypass flow correlated strongly with the number of distal anastomosis in both SQBG (R= 0.403, p<0.002) and MSBG groups (R= 0.346, p<0.004), the postoperative increase in cardiac enzymes (data showed for CK-MB) was not significantly improved neither by higher total bypass flow (R= 0.07, p=0.55 and R= 0.05, p=0.68), better flow per anastomosis (R = 0.06, p=0.62 and R= 0.07, p=0.58) nor by a higher number of distal anastomosis (R= 0.04, p=0.74 and R= 0.09, p=0.47) in none of the two groups respectively.

Conclusion:

Total bypass flow is strongly dependent of the number of distal anastomosis, independently from the type of revascularization technique. Total bypass flow does not influence the postoperative peak values of cardiac enzymes, independently from the revascularization technique. Nonetheless, the changes in cardiac enzymes were independent from the number of distal anastomosis. Last but not least, sequential bypass grafting achieved higher total coronary bypass flows that the multiple single bypass technique - probably due to the higher number of distal anastomosis performed in SQBG - however, without reaching the level of significance.

Peri- and intraoperative risk factors for postoperative delirium after different cardiac surgical procedures

(Abstract ID: 629)

I. Talipov¹, C. Arndt¹, G. Dinges¹, A. Rastan¹, T. Anrasi¹

¹Universitätsklinikum Marburg

Background:

Since there is inconsistency regarding the importance of patientcharacteristics, CPB-inflammation, microembolic load, blood loss, anesthesia, fluidmanagementand cardiac rhythm in the occurrenc e of postoperative delirium, thepresent study aims to analyze risk factors for delirium after different cardiacsurgical procedures with cardiac arrest.

Materials and methods:

Forty-one independent predictors for delirium were retrospec tively evaluated in300 patients undergoing CABG (n=150) or valve/aortic surgery /- CABG(n=150). Postoperative delirium identified by the confusion assessment method,the delirium observational screening scale and administration of haloperidoldidn’t differ between CABG and valve/aortic subgroups (28.7% and 35.33%,p=0.265).

Results:

Delirium patients were older (71.3±8.5 vs. 66.6±9.5, p<0.0001 and 70.0±9.6vs. 62.5±12.6, p<0.001, years) and required more RBCs (4.2±4.2 vs. 2.1±2.1,p<0.001 and 5.4±5.7 vs. 2.5±3.4, p<0.001) and FFPs (8.0±4.2 vs. 6.1±2.9,p<0.001 and 10.9±7.1 vs. 6.4±3.4, p<0.001) than non-delirium patients inboth CABG and valve/aortic groups, respec tively. Alcohol consume, smoking andCTD-percentage were similar.Delirium was associated with longer operation- (298.3±98.4 vs. 250.6±67.8 min,P<0.001), CPB- (171.5±54.9 vs. 140.9±45.8 min, P<0.001) cardiacarrest-(108.8±36.9 vs. 91.3±29.8 min, P<0.01) and reperfusion times (45.7±21.3 vs. 35.3±16.8 min, P<0.01) only in the valve/aortic group (delirium vs.non-delirium, respec tively).Multivariate regression analysis of the entire cohort identified age (p=0.0006,OR=1.1), CPB time (p=0.011, OR=1.0) ventilation time (p=0.015, OR= 1.0), RBCtransfusion (p=0.006, OR=1.2) and new onset of atrial fibrillation (NOAF) (p=0.012,OR=0.4) as independent risk factors for postoperative delirium (AUC80.4%, p<0.0001). Age (p=0.016), CPB time (p=0.005), RBC transfusion(p=0.03) and NOAF (p=0.0006) predicted delirium in the valve/aorticsubgroup (AUC 83.4%, p<0.001), whereas age (p=0.009) and ventilation time(p=0.012) were the only predictors for delirium in patients undergoing isolatedCABG (AUC 77.5%, p<0.001).

Conclusion:

Delirium was predicted by age, ventilation time, CPB time, RBC transfusion andNOAF, and was not predicted by the type of surgical procedure, despite of longeroperation and CPB times of valve/aortic surgeries. Solely age andmechanical ventilation time were identified as independent risk factors forpostoperative delirium after isolated CABG.

ADAR2 Attenuates GABAergic Neuron Function Through Mir-30d In Chronic Post Surgical Pain

(Abstract ID: 809)

D. Li¹, B. Li², Y. Shen², H. Wang², C. J. Bruns¹, Y. Zhao², Y. Xiong²

¹University Hospital Cologne

²Changhai Hospital, Naval Medical University, Shanghai

Background:

Chronic post-surgical pain (CPSP) is pain developing or increasing in intensity after a surgical procedure, in the area of the surgery, persisting beyond the healing process (ie, at least 3 months) and not better explained by another cause such as infection, malignancy, or a pre-existing pain condition. The specific mechanism of CPSP has not yet been clarified. A-to-I RNA editing is a post-transcriptional mechanism that converts the Adenosine (A) into Inosine (I) at the RNA level. This type of RNA editing is the most frequent in humans and is mediated by the Adenosine deaminases that act on RNA (ADAR) enzymes. GAD67 is in charge of GABA synthesis in neurons. This study aimed to elucidate the role and specific mechanism of ADAR2 in the regulation of GABAergic neuron function in CPSP from the rat spinal cord level.

Materials and methods:

The behavioral tests were used to detect the change of paw withdrawal threshold/time caused by Skin/muscle incision and retraction (SMIR). Whole-cell patch clamps were used to record changes in miniature inhibitory postsynaptic currents (mIPSC) of GABAergic neurons in spinal dorsal horn. High-throughput screening and RT-qPCR were used to detect changes in miRNAs after modeling. The dual fluorescent reporter gene detects possible binding sites for miR-30d and GAD67.

Results:

We found the significant hyperalgesia in rats after SMIR surgery and associated with elevated ADAR2 expression in spinal dorsal horn. SMIR surgery also up-regulated the expression of miR-30d expression. Intrathecal injection of ADAR2 siRNA significantly alleviated hyperalgesia and increased mir-30d expression induced by SMIR surgery. Dual fluorescent reporter gene assay confirmed that miR-30d inhibits the expression of GAD67 by interacting with a specific region of the 3'-UTR 1085-1109 of the Gad67 mRNA. The mIPSC frequency of GABAergic neurons decreased after modeling, demonstrating that SMIR impairs presynaptic GABA release by down-regulating GAD67. ADAR2 siRNA can inhibit the upregulation of miR-30d, rescue the GABAergic synaptic function of spinal dorsal horn neurons, up-regulate the expression of GAD67 and alleviate the pain behavior induced by SMIR surgery in rats.

Conclusion:

This study is the first study of in vivo and in vitro experiments on the functional significance and potential molecular mechanisms of ADAR2 involved in CPSP. It provides a new molecular mechanism and prevention target for CPSP.

Impact of gender on the postoperative neurological outcome following surgical replacement of ascending aorta in moderate hypothermic circulatory arrest. A 15 year single center experience

(Abstract ID: 144)

M. Salem¹, C. Friedrich¹, A. Thiem¹, M. A. Salem¹, Y. Erdal¹, K. Huenges¹, B. Panholzer¹, F. Schoeneich¹, J. Schoettler¹, T. Puehler¹, A. M. Eide¹, J. Cremer¹, A. Haneya¹

¹Universitätsklinik Schleswig-Holstein, Kiel

Background:

Gender is known to influence the postoperative outcome in patients undergoing cardiovascular surgery. Various studies focused on the relation between gender and postoperative outcome in coronary artery surgery, as well as valve replacement. However, the impact of gender on neurological outcome in patients undergoing replacement of ascending aorta in moderate hypothermic circulatory arrest (MHCA) down to 20°C is not consistently investigated. The aim of this analysis is to identify the relation between gender and postoperative neurological outcomes in patients undergoing elective replacement of ascending aorta in MHCA.

Materials and methods:

A retrospective analytical study included 905 consecutive patients between 2001 and 2015 (male 66.7% vs. female 33.3%) undergoing elective replacement of ascending aorta using MHCA due to aneurysm or calcification. Patients with type A-dissection of ascending aorta are excluded from the study.

Results:

In preoperatively history, women were significantly older than men (68.4±9.9 vs. 65.8±11.6 years; p=0.002) with significantly higher Euro-score II [4.7% (2.8;7.5) vs. 3.7% (2.2;6.6); p=0.001). Men have a significantly high coronary heart disease (42.9% vs. 30.6%; p<0.001) with more percutaneous coronary intervention (10.9% vs. 6.0%; p=0.016). Intraoperatively, cardiopulmonary bypass time [150 min (118;192) vs. 132 min (105;175); p< 0.001] and time of cross-clamping [97 min (73;129) vs. 83 min (56;110); p<0.001] were longer in males. The DHCA time was similar between both groups. Postoperatively, a significantly high incidence of neurologic complications was noted in men (7.0% vs. 3.3%; p=0.03). Otherwise, no significant differences in complications and major morbidity were observed between the groups. The 30-day mortality (women 4.9% vs. men 3.9%; p = 0.48) showed no statistical significance between both groups. The multivariate logistic regression analysis suggests that postoperative blood transfusion (OR 4.9; p=0.005), age (OR 2.9; p=0.015) and cardiopulmonary bypass time (OR 1.02; p<0.001) were independent risk factors for mortality. However, gender was not an independent risk factor for mortality in these patients.

Conclusion:

Our analysis showed that male gender has high incidence of neurological complications than female gender after surgical replacement of ascending aorta using MHCA in elective replacement of thoracic aorta. However, gender was not an independent risk factor for mortality in these patients.

Experience in the treatment of patients with esophageal cancer

(Abstract ID: 652)

A. Sukhodolia¹, S. Sukhodolia², V. Kernychnyi¹, I. Chubar², V. Mosiychuk², A. Shemchinska²

¹Vinnytsia National Medical University by name M.I. Pirogov, Vinnytsia

²Khmelnytsky Regional Hospital, Khmelnytsky

Background:

Esophageal cancer (EC) is among the 10 most common tumor locations and is one of the leading causes of cancer deaths. According to the National Cancer Registry of Ukraine, in 2017 identified 1,649 cases of disease and 1,312 died, respectively, the incidence rate is 4.6 and the death rate - 3.7. The high mortality rate, which is constantly monitored for a long time, creates the urgency of timely detection and proper treatment of EC. Surgery remains the main method of treatment for patients with EC.

Materials and methods:

During the last 4 years, 25 patients with MS aged 47-84 were operated in the clinic. The average age of patients is 63.8 ± 9.8 years. men were 24, women - 1. I stage n = 1 (4%), stage II n = 8 (32%), stage III n = 9 (36%), stage IV n = 7 (28%).

Results:

All patients were operated on. Structure of surgical interventions: extirpation of esophagus with temporary cervicostomy (fig.1) - 2 (8%), resection of esophagus according to Lewis-Ivory - 9 (36%), laparotomy + left-sided thoracotomy - 7 (28%), abdominal access -1 ( 4%). Palliative surgery was performed in 6 patients: gastrostomy - 5 (20%), enterostomy - 1 (4%).

Postoperative mortality was observed in 2 cases. In the first case, the patient with EC, stage III, with concomitant severe diabetes mellitus, was complicated by postoperative pleural empyema and phlegmon of the subcutaneous tissue of the chest. Despite the Buleau drainage and all the concomitant medical manipulations, the complications were incompatible with life. Otherwise, the patient was diagnosed intraoperatively with tumor germination into the mediastinal pleura and infiltration of the azygos vein. In the postoperative period, acute massive bleeding suddenly developed on the background of partial failure of the esophagogastroanastomosis.

Conclusion:

Patients who have EC diagnosed at stage I-II postoperatively - without complications. Complications in the early postoperative period were observed in patients with stage III and IV EC. These patients, as a rule, with malnutrition, exhausted, decompensated dysphagia and severe concomitant pathology. Based on our data, patients older than 40 years of age are advised to have an endoscopic examination annually for early-stage PC detection prior to clinical signs.

Picture:

 

Perioperative management of pleural empyema in thoracic surgery – experience of a single-center study

(Abstract ID: 731)

A. Gassa¹, V. Lagodka², H. Menghesha¹, G. Schlachtenberger¹, F. Dörr¹, M. Heldwein¹, K. Hekmat¹, T. Wahlers¹

¹Uniklinik Köln, Herzzentrum, Köln

²Universität zu Köln

Background:

Pleural empyema or parapneumonic effusion can result in a lethal state with organized pus, lung restriction and sepsis. In most cases, it is caused by pneumonia. According to EACTS consensus, we differentiate between three stages of pleural empyema. In a single-center study, we retrospectively analyzed patients’ characteristics undergoing video-assisted thoracic surgery (VATS) and thoracotomy in the management of pleural empyema.

Materials and methods:

We retrospectively collected data from 128 patients suffering from pleural empyema undergoing either VATS or thoracotomy from 2010 to May 2019. Both surgical treatments were compared considering perioperative management such as antibiotic therapy, post surgical rinsing or recurrence rate. The study population was evaluated for comorbid conditions, bacteriological culture, length of hospital stay, morbidity and mortality.

Results:

Patients were included when surgery was performed to treat empyema. Hemothorax and pleural effusion not classified as empyema were excluded. A total of 128 patients were retrospectively included into analysis selected from 2,236 patients encoded as pleural effusion undergoing thoracic surgery. 111 patients (86.7%) underwent VATS and 17 patients (13.3%) underwent thoracotomy. Four patients (3.7%) had pleural empyema at stage I, 25 patients (23.4%) at stage II and 75 patients (70.1%) at stage III. In 3.7% cases a conversion to thoracotomy was necessary. A decortication was performed in 82.2% (n = 106) of patients. We found that in 55.1% of cases, germs were detectable in the pleural cavity. In 49 cases (44.5%), postoperative rinsing was performed with Lavasept or Vancomyicin. Antibiotic therapy was performed in 85.7% of all patients.

Conclusion:

Surgical therapy of empyema is a feasible and safe method. In case of chronic empyema, a decortication is indicated to ensure full expansion of the lung. In these cases, an open thoracotomy can be indicated. To prepare patients for surgery, a preoperative rinsing with Lavasept in the pleural cavity by a drain can help facilitate the decortication. There are no randomized trials comparing preoperative and postoperative rinsing.

Effectiveness of VATS – Bullectomy with Partial Pleurectomy in the Management of Primary and Secondary Spontaneous Pneumothorax

(Abstract ID: 1018)

S. N. Fung¹, A. Rehders¹, L. Dizdar¹, H. Ashmawy¹, A. Schauer¹, W. T. Knoefel¹

¹Universitätsklinikum Düsseldorf

Background:

Spontaneous pneumothorax (SP) describes the presence of air without preceding trauma within the pleural space. Depending on the presence of underlying pulmonary disease, spontaneous pneumothorax can be classified either as primary spontaneous pneumothorax (PSP) or secondary spontaneous pneumothorax (SSP). The latter affects patients with pre-existing pulmonary disease, commonly those with chronic obstructive pulmonary disease. Depending on the patients clinical condition and type of SP, emergency therapies vary from simple aspiration, chest tube drainage or observation. Surgical treatment for pneumothorax is mostly considered in case of complications such as persisting air leakage or at recurrence. Common surgical techniques such as video-assisted thoracoscopic surgery (VATS), open thoracotomy and pleurodesis either mechanical or chemical have been reported to affect recurrence rate and length of patient hospitalization. In this study we evaluated underlying clinical features and clinical outcome of patients treated either by chest tube placement (CT) or VATS-bullectomy with partial pleurectomy (VBPP) in our hospital

Materials and methods:

We retrospectively evaluated clinical course and outcome of all patients with SP in a ten years period (january 2009- march 2019) treated at our institution. Patients were stratified according to the underlying condition (PSP vs. SSP). For follow-up, all patients were contacted and assessed with a questionnaire. Primary end point was recurrence free survival. All data were statistically analysed using SPSS.

Results:

A total of 106 patients were included in this study. 71(67%) patients presented with PSP and 35(33%) with SSP. 46.5% of the patients with PSP were treated by chest tube placement (CT), whereas 53.5% subsequently underwent VBPP. PSP patients treated by CT had a recurrence rate of 69.7%. Patients who underwent VBPP recurred only in 13.2%. This difference was highly significant at univariate analysis (P<0.001). This difference was also confirmed at multivariate analysis including potential prognostic parameters (HR 0.063; 95% CI: 0.018 - 0.217; p< 0.001). Likewise, 23(65.7%) patients with SSP were treated with CT and 12 (34.3%) by VBPP. In this subgroup, patients treated with CT recurred in 9/23 (39.10%) and patients who underwent surgery (VBPP) suffered from recurrence in 2/12 (16.60%). However this difference did not reach statistical significance at univariate analysis.Furthermore, we analysed the impact of potential risk factors for SP recurrence. However, there was no correlation between nicotine abuse, gender, side of pneumothorax, Body-Mass-Index (BMI), size of the pneumothorax and age with recurrence rate. The only adverse prognostic parameter in the subgroup of SSP was male gender (HR 5.330; 95% CI: 1.124-25.279; p 0.020).

Conclusion:

Our data suggest that, patients with PSP treated only with CT placement have a significantly increased risk of recurrence. Therefore, liberal indication for surgery in these patients seems advisable. In contrast, patients with SSP should be operated on persistent air leak since long term of surgery is limited.