articleRelative adrenal insufficiency and hemodynamic status in cardiopulmonary bypass surgery patients . A prospective cohort study

Background: The objectives of this study were to determine the risk factors for relative adrenal insufficiency in cardiopulmonary bypass patients and the impact on postoperative vasopressor requirements. Methods: Prospective cohort study on cardiopulmonary bypass patients who received etomidate or not during anesthetic induction. Relative adrenal insufficiency was defined as a rise in serum cortisol ≤ 9 μg/dl after the administration of 250 μg of consyntropin. Plasma cortisol levels were measured preoperatively, immediately before, 30, 60, and 90 minutes after the administration of cosyntropin, and at 24 hours after surgery. Results: 120 elective cardiopulmonary bypass patients were included. Relative adrenal insufficiency (Δcortisol ≤9 μg/ dl) incidence was 77.5%. 78 patients received etomidate and 69 (88%) of them developed relative adrenal insufficiency, (P < 0.001). Controlling for clinical characteristics with a propensity analysis, etomidate was the only independent risk factor associated with relative adrenal insufficiency (OR 6.55, CI 95%: 2.47-17.4; P < 0.001). Relative adrenal insufficiency patients showed more vasopressor requirements just after surgery (P = 0.04), and at 4 hours after surgery (P = 0.01). Pre and post-test plasma cortisol levels were inversely associated with maximum norepinephrine dose (ρ = -0.22, P = 0.02; ρ = -0.18, P = 0.05; ρ = -0.21, P = 0.02; and ρ = -0.22, P = 0.02, respectively). Conclusions: Relative adrenal insufficiency in elective cardiopulmonary bypass patients may induce postoperative vasopressor dependency. Use of etomidate in these patients is a modifiable risk factor for the development of relative adrenal insufficiency that should be avoided. Background Hypothalamic-pituitary-adrenal axis activation is an essential component of the general adaptation to illness and stress and contributes to the maintenance of cellular and organ homeostasis. Relative adrenal insufficiency (RAI) is frequently diagnosed in critically ill patients [13], and its presence is related to poorer prognosis in patients with sepsis. This has led to recommendations for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients [4]. However, the clinical impact and risk factors for RAI have not been clearly determined in cardiopulmonary bypass (CPB) patients. We hypothesized that the appearance of RAI could contribute to more complicated postoperative management in critically ill patients, increasing the use of vasoactive drugs. We aimed to assess risk factors for RAI in patients undergoing CPB, as well as their impact on postoperative vasopressor requirements. Methods Study design and patients A prospective cohort study was performed from January to July 2007 to determine the incidence and identify risk factors associated with the development of postoperative RAI. We included 120 patients who underwent elective cardiac surgery with cardiopulmonary bypass (CPB). To avoid the confounding effect of circadian rhythm on hormone levels, all operations were performed in the morn* Correspondence: joseluis.iribarren@gmail.com 1 Critical Care Department. Hospital Universitario de Canarias. Ofra s/n, La Cuesta. 38320 La Laguna. Tenerife. España Full list of author information is available at the end of the article BioMed Central © 2010 Iribarren et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Iribarren et al. Journal of Cardiothoracic Surgery 2010, 5:26 http://www.cardiothoracicsurgery.org/content/5/1/26 Page 2 of 7 ing, with general anesthesia induced between 8:30 and 9:00 am. Exclusion criteria were: history of adrenal disease, endocarditis, myocardial infarction, preoperative fever or signs of infection, surgery without CPB, emergency operations and corticoid-dependency. Postoperative care took place in a 24-bed polyvalent Critical Care Unit of University Hospital of the Canary Islands (Tenerife, Spain). Local institutional ethics committee approval was given for the study protocol, and informed consent was obtained from all patients before. This study was conducted in accordance with the provisions of the Declaration of Helsinki. Definition of Relative Adrenal Insufficiency (RAI) and Corticotropin test RAI was defined as a rise in serum cortisol ≤9 μg/dl after the administration of 250 μg of corticotropin[4]. All patients underwent a 250 μg corticotropin test (Synacthene®; Novartis Pharma Stein AG, Stein, Switzerland) within the first four hours after surgery. Cortisol levels were measured before the test, at 30, 60 and 90 minutes after the test and finally at 24 hours after surgery. The analysis of serum cortisol was performed by radioimmunoassay (Immulite®; DPC Diagnostic Products, Los Angeles, CA, USA). Perioperative management Anesthesia was induced and maintained by use of a standarized protocol with midazolam (0.1 mg/kg/h) combined with fentanyl (2-5 μg/kg/h) and cis-atracurium (0.06-0.18 mg/kg/h). Etomidate, a short acting intravenous anaesthetic used for the induction of general anaesthesia, was administered according to anesthetist criteria using a dosage of 0.3 mg/kg. Systemic heparinization, CPB, cardioplegic arrest and transfusion policy were performed as previously described[5]. Fluid management was carried out to achieve 8 to 12 mm Hg of central venous pressure or 12 to 15 mmHg of pulmonary artery occlusion pressure at zero positive end-expiratory pressure by infusions of crystalloids and colloids. Catecholamine support, when necessary, was used as follows: Norepinephrine was titrated to achieve a mean arterial pressure of greater or equal to 70 mmHg, and dobutamine was titrated to achieve a cardiac index of greater or equal to 2.5 L/minute per square meter. Amines were tapered off in steps of 0.02 and 1 μg/kg per minute, respectively. Data collection The data collected included demographic variables, comorbidity (renal failure defined as serum creatinine >1.5 mg/dl), type of surgery and postoperative course, including relative adrenal insufficiency, norepinephrine use, ICU stay and mortality. On admission to intensive care, and after 4 and 24 hours during the postoperative period, hemodynamic data were collected using a SwanGanz catheter (Edwards Lifesciences LLC Irvine, CA USA). Surgical risk was calculated using the Parsonnet scale. Statistical analysis Assuming an a priori 60% presentation of the event, with an accuracy of 10% in the estimate and using an asymptotic normal 95% CI, the study required the inclusion of 93 patients. In order to adjust for several confounder variables in the regression analysis, we increased the sample size to 120. Quantitative variables are reported as mean and standard deviation, or median and interquartile range as appropriate (intensive care unit length of stay). Nominal variables are reported as frequencies and percentages. Assumption of normality was tested with Kolmogorov-Smirnov test and homocedasticity with Levene test. Comparisons between groups, (patients with and without RAI) were performed using Pearson's chisquared test or Fischer's exact test for nominal variables, and the Student's t-test or the Mann-Whitney's U test for continuous variables, as appropriate. Propensity score analysis was performed using backward binomial logistic regression analysis. The dependent variable was use of etomidate, and the independent variables were sex, age, beta-blocker treatment, diabetes, renal failure, type of intervention and Parsonnet score. Theses scores were used in a second backward logistic regression analysis. In this analysis, the dependent variable was RAI, and the independent variables were all differences in perioperative variables with a P value < 0.15, preoperative cholesterol levels, and protein levels on arrival, together with the propensity score. Bivariate associations were assessed with Spearman's rank correlation coefficient. A P value of less than 0.05 was considered statistically significant. SPSS 15 (SPSS Inc. Chicago, IL. USA) was used. Results One hundred and twenty from 137 consecutive eligible patients were included. Seventeen of them met criteria for exclusion (8 off-pump, 2 surgical emergencies, 2 with endocarditis, 5 corticoid-dependency), as shown in Figure 1. Demographic variables, comorbidity, medical treatments, perioperative parameters and surgical procedures of the two groups are shown in Table 1. Surgical procedures were: 65 (54.2%) coronary by-pass grafting, 39 (32.5%) valvular replacement, 12 (10%) combined surgery and 4 (3.3%) other procedures. RAI was observed in 93 (77.5%) of the whole patient sample. 78 patients received etomidate and 69 (88%) of them developed RAI, (P < 0.001). Logistic regression analysis including propensity score showed that the use of etomidate was significantly associated with the presence of RAI (OR 6.55, CI Iribarren et al. Journal of Cardiothoracic Surgery 2010, 5:26 http://www.cardiothoracicsurgery.org/content/5/1/26 Page 3 of 7 95%: 2.47-17.4; P < 0.001) after adjusting for preoperative cholesterol levels and protein levels on admission, as shown in Table 2. The exposed attributable fraction due to etomidate was 35% (95%CI: 15-51%). Significantly lower cortisol levels were observed within the 4 h postoperative period (pretest) and at 30, 60, and 90 min post-test in patients who received etomidate as compared with those who did not (Figure 2). Mean arterial pressure (MAP), systemic vascular resistance index (SVRI), systolic volume index(SVI), mixed venous saturation(SvO2) and lactic acid were similar in both groups, although RAI patients required a higher dose of vasoactive drugs on admission to the critical care unit and during the postoperative period (4 h) (Figure 3). Likewise, a tendency to