Dr Afzal Azim,Dr Sukhen Samanta, Dr A. K. Baronia
LUNG ULTRASOUND AS A DIAGNOSTIC TOOL FOR VENTILATOR ASSOCIATED PNEUMONIA: A PROSPECTIVE OBSERVATIONAL STUDY FROM A TERTIARY CARE CENTRE
- Department of Critical Care Medicine, SGPGIMS, Lucknow,India
Objectives:
Early diagnosis of ventilator associated pneumonia (VAP) remains elusive for most of the clinicians. We explored the utility of bedside lung ultrasonography (LUS) based on a scoring system to diagnose VAP.
Methods:
After ethical clearance and informed consent this prospective single centre observational study was conducted from January 2015 to June 2016. 110 patients with suspected VAP were enrolled. We investigated the diagnostic accuracy of LUS using the findings of subpleural consolidation, lobar consolidation, and dynamic air bronchogram. Quantitative mini broncho-alveolar lavage (mBAL) fluid was taken for microbiological diagnosis. We designed a sono-pulmonary infection score (SPIS) based on LUS, clinical and microbiology parameters and used it as a substitute for clinical pulmonary infection score (CPIS). Descriptive data was summarized as mean (SD), median (interquartile range), or number (%). Two groups were compared using Mann – Whitney test and Fisher exact test.
Results:
Prevalence of VAP was 72.7% amongst the suspected patients. Overall LUS performance for diagnosis of VAP (microbiologically confirmed) had a sensitivity, specificity, positive/negative predictive value and positive/negative likelihood ratios of 91.3%, 70%, 89%, 75%, 3 and 0.1 respectively. Adding microbiology to LUS improved diagnostic accuracy with sensitivity, specificity, positive/negative predictive value and positive/negative likelihood ratios of 95%, 90%, 96.2%, 87%, 9.5 and 0.06 respectively. The area under curve for CPIS and modified CPIS (CPIS-microscopy & CPIS-culture) and SPIS and modified SPIS were 0.617, 0.689, 0.801, 0.808, 0.815 and 0.913 respectively.
Conclusions:
Lung ultrasound is valuable tool for early diagnosis of VAP. Larger trials are warranted to establish its utility.
Wan Fadzlina Wan Muhd Shukeri, Azrina Md. Ralib, Mohd Basri Mat-Nor
THE DIAGNOSTIC VALUE OF MODELLED INSULIN SENSITIVITY IN SEPSIS
- International Islamic University of Malaysia,
- Universiti Sains Malaysia
Objective:
Low insulin sensitivity (SI) and sepsis are strongly linked, but the effectiveness of SI as a diagnostic test of sepsis is largely unknown. We aim to study the diagnostic value of modelled SI in sepsis in a mixed cohort of diabetic and non-diabetic critically ill patients.
Methods:
In this cross-sectional study, we analysed SI levels in septic (n = 45) and non-septic (n = 41) patients upon their ICU admission. The SI levels were derived by fitting blood glucose levels, insulin infusion and glucose input rates into the Intensive Control of Insulin-Nutrition-Glucose model.
Results:
In the overall cohort, SI levels were significantly lower in patients with sepsis than those without sepsis (0.560 ± 0.676 versus 1.097 ± 1.473 x 10-4 L/mU/min, P = 0.037). Analysis of the area under the curve (AUC) revealed that modelled SI was a poor diagnostic test of sepsis [AUC 0.588 (95% CI, 0.477-0.693)]. The result from multivariate logistic regression analysis showed that modelled SI could not independently predict sepsis. In a separate analysis among the non-diabetic cohort (n = 19), modelled SI was found to be useful as a diagnostic test of sepsis [AUC 0.911 (95% CI, 0.690-0.992)] with a cut-off of 0.880x 10-4 L/mU/min, sensitivity of 90% and specificity of 78%. There was no significant difference in SI levels among the diabetic cohort.
Conclusion:
Presence of sepsis significantly reduced SI but a low SI could not independently predict sepsis in a mixed cohort of diabetic and non-diabetic critically ill patients. Low SI can equally mark the presence of sepsis, other severe conditions or the result of treatment effects which are indicated in sepsis. Only when applied to the non-diabetic cohort that modelled SI was useful as a diagnostic test of sepsis in the critically ill patients.
Mohd Zulfakar Mazlan, Tengku Abdul Kadir Tengku Zainal Abidin, Saedah Ali, Dr Shubashini Thevadas.Mahamarowi Omar
THE EFFECTS OF PASSIVE LEG RAISING TEST AND FLUID CHALLENGE ON HAEMODYNAMIC PARAMETERS TO ASSESS FLUID RESPONSIVENESS IN INTENSIVE CARE UNIT
- Department of Anesthesiology and Intensive Care, School of Medical Sciences, Health Campus, Universiti Sains Malaysia, 16150, Kubang Kerian, Kelantan, Malaysia.
- Department of Anesthesiology and Intensive Care,Hospital Universiti Sains Malaysia, 16150, Kubang Kerian, Kelantan, Malaysia.
Background:
Passive leg raising (PLR) test and fluid challenge (FC) are two reliable techniques to determine preload responsiveness in Intensive Care Unit (ICU). We evaluated the correlation of cardiac output(CO), heart rate (HR), mean arterial pressure (MAP), stroke volume variation (SVV), stroke volume (SV) during (PLR) and after (FC). The FloTrac/Vigileo system (Edwards Lifesciences, Irvine, CA, USA) is a minimally invasive device used to measure CO in this study.
Methods:
This is a prospective study in general intensive care unit (ICU) and surgical ICU in a university hospital. Thirty-seven septic patients who were mechanically ventilated, deeply sedated, equipped with theFloTrac/Vigileo and decided for fluid bolus were recruited in this study. Recordings of heart rate (HR), mean arterial pressure (MAP), SVV, stroke volume (SV), and CO were obtained at the baseline, after PLR, baseline FC and after FC. Fluid responder was defined as changes of CO (Δco) more than 13%.
Results:
Out of 37 patients, 31 (84%) were in septic shock and receiving vasoactive agents. The median Acute Physiology and Chronic Health II (APACHE II) and Sequential Organ Failure Assessment (SOFA) score were 24 (IQR=13) and 14 (IQR=5) respectively. The haemodynamic changes (pre and post-test) in heart rate (HR) and mean arterial pressure (MAP) were significant in FC group (P=0.013 and <0.001 respectively) but not significant in PLR group (P=0.190 and 0.620). However the changes of SV, SVV and CO were significant for both groups.
Conclusion:
In haemodynamically unstable patients such with higher (APACHE II) and (SOFA) score, the routine use of HR and MAP changes post PLR to assess fluid responsiveness were not accurate as compared to FC methods. However, both groups reveals the use of changes in SV, SVV and CO were useful in assessing fluid responsiveness.