Human cardiovascular dose?response to supplemental oxygen

Acta Physiol 2007, 191, 15–24

Human cardiovascular dose–response to supplemental oxygen Z. Bak,1,2 F. Sjo¨berg,1,2 A. Rousseau,1 I. Steinvall2 and B. Janerot-Sjoberg3 1 Department of Anesthesia and Intensive Care, University Hospital, Linko¨ping, Sweden 2 Departments of Hand and Plastic Surgery and Burn Intensive Care, University Hospital, Linko¨ping, Sweden 3 Department of Clinical Physiology, Heart Center, Linko¨ping University Hospital, Linko¨ping, Sweden

Received 12 November 2006, revision requested 15 January 2007, final revision received 4 March 2007, accepted 13 March 2007 Correspondence: B. JanerotSjoberg, MD, PhD, Department of Clinical Physiology, Linko¨ping University Hospital, SE-581 85 Linko¨ping, Sweden. E-mail: [email protected]

Abstract Aim: The aim of the study was to examine the central and peripheral cardiovascular adaptation and its coupling during increasing levels of hyperoxaemia. We hypothesized a dose-related effect of hyperoxaemia on left ventricular performance and the vascular properties of the arterial tree. Methods: Oscillometrically calibrated arterial subclavian pulse trace data were combined with echocardiographic recordings to obtain non-invasive estimates of left ventricular volumes, aortic root pressure and flow data. For complementary vascular parameters and control purposes whole-body impedance cardiography was applied. In nine (seven males) supine, resting healthy volunteers, aged 23–48 years, data was collected after 15 min of air breathing and at increasing transcutaneous oxygen tensions (20, 40 and 60 kPa), accomplished by a two group, random order and blinded hyperoxemic protocol. Results: Left ventricular stroke volume [86  13 to 75  9 mL (mean  SD)] and end-diastolic area (19.3  4.4 to 16.8  4.3 cm2) declined (P < 0.05), and showed a linear, negative dose–response relationship to increasing arterial oxygen levels in a regression model. Peripheral resistance and characteristic impedance increased in a similar manner. Heart rate, left ventricular fractional area change, end-systolic area, mean arterial pressure, arterial compliance or carbon dioxide levels did not change. Conclusion: There is a linear dose–response relationship between arterial oxygen and cardiovascular parameters when the systemic oxygen tension increases above normal. A direct effect of supplemental oxygen on the vessels may therefore not be excluded. Proximal aortic and peripheral resistance increases from hyperoxaemia, but a decrease of venous return implies extra cardiac blood-pooling and compensatory relaxation of the capacitance vessels. Keywords arterial compliance, end-diastolic area, hyperoxaemia, hyperoxia, normocapnia, stroke volume, vascular resistance.

Oxygen supplementation is a first line treatment for critically ill patients, irrespective of cardiovascular, pulmonary, traumatic, infective or other causes. Apart from the lung (Dantzker et al. 1974) and persistent ductus arteriosus (Hampl et al. 2002) however, the blood vessels react to hyperoxia by arterial vasocon-

striction (Whitehorn et al. 1946, Daly & Bondurant 1962, Eggers et al. 1962, Whalen et al. 1965, Milone et al. 1999, Waring et al. 2003, Rousseau et al. 2005). This is true also for the coronary circulation where a normally functioning endothelium effectively regulates the arterial resistance and thereby the flow, in order to

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

15

Æ Z Bak et al.

Acta Physiol 2007, 191, 15–24

maintain adequate myocardial oxygenation irrespective of activity (Baron et al. 1990, Kemming et al. 2004, McNulty et al. 2005). The vasoconstrictive effects of hyperoxaemia on the vessels in the brain (Watson et al. 2000, Singhal et al. 2005), skeletal muscle (Crawford et al. 1997), retina (Kiss et al. 2002), and skin (Bongard et al. 1992) have been well-documented, and Pittman have shown that there is an almost constant transport of oxygen to the tissue as a result of the hyperoxemic adjustment of the microvascular architecture during increased levels of oxygen (Pittman 1995). The level of central arterial blood pressure results from peripheral resistance and cardiac output (CO). The hyperoxemic-induced peripheral vasoconstriction will elicit compensatory mechanisms resulting in decreasing CO but there is conflict in published data. Some investigators (Benedict & Higgins 1911, Milone et al. 1999, Shibata et al. 2005) have reported a direct effect on heart rate (HR), thereby supporting a hypothesis of altered efferent output from the parasympathetic nervous system. Others (Whitehorn et al. 1946, Eggers et al. 1962, Waring et al. 2003) have described reduced stroke volumes (SV) and CO as the immediate result of increased systemic resistance. The time factor (the oxygen exposition time) is a probable confounding factor when such changes are being examined. From the microvascular perspective, the greatest changes are seen early after a rapid and large change in the oxygen tensions (Rousseau et al. 2005). Reducing any of these factors diminishes the corresponding vascular changes, and a longer adjustment period theoretically reduces the magnitude of any decrease in HR, or SV, or both. The microvascular effects of hyperoxaemia have been claimed to initiate endothelial involvement from increased local vasoconstriction or the decreased actions of dilators (Messina et al. 1994), or effects of mechanisms involving the red blood cell (Ellsworth 2000, Gonzalez-Alonso et al. 2002, McMahon et al. 2002), or both. Breathing 100% oxygen also alters cardiac relaxation and increases filling pressures in patients with coronary artery disease (Haque et al. 1996, Mak et al. 2001). Reduced systolic function has also been reported in healthy volunteers (Frobert et al. 2004).

Although hyperoxia is universally used for support in critically ill patients and has therefore been investigated in numerous studies, the combined dose–response effects on central and peripheral circulatory variables and their coupling have not to our knowledge been evaluated in a controlled way. We hypothesized a doserelated effect of hyperoxaemia on left ventricular performance and the vascular properties of the arterial tree. Simultaneously and non-invasively, we examined the central and peripheral cardiovascular adaptation during increasing levels of hyperoxaemia.

Hyperoxaemic cardiovascular dose–response

Materials and methods Volunteers Nine healthy volunteers, seven men and two women aged 23–48 years (mean  SD 31  8), height 179  9 cm, weight 76  15 kg, and body surface area 1.9  0.2 m2 participated. They were recruited from medical students and co-workers, and none had evidence of hypertension (systolic blood pressure 126  18 mmHg, diastolic 64  10 mmHg) or heart disease in their history, on physical examination, electrocardiography, or transthoracic echocardiography. All gave informed consent to the study, which was approved by the local medical ethics committee.

Study protocol We used a system for oxygen delivery through a nonrebreathing system (Prieur et al. 1998), that had earlier been evaluated by us (Rousseau et al. 2005). The volunteers were in supine position for 15 min before a mask was applied and the study began. They were divided into two groups (protocol I or II) to blind the investigators and participants to the time when increasing levels of oxygen-delivery was started by a third party (Figure 1). This blinding was maintained throughout the study until all the results were available. An oxygen sensor in the circuit indicated the percentage of inspired oxygen levels. Arterial oxygen (PtcO2) and carbon dioxide (PtcCO2) pressure were assessed noninvasively and continuously by transcutaneous meas-

PtcO2 20 kPa PtcO2 40 kPa PtcO2 60 kPa Baseline/air

Baseline/air

I 15 min rest

Mask on

Baseline/air

Baseline/air PtcO2 20 kPa PtcO2 40 kPa PtcO2 60 kPa

II 15 min

5 min

Figure 1 Study protocol for group I (n ¼ 4) and group II (n ¼ 5). Time for measurements are marked in grey. At baseline all subjects were breathing room air. PtcO2: transcutaneously measured oxygen pressure.

16

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

Æ Hyperoxaemic cardiovascular dose–response

Acta Physiol 2007, 191, 15–24

Z Bak et al.

urements (TINA; Radiometer, Copenhagen, Denmark). The sensor was placed on the ventral side of the forearm, avoiding superficial vessels and body hair. Instruments were calibrated between patients according to the manufacturer’s instructions. In parallel with Doppler echocardiography, external subclavian arterial pulse tracing and whole body impedance cardiography were recorded and the time of Doppler measurement was marked in the impedance recording for synchronized recordings and calculations.

(echodisp 3.1; Vingmed Sound). SVDoppler was calculated as:

Doppler echocardiography, and non-invasive estimates of aortic root pressures by external subclavian arterial pulse tracing Central and vascular arterial properties were estimated from the simultaneous calculations of aortic root pressure and flow with a three-element windkessel model; an electric analogue representation of the systemic circulation with three variables (Westerhof et al. 1971). The method has been further developed and evaluated against invasive pressure measurements by Aakhus et al. (1993a,b). Aortic velocity time integral (VTI), left ventricular SV (SVDoppler), CODoppler, fractional left ventricular area change (FAC%), aortic characteristic impedance [ZDoppler; proximal arterial (i.e. Aortic) resistance to LV ejection of blood], the total volume compliance of the arterial tree (ACDoppler), peripheral arteriolar resistance (PRDoppler), and total peripheral resistance (TPRDoppler) were calculated. The sum of PRDoppler and ZDoppler is the TPRDoppler. Short description of method. An ultrasound scanner (CFM 750; Vingmed Sound, Horten, Norway) was used for transthoracic parasternal and apical echocardiographic imaging (frequency 3.25 MHz) and Doppler velocity recording (frequency 2.5 MHz). Simultaneously, an external pulse transducer for pulse tracing of the right subclavian artery (Irex 120–0132; Irex Medical Systems, Ramsey, NJ, USA) was connected through a preamplifier to the scanner and AD-converted at a rate of 200/s. Subclavian pressure (Colan et al. 1985) was calibrated with brachial arterial blood pressures obtained semiautomatically and simultaneously using an oscillometric manometer (Criticon; Dinamap, vital signs monitor 1846, Tampa, FL, USA) and an adult-size cuff on the patient’s upper right arm. The mean average arterial pressure (MAP) was used in subsequent calculations. The ultrasound scanner was interfaced to a computer (Macintosh II series; Apple Computer, Cupertino, CA, USA) for transfer of at least five cardiac cycles with stable and simultaneously recorded digital ultrasound data, pulse pressure data and a single vector electrocardiogram (ECG lead III) for off-line analysis using commercially available software

SVDoppler ¼ VTI  aortic annulus cross-sectional area The latter assumed circular and calculated from a measured diameter (Skjaerpe et al. 1985, Sjoberg & Wranne 1990).

Whole-body impedance cardiography To measure parallel and complementary cardiac and vascular characteristics i.e. SVicg, SVRicg, ACicg, pulse transmission time (PTTicg: the time delay between the two arterial plethysmograms obtained from the heart synchronous impedance wave), and pulse wave velocity (PWVicg: average elastic state of the vessel pathway), we used commercially available wholebody impedance cardiography (ICG) equipment (CircMonTM B202; JR Medical, Tallinn, Estonia). The method has previously been compared with invasive measurements in intensive care units (Koobi et al. 1997a,b, 2001). Short description of method. For central circulatory variables a pair of electrically connected current electrodes (ECG electrodes Blue Sensor type R-00-S; Med¨ lstykke, Denmark) were placed just ico test A/S, O proximal to the wrists and the ankles, and another pair of similar voltage electrodes were placed 5 cm proximally. SVicg was calculated by the equation: SVicg ¼ k  H 2  DZ  C=½Zc Z0 D where H is height (cm), delta Z the amplitude of heart synchronous impedance variation (X), ZC is the calibration factor (0.1X), and Z0 the baseline impedance of the body (X). C is the duration of the cardiac cycle, and D the duration between peak heart synchronous impedance variation and the onset of the next cycle. The coefficient k (X xcm) is derived from the resistivity of the blood, the relationship between the distance of the voltage electrodes and body height, and includes a correction for measured body mass index and a default haematocrit of 0.45. Beat-to-beat- data were saved as files on a personal computer, and a period of 15 s (three to four complete respiratory cycles) was used for calculation of mean COicg. For pulse wave propagation measurements (PWVicg, PTTicg) another two pairs of voltage electrodes were placed in the sternal angle and knee joint regions. SVRicg was conventionally calculated as MAPoscillometric reduced by central vein pressure (CVP, default 3 mmHg) divided by COicg [SVR ¼79.9 · (MAP ) CVP)/CO dyn s cm)5]. ACicg was calculated as the pulse pressure divided by SVicg.

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

17

Hyperoxaemic cardiovascular dose–response

Æ Z Bak et al.

Acta Physiol 2007, 191, 15–24

Statistical analysis

Dose–response studies

Commercially available computer software was used for statistical analysis (minitab; release 13, State College, PA, USA). A P-value 3.40). The comparison between two ‘non-gold standard’ measurements of cardiac and vascular variables, was accomplished as described by Bland & Altman (1986) and with regression analysis.

All variables that changed significantly from hyperoxaemia showed a dose–response relation to increasing oxygen-pressures as analysed by the linear multiple regression model (Table 3 with data for regression equations). For example, the calculated regression equation predicts a mean loss of 4 mL in SVDoppler when PtcO2 is increased by 20 kPa above normal. In all variables a straight-line model described the relation better than a third degree polynom. Similar results were obtained if PtcO2 were grouped either in categories or continuous values. Using inspired air O2 fraction (FiO2) instead in the analysis did not change the results.

Comparative studies There were no significant differences between the data generated by the Doppler echocardiographic compared with those obtained by the whole body impedance method, and correlation coefficients for the regression lines were 0.64 for SV, 0.59 for TPR/SVR, 0.54 for AC and 0.95 for MAPOscillometric/MAPSubclavian; regression lines did not differ significantly from lines of identity. With Altman and Bland analysis there was no bias: Mean SVicg was 8 mL higher than SVDoppler (SDdiff 12.6), and mean TPRDoppler was 220 dyne s cm)5 higher than SVRicg (SDdiff 203).

Discussion Results PtcO2 and PtcCO2 matched the aimed values well, as shown in Table 1. Hyperoxaemia resulted in the central and peripheral haemodynamic changes described in Table 2. CO decreased as a result of decreasing SV (Figure 2) and VTI, without significantly diminishing HR. Behind this we noted a decreasing left ventricular end diastolic area (LVEDA), while left ventricular end systolic area (LVESA) and FAC% did not change significantly. Associated with hyperoxaemia we recorded an increase in SVR or TPR, Z, and PR, while MAP, AC, PTT and PWV remained unchanged.

FiO2 PtcO2 (kPa) PtcCO2 (kPa)

The changes in systemic resistance and cardiac output in the present study are of the magnitude usually seen when healthy volunteers are exposed to increased arterial oxygen tensions – that is, an approximately 10% effect at the highest oxygen tension (Whitehorn et al. 1946, Joachimsson et al. 1996, Harten et al. 2003, Rousseau et al. 2005). We examined the effect of oxygen after early adaptation, and searched the small changes elicited by oxygen in the longer time perspective (Milone et al. 1999). To reduce the need for oxygen and possible metabolites washout periods of debatable duration (Waring et al. 2003), only increasing oxygen

Baseline

20 kPa (low)

40 kPa (medium)

60 kPa (high)

0.21* 10.4  1.2 5.3  0.6

0.45  0.04 18.0  2.4 5.5  0.6

0.75  0.05 36.3  5.5 5.4  0.7

1.00  0.0 57.2  11.4 5.4  0.7

Table 1 Measured oxygen and carbon dioxide pressures for each aimed level

Values are means  1SD. FiO2, inspired air O2 fraction; *, assumed value; PtcO2, trancutaneously measured partial O2 pressure; PtcCO2, transcutaneously measured partial CO2 pressure.

18

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

Acta Physiol 2007, 191, 15–24

Z Bak et al.

Æ Hyperoxaemic cardiovascular dose–response

Table 2 Central circulatory and vascular response to gradually-increased transcutaneously measured oxygen pressure levels (n ¼ 9) Baseline Central circulatory variables HRDoppler (beats min)1) VTI (cm) LVET (ms) SVDoppler (mL) SVicg (mL) CODoppler (l min)1) LVEDA (cm2) LVESA (cm2) FAC% Vascular variables MAPSubclavian (mmHg) ACicg (mL mmHg)1) ACDoppler (mL mmHg)1) PTTicg (ms) PWVicg (m s)1) SVRicg (dyne s cm)5) TPRDoppler (dyne s cm)5) ZDoppler (dyne s cm)5) PRDoppler (dyne s cm)5)

20 kPa

40 kPa

60 kPa

62 21.8 306 86 90 5.5 19.3 9.9 49

        

7 2.0 36 13 16 0.9 4.4 3.0 8.7

60 20.6 301 83 90 5.0 18.6 9.7 48

        

7 1.6 30 11 17 0.8 4.2 2.7 8.6

60 19.4 300 78 88 4.6 18.0 9.8 49

        

9 2.8 21 10 16 0.6 4.3 3.0 6.0

60 18.6 301 75 87 4.4 16.8 9.0 46

        

8 1.3a,b,c 22 9a,b,c 18a 0.6a,b 4.3a 2.0 8.1

90 1.5 1.6 98 10.7 1220 1330 110 1230

        

12 0.3 0.4 15 1.4 197 221 38 231

86 1.5 1.8 97 10.8 1200 1410 118 1310

        

10 0.3 1.0 14 1.6 161 241 27 245

87 1.5 1.4 95 11.1 1260 1530 126 1420

        

9 0.4 0.2 16 1.9 190 289 28 284

87 1.4 1.5 95 11.0 1290 1570 136 1440

        

10 0.3 0.8 16 2.1 215c 187a 19a 186a

Values are means  1SD. HR, heart rate; VTI, aortic velocity time integral; LVET, left ventricular ejection time; SV, stroke volume; CO, cardiac output; LVESA, left ventricular end systolic area; LVEDA, left ventricular end diastolic area; FAC%, fractional area change [(LVEDALVESA)/LVEDA]; MAP, mean arterial pressure; AC, arterial compliance; PTT, pulse transmission time; PWV, pulse wave velocity; SVR, systemic vascular resistance; T, total; PR, peripheral resistance; Z, characteristic impedance; Doppler, obtained with Doppler echocardiography and oscillometrically calibrated arterial subclavian pressure; icg, obtained by whole-body impedance cardiography. Significant change on level P < 0,05: a, significant difference between baseline and 60 kPa; b, between baseline and 40 kPa; c, between 20 kPa and 60 kPa; no further significant differences were found between other levels.

110 100

SV (ml)

90

86 83 78

80

75 70 60 0

Baseline/air

20 kPa

40 kPa

60 kPa

Figure 2 Dose–response to hyperoxaemia. Individual (thin lines) and mean (bold line) stroke volume as measured by Doppler echocardiography (SV) at increasing transcutaneously measured oxygen pressures.

levels were chosen. To avoid observer bias (Hoffmann et al. 1996), we used two separate non-invasive techniques and simultaneous blinding. It is an advantage of

the present study that two parallel methods were used, their intercorrelation was high, and the outcomes were similar. With the experimental outcome of this study it is tempting to discuss the underlying mechanism of the oxygen-induced changes to the vasculature. There is a linear dose–response relationship between arterial oxygen and cardiovascular parameters when the systemic oxygen tension increases above normal. It has been suggested that the oxygen-related effects are mediated either by primary or secondary changes in the haemoglobin. This is not supported by the findings in this study, where the dose relation found at supranormal arterial values was related to the arterial partial pressure of oxygen – where the oxygen saturation of the haemoglobin may further be assumed to be almost constant and close to 100%. A direct effect of supplemental oxygen on the vessels, also suggested by others (Joachimsson et al. 1996, Hanada et al. 2003) may therefore not be excluded. The vasculature in skeletal muscle, where the vascular bed is most important for the regulation of systemic vascular resistance, has also

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

19

Hyperoxaemic cardiovascular dose–response

Itcpt Central circulatory variables HRDoppler (beats min)1) VTI (cm) SVDoppler (mL) SVicg (mL) CODoppler (l min)1) LVEDA (cm2) LVESA (cm2) FAC% Vascular variables MAPoscillometric (mmHg) MAPsubclavian (mmHg) ACicg (mL mmHg)1) ACDoppler (mL mmHg)1) PTTicg (ms) PWVicg (m s)1) SVRicg (dyne s cm)5) TPRDoppler (dyne s cm)5) ZDoppler (dyne s cm)5) PRDoppler (dyne s cm)5)

61.5 22 87.4 90.7 5.45 19.6 10.3 48.2 83.4 88.5 1.51 1.67 97.1 10.8 1190 1326 106 1230

Æ Z Bak et al.

Acta Physiol 2007, 191, 15–24

Slope

SE Coef

P1

)0.027 )0.062 )0.228 )0.070 )0.018 )0.045 )0.023 )0.008

0.039 0.013 0.050 0.032 0.005 0.013 0.013 0.066

0.491 0.000* 0.000* 0.039* 0.002* 0.002* 0.103 0.901

0.017 )0.034 )0.0004 )0.004 )0.030 0.005 1.704 4.52 0.553 3.878

0.028 0.033 0.0012 0.004 0.068 0.007 0.560 1.463 0.142 1.472

0.542 0.305 0.727 0.357 0.658 0.530 0.005* 0.005* 0.001* 0.014*

Table 3 Regression analysis on central circulatory and vascular dose–response to oxygen

Results from analysis of variance of cardiac variables and transcutaneously measured oxygen pressures. (Regression equation: delta variable ¼ intercept + slope*dose). For haemodynamic variables see Table 2. Itcpt, intercept; Slope, gradient of the line; SE Coef, standard error of the slope; P1: the P-value for the slope in a straight linear model. *P < 0.05.

shown a similar relation to the oxygen partial pressure in arterial blood (Thorborg et al. 1988) and a decrease in sympathetic outflow to skeletal muscle resistance vessels during hyperoxia is described (Seals et al. 1991). It has also been shown that supplemental oxygen/ hyperoxaemia increases both the cardiac parasympathetic activity and the arterial–cardiac baroreflex function in a dose-dependent manner secondary to the increased blood pressure (Shibata et al. 2005). This suggests that the vasoconstriction of peripheral arteries is a local primary event preceding these secondary neural consequences. Another important finding of the present study is the reduction in left ventricular end-diastolic volume, which may explain the decrease in CO. There are several tentative explanations for this. The observed rapid decreases in LVEDA and SV without a change in HR indicate a decreased venous return and reduced central blood volume. Sympathetic receptor stimulation may decrease venous return but the results are conflicting (Leenen et al. 1988, Butterworth et al. 2004). Our results, however, indicate pooling of arterial or venous blood, which normally occurs in the capacitance vessels or in the larger veins (Guyton 1996), but for example the splanchnic vasculature may also be involved (Bell

20

et al. 1991). In the lung, hypoxic constriction is described (Dantzker et al. 1974), but not the contrary. A secondary sympathicolysis of the capacitance vessels and larger veins due to increased parasympathetic activity and arterial–cardiac baroreflex function (Shibata et al. 2005) as well as decreased sympathetic outflow to skeletal muscle resistance vessels (Seals et al. 1991) may contribute here. The location of blood pooling is however not indicated by this study, and to our knowledge there have been no studies on hyperoxemic volume effects on veins or pulmonary circulation. A recurring phenomenon in hyperoxia models is the finding of a large interindividual variability in the oxygen response and adaptation. This applies to the magnitude as well as to the time factor. It has previously been speculated that the oxygen-related effects result from the production of oxygen radicals, and it has been shown that different subjects may differ in their genetic predisposition to produce superoxide dismutase (SOD) (McCord & Fridovich 1969, Zelko et al. 2002). Interestingly, a large variability was also found among the present group of volunteers, and a high variability was particularly noted for the highest oxygen level.

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

Æ Hyperoxaemic cardiovascular dose–response

Acta Physiol 2007, 191, 15–24

Z Bak et al.

Early studies of hyperoxaemia claimed that the primary effect of oxygen was to stimulate the parasympathetic nervous system leading to a decrease in HR and in CO, which in turn lead to a secondary increase in SVR. Given the order of development known today, supported by the fact that we were unable to detect any change in HR, this line of reasoning is less likely. Instead, the results of the present study support previous assumptions that the primary, physiological effect of hyperoxaemia is the increase in vascular resistance to which the central circulation slowly adapts. This adaptation is made after a transient increase in blood pressure and a decrease in HR as shown previously (Whitehorn et al. 1946, Eggers et al. 1962, Waring et al. 2003). These effects may be more pronounced but seem to have a large interindividual variability (Joachimsson et al. 1996, Crawford et al. 1997, Milone et al. 1999, Rousseau et al. 2005). The early changes in the vasculature are modified by local and systemic adaptation (for example, autoregulation of local flow and systemic baroreceptor function), normally within 10– 15 min (Rousseau et al. 2005). It is therefore important to stress that the effects measured in this study were obtained after 15 min steady state, and only an increased SVR, and decreased SV and LVEDV occurred. The assumption that the peripheral vascular effect is the initiating event is strengthened by the fact that we were unable to record any direct effects on the heart, as for example, a change in FAC% by the increased oxygen levels. Further evidence supporting that the peripheral vasoconstriction is the primary change caused by hyperoxia and hyperoxaemia is the finding that the vascular changes can be elicited in a ‘heart free’ preparation, such as patients on a heart lung machine (Joachimsson et al. 1996). A significant confounding factor that must be considered in all experimental settings where oxygen tension levels are modified is a change in arterial CO2, as this induces an indirect consequence on the vasculature (Thomson et al. 2006). Theoretically the decrease in CO2 that is seen during hyperoxia elicits vasoconstriction. PtcCO2 was therefore assessed in this study. Transcutaneously measured CO2-levels are considered highly reliable and accurate (Dupeyrat et al. 1998). Contrary to several previous publications (Jedlinska et al. 1998, Mak et al. 2001, Rousseau et al. 2005, Singhal et al. 2005), we were unable to record any effect of hyperoxia on the CO2 levels, which reduces the likelihood of us eliciting a secondary CO2 effect on the vasculature. Possibly, the lack of effects of CO2 in this study may depend on the longer time periods studied after the first exposure to oxygen. Compared with hypoxia studies, hyperoxia studies are less confounded by strong compensatory mechanisms, such as increase in sympathetic tone and breathing stimulus.

Clinical relevance and limitations to the study. In this restricted number of volunteers, we found small but significant changes of cardiovascular parameters at supranormal arterial oxygen levels in a strict protocol where the participants acted as their own control. It needs to be pointed out that the study was performed totally non-invasively in order not to confound the specific effects of oxygen, which are known to be small. Reactive oxygen species, mediators or peripheral vasoregulatory substances like endothelin and NO-products were not measured. The finding of a decreasing CO at rest during hyperoxaemia has not been shown to decrease flow to insufficient values or to reduce oxygen supply, neither in healthy humans (Rousseau et al. 2005) nor in patients undergoing heart surgery (Joachimsson et al. 1996). These findings support that the adjustment process between the peripheral vascular bed and the central circulation normally is harmonized. From a practical perspective, hyperoxia may however be anticipated in a large number of clinical occasions when oxygen is provided to acutely ill patients (Haque et al. 1996, Saadjian et al. 1999, Mak et al. 2001, McDonald et al. 2005). Kemming et al. (2004) found that patients with anaemia, impaired LV function and impaired coronary flow kept their myocardial oxygenation and function when breathing supplemental oxygen. In patients with or without congestive heart failure, Mak et al. (2001) showed impaired cardiac relaxation and increased LV filling pressures after 20 min breathing of 100% oxygen. Haque et al. (1996) also presented that oxygen had detrimental effects on CO, SV, SVR and pulmonary capillary wedge pressure in heart failure. In a randomized, controlled, double blind study of oxygen therapy, patients with uncomplicated myocardial infarction and supplemental oxygen (PaO2 20 kPa for 24 h) tended to have a higher mortality and more ventricular tachycardia than those randomized to air (Rawles & Kenmure 1976). We did not examine patients but investigated the physiological effects of supplemental oxygen given to healthy volunteers. Our findings suggest restrictive use of supplemental oxygen therapy. This is further supported by a recent finding that oxygen treatment may be disadvantageous even in hypoxic patients (McDonald et al. 2005). Further studies. There are however issues other than integrative physiology or patophysiology that make further hyperoxia studies interesting. The mechanisms underlying the functions in which oxygen delivery to the tissues is regulated are still unknown. This applies both to the vasodilatation seen during hypoxia and vasoconstriction in hyperoxia. It is of physiological importance to know if the vasoconstrictive effect seen over the whole oxygen range (from hypoxia to hyperoxia) is one

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

21

Æ Z Bak et al.

Acta Physiol 2007, 191, 15–24

singular mechanism or not and to investigate if the linear dose–response relation that we report is present over the whole oxygen range in order to confirm a direct effect on the vessels, in contrast to a receptor or haemoglobin mediated reaction that should elicit a sigmoid relationship rather than a linear one. Our results that presume peripheral blood-pooling induced by hyperoxaemia further strengthens the complexity of the oxygen-induced reactions. In conclusion, the present study, like others, confirms that hyperoxia causes systemic vasoconstriction (increased systemic vascular resistance) and reduces cardiac output. A stable HR and ejection fraction, but decreasing stroke volume as a result of decreasing left ventricular end diastolic volume, indicates reduced preload and venous return, indicating extracardiac blood-pooling, probably due to compensatory venous sympathicolysis. To our knowledge this has not been previously observed. We have expanded these findings by showing a linear dose–response relation between increasing arterial oxygen and rising peripheral arterial resistance with decreasing stroke volume when the systemic oxygen tension increases above normal. A direct effect of supplemental oxygen on the vessels is thus suggested.

Bell, L., Zaret, B.L. & Rutlen, D.L. 1991. Influence of alphaadrenergic receptor stimulation on splanchnic intravascular volume in conscious humans. Acta Physiol Scand 143, 65–69. Benedict, F. & Higgins, H. 1911. Effects on men at rest of breathing oxygen-rich gas mixtures. Am J Physiol 28, 1–28. Bland, J.M. & Altman, D.G. 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1, 307–310. Bongard, O., Bounameaux, H. & Fagrell, B. 1992. Effects of oxygen inhalation on skin microcirculation in patients with peripheral arterial occlusive disease. Circulation 86, 878– 886. Butterworth, J., Gelman, S. & Mushlin, P.S. 2004. Correpondence: do alpha agonists increase venous return? Anesthesiology 101, 1038–1039. Colan, S.D., Borow, K.M. & Neumann, A. 1985. Use of the calibrated carotid pulse tracing for calculation of left ventricular pressure and wall stress throughout ejection. Am Heart J 109, 1306–1310. Crawford, P., Good, P.A., Gutierrez, E. et al. 1997. Effects of supplemental oxygen on forearm vasodilation in humans. J Appl Physiol 82, 1601–1606. Daly, W.J. & Bondurant, S. 1962. Effects of oxygen breathing on the heart rate, blood pressure, and cardiac index of normal men – resting, with reactive hyperemia, and after atropine. J Clin Invest 41, 126–132. Dantzker, D.R., Wagner, P.D. & West, J.B. 1974. Proceedings: instability of poorly ventilated lung units during oxygen breathing. J Physiol 242, 72P. Dupeyrat, A., Goujard, E., Muret, J. & Ecoffey, C. 1998. Transcutaneous CO2 tension effects of clonidine in paediatric caudal analgesia. Paediatr Anaesth 8, 145–148. Eggers, G.W.N. Jr, Paley, H.W., Leonard, J.J. & Warren, J.V. 1962. Hemodynamic responses to oxygen breathing in man. J Appl Physiol 17, 75–79. Ellsworth, M.L. 2000. The red blood cell as an oxygen sensor: what is the evidence? Acta Physiol Scand 168, 551–559. Frobert, O., Moesgaard, J., Toft, E., Poulsen, S.H. & Sogaard, P. 2004. Influence of oxygen tension on myocardial performance. Evaluation by tissue Doppler imaging. Cardiovasc Ultrasound 2, 22. Gonzalez-Alonso, J., Olsen, D.B. & Saltin, B. 2002. Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery: role of circulating ATP. Circ Res 91, 1046–1055. Guyton, A.C. 1996. Vascular distensibility, and functions of the arterial and venous systems. In: A. Guyton & E. Hall (eds). Textbook of Medical Physiology, 9th edn pp. 171– 173. Saunders, Philadelphia. Hampl, V., Bibova, J., Stranak, Z. et al. 2002. Hypoxic fetoplacental vasoconstriction in humans is mediated by potassium channel inhibition. Am J Physiol Heart Circ Physiol 283, H2440–2449. Hanada, A., Sander, M. & Gonzalez-Alonso, J. 2003. Human skeletal muscle sympathetic nerve activity, heart rate and limb haemodynamics with reduced blood oxygenation and exercise. J Physiol 551, 635–647.

Hyperoxaemic cardiovascular dose–response

Conflict of interest The authors have no conflicts of interest. We thank Dr Svend Aakhus PhD, Department of Cardiology, Rikshospitalet, Oslo, Norway, and Dr Tiit Ko¨o¨bi PhD, Department of Clinical Physiology and Nuclear Medicine, University Hospital Tampere, Finland, for their advice and unselfish help when applying the Doppler and impedance techniques, respectively; and Olle Eriksson, Linko¨ping University, for statistical calculations and advice. Grants were ¨ stergo¨tland and obtained from the County Council of O Linko¨ping University; B Janerot-Sjoberg was partly financed by the Swedish Medical Research Council.

References Aakhus, S., Soerlie, C., Faanes, A. et al. 1993a. Noninvasive computerized assessment of left ventricular performance and systemic haemodynamics by study of aortic root pressure and flow estimates in healthy men, and men with acute and healed myocardial infarction. Am J Cardiol 72, 260–267. Aakhus, S., Torp, H., Haugland, T. & Hatle, L. 1993b. Noninvasive estimates of aortic root pressures: external subclavian arterial pulse tracing calibrated by oscillometrically determined brachial arterial pressures. Clin Physiol 13, 573– 586. Baron, J.F., Vicaut, E., Hou, X. & Duvelleroy, M. 1990. Independent role of arterial O2 tension in local control of coronary blood flow. Am J Physiol 258, H1388–1394.

22

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

Æ Hyperoxaemic cardiovascular dose–response

Acta Physiol 2007, 191, 15–24

Z Bak et al.

Haque, W.A., Boehmer, J., Clemson, B.S., Leuenberger, U.A., Silber, D.H. & Sinoway, L.I. 1996. Hemodynamic effects of supplemental oxygen administration in congestive heart failure. J Am Coll Cardiol 27, 353–357. Harten, J.M., Anderson, K.J., Angerson, W.J., Booth, M.G. & Kinsella, J. 2003. The effect of normobaric hyperoxia on cardiac index in healthy awake volunteers. Anaesthesia 58, 885–888. Hoffmann, R., Lethen, H., Marwick, T. et al. 1996. Analysis of interinstitutional observer agreement in interpretation of dobutamine stress echocardiograms. J Am Coll Cardiol 27, 330–336. Jedlinska, B., Mellstrom, A., Hartmann, M. & Jonsson, K. 1998. Comparison of tissue oxygen-tension measurements by different devices. An experimental study in pigs. Scand J Clin Lab Invest 58, 63–71. Joachimsson, P.O., Sjoberg, F., Forsman, M., Johansson, M., Ahn, H.C. & Rutberg, H. 1996. Adverse effects of hyperoxemia during cardiopulmonary bypass. J Thorac Cardiovasc Surg 112, 812–819. Kemming, G.I., Meisner, F.G., Meier, J. et al. 2004. Hyperoxic ventilation at the critical hematocrit: effects on myocardial perfusion and function. Acta Anaesthesiol Scand 48, 951– 959. Kiss, B., Polska, E., Dorner, G. et al. 2002. Retinal blood flow during hyperoxia in humans revisited: concerted results using different measurement techniques. Microvasc Res 64, 75–85. Koobi, T., Kaukinen, S., Ahola, T. & Turjanmaa, V.M. 1997a. Non-invasive measurement of cardiac output: whole-body impedance cardiography in simultaneous comparison with thermodilution and direct oxygen Fick methods. Intensive Care Med 23, 1132–1137. Koobi, T., Kaukinen, S., Turjanmaa, V.M. & Uusitalo, A.J. 1997b. Whole-body impedance cardiography in the measurement of cardiac output. Crit Care Med 25, 779– 785. Koobi, T., Kaukinen, S. & Kauppinen, P. 2001. Comparison of methods for cardiac output measurement. Crit Care Med 29, 1092. Leenen, F.H., Chan, Y.K., Smith, D.L. & Reeves, R.A. 1988. Effects of beta-1 vs nonselective beta-blockade. Clin Pharmacol Ther 43, 519–528. Mak, S., Azevedo, E.R., Liu, P.P. & Newton, G.E. 2001. Effect of hyperoxia on left ventricular function and filling pressures in patients with and without congestive heart failure. Chest 120, 467–473. McCord, J.M. & Fridovich, I. 1969. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J Biol Chem 244, 6049–6055. McDonald, C.F., Crockett, A.J. & Young, I.H. 2005. Adult domiciliary oxygen therapy. Position statement of the Thoracic Society of Australia and New Zealand. Med J Aust 182, 621–626. McMahon, T.J., Moon, R.E., Luschinger, B.P. et al. 2002. Nitric oxide in the human respiratory cycle. Nat Med 8, 711–717. McNulty, P.H., King, N., Scott, S. et al. 2005. Effects of supplemental oxygen administration on coronary blood flow in

patients undergoing cardiac catheterization. Am J Physiol Heart Circ Physiol 288, H1057–1062. Messina, E.J., Sun, D., Koller, A., Wolin, M.S. & Kaley, G. 1994. Increases in oxygen tension evoke arteriolar constriction by inhibiting endothelial prostaglandin synthesis. Microvasc Res 48, 151–160. Milone, S.D., Newton, G.E. & Parker, J.D. 1999. Hemodynamic and biochemical effects of 100% oxygen breathing in humans. Can J Physiol Pharmacol 77, 124–130. Pittman, R.N. 1995. Influence of microvascular architecture on oxygen exchange in skeletal muscle. Microcirculation 2, 1–18. Prieur, F., Busso, T., Castells, J. et al. 1998. Validity of oxygen uptake measurements during exercise under moderate hyperoxia. Med Sci Sports Exerc 30, 958–962. Rawles, J.M. & Kenmure, A.C. 1976. Controlled trial of oxygen in uncomplicated myocardial infarction. Br Med J 1, 1121–1123. Rousseau, A., Bak, Z., Janerot-Sjoberg, B. & Sjoberg, F. 2005. Acute hyperoxaemia-induced effects on regional blood flow, oxygen consumption and central circulation in man. Acta Physiol Scand 183, 231–240. Saadjian, A., Paganelli, F. & Levy, S. 1999. Hemodynamic response to oxygen administration in chronic heart failure: role of chemoreflexes. J Cardiovasc Pharmacol 33, 144– 150. Seals, D.R., Johnson, D.G. & Fregosi, R.F. 1991. Hyperoxia lowers sympathetic activity at rest but not during exercise in humans. Am J Physiol 260, R873–878. Shibata, S., Iwasaki, K., Ogawa, Y., Kato, J. & Ogawa, S. 2005. Cardiovascular neuroregulation during acute exposure to 40, 70, and 100% oxygen at sea level. Aviat Space Environ Med 76, 1105–1110. Singhal, A.B., Benner, T., Roccatagliata, L. et al. 2005. A pilot study of normobaric oxygen therapy in acute ischemic stroke. Stroke 36, 797–802. Sjoberg, B. & Wranne, B. 1990. Cardiac output determined by ultrasound-Doppler: clinical applications. Clin Physiol 10, 463–473. Skjaerpe, T., Hegrenaes, L. & Hatle, L. 1985. Noninvasive estimation of valve area in patients with aortic stenosis by Doppler ultrasound and two-dimensional echocardiography. Circulation 72, 810–818. Thomson, A.J., Drummond, G.B., Waring, W.S., Webb, D.J. & Maxwell, S.R. 2006. Effects of short-term isocapnic hyperoxia and hypoxia on cardiovascular function. J Appl Physiol 101, 809–816. Thorborg, P., Malmqvist, L.A. & Lund, N. 1988. Surface oxygen pressure distributions in rabbit skeletal muscle: dependence on arterial pO2. Microcirc Endothelium Lymphatics 4, 169–192. Waring, W.S., Thomson, A.J., Adwani, S.H. et al. 2003. Cardiovascular effects of acute oxygen administration in healthy adults. J Cardiovasc Pharmacol 42, 245–250. Watson, N.A., Beards, S.C., Altaf, N., Kassner, A. & Jackson, A. 2000. The effect of hyperoxia on cerebral blood flow: a study in healthy volunteers using magnetic resonance phase-contrast angiography. Eur J Anaesthesiol 17, 152–159.

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x

23

Æ Z Bak et al.

Acta Physiol 2007, 191, 15–24

Westerhof, N., Elzinga, G. & Sipkema, P. 1971. An artificial arterial system for pumping hearts. J Appl Physiol 31, 776– 781. Whalen, R.E., Saltzman, H.A., Holloway, D.H. Jr, McIntosh, H.D., Sieker, H.O. & Brown, I.W. Jr. 1965. Cardiovascular and blood gas responses to hyperbaric oxygenation. Am J Cardiol 15, 638–646. Whitehorn, W.V., Edelmann, A. & Hitchcock, F.A. 1946. The cardiovascular responses to the breathing of 100 per cent

oxygen at normal barometric pressure. Am J Physiol 146, 61–65. Zelko, I.N., Mariani, T.J. & Folz, R.J. 2002. Superoxide dismutase multigene family: a comparison of the CuZn-SOD (SOD1), Mn-SOD (SOD2), and EC-SOD (SOD3) gene structures, evolution, and expression. Free Radic Biol Med 33, 337–349.

Hyperoxaemic cardiovascular dose–response

24

 2007 The Authors Journal compilation  2007 Scandinavian Physiological Society, doi: 10.1111/j.1748-1716.2007.01710.x