Despite a previous meta-analysis that concluded that central venous pressure should not be used to make clinical decisions regarding fluid management, central venous pressure continues to be recommended for this purpose. To perform an updated meta-analysis incorporating recent studies that investigated indices predictive of fluid responsiveness. A priori subgroup analysis was planned according to the location where the study was performed (ICU or operating room). MEDLINE, EMBASE, Cochrane Register of Controlled Trials, and citation review of relevant primary and review articles. Clinical trials that reported the correlation coefficient or area under the receiver operating characteristic curve (AUC) between the central venous pressure and change in cardiac performance following an intervention that altered cardiac preload. From 191 articles screened, 43 studies met our inclusion criteria and were included for data extraction. The studies included human adult subjects, and included healthy controls (n = 1) and ICU (n = 22) and operating room (n = 20) patients. Data were abstracted on study characteristics, patient population, baseline central venous pressure, the correlation coefficient, and/or the AUC between central venous pressure and change in stroke volume index/cardiac index and the percentage of fluid responders. Meta-analytic techniques were used to summarize the data. Overall 57% ± 13% of patients were fluid responders. The summary AUC was 0.56 (95% CI, 0.54-0.58) with no heterogenicity between studies. The summary AUC was 0.56 (95% CI, 0.52-0.60) for those studies done in the ICU and 0.56 (95% CI, 0.54-0.58) for those done in the operating room. The summary correlation coefficient between the baseline central venous pressure and change in stroke volume index/cardiac index was 0.18 (95% CI, 0.1-0.25), being 0.28 (95% CI, 0.16-0.40) in the ICU patients, and 0.11 (95% CI, 0.02-0.21) in the operating room patients. There are no data to support the widespread practice of using central venous pressure to guide fluid therapy. This approach to fluid resuscitation should be abandoned.

Stress hyperglycemia is common in critically ill patients and appears to be a marker of disease severity. Furthermore, both the admission as well as the mean glucose level during the hospital stay is strongly associated with patient outcomes. Clinicians, researchers and policy makers have assumed this association to be causal with the widespread adoption of protocols and programs for tight in-hospital glycemic control. However, a critical appraisal of the literature has demonstrated that attempts at tight glycemic control in both ICU and non-ICU patients do not improve health care outcomes. We suggest that hyperglycemia and insulin resistance in the setting of acute illness is an evolutionarily preserved adaptive responsive that increases the host's chances of survival. Furthermore, attempts to interfere with this exceedingly complex multi-system adaptive response may be harmful. This paper reviews the pathophysiology of stress hyperglycemia and insulin resistance and the protective role of stress hyperglycemia during acute illness.

The clinical determination of the intravascular volume can be extremely difficult in critically ill and injured patients as well as those undergoing major surgery. This is problematic because fluid loading is considered the first step in the resuscitation of hemodynamically unstable patients. Yet, multiple studies have demonstrated that only approximately 50% of hemodynamically unstable patients in the intensive care unit and operating room respond to a fluid challenge. Whereas under-resuscitation results in inadequate organ perfusion, accumulating data suggest that over-resuscitation increases the morbidity and mortality of critically ill patients. Cardiac filling pressures, including the central venous pressure and pulmonary artery occlusion pressure, have been traditionally used to guide fluid management. However, studies performed during the past 30 years have demonstrated that cardiac filling pressures are unable to predict fluid responsiveness. During the past decade, a number of dynamic tests of volume responsiveness have been reported. These tests dynamically monitor the change in stroke volume after a maneuver that increases or decreases venous return (preload) and challenges the patients' Frank-Starling curve. These dynamic tests use the change in stroke volume during mechanical ventilation or after a passive leg raising maneuver to assess fluid responsiveness. The stroke volume is measured continuously and in real-time by minimally invasive or noninvasive technologies, including Doppler methods, pulse contour analysis, and bioreactance.

In patients with acute circulatory failure, the decision to give fluids or not should not be taken lightly. The risk of overzealous fluid administration has been clearly established. Moreover, volume expansion does not always increase cardiac output as one expects. Thus, after the very initial phase and/or if fluid losses are not obvious, predicting fluid responsiveness should be the first step of fluid strategy. For this purpose, the central venous pressure as well as other “static” markers of preload has been used for decades, but they are not reliable. Robust evidence suggests that this traditional use should be abandoned. Over the last 15 years, a number of dynamic tests have been developed. These tests are based on the principle of inducing short-term changes in cardiac preload, using heart–lung interactions, the passive leg raise or by the infusion of small volumes of fluid, and to observe the resulting effect on cardiac output. Pulse pressure and stroke volume variations were first developed, but they are reliable only under strict conditions. The variations in vena caval diameters share many limitations of pulse pressure variations. The passive leg-raising test is now supported by solid evidence and is more frequently used. More recently, the end-expiratory occlusion test has been described, which is easily performed in ventilated patients. Unlike the traditional fluid challenge, these dynamic tests do not lead to fluid overload. The dynamic tests are complementary, and clinicians should choose between them based on the status of the patient and the cardiac output monitoring technique. Several methods and tests are currently available to identify preload responsiveness. All have some limitations, but they are frequently complementary. Along with elements indicating the risk of fluid administration, they should help clinicians to take the decision to administer fluids or not in a reasoned way.

There is overwhelming evidence that sepsis and septic shock are associated with hyperlactatemia (sepsis-associated hyperlactatemia (SAHL)). SAHL is a strong independent predictor of mortality and its presence and progression are widely appreciated by clinicians to define a very high-risk population. Until recently, the dominant paradigm has been that SAHL is a marker of tissue hypoxia. Accordingly, SAHL has been interpreted to indicate the presence of an `oxygen debt’ or `hypoperfusion’, which leads to increased lactate generation via anaerobic glycolysis. In light of such interpretation of the meaning of SAHL, maneuvers to increase oxygen delivery have been proposed as its treatment. Moreover, lactate levels have been proposed as a method to evaluate the adequacy of resuscitation and the nature of the response to the initial treatment for sepsis. However, a large body of evidence has accumulated that strongly challenges such notions. Much evidence now supports the view that SAHL is not due only to tissue hypoxia or anaerobic glycolysis. Experimental and human studies all consistently support the view that SAHL is more logically explained by increased aerobic glycolysis secondary to activation of the stress response (adrenergic stimulation). More importantly, new evidence suggests that SAHL may actually serve to facilitate bioenergetic efficiency through an increase in lactate oxidation. In this sense, the characteristics of lactate production best fit the notion of an adaptive survival response that grows in intensity as disease severity increases. Clinicians need to be aware of these developments in our understanding of SAHL in order to approach patient management according to biological principles and to interpret lactate concentrations during sepsis resuscitation according to current best knowledge.