Cardiopulmonary bypass

Transcription

1 Cardiopulmonary bypass Saran Woods Stephen J Gray Abstract The success of cardiac surgery is the result of revolutionary thinking by those who were unafraid to take risks in the 1950s, when cardiopulmonary bypass was in its infancy. The development of the heartelung machine has moved a long way from the cumbersome screen oxygenator to today s modern disposable hollow-fibre units. Perfusionists are one part of a team of highly skilled professionals dedicated to delivering the best quality care. Perfusion science is going through a number of changes, many of which are focused on receiving recognition from the Health Professions Council. Hospitals can enact local policies enabling perfusionists under the supervision of consultants to administer drugs on bypass. Regulation of parameters on cardiopulmonary bypass remains controversial. Best practice is still evolving with regard to perfusion pressure, pump flow, temperature management and central nervous system monitoring. Most coronary artery bypass surgery is now performed at normothermia or mild hypothermia, making the argument for a-stat or ph-stat blood gas management less critical. Indeed, if the patient has intact cerebral autoregulation (not routinely tested preoperatively), neither pump flow nor pressure influences cerebral blood flow over that autoregulatory range. Keywords a-stat; central nervous system monitoring; drug administration; flow; normothermic; oxygenator; perfusionist; ph-stat; pressure; systemic vasculature resistance History The time line for the development of cardiac surgery is shown in Table 1. The history from the concept of extracorporeal circulation in 1813 by Le Gallois to the current trend of so-called mini-bypass is charted. This claims to reduce the deleterious effects of cardiopulmonary bypass (CPB) by minimizing the bloodeair interface. Blood pumps CPB diverts blood away from the heart (and lungs) and returns it to the systemic arterial system. Therefore, CPB must replace the function of the lungs (gas exchange) and heart (circulation). Advances in efficient atraumatic blood-pumping devices have been slow compared with those in oxygenator technology. Roller and centrifugal pumps are the most commonly used devices. Saran Woods ACPS LCCP BSc(Hons) is Senior Clinical Perfusionist at Cambridge Perfusion Services, UK, with special interests in Ventricular Assist Devices and key educational issues for perfusionists. Conflicts of interest: none declared. Stephen J Gray MBBS BSc FRCA is a Consultant Anaesthetist at Papworth Hospital NHS Trust, UK. His main interest is perioperative transoesophageal echocardiography. Conflicts of interest: none declared. Learning objectives After reading this article, you should: C understand the historical development of cardiopulmonary bypass and the key components in the bypass circuit C know the primary considerations in the way cardiopulmonary bypass is conducted C understand the importance of, and the controversy surrounding, the parameters used to control bypass. Roller pumps These are the most widely used type of pump. These pumps consist of rollers (usually two, 180 apart) positioned on the end of a rotating arm. Forward flow is induced by the rollers compressing tubing mounted in a U-shaped raceway. The flow rate is dependent on the diameter of the tubing, the diameter of the raceway and the rotation rate of the rollers. Roller pumps have the advantage of being simple to set up and prime; however, they are traumatic to blood. Centrifugal pumps Two types predominate: the first uses a nest of smooth plastic cones contained within plastic housing; the second makes use of a vaned impeller. The cones or impellers are magnetically coupled to an electric motor and, when rotated rapidly, impart kinetic energy to the blood, inducing forward flow. These devices are completely non-occlusive, and are dependent on preload and afterload. An electromagnetic flow probe must be attached to the arterial line to determine pump flow. These pumps are more complex, yet less traumatic to the blood elements (reduced haemolysis). Massive air embolism is less likely. The bubble oxygenator Lillehei, in collaboration with Dewall, developed the first bubble oxygenator (Figure 1a). These oxygenators are highly efficient, inexpensive and disposable. Venous blood is passed upwards in a vertical column through which oxygen is simultaneously bubbled. Eddies and vortexes are created whereby oxygen enters the blood and carbon dioxide is released. The column consists of either multiple vertical tubes or a fine mesh that acts as a spoiler to promote mixing of the gas and blood. At the top of the column, the gases and blood form a foam. Defoaming agents cause coalescence of the bubbles. On exiting the column, the arterialized blood passes through a filter and bubble trap (Figure 1b). The membrane oxygenator Membrane oxygenators are more physiological and extensively used. They mimic the pulmonary capillary by interposing a thin membrane between the blood and the gas. There are different types of membrane, including a flat sheet usually arranged as a fan-fold and hollow fibres. These fibres have an internal diameter of 100e200 mm. The total number of fibres ultimately determines the efficiency of the gas exchange. Gas flows on the inside of the fibre and blood flows to the outside, thus maximizing the surface area available for gas exchange. Membrane oxygenators have several advantages: they separate the blood ANAESTHESIA AND INTENSIVE CARE MEDICINE 10:9 416 Ó 2009 Elsevier Ltd. All rights reserved.

2 Key developments in extracorporeal support Date Activity 1813 Le Gallois first proposed the principle of extracorporeal circulation 1916 McLean discovered heparin 1934 De Bakey developed the roller pump 1937 Gibbon performed the world s first cardiopulmonary bypass e on a cat 1953 Gibbon used a film oxygenator with his heartelung machine to perform the first clinical cardiopulmonary bypass on a patient with an atrioseptal defect 1955 Lillehei successfully utilized cross-circulation 1956 Dewall developed the bubble oxygenator 1965 Bodell developed the membrane oxygenator 2003 Concept of the mini-bypass Table 1 and gas phases, create far fewer bubbles, give greater accuracy in blood gas control and make massive gas embolism almost impossible. On the downside, they are more expensive and technically more difficult to set up. Key considerations on the conduct of cardiopulmonary bypass Anticoagulation Adequate systemic anticoagulation is an absolute requirement for CPB. Unfractionated heparin at a dose of 300 iu/kg is used. Activated clotting time (ACT), measured using a portable device, determines the adequacy of heparinization. An ACT >400 seconds is required before going onto bypass. Cannulation strategy The ascending aorta (occasionally femoral artery) is cannulated first. At this stage, avoidance of hypertension (mean arterial pressure (MAP) <60 mmhg) reduces the risk of arterial dissection. Venous cannulation (right atrium or cavae) then follows. Figure 1 a DeWall with the original bubble oxygenator. b A modern hollow-fibre membrane counterpart. Courtesy of University Archives, University of Minnesota, Minneapolis. ANAESTHESIA AND INTENSIVE CARE MEDICINE 10:9 417 Ó 2009 Elsevier Ltd. All rights reserved.

3 CPB is initiated by the perfusionist releasing the clamp on the venous line, thereby diverting blood into the venous reservoir. Concurrently, the speed of the pump is increased to generate a flow of 2.2e2.4 l/min/m 2. When CPB is established, ventilation of the lungs is discontinued. Adequacy of cardiopulmonary bypass This remains an area of controversy. The exact parameters defining optimal CPB are yet to be determined. The perfusionist maintains a MAP between 50 and 70 mmhg, central venous pressure (CVP) between 0 and 5 mmhg, mixed venous oxygenation >65% and a haematocrit >20e25%. Arterial blood is sampled every 20 minutes for blood gas analysis, K þ, haematocrit and glucose. Anaesthesia Total intravenous anaesthesia with propofol (4 mg/kg/h) is the most commonly used method. If inhalational agents are used, they must be recommenced by the perfusionist on initiation of bypass, so as to minimize the likelihood of awareness. Cardioplegia This is one of the prime constituents in myocardial protection. Arrest of the heart is achieved by means of potassium-based solutions. Blood cardioplegia is becoming the norm. This is prepared on bypass with heparinized blood added to a crystalloid base solution. Rewarming Extremely efficient integrated heat exchangers warm the blood returning to the patient. Rewarming should not be too rapid because this can lead to bubble formation, denaturing of plasma proteins and worsening of any cerebral injury. In practice, the difference between arterial inflow and nasopharyngeal temperatures should be less than 4.0 C. Defibrillation Removal of the aortic cross-clamp and accompanying coronary reperfusion can be associated with arrhythmias, in particular ventricular fibrillation. Defibrillation is achieved using internal paddles with a biphasic energy of 4e10 J. Pacing Temporary pacing is instigated by suturing the pacing wires onto the epicardium. These wires are connected to either a singlechamber or a dual-chamber pacing device. Whilst in the operating theatre, fixed rate, dual-chamber pacing (DOO, rate 80e100, AV delay 120e150 ms) is desirable because it is insensitive to radiofrequency interference and cardiac manipulation. Once on the critical care unit, the DDD mode is selected to obviate the risk of pacing-induced ventricular fibrillation. Decannulation The mnemonic TRAVVEL acts as a useful checklist before terminating CPB (Table 2). Calculation of the systemic vasculature resistance (SVR) may be useful, particularly if significant amounts of vasoconstrictor has been used to maintain MAP throughout bypass (see section on calculation of SVR). On reinstituting mechanical ventilation, separation from CPB occurs with the perfusionist gradually occluding venous return to the pump together with incrementally reduced pump flows. With the return of stable heart rate and MAP, the venous cannula is removed. Protamine (1 mg for every 100 iu of heparin) is administered over 2e5 min. Any hypotension accompanying the protamine infusion is treated with transfusion from the venous reservoir. After removal of the aortic cannula, blood within the CPB system is drained and returned to the patient. Calculation of systemic vasculature resistance on cardiopulmonary bypass The SVR increases during the course of CPB. This is related to a gradually reducing vascular cross-sectional area, vasoconstriction accompanying hypothermia, increasing catecholamines (angiotensin II, vasopressin, endothelin) and increasing viscosity: SVRðWood unitsþ¼map CVP=pump flow where the units for MAP are mmhg, those for CVP are mmhg and those for pump flow are l/min; Wood units can be converted to dyne/s/cm 5 by multiplying by 80. TRAVVEL checklist for termination of CPB Mnemonic Stands for Check T Temperature Nasopharyngeal 36e37 C R Rate Stable cardiac rate and rhythm Epicardial pacing might be required A Air Techniques to remove intracardiac air Transoesophageal echocardiography can be used to confirm adequacy V Venting Venting lines either clamped or removed before coming off bypass V Ventilation Mechanical ventilation restarted Left lower lobe expansion visually confirmed (if pleura open) E Electrolytes Normalize metabolic indices Base excess < 5 mmol/l, PO 2 > 10 kpa, PCO 2 w5kpa Haematocrit >20%, K þ > 4.5 mmol/l L Level Operating table Table 2 ANAESTHESIA AND INTENSIVE CARE MEDICINE 10:9 418 Ó 2009 Elsevier Ltd. All rights reserved.

4 Key points for statutory regulation of perfusionists Case for regulation Invasive procedures Clinical intervention with the potential for harm Exercise of judgement by unsupervised professionals, which can substantially affect patients health or welfare Clinical perfusion scientists control, manage and are responsible for one of the most invasive tools used in routine surgery Potential for harm in a perfusionist s routine clinical practice is ever present. Supported by coroner s correspondence. Detailed in full application Evidence used in applying to Health Professions Council confirms that routine judgement of the clinical perfusion scientist can affect patients mortality and morbidity Table 3 Regulation of perfusionists Safety of the patient is paramount. Effective communication between the various teams (surgical, anaesthetic, perfusion and nursing) is vital. Recently, a number of cases in which this communication has broken down have received media attention. This has highlighted the issue of regulation of perfusionists. Perfusion is still not a government-regulated profession. The current governing body is the College of Clinical Perfusion Scientists of Great Britain and Ireland. Application for recognition is currently ongoing with the Health Professions Council (Table 3). Drug administration by perfusionists The Gritten Report, published in 2008, focuses on the issue of drug administration by perfusionists. As a profession, perfusionists routinely give fluids and drugs without prescription (e.g. Hartmann s solution, mannitol, heparin and metaraminol). The perfusionist has no authority to give drugs without a doctor s prescription, as detailed in the 1968 Medicine Act. Some hospitals, after broad consultation, have devised a set of drug administration protocols. Every drug in the perfusionist s armamentarium is detailed in these protocols. If the guidelines are adhered to, the perfusionist can administer drugs on bypass. The consultant anaesthetist must sign the perfusion record sheet at the start of each case. The record sheet contains a declaration stating: I hereby authorize the clinical perfusion scientist to administer medications and fluids in line with the Trust s clinical guidance for drug administration through the bypass machine under my supervision. This is an interim measure until national guidelines are produced. Control of parameters on cardiopulmonary bypass a-stat versus ph-stat Management of arterial blood gases (ABG) during hypothermic CPB remains controversial. Two strategies have evolved: a-stat and phstat. Hypothermia causes alkalosis. With cooling, carbon dioxide becomes more soluble (partial pressure decreases) and the ionization of water decreases. The net effect is a reduction in hydrogen ions. The term a-stat relates to how acidebase regulatory mechanisms function to maintain a constant ratio (a) of dissociated to undissociated forms of the imidazole ring of histidine. The constancy of this charge state is important in the regulation of ph-dependent cellular processes and enzyme function. With a-stat, the uncorrected (37 C) ph is kept at 40 nmol/l and the PCO 2 at 5.3 kpa, so creating a relative alkalosis with cooling. a-stat is credited with preserving cerebral autoregulation, thereby reducing microembolization. With ph-stat (corrected) a ph of 40 nmol/l and a PCO2 of 5.3 kpa is established for that particular temperature. This invariably requires carbon dioxide to be added to the circuit. ph-stat was extensively used in the 1980s in the belief that the potent vasodilatory effects of carbon dioxide would improve cerebral blood flow, thereby decreasing ischaemia on CPB. So which is the preferred strategy? In adults with moderate hypothermia (28e30 C), a-stat is recommended. In a landmark study, Murkin et al. 1 clearly showed the benefits of this approach. However, if deep hypothermic cardiac arrest (DHCA) is used, a cross-over strategy is advocated. Here, ph-stat is used in the initial 10 minutes of cooling, followed by a switch to a-stat. This maximizes cerebral cooling yet avoids the severe acidosis associated with prolonged ph-stat management. In infants, in whom brain injury is more associated with hypoperfusion or the initiation of excitotoxic pathways, ph-stat seems to be advantageous. Pressure versus flow If MAP is kept within the autoregulatory range (50e150 mmhg), cerebral blood flow (CBF) is essentially independent of pressure. This has implications for patients with hypertension, in whom the autoregulatory curve is shifted to the right. Gold et al. 2 provided some direction in a contentious study. Patients were randomized either to lower (50e60 mmhg) or to higher (80e100 mmhg) CPB pressure groups. Using the composite end-point of combined adverse cardiac and neurological outcomes, the higher CPB pressure group had significantly better outcomes 6 months after surgery. Further analysis showed that patients at high risk of stroke (severely atheromatous aorta) were more likely to have a stroke if the MAP was kept in the lower pressure range. The inter-relationship between pump flow and pressure is complex, and is dependent on arterial impedance, temperature, haemodilution and arterial cross-sectional area. Again, it seems that, if cerebral autoregulation remains intact, CBF is unaltered by changes in pump flow. At low-flow states, CBF is likely to be dependent on perfusion pressure. ANAESTHESIA AND INTENSIVE CARE MEDICINE 10:9 419 Ó 2009 Elsevier Ltd. All rights reserved.

5 Normothermic versus hypothermic cardiopulmonary bypass In the 1990s, a number of studies were carried out to investigate whether hypothermia was a requirement for CPB. The Toronto Warm Heart Investigators (1994) showed no increase in neurological complications with the use of normothermic CPB. Conversely, the Atlanta Group demonstrated marked increases in central nervous system CNS problems. Because the study designs were so different, it is not surprising that the results were conflicting. Accumulated evidence indicates that normothermic CPB does not increase the risk of adverse neurological outcomes and that mild hypothermia (34 C) confers additional cerebral protection compared with mild hyperthermia (>37 C), which exacerbates any ischaemic injury. Central nervous system monitoring Adverse cerebral events are the most devastating outcomes associated with cardiac surgery. Factors contributing to brain injury include atheromatous emboli from aortic cannulation, lipid microemboli, gaseous microemboli (air leaks and cavitation), hypoperfusion, hyperperfusion and hyperthermia. Roach et al., 3 studying 2400 patients undergoing elective coronary artery bypass graft in 24 US centres, reported a 6.1% incidence of neurological or neuropsychiatric adverse outcomes. These patients required prolonged hospitalization, with only one in three returning home, and most needed long-term care and rehabilitation. e Near infra-red spectrometry (NIRS): a non-invasive method of determining intravascular regional haemoglobin oxygen saturation. Based on the premise that the skull is translucent to infrared light. The data are displayed as a continuous trend. Remains controversial but gaining acceptance. CNS haemodynamics e Transcranial Doppler ultrasonography: ultrasound probes are placed over the thinnest portion of the temporal bone (acoustic window) and aligned to the middle cerebral artery (MCA). At any one time, the MCA carries 40% of the hemispheric blood. Interpretation of the information is usually operator dependent, and securing the probes is difficult. This technique is sensitive at detecting intracranial blood flow and emboli (particulate and gaseous). Adequately powered prospective studies indicating improved outcomes and economic benefits are sparse. The results of a few outcome studies using multimodal neuromonitoring (EEG, transcranial Doppler, cerebral oximetry) have shown substantial reductions in length of stay and hospital expense compared with other types of monitoring. Interestingly, and perhaps not unexpected, these studies also indicate a benefit to other vital organ systems. A Monitoring neurological function falls into three areas. CNS electrical activity e Electroencephalogram (EEG): measures postsynaptic potentials in cerebral cortical neurones. Exquisitely sensitive but too cumbersome for intraoperative use. e Processed EEG: compressed spectral array, bispectral index (BIS). EEG data are further analysed using fast Fourier transforms, generating a power versus time spectral array. Trends are formed and displayed. Overall, more applicable to the operating theatre. e Evoked potentials (somatosensory, auditory, motor): of some use in thoracic aneurysmal surgery in which there is a threat to the spinal cord. CNS metabolic activity e Jugular bulb venous oximetry: a suitably placed fibreoptic catheter gives a continual global measure of cerebral venous oxygen saturation. Unreliable in profound hypoperfusion and low-flow states. REFERENCES 1 Murkin JM, Martzke JS, Buchan AM, et al. A randomized study of the influence of perfusion technique and ph management strategy in 316 patients undergoing coronary artery bypass surgery. J Thorac Cardiovasc Surg 1995; 110: 349e62. 2 Gold JP, Charlson ME, Williams-Russo P, et al. Improvement of outcomes after coronary artery bypass: a randomized trial comparing intraoperative high versus low mean arterial pressure. J Thorac Cardiovasc Surg 1995; 110: 1302e14. 3 Roach GW, Kanchuger M, Mora Mangano CT, et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996; 335: FURTHER READING Kaplan JA, Reich DL, Lake CL, Konstadt SN. Kaplan s cardiac anesthesia. 5th edn. Philadelphia: Saunders Elsevier; ANAESTHESIA AND INTENSIVE CARE MEDICINE 10:9 420 Ó 2009 Elsevier Ltd. All rights reserved.