Coll. Antropol. 36 (2012) 2: 381–388 Original scientific paper The Relationship between Prolonged Cerebral Oxygen Desaturation and Postoperative Outcome in Patients Undergoing Coronary Artery Bypass Grafting @eljko ^olak1, Marko Borojevi}2, Vi{nja Ivan~an1, Rajka Gabelica1, Bojan Bio~ina2 and Vi{nja Majeri}-Kogler1 1 2 University of Zagreb, Zagreb University Hospital Center, Department of Anesthesiology, Reanimatology and Intensive Care, Zagreb, Croatia University of Zagreb, Zagreb University Hospital Center, Department of Cardiac Surgery, Zagreb, Croatia ABSTRACT 58 patients who underwent on-pump coronary artery bypass graft surgery were evaluated for changes in regional cerebral oxygen saturation (rSO2) measured by near infrared spectroscopy (NIRS). If rSO2 during the operation fell to more than 20% under the baseline, standardized interventions were undertaken to maintain rSO2. Despite those interventions, in some cases we observed inability to maintain rSO2 above this threshold. Therefore we divided patients in two subgroups: 1. without prolonged rSO2 desaturation; 2. with prolonged rSO2 desaturation (area under the curve >150 min% for rSO2<20% of baseline and >50 min% for rSO2<50% of absolute value). The data were analyzed to determine whether there were major differences in outcome of these two groups. 18 out of 58 patients (31%) had prolonged rSO2 desaturation during operation. There was significantly higher number of diabetic patients in group with prolonged rSO2 desaturation (p=0.02). Intraoperative data revealed significantly more blood consumption during cardiopulmonary bypass (p=0.007) and the need for inotropes (p=0.04) in desaturation group. Three patients in prolonged desaturation group and no one in another group had stroke, coma or stupor (p=0.03). Logistic regression analysis revealed diabetes mellitus and age as predictors for prolonged rSO2 desaturation. We concluded that prolonged intraoperative rSO2 desaturation is significantly associated with worse neurological outcome in patients – nonresponders to standardized interventions for prevention of rSO2 desaturation. Key words: cerebral oximetry, near-infrared spectroscopy, on-pump coronary artery bypass, neurologic outcome Introduction Neurological injury in patients undergoing coronary revascularization remains one of the major causes of morbidity after cardiac surgery. The incidence of cerebral injury is still high (up to 6.1%) because the number of aged patients with numerous comorbidities has significantly increased in cardiosurgical population. Several etiologic factors have been proposed including embolisation, hypoxic insult, compromised cerebral blood flow and systemic inflammatory response, but also previous unrecognized neurological abnormality1–3. There is a need for cerebral monitoring during cardiac surgery with the aim to reduce postoperative neurological complications4–6. Transcranial near-infrared spectroscopy (NIRS) provides continuous noninvasive measurement of regional cerebral oxygen saturation (rSO2) in small local area of cortex. It is influenced by the actual oxygen supply and demand balance and serves as a sensitive indicator of global cerebral hypoperfusion7,8. Numerous studies reported that cerebral desaturation during cardiac surgery is associated with early neurological dys- Received for publication November 29, 2010 381 @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 function and other postoperative adverse events. Intraoperative monitoring of cerebral oxygenation with NIRS and maintaining of balance in oxygen supply-demand can reduce postoperative neurologic and systemic complications9–14. The aim of the study was to examine usefulness of intraoperative cerebral oxymetry monitoring and efficacy of interventions performed to maintain regional rSO2 above the baseline value by determining a relationship between intraoperative cerebral oxygen desaturation and adverse postoperative events. Patients and Methods Study population The data were obtained from University Hospital Center Zagreb and the study was approved by the hospital Ethical committee. The total number of 58 patients, who underwent on-pump coronary artery bypass graft surgery (CABG) and signed informed consent, was included in the study. These patients were recruited between June 2009 and February 2010. Patients with significant carotid artery stenosis, previous stroke or head injury, seizure, psychiatric illness, decompensate heart failure (NYHA III/IV), emergency cardiac surgery, off-pump CABG, and severely impaired renal or liver function were excluded from the study. Operation protocol The same anesthesia protocol was applied to all patients and included premedication with intramuscular morphine 0.1 mg/kg and induction with midazolam 0.1 mg/kg, hypnomidate 0.2 mg/kg, sufentanyl 1 mcg/kg and pancuronium 0.1 mg/kg. After tracheal intubation anesthesia was maintained with sufentanyl-continuous infusion of 0.8 mcg/kg/h, propofol-5 mg/kg/h (only during cardiopulmonary bypass), sevoflurane 0.5–2 minimal alveolar concentration (MAC) and pancuronium 2 mg if necessary. Minute ventilation was adjusted to maintain arterial CO2 partial pressure (PaCO2) between 4.5 and 6 kPa and FiO2 of 0.5. Measurements of cardiac index and other hemodynamic parameters were provided using Swan-Ganz catheter from the beginning of the operation until 24 hours after arrival in ICU. Inotropic agents were used to maintain a cardiac index ³ 2.0 L/min/m2 and a systolic blood pressure ³ 90 mmHg after preload was adjusted according to hemodynamic parameters obtained by Swan-Ganz catheter and transesophageal ultrasound. The first inotrop of choice was dobutamin. No other monitoring of neurologic function was used. BIS monitoring was not used. After heparinization (400 i.j./kg) cardiopulmonary bypass (CPB) was established with a moderate hemodilution and systemic hypothermia (28–32 °C). Heart was protected with blood cardioplegia (Croatian Institute for Transfusion Medicine, Zagreb) during cross-clamping of aorta. Generally a single clamp technique was used. Standard flow rates of 2–2.5 L/minute/m2 were main382 tained with hematocrit above 22%, mean arterial pressure over 60 mmHg, partial pressure of carbon dioxide between 4.5 and 6 kPa by alpha-stat management15. The following parameters were recorded during operation: number of grafts, cardiopulmonary bypass time (CPB time), aortic cross-clamping time (ACC time), duration of operation, cardiac index (CI) before and after cardiopulmonary bypass, the amount of blood given during and after cardiopulmonary bypass, number of patients who needed inotropic support and parameters about rSO2. Cerebral oxymetry All patients were monitored using the INVOS system (INVOS 5100C; Somanetics Corp, Troy, MI, USA). The INVOS system is based on near-infrared spectroscopy (NIRS) for noninvasive and continuous measurement of changes in cerebral oxygen saturation. Before induction in anesthesia the probes for INVOS cerebral oxygen monitoring were attached bilaterally on the patient’s forehead overlying the frontal-temporal region. The light in the near infrared range penetrates the skin, skull and deeper cerebral tissues. Two values were measured with two detectors at different distances (3 cm and 4 cm) from the light source. The signals common to both measurements (from extracranial tissue) are eliminated and oxygen saturation of blood in the cerebral cortex beneath the sensors is displayed. NIRS doesn’t need pulsatile flow so it can be used during cardiopulmonary bypass. Probes were attached in awaken patients who breathe 6 L/min of oxygen by nasal catheter. Several minutes after the probes were attached the baseline regional cerebral oxygenation (rSO2) value was determined for each side of the brain. During the entire surgery rSO2 values were displayed on a screen and recorded on the USB memory stick. If the rSO2 decreased to more than 20% of baseline during the operation, we responded with standardized interventions to maintain rSO2 above the baseline value. These interventions (after checking head and perfusion cannulae position) included maximizing of cerebral oxygen delivery (increasing of oxygen concentration – FiO2, arterial CO2 partial pressure – PaCO2, mean arterial pressure, cardiac output or pump flow and hematocrit) or reduction in cerebral oxygen consumption (raising of anesthetic depth and reduction of temperature)16. If systemic oxygen saturation fell we increased the fraction of inspired oxygen (FiO2). We try to maintain mean arterial pressure within 15% of baseline using vasopressors as required. Arterial CO2 partial pressure (PaCO2) has strong impact on cerebral blood flow. Hypocapnia causes decrease and hypercapnia increase in cerebral blood flow and rSO2, so one of the common interventions was normalization of hypocapnia or increasing of PaCO2 to higher normal values. rSO2 desaturation associated with hematocrit reduction to < 22% was the indication for transfusion. We undertook certain steps (preload and afterload management, inotropes) to increase cardiac output or to increase pump flow rate to correct rSO2 desaturation. We avoided hyperthermia after rewar- @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 ming during cardiopulmonary bypass because it would be associated with increase in cerebral metabolic rate of oxygen (CMRO2) and with rSO2 reduction. The rewarming rate during cardiopulmonary bypass was slow, and it took approximately half an hour. One of the measures to improve rSO2 was deepening of anesthesia to reduce CMRO2. The order of performed interventions was at the anesthesiologist’s discretion according to estimation of decline degree of each parameter from acceptable range. Usually the parameter with the worst decline from normal was first modified. After the surgery, areas under the curve (AUC) were calculated for desaturations; more than 20% of baseline value (rSO2 AUC<20% baseline), more than 50% (rSO2 AUC<50% absolute) and more than 40% of absolute value (rSO2 AUC< 40% absolute). Areas under the curve account for both depth and duration of desaturation below these thresholds so they are expressed as min% values. We divided patients for analysis in two subgroups: 1. without prolonged cerebral desaturation; 2. with prolonged cerebral desaturation. Prolonged desaturation was defined as rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min%, based on the Murkin12 and Slater14 studies. The data were analyzed to determine whether there were major differences in outcome of both groups. Postoperative management and patients’ outcome During postoperative period all complications were recorded. ECGs were obtained before CABG, immediately after arrival in ICU and each postoperative day in ICU. Patients were monitored by telemetry during their postoperative hospital stay and the incidence of new atrial fibrillation from ICU admission until hospital discharge was recorded. Troponin T levels were measured immediately after surgery and on the first and second postoperative day. The presence of neurological deficit such as coma, stupor, stroke or delirium was observed17. Coma is defined as a profound state of unconsciousness without response to verbal call, pain or any other stimulus. Stupor is defined as a state of unconsciousness from which patient can be aroused only by vigorous physical stimulation. Stroke is defined as a syndrome characterized by the acute onset of a neurologic deficit that persists for at least 24 hours and reflects focal involvement of the central nervous system. Delirium or encephalopathy is characterized by confusion, agitation, disorientation, decreased alertness, sleep disturbances, memory deficit or seizure but without an obvious focal neurological deficit. Ventilatory time was recorded in hours from the admission to ICU until the moment of extubation. Respiratory insufficiency is defined as a need for prolonged mechanical ventilation for more than 48 hours. Infection is defined as a clinically manifested and microbiologically confirmed infection which required antibiotics therapy. Patients who developed acute renal insufficiency and needed dialysis treatment were detected. The length of stay in the ICU was defined as time from ICU arrival un- til transfer to the floor. Before transfer, patients had to be extubated with stable vital signs and without any inotropic support. Patients in our institution usually spend up to seven days in the hospital after surgery. The need for prolonged hospitalization of more than seven days was recorded. Criteria for discharge included stable cardiac rhythm, temperature < 37 °C, a well healed incision, and oxygen saturation >90 % on the room air. Logistic value of EuroSCORE was calculated for all patients. EuroSCORE is a simple, objective and up-to-date scoring system for assessment of heart surgery, based on one of the largest, most complete and accurate databases in European cardiac surgery18. SAPS II (Simplified Acute Physiology Score) values were calculated in the moment of entrance and discharge from ICU. SAPS II is a severity of disease classification system for patients admitted to intensive care units19. All patients older than 65 years and diabetic patients older than 60 years were routinely preoperatively screened for evidence of carotid artery disease. Statistical analysis Descriptive statistics were calculated for all analyzed variables. Continuous data are presented as mean±standard deviation (X±SD). Categorical variables are presented as frequencies numbers and percentages. For all analysis type I error of 5% was considered statistically significant. Categorical data were analyzed using the c2test or Fisher’s exact test. For continuous variables, comparison between groups was performed using a t-test or non parametric Mann-Whitney U test as appropriate. Multivariate (stepwise procedure) logistic regression was used to reveal predictors of prolonged rSO2 desaturation. All statistical analyses were performed using Statistica 7.1 and SAS 8.2 software. Results The pattern of rSO2 changes during the operation was similar. During the period before CPB, rSO2 desaturation rarely occurred. Most cases of desaturation were observed during CPB; immediately after initiation of CPB, after aortic cross-clamp and releasing of the clamp and during rewarming. After the termination of CPB, rSO2 usually increased to preoperative values. 18 out of 58 patients had prolonged rSO2 desaturation during procedure. Considering preoperative data, it is important to notice that there were significantly more patients with diabetes mellitus (p=0.02) in group with prolonged desaturation (Table 1). Patients in desaturation group were older (Figure 1) and had higher EuroSCORE values but this was not statistically significant. The analyses of intraoperative parameters show that the use of blood during CPB (p=0.007) and the need for inotropes (p=0.04) were significantly higher in desaturation group (Table 2). Although baseline rSO2 values were lower in desaturation group, that parameter has not reached statisti383 @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 higher in desaturation group (p<0.001) as it was the number of patients who got blood transfusion (p<0.001) to maintain adequate rSO2 (Table 2). 80 75 Stroke, coma or stupor was recognized in three patients in rSO2 desaturation group but no one in another group, which has reached statistical significance (p= 0.03). There was no statistically significant difference in the incidence of delirium between groups (Table 3). Age (years) 70 65 The duration of ICU staying was not statistically different between groups. The percentage of patients who needed hospitalization for more than seven days after the surgery was higher in desaturation group (67 vs. 42%) but this wasn’t statistically significant (p=0.09). 60 55 50 Mean Mean±SD Mean±1,96*SD 45 0 1 Desaturation status Fig. 1. Desaturation status and age. Desaturation status: 0 – without prolonged desaturation, 1 – with prolonged desaturation. Prolonged desaturation – rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min. cal significance, but the lowest values of rSO2 during operation were significantly deeper in desaturation group (p<0.001). The number of interventions was significantly In multivariate logistic regression model independent preoperative variables were age, sex, diabetes, hypertension, EuroSCORE, ejection fraction, and perioperative variables CPB time, ACC time, baseline rSO2 and ICU staying, Dichotomous variable – prolonged rSO2 desaturation was dependent variable. In the multivariate logistic regression model with the use of stepwise procedure the variables – age (p=0.04) with odds ratio (OR)=1.10 (95% confidence interval [95%CI];1.004–1.200) and diabetes mellitus (p=0.02) with OR=5.07 (95%CI;1.37–18.82) were recognized as statistically significant predictors of prolonged rSO2 desaturation during coronary artery bypass grafting. TABLE 1 PATIENTS´ CHARACTERISTICS AND PREOPERATIVE DATA Parameter Without prolonged desaturation* N=40 With prolonged desaturation* N=18 p-value Odds ratio (95%CI) Age – years 61.3±7.4 65.1±6.6 p=0.07 OR=1.10 (1.004–1.20) Female sex 8 (20) 6 (33) p=0.32 EuroSCORE – % 2.0±1.6 2.6±2.3 p=0.27 EF – % 56±10 56±10 p=0.94 8 (20) 9 (50) p=0.02** OR=5.07 (1.37–18.82) Hypertension 33 (83) 12 (67) p=0.2 IM < 1 month 4 (10) 0 p=0.3 COPB 4 (10) 3 (17) p=0.67 Comorbidity Diabetes mellitus Atrial fibrillation 2 (5) 1 (5) p=1.0 Current smoker 10 (25) 5 (28) p=1.0 History of smoking 14 (35) 3 (17) p=0.16 Medications b-blockers 37 (92) 15 (83) p=0.36 ACE inhibitors 30 (75) 12 (67) p=0.2 Calcium channel blockers 13 (32) 5 (28) p=0.72 Antilipemics 34 (85) 15 (83) p=1.0 Data are expressed as frequencies numbers (%) or X±SD (mean±standard deviation) OR – odds ratio, 95%CI – 95% confidence interval * prolonged desaturation – rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min% ** statistically significant value EF – ejection fraction, IM < 1 month – myocardial infarction within 1 month, COPB – chronic obstructive pulmonary disease 384 @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 TABLE 2 INTRAOPERATIVE PARAMETERS Parameter Without prolonged desaturation* N=40 With prolonged desaturation* N=18 p-value Number of grafts 2.4±0.6 2.5±0.5 p=0.97 CPB time – min 90±36 97±27 p=0.46 ACC time – min 62±24 68±21 p=0.35 306±69 319±52 p=0.46 Duration of operation – min CI before CPB – L/min/m2 CI after CPB – L/min/m2 Single clamp – number of patients 2.4±0.4 2.4±0.3 p=0.13 2.8±0.6 2.8±0.5 p=0.85 p=0.42 33 (83) 17 (94) Blood during CPB – ml 216±254 493±380 p=0.007** Blood after CPB – ml p=0.39 128±211 178±201 Inotropes – number of patients 15 (37) 12 (67) p=0.04** Baseline rSO2 left – % 74±6 70±9 p=0.11 Baseline rSO2 right – % 72±7 69±9 p=0.12 51.9±7 42.5±7 p<0.001** The lowest rSO2 – % Desaturation AUC – min% rSO2 AUC<20% baseline 42±42 617±997 p<0.001** rSO2 AUC<50% absolute 5±11 150±202 p<0.001** rSO2 AUC<40% absolute p<0.001** Number of interventions Blood transfusion as an intervention – number of patients 0 13±29 4.4±5.4 8.8±5.6 p<0.001** 8 (20) 13 (72) p<0.001** Data are expressed as frequencies numbers (%) or X±SD (mean±standard deviation) * prolonged desaturation – rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min% ** statistically significant values CPB – cardiopulmonary bypass, ACC time – aortic cross-clamping time, CI – cardiac index, rSO2 – regional cerebral oxygen saturation, AUC – area under the curve In this study we observed the consequences of cerebral oxygen desaturation measured by INVOS system during coronary bypass graft surgery. We introduced INVOS 5100 (Somanetics, Troy, USA) monitoring in our cardiac anesthesia department in 2009. All interventions undertaken to maintain cerebral oxygenation above threshold of 20% under the baseline value were in the range of good clinical practice. In some cases, despite very strict approach in desaturation treatment, there was no improvement in rSO2 values. Preoperative data revealed age as important predictor of prolonged rSO2 desaturation. Age is earlier identified as a major risk factor for neurologic impairment after cardiac surgery because of aberrations in cerebral autoregulation associated with reduced cerebral blood flow20,21. According to results of current study diabetes mellitus is also strong predictor of prolonged rSO2 desaturation during coronary artery bypass grafting and the prolonged rSO2 desaturation is five times more common among patients with diabetes compared to patients without diabetes. Diabetes mellitus is well-known risk factor for the development of coronary artery occlusive disease, but also one of the major predisposing factors for development of postoperative neurological impairment after coronary artery bypass graft surgery. Postoperative neurological deficits in patients with diabetes mellitus are considered to be related to the impairment of cerebral perfusion and autoregulation during cardiac surgery under cardiopulmonary bypass22–24. It is reported that CPB 35 30 25 Count Discussion and Conclusion 20 15 10 5 No 0 Yes Yes Prolonged desaturation No Diabetes mellitus Fig. 2. Desaturation status and diabetes mellitus. Prolonged desaturation – rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min%. 385 @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 TABLE 3 POSTOPERATIVE PARAMETERS Parameter Without prolonged desaturation* N=40 With prolonged desaturation* N=18 Blood in ICU – ml 205±269 182±284 p=0.78 CI – arrival in ICU – L/min/m2 2.9±0.6 2.9±0.6 p=0.96 CI – after 24 h – L/min/m2 Troponin T – arrival in ICU – ng/ml p-value 3.1±0.5 3.2±0.5 p=0.45 0.46±0.3 0.92±0.9 p=0.07 Troponin T POD 1– ng/ml 1.17±1.1 1.08±0.9 p=0.76 Troponin T POD 2– ng/ml 0.63±0.8 0.63±0.4 p=0.46 Ventilatory time – hours 10.9±9.3 15.1±12.7 p=0.13 Postoperative complications – number of patients (%) Stroke, coma, stupor 0 3 (16) p=0.03** 4 (10) 5 (27) p=0.11 Atrial fibrillation 5 (12) 6 (33) p=0.08 Respiratory insufficiency 1 (2.5) 2 (11) p=0.22 Dialysis 0 1 p=0.31 Infection 6 (15) 2 (11) p=1.0 Delirium ICU length of stay – days Hospital staying longer than 7 days – number of patients (%) 2±0.6 3.9±6.6 p=0.17 17 (42) 12 (67) p=0.09 SAPS II – arrival in ICU 18.0±6.8 20.0±4.4 p=0.53 SAPS II – ICU discharge 10.5±4 12.4±4 p=0.11 Data are expressed as frequencies numbers (%) or X±SD (mean±standard deviation) * prolonged desaturation – rSO2 AUC<20% baseline of more than 150 min% or rSO2 AUC<50% absolute of more than 50 min% ** statistically significant value ICU – intensive care unit, CI – cardiac index, POD – postoperative day, SAPS – Simplified Acute Physiology Score can alter the cerebral endothelial and nitric oxide function more extensively in diabetic patients than in non-diabetic patients. Cerebrovascular CO2 reactivity, which is very important for cerebral flow autoregulation, can be reduced in diabetic patients with consecutive impairment in cerebral oxygenation25. Miyoshi et al. reported that diabetic patients experienced cerebral desaturation during CPB, assessed by jugular bulb oxygen saturation (SjvO2), more often than non-diabetic patients26. On the contrary, regarding off-pump coronary artery bypass grafting, Oh et al. reported that cerebral oxygenation in diabetic patients was similar to that of non-diabetic patients27. Normal absolute value for rSO2 has not been established but group mean rSO2 value of 67±10 has been observed in conscious healthy volunteers and cardiac patients28. The degree of cerebral desaturation which can be followed by higher risk for postoperative complications is still not adopted. Yao et al. revealed that nadir rSO2<35% or rSO2 AUC<40% of more than 10 min% significantly increased the incidence of neuropsychological decline in patients undergoing cardiac surgery with cardiopulmonary bypass11. Slater et al. defined a prolonged rSO2 desaturation as rSO2 score greater than 50 min% below 50% saturation threshold14. These values of prolonged rSO2 desaturation were associated with increased risk of neurocognitive decline. Murkin et al. showed that 386 patients having prolonged desaturation, defined as area under the curve (AUC) below 70% of baseline for more than 150 min%, tended to have more frequently major organ morbidity and mortality compared to patients without such prolonged desaturation12. Based on these studies, we defined prolonged desaturation rSO2 as an AUC of more than 150 min% under 20% of baseline value, or rSO2 AUC of more than 50 min% under 50% of absolute value. 31% of our patients had prolonged rSO2 desaturation despite of interventions undertaken to maintain rSO2 above the threshold of 20% under the baseline value. Slater et al. in their prospective study randomized CABG patients monitored with INVOS to a blinded control group and unblinded intervention group14. They observed almost identical desaturation rates in both control (30%) and intervention group (26%) and explained the results as a poor compliance to the treatment protocol. On the contrary, Murkin et al. concluded that intervention protocol undertaken to return rSO2 to baseline, in most cases resulted in a rapid improvement of rSO2 without added risk to the patient. The overall success rate was about 80%12. Neither of these studies analyzed patients without appropriate response to interventions used to maintain rSO2 in normal range. The most frequent interventions (after repositioning the head or perfusion cannulae) to maintain rSO2 in our @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 patients were increasing MAP and FiO2, increasing arterial CO2 partial pressure – PaCO2, blood transfusion and increasing pump flow. There were significantly more interventions in desaturation group (p<0.001). The results of our study show significantly higher incidence of blood transfusion as intervention to maintain cerebral saturation in desaturation group compared to another group. The blood transfusion was assumed as intervention only in case when cerebral desaturation was followed by decreasing of hematocrit value to less than 22%, so the indication for transfusion was in accordance with recommendations of good clinical practice. Murkin et al. concluded in their study that patients undergoing cerebral oximetry were not at increased risk of receiving blood transfusion as a consequence of monitoring rSO2, because there was no difference in blood transfusion rates between intervention and control INVOS group12. Torella et al. proposed that NIRS could be used as a monitor of blood loss and they observed good correlation between blood loss and NIRS parameters29. In our study there were significantly more patients in desaturation group who needed inotropic support (p= 0.04), although »inotropes« were relatively seldom used as intervention to maintain rSO2. Inotropic support was more related to hemodynamic parameters (cardiac index less than 2 L/min/m2) and data obtained with transesophageal echocardiography (contractility, ejection fraction). This finding is in correlation with higher values of troponin T after admission in ICU in desaturation group, although statistical significance is not achieved (p=0.07). The results suggest that cerebral desaturation may be in relation with impaired cardiac function. There are only few studies about relationship between brain oximetry and cardiac function30,31. The incidence of coma, stupor and stroke in our study was significantly higher in desaturation group (p=0.03). This findings confirm the results of previous studies about significant association of cerebral desaturation with adverse neurological outcome. Goldman et al. demonstrated a decrease in the stroke rate in cardiac surgical patients after implementation of cerebral oximetry but they didn’t verify if patients in rSO2 monitored group who developed stroke, had or had not prolonged desaturation10. Statistical significance between groups for other postoperative complications was not detected in the circumstances of our study. Interventions carried out during the use of cerebral oximetry in most patients resolved decrease in rSO2, but in some patients there was no success and prolonged desaturation occurred. We presume that those patients have some problems with cerebral vascular autoregulation and may deserve more attention during surgery and early postoperative period. Some of limitations of this study are small sample size and the fact that there was a lack of rSO2 monitoring during early postoperative period when cerebral desaturation and some of complications can occur. Nevertheless, patients were treated according to same standard protocol during the postoperative period. In conclusion, although monitoring of cerebral oxygenation gives us possibility to maintain balance in oxygen supply and demand for the purpose of patients’ clinical benefit, we observed some patients who didn’t respond to our interventions. The consecutive prolonged rSO2 desaturation in our study was associated with higher incidence of neurological complications. The question is how to manage those patients. There is no data about those patients in literature and about difference in outcome between »responders« and »nonresponders« to interventions undertaken to prevent prolonged desaturation. We hope that our study will prompt other researchers to carefully analyze data based on larger number of patients – nonresponders to interventions for maintenance of rSO2. Acknowledgements The authors express their gratitude to all anesthesiology and intensive care, cardiac surgery and perfusion staff for excellent assistance and support. We specially thank to all participants of our study. REFERENCES 1. NEWMAN MF, MATHEW JP, GROCOTT HP, MACKENSEN GB, MONK T, WELSH-BOHMER KA, BLUMENTHAL JA, LASKOWITZ DT, MARK DB, Lancet, 368 (2006) 694. DOI: 10.1016/S0140-6736(06) 69254-4. — 2. ARROWSMITH JE, GROCOTT HP, REVES JG, NEWMAN MF, Br J Anaesth, 84 (2000) 378. DOI: 10.1093/oxfordjournals.bja.a013444. — 3. ROACH GW, KANCHUGER M, MANGANO CM, NEWMAN M, NUSSMEIER N, WOLMAN R, AGGARWAL A, MARSCHALL K, GRAHAM SH, LEY C, OZANNE G, MANGANO DT, N Engl J Med, 335 (1996) 1857. DOI: 10.1056/NEJM199612193352501. — 4. GUARRACINO F, Curr Opin Anaesthesiol, 21 (2008) 50. DOI: 10.1097/ACO.0b013e3282f3f499. — 5. SCHWARZ G, LITSCHER G, Eur J Anaesthesiol, 19 (2002) 543. DOI: 10.1097/00003643-200208000-00001. — 6. EDMONDS HL JR, Heart Surg Forum, 5 (2002) 225. — 7. MURKIN JM, ARANGO M, Br J Anaesth, 103 (2009) 3. DOI: 10.1093/bja/aep299. — 8. TAILLEFER MC, DENAULT AY, Can J Anesth, 52 (2005) 79. DOI: 10.1007/BF03018586. — 9. VOHRA HA, MODI A, OHRI SK, Interact CardioVasc Thorac Surg, 9 (2009) 318. DOI: 10.1510/icvts.2009.206367. — 10. GOLDMAN S, SUTTER F, FERDINAND F, TRACE C, Heatr Surg Forum, 7 (2004) 376. DOI: 10.1532/HFS98. 20041062. — 11. YAO FSF, TSENG CCA, HO CYA, LEVIN SK, ILLNER P, J Cardiothorac Vasc Anesth, 5 (2004) 552. — 12. MURKIN JM, ADAMS SJ, NOVICK RJ, QUANTZ M, BAINBRIDGE D, IGLESIAS I, CLELAND A, SCHAEFER B, IRWIN B, FOX S, Anesth Analg, 104 (2007) 51. DOI: 10.1213/01.ane.0000246814.29362.f4. — 13. HONG SW, SHIM JK, CHOI YS, KIM DH, CHANG BC, KWAK YL, Eur J Cardiothorac Surg, 33 (2008) 560. DOI: 10.1016/j.ejcts.2008.01.012. — 14. SLATER JP, GUARINO T, STACK J, VINOD K, BUSTAMI RT, BROWN JM, RODRIGUEZ AL, MAGOVERN CJ, ZAUBLER T, FREUNDLICH K, PARR GVS, Ann Thorac Surg, 87 (2009) 36. DOI: 10.1016/j.athoracsur.2008.08.070. — 15. NAUPHAL M, EL-KHATIB M, TAHA S, HAROUN-BIZRI S, ALAMEDDINE M, BARAKA A, Eur J Anaesthesiol, 24 (2007) 15. DOI: 10.1017/ S02650215060000998. — 16. DENAULT A, DESCHAMPS A, MURKIN JM, Semin Cardiothorac Vasc Anesth, 11 (2007) 274. — 17. TAGGART DP, WESTABY S, Curr Opinion in Card, 16 (2001) 271. DOI: 10.1097/ 00001573-200109000-00003. — 18. NASHEF SAM, ROQUES F, MICHEL P, GAUDUCHEAU E, LEMESHOW S, SALAMON R, THE EUROSCORE STUDY GROUP, Eur J Cardiothorac Surg, 16 (1999) 9. DOI: 10.1016/ 387 @. ^olak et al.: Cerebral Oxygen Desaturation in Coronary Surgery, Coll. Antropol. 36 (2012) 2: 381–388 S1010-7940(99)00134-7. — 19. LE GALL JR, LEMESHOW S, SAULNIER F, JAMA, 270 (1993) 2957. DOI: 10.1001/jama.270.24.2957. — 20. BRUSINO PG, REVES JG, SMITH LR, PROUGH DS, STUMP DA, MCINTYRE RW, J Thorac Cardiovasc Surg, 97 (1989) 541. — 21. SCHELL RM, KERN FH, GREELEY WJ, SCHULMAN SR, FRASCO PE, CROUGHWELL ND, NEWMAN M, REVES JG, Anesth Analg, 76 (1993) 849. — 22. LAVI S, GAITINI D, MILLOUL V, JACOB G, Am J Physiol Heart Circ Physiol, 291 (2006) 1856. DOI: 10.1152/ajpheart.00014.2006. — 23. PALLAS F, LARSON DF, Perfusion, 11 (1996) 363. DOI: 10.1177/ 026765919601100502. — 24. KADOI Y, HINOHARA H, KUNIMOTO F, SAITO S, IDE M, HIRAOKA H, KAWAHARA F, GOTO F, Stroke, 34 (2003) 2399. DOI: 10.1161/01.STR.0000090471.28672.65. — 25. SRINIVASAN AK, GRAYSON AD, FABRI BM, Ann Thorac Surg, 78 (2004) 1604. DOI: 10.1016/j.athoracsur.2004.04.080. — 26. MIYOSHI S, MORITA T, KADOI Y, GOTO F, Surg Today, 35 (2005) 530. DOI: 10.1007/s00595-004-2977-0. — 27. OH YJ, KIM JY, SHIM JK, YOO KJ, LEE JW, KWAK YJ, Circ J, 72 (2008) 1259. DOI: 10.1253/circj.72.1259. — 28. EDMONDS HL JR, GANZEL BL, AUSTIN EH III, Semin Cardiothorac Vasc Anesth, 8 (2004) 147. — 29. TORELLA F, HAYNES SL, MCCOLLUM CN, J Surg Res, 110 (2003) 217. DOI: 10.1016/S0022-4804(03)00037-4. — 30. PAQUET C, DESCHAMPS A, DENAULT AY, COUTURE P, CARRIER M, BABIN D, LEVESQUE S, PIQUETTE D, LAMBERT J, TARDIF JC, J Cardiothorac Vasc Anesth, 22 (2008) 840. DOI: 10.1053/j.jvca.2008.02.013. — 31. MORITZ S, ROCHON J, VÖLKEL S, HILKER M, HOBBHAHN J, GRAF BM, ARLT M, Eur J Anaesthesiol, 27 (2010) 542. @. ^olak University of Zagreb, Zagreb University Hospital Center, Ki{pati}eva 12, 10000 Zagreb, Croatia e-mail: zcolak2000@yahoo.com UTJECAJ DUGOTRAJNE CEREBRALNE DESATURACIJE KISIKOM NA POSLIJEOPERACIJSKI ISHOD KOD BOLESNIKA PODVRGNUTIH KIRUR[KOJ REVASKULARIZACIJI MIOKARDA SA@ETAK 58 bolesnika podvrgnutih kirur{koj revaskularizaciji miokarda uz upotrebu kardiopulmonalnog bypassa, pra}eno je bliskom infracrvenom spektroskopijom (near infrared spectroscopy- NIRS) glede promjena u regionalnoj cerebralnoj saturaciji kisikom (rSO2). U slu~aju pada rSO2 tijekom operacije za vi{e od 20% od bolesnikove bazalne vrijednosti, poduzete su standardizirane intervencije radi odr`avanja rSO2. Usprkos tim intervencijama, kod nekih bolesnika smo primijetili nemogu}nost odr`avanja rSO2 iznad te grani~ne vrijednosti. Bolesnike smo stoga radi analize podataka podijelili u dvije grupe: 1. bez dugotrajne rSO2 desaturacije; 2. s dugotrajnom rSO2 desaturacijom (povr{ina ispod krivulje >150 min% za rSO2<20% od bazalne vrijednosti te >50 min% za rSO2<50% apsolutne vrijednosti). Podaci su analizirani da bi se utvrdilo da li postoje zna~ajne razlike u kona~nom ishodu ovih dviju grupa. 18 od ukupno 58 bolesnika (31%) imalo je dugotrajnu rSO2 desaturaciju za vrijeme operacije. U grupi bolesnika sa dugotrajnom rSO2 desaturacijom utvr|en je zna~ajno ve}i broj onih koji boluju od dijabetesa (p=0,02). Ujedno, u grupi bolesnika sa dugotrajnom rSO2 desaturacijom, podaci vezani uz operaciju pokazuju zna~ajno ve}u potro{nju krvi tijekom kardiopulmonalnog bypassa (p=0,007) i ve}u potrebu za inotropnom terapijom (p=0,04). Tri su bolesnika iz navedene grupe imala mo`dani udar, komu ili stupor, dok u drugoj grupi niti jedan (p=0,03). Analizom podataka metodom logisti~ke regresije utv|eno je da su dijabetes i godine starosti zna~ajni prediktori za dugotrajnu rSO2 desaturaciju. Zaklju~ili smo da je dugotrajna rSO2 desaturacija tijekom operacije povezana sa zna~ajno lo{ijim neurolo{kim ishodom kod bolesnika kod kojih primjena standardiziranih postupaka nije dovela do mogu}nosti odr`avanja rSO2 iznad spomenutih vrijednosti. 388