Secondary Logo

Journal Logo

Original Article

The cardiovascular effects of inspired oxygen fraction in anaesthetized patients

Anderson, K. J.*; Harten, J. M.*; Booth, M. G.*; Kinsella, J.*

Author Information
European Journal of Anaesthesiology: June 2005 - Volume 22 - Issue 6 - p 420-425
doi: 10.1017/S0265021505000712

Abstract

It is standard practice to administer an inspired oxygen (O2) fraction (FiO2) of 1.0 during induction of and emergence from anaesthesia. This practice is perceived to benefit patients as it increases the O2 reservoir during manipulation of the airway. In contrast, the possible detrimental cardiovascular effects of O2 therapy have been highlighted [1]. Animal work has demonstrated that O2 reduces cardiac output through a decreased heart rate (HR) and increases systemic vascular resistance [2]. These findings have been confirmed in healthy volunteers using invasive [3] and non-invasive [4] cardiac output monitoring techniques. In patients with congestive heart failure, these effects were more pronounced [5]. The effects of O2 on haemodynamic changes in subjects under general anaesthesia have not been studied.

Although the systemic haemodynamic effects of sevoflurane and propofol anaesthesia are similar, the effects of O2 on pulmonary circulation are different under these two types of anaesthesia [6-10]. This raises the possibility that the effects of O2 on systemic haemodynamics under sevoflurane and propofol anaesthesia may also be different.

Ethically it is difficult to study the effect of drugs on cardiac output in healthy individuals using invasive monitoring. Transthoracic electrical bio-impedance estimates cardiac output by measuring non-invasively changes in resistance to a small current applied across the thorax over time during the cardiac cycle. The non-invasive nature of transthoracic electrical bio-impedance may be ethically more acceptable [11].

The aim of this study was to measure the systemic haemodynamic effects of changing the FiO2 in patients, awake and during anaesthesia with sevoflurane or propofol.

Methods

Following approval by the North Glasgow Hospital University Trust Research and Ethics Committee, and after informed consent, we studied 15 patients in the propofol study and 15 patients in the separate sevoflurane study. Exclusion criteria included age <18 or >70 yr, significant cardio-respiratory disease (American Society of Anesthesiologists (ASA) III-V), current treatment with cardio-active drugs and pregnancy. Patients did not receive any sedative premedication, and lay supine for 5 min prior to the study. A 20-G catheter was inserted into an upper limb vein, and all monitoring applied. No intravenous (i.v.) fluids or vaso-active drugs were given during the study period.

The patients were connected to the transthoracic bio-impedance monitor (BioZ System 1.52; Cardiodynamics International Corporation, San Diego, CA, USA) in accordance with the manufacturer's instructions. The patients' weight and height were entered into the machine allowing measurements to be indexed for body surface area. The monitor displays an electrocardiogram, measures stroke volume and non-invasive blood pressure (BP) and calculates cardiac output, cardiac index, stroke index and systemic vascular resistance index. The measurement of cardiac index is continuous, and was averaged over 20 cardiac cycles. BP was measured at each measurement point. In addition, arterial O2 saturation (SPO2) and the FiO2 and expired fractional concentration of O2 (FEO2) were measured with a combined spectrophotometric gas analyser and pulse oximeter (Capnomac Ultima; Datex-Ohmeda Ltd, Hatfield, Herts, UK). Baseline cardiovascular measurements were made with patients breathing medical air via a circle anaesthetic system and tight fitting facemask. The same measurements were made 5 min after equilibration at each stage of the study (Fig. 1) to allow the cardiovascular effects to have occurred [3]. Room temperature was kept constant between 22°C and 24°C.

Figure 1.
Figure 1.:
Time line showing protocol design. The solid line at the top shows time from the start to the end of the study. The FiO2 at each stage is shown on the second solid line. For the third line, the dotted line represents the phase of equilibration with the new FiO2 and the continuous line the time when the FiO2 had equilibrated (when FEO2 had reached a stable value). Measurement points are indicated by arrows; these were taken 5 min after equilibration of FiO2. The end of the study was completed before surgery commenced. I: induction of anaesthesia; L: insertion of laryngeal mask.

In the sevoflurane group, anaesthesia was induced by inhalational induction with 8% sevoflurane in 92% O2. When the patient was adequately anaesthetized a laryngeal mask was placed, and its correct positioning confirmed. The inspired fraction of sevoflurane was reduced to 2%. When the end-tidal fraction of sevoflurane had reduced to 2%, the inspired sevoflurane was adjusted to maintain the end-tidal concentration at 2% throughout the study. At this stage, sevoflurane was assumed to have equilibrated.

In the propofol group, anaesthesia was administered via a Graseby 3500 (Graseby Medical Limited, Colonial Way, Watford, Hertfordshire, UK) target-controlled infusion pump with inbuilt Diprifusor® module. The initial target plasma concentration was set between 4 and 8 μg mL−1. When the patient was adequately anaesthetized a laryngeal mask was placed, and its correct positioning confirmed. At this stage the target plasma concentration of propofol was reduced to 4.5 μg mL−1, and when the effect-site concentration had reached this value propofol was assumed to have equilibrated.

Measurements were taken 5 min following equilibration for each FiO2, which was defined as the time point when FEO2 was stable (Fig. 1). The haemodynamic effects of carbon dioxide (CO2) [12] were controlled for by manually ventilating the patient to the baseline awake end-tidal CO2 concentration. Cardiovascular measurements were made immediately on stopping manual ventilation, to avoid the confounding effects of changing intrathoracic pressure. All measurements were made before the start of surgery.

Sample size was calculated based on previously published data [4]. To detect a clinically significant difference in cardiac output of 0.36 L min−1 (standard deviation (SD), 0.55), the sample size required was 15 assuming a power of 0.80 and α of 0.05. Data was continuous, matched and normally distributed and was analysed by paired t-tests using SPSS for Windows (SPSS Inc., Chicago, IL, USA).

Results

All 30 patients completed the study and were included in the analysis. No adverse events occurred during the study period. The patient characteristic details of the patients and their baseline cardiovascular measurements breathing medical air are given in Table 1.

Table 1
Table 1:
Baseline patient data. Values are expressed as mean (SD) unless otherwise stated.

Summary data showing the effect of FiO2 on cardiovascular measurements before and after anaesthesia are demonstrated for sevoflurane patients in Table 2, and for propofol patients in Table 3.

Table 2
Table 2:
Effect of changing FiO2 on measured and calculated cardiovascular parameters in awake and sevoflurane anaesthetized patients. Absolute results are expressed as mean (SD).
Table 3
Table 3:
Effect of changing FiO2 on measured and calculated cardiovascular parameters in propofol anaesthetized patients. Absolute results are expressed as mean (SD).

Both groups had the FiO2 increased from 0.3 to 1.0 during preoxygenation; these results were combined to give the effect of preoxygenation in all 30 patients (Table 4).

Table 4
Table 4:
The effect of preoxygenation in both the sevoflurane and propofol groups combined (n = 30). Results are expressed as mean (SD). Paired t-test comparisons for patients breathing at FiO2 0.21 and 1.0 are shown (P-value).

The effect of anaesthesia with 2% sevoflurane can be extracted from Table 2 by controlling for FiO2. For patients who were breathing an FiO2 of 0.3, awake vs. anaesthetized comparisons showed that there was no significant change in mean HR (69.1 ± 8.4 vs. 69.1 ± 7.5 beats per minute (bpm); P = 0.97), mean stroke index (49.8 ± 9.8 vs. 46.7 ± 8.3 mL m−2; P = 0.05), or mean cardiac index (3.40 ± 0.65 vs. 3.25 ± 0.56 L min−1 m−2; P = 0.11). Mean arterial pressure was significantly reduced (88.4 ± 9.0 vs. 71.4 ± 8.7 mmHg; P < 0.001) as was systemic vascular resistance index (2022 ± 315 vs. 1735 ± 388 dyn s−1 cm−5 m−2; P = 0.008) and SPO2 (98.5 ± 0.9 vs. 97.1 ± 1.4%; P = 0.003).

For patients anaesthetized with sevoflurane, who were breathing an FiO2 of 1.0, awake vs. anaesthetized comparisons showed that there was no significant change in mean HR (66.3 ± 7.8 vs. 65.1 ± 7.8 bpm; P = 0.32), mean stroke index (48.8 ± 9.5 vs. 47.4 ± 9.4 mL m−2; P = 0.33), or mean cardiac index (3.21 ± 0.61 vs. 3.06 ± 0.57 L min−1 m−2; P = 0.08). Mean arterial pressure was significantly reduced (90.1 ± 9.1 vs. 74.8 ± 8.7 mmHg; P < 0.001) as was the mean systemic vascular resistance index (2195 ± 423 vs. 1883 ± 329 dyn s cm−5 m−2; P = 0.008) and SPO2 (99.2 ± 0.4 vs. 98.7 ± 0.7%; P = 0.03).

The effect of anaesthesia with propofol at a steady state effect-site concentration of 4.5 μg mL−1 can be extracted from Table 3 by controlling for FiO2. For patients who were breathing an FiO2 of 0.3, comparing awake vs. anaesthetized comparisons showed that there was an increase in mean HR (61.6 ± 9.4 vs. 72.7 ± 11.6 bpm; P < 0.001) a reduction in mean stroke index (48.9 ± 9.5 vs. 40.5 ± 8.4 mL m−2; P < 0.001) and overall no change in mean cardiac index (2.94 ± 0.40 vs. 2.89 ± 0.42 L min−1 m−2; P = 0.42). Mean arterial pressure was significantly reduced (82.7 ± 10.0 vs. 66.5 ± 6.8 mmHg; P < 0.001) as was systemic vascular resistance index (2153 ± 333 vs. 1771 ± 259 dyn s−1 cm−5 m−2; P < 0.001) and SPO2 (98.9 ± 0.7 vs. 97.2 ± 1.5%; P = 0.001).

For patients who were breathing an FiO2 of 1.0, awake vs. anaesthetized comparisons showed that there was an increase in mean HR (59.3 ± 10.9 vs. 67.5 ± 11.8 bpm; P = 0.007) and a reduction in mean stroke index (48.2 ± 7.9 vs. 42.2 ± 8.6 mL m−2; P = 0.001). Overall there was no significant change in mean cardiac index (2.86 ± 0.51 vs. 2.76 ± 0.46 L min−1 m−2; P = 0.34). Mean arterial pressure was significantly reduced (84.3 ± 10.7 vs. 72.1 ± 8.7 mmHg; P < 0.001) as was the mean systemic vascular resistance index (2245 ± 349 vs. 2015 ± 369 dyn s−1 cm−5 m−2; P = 0.03) and SPO2 (99.5 ± 0.5 vs. 98.8 ± 0.7; P = 0.003).

Discussion

When measuring the haemodynamic effects of changing the FiO2 from 0.3 to 1.0 we confirmed previous findings in awake subjects [4,5] of a decrease in HR and cardiac index, an increase in systemic vascular resistance and an unchanged mean arterial pressure.

A number of conclusions can be drawn from our data in anaesthetized patients. Firstly, the change of FiO2 from 1.0 to 0.3 reversed the changes observed in preoxygenating the awake subjects. Changing FiO2 therefore appears to cause cardiovascular effects both when awake and anaesthetized. In anaesthetized patients, these effects are similar for those receiving propofol and sevoflurane. Secondly, reducing FiO2 in the anaesthetized patients, as routinely performed following induction of anaesthesia, also decreased mean arterial pressure. Thirdly, varying FiO2 did not change stroke index in the awake patients, but caused a small reduction when patients were anaesthetized with propofol.

For this study, we used thoracic electrical bio-impedance to measure cardiovascular parameters. Transthoracic electrical bio-impedance estimates cardiac output by measuring non-invasively changes in resistance to a small current applied across the thorax over time during the cardiac cycle. We do not claim transthoracic electrical bio-impedance to be a gold-standard measurement tool for cardiac output. Indeed, early transthoracic bio-impedance monitors correlated poorly with thermodilution and dye dilution cardiac output measurement [13]. However, more recent models have demonstrated a better agreement [11]. Using this system, we observed a similar magnitude of cardiovascular changes in response to hyperoxia in awake volunteers to those previously demonstrated using a gold-standard technique [3,4]. Transthoracic bio-impedance therefore appears to be a practical tool to study the cardiovascular effects associated with a change of FiO2 in healthy patients (with no clinical or ethical reason to use invasive techniques) under general anaesthesia.

The beneficial outcomes following intraoperative hyperoxia have been established [14-16], though some have been recently challenged [17]. To our knowledge, no previous work has examined the effects of O2 on cardiovascular parameters during general anaesthesia. The observed changes, although statistically significant, were of limited clinical significance. In this group of healthy anaesthetized patients, this would present little risk of detrimental effects. This could be different for patients with co-morbidity. In awake patients with heart failure [5], hyperoxia induces larger changes of cardiac index and systemic vascular resistance than in healthy volunteers [3,4]. We hypothesise that the effects of hyperoxia in anaesthetized patients with cardiac co-morbidity may therefore be larger. These patients could be more susceptible to any negative inotropic effects. Before subjecting more sick patients to a similar study, it would seem wise to characterise the effects in more depth in healthy individuals.

When designing the study we considered randomizing the allocation of the low and high O2 treatment to reduce any potential bias. However, owing to clinical time constraints, we decided, in the first instance, to study the standard anaesthetic practice of preoxygenation followed by administration of a lower FiO2 after securing the airway. The observed effect of FiO2 on the cardiovascular system could be firstly a genuine effect of O2 or secondly merely a time effect. The effects of increasing FiO2 when awake were reversed when reducing FiO2 when anaesthetized and would not support a time effect. However, we accept that a study where FiO2 was given in a randomized manner, rather than the standard clinical order, would be necessary to confirm our findings. Furthermore, it would be advisable to repeat the measurements in patients with controlled ventilation, this would control better the arterial CO2 tension and avoid the possible cardiovascular changes caused by changing between spontaneous and manual ventilation modes.

Considering its established clinical use, there are sparse published data, with low patient numbers, of the haemodynamic effects of sevoflurane or propofol on healthy human being individuals [6-8]. This may reflect the ethical dilemmas of using invasive methodology to measure cardiac output in patients where it is not clinically justified. The cardiovascular effects we observed under sevoflurane and propofol anaesthesia were a stable cardiac index, a reduced systemic vascular resistance, mean arterial pressure and O2 saturation. These were similar to previously published work (Table 5) [6-8]. The quantitative differences between our results and those from previous data could reflect differences in the anaesthetic technique used in the different studies, sample size, patient groups or the measurement methodology.

Table 5
Table 5:
The percentage changes of cardiovascular parameters attributable to anaesthesia with propofol or sevoflurane. Those measured by thoracic electrical bio-impedance are from the present study, those using thermodilution via pulmonary artery catheter are from previous studies. The percentage change indicates the differences between the awake and anaesthetized state. For the present study, where a range is given this represents the figures measured whilst breathing an FiO2 of 0.3 and 1.0. The arrows indicate an increase (↑) or decrease (↓).

Comparing the effects of propofol and sevoflurane should be done with caution because patients were not randomly allocated to receive either anaesthetic agent. It is interesting however, that although cardiac index was unchanged by both agents, HR was increased and stroke index decreased with propofol, but unchanged with sevoflurane.

This observational study has demonstrated the cardiovascular effects of O2 during anaesthesia with propofol or sevoflurane in healthy patients. The haemodynamic effects of O2 in anaesthetized patients with cardiac co-morbidity remain to be established.

References

1. Thomson AJ, Webb DJ, Maxwell SR, Grant IS. Oxygen therapy in acute medical care. BMJ 2002; 324: 1406-1407.
2. Lodato RF. Decreased oxygen consumption and cardiac output during normobaric hyperoxia in conscious dogs. J Appl Physiol 1989; 67: 1551-1559.
3. Daly WJ, Bondurant S. The effects of oxygen breathing on heart rate, blood pressure, and cardiac index of normal men-resting, with reactive hyperaemia, and after atropine. J Clin Invest 1962; 41: 126-132.
4. Harten JM, Anderson KJ, Angerson WJ, Booth MG, Kinsella J. Normobaric hyperoxia reduces cardiac index in healthy awake volunteers. Anaesthesia 2003; 58: 885-888.
5. Haque WA, Boehmer J, Clemson BS, Leuenberger UA, Silber DH, Sinoway LI. Hemodynamic effects of supplemental oxygen administration in congestive heart. J Am Coll Cardiol 1996; 27: 353-357.
6. Claeys MA, Gepts E, Camu F. Haemodynamic changes during anaesthesia induced and maintained with propofol. Br J Anaesth 1983; 60: 3-9.
7. Malan TP, DiNardo JA, Isner RJ, et al. Cardiovascular effects of sevoflurane compared with those of isoflurane in volunteers. Anesthesiology 1995; 83: 918-928.
8. Ebert TJ, Muzi M, Lopatka CW. Neurocirculatory responses to sevoflurane in humans. A comparison to desflurane. Anesthesiology 1995; 83: 88-95.
9. Van Keer L, Van Aken H, Vandermeersch E. Propofol does not inhibit hypoxic pulmonary vasoconstriction in humans. J Clin Anesth 1989; 1: 284-288.
10. Beck DH, Doepfmer UR, Sinemus C, Bloch A, Schenk MR, Kox WJ. Effects of sevoflurane and propofol on pulmonary shunt fraction during one-lung ventilation for thoracic surgery. Br J Anaesth 2001; 86: 38-43.
11. Sageman WS, Riffenburgh RH, Spiess BD. Equivalence of bioimpedance and thermodilution in measuring cardiac index after cardiac surgery. J Cardiothorac Vasc Anesth 2002; 16: 8-14.
12. Akca O, Doufas AG, Morioka N, Iscoe S, Fisher J, Sessler DI. Hypercapnia improves tissue oxygenation. Anesthesiology 2002; 97: 801-806.
13. Raaijmakers E, Faes TJ, Scholten RJ, Goovaerts HG, Heethaar RM. A meta-analysis of three decades of validating thoracic impedance cardiography. Crit Care Med 1999; 27: 1203-1213.
14. Goll V, Akca O, Greif R, Sessler DI. Ondansetron is no more effective than supplemental intraoperative oxygen for prevention of postoperative nausea and vomiting. Anesth Analg 2001; 92: 112-117.
15. Greif R, Akca O, Horn EP, Kurz A, Sessler DI. Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection. New Engl J Med 2000; 342: 161-167.
16. Greif R, Laciny S, Rapf B, Hickle RS, Sessler DI. Supplemental oxygen reduces the incidence of postoperative nausea and vomiting. Anesthesiology 1999; 91: 1246-1252.
17. Pryor KO, Fahey III TJ, Lien CA, Goldstein PA. Surgical site infection and the routine use of perioperative hyperoxia in a general surgical population: a randomized controlled trial. JAMA 2004; 291: 79-87.
Keywords:

OXYGEN; HAEMODYNAMIC PHENOMENA; cardiac output; heart rate; vascular resistance; IMPEDANCE CARDIOGRAPHY; PROPOFOL; SEVOFLURANE

© 2005 European Society of Anaesthesiology