Skip to main content

Low central venous saturation predicts poor outcome in patients with brain injury after major trauma: a prospective observational study



Continuous monitoring of central venous oxygen saturation (ScvO2) has been proposed as a prognostic indicator in several pathological conditions, including cardiac diseases, sepsis, trauma. To our knowledge, no studies have evaluated ScvO2 in polytraumatized patients with brain injury so far. Thus, the aim of the present study was to assess the prognostic role of ScvO2 monitoring during first 24 hours after trauma in this patients' population.


This prospective, non-controlled study, carried out between April 2006 and March 2008, was performed in a higher level Trauma Center in Florence (Italy). In the study period, 121 patients affected by major brain injury after major trauma were recruited. Inclusion criteria were: 1. Glasgow Coma Scale (GCS) score ≤ 13; 2. an Injury Severity Score (ISS) ≥ 15. Exclusion criteria included: 1. pregnancy; 2. age < 14 years; 3. isolated head trauma; 4. death within the first 24 hours from the event; 5. the lack of ScvO2 monitoring within 2 hours from the trauma. Demographic and clinical data were collected, including Abbreviated Injury Scale (AIS), Injury Severity Score (ISS), Simplified Acute Physiologic Score II (SAPS II), Marshall score. The worst values of lactate and ScvO2 within the first 24 hours from trauma, ICU length of stay (LOS), and 28-day mortality were recorded.


Patients who deceased within 28 days showed higher age (53 ± 16.6 vs 43.8 ± 19.6, P = 0.043), ISS core (39.3 ± 14 vs 30.3 ± 10.1, P < 0.001), AIS score for head/neck (4.5 ± 0.7 vs 3.4 ± 1.2, P = 0.001), SAPS II score (51.3 ± 14.1 vs 42.5 ± 15, P = 0.014), Marshall Score (3.5 ± 0.7 vs 2.3 ± 0.7, P < 0.001) and arterial lactate concentration (3.3 ± 1.8 vs 6.7 ± 4.2, P < 0.001), than survived patients, whereas ScvO2 resulted significantly lower (66.7% ± 11.9 vs 70.1% ± 8.9 vs, respectively; P = 0.046). Patients with ScvO2 values ≤ 65% also showed higher 28-days mortality rate (31.3% vs 13.5%, P = 0.034), ICU LOS (28.5 ± 15.2 vs 16.6 ± 13.8, P < 0.001), and total hospital LOS (45.1 ± 20.8 vs 33.2 ± 24, P = 0.046) than patients with ScvO2 > 65%.


ScvO2 value less than 65%, measured in the first 24 hours after admission in patients with major trauma and head injury, was associated with higher mortality and prolonged hospitalization.


Organ and tissue damages caused by a trauma impact lead to the development of systemic inflammatory response syndrome (SIRS) [1]. The local and systemic release of inflammation mediators produces oxidative stress, capillary leakage, microcirculatory disturbances, metabolic alteration, imbalance of pro- and anti-inflammatory mechanisms, ischaemia/reperfusion injury, coagulation disturbances [1]. In major trauma, occult tissue hypoperfusion within the first 24 hours after event precedes multiple organ dysfunction syndrome (MODS) [2]. Despite the unquestionable usefulness of routinely monitoring of hemodynamic parameters, such as arterial lactate, central venous pressure, blood pressure, heart rate, urinary volume, hypoxia may exist despite normal clinical values [3]. Continuous monitoring of central venous oxygen saturation (ScvO2) has been proposed as an indicator of tissue hypoperfusion. Although the reliability of ScvO2, compared with mixed venous oxygen saturation (SvO2), is still under discussion, particularly under shock conditions [3], the prognostic importance of low level (≤ 65%) of ScvO2 has been highlighted in severe sepsis [4, 5], myocardial infarction [6], cardiac failure [7], and trauma [8]. Moreover, ScvO2 can be easily monitored with a central venous line, whereas SvO2 requires a pulmonary artery catheterization.

In a recent investigation about the cause of death in trauma Soreige K et al.[9] found a high percentage of head trauma (67%) and concluded that focus on brain injuries prevention is imperative.

To our knowledge, no studies have evaluated ScvO2 in patients affected by brain injury after major trauma so far. Thus, the aim of the present investigations was to assess the useful of an early monitoring (first 24 hours) of ScvO2 in patients with major trauma and head injury.


Data collection

The study population was recruited from the Intensive Care Unit (ICU) of Careggi Teaching Hospital (Florence, Italy) from 1st April 2006 to 31st March 2008. The Careggi Teaching Hospital is an a university-affiliated tertiary care hospital of Tuscany (IT) and acts as the provincial higher level trauma center for trauma. The ICU is a 10-single bed multidisciplinary medical/surgical/trauma unit that cares primarily patients of Emergency Department.

Patients were treated according to international standard care [10] and to Advanced Trauma Life Support guide lines (see Additional file 1), and were hemodynamically stabilized according to vital signs such as heart rate, blood pressure, and central venous pressure. Other parameters considered in the evaluation of severity of hypovolemia were the need for vasoactive drugs after hemodynamic resuscitation and the number of blood units administered. Inclusion criteria were: 1. admission diagnosis for major trauma with head injury; 2. a Glasgow Coma Scale (GCS) score ≤ 13, evaluated in the pre-hospital setting before sedation; 3. an Injury Severity Score (ISS) ≥ 15; 4. Exclusion criteria included: 1. pregnancy; 2. age < 14 years; 3. isolated head trauma; 4. death within the first 24 hours from the event; 5. the lack of ScvO2 monitoring within 2 hours from the trauma. This prospective observational study, followed the principles of the Helsinki declaration and national ethical guidelines. The study was authorized by the Local Scientific committee and the Local Ethic Committee, which waved the need of the informed consent because it did not modify the therapeutic approach provided by the current guide lines and internal protocols.

Patients' demographic and clinical characteristics were extracted from institutional ICU-database (software: FileMaker Pro 5.5v2, FileMaker, Inc, USA). All data have been collected prospectively by two trained data collectors. Patients has been anonymized, thus privacy was guarantee. Severity of disease was estimated by the Simplified Acute Physiology Score II (SAPS II) [11], a gravity score validated in 1993, submitted to a recent revision [12] (whose score is easily calculable on the site, GCS, ISS, and AIS. The severity of the intracranial lesions was evaluated by the Marshall radiological score [13] (table 1) from the first CT-scan performed at hospital admission. Continuous monitoring of ScvO2 was performed with Vigileo system (Edwards Lifesciences, Irvine, California, USA). Central catheter was placed within the first two hours after admission in all patients enrolled. During the first 24 hours after trauma, both minimum value of ScvO2, measured for at least 15 minutes during the first 24 hours after trauma, and highest arterial lactate concentration were recorded. The data collectors verified that these values were not affected by episodic hypoxia during tracheal tube change, fibrobronchoscopy, occasional hypotension.

Table 1 Marshall Score [13].

Statistical Analysis

All analyses were performed using SPSS 10.0 statistical software package (SPSS, Chicago, Illinois, USA). Student's t-test was used for the numerical data, and two-tailed P value was considered significant if less than 0.05. Categorical data have been analysed with Fisher's exact test, with two-sided P significant if < 0.05. The cut-off values have been chosen based on statistically significant mean values of subgroups analysis. Multiple logistic regression has been performed on demographic and clinical parameters. A value of < 0.05 was considered significant.


Overall population

In the study period, 121 patients met the inclusion criteria. 255 patients were admitted to the Emergency Room with ISS > 15 and GCS < 13; of these 21 died in the first 24 hours, 63 did not confirm a low GCS at the secondary evaluation and, in 50 of these the insertion of venous central catheter in the first 2 hours from trauma was not possible. Overall patients' demographic and clinical characteristics and comparison between 28-days survived and deceased patients are summarised in Table 2. Patients enrolled had a prevalence of head/neck and chest injury, as resulted from AIS score (Table 2). Male sex was predominant in both groups and represents 78% of overall population. This gender prevalence is in accordance on what usually observed in our clinical practice. On average, deceased patients showed a significant higher age (53 ± 16.6 vs 43.8 ± 19.6, P = 0.043), ISS core (39.3 ± 14 vs 30.3 ± 10.1, P < 0.001), AIS score for head/neck (4.5 ± 0.7 vs 3.4 ± 1.2, P = 0.001), SAPS II score (51.3 ± 14.1 vs 42.5 ± 15, P = 0.014), and Marshall Score (3.5 ± 0.7 vs 2.3 ± 0.7, P < 0.001), if compared with survived patients. Consistently with these observations, survived and deceased patients resulted significantly different in arterial lactate concentration (3.3 ± 1.8 vs 6.7 ± 4.2, respectively; P < 0.001) and ScvO2 (70.1% ± 8.9 vs 66.7% ± 11.9, respectively; P = 0.046) (Table 2).

Table 2 Demographic and clinical characteristics of overall population and comparison between survived and deceased patients at 28 days.


The observed overall 28-days mortality rate was 18.2%. According with the differences between survived and deceased patients, contingency analysis showed that the significant relative risks for death were identified in age > 40 yrs, AIS (head/neck) ≥ 4, ISS score above 30, SAPS II score above 45, Marshall score > 2, lactate concentration > 3, and ScvO2 ≤ 65% (Table 3). A multiple regression analysis performed on parameters with a higher correlation with 28-days mortality showed that variables which mostly contributed to the prediction of mortality were Marshall score (P = 0.001) and lactate concentration (P = 0.002). The calculation of concordance rate between two multiple regression analysis showed that adding ScvO2 to the other variables considered (age, AIS for head/neck, ISS, SAPS II, Marshall score and lactate concentration), raised the rate of correct predictions of 1% (from 86.6% to 87.6%).

Table 3 Results of the relative risk analysis (95% confidence interval) related to 28-day mortality.

Subgroup analysis

Subgroup analysis of overall population, based on an ScvO2 value of ≤ 65%, showed that patients with worst ScvO2 had a significantly higher 28-days mortality rate (31.3% vs 13.5%, P = 0.034) respect to the comparison group (Table 4). Also total intra-ICU mortality (34.4% vs 13.5%, P = 0.017) and total intra-hospital mortality (37.5% vs 14.6%, P = 0.010) resulted significantly higher in ScvO2 ≤ 65% group respect to the ScvO2 > 65% group (Table 4). Moreover, the ScvO2 ≤ 65% group significantly differed in age (54.5 ± 18.3 vs 42.2 ± 18.8, P = 0.002), ISS score (36.1 ± 10.3 vs 30.5 ± 11.5, P = 0.018), total AIS score (12.94 ± 2.9 vs 11.3 ± 3.3, P = 0.016), AIS score for extremity (2.2 ± 1.7 vs 1.3 ± 1.2, P = 0.001), SAPS II score (51.1 ± 15.0 vs 41.6 ± 14.5, P = 0.002), and GCS score (5.7 ± 2.7 vs 7.8 ± 3.0, P = 0.001) from patients with ScvO2 > 65% (Table 4). Conversely, Marshall score and lactate values did not differ significantly between the two groups (Table 4).

Table 4 Comparison of demographic and clinical characteristics of overall patients.


The aim of the present investigation was to assess the role of early ScvO2 monitoring in trauma patients with associated head injury. We assumed that a persistent low ScvO2 could be the expression of the dangerous effects of generalized and, concomitant, brain hypoperfusion in patients with brain injury and major trauma and, therefore, that it could predict a worst evolution.

The prognostic importance of low level (≤ 65%) of ScvO2 has been emphasized in severe sepsis [4, 5], myocardial infarction [6], and cardiac failure [7]. The significance of ScvO2 in patients with major trauma has been previously reported in 26 injured patients [8]. In their investigation, Scalea and co-workers found that ScvO2 measurement was a trustworthy parameter to estimate blood loss, especially in patients who showed an ScvO2 below 65% despite stable clinical signs [8]. In opposition with this observation, a subsequent study did not confirm the reliability of ScvO2 as indicator of blood loss in 40 trauma patients [14]. According with Scalea et al., the lack of linear correlation between low ScvO2 and stable clinical signs has been also corroborated by Rady and colleagues in a nonrandomized study on sixteen patients presenting to the emergency department in shock conditions, who presented ScvO2 below 65% in almost 50% of cases, despite the hemodynamic stabilization [5].

The ScvO2 cut-off value of ≤ 65% has been chosen in the present investigation based on existing literature [5, 8, 14] and on the average we found in our patients deceased at 28 days (66.7%, Table 2). The main finding of this study was the relationship between an ScvO2 values ≤ 65%, for at least 15 minutes during the first 24 hours from trauma, and mortality rates (Table 4). It must be noted that in the group with lower ScvO2 also injury scores and lactate levels were worst. In regard, the absolute significance of ScvO2 in predicting mortality in the multivariate analysis was lower than Marshall score and lactate levels in our investigation. However, the inclusion of ScvO2 in a multiple regression model showed that mortality prediction can be improved from 86.6% to 87.6%: this improvement, whereas not ample, could be relevant in the day to day clinical practice. In consideration of our results, ScvO2should be never considered as a substitute of lactate monitoring or injury scores in predicting outcome. In addition to these observations, patients with ScvO2 values > 65% discharged alive from ICU, showed significantly reduction in both ICU and total hospital LOS (Table 4). Our findings are partially in accordance with a recent study performed on 96 critically ill patients admitted in a multidisciplinary ICU, in which an ScvO2 value below 60% at admission was related with high mortality rate but not with a prolonged ICU LOS [15]. On regard, it must be considered that the study of Bracht and co-workers was performed on unselected critically ill patients, with a not specified number of central nervous system injured patients (less than 30% of total patients enrolled), whereas our population included polytraumatized patients with head injury. In our study, the association between low level of ScvO2 and ICU/hospital LOS appeared strong; however, this finding could be also related with the differences in age and injuries observed in the two subgroups (Table 4).

The most numerous and extensive studies on ScvO2 optimization have been performed on septic patients. To our knowledge, no study has investigated the usefulness of ScvO2 monitoring to guide hemodynamic stabilization and to improve outcome in patients with major trauma [3]. The pathogenic changes following major trauma include SIRS, oxidative stress, capillary leakage, metabolic alterations, and diffuse tissue hypoxia as a result of circulatory abnormalities [1]. This pathophysiological process is somehow very similar with what observed in septic patients, although sepsis presents a different aetiology, evolution and possibility to be treated with antibiosis. In our investigation, ScvO2 was not included as a target parameter for hemodynamic optimization, so none can be gathered from our data on application of early-goal directed therapy protocol in major trauma. Nevertheless, considering the good evidence in patients with severe sepsis/septic shock, the use of ScvO2 to guide hemodynamic optimization in polytraumatized patients would merit to be adequately investigated.

Limits of present findings merit considerations. 1) We did not included data on fluids administration in the first 24 hours, brain damages evolutions, intracranial pressure, and their relationship with ScvO2 levels. Second, the exclusion of patients who died in the first 24 hours from the study protocol might be a potential bias. The reason of this choice depended from our goal to investigate the role of ScvO2 in the first 24 hours, which are considered critical in the evolution of MODS in polytrauma [2]. Third, we were not blinded for the primary end points (but in an emergency setting is not so easy), but the collection of data was made by an external investigator that was blind for the end points. Finally, the lack of a control group with a pulmonary artery catheter with SvO2 monitoring did not allow a comparison between ScvO2 and SvO2, although previous studies showed that changes of both values proceed in a parallel manner, even under shock conditions [3, 16].


In conclusion, an early monitoring of ScvO2 could aid to predict the outcome in patients with major trauma and head injury. In particular, we found that a value less than 65%, measured during the first 24 hours after admission, was associated with higher mortality rate and prolonged ICU and hospital LOS. So, our results seem to suggest the use of ScvO2 as a target of resuscitation in patients with brain injury after major trauma. The validation of ScvO2 in monitoring of cerebral metabolism and as a target for hemodynamic resuscitation in polytraumatized patients remains to be investigated.

Key messages

  • ScvO2 should be monitored in patients with major trauma, especially if head injury coexists.

  • ScvO2 value less than 65%, measured in the first 24 hours after admission, is associated with higher mortality and prolonged ICU/hospital LOS.

  • ScvO2 as a prognostic indicator should be considered in its relationship with typology of injuries and the other illness scores.



abbreviated injury scale


Glasgow coma scale


intensive care unit


injuryseverity score


length of stay, MODS: multiple organ dysfunction syndrome


simplified acute physiology score II


systemic inflammatory response syndrome


central venous oxygen saturation


mixed venous oxygen saturation.


  1. Keel M, Trentz O: Pathophysiology of polytrauma. Injury. 2005, 36: 691-709. 10.1016/j.injury.2004.12.037.

    Article  PubMed  Google Scholar 

  2. Blow O, Magliore L, Claridge JA, Butler K, Young JS: The golden hour and the silver day: detection and correction of occult hypoperfusion within 24 hours improves outcome from major trauma. J Trauma. 1999, 47: 964-969. 10.1097/00005373-199911000-00028.

    Article  CAS  PubMed  Google Scholar 

  3. Reinhart K, Bloos F: The value of venous oximetry. Curr Opin Crit Care. 2005, 11: 259-263. 10.1097/

    Article  PubMed  Google Scholar 

  4. Krafft P, Steltzer H, Hiesmayr M, Klimscha W, Hammerle AF: Mixed venous oxygen saturation in critically ill septic shock patients. The role of defined events. Chest. 1993, 103: 900-906. 10.1378/chest.103.3.900.

    Article  CAS  PubMed  Google Scholar 

  5. Rady MY, Rivers EP, Martin GB, Smithline H, Appelton T, Nowak RM: Continuous central venous oximetry and shock index in the emergency department: use in the evaluation of clinical shock. Am J Emerg Med. 1992, 10: 538-541. 10.1016/0735-6757(92)90178-Z.

    Article  CAS  PubMed  Google Scholar 

  6. Astiz ME, Rackow EC, Kaufman B, Falk JL, Weil MH: Relationship of oxygen delivery and mixed venous oxygenation to lactic acidosis in patients with sepsis and acute myocardial infarction. Crit Care Med. 1988, 16: 655-658. 10.1097/00003246-198807000-00001.

    Article  CAS  PubMed  Google Scholar 

  7. Ander DS, Jaggi M, Rivers E, Rady MY, Levine TB, Levine AB, Masura J, Gryzbowski M: Undetected cardiogenic shock in patients with congestive heart failure presenting to the emergency department. Am J Cardiol. 1998, 82: 888-891. 10.1016/S0002-9149(98)00497-4.

    Article  CAS  PubMed  Google Scholar 

  8. Scalea TM, Hartnett RW, Duncan AO, Atweh NA, Phillips TF, Sclafani SJ, Fuortes M, Shaftan GW: Central venous oxygen saturation: a useful clinical tool in trauma patients. J Trauma. 1990, 30: 1539-1543.

    Article  CAS  PubMed  Google Scholar 

  9. Søreide K, Krüger AJ, Vårdal AL, Ellingsen CL, Søreide E, Lossius HM: Epidemiology and contemporary patterns of trauma deaths: changing place, similar pace, older face. World J Surg. 2007, 31: 2092-103. 10.1007/s00268-007-9226-9.

    Article  PubMed  Google Scholar 

  10. Fowler RA, Adhikari NK, Scales DC, Lee WL, Rubenfeld GD: Update in critical care 2007. Am J Respir Crit Care Med. 2008, 177: 808-819. 10.1164/rccm.200801-137UP.

    Article  PubMed  Google Scholar 

  11. Le Gall JR, Lemeshow S, Saulnier F: A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA. 1993, 270: 2957-63. 10.1001/jama.270.24.2957.

    Article  CAS  PubMed  Google Scholar 

  12. Le Gall JR, Neumann A, Hemery F, Bleriot JP, Fulgencio JP, Garrigues B, Gouzes C, Lepage E, Moine P, Villers D: Mortality prediction using SAPS II: an update for French intensive care units. Crit Care. 2005, 9: 636-7. 10.1186/cc3821.

    Article  Google Scholar 

  13. Marshall LF, Marshall SB, Klauber MR, Van Berkum Clark M, Eisenberg H, Jane JA, Luerssen TG, Marmarou A, Foulkes MA: The diagnosis of head injury requires a classification based on computed axial tomography. J Neurotrauma. 1992, 9 (Suppl 1): S287-S292.

    PubMed  Google Scholar 

  14. Bannon MP, O'Neill CM, Martin M, Ilstrup DM, Fish NM, Barrett J: Central venous oxygen saturation, arterial base deficit, and lactate concentration in trauma patients. Am Surg. 1995, 61: 738-745.

    CAS  PubMed  Google Scholar 

  15. Bracht H, Hanggi M, Jeker B, Wegmuller N, Porta F, Tuller D, Takala J, Jakob SM: Incidence of low central venous oxygen saturation during unplanned admissions in a multidisciplinary intensive care unit: an observational study. Crit Care. 2007, 11: R2-10.1186/cc5144.

    Article  PubMed Central  PubMed  Google Scholar 

  16. Reinhart K, Kuhn HJ, Hartog C, Bredle DL: Continuous central venous and pulmonary artery oxygen saturation monitoring in the critically ill. Intensive Care Med. 2004, 30: 1572-1578. 10.1007/s00134-004-2337-y.

    Article  PubMed  Google Scholar 

Download references


The authors acknowledge Dr. Paola Nicoletti for her contributions to the revising and editing of the manuscript. The study was supported by institutional funds.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Alessandro Di Filippo.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

ADF and AP planned the study and reviewed the literature. ADF, CG and AP wrote the manuscript. MC and GZ performed the statistical analyses. GFG, RS, LP and CG participated in the design ofthe study. LP and CG collected data and helped draft the manuscript. GZ revised the manuscript. All authors have seen and approved the final revised version.

Electronic supplementary material


Additional file 1: Appendix on trauma. Major diagnostic and therapeutic procedures in the treatment of trauma in the Tuscany Region, Italy. (DOC 32 KB)

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Di Filippo, A., Gonnelli, C., Perretta, L. et al. Low central venous saturation predicts poor outcome in patients with brain injury after major trauma: a prospective observational study. Scand J Trauma Resusc Emerg Med 17, 23 (2009).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: