Kevin Pilarczyk

Preoperative intra-aortic counterpulsation in high-risk patients undergoing cardiac surgery: a meta-analysis of randomized controlled trials

In contrast to the results of previous studies, recent randomized controlled trials (RCTs) failed to show a benefit of prophylactic aortic counterpulsation in high-risk patients undergoing cardiac surgery. The present analysis aims to redefine the effects of this treatment modality in the light of this new evidence. MEDLINE, EMBASE, CENTRAL/CCTR, Google Scholar and reference lists of relevant articles were searched for full-text articles of RCTs in English or German. Assessments for eligibility, relevance, study validity and data extraction were performed by two reviewers independently using prespecified criteria. The primary outcome was hospital mortality. A total of nine eligible RCTs with 1171 patients were identified: 577 patients were treated preoperatively with intra-aortic balloon pump (IABP) and 594 patients served as controls. The pooled odds ratio (OR) for hospital mortality (22 hospital deaths in the intervention arm, 54 in the control group) was 0.381 (95% CI 0.230-0.629; P < 0.001). The pooled analyses of five RCTs including only patients undergoing isolated on-pump coronary artery bypass grafting (n[IABP] = 348, n[control] = 347) also showed a statistically significant improvement in mortality for preoperative IABP implantation (fixed-effects model: OR 0.267, 95% CI 0.129-0.552, P < 0.001). The pooled OR for hospital mortality from two randomized off-pump trials was 0.556 (fixed-effects model, 95% CI 0.207-1.493, P = 0.226). Preoperative aortic counterpulsation was associated with a significant reduction in low cardiac output syndrome (LCOS) in the total population (fixed-effects model: OR 0.330, 95% CI 0.214-0.508, P < 0.001) as well as in the subgroup of CAGB patients (fixed-effects model: OR 0.113, 95% CI 0.056-0.226, P < 0.001), whereas there was no benefit in the off-pump population (fixed-effects model: OR 0.555, 95% CI 0.209-1.474, P = 0.238). Preoperative IABP implantation was associated with a reduction of intensive care unit (ICU) stay in all investigated populations with a greater effect in the total population [fixed-effects model: standard mean difference (SMD) -0.931 ± 0.198, P < 0.001] as well as in the subgroup of CAGB patients (fixed-effects model: SMD -1.240 ± 0.156, P < 0.001), compared with the off-pump group (fixed-effects model: SMD -0.723 ± 0.128, P < 0.001). Despite contradictory results from recent trials, the present study confirms the findings of previous meta-analyses that prophylactic aortic counterpulsation reduces hospital mortality, incidence of LCOS and ICU requirement in high-risk patients undergoing on-pump cardiac surgery. However, owing to small sample sizes and the lack of a clear-cut definition of high-risk patients, an adequately powered, prospective RCT is necessary to find a definite answer to the question, if certain groups of patients undergoing cardiac surgery benefit from a prophylactic IABP insertion.

Use of donors who have suffered cardiopulmonary arrest and resuscitation in lung transplantation

OBJECTIVES Shortage of donors is one of the major limitations in lung transplantation (LuTX) and an aggressive expansion of criteria for donor selection has been proposed. This study evaluates the outcome of recipients of pulmonary grafts coming from resuscitated donors when compared with recipients of non-resuscitated donors. METHODS We retrospectively analyzed the donor and recipient charts of all double LuTX performed at our institution between 2000 and 2008 with regard to the performance of donor-cardiopulmonary resuscitation (CPR). RESULTS Out of 186 eligible transplants, 22 patients (11.8%) received lungs from donors who have suffered cardiac arrest (CA) and subsequent CPR. Mean duration of CPR was 15.2 ± 11.3 min. Terminal laboratory profiles of CPR donors and non-CPR donors were similar as were ventilation time and paO(2)/FiO(2) ratio before organ harvesting or chest X-ray. CPR-donor status did not affect the following indices of graft function: length of postoperative ventilation, paO(2)/FiO(2) ratio up to 48 h and lung function up to 60 months. Length of intensive care and hospital stay, need for inotropic support and 30-day mortality were not significantly different for the transplantation of CPR or no-CPR donor lungs. One- and 3-year survival rates were comparable as well with 84.4% and 66.3% for CPR donors versus 88.5% and 69.8% no-CPR donors. CONCLUSIONS This study indicates that transplantation of lungs from resuscitated donors may not affect outcome after LuTX. Therefore, donor history of CA should not automatically preclude LuTX.

Tracheostomy After Cardiac Surgery With Median Sternotomy and Risk of Deep Sternal Wound Infections: Is It a Matter of Timing?

OBJECTIVE To assess the impact of timing of percutaneous dilatational tracheotomy (PDT) on incidence of deep sternal wound infections (DSWI) after cardiac surgery with median sternotomy. DESIGN Retrospective study between 2003 and 2013. SETTING Single-center university hospital. PARTICIPANTS Eight hundred seventy-nine patients after cardiac surgery with extracorporeal circulation and median sternotomy. INTERVENTIONS PDT using the Ciaglia-technique with direct bronchoscopic guidance. MEASUREMENT AND MAIN RESULTS Mean time from surgery and (re)intubation to PDT was 6.7±9.9 and 3.8±3.3 days, respectively. Incidence of DSWI was 3.9% (34/879). The incidence of DSWI was comparable between patients with PDT performed before postoperative day (POD) 10 and those with PDT after POD 10 (29/755 [3.8%] v 5/124 [4.0%], p = n.s.). However, the authors observed an association of timing of PDT and DSWI: The incidence of DSWI was significantly higher in patients with PDT performed≤POD 1 compared to those with PDT after POD 2 (12/184 [6.52%] v 22/695 [3.16%], p = 0.046). In multivariate analysis, obesity, use of bilateral internal mammary arteries, ICU stay>30 days and PDT<48 hours after surgery (OR 3.519, 95% CI 1.242-9.976, p = 0.0018) were independent predictors of DSWI. In 15/34 patients (44.1%), similarity of microorganisms between sternotomy site and tracheal cultures was observed, indicating a possible cross-contamination. CONCLUSIONS PDT within the first 10 postoperative days after cardiac surgery with median sternotomy can be performed safely without an increased risk of DSWI. In contrast, very early PDT within 48 hours after surgery is associated with an increased risk of mediastinitis and should, therefore, be avoided.

Dissectometer - a new device for tensile strength testing of the vascular wall

Abstract Objectives: The underlying mechanisms leading to aortic dissection are poorly understood. The present paper describes a unique device for aortic wall testing during open heart surgery which allows to directly predict the risk of aortic dissection. Material and methods: Samples of the aortic wall were taken during surgery of the aortic valve and ascending aorta in 35 patients. Disruption of the samples was performed by using the dissectometer (patent pending). The output signal of the actuator was visualized by a “tensile strain curve”. Seven destinctive curve parameters were compared by parametric or nonparametric evaluation with regard to the results of the normality test. Results: Variations of all seven parameters in each patient differed not significantly; however, significant differences could be observed between the 35 patients. All parameters proved to be consistent and reproducible intra-individually. Histological analysis showed the integrity of the surface and localization of dissection in the external layer of tunica media. Conclusions: The dissectometer proved to be a suitable device to predict aortic wall consistence intraoperatively. Statistical analysis confirmed that a single sample of aorta is sufficient for assessment of aortic wall quality. A valid statistical discrimination between “healthy” and pathological aortas could only be assumed by generation of a large database of patients.

Ascending Aortic Wall Cohesion: Comparison of Bicuspid and Tricuspid Valves

Objectives. Bicuspid aortic valve (AV) represents the most common form of congenital AV malformation, which is frequently associated with pathologies of the ascending aorta. We compared the mechanical properties of the aortic wall between patients with bicuspid and tricuspid AV using a new custom-made device mimicking transversal aortic wall shear stress. Methods. Between 03/2010 and 07/2011, 190 consecutive patients undergoing open aortic valve replacement at our institution were prospectively enrolled, presenting either with a bicuspid (group 1, n = 44) or a tricuspid (group 2, n = 146) AV. Aortic wall specimen were examined with the “dissectometer” resulting in nine specific aortic-wall parameters derived from tensile strength curves (TSC). Results. Patients with a bicuspid AV showed significantly more calcified valves (43.2% versus 15.8%, P < 0.001), and a significantly thinner aortic wall (2.04 ± 0.42 mm versus 2.24 ± 0.41 mm, P = 0.008). Transesophageal echocardiography diameters (annulus, aortic sinuses, and sinotubular junction) were significantly larger in the bicuspid group (P = 0.003, P = 0.02, P = 0.01). We found no difference in the aortic wall cohesion between both groups as revealed by shear stress testing (P = 0.72, P = 0.40, P = 0.41). Conclusion. We observed no differences of TSC in patients presenting with tricuspid or bicuspid AVs. These results may allow us to assume that the morphology of the AV and the pathology of the ascending aorta are independent.

Is universal antifungal prophylaxis mandatory in adults after lung transplantation? A review and meta-analysis of observational studies.

Lung transplant (LTX) recipients are at high risk of invasive Aspergillus infections (IAI). However, no randomized‐controlled trials (RCT) or international guidelines on antifungal prophylaxis (AFP) in the LTX population exist.

Safety and Efficiency of Percutaneous Dilatational Tracheostomy With Direct Bronchoscopic Guidance for Thoracic Transplant Recipients

BACKGROUND: Percutaneous dilatational tracheostomy (PDT) is the standard airway access in critically ill patients who require prolonged mechanical ventilation. However, the literature lacks reports about the effectiveness and safety of this procedure in thoracic organ transplant recipients, who have increased risks of bleeding and infection. METHODS: We retrospectively reviewed the records of subjects who underwent thoracic organ transplantation at our institution between January 2004 and March 2011 followed by PDT (using the Ciaglia Blue Rhino technique with direct bronchoscopic guidance). RESULTS: From a total of 312 thoracic transplant recipients, we identified 93 (29.8%) subjects with PDT. Of these, 79 had undergone double lung transplant, 11 had undergone heart transplant, 2 had undergone combined heart-lung transplant, and 1 had undergone combined heart-kidney transplant. Mean age was 49.5 ± 11.2 y, and 58% of subjects were female. The mean time from intubation to PDT was 3.7 ± 3.4 d, and mean time from transplant to PDT was 12.6 ± 28.3 d. Thirty-two subjects (34.4%) underwent PDT after re-intubation. Thirty-nine subjects were receiving renal replacement therapy (41.9%), and 28 had a coagulopathy (30.1%). Moderate but not significant bleeding was observed in 3 subjects. There were no major complications during PDT procedures. Forty-five subjects (48.4%) could be weaned successfully from the ventilator and the tracheostoma could be removed. Forty-eight subjects (51.6%) died due to sepsis, multi-organ failure, or transplant failure. No procedure-related deaths were noted. There were no significant late complications. Among the 45 who survived their stay in the ICU, the functional and cosmetic outcomes of PDT were excellent. CONCLUSIONS: PDT can be safely performed on patients with acute respiratory failure after thoracic organ transplantation. Therefore, we recommend the use of this technique for prolonged airway management in these patients.

Comparison of ascending aortic cohesion between patients with bicuspid aortic valve stenosis and regurgitation

OBJECTIVES A bicuspid aortic valve (BAV) is commonly associated with aortic wall abnormalities, including dilatation of the ascending aorta and increased potential for aortic dissection. We compared the mechanical properties of the aortic wall of BAV patients with aortic valve stenosis (AS) and regurgitation (AR) using a dissectometer, a device mimicking transverse aortic wall shear stress. METHODS Between March 2010 and February 2013, 85 consecutive patients with bicuspid aortic valve undergoing open aortic valve replacement at our institution were prospectively enrolled, presenting either with stenosis (Group 1, n = 58) or regurgitation (Group 2, n = 27). Aortic wall cohesion measured by the dissectometer (Parameters P7, P8 and P9), aortic diameters measured by transoesophageal echocardiography (TOE) and thickness of the wall were compared. One patient presenting with the Marfan syndrome was excluded from the study. RESULTS Patients with aortic regurgitation were significantly younger (48.2 ± 15.8 vs 64.7 ± 10.7, P < 0.001), and had a significantly thicker aortic wall (2.30 ± 0.49 mm vs 2.06 ± 0.35 mm, P = 0.029). Transoesophageal echocardiography diameters (annulus, aortic sinuses and sinotubular junction) were significantly larger in the AR group (27.3 ± 3.6 vs 25.5 ± 2.4, P = 0.008; 41.1 ± 7.7 vs 36.7 ± 8.0, P = 0.011; 37.6 ± 9.7 vs 33.8 ± 9.1, P = 0.049). The ascending aortic diameter did not differ (43.2 ± 10.6 vs 40.3 ± 9.1, P = 0.292). Patients with AR had significantly worse aortic cohesion, as measured by shear stress testing (P7: 97.2 ± 45.0 vs 145.5 ± 84.9, P = 0.015; P8: 2.00 ± 0.65 vs 3.82 ± 1.56, P < 0.001; P9: 2.96 ± 0.82 vs 4.98 ± 1.80, P < 0.001) compared with those with AS. CONCLUSIONS We observed significantly worse aortic wall cohesion, a thicker aortic wall and a larger aortic root in patients presenting with bicuspid AR compared with patients with AS. These results suggest that bicuspid AR represents a different disease process with possible involvement of the ascending aorta, as demonstrated by dissectometer examination.

Does Traumatic Donor Cause of Death Influence Outcome after Lung Transplantation? A Single-Centre Analysis

Background Owing to the shortage of donor organs in lung transplantation (LuTX), liberalization of donor selection criteria has been proposed. However, some studies suggested that donor traumatic brain damage might influence posttransplantation allograft function. This article aimed to investigate the association of donor cause of death (DCD) and outcome after LuTX. Methods A retrospective analysis of 186 consecutive double LuTXs at our institution from January 2000 to December 2008 was performed. DCD was categorized into traumatic brain injury (TBI) and nontraumatic brain injury (NTBI). In addition, NTBI was sub classified as spontaneous intracerebral bleeding (B), hypoxic brain damage (H), and intracerebral neoplasia (N). Results DCD was classified as TBI in 50 patients (26.9%) and NTBI in 136 patients (73.1%): B in 112 patients (60.2%), H in 21 patients (11.3%), and N in 3 patients (1.6%). Young male donors predominated in group TBI (mean age 36.0 ± 14.5 vs. 42.8 ± 10.7, p < 0.01; 29 males in the TBI group [58.0%] vs. 48 males in the NTBI group [35.3%], p < 0.01). Groups of DCD did not differ significantly by recipient age or gender, recipient diagnosis, donor ventilation time, or paO2 /FiO2 before harvesting. TBI donors received significantly more blood (3.4 ± 3.8 vs. 1.8 ± 1.9, p = 0.03). A chest trauma was evident only in group T (n = 7 [3.7%] vs. 0 [0%], p < 0.001). Mode of donor death did not affect the following indices of graft function: length of postoperative ventilation, paO2 /FiO2 ratio up to 48 hours, and lung function up to 36 months. One‐ and three‐year survival was comparable with 84.4 and 70.4% for TBI donors versus 89.4% and 69.2% for NTBI donors. Five‐year survival tended to be lower in the TBI group but did not reach statistical significance (43.4 vs. 53.9%). Conclusion This study indicates that traumatic DCD does not affect outcome after LuTX. These results can be achieved with an ideal donor management combined with an individual case‐to‐case evaluation by an experienced LuTX surgeon.

Extracorporeal membrane oxygenation for acute respiratory distress syndrome in adults: an analysis of differences between survivors and non-survivors

Objectives: Over the last decade, extracorporeal membrane oxygenation (ECMO) has become a promising option for patients with severe acute respiratory distress syndrome (ARDS). In this single-center observational cohort study, data from a patient group with severe ARDS treated with ECMO was analyzed. Methods: Data from 46 patients [median age 54 years (18 to 72), male: 65.2%] were evaluated retrospectively between January 2009 and September 2015. Results: Diagnosis leading to ARDS was pneumonia in 63.1% of the patients. The median SOFA Score was 13 (10 to 19) and the median LIS was 3.5 (2.67 to 4). The median duration of ECMO support was 12 days (1 to 86). Twenty-eight patients (60.9%) were successfully weaned from ECMO and 22 patients survived (47.8%). Non-survivors needed significantly more frequent renal replacement therapy (37.5% vs. 18.2%; p<0.01) and transfusion of red blood cell concentrates [0.4 units (0.3 to 1.2) vs. 0.9 units (0.5 to 1.6); p<0.01] during ECMO support compared to patients who survived. Conclusion: This report suggests that ECMO currently allows treatment of severe ARDS with presumed improved survival. The incidence rate of acute kidney injury and transfusion are associated with adverse outcomes.