Assessing Baseline Cortisol Levels in Patients Admitted With Septic Shock in Intensive Care Unit

Overview

Critical illness-related corticosteroid insufficiency (CIRCI), a term coined since 2008 by Society of Critical Care Medicine (SCCM), and is characterized by inflammation resulting from inadequate intracellular glucocorticoid-mediated anti-inflammatory activity leading to increased morbidity and mortality in Intensive Care Unit (ICU) patients.1 Severe Sepsis with shock is a common reason for admission to ICU/hospital and may require ionotropic support.2 The current guidelines from SCCM in 2017 suggest using either random cortisol of \< 10 ug/dL (\<276 nmol/L) or change in cortisol at 60 min after cosyntropin (250 µg) administration from baseline cortisol of \<9 µg/dl (\<248 nmol/L) to assess of presence of CRCI and recommend use of hydrocortisone in these patients.3 There have been studies done to look at baseline cortisol in patient with severe pneumonia requiring ICU and they have found cortisol level of \< 15 ug/dl (\<414 nmol/L) can predict CIRCI.4 However, there is no study on assessment of baseline random cortisol levels in patients with septic shock in our local population. The current guidance from Surviving Sepsis campaign suggests a more clinical approach of adding IV corticosteroids only if there is ongoing requirement for vasopressors, which is a new change in contrast to 2016 guidelines.5 This study aims to look the available mean baseline cortisol in these patients to create a reference data for local population.

This is the first study to ascertain the baseline cortisol in patients with septic shock in local Emirati population. This will help provide more evidence towards diagnosing/ruling out CIRCI reliably. At present there is no consensus data globally on expected random cortisol levels in patients having septic shock and no data expected random cortisol levels in UAE/Gulf region in these patients which makes diagnosis of CIRCI difficult. Objective: Primary: The aim of this study is to estimate the mean random cortisol levels for patient admitted to ICU with septic shock requiring ionotropic support (who did not require any hydrocortisone during ICU stay) at Tawam hospital, United Arab Emirates to obtain a reference for our local population. Secondary: 1. To estimate the mean random cortisol levels for patient admitted to ICU with septic shock requiring ionotropic support (who also required hydrocortisone) during ICU stay to obtain a reference for our local population with CIRCI. 2. To assess the outcome of patients treated as CIRCI. 3. To assess the duration of hydrocortisone-use during ICU/hospital. 4. To see if the steroids were continued post discharge from ICU/hospital. 5. To compare biochemical profile of the patients treated with suspected CRICI in comparison to patients not requiring any steroids. Study Type: Retrospective observational study in which chart review of records of all patients admitted to ICU with diagnosis of severe sepsis will be done Study location ICU, Tawam Hospital, Al Ain Study duration The retrospective chart review will be done for patient admitted into ICU between 01/06/2012 to 01/06/2022 Sample Size: Investigators plan to include minimum 400 patients based on study power calculation of minimum 386 patients. INCLUSION CRITERIA: * Age \> 18; both male or female * Any nationality * Patients admitted to ICU with diagnosis of severe sepsis / septic shock bearing ICD-10 codes R65.20 or R65.21, AND * who required ionotropic support within 24-48 hours of admission, AND * had a valid cortisol measurement done during this admission/encounter EXCLUSION CRITERIA: * Patients with additional diagnosis of polytrauma or haemorrhagic shock or other causes of shock other than septic shock. * Patient who received steroids within 24 hours prior to sending cortisol sample * Patient with known adrenal insufficiency. * Patient on oral/inhaled steroids (home medications) prior to this admission for any reason (equivalent to 7.5 mg prednisolone or above) in the last 6 weeks Sample Collection Process: This is a retrospective observational study in which records of all patients admitted to ICU with diagnosis of severe sepsis / septic shock bearing ICD-10 codes R65.20 or R65.21, who required ionotropic support within 24 hours of admission and had a valid cortisol measurement done during this admission/encounter, will be reviewed. The biochemical workup (Full blood count, liver function and renal function test), length of stay in ITU and outcome of the patient including use of steroids during hospital or discharge will be obtained. Consenting Process: It will an observational study (no intervention) with retrospective analysis of patient records on SalamTek. No patient sensitive or identifiable information will be kept on records. No individual patient consenting required. Data Handling/confidentiality \& Statistical Analysis: The data will be collected on an excel sheet in secure CMHS network with no identifiable patient information. This excel sheet will be transferred onto SPSS package for Microsoft Windows version 12.0. The variables of demography (age, sex), history of present disease (duration, severity), positive signs and routine investigations will be presented as simple descriptive statistics. The means and standard deviations of the numerical data such as age, Haemoglobin, creatinine, etc will be calculated. The mean first cortisol level (prior to any corticosteroid treatment) for patients without diagnosis of CRICI (who responded to ionotropes) will be obtained to see the baseline cortisol level in these patients. For patient treated as CIRCI, the mean cortisol level will be obtained and comparison analysis via receiver operator curves will be obtained to see if a value of random cortisol can predict absence of CIRCI with at least 95% sensitivity.