The authors propose that for a moderate reduction of perfusion during progressive irreversible ischemia, oxygen extraction increases to maintain aerobic metabolism, and arteriojugular oxygen difference (AJDo2) increases. Because of reduced carbon dioxide washout, venoarterial difference in carbon dioxide tension (DPco2) increases, with no change in the DPco2/AJDo2 ratio. With further reduction of cerebral perfusion, the aerobic metabolism will begin to decrease, AJDo2 will decrease while DPco2 will continue to increase, and the ratio will increase. When brain infarction develops, the metabolism will be abated, no oxygen will be consumed, and no carbon dioxide will be produced.


The authors studied 12 patients with acute cerebral damage that evolved to brain death and collected intermittent arterial and jugular blood samples.


Four patterns were observed: (1) AJDo2 of 4.1 +/- 0.7 vol%, DPco2 of 6.5 +/- 1.9 mmHg, and a ratio of 1.55 +/- 0.3 with cerebral perfusion pressure of 62.5 +/- 13.4 mmHg; (2) a coupled increase of AJDo2 (5.8 +/- 0.7 vol%) and DPco2 (10.1 +/- 1.0 mmHg) with no change in ratio (1.92 +/- 0.14) and cerebral perfusion pressure (57.9 +/- 5.8 mmHg); (3) AJDo2 of 4.7 +/- 0.4 vol% with an increase in DPco2 (11.8 +/- 1 mmHg) and correspondingly higher ratio (2.7 +/- 0.2); in this phase, cerebral perfusion pressure was 39.7 +/- 10.5 mmHg; (4) immediately before diagnosis of brain death (cerebral perfusion pressure, 17 +/- 10.4 mmHg), there was a decrease of AJDo2 (1.1 +/- 0.1 vol%) and of DPco2 (5.3 +/- 0.6 mmHg) with a further ratio increase (5.1 +/- 0.8).


Until compensatory mechanisms are effective, AJDo2 and DPco2 remain coupled. However, when the brain's ability to compensate for reduced oxygen delivery is exceeded, the ratio of DPco2 to AJDo2 starts to increase.

ISCHEMIA is a common pathway of damage after acute cerebral injury.1,2Different techniques have been used to detect ischemia at the bedside, including continuous cerebral perfusion pressure (CPP) monitoring and measurements of the arteriojugular oxygen difference (AJDo2). CPP is the physiologic variable that defines the pressure gradient driving cerebral blood flow (CBF). AJDo2provides information about the ratio between the cerebral metabolic rate of oxygen and CBF. High AJDo2(> 8.5 vol%) is indicative of ischemia. However, cerebral ischemia and infarction can develop even at a low AJDo2,3highlighting the need for better bedside measurements. Studies in the systemic circulation have shown that ischemia can also be revealed by changes in the venoarterial carbon dioxide tension difference, because when blood flow decreases, the difference between mixed venous and arterial carbon dioxide tension (Pco2) increases.4,5Progression to brain death represents a condition of progressive irreversible global cerebral ischemia. Combining the measurements of venoarterial Pco2(DPco2) and AJDo2, various phases can be hypothesized in the progression to brain death. For moderate cerebral perfusion reduction, aerobic metabolism will be preserved; oxygen extraction will increase, with a subsequent increase in AJDo2. Similarly, because the same amount of carbon dioxide will be cleared by less flow, DPco2will increase proportionally; in this phase, AJDo2and DPco2changes should therefore remain coupled, and their ratio should remain stable. With a further decrease in cerebral perfusion, the aerobic metabolism will begin to fail, so AJDo2will start to decrease while DPco2will continue to increase, and the ratio will increase. When brain death develops, a condition of CBF arrest identified with the irreversible cessation of all brain functions, the metabolism will be abated, no oxygen will be consumed, and no carbon dioxide will be produced, and in this state, because there is no CBF, we hypothesize that jugular sampling will no longer be indicative of the venous cerebral drainage.

Demographic Characteristics

Patients (n = 12) admitted to the Neuroscience intensive care unit of the Ospedale Maggiore Policlinico of Milan, Italy, from August 1998 to December 2003 whose cases evolved to brain death were studied. Brain death was identified as the irreversible cessation of all brain functions and was diagnosed according to national law, which requires the following criteria: (1) deep coma (Glasgow Coma Scale score of 3); (2) absence of brainstem reflexes; (3) positive apnea test result (absence of respiratory drive with Pco2> 60 mmHg and pH < 7.4); and (4) electroencephalography documenting the absence of spontaneous or induced brain electric activity in a normothermic, nonsedated patient. Brain death was confirmed over a 6-h interval, by a committee consisting of a neurologist with special expertise in electro-encephalography, a forensic medicine doctor, and an anesthesiologist.6 

Patient characteristics are shown in table 1.

Table 1. Patient Characteristics 

Table 1. Patient Characteristics 
Table 1. Patient Characteristics 

This study was approved by the local human studies committee (Institutional Review Board, Milan, Italy), and patients' relatives were informed that the data collected would be used for research purposes.

Clinical Management

All patients were treated according to a predefined protocol.7Space-occupying lesions were evacuated promptly by surgery; serial computerized tomography scans were repeated every 2 days or in case of clinical deterioration. All patients were intubated and mechanically ventilated. Neurologic examinations were performed hourly by nursing personnel and at least three times daily by the medical staff. Intracranial pressure, mean arterial pressure, and CPP were continuously monitored in all except two patients. During the study, none of the patients were given sedatives.

Catecholamines (noradrenaline up to 0.2 μg · kg−1· min−1and dopamine up to 10 μg · kg−1· min−1) were used in euvolemic patients to avoid hypotension. Anemia (hemoglobin < 9 g/dl) was corrected by erythrocyte transfusions.

Data Collection

All patients had an arterial line (usually radial), and cerebral venous blood was obtained through a catheter inserted into the superior jugular bulb, usually on the side of the main brain lesion.8,9In patients with diffuse injury, the right side was preferred.10In case of bone decompression (two patients), the jugular vein contralateral to the bone flap was cannulated. The proper position of the catheter was always checked by x-ray in the anteroposterior and lateral positions.

Arterial and jugular blood was sampled intermittently (mean interval, 125 ± 9 min) and analyzed by an IL-1620 Blood Gas Analyzer and an IL-682 Co-Oximeter (both from Instrumentation Laboratory, Lexington, MA). The following parameters were measured for arterial (a) and jugular (j) samples: (1) pH (apH/jpH), (2) oxygen tension (Pao2/Pjo2), (3) carbon dioxide tension (Paco2/Pjco2), and (4) oxygen saturation (Sao2/Sjo2). The following parameters were calculated: AJDo2=[(Sao2− Sjo2) × hemoglobin × 1.34 + (Pao2− Pjo2) × 0.003], DPco2= Pjco2− Paco2, ratio = DPco2/AJDo2.

In a separate group of six brain-injured patients, the external jugular vein was cannulated. Blood samples were taken in triplicate to determine the external jugular (ej) saturation and carbon dioxide and to calculate AejDO2, DPejco2, and the ej ratio, as for the internal jugular vein.

Observation Time

It is rarely possible to prespecify which patients, despite active treatment, will proceed to brain death. Therefore, we must clarify as we proceed. We collect measurements prospectively in all severely brain-injured patients. Goals of this data collection are oriented to the issue of oxygen consumption and tissue viability11in survivors while the goals are oriented to define patterns of progressive hypoperfusion in patients who proceed to brain death. Even if this is a prospective data collection, we can study the progression toward brain death only afterward. For the purpose of this study, we only considered samples collected in the 72 h preceding brain death.


We established four patterns. Based on previous data, we classified as “normal pattern” the concurrent presence of AJDo2of 4.56 ± 0.20 vol%, and DPco2of 8.05 ± 0.29 mmHg.11In the absence of references for numeric ranges, we used qualitative criteria to define “compensated hypoperfusion,”“uncompensated hypoperfusion,” and “death” pattern as follows: A coupled increase of AJDo2and DPco2above these values was classified as compensated hypoperfusion. A DPco2increase not accompanied by a corresponding increase in AJDo2was classified as uncompensated hypoperfusion. The death pattern involved a decrease of both AJDo2and DPco2.

Statistical Analysis

Continuous variables are expressed as mean ± SD. Changes in AJDo2, DPco2, their ratio, and CPP were compared across the normal, compensated, and uncompensated patterns by analysis of variance. Blocks were used to diminish the effect of variance among subjects (SAS System, version 8; SAS Institute, Inc., Cary, NC); the block effect was considered random. In case of significance (P < 0.05), the Tukey-Kramer test was used for post hoc  analysis. We assumed that because of the CBF arrest, jugular sampling during the death pattern would no longer be indicative of the mixed cerebral blood but of extracerebral venous drainage. The death pattern was therefore considered separately, the unpaired t  test being used to assess differences between the death pattern and external jugular samples.

On average, monitoring started 36 h before brain death (range, 3–72 h). A total of 206 arterial and jugular samples were collected.

AJDo2, DPco2, and Their Ratio

We observed four common patterns of AJDo2, DPco2, and their ratio: (1) AJDo2of 4.1 ± 0.7 vol%, DPco2of 6.5 ± 1.9 mmHg, and a ratio of 1.55 ± 0.3, with CPP of 62.5 ± 13.4 mmHg, corresponding to the normal pattern11; (2) a coupled increase of AJDo2(5.8 ± 2.2 vol%) and DPco2(10.1 ± 3.5 mmHg) with no change in the ratio (1.92 ± 0.4), obtained with CPP of 57.9 ± 15.4 mmHg, corresponding to the compensated hypoperfusion pattern; (3) AJDo24.7 ± 1.5 vol% with an increase in DPco2(11.8 ± 3.5 mmHg) and a subsequently higher ratio (2.7 ± 0.8) obtained with CPP of 34.7 ± 29.5 mmHg, corresponding to the uncompensated hypoperfusion pattern; (4) immediately before brain death was diagnosed, there was a decrease of AJDo2(1.1 ± 0.4 vol%) and (to a lesser degree) DPco2(5.3 ± 1.9 mmHg), with a further increase in the ratio (5.1 ± 2.6). In this phase (brain death pattern), CPP was 17 ± 27.6 mmHg.

Distribution of Patterns

Seven of the 12 patients showed at least one normal pattern; compensated hypoperfusion was identified in 10 patients, and all patients experienced uncompensated hypoperfusion. The death pattern was consistently observed immediately before brain death and was detected in all except one patient in whom this drop was missed, probably because of the intermittent sampling.

Time Course

Normal and compensated hypoperfusion patterns occurred close to each other, 34.6 ± 13.3 and 29.9 ± 14.6 h before brain death, respectively. Uncompensated hypoperfusion and the death pattern occurred 14.57 ± 8.27 and 2.07 ± 2.94 h before brain death. The different patterns progressed sequentially over time from compensated hypoperfusion to uncompensated hypoperfusion and ultimately brain death.

Progression to Uncompensated Hypoperfusion

Arteriojugular oxygen difference showed no significant changes across the various patterns. DPco2showed a significant increase from normal to compensated (P = 0.03) and uncompensated (P < 0.01) hypoperfusion. The ratio increased steeply in the uncompensated phase, with a significant difference from both the normal state (P < 0.01) and compensated hypoperfusion (P < 0.01). CPP showed no significant changes between the normal and compensated hypoperfused states, but all patients in the uncompensated state had a critical CPP (P < 0.05; fig. 1).

Fig. 1. Mean (± SD) arteriojugular oxygen difference (AJDo2) (  A ), venoarterial difference in carbon dioxide tension (DPco2) (  B ), DPco2/AJDo2ratio (ratio) (  C ), and cerebral perfusion pressure (CPP) (  D ) across the different patterns: normal (N), compensated hypoperfusion (C), uncompensated hypoperfusion (U), death (D). *  P < 0.05, **  P < 0.01. 

Fig. 1. Mean (± SD) arteriojugular oxygen difference (AJDo2) (  A ), venoarterial difference in carbon dioxide tension (DPco2) (  B ), DPco2/AJDo2ratio (ratio) (  C ), and cerebral perfusion pressure (CPP) (  D ) across the different patterns: normal (N), compensated hypoperfusion (C), uncompensated hypoperfusion (U), death (D). *  P < 0.05, **  P < 0.01. 

Close modal

Brain Death

Arteriojugular oxygen difference, DPco2, and ratios during brain death were no different from those measured in the external jugular vein of a separate group of brain-injured patients (fig. 2). In this group, AejDO2was 1.33 ± 0.33 vol%, DPejco2was 5.72 ± 2.27 mmHg, and the ej ratio was 4.9 ± 1.7. The high ratio is the effect of an extremely low AJDo2despite DPco2in the normal range.

Fig. 2. Mean (± SD) arteriojugular oxygen difference (AJDo2) (  A ), venoarterial difference in carbon dioxide tension (DPco2) (  B ), and DPco2/AJDo2ratio (ratio) (  C ) of the internal jugular vein (  left gray column ) and external jugular vein (  central gray column ) of the control patients and internal jugular vein of brain-death patients (  white column ). *  P < 0.05. 

Fig. 2. Mean (± SD) arteriojugular oxygen difference (AJDo2) (  A ), venoarterial difference in carbon dioxide tension (DPco2) (  B ), and DPco2/AJDo2ratio (ratio) (  C ) of the internal jugular vein (  left gray column ) and external jugular vein (  central gray column ) of the control patients and internal jugular vein of brain-death patients (  white column ). *  P < 0.05. 

Close modal

The main finding of this study is that the progression to brain death is marked by distinct patterns of AJDo2, DPco2, and their ratio. When cerebral perfusion initially decreases, there is a coupled increase of AJDo2and DPco2, with no change in the ratio. A further, critical decrease of cerebral perfusion, however, is indicated by an increase in the ratio. These findings are consistent with the transition from compensated to uncompensated hypoperfusion leading to changes in the redox state of the cell and anaerobic metabolism. At this stage, changes in pH, reflecting changes in buffer base, account for most of the increase in jugular Pco2and DPco2.12,13Therefore, although AJDo2may indicate the adequacy of CBF relative to oxidative cerebral metabolic requirements as long as oxygen delivery reductions can be compensated by increases in cerebral oxygen extraction, the ratio between DPco2and AJDo2may detect uncompensated hypoperfusion.

Studies in the systemic circulation have already highlighted the importance of the venoarterial carbon dioxide difference during ischemia.5,14In addition, the ratio of venoarterial Pco2difference over arteriovenous oxygen content has been widely used to assess the severity of systemic organ hypoxia and hypoperfusion. In 89 critically ill patients with a pulmonary artery catheter, the ratio, compared with DPco2, AJDo2, and SVo2, was the best predictor of hyperlactatemia (arterial lactate > 2 mm), with 79% sensitivity and 84% specificity.

In the experimental animal setting, our group first described AJDo2and DPco2in the progression of global cerebral ischemia in a swine model of stepwise CBF reduction, where a small reduction of CBF (50–60% of baseline) was associated with increases of both AJDo2(from 5.9 to 7.01 vol%) and DPco2(from 10 to 14.5 mmHg). When CBF was further reduced (20–30% of baseline), there was only a slight increase in AJDo2(from 7.01 to 8.17 vol%) but a steep increase in DPco2. Although the ratio was not calculated in that article, the numbers point to a slight increase in the ratio (22%) during mild CBF impairment, and a definite increase (85%) when CBF was severely reduced.15,16In humans, a DPco2increase indicative of global cerebral ischemia was described in a case report by Chieregato et al.  17Similar results (Pco2increases) in the cerebral tissue during progressive brain ischemia have been shown with focal gas tissue measurement techniques.18 

To our knowledge, however, this is the first study combining AJDo2, DPco2, and their ratio. A weakness of this study is that we did not measure CBF, so most of our interpretation is speculative. We purposely selected a group of patients in whom a “normal” baseline could be identified, with data suggesting adequacy of CBF, and who deteriorated, through progressive CPP reduction, to a state of no CBF, as documented by the instrumental and clinical signs of brain death. Therefore, even without CBF measurement, there was presumably a progression from preserved flow to no flow in these patients.

Death Pattern

In 1973, Minami et al.  19studied 40 patients with severe head trauma. The 19 who survived had higher AJDo2(4.41%) than patients in a state of brain death (1.86 vol%). In addition, the average DPco2was 7.9 mmHg in the survivors and 4.8 mmHg in the brain death group. Our data confirm these findings: Average AJDo2in the brain death state was never higher than 1.78 vol%, and DPco2was significantly lower than in the uncompensated state. When the brain dies, jugular sampling becomes representative of extracerebral venous drainage, with an extremely low AJDo2but DPco2in the normal range, giving a ratio above the physiologic range for brain tissue. Interestingly, the ratio calculated in the external jugular vein was above 2.1, which is usually considered as the upper physiologic limit. A high ratio can be explained by either a high DPco2or a very low AJDo2. It follows that physiologic ranges should be characterized depending on the metabolic rate of the tissue explored.

The adequacy of cerebral perfusion is better disclosed combining CPP with AJDo2, DPco2, and their ratio. Until compensatory mechanisms come into effect, AJDo2and DPco2remain coupled, but when the brain's ability to compensate for reduced oxygen delivery is exceeded, the ratio between DPco2and AJDo2starts to increase. These two “new” indicators of decompensated ischemia could be useful for early bedside monitoring of patients with severe cerebral damage.

The authors thank Angelo Colombo, M.D. (Staff Anesthesiologist, Department of Anesthesia and Critical Care Medicine, Ospedale Maggiore Policlinico IRCCS, Milano, Italy), for his skillful contribution to the statistical analysis.

Graham DI, Adams JH, Doyle D: Ischaemic brain damage in fatal non-missile head injuries. J Neurol Sci 1978; 39:213–34
Jones PA, Andrews PJ, Midgley S, Anderson SI, Piper IR, Tocher JL, Housley AM, Corrie JA, Slattery J, Dearden NM: Measuring the burden of secondary insults in head-injured patients during intensive care. J Neurosurg Anesthesiol 1994; 6:4–14
Robertson CS, Narayan RK, Gokaslan ZL, Pahwa R, Grossman RG, Caram Jr P, Allen E: Cerebral arteriovenous oxygen difference as an estimate of cerebral blood flow in comatose patients. J Neurosurg 1989; 70:222–30
Mekontso-Dessap A, Castelain V, Anguel N, Bahloul M, Schauvliege F, Richard C, Teboul JL: Combination of venoarterial PCO2 difference with arteriovenous O2 content difference to detect anaerobic metabolism in patients. Intensive Care Med 2002; 28:272–7
Zhang H, Vincent JL: Arteriovenous differences in PCO2 and pH are good indicators of critical hypoperfusion. Am Rev Respir Dis 1993; 148:867–71
Wijdicks EF: The diagnosis of brain death. N Engl J Med 2001; 344:1215–21
Stocchetti N, Penny KI, Dearden M, Braakman R, Cohadon F, Iannotti F, Lapierre F, Karimi A, Maas Jr A, Murray GD, Ohman J, Persson L, Servadei F, Teasdale GM, Trojanowski T, Unterberg A: Intensive care management of head-injured patients in Europe: A survey from the European brain injury consortium. Intensive Care Med 2001; 27:400–6
Metz C, Holzschuh M, Bein T, Woertgen C, Rothoerl R, Kallenbach B, Taeger K, Brawanski A: Monitoring of cerebral oxygen metabolism in the jugular bulb: Reliability of unilateral measurements in severe head injury. J Cereb Blood Flow Metab 1998; 18:332–43
Shenkin HA, Harmel MH, Kety SS: Dynamic anatomy of the cerebral circulation. Arch Neurol Psychiatr 1948; 60:240–54
Cormio M, Robertson CS: Ultrasound is a reliable method for determining jugular bulb dominance. J Neurosurg Anesthesiol 2001; 13:250–4
Stocchetti N, Canavesi K, Magnoni S, Valeriani V, Conte V, Rossi S, Longhi L, Zanier ER, Colombo A: Arterio-jugular difference of oxygen content and outcome after head injury. Anesth Analg 2004; 99:230–4
Sun XG, Hansen JE, Stringer WW, Ting H, Wasserman K: Carbon dioxide pressure-concentration relationship in arterial and mixed venous blood during exercise. J Appl Physiol 2001; 90:1798–810
Siesjo BK, Ekholm A,Katsura K, Theander S: Acid-base changes during complete brain ischemia. Stroke 1990; 21:III194–9
Johnson BA, Weil MH: Redefining ischemia due to circulatory failure as dual defects of oxygen deficits and of carbon dioxide excesses. Crit Care Med 1991; 19:1432–8
Rossi S, Stocchetti N, Longhi L, Balestreri M, Spagnoli D, Zanier ER, Bellinzona G: Brain oxygen tension, oxygen supply, and oxygen consumption during arterial hyperoxia in a model of progressive cerebral ischemia. J Neurotrauma 2001; 18:163–74
Rossi S, Colombo A, Magnoni S, Roncati EZ, Conte V, Stocchetti N: Cerebral veno-arterial pCO2 difference as an estimator of uncompensated cerebral hypoperfusion. Acta Neurochir Suppl 2002; 81:201–4
Chieregato A, Zoppellari R, Targa L: Cerebral arteriovenous PCO2 difference and early global cerebral ischemia in a patient with acute severe head injury. J Neurosurg Anesthesiol 1997; 9:256–62
Hoffman WE, Charbel FT, Edelman G: Brain tissue oxygen, carbon dioxide, and pH in neurosurgical patients at risk for ischemia. Anesth Analg 1996; 82:582–6
Minami T, Ogawa M, Sugimoto T, Katsurada K: Hyperoxia of internal jugular venous blood in brain death. J Neurosurg 1973; 39:442–7