Evaluating the adequacy of fluid resuscitation in patients with septic shock: controversies and future directions

Fluid resuscitation is a cornerstone in the treatment of severe sepsis and septic shock. However, there is little evidence to guide clinicians in its administration. Current guidelines recommend targeting fluid therapy based on measurements of cardiac filling pressures, such as central venous pressure. Static pressures are poor predictors of a patient's response to fluid. Such response can be better predicted by measuring changes in hemodynamic parameters caused by positive pressure ventilation or maneuvers designed to simulate increased preload. These changes can be measured by analysis of arterial waveforms, echocardiography or Doppler, or with emerging noninvasive technologies. This article reviews the current role of fluid replacement strategies and the use of monitoring systems in the overall resuscitation of patients with severe sepsis and septic shock.

Antibiotic stewardship in the intensive care unit

The rapid emergence and dissemination of antimicrobial-resistant microorganisms in ICUs worldwide constitute a problem of crisis dimensions. The root causes of this problem are multifactorial, but the core issues are clear. The emergence of antibiotic resistance is highly correlated with selective pressure resulting from inappropriate use of these drugs. Appropriate antibiotic stewardship in ICUs includes not only rapid identification and optimal treatment of bacterial infections in these critically ill patients, based on pharmacokinetic-pharmacodynamic characteristics, but also improving our ability to avoid administering unnecessary broad-spectrum antibiotics, shortening the duration of their administration, and reducing the numbers of patients receiving undue antibiotic therapy. Either we will be able to implement such a policy or we and our patients will face an uncontrollable surge of very difficult-to-treat pathogens.

Intraoperative management of heart-lung interactions: "from hypothetical prediction to improved titration"

Extensive literature describes the suitability of dynamic parameters to predict responsiveness in fluid. However, based on heart-lung interactions, these parameters can have serious limitations, including the use of protective lung ventilation. Although the latter seems to be beneficial for healthy patients undergoing high-risk surgery, the intraoperative interpretation of dynamic parameters to predict fluid responsiveness can be hazardous. In this context, the attending physician could, alternatively, titrate the need of fluids with a small fluid challenge, which remains unaffected by low tidal volume, the presence of arrhythmia, or the presence of spontaneous ventilation. When intraoperative prediction of fluid responsiveness is required in mechanically ventilated patients, "improved" titration should be preferred to a hypothetical prediction.

Dynamic variables and fluid responsiveness in patients for aortic stenosis surgery

BACKGROUND

Aortic stenosis is the most common valvular disease in developed countries, but it carries an increased mortality during non-cardiac surgery underscoring the importance of adequate hemodynamic management. Further, haemodynamic management of patients immediately after surgery for aortic stenosis can be challenging. Prediction of fluid responsiveness using dynamic variables has not been sufficiently studied in patients for aortic stenosis surgery.

METHODS

Observational study evaluating fluid responsiveness on 32 (31 analysed) patients scheduled for aortic valve replacement due to aortic stenosis on mechanical ventilation before and after valve replacement. Increase in stroke volume (oesophagus Doppler) ≥ 15% to a fluid challenge defined fluid responders.

RESULTS

Before surgery (31 fluid loads performed in 31 patients), areas under receiver operating characteristics curves (95% confidence intervals) were stroke volume variation (from arterial pulse contour analysis) 0.77 (0.58-0.90), pulse pressure variation 0.75 (0.54-0.90) and Pleth variability index 0.51 (0.31-0.69). After aortic valve replacement (31 fluid loads performed in 23 patients) the values were stroke volume variation 0.90 (0.74-0.98), pulse pressure variation 0.95 (0.80-1.0) and Pleth variability index 0.72 (0.52-0.87).

CONCLUSIONS

The arterial pressure-based variables had moderate predictive values before valve replacement, but it predicted fluid responsiveness well postoperatively. Pleth variability index did not predict fluid responsiveness preoperatively, and it had a moderate predictive value postoperatively. These results indicate that arterial pressure-based dynamic variables have limited potential to guide fluid therapy in patients with aortic stenosis. Their ability to guide fluid therapy after aortic valve replacement seems better.

Abdominal infections in the intensive care unit: characteristics, treatment and determinants of outcome

BACKGROUND

Abdominal infections are frequent causes of sepsis and septic shock in the intensive care unit (ICU) and are associated with adverse outcomes. We analyzed the characteristics, treatments and outcome of ICU patients with abdominal infections using data extracted from a one-day point prevalence study, the Extended Prevalence of Infection in the ICU (EPIC) II.

METHODS

EPIC II included 13,796 adult patients from 1,265 ICUs in 75 countries. Infection was defined using the International Sepsis Forum criteria. Microbiological analyses were performed locally. Participating ICUs provided patient follow-up until hospital discharge or for 60 days.

RESULTS

Of the 7,087 infected patients, 1,392 (19.6%) had an abdominal infection on the study day (60% male, mean age 62 ± 16 years, SAPS II score 39 ± 16, SOFA score 7.6 ± 4.6). Microbiological cultures were positive in 931 (67%) patients, most commonly Gram-negative bacteria (48.0%). Antibiotics were administered to 1366 (98.1%) patients. Patients who had been in the ICU for ≤ 2 days prior to the study day had more Escherichia coli, methicillin-sensitive Staphylococcus aureus and anaerobic isolates, and fewer enterococci than patients who had been in the ICU longer. ICU and hospital mortality rates were 29.4% and 36.3%, respectively. ICU mortality was higher in patients with abdominal infections than in those with other infections (29.4% vs. 24.4%, p < 0.001). In multivariable analysis, hematological malignancy, mechanical ventilation, cirrhosis, need for renal replacement therapy and SAPS II score were independently associated with increased mortality.

CONCLUSIONS

The characteristics, microbiology and antibiotic treatment of abdominal infections in critically ill patients are diverse. Mortality in patients with isolated abdominal infections was higher than in those who had other infections.

Vancomycin and piperacillin-tazobactam against methicillin-resistant Staphylococcus aureus and vancomycin-intermediate Staphylococcus aureus in an in vitro pharmacokinetic/pharmacodynamic model

PURPOSE

Synergy between β-lactams and vancomycin against methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-intermediate Staphylococcus aureus (VISA) has been observed in vitro and in vivo. However, studies investigating piperacillin-tazobactam with vancomycin against MRSA and VISA are limited despite broad clinical use of these antibiotics in combination. This study evaluated vancomycin and piperacillin-tazobactam against MRSA and VISA by using an in vitro pharmacokinetic/pharmacodynamic model.

METHODS

Two clinical MRSA strains (M3425 and M494) and one VISA strain (Mu50) were tested in duplicate by using a 72-hour, 1-compartment pharmacokinetic/pharmacodynamic model with the following exposures: growth control, vancomycin only, piperacillin-tazobactam only, and vancomycin with piperacillin-tazobactam. Vancomycin 1 g every 12 hours (free trough concentration, 8.75 mg/L; Cmin, 17.5 mg/L) and piperacillin-tazobactam 13.5 g per 24 hours' continuous infusion (free steady-state concentration, 27 mg/L) were simulated. Time-kill curves were constructed, and reductions in log10 CFU/mL at all time points were compared between regimens tested.

FINDINGS

Vancomycin and piperacillin-tazobactam MICs for M494, M3425, and Mu50 were 1, 1, and 4 and 1.5, 32, and >256 mg/L, respectively. All isolates had an oxacillin MIC ≥ 4 mg/L. Against all 3 isolates, vancomycin with piperacillin-tazobactam achieved a significant reduction in inoculum at 72 hours compared with vancomycin alone (all, P ≤ 0.015). The superiority of vancomycin with piperacillin-tazobactam compared with vancomycin alone became detectable at 8 hours for M3425 (P < 0.001) and at 24 hours for M494 and Mu50 (both, P ≤ 0.008). Although vancomycin with piperacillin-tazobactam achieved enhanced antibacterial activity at 72 hours against M3425 compared with vancomycin alone, bacterial regrowth occurred. Reduced susceptibility to vancomycin at 72 hours for M3425 was confirmed by using population analysis profile/AUC analysis. At 72 hours, M3425 had a PAP/AUC ratio of 0.77 compared to 0.51 at baseline.

IMPLICATIONS

Vancomycin with piperacillin-tazobactam demonstrated enhanced antimicrobial activity against MRSA and VISA compared with vancomycin alone. These results further enhance existing data that support using vancomycin in combination with a β-lactam for invasive MRSA infections. Combination therapy with vancomycin and a β-lactam against MRSA warrants clinical consideration.

Risk factors for target non-attainment during empirical treatment with β-lactam antibiotics in critically ill patients

PURPOSE

Risk factors for β-lactam antibiotic underdosing in critically ill patients have not been described in large-scale studies. The objective of this study was to describe pharmacokinetic/pharmacodynamic (PK/PD) target non-attainment envisioning empirical dosing in critically ill patients and considering a worst-case scenario as well as to identify patient characteristics that are associated with target non-attainment.

METHODS

This analysis uses data from the DALI study, a prospective, multi-centre pharmacokinetic point-prevalence study. For this analysis, we assumed that these were the concentrations that would be reached during empirical dosing, and calculated target attainment using a hypothetical target minimum inhibitory concentration (MIC), namely the susceptibility breakpoint of the least susceptible organism for which that antibiotic is commonly used. PK/PD targets were free drug concentration maintained above the MIC of the suspected pathogen for at least 50 % and 100 % of the dosing interval respectively (50 % and 100 % f T (>MIC)). Multivariable analysis was performed to identify factors associated with inadequate antibiotic exposure.

RESULTS

A total of 343 critically ill patients receiving eight different β-lactam antibiotics were included. The median (interquartile range) age was 60 (47-73) years, APACHE II score was 18 (13-24). In the hypothetical situation of empirical dosing, antibiotic concentrations remained below the MIC during 50 % and 100 % of the dosing interval in 66 (19.2 %) and 142 (41.4 %) patients respectively. The use of intermittent infusion was significantly associated with increased risk of non-attainment for both targets; creatinine clearance was independently associated with not reaching the 100 % f T( >MIC) target.

CONCLUSIONS

This study found that-in empirical dosing and considering a worst--case scenario--19 % and 41 % of the patients would not achieve antibiotic concentrations above the MIC during 50 % and 100 % of the dosing interval. The use of intermittent infusion (compared to extended and continuous infusion) was the main determinant of non-attainment for both targets; increasing creatinine clearance was also associated with not attaining concentrations above the MIC for the whole dosing interval. In the light of this study from 68 ICUs across ten countries, we believe current empiric dosing recommendations for ICU patients are inadequate to effectively cover a broad range of susceptible organisms and need to be reconsidered.

High-dose continuous oxacillin infusion results in achievement of pharmacokinetics targets in critically ill patients with deep sternal wound infections following cardiac surgery

Knowledge regarding antimicrobial therapy strategies in deep sternal wound infections (DSWI) following cardiac surgery is limited. Therefore, we aimed to determine the steady-state plasma and mediastinal concentrations of oxacillin administered by continuous infusion in critically ill patients with DSWI and to compare these concentrations with the susceptibility of staphylococci recovered. A continuous infusion of oxacillin (150 to 200 mg/kg of body weight/24 h) was administered after a loading dose (50 mg/kg). Plasma and mediastinal concentrations of total and unbound oxacillin were determined 4 h after the loading dose (H4) and then at day 1 (H24) and day 2 (H48). Twelve patients were included. Nine patients exhibited bacteremia, 5 were in septic shock, 8 were positive for Staphylococcus aureus, and 4 were positive for coagulase-negative staphylococci. The median MIC (first to third interquartile range) was 0.25 (0.24 to 0.41) mg/liter. Median plasma concentrations of total and unbound oxacillin at H4, H24, and H48 were, respectively, 64.4 (41.4 to 78.5) and 20.4 (12.4 to 30.4) mg/liter, 56.9 (31.4 to 80.6) and 21.7 (6.5 to 27.3) mg/liter, and 57.5 (32.2 to 85.1) and 20 (14.3 to 35.7) mg/liter. The median mediastinal concentrations of total and unbound oxacillin at H4, H24, and H48 were, respectively, 2.3 (0.7 to 25.9) and 0.9 (<0.5 to 15) mg/liter, 29.1 (19.7 to 38.2) and 12.6 (5.9 to 19.8) mg/liter, and 31.6 (14.9 to 42.9) and 17.1 (6.7 to 26.7) mg/liter. High-dose oxacillin delivered by continuous infusion is a valuable strategy to achieve our pharmacokinetic target (4× MIC) at the site of action at H24. But concerns remain in cases of higher MICs, emphasizing the need for clinicians to obtain the MICs for the bacteria and to monitor oxacillin concentrations, especially the unbound forms, at the target site.

Beta-lactam antibiotic dosing during continuous renal replacement therapy: how can we optimize therapy?

Correct antibiotic treatment is of utmost importance to treat infections in critically ill patients, not only in terms of spectrum and timing but also in terms of dosing. However, this is a real challenge for the clinician because the pathophysiology (such as shock, augmented renal clearance, and multiple organ dysfunction) has a major impact on the pharmacokinetics of hydrophilic antibiotics. The presence of extra-corporal circuits, such as continuous renal replacement therapy, may further complicate this difficult exercise. Standard dosing may result in inadequate concentrations, but unadjusted dosing regimens may lead to toxicity. Recent studies confirm the variability in concentrations, and the wide variation in dialysis techniques used certainly contributes to these findings. Well-designed clinical studies are needed to provide the data from which robust dosing guidance can be developed. In the meantime, non-adjusted dosing in the first 1 to 2 days of antibiotic therapy during continuous renal replacement therapy followed by dose reduction later on seems to be a prudent approach.

Low flucloxacillin concentrations in a patient with central nervous system infection: the need for plasma and cerebrospinal fluid drug monitoring in the ICU

OBJECTIVE

To report the difficulty in achieving and maintaining target antibiotic exposure in critically ill patients with deep-seeded infections.

CASE SUMMARY

We present a case of a 36-year-old man who was admitted to the intensive care unit with diffuse central nervous system and peripheral methicillin-sensitive Staphylococcus aureus infection (minimum inhibitory concentration; MIC, 1 µg/mL). Owing to the complicated nature of the infection, sequential concentrations of free flucloxacillin were measured in plasma and cerebrospinal fluid (CSF) and used to direct antibiotic dosing. Unsurprisingly, the trough plasma concentrations of flucloxacillin were below the MIC (0.2-0.4 µg/mL), and the corresponding CSF concentrations were undetectable (<0.1 µg/mL) with standard intermittent bolus dosing of 2 g every 4 hours. By administering flucloxacillin by continuous infusion (CI) and increasing the dose to 20 g daily, the plasma (2.2-5.7 µg/mL) and CSF (0.1 µg/mL) levels were increased, albeit lower than the predefined targets (plasma, 40 µg/mL; CSF, 4 µg/mL).

DISCUSSION

The presence of physiological changes associated with critical illness-namely, hypoalbuminemia and augmented renal clearance-may significantly alter antibiotic pharmacokinetics, and this phenomenon may lead to suboptimal antibiotic exposure if they are not accounted for. This case also highlights the value of applying CI in such patient groups and demonstrates the significance of monitoring plasma and CSF drug concentrations in optimizing antibiotic delivery.

CONCLUSIONS

Future research should aim to evaluate the utility of such drug monitoring with regard to patient outcomes and cost-effectiveness.

Respiratory variations in the arterial pressure during mechanical ventilation reflect volume status and fluid responsiveness

Optimal fluid management is one of the main challenges in the care of the critically ill. However, the physiological parameters that are commonly monitored and used to guide fluid management are often inadequate and even misleading. From 1987 to 1989 we published four experimental studies which described a method for predicting the response of the cardiac output to fluid administration during mechanical ventilation. The method is based on the analysis of the variations in the arterial pressure in response to a mechanical breath, which serves as a repetitive hemodynamic challenge. Our studies showed that the systolic pressure variation and its components are able to reflect even small changes in the circulating blood volume. Moreover, these dynamic parameters provide information about the slope of the left ventricular function curve, and therefore predict the response to fluid administration better than static preload parameters. Many new dynamic parameters have been introduced since then, including the pulse pressure (PPV) and stroke volume (SVV) variations, and various echocardiographic and other parameters. Though seemingly different, all these parameters are based on measuring the response to a predefined preload-modifying maneuver. The clinical usefulness of these 'dynamic' parameters is limited by many confounding factors, the recognition of which is absolutely necessary for their proper use. With more than 20 years of hindsight we believe that our early studies helped pave the way for the recognition that fluid administration should ideally be preceded by the assessment of "fluid responsiveness". The introduction of dynamic parameters into clinical practice can therefore be viewed as a significant step towards a more rational approach to fluid management.

Functional haemodynamic monitoring

PURPOSE OF REVIEW

Functional haemodynamic monitoring is the assessment of the dynamic interactions of haemodynamic variables in response to a defined perturbation.

RECENT FINDINGS

Fluid responsiveness can be predicted during positive pressure breathing by variations in venous return or left ventricular output using numerous surrogate markers, such as arterial pulse pressure variation (PPV), left ventricular stroke volume variation (SVV), aortic velocity variation, inferior and superior vena cavae diameter changes and pulse oximeter pleth signal variability. Similarly, dynamic changes in cardiac output to a passive leg raising manoeuvre can be used in any patient and measured invasively or noninvasively. However, volume responsiveness, though important, reflects only part of the overall spectrum of functional physiological variables that can be measured to define physiologic state and monitor response to therapy. The ratio of PPV to SVV defines central arterial elastance and can be used to identify those hypotensive patients who will not increase their blood pressure in response to a fluid challenge despite increasing cardiac output. Dynamic tissue O2 saturation (StO2) responses to complete stop flow conditions, as can be created by measuring hand StO2 and occluding flow with a blood pressure cuff, assesses cardiovascular sufficiency and micro-circulatory blood flow distribution. They can be used to identify those ventilator-dependent individuals who will fail a spontaneous breathing trial or trauma patients in need of life-saving interventions.

SUMMARY

Functional haemodynamic monitoring approaches are increasing in numbers, conditions in which they are useful and resuscitation protocol applications. This is a rapidly evolving field whose pluripotential is just now being realized.

A 10-second fluid challenge guided by transthoracic echocardiography can predict fluid responsiveness

Introduction

The accurate assessment of intravascular volume status for the therapy of severe hypovolemia and shock is difficult and critical to critically ill patients. Non-invasive evaluation of fluid responsiveness by the rapid infusion of a very limited amount of volume is an important clinical goal. This study aimed to test whether echocardiographic parameters could predict fluid responsiveness in critically ill patients following a low-volume (50-ml crystalloid solution) infusion over 10 seconds.

Methods

We prospectively studied 55 mechanically ventilated patients. Echocardiography was performed during a 50-ml infusion of crystalloid solution over 10 seconds and a further 450 ml over 15 minutes. Cardiac output (CO), stroke volume (SV), aortic velocity time index (VTI), and left ventricular ejection fraction (LVEF) were recorded. Patients were classified as responders (Rs) if CO increased by at least 15% following the 500-ml volume expansion or were classified as non-responders (NRs) if CO increased by less than 15%. Area under the receiver operating characteristic curves (AUC) compared CO variations after 50 ml over 10 seconds (∆CO50) and 500 ml over 15 minutes (∆CO500) and the variation of VTI after infusion of 50 ml of fluid over 10 seconds (∆VTI50).

Results

In total, 50 patients were enrolled, and 27 (54%) of them were Rs. General characteristics, LVEF, heart rate, and central venous pressure were similar between Rs and NRs. In the Rs group, the AUC for ∆CO50 was 0.95 ± 0.03 (P <0.01; best cutoff value, 6%; sensitivity, 93%; specificity, 91%). Moreover, ∆CO50 and ∆CO500 were strongly correlated (r = 0.87; P <0.01). The AUC for ∆VTI50 was 0.91 ± 0.04 (P <0.01; best cutoff value, 9%; sensitivity, 74%; specificity, 95%). ∆VTI50 and ∆CO500 were positively correlated (r = 0.72; P <0.01).

Conclusion

In critically ill patients, the variation of CO and VTI after the administration of 50-ml crystalloid solution over 10 seconds (∆CO50 and ∆VTI50) can accurately predict fluid responsiveness.

Trial registration

Current Controlled Trials ISRCTN10524328. Registered 12 December 2013.

Does contemporary vancomycin dosing achieve therapeutic targets in a heterogeneous clinical cohort of critically ill patients? Data from the multinational DALI study

Introduction

The objective of this study was to describe the pharmacokinetics of vancomycin in ICU patients and to examine whether contemporary antibiotic dosing results in concentrations that have been associated with favourable response.

Methods

The Defining Antibiotic Levels in Intensive Care (DALI) study was a prospective, multicentre pharmacokinetic point-prevalence study. Antibiotic dosing was as per the treating clinician either by intermittent bolus or continuous infusion. Target trough concentration was defined as ≥15 mg/L and target pharmacodynamic index was defined as an area under the concentration-time curve over a 24-hour period divided by the minimum inhibitory concentration of the suspected bacteria (AUC_0–24/MIC ratio) >400 (assuming MIC ≤1 mg/L).

Results

Data of 42 patients from 26 ICUs were eligible for analysis. A total of 24 patients received vancomycin by continuous infusion (57%). Daily dosage of vancomycin was 27 mg/kg (interquartile range (IQR) 18 to 32), and not different between patients receiving intermittent or continuous infusion. Trough concentrations were highly variable (median 27, IQR 8 to 23 mg/L). Target trough concentrations were achieved in 57% of patients, but more frequently in patients receiving continuous infusion (71% versus 39%; P = 0.038). Also the target AUC_0–24/MIC ratio was reached more frequently in patients receiving continuous infusion (88% versus 50%; P = 0.008). Multivariable logistic regression analysis with adjustment by the propensity score could not confirm continuous infusion as an independent predictor of an AUC_0–24/MIC >400 (odds ratio (OR) 1.65, 95% confidence interval (CI) 0.2 to 12.0) or a C_min ≥15 mg/L (OR 1.8, 95% CI 0.4 to 8.5).

Conclusions

This study demonstrated large interindividual variability in vancomycin pharmacokinetic and pharmacodynamic target attainment in ICU patients. These data suggests that a re-evaluation of current vancomycin dosing recommendations in critically ill patients is needed to more rapidly and consistently achieve sufficient vancomycin exposure.

Variability in protein binding of teicoplanin and achievement of therapeutic drug monitoring targets in critically ill patients: lessons from the DALI Study

The aims of this study were to describe the variability in protein binding of teicoplanin in critically ill patients as well as the number of patients achieving therapeutic target concentrations. This report is part of the multinational pharmacokinetic DALI Study. Patients were sampled on a single day, with blood samples taken both at the midpoint and the end of the dosing interval. Total and unbound teicoplanin concentrations were assayed using validated chromatographic methods. The lower therapeutic range of teicoplanin was defined as total trough concentrations from 10 to 20 mg/L and the higher range as 10-30 mg/L. Thirteen critically ill patients were available for analysis. The following are the median (interquartile range) total and free concentrations (mg/L): midpoint, total 13.6 (11.2-26.0) and free 1.5 (0.7-2.5); trough, total 11.9 (10.2-22.7) and free 1.8 (0.6-2.6). The percentage free teicoplanin for the mid-dose and trough time points was 6.9% (4.5-15.6%) and 8.2% (5.5-16.4%), respectively. The correlation between total and free antibiotic concentrations was moderate for both the midpoint (ρ = 0.79, P = 0.0021) and trough (ρ = 0.63, P = 0.027). Only 42% and 58% of patients were in the lower and higher therapeutic ranges, respectively. In conclusion, use of standard dosing for teicoplanin leads to inappropriate concentrations in a high proportion of critically ill patients. Variability in teicoplanin protein binding is very high, placing significant doubt on the validity of total concentrations for therapeutic drug monitoring in critically ill patients.

How to approach and treat VAP in ICU patients

Background

Ventilator-associated pneumonia (VAP) is one of the most frequent clinical problems in ICU with an elevated morbidity and costs associated with it, in addition to prolonged MV, ICU-length of stay (LOS) and hospital-length of stay. Current challenges in VAP management include the absence of a diagnostic gold standard; the lack of evidence regarding contamination vs. airway colonization vs. infection; and the increasing antibiotic resistance. We performed a Pubmed search of articles addressing the management of ventilator-associated pneumonia (VAP). Immunocompromised patients, children and VAP due to multi-drug resistant pathogens were excluded from the analysis. When facing a patient with VAP, it’s important to address a few key questions for the patient’s optimal management: when should antibiotics be started?; what microorganisms should be covered?; is there risk for multirresistant microorganisms?; how to choose the initial agent?; how microbiological tests determine antibiotic changes?; and lastly, which dose and for how long?. It’s important not to delay adequate treatment, since outcomes improve when empirical treatment is early and effective. We recommend short course of broad-spectrum antibiotics, followed by de-escalation when susceptibilities are available. Individualization of treatment is the key to optimal management.

Individualised antibiotic dosing for patients who are critically ill: challenges and potential solutions

Infections in critically ill patients are associated with persistently poor clinical outcomes. These patients have severely altered and variable antibiotic pharmacokinetics and are infected by less susceptible pathogens. Antibiotic dosing that does not account for these features is likely to result in suboptimum outcomes. In this Review, we explore the challenges related to patients and pathogens that contribute to inadequate antibiotic dosing and discuss how to implement a process for individualised antibiotic therapy that increases the accuracy of dosing and optimises care for critically ill patients. To improve antibiotic dosing, any physiological changes in patients that could alter antibiotic concentrations should first be established; such changes include altered fluid status, changes in serum albumin concentrations and renal and hepatic function, and microvascular failure. Second, antibiotic susceptibility of pathogens should be confirmed with microbiological techniques. Data for bacterial susceptibility could then be combined with measured data for antibiotic concentrations (when available) in clinical dosing software, which uses pharmacokinetic/pharmacodynamic derived models from critically ill patients to predict accurately the dosing needs for individual patients. Individualisation of dosing could optimise antibiotic exposure and maximise effectiveness.

Pulse pressure variation is comparable with central venous pressure to guide fluid resuscitation in experimental hemorrhagic shock with endotoxemia

INTRODUCTION

Pulse pressure variation (PPV) has been proposed as a promising resuscitation goal, but its ability to predict fluid responsiveness has been questioned in various conditions. The purpose of this study was to assess the performance of PPV in predicting fluid responsiveness in experimental hemorrhagic shock with endotoxemia, while comparing it with goals determined by a conventional set of guidelines.

METHODS

Twenty-seven pigs were submitted to acute hemorrhagic shock with intravenous infusion of endotoxin and randomized to three groups: (i) control; (ii) conventional treatment with crystalloids to achieve and maintain central venous pressure (CVP) 12 to 15 mmHg, mean arterial pressure of 65 mmHg or greater, and SvO2 (mixed venous oxygen saturation) of 65% or greater; (iii) treatment to achieve and maintain PPV of 13% or less. Parametric data were analyzed by two-way analysis of variance and Tukey test and differences in crystalloid volumes by t test. Predictive values of variables regarding fluid responsiveness were evaluated by receiver operating characteristic curves and multiple logistic regression.

RESULTS

Both treatments produced satisfactory hemodynamic recovery, without statistical differences in fluid administration (P = 0.066), but conventional treatment induced higher CVP (P = 0.001). Areas under receiver operating characteristic curves were larger for CVP (0.77; 95% confidence interval, 0.68-0.86) and PPV (0.74; 95% confidence interval, 0.65-0.83), and these variables were further selected by multiple logistic regression as independent predictors of responsiveness. Optimal PPV cutoff was 15%, with false-positive results involving mean pulmonary arterial pressure of 27 mmHg or greater.

CONCLUSIONS

Acute resuscitation guided by PPV was comparable with the strategy guided by CVP, mean arterial pressure, and SvO2. Central venous pressure and PPV were individually limited but independently predictive of fluid responsiveness.

Pulse pressure variations to guide fluid therapy in donors: a multicentric echocardiographic observational study

PURPOSE

Preload responsiveness parameters could be useful in the hemodynamic management of septic shock.

METHODS

A multicentric prospective echocardiographic observational study was conducted from March 2009 to August 2011. Clinically brain-dead subjects were included. Pulse pressure variations (ΔPPs) were recorded. Cardiac index, variation of the maximum flow velocity of aortic systolic blood flow, and right ventricular function parameters were evaluated via transthoracic echocardiography. Fluid responsiveness was defined by at least 15% cardiac index increase, 30 minutes after a 500-mL colloid solution infusion. The number of organs harvested was recorded.

RESULTS

Twenty-five subjects were included. Pulse pressure variation could not discriminate responders (n=15) from nonresponders (n=10). The best ΔPP threshold (20%) could discriminate responders with a sensitivity of 100% and a specificity of 40%. Variation of the maximum flow velocity of aortic systolic blood flow, tricuspid annular plane systolic excursion, and right ventricle dilation could not discriminate responders from nonresponders. Eighteen subjects underwent organ harvesting. The number of organs harvested was higher in responders (3.5 [3-5]) than in nonresponders (2.5 [2-3]; P=.03).

CONCLUSIONS

A ΔPP threshold of 13% is insufficient to guide volume expansion in donors. The best threshold is 20%. Fluid responsiveness monitoring could enhance organ harvesting.

Evaluation of pulse pressure variation validity criteria in critically ill patients: a prospective observational multicentre point-prevalence study

BACKGROUND

Respiratory variation in pulse pressure (ΔPP) is commonly used to predict the fluid responsiveness of critically ill patients. However, some researchers have demonstrated that this measurement has several limitations. The present study was designed to evaluate the proportion of patients satisfying criteria for valid application of ΔPP at a given time-point.

METHODS

A 1 day, prospective, observational, point-prevalence study was performed in 26 French intensive care units (ICUs). All patients hospitalized in the ICUs on the day of the study were included. The ΔPP validity criteria were recorded prospectively and defined as follows: (i) mechanical ventilation in the absence of spontaneous respiration; (ii) regular cardiac rhythm; (iii) tidal volume ≥8 ml kg(-1) of ideal body weight; (iv) a heart rate/respiratory rate ratio >3.6; (v) total respiratory system compliance ≥30 ml cm H2O(-1); and (vi) tricuspid annular peak systolic velocity ≥0.15 m s(-1).

RESULTS

The study included 311 patients with a Simplified Acute Physiology Score II of 41 (39-43). Overall, only six (2%) patients satisfied all validity criteria. Of the 170 patients with an arterial line in place, only five (3%) satisfied the validity criteria. During the 24 h preceding the study time-point, fluid responsiveness was assessed for 79 patients. ΔPP had been used to assess fluid responsiveness in 15 of these cases (19%).

CONCLUSIONS

A very low percentage of patients satisfied all criteria for valid use of ΔPP in the evaluation of fluid responsiveness. Physicians must consider limitations to the validity of ΔPP before using this variable.

Efficacy of dynamic indices in predicting fluid responsiveness in patients with obstructive jaundice

Previous studies have shown that the stroke volume variation (SVV), the pulse pressure variation (PPV) and the pleth variability index (PVI) could be successfully used for predicting fluid responsiveness (FR) in surgical patients. The aim of this study was to validate the ability of SVV, PPV and PVI to predict intraoperative FR in mechanically ventilated patients with obstructive jaundice (OJ). Thirty-two patients with OJ (mean serum total bilirubin 190.5 ± 95.3 µmol L(-1)) received intraoperative volume expansion (VE) with 250 ml colloids immediately after an exploratory laparotomy had been completed and after a 5 min period of hemodynamic stability. Hemodynamic variables were recorded before and after VE. FR was defined as an increase in stroke volume index > 10% after VE. The ability of SVV, PPV and PVI to predict FR was assessed by calculation of the area under the receiver operating characteristic curve. Eleven (34%) patients were responders and 21 patients were nonresponders to VE. The PPV was the unique dynamic index that had the moderate ability to predict FR during surgical procedures, the area under the curve was 0.71 (95% CI, 0.523 to 0.856; P = 0.039) and the threshold (sensitivity and specificity) discriminated responders was 7.5% (63.6%/71.4%). The present study concluded that SVV and PVI were not reliable predictors of FR, but PPV has some value predicting FR in patients with OJ intraoperatively.

Endpoints of resuscitation: what are they anyway?

Hemodynamic optimization of surgical patients during and after surgery in the Surgical Intensive Care Unit is meant to improve outcomes. These outcomes have been measured by Length Of Stay (LOS), rate of infection, days on ventilator, etc. Unfortunately, the adaptation of modern technology to accomplish this has been slow in coming. Ever since Shoemaker described in 1988 using a pulmonary artery catheter (PAC) to guide fluid and inotropic administration to deliver supranormal tissue oxygenation, many authors have written about different techniques to achieve this "hemodynamic optimization". Since the PAC and CVC have both gone out of favor for utilization to monitor and improve hemodynamics, many clinicians have resorted using the easy to use static measurements of blood pressure (BP), heart rate (HR), and urine output. In this paper, the authors will review why these static measurements are no longer adequate and review some of the newer technology that have been studied and proven useful. This review of newer technologies combined with laboratory measurements that have also proven to help guide the clinician, may provide the impetus to adopt new strategies in the operating rooms (OR) and SICU.

Ventricular diastolic abnormalities in the critically ill

PURPOSE OF REVIEW

Left-ventricular diastolic dysfunction is associated with various conditions frequently encountered in ICU patients. Due to prolonged relaxation and increased left-ventricular stiffness, patients with diastolic dysfunction are at high risk of developing abrupt pulmonary venous congestion. The present review describes the clinical spectrum of left-ventricular diastolic abnormalities in ICU patients.

RECENT FINDINGS

Left-ventricular diastolic dysfunction is associated with a preserved ejection fraction in half of the patients presenting with acute pulmonary edema. These patients may have dramatic presentation, such as flash pulmonary edema during a hypertensive crisis. Left-ventricular diastolic dysfunction is frequently involved in patients who fail extubation and may trigger weaning pulmonary edema. Sepsis and myocardial ischemia may also be associated with left-ventricular diastolic dysfunction. The diagnosis of left-ventricular diastolic dysfunction practically relies on two-dimensional and Doppler echocardiography. Further large-scale clinical studies are needed to better characterize the prevalence, the clinical relevance and time-course of left-ventricular diastolic dysfunction in ICU patients.

SUMMARY

Left-ventricular diastolic dysfunction accounts for a growing proportion of cardiogenic pulmonary edema and weaning failure in ICU patients. It may be reversible when induced by sepsis or myocardial ischemia. Its prognostic value in the ICU settings remains to be further investigated.

Pulmonary Arterial Hypertension in Intensive Care Unit

For the intensive care unit (ICU) physician, the diagnosis of pulmonary arterial hypertension (PAH) is difficult as it can easily be confounded with other forms of pulmonary hypertension (PH). The key issue is that PAH is a form of PH. On the opposite, PH does not automatically imply PAH. Pulmonary arterial hypertension must be differentiated from other causes of PH that are frequently seen in ICU. It was recently emphasized that pulmonary veno-occlusive disease (PVOD) must be differentiated from PH and PAH. The prognosis of PAH was consistently improved in the ten past years by introduction of selective pulmonary vasodilators and management by highly specialized medical teams. In ICU patients, PAH remains a severe disease with a high mortality rate. When PAH is suspected, a systematic diagnosis approach is of particular importance in order to rapidly eliminate left cardiac, thromboembolic and pulmonary causes of PH. Left cardiac disease is the most common cause of PH. Early recognition of PAH allows a rapid introduction of selective pulmonary vasodilators that can improve outcome. Idiopathic PAH is the most frequent cause but it can also be associated with scleroderma, HIV infection, anorexigen toxicity, thyroid disease, cirrhosis. Pulmonary vasodilators should be only a part of a general management including treatment of triggering factors, optimization of fluid balance, decrease of RV afterload by using pulmonary vasodilators while maintaining cardiac output and mean arterial pressure. The early contact of PH referral center or specialized physician is of particular importance.

Basic concepts of fluid responsiveness

Predicting fluid responsiveness, the response of stroke volume to fluid loading, is a relatively novel concept that aims to optimise circulation, and as such organ perfusion, while avoiding futile and potentially deleterious fluid administrations in critically ill patients. Dynamic parameters have shown to be superior in predicting the response to fluid loading compared with static cardiac filling pressures. However, in routine clinical practice the conditions necessary for dynamic parameters to predict fluid responsiveness are frequently not met. Passive leg raising as a means to alter biventricular preload in combination with subsequent measurement of the change in stroke volume can provide a fast and accurate way to guide fluid management in a broad population of critically ill patients.

Pavane for a pulse pressure variation defunct

Hemodynamic management of critically ill patients in the ICU or high-risk patients in the operating room has paradoxically shown progress in terms of outcome after the systematic application of volume responsiveness/flow optimization based on pulse pressure variation and/or stroke volume variation during controlled, positive-pressure ventilation in patients without spontaneous respiratory efforts. This assessment of circulatory optimization should ideally be based on an exhaustive, predictive and coherent physiological understanding of the cardiovascular system model. This paper sketches the extremely complex physiological background of the concept of volume responsiveness, concluding that it is not a reliable means of guiding hemodynamic optimization because it is based on a nonexhaustive, nonpredictive and incoherent physiological model.

Predicting fluid responsiveness during infrarenal aortic cross-clamping in pigs

OBJECTIVE

Infrarenal aortic cross-clamping (ACC) induces hemodynamic disturbances that may affect respiratory-induced variations in stroke volume and, therefore, affect the ability of dynamic parameters such as pulse-pressure variation (PPV) to predict fluid responsiveness. Since this issue has not been investigated yet to authors' knowledge, the hypothesis was tested that ACC may change PPV and impair its ability to predict fluid responsiveness.

DESIGN

Prospective laboratory experiment.

SETTING

A university research laboratory.

PARTICIPANTS

Nineteen anesthetized and mechanically ventilated pigs.

INTERVENTIONS

Two courses of volume expansion were performed using 500 mL of saline before and during ACC. Animals were monitored using a systemic arterial catheter, and a pulmonary arterial catheter (stroke volume, central venous pressure, pulmonary arterial occlusion pressure). Animals were defined as responders to volume expansion if stroke volume increased ≥ 15%.

RESULTS

Before ACC, 13 animals were responders. Fluid responsiveness was predicted by a PPV ≥ 14% with a sensitivity of 77% (95% CI = 46%-95%) and a specificity of 83% (95% CI = 36%-97%). The area under the receiver operating characteristic curve was 0.90(95% CI = 0.67-0.99) and was higher than those generated for central venous pressure and pulmonary arterial occlusion pressure. ACC induced an increase in PPV (p<0.0005). During ACC, 8 animals were responders. An 18% PPV threshold discriminated between responders and non-responders to volume expansion, with a sensitivity of 87% (95% CI = 47%-98%) and a specificity of 54% (95% CI = 23%-83%). The area under the receiver operating characteristic curve was 0.72 (95% CI = 0.47-0.90) and was not different from those generated for central venous pressure and pulmonary arterial occlusion pressure.

CONCLUSIONS

ACC induced a significant increase in PPV and reduced its ability to predict fluid responsiveness.

Early diastolic dysfunction is associated with intensive care unit mortality in cancer patients presenting with septic shock

BACKGROUND

Cancer patients present a high risk of sepsis and are exposed to cardiotoxic drugs during chemotherapy. Myocardial dysfunction is common during septic shock and can be evaluated at bedside using echocardiography. The aim of this study was to identify early cardiac dysfunctions associated with intensive care unit (ICU) mortality in cancer patients presenting with septic shock.

METHODS

Seventy-two cancer patients admitted to the ICU underwent echocardiography within 48 h of developing septic shock. History of malignancies, anticancer treatments, and clinical characteristics were prospectively collected.

RESULTS

ICU mortality was 48%. Diastolic dysfunction (e' ≤8 cm s(-1)) was an independent echocardiographic parameter associated with ICU mortality {odds ratio (OR) 7.7 [95% confidence interval (CI), 2.58-23.38]; P<0.001}. Overall, three factors were independently associated with ICU mortality: sepsis-related organ failure assessment score at admission [OR 1.35 ( 95% CI, 1.05-1.74); P=0.017], occurrence of diastolic dysfunction [OR 16.6 (95% CI, 3.28-84.6); P=0.001], and need for conventional mechanical ventilation [OR 16.6 (95% CI, 3.6-77.15); P<0.001]. Diastolic dysfunction was not associated with exposure to cardiotoxic drugs.

CONCLUSIONS

Early diastolic dysfunction is a strong and independent predictor of mortality in cancer patients presenting with septic shock. It is not associated with exposure to cardiotoxic drugs. Further studies incorporating monitoring of diastolic function and therapeutic interventions improving cardiac relaxation need to be evaluated in cancer patients presenting with septic shock.

Fluid challenge: tracking changes in cardiac output with blood pressure monitoring (invasive or non-invasive)

PURPOSE

To assess whether invasive and non-invasive blood pressure (BP) monitoring allows the identification of patients who have responded to a fluid challenge, i.e., who have increased their cardiac output (CO).

METHODS

Patients with signs of circulatory failure were prospectively included. Before and after a fluid challenge, CO and the mean of four intra-arterial and oscillometric brachial cuff BP measurements were collected. Fluid responsiveness was defined by an increase in CO ≥10 or ≥15% in case of regular rhythm or arrhythmia, respectively.

RESULTS

In 130 patients, the correlation between a fluid-induced increase in pulse pressure (Δ500mlPP) and fluid-induced increase in CO was weak and was similar for invasive and non-invasive measurements of BP: r² = 0.31 and r² = 0.29, respectively (both p < 0.001). For the identification of responders, invasive Δ500mlPP was associated with an area under the receiver-operating curve (AUC) of 0.82 (0.74-0.88), similar (p = 0.80) to that of non-invasive Δ500mlPP [AUC of 0.81 (0.73-0.87)]. Outside large gray zones of inconclusive values (5-23% for invasive Δ500mlPP and 4-35% for non-invasive Δ500mlPP, involving 35 and 48% of patients, respectively), the detection of responsiveness or unresponsiveness to fluid was reliable. Cardiac arrhythmia did not impair the performance of invasive or non-invasive Δ500mlPP. Other BP-derived indices did not outperform Δ500mlPP.

CONCLUSIONS

As evidenced by large gray zones, BP-derived indices poorly reflected fluid responsiveness. However, in our deeply sedated population, a high increase in invasive pulse pressure (>23%) or even in non-invasive pulse pressure (>35%) reliably detected a response to fluid. In the absence of a marked increase in pulse pressure (<4-5%), a response to fluid was unlikely.

Noninvasive assessment of hemodynamic response to a fluid challenge using femoral Doppler in critically ill ventilated patients

PURPOSE

The purpose of the study is to determine if femoral artery blood flow Doppler parameters can assess cardiac response to a fluid challenge (FC).

MATERIALS AND METHODS

We prospectively recorded in 52 critically ill ventilated patients' velocity time integral variation (%VTIf) and maximal systolic velocity variation (%Vfmax) derived from femoral Doppler analysis and aortic velocity time integral variation registered on transthoracic echocardiography before and after an FC of 500-mL saline.

RESULTS

According to Pearson coefficient, %Vfmax and %VTIf were found to be positively correlated with aortic velocity time integral variation (r(2) = 0.46 and 0.51, respectively; P < .0001) and were significantly different between responder patients and nonresponders (11% ± 3.4% vs 5.9% ± 4.3% and 14.9% ± 4.2% vs 5.5% ± 5.5%, respectively; P < .0001). Increase of %VTIf 10% or higher and %Vfmax 7% or higher after an FC showed a sensitivity of 80% and 84%, a specificity of 85% and 73%, and an area under the curve of 0.905 and 0.851, respectively, for discriminating responder and nonresponder patients.

CONCLUSION

Variation of femoral Doppler parameters before and after FC mirrors cardiac response to fluid loading. This tool could be considered as an alternative to transthoracic echocardiography in case of poor thoracic insonation.

Assessment of changes in cardiac index and fluid responsiveness: a comparison of Nexfin and transpulmonary thermodilution

BACKGROUND

The Nexfin device uses non-invasive photoplethysmography to monitor cardiac output and respiratory variations in pulse pressure and stroke volume. The aim of this study was to compare rapid changes in cardiac index after fluid challenge between Nexfin and bolus transpulmonary thermodilution and the ability to predict fluid responsiveness of dynamic indices given by Nexfin.

METHODS

Simultaneous comparative cardiac index were collected from transpulmonary thermodilution and Nexfin before and after fluid challenge in 45 patients following conventional cardiac surgery. Correlations, Bland-Altman analyses and percentage errors were calculated. Pulse pressure variations and stroke volume variations before fluid challenge were collected to assess their discrimination in predicting fluid responsiveness.

RESULTS

Eight (18%) patients were excluded. A weak positive relationship was found between rapid changes in cardiac index after fluid challenge given by both technologies (n = 37, r = 0.39, P = 0.019). Bias, precision and limits of agreements were 0.20 l/min/m(2) (95% confidence interval (CI) 0.02-0.40), 0.57 l/min/m(2) and ± 1.12 l/min/m(2) before fluid challenge, and 0.01 l/min/m(2) (95% CI -0.24 to 0.26), 0.74 l/min/m(2) and ± 1.45 l/min/m(2) after fluid challenge. Percentage errors between Nexfin and transpulmonary thermodilution were 55% and 58% before and after fluid challenge, respectively. Pulse pressure variations and stroke volume variations given by Nexfin were not discriminant to predict fluid responsiveness: areas under receiver operating characteristics curves 0.57 (95% CI 0.40-0.73) and 0.50 (0.33-0.67), respectively.

CONCLUSIONS

The Nexfin cannot be used to measure rapid changes in cardiac index following fluid challenge and to predict fluid responsiveness after cardiac surgery.

Goal-directed fluid optimization based on stroke volume variation and cardiac index during one-lung ventilation in patients undergoing thoracoscopy lobectomy operations: a pilot study

OBJECTIVES

This pilot study was designed to utilize stroke volume variation and cardiac index to ensure fluid optimization during one-lung ventilation in patients undergoing thoracoscopic lobectomies.

METHODS

Eighty patients undergoing thoracoscopic lobectomy were randomized into either a goal-directed therapy group or a control group. In the goal-directed therapy group, the stroke volume variation was controlled at 10%±1%, and the cardiac index was controlled at a minimum of 2.5 L.min-1.m-2. In the control group, the MAP was maintained at between 65 mm Hg and 90 mm Hg, heart rate was maintained at between 60 BPM and 100 BPM, and urinary output was greater than 0.5 mL/kg-1/h-1. The hemodynamic variables, arterial blood gas analyses, total administered fluid volume and side effects were recorded.

RESULTS

The PaO2/FiO2-ratio before the end of one-lung ventilation in the goal-directed therapy group was significantly higher than that of the control group, but there were no differences between the goal-directed therapy group and the control group for the PaO2/FiO2-ratio or other arterial blood gas analysis indices prior to anesthesia. The extubation time was significantly earlier in the goal-directed therapy group, but there was no difference in the length of hospital stay. Patients in the control group had greater urine volumes, and they were given greater colloid and overall fluid volumes. Nausea and vomiting were significantly reduced in the goal-directed therapy group.

CONCLUSION

The results of this study demonstrated that an optimization protocol, based on stroke volume variation and cardiac index obtained with a FloTrac/Vigileo device, increased the PaO2/FiO2-ratio and reduced the overall fluid volume, intubation time and postoperative complications (nausea and vomiting) in thoracic surgery patients requiring one-lung ventilation.