10 

Should I Monitor Jugular Venous Oxygen Saturation?

Nino Stocchetti

BRIEF ANSWER

Background

Pathophysiology

A normal adult human brain has a mean cerebral blood flow (CBF) of around 50 to 60 mL/100 g/min. If its normal hemoglobin (Hgb) content is fully saturated with oxygen, arterial blood carries ~20 mL of oxygen per deciliter (dL). The oxygen content of venous blood draining the brain varies. Under normal conditions, internal jugular (IJ) hemoglobin oxygen saturation ranges between 55% and 69%. Therefore, the normal arteriojugular difference in O₂ content (AJDO₂) is 6.3 2.4 mL/dL. According to the Fick principle, AJDO₂ is proportional to cerebral metabolic rate of oxygen consumption (CMRO₂) and inversely proportional to CBF.^1–4

Under normal conditions, AJDO₂ remains stable. If flow decreases or CMRO₂ increases, AJDO₂ rises as well, indicating that the brain is extracting more oxygen from the blood flowing through the cerebral circulation. When flow increases relative to cerebral oxygen consumption, AJDO₂ drops. High AJDO₂ (and, somewhat less accurately, low SjvO₂) has thus been interpreted as synonymous with cerebral ischemia, whereas low AJDO₂ is taken as a clear indicator of hyperemia. However, this is an oversimplification, and some points need to be clarified.

AJDO₂, Not Simply SjvO₂, Is Related to Flow and Metabolism

Although it is easier to use SjvO₂ instead of AJDO₂ for estimating the relationship between CMRO₂ and CBF, jugular saturation also depends on the Hgb concentration. Therefore, IJ saturation cannot be used independently of the Hgb value for estimating the status of cerebral metabolism, and AJDO₂ must be calculated. For example, low levels of jugular saturation may simply indicate anemia (in which the brain may extract more oxygen from the diluted blood), not inadequate flow.

There are heterogeneities in the distribution of flow and in oxygen consumption in the brain. AJDO₂ cannot detect problems in small areas of the brain, and it can miss important events, such as cerebral infarction, when they are of limited extent (class III data).5

Oxygen May Still Be Needed Even If It Is Not Extracted

There are limits to the brain’s ability to extract oxygen. When these capabilities are exhausted, no more oxygen can be removed from the blood, so AJDO₂ does not rise further. That does not imply there is no need for more oxygen. In this situation, AJDO₂ does not increase, and ischemia develops.

Oxygen extraction may be impaired, as in cases of mitochondrial dysfunction. AJDO₂ may be very low in this situation, or when a large portion of the brain does not extract oxygen at all (as in cerebral infarction).

For all these reasons, the simple extrapolation “low AJDO₂ indicates hyperemia” is simply not true.

Literature Review

Since the landmark paper by Obrist et al⁶ and further work by Robertson et al,² AJDO₂ has been used increasingly in monitoring severe head injury. It has been both praised as the optimal monitoring guide in head injury management⁷ and rejected as still unproven and experimental.⁸

Technical Aspects of Monitoring Jugular Saturation

ON WHICH SIDE SHOULD THE JUGULAR CATHETER BE INSERTED?

Jugular saturation may vary depending on whether the jugular catheter is positioned on the right or left side. Bilateral sampling has confirmed that the two sides generally provide comparable data. Nevertheless, differences that may sometimes be clinically relevant have been detected in a substantial percentage of patients (class II data).⁹

Such differences, which arise from drainage of cerebral areas with different values for blood flow and CMRO₂, likely occur more frequently when focal masses are present. For this reason, some clinicians choose to insert SjvO₂ catheters on the side of predominant intracerebral pathology.

For cases of diffuse damage, however, use of the larger jugular vein or “dominant IJ” has been proposed. Cannulation of the dominant IJ should guarantee that sampling takes place on the side of greater cerebral venous drainage. Different methods of identifying the dominant IJ have been reported, including unilateral compression of the neck during intracranial pressure (ICP) recording, with the side of the larger ICP increase considered to be the side of predominant venous drainage,³ or identification by computed tomography (CT) scan of the side with the larger jugular foramen.^9,10 In a randomized trial, ultrasound measurement was found useful for increasing the likelihood of successful jugular vein cannulation (class I data).¹¹ Although this study tested the IJ for central venous catheter placement, the findings can probably be applied to the retrograde jugular cannulation necessary for determining AJDO₂.

WHERE SHOULD THE TIP OF THE JUGULAR CATHETER BE PLACED?

The tip of the catheter must be in the upper portion of the superior jugular bulb. When the tip lies too inferiorly, samples are contaminated with blood draining the neck and the face, rather than the brain. On the other hand, even if the tip is advanced too far, reliable AJDO₂ data can still be obtained. A lateral skull radiograph is helpful for determining the location of the catheter tip.

Venous oxygen saturation in the sinuses is generally slightly lower than that in the jugular bulb.⁹ If a catheter is erroneously positioned in the inferior petrosal sinus, it may produce misleading data. The inferior petrosal sinus has a higher Hgb saturation than the jugular bulb, probably because it drains the anterior pituitary lobe and carries some extracranial blood, for example, from the superior ophthalmic vein.¹²

Thus, in practical terms, two aspects must be considered. When conspicuous unilateral damage exists, it may be reasonable to attempt to cannulate the ipsilateral jugular vein; this is not the case with diffuse damage (level III recommendation). The second crucial point is that after placement of the catheter, the position of its tip should be checked by x-ray.¹³

SLOW SAMPLING IS NECESSARY FOR ACCURATE MEASUREMENT OF SATURATION

When intermittent samples are used, the rate of blood withdrawal may affect IJ saturation. With fast withdrawal, higher saturations are detected, probably as a result of extracranial contamination (class II data). Thus, blood samples should be withdrawn slowly (level II recommendation).¹⁴

When optical catheters are used for continuous reading, they provide reliable data only if carefully maintained. Close attention must be paid to their position, light intensity, and other details. Although optical catheters have been reported to give a very low percentage of reliable data in some centers, others have described higher levels of confidence.4,15

COMPLICATIONS

Complications of jugular venous monitoring are rare (class III data). Carotid puncture has an incidence of 1 to 4.5%.^4,16 Infection is very rare, and no cases of severe infection due to jugular cannulation for jugular venous monitoring have been reported. Damage to the vein is more difficult to assess and to quantify, perhaps explaining the paucity of reports of this complication. In 20 patients systematically investigated with ultrasonography, eight (40%) had nonobstructive subclinical internal jugular vein thrombi after jugular bulb catheter monitoring (95% confidence interval, 19–61%). In this single series, therefore, the incidence of subclinical internal jugular vein thrombosis after jugular bulb catheter monitoring was considerable.¹⁶

Clinical Use of Jugular Saturation

JUGULAR SATURATION IN THE EARLY POSTINJURY PHASE

Episodes of high AJDO₂ are usually detected during the early phases after head injury. Although CMRO₂ is likely to be low during this period, CBF is depressed by an even greater extent. Hyperventilation is also frequently used, sometimes inadvertently.¹⁷ During rescue and transport to the trauma center, in fact, ventilation is not likely to be carefully titrated. Due to the combination of intra- and extracranial injuries, anemia is also frequent and lowers jugular saturation even if CBF is preserved.

Expansion of intracranial masses that cause intracranial hypertension and reduced cerebral perfusion can be detected by SjvO₂ monitoring. In 25 severely head-injured patients who underwent removal of intracranial traumatic hematomas, jugular saturation was low preoperatively but improved after surgical evacuation (class II data).¹⁸

HYPERVENTILATION

Hyperventilation is a powerful tool for controlling increased intracranial pressure. It lowers the intracranial blood volume by reducing cerebral blood flow.

Since avoiding cerebral ischemia is essential in head-injured patients, there seems to be conflict between the potential improvement in cerebral perfusion pressure obtained by hypocapnia and the undesirable side effect of flow reduction. Because AJDO₂ is very sensitive to changes in cerebrovascular resistance induced by hypocapnia, jugular vein monitoring may reveal inappropriate reductions of CBF caused by hyperventilation. Therefore, SjvO₂ monitoring should be used for titrating the level of hyperventilation (level III recommendation).¹⁹

ARTERIOVENOUS FISTULAS

Because arteriovenous fistulas are characterized by a rapid transit of arterial blood to the venous circulation, AJDO₂ measurements are not useful in such cases. However, extremely high values of jugular venous oxygen saturation, approaching arterial levels, have prompted cerebral angiography to confirm clinical suspicions that traumatic carotid-cavernous fistulas were present.^20,21

JUGULAR SATURATION, AJDO₂, AND THEIR ASSOCIATION WITH OUTCOME

Jugular venous desaturation (to less than 50%) has been correlated with poor outcome (class II data).²² If desaturation occurs while AJDO₂ is high, flow may be inadequate. It seems reasonable, therefore, to attempt to relate inadequacy of flow and brain damage. In fact, repeated episodes of jugular desaturation have been found to correlate with unfavorable outcome.²² However, when cerebral metabolism is very depressed, as in cases with profound damage to neuronal cells, oxygen consumption is very low.²³ In this situation AJDO₂ is very low too, suggesting impaired cerebral activity rather than excessive flow. Patients with unfavorable outcome have a lower AJDO₂ than patients with good results (class II data).^24,25