Physical examination, central venous pressure, and chest radiography for the prediction of transpulmonary thermodilution–derived hemodynamic parameters in critically ill patients: A prospective trial

Hemodynamic assessment in the contemporary intensive care unit: a review of circulatory monitoring devices

The assessment of the circulating volume and efficiency of tissue perfusion is necessary in the management of critically ill patients. The controversy surrounding pulmonary artery catheterization has led to a new wave of minimally invasive hemodynamic monitoring technologies, including echocardiographic and Doppler imaging, pulse wave analysis, and bioimpedance. This article reviews the principles, advantages, and limitations of these technologies and the clinical contexts in which they may be clinically useful.

Indexation of cardiac output to biometric parameters in critically ill patients: A systematic analysis of a transpulmonary thermodilution-derived database

PURPOSE

Cardiac output (CO) (liters per minute) is usually normalized (ie, indexed) to the patient's body surface area (BSA) resulting in the hemodynamic variable cardiac index (CI) (liters per minute per square meter). We aimed (1) to evaluate the impact of different body weight-based CO indexations on the resulting CI values and (2) to identify biometric parameters independently associated with CO in critically ill patients.

MATERIALS AND METHODS

The study is an analysis of a database containing transpulmonary thermodilution-derived hemodynamic variables of 234 medical intensive care unit patients.

RESULTS

Cardiac index indexed to actual BSA was statistically significantly lower compared with CI indexed to predicted BSA in the totality of patients and in the subgroups of patients with body mass index greater than or equal to 25 kg/m(2) but less than 30 kg/m(2) and body mass index greater than or equal to 30 kg/m(2) (with a statistically significant difference in the proportion of low and high CI measurements). Multivariate analysis of the first CO measurement of each patient demonstrated that CO was independently associated with age (P < .001), height (P = .001), and actual body weight (BWact) (P = .030). Multivariate analysis of the mean of the patients' CO measurements confirmed age (P < .001), height (P = .001), and BWact (P < .001) as biometric factors independently associated with CO. Age was identified as the most important factor with each year of age decreasing CO by 66 mL/min (95% confidence interval, 47-86 mL/min).

CONCLUSIONS

The indexation of CO to BSA is highly dependent on the body weight estimation formula used to calculate BSA. Cardiac output is independently associated with the biometric factors age, height, and BWact. These factors might be considered for indexation of CO.

Cardiac preload responsiveness in children with cardiovascular dysfunction or dilated cardiomyopathy: a multicenter observational study

OBJECTIVES

To characterize cardiac preload responsiveness in pediatric patients with cardiovascular dysfunction and dilated cardiomyopathy using global end-diastolic volume index, stroke volume index, cardiac index, and extravascular lung water index.

DESIGN

Prospective multicenter observational study.

SETTING

Medical/surgical PICUs of seven Spanish University Medical Centers.

PATIENTS

Seventy-five pediatric patients (42 male, 33 female), median age 36 months (range, 1-207 mo), were divided into three groups: normal cardiovascular status, cardiovascular dysfunction, and dilated cardiomyopathy.

INTERVENTIONS

All patients received hemodynamic monitoring with PiCCO2 (Pulsion Medical System SE, Munich, Germany). We evaluated 598 transpulmonary thermodilution sets of measurements. In 40 patients, stroke volume index, cardiac index, and global end-diastolic volume index were measured before and after 66 fluid challenges and loadings to test fluid responsiveness at different preload levels.

MEASUREMENTS AND MAIN RESULTS

Global end-diastolic volume versus predicted body surface area exhibits a power-law relationship: Global end-diastolic volume = 488.8·predicted body surface area (r = 0.93). Four levels of cardiac preload were established from the resulting "normal" global end-diastolic volume index (= 488.8·predicted body surface area). Stroke volume index and cardiac index versus global end-diastolic volume index/normal global end-diastolic volume index built using a linear mixed model analysis emulated Frank-Starling curves: in cardiovascular dysfunction group, stroke volume index (geometric mean [95% CI]) was 27 mL/m (24-31 mL/m) at "≤ 0.67 times normal global end-diastolic volume index," 37 mL/m (35-40 mL/m) at "> 0.67 ≤ 1.33 times normal global end-diastolic volume index" (Δ stroke volume index = 35%; p < 0.0001; area under the receiver-operating characteristic curve = 75%), 45 mL/ m (41-49 mL/m) at "> 1.33 ≤ 1.51 times normal global end-diastolic volume index" (Δ stroke volume index = 21%; p < 0.0001; area under the receiver-operating characteristic curve = 73%), and 47 mL/m (43-51 mL/m) at "> 1.51 times normal global end-diastolic volume index" (Δ stroke volume index = 4%; p = 1; area under the receiver-operating characteristic curve = 54%). In dilated cardiomyopathy group, stroke volume index was 21 mL/m (17-26 mL/m) at "> 0.67 ≤ 1.33 times normal global end-diastolic volume index," 27 mL/m (21-34 mL/ m) at "> 1.33 ≤ 1.51 times normal global end-diastolic volume index" (Δ stroke volume index = 29%; p = 0.005; area under the receiver-operating characteristic curve = 64%), and 25 mL/m (20-32 mL/m) at "> 1.51 times normal global end-diastolic volume index" (Δ stroke volume index = -8%; p = 1; area under the receiver-operating characteristic curve = 54%).

CONCLUSIONS

This study provides "normal" values for global end-diastolic volume index and limits of cardiac preload responsiveness in pediatric patients with cardiovascular dysfunction and dilated cardiomyopathy: 1.33 times normal global end-diastolic volume index represents the upper limit of patent cardiac preload responsiveness, with the highest expected responsiveness being below 0.67 times normal global end-diastolic volume index. The maximum response of the Frank-Starling relationship and therefore the level of no additional preload reserve is 1.33 to 1.51 times normal global end-diastolic volume index. Above 1.51 times normal global end-diastolic volume index preload responsiveness is unlikely, and the risk of pulmonary edema is maximal.

The fluid management of adults with severe malaria

Fluid resuscitation has long been considered a key intervention in the treatment of adults with severe falciparum malaria. Profound hypovolemia is common in these patients and has the potential to exacerbate the acidosis and acute kidney injury that are independent predictors of death. However, new microvascular imaging techniques have shown that disease severity correlates more strongly with obstruction of the microcirculation by parasitized erythrocytes - a process termed sequestration. Fluid loading has little effect on sequestration and increases the risk of complications, particularly pulmonary edema, a condition that can develop suddenly and unpredictably and that is frequently fatal in this population. Accordingly, even if a patient is clinically hypovolemic, if there is an adequate blood pressure and urine output, there may be little advantage in infusing intravenous fluid beyond a maintenance rate of 1 to 2 mL/kg per hour. The optimal agent for fluid resuscitation remains uncertain; significant anemia requires blood transfusion, but colloid solutions may be associated with harm and should be avoided. The preferred crystalloid is unclear, although the use of balanced solutions requires investigation. There are fewer data to guide the fluid management of severe vivax and knowlesi malaria, although a similar conservative strategy would appear prudent.

Guiding principles of fluid and volume therapy

Fluid therapy is a core concept in the management of perioperative and critically ill patients for maintenance of intravascular volume and organ perfusion. Recent evidence regarding the vascular barrier and its role in terms of vascular leakage has led to a new concept for fluid administration. The choice of fluid used should be based on the fluid composition and the underlying pathophysiology of the patient. Avoidance of both hypo- and hypervolaemia is essential when treating circulatory failure. In daily practice, the assessment of individual thresholds in order to optimize cardiac preload and avoid hypovolaemia or deleterious fluid overload remains a challenge. Liberal versus restrictive fluid management has been challenged by recent evidence, and the ideal approach appears to be goal-directed fluid therapy.

Effects of paracentesis on hemodynamic parameters and respiratory function in critically ill patients

BACKGROUND

Ascites is a major and common complication of liver cirrhosis. Large or refractory ascites frequently necessitates paracentesis. The aim of our study was to investigate the effects of paracentesis on hemodynamic and respiratory parameters in critically ill patients.

METHODS

Observational study comparing hemodynamic and respiratory parameters before and after paracentesis in 50 critically ill patients with advanced hemodynamic monitoring. 28/50 (56%) required mechanical ventilation.Descriptive statistics are presented as mean ± standard deviation for normally distributed data and median, range, and interquartile range (IQR) for non-normally distributed data. Comparisons of hemodynamic and respiratory parameters before and after paracentesis were performed by Wilcoxon signed-rank tests. Bivariate relations were assessed by Spearman's correlation coefficient and univariate regression analyses.

RESULTS

Median amount of ascites removed was 5.99 L (IQR, 3.33-7.68 L). There were no statistically significant changes in hemodynamic parameters except a decrease in mean arterial pressure (-7 mm Hg; p = 0.041) and in systemic vascular resistance index (-116 dyne·sec/cm5/m2; p = 0.016) when measured 2 hours after paracentesis. In all patients, oxygenation ratio (PaO2/FiO2; median, 220 mmHg; IQR, 161-329 mmHg) increased significantly when measured immediately (+58 mmHg; p = 0.001), 2 hours (+9 mmHg; p = 0.004), and 6 hours (+6 mmHg); p = 0.050) after paracentesis. In mechanically ventilated patients, lung injury score (cumulative points without x-ray; median, 6; IQR, 4-7) significantly improved immediately (5; IQR, 4-6; p < 0.001), 2 hours (5; IQR, 4-7; p = 0.003), and 6 hours (6; IQR 4-6; p = 0.012) after paracentesis.

CONCLUSION

Paracentesis in critically ill patients is safe regarding circulatory function and is related to immediate and sustained improvement of respiratory function.

Haemodynamic monitoring using arterial waveform analysis

PURPOSE OF REVIEW

To describe the theory behind arterial waveform analysis, the different variables that may be obtained using this method, reliability of measurements and their clinical relevance. Areas for future research are identified.

RECENT FINDINGS

The precision of cardiac output (CO) measurements varies considerably and deteriorates during haemodynamic instability. Significant device-to-device differences exist. Nevertheless, most are sufficiently accurate for tracking changes in CO. Targeted intervention guided by haemodynamic monitoring reduces mortality and morbidity in high-risk surgical patients. Dynamic changes in variables such as systolic pulse variation, pulse pressure variation (PPV) and stroke volume variation (SVV) may be useful for evaluating fluid responsiveness, although important caveats exist. Newer indices such as PPV : SVV ratio may be useful in identifying preload and vasopressor-dependent patients. Peripheral arterial dP/dt has not been validated in critically ill patients and requires further investigation.

SUMMARY

Despite significant limitations in measurement accuracy and inter-device differences, arterial waveform analysis is a potentially useful tool for monitoring the central circulation in critically ill patients. Future studies investigating the effects of haemodynamic management guided by arterial waveform variables in critically ill patients are urgently needed. The evaluation of cardiopulmonary interactions and usefulness of dP/dt are other areas that require further investigation.

The reliability of the physical examination to guide fluid therapy in adults with severe falciparum malaria: an observational study

Background

Adults with severe malaria frequently require intravenous fluid therapy to restore their circulating volume. However, fluid must be delivered judiciously as both under- and over-hydration increase the risk of complications and, potentially, death. As most patients will be cared for in a resource-poor setting, management guidelines necessarily recommend that physical examination should guide fluid resuscitation. However, the reliability of this strategy is uncertain.

Methods

To determine the ability of physical examination to identify hypovolaemia, volume responsiveness, and pulmonary oedema, clinical signs and invasive measures of volume status were collected independently during an observational study of 28 adults with severe malaria.

Results

The physical examination defined volume status poorly. Jugular venous pressure (JVP) did not correlate with intravascular volume as determined by global end diastolic volume index (GEDVI; r_s = 0.07, p = 0.19), neither did dry mucous membranes (p = 0.85), or dry axillae (p = 0.09). GEDVI was actually higher in patients with decreased tissue turgor (p < 0.001). Poor capillary return correlated with GEDVI, but was present infrequently (7% of observations) and, therefore, insensitive. Mean arterial pressure (MAP) correlated with GEDVI (r_s = 0.16, p = 0.002), but even before resuscitation patients with a low GEDVI had a preserved MAP. Anuria on admission was unrelated to GEDVI and although liberal fluid resuscitation led to a median hourly urine output of 100 ml in 19 patients who were not anuric on admission, four (21%) developed clinical pulmonary oedema subsequently. MAP was unrelated to volume responsiveness (p = 0.71), while a low JVP, dry mucous membranes, dry axillae, increased tissue turgor, prolonged capillary refill, and tachycardia all had a positive predictive value for volume responsiveness of ≤50%. Extravascular lung water ≥11 ml/kg indicating pulmonary oedema was present on 99 of the 353 times that it was assessed during the study, but was identified on less than half these occasions by tachypnoea, chest auscultation, or an elevated JVP. A clear chest on auscultation and a respiratory rate <30 breaths/minute could exclude pulmonary oedema on 82% and 72% of occasions respectively.

Conclusions

Findings on physical examination correlate poorly with true volume status in adults with severe malaria and must be used with caution to guide fluid therapy.

Trial registration

Clinicaltrials.gov identifier: NCT00692627

Comparison of thermodilution measured extravascular lung water with chest radiographic assessment of pulmonary oedema in patients with acute lung injury

Background

Acute lung injury and the acute respiratory distress syndrome (ALI/ARDS) are characterized by pulmonary oedema, measured as extravascular lung water (EVLW). The chest radiograph (CXR) can potentially estimate the quantity of lung oedema while the transpulmonary thermodilution method measures the amount of EVLW. This study was designed to determine whether EVLW as estimated by a CXR score predicts EVLW measured by the thermodilution method and whether changes in EVLW by either approach predict mortality in ALI/ARDS.

Methods

Clinical data were collected within 48 hours of ALI/ARDS diagnosis and daily up to 14 days on 59 patients with ALI/ARDS. Two clinicians scored each CXR for the degree of pulmonary oedema, using a validated method. EVLW indexed to body weight was measured using the single indicator transpulmonary thermodilution technique.

Results

The CXR score had a modest, positive correlation with the EVLWI measurements (r = 0.35, p < 0.001). There was a 1.6 ml/kg increase in EVLWI per 10-point increase in the CXR score (p < 0.001, 95% confidence interval 0.92-2.35). The sensitivity of a high CXR score for predicting a high EVLWI was 93%; similarly the negative predictive value was high at 94%; the specificity (51%) and positive predictive value (50%) were lower. The CXR scores did not predict mortality but the EVLW thermodilution did predict mortality.

Conclusion

EVLW measured by CXR was modestly correlated with thermodilution measured EVLW. Unlike CXR findings, transpulmonary thermodilution EVLWI measurements over time predicted mortality in patients with ALI/ARDS.

How to perform indexing of extravascular lung water: a validation study

BACKGROUND

Extravascular lung water is a quantitative marker of the amount of fluid in the thoracic cavity besides the vasculature. Indexing to both predicted and actual body weight have been proposed to compare different individuals and provide a uniform range of normal.

OBJECTIVE

We explored extravascular lung water measured by single-indicator transpulmonary thermodilution in a large cohort of patients without cardiopulmonary instability, in order to evaluate current and alternative indexing methods.

DESIGN

Prospective, observational.

SETTING

Neurosurgical ICU in a tertiary referral academic teaching hospital.

PATIENTS

One hundred and one consecutive patients requiring elective brain tumor surgery and postoperative ICU surveillance.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Indexed to predicted body weight, females had a mean extravascular lung water of 9.1 (SD=3.1, range: 5-23) mL/kg and males of 8.0 (SD=2.0, range: 4-19) mL/kg (p<0.001). Values indexed to predicted body weight were inversely correlated with the patient's height (p<0.001). Indexed to the traditionally used actual body weight, data showed a significant relationship to weight (p<0.001) and gender (p<0.05). In contrast, indexing to body height presented a method without dependencies on height, weight, or gender, yielding a uniform 95% confidence interval of 218-430 mL/m. Extravascular lung water increased with positive perioperative fluid balance (p=0.04).

CONCLUSIONS

Using either predicted or actual body weight for indexing extravascular lung water does not lead to independence of height, weight, and gender of the patient. Specifying a fixed range of normal or a uniform upper threshold for all patients is misleading for either method, despite widespread use. Our data suggest that indexing extravascular lung water to height is superior to weight-based methods. As we are not aware of any abnormal hemodynamic profile for brain tumor patients, we propose our findings to be a close approximation to normal values.

Effects of red blood cell transfusion on hemodynamic parameters: a prospective study in intensive care unit patients

Background

The aim of the study was to investigate the effect of red blood cell (RBC) transfusion on hemodynamic parameters including transpulmonary thermodilution (TPTD)-derived variables.

Methods

We compared hemodynamic parameters obtained before and after RBC transfusion (2 RBC units) in 34 intensive care unit (ICU) patients.

Results

Directly after RBC transfusion, we observed a significant increase in hematocrit (28 ± 3 vs. 22 ± 2%, p < 0.001), hemoglobin (9.4 ± 0.9 vs. 7.6 ± 0.8 g/dL, p < 0.001), arterial oxygen content (CaO₂) (12.2 ± 1.2 vs. 9.9 ± 1.0 mL/dL, p < 0.001), and oxygen delivery (DO₂) (1073 ± 369 vs. 934 ± 288 mL/min, p < 0.001) compared with baseline. Cardiac output (CO) (8.89 ± 3.06 vs. 9.42 ± 2.75 L/min, p = 0.020), cardiac index (CI) (4.53 ± 1.36 vs. 4.82 ± 1.21 L/min/m², p = 0.016), and heart rate (91 ± 16 vs. 95 ± 14 bpm, p = 0.007) were significantly lower following RBC transfusion while no significant change in stroke volume (SV) was observed. Mean arterial pressure (MAP) (median 87 vs. 78 mmHg, p < 0.001) and systemic vascular resistance index (SVRI) (median 1212 vs. 1103 dyn*s*cm^-5*m², p = 0.001) significantly increased directly after RBC transfusion. Global end-diastolic volume index (GEDVI), extravascular lung water index (EVLWI), and pulmonary vascular permeability index (PVPI) did not significantly change.

Conclusions

In ICU patients, the transfusion of 2 RBC units induces a significant decrease in CO and CI because of a significant decrease in heart rate (while SV remains unchanged). Despite the decrease in CO, DO₂ significantly increases because of a significant increase in CaO₂. In addition, RBC transfusion results in a significant increase in MAP and SVRI. No significant changes in TPTD-parameters reflecting cardiac preload (GEDVI), pulmonary edema (EVLWI), and pulmonary vascular permeability (PVPI) are observed following RBC transfusion.

Assessment of intravascular volume status and volume responsiveness in critically ill patients

Accurate assessment of a patient's volume status, as well as whether they will respond to a fluid challenge with an increase in cardiac output, is a critical task in the care of critically ill patients. Despite this, most decisions regarding fluid therapy are made either empirically or with limited and poor data. Given recent data highlighting the negative impact of either inadequate or overaggressive fluid therapy, understanding the tools and techniques available for accurate volume assessment is critical. This review highlights both static and dynamic methods that can be utilized to help in the assessment of volume status.

Prediction of fluid responsiveness in patients admitted to the medical intensive care unit

PURPOSE

Accurate prediction of fluid responsiveness is of importance in the treatment of patients admitted to the intensive care unit (ICU). We investigated whether physical examination, central venous pressure (CVP), central venous oxygen saturation (ScvO2), passive leg raising (PLR) test, and transpulmonary thermodilution (TPTD)-derived parameters can predict volume responsiveness in patients admitted to the ICU.

MATERIALS AND METHODS

In this prospective study, structured clinical examination, measurement of CVP and ScvO2, a PLR test, and TPTD measurements were performed in 31 patients. A fluid challenge test was performed in 24 patients (fluid responsiveness was defined as a cardiac index [CI] increase of ≥ 15%).

RESULTS

Physical examination, CVP, ScvO2, the PLR test, and the TPTD-derived volumetric preload parameter global end-diastolic volume index showed poor prognostic capabilities regarding prediction of fluid responsiveness. Twenty-nine percent of patients were fluid responsive. There was a statistically significant correlation between the fluid challenge-induced increase in CI and changes in global end-diastolic volume index (r = 0.666, P < .001). In only 17% of patients, CI did not increase after fluid loading.

CONCLUSIONS

Prediction of fluid responsiveness is difficult using physical examination, CVP, ScvO2, PLR maneuver, or TPTD-derived variables in critically ill patients. A volume challenge test should be considered for the assessment of fluid responsiveness in critically ill patients admitted to the ICU.

Advanced hemodynamic monitoring before and after transjugular intrahepatic portosystemic shunt: implications for selection of patients--a prospective study

PURPOSE

To investigate immediate and short-term effects of transjugular intrahepatic portosystemic shunt (TIPS) on cardiocirculatory, hepatic, and renal function and characterize predictors for TIPS outcome in terms of organ function after TIPS.

MATERIALS AND METHODS

This prospective study was approved by the ethics committee at a university hospital and was conducted in a medical intensive care unit. Informed consent was obtained. Twenty patients with indication for TIPS were enrolled. Monitoring of hemodynamic and hepatic function (transpulmonary thermodilution, indocyanine green plasma disappearance rate [ICG-PDR]) was performed. Biochemical markers of organ function were obtained. Statistical analysis (Wilcoxon test, Spearman correlation, multivariate linear regression analysis, receiver operating characteristic [ROC] analysis) was performed.

RESULTS

After TIPS, central venous pressure (median, 11 vs 15 cm H(2)O; P < .001), cardiac index (3.4 vs 3.8 L/min/m(2); P = .001), and global end-diastolic volume index (GEDVI) (726 vs 775 mL/m(2); P = .003) increased significantly. Portosystemic pressure gradient (28 vs 11 cm H(2)O; P < .001) and systemic vascular resistance index (1610 vs 1384 dyn · sec · cm(-5) · m(2); P = .015) decreased significantly. Creatinine (1.1 vs 1.1 mg/dL; P = .008) and blood urea nitrogen (BUN) (27 vs 21 mg/dL; P = .006) decreased significantly. Bilirubin (1.8 vs 2.2 mg/dL; P = .032) and international normalized ratio (1.4 vs 1.5; P = .022) increased significantly. ICG-PDR significantly deteriorated after TIPS (P = .006). Higher baseline creatinine was independently associated with a decrease in creatinine after TIPS (R = 0.816, P < .001). ROC analysis identified baseline BUN (P = .026, area under ROC curve [AUC] = 0.818), cystatin C (P = .033, AUC = 0.805), and creatinine (P = .052, AUC = 0.779) as predictors of a decrease in creatinine of 0.5 mg/dL or greater and/or 25% or greater. An increase in bilirubin of 1 mg/dL or greater 1 week after TIPS was significantly associated with high baseline BUN (P = .007, AUC = 0.893) and high central venous pressure (P = .040, AUC = 0.800). Lower baseline alanine aminotransferase (P = .002, AUC = 1.000) and cardiac power index · GEDVI (P = .005, AUC = 0.960) predicted favorable TIPS outcome (creatinine decrease of ≥ 0.2 mg/dL without model for end-stage liver disease score increase of more than one point).

CONCLUSION

Patients with renal insufficiency, compensated hepatocellular function, decreased cardiac preload, and decreased cardiac performance benefit most from TIPS.

Computed tomography to estimate cardiac preload and extravascular lung water. A retrospective analysis in critically ill patients

Background

In critically ill patients intravascular volume status and pulmonary edema need to be quantified as soon as possible. Many critically ill patients undergo a computed tomography (CT)-scan of the thorax after admission to the intensive care unit (ICU). This study investigates whether CT-based estimation of cardiac preload and pulmonary hydration can accurately assess volume status and can contribute to an early estimation of hemodynamics.

Methods

Thirty medical ICU patients. Global end-diastolic volume index (GEDVI) and extravascular lung water index (EVLWI) were assessed using transpulmonary thermodilution (TPTD) serving as reference method (with established GEDVI/EVLWI normal values). Central venous pressure (CVP) was determined. CT-based estimation of GEDVI/EVLWI/CVP by two different radiologists (R1, R2) without analyzing software. Primary endpoint: predictive capabilities of CT-based estimation of GEDVI/EVLWI/CVP compared to TPTD and measured CVP. Secondary endpoint: interobserver correlation and agreement between R1 and R2.

Results

Accuracy of CT-estimation of GEDVI (< 680, 680-800, > 800 mL/m²) was 33%(R1)/27%(R2). For R1 and R2 sensitivity for diagnosis of low GEDVI (< 680 mL/m²) was 0% (specificity 100%). Sensitivity for prediction of elevated GEDVI (> 800 mL/m²) was 86%(R1)/57%(R2) with a specificity of 57%(R1)/39%(R2) (positive predictive value 38%(R1)/22%(R2); negative predictive value 93%(R1)/75%(R2)). Estimated CT-GEDVI and TPTD-GEDVI were significantly different showing an overestimation of GEDVI by the radiologists (R1: mean difference ± standard error (SE): 191 ± 30 mL/m², p < 0.001; R2: mean difference ± SE: 215 ± 37 mL/m², p < 0.001). CT GEDVI and TPTD-GEDVI showed a very low Lin-concordance correlation coefficient (ccc) (R1: ccc = +0.20, 95% CI: +0.00 to +0.38, bias-correction factor (BCF) = 0.52; R2: ccc = -0.03, 95% CI: -0.19 to +0.12, BCF = 0.42). Accuracy of CT estimation in prediction of EVLWI (< 7, 7-10, > 10 mL/kg) was 30% for R1 and 40% for R2. CT-EVLWI and TPTD-EVLWI were significantly different (R1: mean difference ± SE: 3.3 ± 1.2 mL/kg, p = 0.013; R2: mean difference ± SE: 2.8 ± 1.1 mL/kg, p = 0.021). Again ccc was low with -0.02 (R1; 95% CI: -0.20 to +0.13, BCF = 0.44) and +0.14 (R2; 95% CI: -0.05 to +0.32, BCF = 0.53). GEDVI, EVLWI and CVP estimations of R1 and R2 showed a poor interobserver correlation (low ccc) and poor interobserver agreement (low kappa-values).

Conclusions

CT-based estimation of GEDVI/EVLWI is not accurate for predicting cardiac preload and extravascular lung water in critically ill patients when compared to invasive TPTD-assessment of these variables.