Landi M. Parish

A multi-feature and multi-channel univariate selection process for seizure prediction

OBJECTIVE To develop a prospective method for optimizing seizure prediction, given an array of implanted electrodes and a set of candidate quantitative features computed at each contact location. METHODS The method employs a genetic-based selection process, and then tunes a probabilistic neural network classifier to predict seizures within a 10 min prediction horizon. Initial seizure and interictal data were used for training, and the remaining IEEG data were used for testing. The method continues to train and learn over time. RESULTS Validation of these results over two workshop patients demonstrated a sensitivity of 100%, and 1.1 false positives per hour for Patient E, using a 2.4s block predictor, and a failure of the method on Patient B. CONCLUSIONS This study demonstrates a prospective, exploratory implementation of a seizure prediction method designed to adapt to individual patients with a wide variety of pre-ictal patterns, implanted electrodes and seizure types. Its current performance is limited likely by the small number of input channels and quantitative features employed in this study, and segmentation of the data set into training and testing sets rather than using all continuous data available. SIGNIFICANCE This technique theoretically has the potential to address the challenge presented by the heterogeneity of EEG patterns seen in medication-resistant epilepsy. A more comprehensive implementation utilizing all electrode sites, a broader feature library, and automated multi-feature fusion will be required to fully judge the methods potential for predicting seizures.

Aortic size in acute type A dissection: implications for preventive ascending aortic replacement §

OBJECTIVE Elective ascending aortic replacement is recommended to prevent acute type A aortic dissection when any segment of the proximal aorta is greater than 5.5 cm. However, little data exist that meticulously describe the size of the ascending aorta at multiple levels in patients who suffer acute type A dissections. We sought to definitively characterize the size distribution of the proximal aorta in this patient population. METHODS Preoperative transesophageal echocardiography was used to measure the diameter of the proximal aorta at the aortic annulus, in the sinus segment, at the sinotubular junction and in the ascending aorta in 177 non-Marfan patients with tricuspid aortic valves who presented to one institution over a 10-year period with an acute type A dissection. Predicted aortic diameters for each patient based on the individuals age, gender and body size were also calculated at all four aortic positions using previously published regression equations derived from a large cohort of normal patients. RESULTS Sixty patients were female (33.9%; aged 67+/-12 years) and 117 were male (66.1%; aged 60+/-17 years). Sixty-two percent of all patients had maximum aortic diameters less than 5.5 cm at time of dissection and 42% of patients had maximum aortic diameters less than 5.0 cm. Over 20% of all patients had maximal aortic dimensions of less than 4.5 cm. In women, 12% of the dissected aortas had a maximal dimension less than 4.0 cm. CONCLUSIONS The majority of patients with acute type A aortic dissection present with aortic diameters <5.5 cm and thus do not fall within current guidelines for elective ascending aortic replacement. Methods other than size measurement of the ascending aorta are needed to identify patients at risk for dissection. Aggressive medical management of patients with ascending aortic diameters over 4 cm is warranted. Preventative replacement of the ascending aorta at 4.5 cm should be considered especially at high volume aortic surgery centers and patients having cardiac surgery for other indications.

Mild Hypothermia to Limit Myocardial Ischemia-Reperfusion Injury: Importance of Timing

BACKGROUND Hypothermia during ischemia has been shown to reduce myocardial reperfusion injury. We sought to establish the cardioprotective effect of very mild total-body hypothermia (<or= 2.5 degrees C) and to determine whether the application of hypothermia at different points during the ischemia-reperfusion period influenced the degree of myocardial salvage. METHODS Rabbits were subjected to 30 minutes of myocardial ischemia followed by 3 hours of reperfusion. Twenty-five animals were maintained at normal temperature (39.5 degrees C) throughout the experiment (W-W-W group). All other animals were cooled to reduce left atrial temperature 2.0 degrees C to 2.5 degrees C. Eleven animals reached goal temperature before coronary occlusion (C-C-C group), in 14 animals cooling was initiated at coronary occlusion (W-C0-C group), in 8 animals cooling was initiated 15 minutes after coronary occlusion (W-C15-C group), in 5 animals cooling was initiated 25 minutes after coronary occlusion (W-C25-C group), and in 13 animals cooling was started concurrently with reperfusion (W-W-C group). Infarct size as a percentage of the risk area (I/AR) was determined by a double staining-planimetry technique. RESULTS Goal temperature was achieved before reperfusion in the C-C-C and W-C0-C groups but was not achieved until the reperfusion period in the other treatment groups. Infarct size was 59.0 +/- 1.2% in the W-W-W group and was reduced in all cooling groups (C-C-C = 30.4 +/- 4.9%; W-C0-C = 33.4 +/- 5.0%; W-C15-C = 42.4 +/- 1.4%; W-C25-C = 44.1 +/- 2.3%; W-W-C = 50.5 +/- 4.1%). The temperature at reperfusion correlated most strongly with infarct size (r = 0.72, p < 1 x 10(-12)). CONCLUSIONS Very mild hypothermia affords a significant cardioprotective effect. Temperature at the time of reperfusion most strongly correlates with the degree of myocardial salvage.

Very Mild Hypothermia During Ischemia and Reperfusion Improves Postinfarction Ventricular Remodeling

BACKGROUND Mild hypothermia (< 4 degrees C) improves myocardial salvage after infarct reperfusion in animals and in early clinical studies. In this experiment the effect of mild hypothermia during ischemia and early reperfusion on long-term postinfarction left ventricular (LV) remodeling was assessed in an ovine infarct model. METHODS In the initial phase of the experiment the effect of progressive degrees of hypothermia on infarct size was quantified. Thirty-eight male sheep were subjected to 1 hour of ischemia using a standardized anteroapical infarct followed by 3 hours of reperfusion. Temperature was maintained at either 39.5 degrees C (n = 11), 38.5 degrees C (n = 7), 37.5 degrees C (n = 7), 36.5 degrees C (n = 7), or 35.5 degrees C (n = 6) for the entire period of ischemia and reperfusion. The area at risk (AR) and infarct size as a percentage of AR (I/AR) were determined with a double staining and planimetry technique. In the second phase of the study, chronic post-infarction remodeling was assessed in animals with nonreperfused infarcts (n = 6), 1 hour of ischemia followed by reperfusion at 39.5 degrees C (n = 6) and 1 hour of ischemia followed by reperfusion at 37.5 degrees C (n = 6). Remodeling was determined at 8 weeks after infarction using echocardiography. RESULTS The I/AR in the 39.5 degrees C, 38.5 degrees C, 37.5 degrees C, 36.5 degrees C, and the 35.5 degrees C groups was 71.8 +/- 3.0%, 63.1 +/- 1.9%, 49.4 +/- 1.4%, 38.7 +/- 1.4%, and 21.7 +/- 2.2%, respectively (p < 0.05 between all groups). In the chronic study LV end systolic volume at 8 weeks after infarction was 81 +/- 8 mL in the nonreperfused group, 57 +/- 4 mL in the 39.5 degrees C reperfusion group, and 41 +/- 3 mL in the 37.5 degrees C reperfusion group (p < 0.05 for between group differences). CONCLUSIONS Subtle degrees of hypothermia can significantly improve immediate myocardial salvage and long-term LV remodeling after infarct reperfusion.

Regional Heterogeneity of Myocardial Reperfusion Injury: Effect of Mild Hypothermia

BACKGROUND Mild hypothermia confers a myocardial protective effect that may make it a useful adjunct to reperfusion therapy for myocardial infarction (MI). The effect of temperature on the extent and distribution of myocardial reperfusion injury in a collateral deficient ovine model was studied. METHODS Topical cooling maintained left atrial temperature at 39.5 degrees C (n = 8), 38.5 degrees C (n = 5), 37.5 degrees C (n = 6), 36.5 degrees C (n = 6), or 35.5 degrees C (n = 5) in sheep prior to 1 hour of coronary occlusion to produce an anteroapical myocardial risk area (AR) followed by 3 hours of reperfusion. A dual staining and planimetry technique was used to assess infarct size as a percentage of the AR in 3 myocardial short axis slices that included the entire AR (slice 1= most apical; slice 3= most basal). The subendocardial, midmyocardial, and subepicardial extent in short axis of the infarct was also assessed in each slice. Microspheres assessed transmural blood flow. RESULTS At 39.5 degrees C there was a long-axis gradient in myocardial injury that was most severe at the apex and lessened toward the base. The midmyocardial region was most susceptible to injury at all long axis levels. Temperature reduction (as little as 1 degrees C) was associated with improved salvage that was most pronounced in the apical subendocardium and least in the basilar midmyocardium. Reperfusion at 39.5 degrees C resulted in severe transmural microvascular injury (no-reflow) that was completely obviated at temperatures below 38.5 degrees C. CONCLUSIONS Myocardial reperfusion injury varies over the long and short LV axes. Mild hypothermia preferentially improves myocardial salvage at the LV apex. Small temperature changes can dramatically affect microvascular integrity.

Endocarditis with massive aortic root abscess and atrioventricular septal destruction

Endocarditis involving the aortic root and intervalvular fibrous skeleton presents a reconstructive dilemma. We report a case of endocarditis involving the aortic root and tricuspid valve with extensive destruction of the atrioventricular septum. Debridement necessitated resection of the aortic root, aortic valve, tricuspid valve, and a large portion of atrioventricular septum, leaving the right atrium, right ventricle, left ventricle and aorta in open communication. Reconstruction was accomplished by separating the left and right hearts with a Dacron patch, tricuspid valve replacement, and aortic root replacement. Proper planar localization of the aortic root was necessary to avoid left ventricular outflow obstruction and coronary torsion.