Secondary Logo

Journal Logo

Original Article

Changes of jugular venous blood temperature associated with measurements of cerebral blood flow using the transcerebral double-indicator dilution technique

Mielck, F.*; Bräuer, A.*; Radke, O.*; Hanekop, G.*; Loesch, S.*; Friedrich, M.; Hilgers, R.; Sonntag, H.*

Author Information
European Journal of Anaesthesiology: April 2004 - Volume 21 - Issue 4 - p 289-295


The transcerebral double-indicator dilution (TCID) technique is a recently developed method to measure global cerebral blood flow (CBF) at the bedside. In previous investigations good validity, accuracy and reliability for clinical purposes have been demonstrated for this method [1-3]. This method is based on bolus injection of ice-cold indocyanine green dye via a central venous line and the simultaneous recording of resulting thermo- and dye-dilution curves in the aorta and the jugular bulb using combined fibreoptic-thermistor catheters. CBF is calculated from the mean transit times of the indicators through the brain.

As mean transit times are derived from the indicator curves, only relative temperature changes are required for measurements. However, with this method a large amount of ice-cold solution is administered as a bolus. Presently, no data are available concerning the absolute blood temperatures at the respective measurement sites. Therefore, this prospective clinical study was performed to elucidate three objectives. First, is there a decrease in absolute blood temperature after the administration and recording procedure? Second, if a decrease in absolute blood temperature is detected, is there a recovery to baseline conditions within a definite time limit? Third, do possible alterations in absolute blood temperature influence CBF and metabolism?

Answers to these questions additionally may contribute to the understanding of the characteristic temperature drift artefact problem associated with the TCID method, which occasionally leads to the rejection of thermo-dilution curves from evaluation.


The study was approved by the local Ethics Committee and written informed consent was obtained from each patient. Nine male patients scheduled for elective coronary artery bypass grafting were investigated. Mean age was 62 yr (range 49-74 yr), body weight was 85 ± 13 kg and height was 173 ± 9 cm, respectively (mean ± SD). According to clinical and duplex-ultrasonic investigation of intra- and extracranial arteries, none of the patients showed evidence of cerebrovascular disease. Patients with a history or laboratory evidence of hepatic, renal or nervous system disease were excluded from the study. None of the patients had general anaesthesia 30 days prior to surgery. Individual cardiac medication such as antihypertensive agents, nitroglycerin or β-adrenoceptor blocking agents, or both, was continued until the day of surgery. Flunitrazepam (2 mg orally) premedication was given on the evening before and on the morning of surgery.

Anaesthesia and catheterization procedure

Before induction of anaesthesia, routine haemodynamic monitoring was established including electrocardiography (leads II and V5), and arterial and central venous catheterization. Induction of anaesthesia was performed by intravenous (i.v.) administration of fentanyl 7 μg kg−1 and midazolam 0.2 mg kg−1; tracheal intubation was facilitated by administration of pancuronium 0.15 mg kg−1. Anaesthesia was maintained with fentanyl 10 μg kg−1 h−1 and midazolam 0.15 μg kg−1 h−1. A jugular bulb catheter (4-FG, PV-2024; Pulsion Medical Systems, Munich, Germany) was inserted by retrograde cannulation of the right internal jugular vein in order to measure pressure, venous thermo- and dye-dilution kinetics, and to obtain blood samples simultaneously. The catheter was inserted until the resistance of the catheter tip meeting the roof of the jugular bulb was felt. Then the catheter was withdrawn 0.5 cm. The correct catheter position was verified by fluoroscopy. In addition, an arterial fibreoptic-thermistor catheter (4-FG, PV-2024, Pulsion Medical Systems) was placed in the thoracic aorta via the left femoral artery for measurement of the arterial dye- and thermo-dilution kinetics. Mechanical ventilation of the lungs was performed using a volume controlled ventilator (Cato, Dräger AG, Lübeck, Germany) with a gas flow rate of 6 L min−1 at an inspired oxygen fraction (FiO2) of 0.3 in air. A tidal volume of 10 mL kg−1 body weight and a respiratory rate of 10 breaths min−1 was initiated and a positive end-expiratory pressure of 7 cm H2O was applied in all patients. To avoid additional pharmacologically induced alterations of CBF and cerebral metabolism neither vasoactive drugs nor inhalational anaesthetics were given.

Study design

After induction of anaesthesia and the catheterization procedure a 20 min period for anaesthetic, haemodynamic and ventilatory stabilization, with a target partial pressure for arterial carbon dioxide (PaCO2) of approximately 5.3 kPa, was imposed before baseline measurements (measurement period, MP 1). Thereafter, consecutive measurements were repeated every 20 min (MP 2-MP 6) before the start of the operation. This 20 min interval prior to the next measurement was determined by the CBF calculation procedure, inspection of haemodynamic and blood-gas stability, and consecutive adjustments when necessary. CBF measurement was performed using the TCID technique. Absolute blood temperature was measured in jugular bulb and aorta before and after every CBF measurement. Before every CBF measurement arterial and central venous pressures were obtained. Thermo-dilution measurements of transpulmonary cardiac output (CO) were derived from every CBF measurement at random times during the respiratory cycle. Arterial and jugular venous blood samples were drawn for the determination of cerebral metabolism. Cerebral metabolic rate of oxygen (CMRO2) and glucose (CMRglc) were calculated according to the standard formula. To prevent hypothermia, a forced air warming system (WarmTouch® 5800) with CareQuilt® full body blanket (Mallinckrodt Medical Inc., St. Louis, MO, USA) was applied prior to induction of anaesthesia and during the complete period of measurements. The body temperature of the patients was measured in the urine bladder by a Mon-a-therm® Foley-temp 14-Fr/Ch catheter (Mallinckrodt Medical Inc.).

Measurement of absolute blood temperature

The TCID methodology is based on the digital recording of resulting thermo- and dye-dilution curves - via combined fibreoptic-thermistor catheters over opto-electronic devices from the aorta (COLD Z-021®; Pulsion Medical Systems) and the jugular bulb (IVF-5®; Pulsion Medical Systems) - and the determination of mean transit times for the dye and the thermal indicator dilution kinetics (software version COLD® V 5.4.3 CBF 2.0; Pulsion Medical Systems). Due to very small relative, absolute, temperature changes in the jugular bulb, the temperature stability of the measuring device is essential. Therefore, the accuracy of the complete device consisting of COLD® systems and fibreoptic-thermistor catheters was verified in vitro prior to this clinical investigation. For the range of 38.00-35.00°C measurements were performed in a temperature controlled water bath with a step-by-step temperature decrease of ΔT = 0.05°C in comparison to a standard certified mercury thermometer. The COLD® systems including catheters yielded a bias of −0.02°C and a precision of ±0.03°C. All clinically measured values were well within the range of linearity.

The CBF measurement device permits only the display of the absolute blood temperature on the COLD® Z-021 monitor. In order to simultaneously measure and display the absolute blood temperature in both locations a second COLD® Z-021 was connected to thermistor port of the catheter in the jugular bulb. This procedure was performed exactly before and after the TCID measurement.

Transcerebral double-indicator dilution technique

The measurement of CBF using the TCID technique is based on the simultaneous bolus application of the two indicators 'negative heat' and indocyanine green into a central vein (25 mg, 40 mL H2O, <5°C). Using the two fibreoptic-thermistor catheters placed in the thoracic aorta and the jugular bulb (see section on Anaesthesia and catheterization procedure), the resulting thermo- and dye-dilution curves were digitally recorded simultaneously at the arterial input and cerebrovenous output of the brain, and stored on a hard computer disc. Further analysis of the indicator dilution curves was performed off-line on a microcomputer using a Pascal software package [1].

Calculation of cerebral blood flow

The newly developed technique for measurement of CBF by double-indicator dilution is based on the mean transit time principle. Basically, CBF is calculated from the transcerebral mean transit time (mtttc) of a diffusible indicator and the partition coefficient λ between blood and brain tissue of the respective diffusible indicator:

In the current methodology, negative heat (i.e. ice-cold solution) is used as a highly diffusible indicator, which equilibrates nearly instantaneously with the brain tissue. With the thermo-dilution curve at the inflow of the brain (measured in the thoracic aorta) and the thermo-dilution curve at the outflow of the brain (measured in the jugular bulb), the transcerebral mean transit time can be calculated using a 'black box' analysis approach of both thermodilution curves [1]. As the water content of the brain is very high [4], the partition coefficient λtherm between the brain tissue and blood was assumed to be equal 1 mL g−1 and CBF was calculated as:


The calculation of CBF has been described in detail by Wietasch and colleagues [1].

Statistical analysis

All values presented in Tables and Figures are given as mean ± SD. An analysis of variance (ANOVA) for repeated measures was performed for differences between MPs considering all variables other than blood temperature. An advanced linear model for multiple within-subject repeated measures and multiple-dependent variables was performed for differences between MPs (MP 1-MP 6) and for differences between pre- and post-TCID as well as jugular-venous and aortic blood temperature. Results were considered significant at P < 0.05. Additional descriptive analysis was performed using the Statistica® software package (Kernel-Version 6.0.591®; StatSoft Inc., Tulsa, OK, USA).


In this investigation none of the nine patients experienced any complications during the measurement procedures. A total of 54 sets of data were accomplished. Seven CBF measurements (13%) were rejected because they were unsuitable for evaluation because of temperature drift artefact in the jugular bulb. Measured and calculated haemodynamic, metabolic, blood-gas and temperature data are presented in Table 1. During the complete investigated period MP 1-MP 6 heart rate (HR), CO and haemoglobin (Hb) concentration remained unchanged.

Table 1
Table 1:
Data of nine patients presented as mean ± SD.

With MP 1 regarded as baseline, mean arterial pressure (MAP) and systemic vascular resistance (SVR) showed a significant increase at MP 6. Although meticulous care was taken to stabilize the arterial carbon dioxide partial pressure (PaCO2), a marginal, yet significant increase was found at MP 4 (Table 1).

The temperature of the injected solution was unchanged throughout the complete periods of measurement and the temperature in the urinary bladder remained unchanged throughout the investigated period (Table 1).

The time course of the jugular venous blood temperature of every patient before the TCID measurement is demonstrated in Figure 1a. While three patients showed a steady but slow reduction in jugular venous blood temperature the remaining patients demonstrated an almost unchanged temperature trend. An almost identical observation can be displayed for the jugular venous blood temperature time course after the TCID measurement (Fig. 1b).

Figure 1
Figure 1:
Individual time course of absolute jugular venous blood temperature of nine patients (1-9) over MP 1-MP 6 (a) before and (b) after TCID measurement.

Statistical analysis revealed that the absolute blood temperature in the aorta was always significantly lower than in the jugular bulb. Furthermore, blood temperature in the aorta as well as the jugular bulb was always significantly higher before than after TCID measurement (Table 2, Fig. 2).

Table 2
Table 2:
Results of statistical analysis for multiple-dependent variables (R1) and multiple within-subject repeated measures, MP 1-MP 6 (R2).
Figure 2
Figure 2:
Absolute blood temperatures of nine patients at MP 1-MP 6. Demonstrated are: aorta: ●, mean - SD before TCID measurement; ○, mean - SD after TCID measurement; jugular bulb: ▾, mean + SD before TCID measurement; ▿, mean + SD after TCID measurement.

During the time interval prior to the next ice-cold bolus administration absolute blood temperature showed almost complete recovery from the preceding administration, however baseline level (MP 1) was never reached (Table 1, Fig. 2).

Individual CBF data do not present any systematic changes in association to blood temperature changes in the jugular bulb (Fig. 3). Mean CBF, CMRO2, CMRglc and jugular venous oxygen saturation (SjvO2) remained unchanged during the investigated period (Table 1).

Figure 3
Figure 3:
Individual time course of CBF of nine patients (1-9) over MP 1-MP 6.


This prospective clinical study was performed in order to elucidate the effects of repeated measurements using the TCID technique on absolute blood temperature, CBF and metabolism. The results of this study demonstrate that during the investigated time course the blood temperature in the jugular bulb was significantly higher when compared to the aorta, irrespective of the indicator administration. Furthermore, the administration of the ice-cold bolus reduces the blood temperature in the jugular bulb as well as the aorta after the TCID measurement procedure with a recovery almost to baseline conditions during the following time period. A clinical impact from these observations could not be demonstrated because global CBF and metabolism measurements remain unaltered.

Methodological considerations

It was a necessary prerequisite of this investigation to achieve steady-state conditions after induction of anaesthesia over a definite period of time, long enough to assess changes in absolute blood temperature and in a trend course after repetitive administration of ice-cold indicator boluses. Although meticulous care was taken to achieve a constant normocapnic condition by adjusting the ventilator setting throughout the study period, this intention was not completely successful as we noticed a slight PaCO2 increase in MP 4 which reached statistical significance. However, accuracy and resolution of the TCID technique is not high enough to detect an increase of CBF due to an elevation in PaCO2 in the order of 0.3 kPa. Furthermore, this alteration may be understood as minor impact, as the cerebrovenous oxygen saturation did not change in the corresponding period.

MAP increased during MP 6 in comparison to baseline measurement. A possible explanation for this observation may be seen in the slightly fading anaesthetic depth over time, although continuous anaesthetic drug administration explicitly remained unchanged. At this point a discrete increase in SVR was noticed, without a measurable effect on other global haemodynamic variables, e.g. HR and CO which remained unchanged. Nevertheless, this situation was estimated steady enough for a clinical investigation with these objectives.

Due to the fact that very small temperature amplitudes have to be detectable in the jugular bulb, the temperature stability of the measuring device is essential. Therefore, the accuracy of the complete device consisting of COLD® systems and fibreoptic-thermistor catheters was verified in vitro prior to this clinical investigation. The COLD® systems including catheters yielded a bias of −0.02°C and a precision of ±0.03°C. All clinically measured values were well within the range of linearity. This precision was estimated satisfactory for this investigation with a clinical objective.

One of the reasons for the rejection of 13% of CBF measurements, due to distorted curves, may be seen in the fact that to some extent the impairment of signal-to-noise ratio is probably caused by malposition or intermittent wall contact of the catheter tip in the jugular bulb. Although the correct position of the catheter was verified by fluoroscopy, we found that the minor adjustment of the catheter by 1-2 mm resulted in a better intraluminal position and an improvement of the dye as well as the temperature signal. Intrathoracic pressure and venous blood flow alterations may be the other reasons for temperature drifts in the jugular bulb flow changes during artificial ventilation of the lungs. During the ventilatory cycle, positive pressure ventilation causes changes in intrathoracic pressure that in turn may lead to inconsistent or reduced jugular blood flow and blood drainage out of the jugular bulb. Therefore, minor temperature fluctuations in the jugular bulb cannot be excluded as reason for impaired thermo-dilution curves.

Jugular bulb blood temperature

Cerebrovenous blood temperature is mainly influenced by the temperature of the arterial blood, the temperature of the brain determined by local heat production and the CBF. In clinical as well as experimental investigations the non-injured brain has been found to be 0.2-0.4°C warmer than the body core temperature [5-8]. This observation can be confirmed by results from this investigation where during the complete time course, the blood temperature in the jugular bulb was significantly higher than in the aorta. The repeated measurement of CBF by TCID therefore does not change the principle of the brain as a heat-producing organ.

The blood temperature in the jugular bulb after TCID measurement was lower than before the procedure. However, the temperature increased during the subsequent time interval of 20 min almost to baseline level prior to the next TCID measurement. The maintenance of a constant brain temperature is essentially dependent on a balance between energy metabolism and blood flow. An effect of ice-cold bolus application on the CMRO2 or CMRglc and CBF cannot be derived from the results of this investigation. Global CBF, as well as jugular venous oxygen saturation, is not found to be impaired by indicator administration.

As the jugular bulb is relatively easy to access and as 99% of the jugular bulb blood is drained from the intracerebral vasculature, blood sampling is considered to reflect accurately the oxygen content and cerebral metabolic state of the total brain tissue [5,9]. Consequently, the cerebrovenous blood may be estimated to be an appropriate temperature medium to evaluate global cerebral temperature. Nevertheless, there are conflicting reports showing jugular vein temperature not to be a good measurement of brain temperature [10-12]. However, some of these results have been accomplished in patients with neuropathological conditions, which as such may be the reason for this discrepancy. Patients with significant intracranial pathology were excluded from this investigation and measurements were performed under steady global and cerebral haemodynamic conditions. We therefore see no reason to refrain from the use of cerebrovenous blood temperature as an estimate for global brain temperature in order to assess the effects of repeated TCID measurements. In addition the recovery of the absolute jugular bulb blood temperature in the time interval before the next TCID measurement therefore may be understood as an indirect indicator that the brain tissue temperature is not decreased by this procedure. Furthermore, the TCID method is based on the mass conservation principle, which relies on the fact that the indicator amount administered before the organ is equal to the amount detected after the organ within a definite time interval. However, although global CBF can be measured by this technique, conclusions on the regional CBF are very limited, because especially in the injured brain regional CBF as well as temperature distribution may differ considerably.

The measurements for CMR and CBF presently cannot be performed continuously, as a single TCID measurement requires a time period of 4 min. This is basically not in accordance with the principles of online monitoring. Nevertheless, as repeated TCID measurements have been shown to be reliable [3], this technique provides the possibility of intermittent and frequent assessment of global CBF at the bedside.

The mean cerebrovenous blood temperature difference between pre- and post-TCID measurements, although significant, is very small indeed. Theoretically, cerebral metabolism is decreased by 5-10% per degree Celsius of temperature reduction [13,14]. So far, to our knowledge a linear relationship between blood temperature and metabolism has been demonstrated from normothermia to mild hypothermia, to the restriction that only 1°C intervals have been investigated. The relationship between cerebral metabolism and blood temperature changes of less than 1°C remain unknown. However, it becomes obvious that the resolution of the used tools and measurement procedures to assess physiological variables is not high enough to detect minor changes of this degree. Therefore, short and transient alterations of the variables CBF and CMR may be possible and consequently may be overseen during the non-MP. Next to these limitations the TCID technique clearly has the disadvantages of occasional distorted curves, the limited assessment of regional CBF and off-line monitoring. However, although this method is an intriguing tool to measure cerebral circulation and metabolism at bedside, additional investigations are necessary to increase the experience with this technique especially in patients with intracranial pathology.

In conclusion, this clinical investigation showed that during total i.v. anaesthesia and steady-state conditions, absolute jugular bulb blood temperature is higher compared to absolute aortic blood temperature. Furthermore, performing CBF measurements using the TCID technique with bolus administration of ice-cold indicator results in a reduction of blood temperature in jugular bulb as well as in the aorta after the 4 min registration period. A 20 min time interval following the TCID registration showed an almost complete recovery of the induced blood temperature decrease.

Finally, this study showed that repeated measurements with the TCID technique do not affect global CBF and metabolism. The TCID technique has proven to be a feasible tool for monitoring cerebral circulation. The implementation of this monitoring device at the bedside not only offers the opportunity to intermittently measure CBF, but also permits continuous oximetry and sampling of blood for cerebral metabolic monitoring as well as continuous detection of jugular bulb blood temperature. The integrated interpretation of these parameters may prove beneficial in the treatment of cardiosurgical and neurotraumatized patients.


Preliminary results of this investigation were presented in part at the 15th Annual Meeting of the European Association of Cardiothoracic Anaesthetists, June 2000; abstract A6 in Br J Anaesth 2000; 84 (Suppl. 1): 3.


1. Wietasch G, Mielck F, Scholz M, v. Spiegel T, Stephan H, Hoeft A. Bedside assessment of cerebral blood flow by double-indicator dilution technique. Anesthesiology 2000; 92: 367-375.
2. Keller E, Wietasch G, Ringleb P, et al. Bedside monitoring of cerebral blood flow in patients with acute hemispheric stroke. Crit Care Med 2000; 28: 511-516.
3. Mielck F, Wietasch G, Weyland A, et al. Reliability of cerebral blood flow measurements by transcerebral double-indicator dilution technique. Eur J Anaesthesiol 2001; 18: 653-661.
4. Herscovitch P, Raichle ME. What is the correct value for the brain-blood partition coefficient for water? J Cereb Blood Flow Metab 1985; 5: 65-69.
5. Crowder CM, Tempelhoff R, Theard A, Cheng MA, Todorov A, Dacey R. Jugular bulb temperature: comparison with brain surface and core temperature in neurosurgical patients during mild hypothermia. J Neurosurg 1996; 85: 98-103.
6. Clausen T, Rieger A, Roth S, et al. Cerebrovenous blood temperature-influence of cerebral perfusion pressure changes and hyperventilation: evaluation in a porcine study and in man. J Neurosurg Anaesth 2000; 12: 2-9.
7. Lanier WL, Iaizzo PA, Murray MJ. The effects of convective cooling and rewarming on systemic and central nervous system physiology in isoflurane-anesthetized dogs. Resuscitation 1992; 23: 121-136.
8. Laptook AR, Corbett RJ, Sterett R, Burns DK, Tollefsbol G, Garcia D. Modest hypothermia provides partial neuroprotection for ischemic neonatal brain. Pediatr Res 1994; 35: 436-442.
9. Robertson CS, Narayan RK, Gokaslan ZL, et al. Cerebral arteriovenous oxygen difference as an estimate of cerebral blood flow in comatose patients. J Neurosurg 1989; 70: 222-230.
10. Stone JG, Young WL, Smith CR, et al. Do standard monitoring sites reflect true brain temperature when profound hypothermia is rapidly induced and reversed? Anesthesiology 1995; 82: 344-351.
11. Rumana CS, Gopinath SK, Uzura M, Valadka AB, Robertson CS. Brain temperature exceeds systemic temperature in head-injured patients. Crit Care Med 1998; 26: 562-567.
12. Sandstrom K, Nilsson K, Andreasson S, Larsson LE. Jugular bulb temperature compared with non-invasive temperatures and cerebral arteriovenous oxygen saturation differences during open heart surgery. Paediatr Anaesth 1999; 9: 123-128.
13. Schell RM, Kern FH, Greeley WJ, et al. Cerebral blood flow and metabolism during cardiopulmonary bypass. Anesth Analg 1993; 76: 849-865.
14. Marion DW, Penrod LE, Kelsey SF, et al. Treatment of traumatic brain injury with moderate hypothermia. New Engl J Med 1997; 336: 540-546.

HAEMODYNAMICS, regional blood flow, cerebral blood flow; INVESTIGATIVE TECHNIQUES, indicator dilution techniques, dye-dilution technique, thermo-dilution; TEMPERATURE, jugular bulb

© 2004 European Academy of Anaesthesiology