Global tissue hypoxia as a result from systemic inflammatory response or circulatory failure is an important indicator of shock preceding multiple organ dysfunction syndrome (MODS). The development of MODS predicts outcome of the septic patient 1. Unrecognized and untreated global tissue hypoxia increases morbidity and mortality. Accurate detection of global tissue hypoxia is therefore of vital importance. Physical findings, vital signs, measuring central venous pressure (CVP) and urinary output are of the utmost importance, but not always sufficient for accurate detection of global tissue hypoxia 23.

It is now generally accepted that a decreased central venous oxygen saturation (ScvO₂) obtained from a central venous catheter, can reveal a mismatch between oxygen supply and oxygen demand, hence global tissue hypoxia 1. Decreased values predict poor prognosis after cardiovascular surgery 4, in severe cardiopulmonary disease 5, and in septic or cardiogenic shock 67. ScvO₂ and SvO₂ (mixed venous oxygen saturation) can therefore be used as hemodynamic goals during resuscitation. According to Rivers et al. 8, hemodynamic optimization demands 'early goal-directed therapy' (EGDT), including ScvO₂-guided treatment. It was concluded that goal-oriented manipulation of cardiac preload, afterload and contractility, to achieve a balance between systemic oxygen delivery and oxygen demand, is a valuable strategy in patients with sepsis or septic shock during the resuscitation period in the emergency department (ED) 8. More recently, as a result of this study, an EGDT treatment protocol was included in the 'Surviving Sepsis Campaign' guidelines 1. Also, several studies on implementation of such a protocol, partially in combination with other recommendations, have been published over the last years 91011.

Based on clinical experience it seemed that the syndrome targeted in the EGDT study 8 was not very common in intensive care units (ICUs) in The Netherlands, and thus EGDT was not commonly indicated. The main purpose of this study was to determine the incidence of low ScvO₂ values in our geographical setting. We monitored a heterogeneous group of critically ill patients during unplanned admission in three Dutch multidisciplinary ICUs. Also, illustratively, we compared the subgroup of septic patients with the population of septic patients as described by Rivers et al. 8, with respect to ScvO₂ and other baseline characteristics.

The main result of this present multi-center observational study is the low incidence of low ScvO₂ values (< 50%) in septic patients being only 1%. Secondary findings were the normal mean ScvO₂ values and normal mean SvO₂ values in critically ill patients, including patients with severe sepsis or septic shock, on admission in the three ICUs.

Development of severe sepsis and septic shock involves several pathogenic changes, including global tissue hypoxia as a result of circulatory abnormalities 14. In particular, hemodynamic optimization as a therapeutic target has been studied over the last decade 2891516. Based on promising results from earlier studies 2, Rivers et al. randomized patients with severe sepsis or septic shock to standard therapy or EGDT. The latter resulted in an absolute reduction in 28-day mortality of 16% 8. Improvement of the balance between oxygen delivery (DO₂) and oxygen demand (VO₂) played an important role. Other studies, however, found no reduction of morbidity or mortality as a result of aggressive hemodynamic optimization, despite higher central venous oxygenation or lower lactate concentrations 1516. Studies that enrolled patients admitted into the ICU were unable to show a decrease in mortality rate after aggressive hemodynamic optimization 1617, in contrast to studies that implemented certain treatment protocols, including antibiotic therapy, in the emergency department 8911. In this ICU study we found mean ScvO₂ and SvO₂ values in the normal range. Similar figures are described previously in the later stage of sepsis and in ICU patients 1819. This is in concordance with the findings by Gattinoni et al. (67, 3–69, 7%) 15 and Bracht et al. (70%) 20.

ScvO₂ is a surrogate for SvO₂: a significant correlation between the two has been described 21. Although it might still be debatable whether central venous and mixed venous oxygen saturation are equivalent or not 181921, the clinical importance of both measurements seems not to be an issue. The Surviving Sepsis Campaign recognizes such in the resuscitation portion of its severe sepsis and septic shock information 1. Our study design does not allow for any statistical evaluation of ScvO₂ compared to SvO₂.

APACHE II scores were similar in comparison to the population described in the EGDT study 8. This suggests equal mortality rate predictions. However, physiological scores such as APACHE II are dependent on variables that reflect the progression or reversal of organ dysfunction. Treatment in the ED or operating theatre prior to ICU admission influences calculation of the physiological scores. Consequently, the physiological scores at our ICUs could partially be underestimated. The significantly higher lactate plasma levels in the EGDT study suggest a more severe tissue hypoperfusion in that group. However, it is the clearance rate that is associated with less organ failure and improved survival 22.

Unlike significantly lower mean arterial pressures, the higher CVP and the lower hematocrit suggest that the septic patients were less hypovolemic compared to the EGDT population. Relatively high mean blood pressure in the EGDT population suggests an earlier stage of sepsis with predominating vasoconstriction, or pre-existing hypertension. The higher CVP in the subgroup with septic patients (n = 125) is partially the result of high percentage of endotracheal intubation and thus increased intrathoracic pressure before measurement (83%). In the EGDT study, less than 55% needed intubation at admission.

As mentioned earlier, in the present study the patients were treated in the ED, or elsewhere, before admission to the ICU. This treatment was different from the treatment given in the EGDT study. Nevertheless, our patients received some fluid therapy. Transfusion of red blood cells in our EDs was based on clinical suspicion or evidence of severe blood loss and not on low hematocrit only. Also, a main principle of treatment is to improve oxygen delivery and this could contribute to higher ScvO₂ values in the ICU population compared to the patients described in the EGDT study and other ED studies. Other interventions such as sedation and analgesia, most often to facilitate endotracheal intubation and ventilation, might have been beneficial for the balance between oxygen delivery and oxygen demand. Also, the trend of changing ScvO₂ and other physiological values influencing outcome 2324 is not taken into account in our study. Of course, all these factors are important differences between ER populations and ICU populations, but are not predominating. We are aware that comparison between those populations is limited by the above-mentioned differences.

As a result of the study design, statements about cut-off S(c)vO₂ values for outcome prediction 2024 or impact on therapeutic intervention are not possible. Also, we did not investigate the use of vital signs as indicator of tissue oxygenation in comparison to mixed or central venous saturation. Lack of clear insight of treatment and time spent at the different EDs, operating theatres or wards is a limitation of our study as well. Nevertheless, since we also aimed at the usefulness of measuring ScvO₂ or SvO₂ on ICU admission, we think these factors are not pertinent to the results. For example, Bracht et al. 20 found no correlation between ScvO₂ values and length of hospital stay before unplanned ICU admission.

Comparing our sepsis population with the ED population described in the important study by Rivers 8 is meant to be purely illustrative. Obviously, as we described, there are differences between ED and ICU populations in general. But there are also, depending on geographical setting, important differences between populations and health care systems. Therefore, we subscribe to the view of Ho et al. 25 that the syndrome described in the EGDT trial may be relatively uncommon depending on geographical setting and health care system. However, this does not undermine the importance of early identification of patients at high risk for cardiovascular collapse. For example, in our study of the 14 patients with a ScvO₂ < 50%, the in-hospital mortality was 57%.

Finally, the in-hospital mortality in our study was 32.0% for the total population and 27.0% for the patients with severe sepsis or septic shock. Again this reflects recent findings by others: Ho et al. (30.2%) 25 and Shapiro et al. (26.9%) 26.

In conclusion, the incidence of low ScvO₂ values for acutely admitted critically ill patients is low in Dutch ICUs. This is especially true for patients with sepsis/septic shock.

The incidence of low ScvO₂ values of acutely admitted critically ill patients is low in Dutch ICUs. This is especially true for patients with sepsis or septic shock.

In our setting, use of ScvO₂-guided resuscitation may only be helpful in a small subset of sepsis.

Mean SvO₂ values and mean ScvO₂ values in acutely admitted critically ill patients, including patients with severe sepsis or septic shock, were normal in our ICUs.

AMC = Amsterdam Medical Center; APACHE = acute physiology, age and chronic health evaluation; CCO = continuous cardiac output; CI = confidence interval; CVC = central venous catheter; CVP = central venous pressure; DO₂ = (systemic) oxygen delivery; ED = emergency department; EGDT = early goal directed therapy; GH = Gelre Hospital; Hct = hematocrit; ICU = intensive care unit; MAP = mean arterial pressure; MCL = Medical Center Leeuwarden; MODS = multiple organ dysfunction syndrome; PAC = pulmonary artery catheter; SAP = systolic arterial pressure; SAPS = simplified acute physiology score; ScvO₂ = central venous oxygen saturation; SvO₂ = mixed venous oxygen saturation; S(c)vO₂ = mixed/central venous oxygen saturation; SOFA = sequential organ failure assessment; VO₂ = (systemic) oxygen consumption.

The authors declare that they have no competing interests.

PB drafted the manuscript, participated in coordination, and performed statistical analysis. JH was responsible for acquisition of patient data in AMC. MS participated in the design of the study and helped to draft the manuscript. CB provided general support. PS participated in the design of the study, was responsible for acquisition of patient data. in GH and helped to draft the manuscript. MK conceived of the study and participated in its design and coordination and helped to draft the manuscript.