G. Burkhard Mackensen

Central nervous system injury associated with cardiac surgery

Millions of individuals with coronary artery or valvular heart disease have been given a new chance at life by heart surgery, but the potential for neurological injury is an Achilles heel. Technological advancements and innovations in surgical and anaesthetic technique have allowed us to offer surgical treatment to patients at the extremes of age and infirmity-the group at greatest risk for neurological injury. Neurocognitive dysfunction is a complication of cardiac surgery that can restrict the improved quality of life that patients usually experience after heart surgery. With a broader understanding of the frequency and effects of neurological injury from cardiac surgery and its implications for patients in both the short term and the long term, we should be able to give personalised treatments and thus preserve both their quantity and quality of life. We describe these issues and the controversies that merit continued investigation.

β2-Adrenergic and Several Other G Protein–Coupled Receptors in Human Atrial Membranes Activate Both Gs and Gi

Cardiac G protein–coupled receptors that couple to G&agr;s and stimulate cAMP formation (eg, &bgr;-adrenergic, histamine, serotonin, and glucagon receptors) play a key role in cardiac inotropy. Recent studies in rodent cardiac myocytes and transfected cells have revealed that one of these receptors, the &bgr;2-adrenergic receptor (AR), also couples to the inhibitory G protein G&agr;i (activation of which inhibits cAMP formation). If &bgr;2ARs could be shown to couple to G&agr;i in the human heart, it would have important ramifications, because levels of G&agr;i increase with age and in failing human heart. Therefore, we investigated whether &bgr;2ARs in the human heart activate G&agr;i. By photoaffinity labeling human atrial membranes with [32P]azidoanilido-GTP, followed by immunoprecipitation with antibodies specific for G&agr;i, we found that G&agr;i is activated by stimulation of &bgr;2ARs but not of &bgr;1ARs. In addition, we found that other G&agr;s-coupled receptors also couple to G&agr;i, including histamine, serotonin, and glucagon. When coupling of these receptors to G&agr;i is disrupted by pertussis toxin, their ability to stimulate adenylyl cyclase is enhanced. These data provide the first evidence that &bgr;2AR and many other G&agr;s-coupled receptors in human atrium also couple to G&agr;i and that abolishing the coupling of these receptors to G&agr;i increases the receptor-mediated adenylyl cyclase activity.

A comparison of strain-related susceptibility in two murine recovery models of global cerebral ischemia

Genetically engineered mice are increasingly important in stroke research. The strains on which these constructs are built are known to have inherent differential sensitivities to ischemic insults. This has been largely attributed to differences in vascular anatomy. This study compared the outcome from forebrain ischemia in two common murine background strains using two different types of ischemic insult. C57Bl/6 and SV129 mice were subjected to two vessel (bilateral carotid) occlusion (2VO) or 2VO plus systemic hypotension (2VO+Hypo; mean arterial pressure=30+/-2 mmHg) for 10-20 min. Ventilation and pericranial temperature were controlled. Cerebral blood flow (CBF) was determined by 14C-iodoantipyrine autoradiography. Histologic damage in forebrain structures was measured 3 days post-ischemia. During 2VO+Hypo, the EEG became isoelectric in all animals. During 2VO alone, EEG isoelectricity occurred in 73% of C57Bl/6 and 50% of SV129 mice. Forebrain CBF was reduced to a similar extent in both strains. Greater CBF variability was seen with 2VO alone versus 2VO+Hypo. CBF was less in the 2VO+Hypo model. SV129 mice had wider posterior communicating but smaller basilar artery diameters. With or without hypotension, SV129 mice had markedly less severe histologic damage than C57Bl/6 mice. A time-dependent increase in histologic damage was demonstrated in the 2VO+Hypo model but not with 2VO alone. The 2VO and 2VO+Hypo models produced similar magnitudes of histologic injury in C57Bl/6 mice subjected to 10-min ischemia. SV129 mice were resistant to ischemia in either model. The 2VO+Hypo model produced a more uniform severity of ischemia as defined by CBF and EEG examination. Despite this, the murine strain had a substantially greater impact on histologic outcome than did cerebrovascular anatomy or the type of model used to produce the ischemic insult.

American College of Chest Physicians Consensus Statement on the Use of Topical Anesthesia, Analgesia, and Sedation During Flexible Bronchoscopy in Adult Patients

BACKGROUND Optimal performance of bronchoscopy requires patients comfort, physicians ease of execution, and minimal risk. There is currently a wide variation in the use of topical anesthesia, analgesia, and sedation during bronchoscopy. METHODS A panel of experts was convened by the American College of Chest Physicians Interventional/Chest Diagnostic Network. A literature search was conducted on MEDLINE from 1969 to 2009, and consensus was reached by the panel members after a comprehensive review of the data. Randomized controlled trials and prospective studies were given highest priority in building the consensus. RESULTS In the absence of contraindications, topical anesthesia, analgesia, and sedation are suggested in all patients undergoing bronchoscopy because of enhanced patient tolerance and satisfaction. Robust data suggest that anticholinergic agents, when administered prebronchoscopy, do not produce a clinically meaningful effect, and their use is discouraged. Lidocaine is the preferred topical anesthetic for bronchoscopy, given its short half life and wide margin of safety. The use of a combination of benzodiazepines and opiates is suggested because of their synergistic effects on patient tolerance during the procedure and the added antitussive properties of opioids. Propofol is an effective agent for sedation in bronchoscopy and can achieve similar sedation, amnesia, and patient tolerance when compared with the combined administration of benzodiazepines and opiates. CONCLUSIONS We suggest that all physicians performing bronchoscopy consider using topical anesthesia, analgesic and sedative agents, when feasible. The existing body of literature supports the safety and effectiveness of this approach when the proper agents are used in an appropriately selected patient population.

Mice Overexpressing Extracellular Superoxide Dismutase Have Increased Resistance to Global Cerebral Ischemia

Transgenic mice, which exhibit a fivefold increase in brain parenchymal extracellular superoxide dismutase (EC-SOD) activity, were used to investigate the role of EC-SOD in global ischemic brain injury. Halothane-anesthetized normothermic wild-type (n = 22) and transgenic (n = 20) mice underwent 10 min of near-complete forebrain ischemia induced by bilateral carotid artery occlusion and systemic hypotension (mean arterial pressure = 30 mm Hg). After 3 days of recovery, the brains were histologically examined. Other mice underwent autoradiographic determination of regional CBF 10 min prior to, during, and 30 min after forebrain ischemia. Histologic injury in the cortex and caudoputamen was minimal in both groups. The percentage of dead hippocampal CA1 neurons was reduced in the EC-SOD transgenic group (wild type = 44 +/- 28%; EC-SOD transgenic = 23 +/- 21%, mean +/- SD, P = 0.015). CBF was similar between groups prior to ischemia. The intraischemic blood flow was severely reduced in forebrain structures and was similar between groups. Blood flow at 30 min postischemia had recovered to 50-60% of baseline values in both groups. These results indicate that EC-SOD can play an important role in defining the magnitude of selective neuronal necrosis resulting from near-complete forebrain ischemia. This implicates involvement of extracellular superoxide anions in the pathologic response to global cerebral ischemia.

Cardiopulmonary Bypass Induces Neurologic and Neurocognitive Dysfunction in the Rat

Background Neurocognitive dysfunction is a common complication of cardiac surgery using cardiopulmonary bypass (CPB). Elucidating injury mechanisms and developing neuroprotective strategies have been hampered by the lack of a suitable long-term recovery model of CPB. The purpose of this study was to investigate neurologic and neurocognitive outcome after CPB in a recovery model of CPB in the rat. Methods Fasted rats (n = 10) were subjected to 60 min of normothermic (37.5°C) nonpulsatile CPB using a roller pump and a membrane oxygenator. Sham-operated controls (n = 10) were not subjected to CPB. Neurologic outcome was assessed on days 1, 3, and 12 after CPB using standardized functional testing. Neurocognitive outcome, defined as the time (or latency) to finding a submerged platform in a Morris water maze (an indicator of visual–spatial learning and memory), was evaluated daily from post-CPB days 3–12. Histologic injury in the hippocampus was also evaluated. Results Neurologic outcome was worse in the CPB versus the sham-operated controls at all three measurement intervals (P < 0.001). The CPB group also had longer water maze latencies compared with the sham-operated controls (P = 0.004), indicating significant neurocognitive dysfunction after CPB. No difference in histologic injury between groups was observed. Conclusions CPB caused both neurologic and neurocognitive impairment in a rodent recovery model. This model could potentially facilitate the investigation of CPB-related injury mechanisms and possible neuroprotective interventions.

Effects of extreme hemodilution during cardiac surgery on cognitive function in the elderly

Background:Strategies for neuroprotection including hypothermia and hemodilution have been routinely practiced since the inception of cardiopulmonary bypass. Yet postoperative neurocognitive deficits that diminish the quality of life of cardiac surgery patients are frequent. Because there is uncertainty regarding the impact of hemodilution on perioperative organ function, the authors hypothesized that extreme hemodilution during cardiac surgery would increase the frequency and severity of postoperative neurocognitive deficits. Methods:Patients undergoing coronary artery bypass grafting surgery were randomly assigned to either moderate hemodilution (hematocrit on cardiopulmonary bypass ≥27%) or profound hemodilution (hematocrit on cardiopulmonary bypass of 15–18%). Cognitive function was measured preoperatively and 6 weeks postoperatively. The effect of hemodilution on postoperative cognition was tested using multivariable modeling accounting for age, years of education, and baseline levels of cognition. Results:After randomization of 108 patients, the trial was terminated by the Data Safety and Monitoring Board due to the significant occurrence of adverse events, which primarily involved pulmonary complications in the moderate hemodilution group. Multivariable analysis revealed an interaction between hemodilution and age wherein older patients in the profound hemodilution group experienced greater neurocognitive decline (P = 0.03). Conclusions:In this prospective, randomized study of hemodilution during cardiac surgery with cardiopulmonary bypass in adults, the authors report an early termination of the study because of an increase in adverse events. They also observed greater neurocognitive impairment among older patients receiving extreme hemodilution.

Impact of Tricuspid Valve Regurgitation in Patients Treated With Implantable Left Ventricular Assist Devices

BACKGROUND The progression of tricuspid valve regurgitation (TR) and the impact of preoperative TR on postoperative outcomes in patients having left ventricular assist device (LVAD) implantation has not been studied. METHODS One hundred seventy-six consecutive implantable LVAD procedures were retrospectively reviewed. A total of 137 patients comprised the final study group with complete preimplant characteristics, before and after echocardiogram assessment of TR, and outcomes data. Patients were divided into two groups: insignificant TR (iTR) consisting of those with preimplant TR grades of none, trace, and mild; and significant TR (sTR) consisting of those with moderate and severe TR grades. RESULTS Relative to patients with iTR, patients with sTR were younger (53.6±12.8 versus 58.4±10.0 years, p=0.02) and more commonly had nonischemic cardiomyopathies (69% versus 38%, p<0.001). The preimplant incidence of iTR and sTR was 51% and 49%. Immediately after the LVAD implant procedure, TR did not significantly change. At late follow-up (156±272 days), 32% had moderate or severe TR. Also, 41% of the original sTR group persisted with moderate or severe TR. Relative to patients with iTR, patients with sTR required longer postimplant intravenous inotropic support (8.5 versus 5.0 days, p=0.02), more commonly required a temporary right ventricular assist device, and had a longer postimplant length of hospital stay (27.0 versus 20.0 days, p=0.03). There was also a trend toward decreased survival for sTR versus iTR (log rank=0.05). CONCLUSIONS Tricuspid regurgitation is not reduced immediately after LVAD implantation. Significant TR is associated with longer postimplant inotropic support and length of hospital stay.

Clinical Impact of Concomitant Tricuspid Valve Procedures During Left Ventricular Assist Device Implantation

BACKGROUND Almost 50% of patients referred for implantable left ventricular assist device (LVAD) have significant tricuspid regurgitation (TR). Preoperative TR is associated with negative outcomes but the clinical benefit of concomitant tricuspid valve procedures has not been extensively studied. METHODS One hundred fifteen patients, undergoing implantable LVADs, were identified as having significant TR by echocardiography prior to their surgical procedure. Patients underwent either LVAD alone (n = 81) versus LVAD plus concomitant tricuspid procedures (n = 34) (29 annuloplasty ring repairs and 5 bioprosthetic replacements.) Preoperative characteristics and hemodynamics, as well as TR severity and clinical outcomes were retrospectively determined from chart and database review and compared for the two groups. RESULTS Preoperative characteristics and hemodynamics were similar for the two groups. Postoperative TR was markedly reduced for the group undergoing concomitant procedures versus LVAD alone. A temporary right ventricular assist device was required for only one of the 34 cases in which concomitant tricuspid procedures were performed; for patients undergoing LVAD alone, 8 of 81 required right ventricular assist devices. Mean duration of postoperative inotrope utilization was increased for the LVAD alone group versus the group with concomitant tricuspid procedures (10.0 vs 8.0 days, respectively, p = 0.04). The incidence of postoperative renal dysfunction was increased for the LVAD alone group (39%) versus concomitant procedures (21%) (p = 0.05). The LVAD alone group also had a greater mean postimplant length of hospitalization versus the concomitant procedures group (26.0 vs 19.0 days, p = 0.02). Finally, there was a trend toward improved survival for the group with concomitant tricuspid procedures versus LVAD alone. CONCLUSIONS For patients with significant TR undergoing implantable LVAD procedures, concomitant tricuspid procedures are associated with improved early clinical outcomes.

Utility of a Simple Algorithm to Grade Diastolic Dysfunction and Predict Outcome After Coronary Artery Bypass Graft Surgery

BACKGROUND Inclusion of a measure of left ventricular diastolic dysfunction (LVDD) may improve risk prediction after cardiac surgery. Current LVDD grading guidelines rely on echocardiographic variables that are not always available or aligned to allow grading. We hypothesized that a simplified algorithm involving fewer variables would enable more patients to be assigned a LVDD grade compared with a comprehensive algorithm, and also be valid in identifying patients at risk of long-term major adverse cardiac events (MACE). METHODS Intraoperative transesophageal echocardiography data were gathered on 905 patients undergoing coronary artery bypass graft surgery, including flow and tissue Doppler-based measurements. Two algorithms were constructed to categorize LVDD: a comprehensive four-variable algorithm, A, was compared with a simplified version, B, with only two variables-transmitral early flow velocity and early mitral annular tissue velocity-for ease of grading and association with MACE. RESULTS Using algorithm A, only 563 patients (62%) could be graded, whereas 895 patients (99%) received a grade with algorithm B. Over the median follow-up period of 1,468 days, Cox modeling showed that LVDD was significantly associated with MACE when graded with algorithm B (p=0.013), but not algorithm A (p=0.79). Patients with the highest incidence of MACE could not be graded with algorithm A. CONCLUSIONS We found that an LVDD algorithm with fewer variables enabled grading of a significantly greater number of coronary artery bypass graft patients, and was valid, as evidenced by worsening grades being associated with MACE. This simplified algorithm could be extended to similar populations as a valid method of characterizing LVDD.