Panton-Valentine leukocidin expressing Staphylococcus aureus pneumonia managed with extracorporeal membrane oxygenation: experience and outcome.

OBJECTIVE: Panton-Valentine leukocidin expressing Staphylococcus aureus pneumonia, an infection that affects predominantly young people, has a mortality rate of > 70% despite aggressive conventional management. Little information is available on the management of patients with Panton-Valentine leukocidin expressing S. aureus pneumonia with extracorporeal membrane oxygenation support. As a large extracorporeal membrane oxygenation center, we reviewed our experience and outcomes with Panton-Valentine Leukocidin expressing S. aureus pneumonia. DATA SOURCES: Locally held register of all extracorporeal membrane oxygenation patients at Glenfield Hospital. STUDY SELECTION: Retrospective study including all patients with sputum-positive Panton-Valentine leukocidin expressing S. aureus pneumonia managed with extracorporeal membrane oxygenation support at a single extracorporeal membrane oxygenation center. DATA SYNTHESIS: On review of our database held from September 1989 until date, there were four patients with sputum-confirmed Panton-Valentine leukocidin expressing S. aureus pneumonia managed with extracorporeal membrane oxygenation. Refractory hypoxemia and/or uncompensated hypercapnia despite optimal conventional management were the indications for extracorporeal membrane oxygenation. After varying periods on extracorporeal membrane oxygenation with appropriate antibiotic and ancillary care, all four patients were discharged home. CONCLUSIONS: Panton-Valentine leukocidin expressing S. aureus pneumonia can cause severe, necrotizing pneumonia associated with acute respiratory distress syndrome, which can be particularly challenging to manage. Extracorporeal membrane oxygenation support permits low pressure lung ventilation, avoiding barotrauma to lungs made friable by Panton-Valentine leukocidin expressing S. aureus infection. Although this is a small number of patients, the results are encouraging.

Identification of Streptococcus gallolyticus subsp. macedonicus as the etiological agent in a case of culture-negative multivalve infective endocarditis by 16S rDNA PCR analysis of resected valvular tissue.

Today, PCR using broad-range primers is being used increasingly to detect pathogens from resected heart valves. Herein is described the first case of multivalve infective endocarditis where 16S rDNA PCR was used to detect a single pathogen from two affected valves in a 61-year-old man. Triple heart valve replacement was required despite six weeks of appropriate antimicrobial therapy. The organism was confirmed as Streptococcus gallolyticus subsp. macedonicus, a member of the 'S. equinus/S. bovis' complex. To date, only one report has been made of human infection due to this organism. This may be due to the limited resolution of the routine diagnostic methods used and/or as a consequence of the complex nomenclature associated with this group of organisms.

Factors influencing the outcome of paediatric cardiac surgical patients during extracorporeal circulatory support.

BACKGROUND: Veno-arterial extracorporeal membrane oxygenation (ECMO) is a common modality of circulatory assist device used in children. We assessed the outcome of children who had ECMO following repair of congenital cardiac defects (CCD) and identified the risk factors associated with hospital mortality. METHODS: From April 1990 to December 2003, 53 patients required ECMO following surgical correction of CCD. Retrospectively collected data was analyzed with univariate and multivariate logistic regression analysis. RESULTS: Median age and weight of the patients were 150 days and 5.4 kgs respectively. The indications for ECMO were low cardiac output in 16, failure to wean cardiopulmonary bypass in 13, cardiac arrest in 10 and cardio-respiratory failure in 14 patients. The mean duration of ECMO was 143 hours. Weaning off from ECMO was successful in 66% and of these 83% were survival to hospital-discharge. 37.7% of patients were alive for the mean follow-up period of 75 months. On univariate analysis, arrhythmias, ECMO duration >168 hours, bleeding complications, renal replacement therapy on ECMO, arrhythmias and cardiac arrest after ECMO were associated with hospital mortality.On multivariate analysis, abnormal neurology, bleeding complications and arrhythmias after ECMO were associated with hospital mortality. Extra and intra-thoracic cannulations were used in 79% and 21% of patients respectively and extra-thoracic cannulation had significantly less bleeding complications (p = 0.031). CONCLUSION: ECMO provides an effective circulatory support following surgical repair of CCD in children. Extra-thoracic cannulation is associated with less bleeding complications. Abnormal neurology, bleeding complications on ECMO and arrhythmias after ECMO are poor prognostic indicators for hospital survival.

Venovenous extracorporeal membrane oxygenation for respiratory failure in inotrope dependent neonates.

It is often stated that venovenous extracorporeal membrane oxygenation (VV ECMO) should not be used in inotrope dependent patients. It is our practice to use VV ECMO in most patients with respiratory failure even though many of these patients are receiving significant doses of inotropes. Our objective was to review the mode of ECMO in relation to precannulation doses of inotropes administered to neonates treated with ECMO for respiratory failure. Forty-three consecutive case notes were reviewed. Data were collected for basic demographic and ECMO parameters. Inotropic doses were converted to a single score for ease of comparison, with one point equivalent to 1 microg/kg/min dopamine. Forty-three neonates were studied; 37(86%) were treated with VV ECMO and 6 (14%) were treated with VA ECMO. Significant pre-ECMO inotropic support (score > 10) was present in 30 (70%) of the 43 cases. Of these patients, 26 were treated via VV ECMO with a survival rate of 84%, while 4 were treated with VA ECMO with a survival of 75%. Inotrope scores fell to nonsignificant levels (< 10) within 24 hours, regardless of ECMO mode. Mean arterial blood pressure remained above precannulation levels in both groups. VV ECMO allows safe treatment of neonatal respiratory failure in the presence of significant inotropic support. We recommend VV ECMO for neonatal respiratory failure in all cases except where double lumen cannulation is impossible or when septic shock is refractory to inotropic support (i.e., mean blood pressure < 35 mm Hg despite inotrope score of > 100).

Does the type of prosthesis influence early left ventricular mass regression after aortic valve replacement? Assessment with magnetic resonance imaging.

BACKGROUND: Debate exists regarding selection of the prosthesis type most likely to maximize early left ventricular (LV) mass regression after aortic valve replacement (AVR) for stenotic valvular disease. The aim of this study was to compare the degree of LV mass regression measured by MRI 6 months after prospectively randomized valve implantation for two biological prostheses, stented and stentless, and for two mechanical valves, tilting disc and bileaflet. METHODS: Thirty-nine consecutive patients with predominant aortic stenosis accepted for elective AVR were studied. Twenty patients requiring a tissue prosthesis were randomly assigned to receive either a Freestyle or Mosaic valve. The remaining 19 patients in whom mechanical prosthesis was indicated were randomly assigned to receive either an Ultracor or an ATS valve. RESULTS: There was no difference in valve size implanted between the compared groups. LV mass measurements were performed with MRI (1.5-T Vision, Siemens, Germany) immediately before and 6 months after surgery. All valve types produced significant postoperative reduction in LV mass compared with preoperative values (P <.01). Percent change in LV mass regression was similar between the two porcine valve types, Mosaic (24.4% +/- 11.1%) and Freestyle (21.1% +/- 16.7%), and between the two mechanical valve designs, Ultracor (19.3% +/- 9.5%) and ATS (26.3% +/- 10.8%), respectively. CONCLUSIONS: Significant LV remodeling occurs early after AVR for aortic stenosis. The degree of regression in LV mass is independent of prosthesis type implanted.