Introduction An increasing proportion of the patients undergoing cardiac surgery have diabetes mellitus, in particular type II diabetes. In spite of this, diabetic patients have received limited attention in this setting. Although diabetes is a metabolic disease cardiac metabolism in association with surgery has previously not been explored in diabetics. This investigation was carried out to describe the metabolic state of the heart in diabetics after cardiac surgery and to study if it is accessible to metabolic intervention with high-dose GIK. Also, the potential hazards associated with such a regime in clinical practice were evaluated. Furthermore, a comparison of the outcome in diabetic and nondiabetic patients after coronary surgery was done. Methods Myocardial metabolism and how it was influenced by high-dose GIK was assessed with coronary sinus catheter technique in a prospective randomized study on 20 type II diabetic patients undergoing CABG (paper I, II). Safety issues concerning high-dose GIK were assessed in two retrospective studies. The potential role of metabolic interventions for neurological injury was assessed in a cohort of 775 consecutive patients undergoing CABG or combined CABG + valve surgery, in whom metabolic interventions gradually replaced traditional treatment for postoperative heart failure (paper III). A detailed analysis of blood glucose and electrolyte control was done in all cases (n=89) receiving high-dose GIK during one year (paper IV). The hemodynamic impact of highdose GIK was assessed with standard postoperative monitoring including Swan-Ganz catheters (paper II, IV). Outcome and prognosis after CABG in diabetic patients (n=540) were compared with nondiabetics (n=2239) with the aid of the institutional database comprising all isolated CABG procedures from 1995-1999 (paper V). Results The metabolism of the diabetic heart after CABG was characterized by predominant uptake of FFA and restricted uptake of carbohydrate substrates. A high extraction rate of beta-hydroxybutyric acid and glutamate was also found. Alanine was released from the heart (paper I). High-dose GIK induced a shift towards uptake of carbohydrates, in particular lactate, at the expense of FFA and betahydroxybutyric acid (paper II). A substantial systemic glucose uptake was found during high-dose GIK treatment but the uptake tended to be lower and blood glucose higher if adrenergic drugs were used or/and if the patient was a diabetic (paper IV). High-dose GIK was associated with beneficial effects on cardiac output both in the prospective and retrospective analyses (paper II, IV). No evidence for untoward neurological effects associated with GIK treatment was found. History of cerebrovascular disease was the most important risk factor for postoperative cerebral complications and in general markers for advanced atherosclerotic disease were found to be of importance (paper III). High-dose GIK in clinical practice was associated with acceptable blood glucose and electrolyte control and no serious adverse events were recorded (paper IV). Patients with diabetes undergoing CABG had an acceptable short-term mortality that did not differ significantly from non-diabetic patients. However, diabetic patients had a higher early postoperative morbidity particularly with regard to stroke, renal- and infectious complications. Also, long-term survival was markedly reduced in diabetic patients, particularly in insulin treated patients (paper V). Comments FFA were the main source of energy for the heart in type II diabetics after CABG whereas the uptake of carbohydrates was restricted. The high extraction rates of beta-hydroxybutyric acid and glutamate may represent an adaptation to the unfavorable metabolic situation of the post-ischemic diabetic heart. High-dose GIK can be used in type II diabetic patients after cardiac surgery to promote carbohydrate uptake at the expense of FFA and beta-hydroxybutyric acid. The magnitude of this shift was sufficient to account for the entire myocardial oxygen consumption assuming that the substrates extracted were oxidized. This could have implications for the treatment of the diabetic heart in association with surgery and ischemia. Provided careful monitoring high-dose GIK can be safely used in clinical practice and this treatment deserves further evaluation in the treatment of postoperative heart failure. High-dose GIK also provides a means for strict blood glucose control and as substantial amounts of glucose can be infused even in critically ill patients, it may prove useful for nutrition in critical care. Several of the risk factors for neurological injury identified constitute markers for advanced atherosclerotic disease, thus, also providing an explanation for the increased risk of neurological injury in diabetics after cardiac surgery. Short-term mortality was acceptable in diabetics after CABG. However, further efforts are warranted to address postoperative morbidity and late outcome. This represents a challenge as diabetic patients are accounting for an increasing proportion of the patients undergoing CABG. / On the day of the public defence the status of article IV was: Submitted and the title of article IV was in the printed version: High-dose GIK in cardiac surgery - clinical safety issues and lessons learned.

Introduction: Neurological complications and stroke in association with cardiac surgery is a serious problem. The stroke event can occur during surgery (early stroke) or in the postoperative period with a symptom free interval (delayed stroke). Particle embolization due to aortic manipulation during surgery has been suspected as a mechanism for early stroke. The present thesis address mechanisms and survival effects of stroke both clinically (I-III) and experimentally (IV-V). Methods: Study I) Within a cohort of 2641 consecutive cases, a group of cardiac surgery patients with stroke and evaluated by computed tomography (CT) were studied (n=77). CT-findings were analyzed in relation to stroke symptoms. Study II) Data from 9122 patients undergoing coronary surgery were analyzed. Records of patients with any signs of neurological complications were reviewed to extract 149 subjects with stroke at extubation (early, 1.6%) versus 99 patients having a free interval (delayed, 1.1%). Early and delayed stroke were evaluated separately. Independent risk factors for stroke were analyzed by logistic regression and survival by Cox regression (9.3 years median follow-up). Study III) Patients with early (n=223) and delayed stroke (n=116) were identified among 10809 patients undergoing cardiac and aortic surgery, both groups exposed to cardiopulmonary bypass. Stroke patients were subdivided by the hemispheric location of lesions. Subgroups were compared and their associated pre- and peroperative variables and survival were analyzed. Study IV) Aortic cross-clamp manipulation was studied in a human cadaveric perfusion model. The pressurized aorta was repeatedly cross-clamped and washout samples were collected before and after clamp maneuvers. Particles in the washout samples were evaluated by microscopy and by digital image analysis. Study V) Pig aortas were pressurized and cannulated. Washout samples were collected before and after cannulation (n = 40). Particles were deposited onto a 10-μm filter to be evaluated by microscopy and digital image analysis. Results: Study I) In the group of patients exposed to routine cardiac surgery (i.e., clamping and cannulation) and with early stroke, right-hemispheric lesions were more frequent than of the contra-lateral side (P=0.005). Patients with aortic dissections had a strong dominance of bilateral findings, which was different from the unilateral pattern in the routine-surgery group (P<0.001). Study II) Early and delayed stroke did not share any risk factors. Both early and delayed stroke explained mortality in the early postoperative period (P<0.001, P<0.001 respectively) but also at long term follow-up (P=0.008, P<0.001 respectively). For patients surviving their first postoperative year, delayed but not early stroke influenced long-term mortality (P=0.001 and P=0.695, respectively). Study III) Stroke lesions in association to cardiac surgery were near exclusively ischemic. Early stroke had a preponderance for right-hemispheric lesions (P=0.009). In contrast, patients with early stroke that had undergone surgery of the aorta with circulatory arrest showed a pattern with more bilateral lesions compared to ‘cardiac-type’ operations (P<0.001). Patients with bilateral lesions had a dramatically impaired survival compared to those with unilateral lesions (P<0.001). Study IV) In the cadaveric perfusion model, cross-clamping produced a significant output of particles, which was seen for size intervals of 1 mm and smaller (P=0.002 to P=0.022). In all size intervals the particle output correlated with the degree of overall aortic calcification (P =0.002 to P=0.025). Study V) At cannulation of the pig aorta, more particles were noted after cannulation compared to before the maneuver (P<0.001). This increase included small (<0.1 mm, P<0.001) and intermediate-size particles (0.1-0.5 mm, P< 0.001). Particles above 0.5 mm were few and were not associated with cannulation. Conclusions: The influence of stroke on mortality was devastating, for both early and delayed stroke. These two stroke groups had obvious differences in both their risk factors and their hemispheric distribution. It is here emphasized that early and delayed stroke should be considered as two separate entities with suggested mechanistic differences. Ischemic lesions accounted for near all stroke events seen in association to cardiac surgery. For early stroke, these were mostly located within the right hemisphere. Results from the experimental studies underscore microembolic risks associated with aortic manipulation.

ABSTRACT In the modern era, the most common post-operative complications following cardiopulmonary bypass (CPB) are neurocognitive deficits (NCD) and atrial fibrillation (AF). Both morbidities have been linked to inflammation resulting from surgery, anesthesia and CPB. Microemboli, inadequate oxygen delivery and the inflammatory response consequent to blood contacting artificial components of the CPB circuit have all been linked to postoperative NCD and to a lesser extent post-operative AF. The artificial components of the CPB circuit consist of stainless steel, polyvinylchloride (PVC), polycarbonate and other carbon-based plastics. In order to attenuate the negative sequelae of blood-circuit contact related inflammatory response, industry developed the biocompatible circuit (BCC) coating for the disposable CPB circuits. Four such coatings were studied and compared to an uncoated control group in a total of 101 patients undergoing routine CPB-assisted cardiac surgical procedures. Soluble adhesion molecule (SAM) activation was studied at different time points and common clinical outcomes such as white blood cell activation, serum renal function parameters urea and creatinine, postoperative bleeding, transfusion requirements, intensive care and hospital length of stay, CPB pump volume balances, changes in weight, postoperative serum lactate and glucose and the development of AF postoperatively, were compared. Additionally, postoperative neurocognitive testing was performed using a simple bedside neurocognitive test called the antisaccadic eye movement test. The patients in all groups were tested for comparison preoperatively and 72 hr postoperatively. Results: The mandate of BCC coating development and manufacture is to attenuate the well-documented and demonstrated inflammatory response consequent to the contact of blood with artificial CPB surfaces. The studied BCCs significantly decreased platelet transfusions in females. In addition, the BCCs decreased the concentrations of 2 SAMs when measured 6 hours after surgery and CPB. The difference in SAM expression seen between the coated and uncoated groups at 6 hr was no longer apparent at 72 hr. Very little difference was noted between the four BCC groups. Patients who developed AF postoperatively seemed predisposed to do so as the serum levels of soluble vascular cell adhesion molecule was significantly higher at baseline and remained so at 6 and 72 hr. The decreased platelet transfusions in females resulting from BCC use is a highly significant finding within this high-risk group of patients. As most platelet transfusions occur soon after the patient is disconnected from CPB, the short-term decrease in SAM activation can be linked to this improved clinical finding. The studied BCC coatings have achieved limited success in their intended mandate to attenuate inflammatory response in terms of improved clinical and laboratory desired outcomes.

ABSTRACT In the modern era, the most common post-operative complications following cardiopulmonary bypass (CPB) are neurocognitive deficits (NCD) and atrial fibrillation (AF). Both morbidities have been linked to inflammation resulting from surgery, anesthesia and CPB. Microemboli, inadequate oxygen delivery and the inflammatory response consequent to blood contacting artificial components of the CPB circuit have all been linked to postoperative NCD and to a lesser extent post-operative AF. The artificial components of the CPB circuit consist of stainless steel, polyvinylchloride (PVC), polycarbonate and other carbon-based plastics. In order to attenuate the negative sequelae of blood-circuit contact related inflammatory response, industry developed the biocompatible circuit (BCC) coating for the disposable CPB circuits. Four such coatings were studied and compared to an uncoated control group in a total of 101 patients undergoing routine CPB-assisted cardiac surgical procedures. Soluble adhesion molecule (SAM) activation was studied at different time points and common clinical outcomes such as white blood cell activation, serum renal function parameters urea and creatinine, postoperative bleeding, transfusion requirements, intensive care and hospital length of stay, CPB pump volume balances, changes in weight, postoperative serum lactate and glucose and the development of AF postoperatively, were compared. Additionally, postoperative neurocognitive testing was performed using a simple bedside neurocognitive test called the antisaccadic eye movement test. The patients in all groups were tested for comparison preoperatively and 72 hr postoperatively. Results: The mandate of BCC coating development and manufacture is to attenuate the well-documented and demonstrated inflammatory response consequent to the contact of blood with artificial CPB surfaces. The studied BCCs significantly decreased platelet transfusions in females. In addition, the BCCs decreased the concentrations of 2 SAMs when measured 6 hours after surgery and CPB. The difference in SAM expression seen between the coated and uncoated groups at 6 hr was no longer apparent at 72 hr. Very little difference was noted between the four BCC groups. Patients who developed AF postoperatively seemed predisposed to do so as the serum levels of soluble vascular cell adhesion molecule was significantly higher at baseline and remained so at 6 and 72 hr. The decreased platelet transfusions in females resulting from BCC use is a highly significant finding within this high-risk group of patients. As most platelet transfusions occur soon after the patient is disconnected from CPB, the short-term decrease in SAM activation can be linked to this improved clinical finding. The studied BCC coatings have achieved limited success in their intended mandate to attenuate inflammatory response in terms of improved clinical and laboratory desired outcomes.

Thoraxkirurgiska ingrepp har väldokumenterade ökade risker för svåra smärtor postoperativt och tidigare forskning visar om den postoperativa smärtlindringen underbehandlas ökar risken att utveckla en kronisk smärta (CPSP) som drabbar 30-50% av alla som genomgår thoraxkirurgi. Detta leder till högre vårdkostnader för samhället, en kraftig försämring av patientens livskvalitet och ett vårdlidande för patienten. Sjukvården borde fokusera på att möta patienternas lidande och ge redskap och möjlighet för vårdarna att lindra lidandet och stärka patientens förmåga till egenvård av smärta i hemmet. Syftet med denna litteraturstudie var att belysa sjuksköterskans upplevelse av vad som behövs för att ge en bra smärtbehandling till patienter som genomgått större kirurgi exempelvis thoraxkirurgi och vilka hinder som måste överstigas för att lyckas. Examensarbetet genomfördes som en systematisk litteraturstudie där innehållet från elva artiklar analyserades och sammanställdes. I analysen av resultatet framkom två teman; Upplevda hinder och sjuksköterskans strategier. De beskriver hur sjuksköterskan på olika sätt arbetar med smärtskattning och vilka hinder som kan uppstå. Resultatet visar på stora problem i smärtbehandling, både vad gäller att smärtskatta som att lindra lidandet och berodde bland annat på bristande kunskap hos både patient och sjuksköterska. Slutsatsen är att sjukvården borde satsa på mer individualiserad patientinformation både preoperativt och innan hemgång, så patienten förstår varför smärtlindring är viktigt. Om detta görs kan man förhindra postoperativa komplikationer och kronisk smärta, som leder till lägre vårdkostnader och minst lika viktigt att bibehålla en patient med god livskvalitet.

Hyperglycemia in adult cardiac surgery may result in post-operative sternal wound infections, pneumonia, renal failure, increased length of stay, and cost. The Surgical Care Improvement Project (SCIP) (2006) requires blood glucose control in cardiac surgery at 6:00 am on post-operative day one (POD1) and post-operative day two (POD2) to be 200mg/dL or less. Enhanced Recovery After Surgery (ERAS) guidelines use a Maltodextrin 12.5% carbohydrate beverage six hours and two hours pre-operative of general surgery to improve post-operative outcomes, cost, and length of stay. One study replicated ERAS guidelines in adult coronary artery bypass grafting surgery patients finding patients had decreased length of stay and improved glycemic control six hours postoperatively. The purpose of this quality improvement project is to outline a proactive approach to the modifiable risk factor of pre-operative fasting. This quality improvement project describes a pre-operative fasting carbohydrate protocol for non-emergent, adults, scheduled for cardiac surgery at 10:00am or later, with a hemoglobin A1C of 8.4% or less, and a body mass index of 35 or less. The protocol is named the Hungry Sweet Heart Protocol and an implementation plan is described for a community hospital located in Tucson, AZ. Updating practices of strict NPO status prior to cardiac surgery is a proactive measure to improve glycemic control and adherence to SCIP guidelines post-operatively. Interdisciplinary teams, including DNPs, are perfectly suited to guide this implementation.

A substantial number of patients undergoing cardiac surgery are on dual anti-platelet treatment with clopidogrel and aspirin. A disadvantage with this treatment is increased risk of bleeding. Bleeding is a complication of major concern associated with adverse outcome for the patient and increased hospital resource utilization. Great variability in individual response to clopidogrel has been reported. If in vitro measurements of platelet reactivity would correlate with clinical bleeding parameters, potential bleeders could be identified preoperatively. The aims of this thesis were: (1) to describe the degree of pre-operative platelet inhibition in patients scheduled for primary isolated coronary artery bypass graft surgery; (2) to prospectively investigate whether the pre-operative platelet inhibition correlated with intra- and postoperative bleeding and transfusion requirements; and (3) to test the ability of clinically relevant risk factors to predict re-exploration for bleeding. (4) In addition, a cost analysis was performed on patients re-explored for bleeding, to analyse the magnitude of added resource utilization and costs. Based on this, a cost model of prophylactic treatment with haemostatic drugs was calculated. Platelet function tests investigated were: (1) flow cytometry, (2) VASP, (3) VerifyNowSystem, (4) PlateletMapping (a modified TEG), and (5) PFA-100. Clinical risk factors for re-exploration and the influence of antiplatelet and antifibrinolytic therapy were evaluated in a retrospective analysis. Cost analysis at three cardiothoracic centres was performed in a case-control study. In conclusion, there was no clinically useful correlation between preoperative assessment of platelet inhibition and blood loss or transfusion requirements during coronary artery bypass surgery. Furthermore, there was only modest agreement between the methods evaluating ADP-receptor blockade. Pre-operative treatment with the P2Y12-receptor inhibitor clopidogrel was an essential risk factor for re-exploration due to bleeding. Except for clopidogrel, no strong clinical factor to predict the risk of re-exploration was identified. The resource utilisation costs were 47% higher in patients requiring re-exploration due to bleeding than in those not requiring re-exploration. Prolonged stay in the ICU and recovery ward accounted for half of the added cost, a third was due to the costs of surgery, one fifth due to increased cost of transfusions, and <2% was due to haemostatic drug treatment.

Evidence suggests that there is an immunological response of the recipient to xenograft bioprosthetic heart valves. Information on the impact of HMG-CoA reductase inhibitors (statins) and their anti-inflammatory properties on bioprosthetic valve failure remains limited. We sought to examine the efficacy of statin therapy in a rodent model of bioprosthetic valve implantation. To mimic the human scenario, fresh or glutaraldehyde-fixed aortic valve root conduits from Lewis rats or Hartley guinea pigs were microsurgically implanted intravascularly into the infra-renal aorta of Lewis rats. The syngeneic control group consisted of a fresh rat valve conduit implanted into a rat. The xenogeneic control group consisted of a glutaraldehyde-fixed guinea pig valve conduit implanted into a rat. Treatment groups consisted of xenogeneic groups treated with either daily steroids or statins. Overall, steroid treatment attenuated the inflammatory response observed within the xenogeneic glutaraldehyde-fixed valve conduits. Treatment with statins did not decrease this inflammatory response.

The purpose of this study was to investigate the effects of three different glycemic control conditions (tight, conventional, and standard) in the intraoperative period on: 1) postoperative surgical site infections, and 2) postoperative procalcitonin, and C-reactive protein levels in patients undergoing open-heart surgery. Secondary aims of the study were to investigate the effects of the three glycemic treatment conditions on: 1) intraoperative blood glucose; 2) intraoperative glycemic stability; and 3) intensive care unit length of stay, in patients undergoing open-heart surgery. An experimental design with a multilevel, single factor, within-subjects design was utilized. Patients were nested within anesthesia provider teams. The design was counterbalanced by means of a Latin square, where each of three anesthesia provider teams dispensed each of three glycemic control conditions once. Thirty-seven participants were randomized to either tight glycemic control (n =15), which maintained blood glucose 110-149 mg/dl via continuous intravenous insulin infusion, conventional glycemic control (n = 11), which maintained blood glucose 150-180 mg/dl via continuous intravenous insulin infusion, or standard glycemic control (n =11) which maintain blood glucose 150-180 mg/dl via intravenous bolus injections of insulin. The main findings of this study were that there were no significant differences between the three glycemic interventional treatment groups in 1) thirty-day surgical site infections, 2) postoperative C-reactive protein or procalcitonin concentrations 3) intensive care unit length of stay, 4) intraoperative blood glucose levels, or 5) glycemic stability. An association between intraoperative peak blood glucose and surgical site infection was established. Participants that experienced higher peak blood glucose levels intraoperatively exhibited increased surgical site infections. Procalcitonin levels were significantly elevated in participants that experienced a surgical site infection, but C-reactive protein showed no significant difference between participants with or without a surgical site infection. Coronary artery bypass graft surgery concomitant with valve replacement surgery was associated with a higher rate of surgical site infections compared coronary artery bypass graft surgery or valve surgery independently. In conclusion, an association was found between higher peak intraoperative blood glucose levels and increased surgical site infections, therefore maintaining intraoperative blood glucose levels below 180 mg/dl via a continuous intravenous infusion of insulin, may reduce postoperative surgical site infections in the open-heart patient. The use of tight glycemic control during the intraoperative period can be achieved safely, with the use of judicious protocols, but its benefits remain unproven. Inflammatory biomarker procalcitonin was predictive of infection, where C-reactive protein was not. The addition of procalcitonin to routine postoperative blood work, in open-heart patients, may benefit providers in the diagnosis and early treatment of surgical site infections. This study was underpowered. Further studies with appropriate sample size, may be able to determine if tight glycemic control, compared to moderate glycemic control, in the intraoperative period is of benefit to patients undergoing open-heart surgery.

Evidence suggests that there is an immunological response of the recipient to xenograft bioprosthetic heart valves. Information on the impact of HMG-CoA reductase inhibitors (statins) and their anti-inflammatory properties on bioprosthetic valve failure remains limited. We sought to examine the efficacy of statin therapy in a rodent model of bioprosthetic valve implantation. To mimic the human scenario, fresh or glutaraldehyde-fixed aortic valve root conduits from Lewis rats or Hartley guinea pigs were microsurgically implanted intravascularly into the infra-renal aorta of Lewis rats. The syngeneic control group consisted of a fresh rat valve conduit implanted into a rat. The xenogeneic control group consisted of a glutaraldehyde-fixed guinea pig valve conduit implanted into a rat. Treatment groups consisted of xenogeneic groups treated with either daily steroids or statins. Overall, steroid treatment attenuated the inflammatory response observed within the xenogeneic glutaraldehyde-fixed valve conduits. Treatment with statins did not decrease this inflammatory response.