Cerebral Oxygenation During Exercise in Cardiac Patients

Cerebral Oxygenation During Exercise in Cardiac Patients* Akira Koike, MD; Haruki Itoh, MD; Reiko Oohara, BS; Masayo Hoshimoto; Akihiko Tajima, BS; Tadanori Aizawa, MD; and Long Tai Fu, MD

Background: Until recently, compensatory mechanisms have been believed to regulate adequately cerebral blood flow in humans. However, this has been called into question by a series of new investigations suggesting that patients with left ventricular dysfunction suffer from cerebral hypoperfusion. We compared cerebral oxygenation during incremental exercise between patients with valvular heart disease and normal subjects. Methods: Thirty-three patients with valvular disease and 33 normal subjects performed a symptomlimited incremental exercise test using a cycle ergometer. Oxyhemoglobin at the forehead was continuously monitored during exercise using near-infrared spectroscopy. Respiratory gas measurements were performed on a breath-by-breath basis. Results: The increase in oxyhemoglobin during exercise was significantly lower in the patients with valvular disease than in normal subjects. The change in oxyhemoglobin during exercise (⌬O2Hb) at the forehead was negatively correlated with the slope of the increase in minute ventilation to the increase in carbon dioxide output (⌬V˙E/⌬V˙CO2), and positively correlated with the peak oxygen uptake (V˙O2), gas exchange threshold (GET), and slope of the increase in V˙O2 to the increase in the work rate (⌬V˙O2/⌬WR). Among the patients with valvular disease, 15 patients showed a decrease in oxyhemoglobin at the forehead during exercise. When compared with the patients with increased oxyhemoglobin, those with decreased levels exhibited a higher ⌬V˙E/⌬V˙CO2 and a lower peak V˙O2, GET, and ⌬V˙O2/⌬WR. Conclusions: The present findings strongly suggest that cerebral oxygenation during exercise is dependent on the cardiovascular and pulmonary systems. The study also indicated the presence of cerebral hypoperfusion during exercise in cardiac patients whose cardiac output fails to increase normally. (CHEST 2004; 125:182–190) Key words: brain; cerebrovascular circulation; exercise Abbreviations: GET ⫽ gas exchange threshold; GET Vo2 ⫽ gas exchange threshold carbon dioxide output; GET WR ⫽ gas exchange threshold work rate; NIRS ⫽ near-infrared spectroscopy; ⌬O2Hb ⫽ change in oxyhemoglobin during exercise; Petco2 ⫽ end-tidal carbon dioxide partial pressure; Peto2 ⫽ end-tidal oxygen partial pressure; Spo2 ⫽ pulse oximetric saturation; TOI ⫽ tissue oxygenation index; V˙co2 ⫽ carbon dioxide output; V˙e ⫽ minute ventilation; ⌬V˙e/⌬V˙co2 ⫽ slope of the increase in minute ventilation to the increase in carbon dioxide output; V˙o2 ⫽ oxygen uptake; ⌬V˙o2/⌬WR ⫽ slope of the increase in oxygen uptake to the increase in the work rate; WR ⫽ work rate

recently, complex compensatory mechaU ntil nisms have been believed to regulate adequately the blood flow to vital organs, especially to the brain.1,2 However, a new series of investigations using magnetic resonance spectroscopy has suggested that cardiac patients with left ventricular dysfunction may suffer from cerebral hypoperfusion.3 During rest, the blood flow to *From The Cardiovascular Institute, Tokyo, Japan. This study was supported in part by a grant from the Takeda Science Foundation. Manuscript received December 27, 2002; revision accepted June 24, 2003. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: Akira Koike, MD, The Cardiovascular Institute, 3-10, Roppongi 7-chome, Minato-ku, Tokyo 106-0032, Japan; e-mail: [email protected] 182

the main organs is sufficiently maintained even in these patients. During exercise, however, the oxygen demand by muscle cells increases up to 10 to 15 times that in the resting condition. To meet this sudden surge in oxygen demand, the blood flow to the muscle cells must increase tremendously. Given that the blood flow to each organ is determined by cardiac output, increased distribution of blood flow to muscle may result in relative hypoperfusion in other organs. As a consequence, the cerebral circulation may become insufficient, especially in cardiac patients in whom the cardiac output fails to increase normally. In spite of this peril, only limited data has been collected on cerebral oxygenation in these patients. The recent development of near-infrared spectroscopy (NIRS) has expanded the diagnostic assessment for tissue oxygenation.4 –11 In the present study, Clinical Investigations

we hypothesized that cerebral oxygenation may become insufficient during exercise in cardiac patients. To test this hypothesis, we used an NIRS system to continuously measure the change in cerebral oxygenation during incremental exercise both in patients with valvular heart disease and normal subjects. We also compared the indexes obtained from NIRS with the parameters of cardiopulmonary exercise testing in order to determine the factors influencing cerebral oxygenation. Materials and Methods Study Patients Thirty-three consecutive patients with valvular heart disease (age, 64.7 ⫾ 12.2 years [mean ⫾ SD]) were studied between

February 2001 and December 2001 (Table 1). Those with cerebrovascular disease diagnosed based on clinical documentation were excluded from the study. All patients were in clinically stable condition at the time of the study. Twenty-one patients were in sinus rhythm, and 12 patients were in atrial fibrillation. Medications influencing hemodynamic variables included diuretics prescribed in 19 cases, digitalis in 12 cases, calcium-channel blockers in 9 cases, nitrates in 9 cases, angiotensin-receptor blockers in 7 cases, angiotensin-converting enzyme inhibitors in 6 cases, and ␤-blockers in 3 cases. The control group was made up of 33 subjects ⱖ 50 years old who were recruited from a medical screening clinic during same period and confirmed to be free of any significant heart disease on the basis of history, physical examination, chest radiograph, 12-lead ECG, and echocardiography. All of the control subjects had a normal ECG response on a maximal ergometer exercise. The Human Subjects Committee approved the protocol and procedures for the exercise testing. The purposes and risks of the study were explained to the patients, and written informed consent was obtained from each patient.

Table 1—Clinical Characteristics and Diagnosis in Each Patient* Patient No.

Age, yr

With ⌬O2Hb ⬎ 0 1 52 2 33 3 42 4 45 5 54 6 70 7 75 8 51 9 68 10 73 11 67 12 66 13 69 14 70 15 75 16 79 17 63 18 74 With ⌬O2Hb ⬍ 0 19 54 20 53 21 64 22 62 23 84 24 76 25 58 26 77 27 73 28 62 29 87 30 61 31 66 32 56 33 75 Mean 62.6 SD 13.2

Gender

Height, cm

Weight, kg

LVEF, %

Peak V˙ o2, mL/min/kg

Cardiac Diagnosis

Male Male Male Male Female Female Male Male Female Female Male Male Male Male Female Female Female Male

168.1 168.6 180.0 170.0 159.0 158.5 167.0 160.5 155.0 144.0 172.8 171.0 167.3 163.6 142.2 149.5 156.5 171.0

77.0 64.5 64.0 63.0 76.5 50.5 49.5 63.5 52.9 42.8 66.8 66.5 71.5 68.6 45.7 53.5 51.0 75.0

45.0 62.0 54.0 68.0 61.0 68.0 67.0 62.0 45.0 76.0 59.0 54.0 70.0 72.0 63.0 73.0 69.0 57.0

18.5 23.1 21.1 22.2 14.7 21.1 23.2 30.7 11.3 15.0 15.4 22.3 23.1 15.4 17.0 14.9 25.5 18.0

MR AR AR MR MR AR, MR AR ASR ASR AR, MR, TR MR, TR AR MR, TR TR ASR, MR ASR AR, MR AR, MR

Male Female Male Female Male Female Female Male Female Male Female Female Female Male Male

175.3 153.1 160.3 142.0 170.0 157.6 147.3 167.5 149.0 175.6 151.5 157.6 152.0 167.6 161.6 162.5 10.2

50.0 43.0 62.0 44.9 51.0 50.0 42.0 59.5 39.7 75.3 55.0 49.0 44.0 69.5 64.0 61.3 10.8

16.0 65.0 72.0 65.0 73.0 53.0 64.0 33.0 68.0 61.0 55.0 56.0 51.0 54.0 29.0 62.5 9.0

6.2 8.0 13.7 13.5 12.7 14.0 15.9 13.4 12.2 11.1 10.4 22.6 10.5 10.6 14.6 19.6 4.8

MR MR MSR MSR, TR TR ASR, MR TR AR, TR TR MR, TR MR AR MR, TR AR, MR, TR AR

*LVEF ⫽ left ventricular ejection fraction; AR ⫽ aortic regurgitation; ASR ⫽ aortic stenoregurgitation; MR ⫽ mitral regurgitation; MSR ⫽ mitral stenoregurgitation; TR ⫽ tricuspid regurgitation. www.chestjournal.org

CHEST / 125 / 1 / JANUARY, 2004

183

Exercise Testing An incremental symptom-limited maximal exercise test was performed using an upright, electromagnetically braked cycle ergometer (Corival 400; Lode; Groningen, Holland). After beginning with a 4-min warm-up at 20 W of 60 revolutions per minute, the exercise load was increased incrementally by 1 W every 6 s (10 W/min). ECG was monitored continuously during the test (System ML-6500; Fukuda Denshi; Tokyo, Japan). Cuff BP was measured at rest on a cycle ergometer, and then every minute during exercise testing with an automatic indirect manometer (STBP-780; Nippon Colin; Aichi, Japan).12 Pulse oximetric saturation (Spo2) was monitored using a pulse oximeter (DDG-2001; Nihon Kohden; Tokyo, Japan) attached at the left earlobe. NIRS Monitoring Cerebral oxygenation was monitored using a commercially available NIRS system (NIRO-300; Hamamatsu Photonics KK; Hamamatsu, Japan). A probe holder containing an emission probe and detection probe was attached at the left side of forehead with a distance of 5 cm between the probes. The methodology of this system has been described in detail in previous reports.7–11 NIRO300 measures the concentration changes of oxyhemoglobin and deoxyhemoglobin using a modified Beer-Lambert law.9,11 It gives an absolute unit (micromoles per liter) for the changes in oxyhemoglobin and deoxyhemoglobin by incorporating an optical path length. For the brain, this path length is 30 cm when the distance between the emission probe and detection probe is set at 5 cm.5,7 NIRO-300 also measures a tissue oxygenation index (TOI), which can be expressed as oxyhemoglobin/(oxyhemoglobin ⫹ deoxyhemoglobin) ⫻ 100 (expressed as percentage), using a photon-diffusion theory.6,9 Oxyhemoglobin, deoxyhemoglobin, and TOI were measured every 2 s from 4 min before the start of exercise until the end of exercise, and expressed as the magnitude of the change from the initial value. Variables of NIRS at rest were determined as the averages of values obtained as the subjects sat on the ergometer over a 4-min period before the start of the exercise test. Each variable at peak exercise was defined as the average value obtained during the last 30 s of incremental exercise. The change in each variable during exercise was defined as the peak exercise value – resting value. Respiratory Gas Analysis Oxygen uptake (V˙ o2), carbon dioxide output (V˙ co2), and minute ventilation (V˙ e) were measured throughout the test using an Aeromonitor AE-300S (Minato Medical Science; Osaka, Japan).13,14 End-tidal oxygen partial pressure (Peto2) and end-tidal carbon dioxide partial pressure (Petco2) were also measured using this device. Prior to calculating the parameters from respiratory gas analysis, a 5-point moving average of the breathby-breath data were performed. Peak V˙ o2 was defined as the average value obtained during the last 15 s of incremental exercise. The gas exchange threshold (GET) was determined by the V-slope method,15–17 and expressed as both V˙ o2 (GET V˙ o2) and the work rate (WR) [GET WR] at the threshold point. The slope of the increase in V˙ o2 to the increase in the WR (⌬V˙ o2/⌬WR) was calculated from the data recorded between 30 s after the start of incremental exercise to 30 s before the end of exercise by least-squares linear regression.18 The slope of the increase in ventilation to the increase in carbon dioxide output (⌬V˙ e/⌬V˙ co2) was calculated from the start of incremental exercise to the respiratory compensation point by least-squares linear regression.18,19 184

Reproducibility of Cerebral Oxyhemoglobin Values During Exercise Reproducibility of the change in oxyhemoglobin during exercise (⌬O2Hb) at the forehead was assessed in 12 patients with stable chronic heart disease (11 patients with coronary artery disease and 1 patient with valvular heart disease). ⌬O2Hb values were compared between two incremental symptom-limited exercise tests with the same protocol. The tests were performed at an interval of 11.3 ⫾ 4.6 days. Statistics Data are presented as the mean ⫾ SD. Comparisons of variables between the patients and normal subjects and those among the subgroups of patients were made by the unpaired t test or ␹2 analysis where appropriate. Linear regression analysis was used to correlate the measured variables. The reproducibility of oxyhemoglobin values during exercise was assessed both by linear regression analysis and the method of Bland and Altman.20 For all comparisons, p ⬍ 0.05 was considered statistically significant.

Results Figure 1 represents the changes in the measures of cerebral oxygenation during exercise in two representative subjects: a normal subject without heart disease (left, A) and a patient with valvular heart disease (right, B). In the normal subject, oxyhemoglobin and TOI at the forehead increased during exercise, while deoxyhemoglobin showed no consistent change (decreased initially and then increased slightly). In the patient with valvular disease, oxyhemoglobin at the forehead gradually decreased during exercise, while deoxyhemoglobin increased. ⌬O2Hb at the forehead was determined for 12 cardiac patients from two incremental exercise tests conducted on separate test days. As shown in Figure 2, there was good reproducibility in the measurements of ⌬O2Hb during exercise (r ⫽ 0.88, p ⬍ 0.0001). There was a significant negative correlation between ⌬O2Hb and age in normal subjects (r ⫽ ⫺ 0.64, p ⬍ 0.0001), indicating a lower increase or even a decrease in oxyhemoglobin during exercise at higher age. Comparison of Cardiopulmonary Indexes and Cerebral ⌬O2Hb Between Cardiac Patients and Normal Subjects Table 2 demonstrates NIRS, hemodynamic, and respiratory gas variables in normal subjects and patients with valvular heart disease. The patients with valvular disease had significantly lower left ventricular ejection fraction than normal subjects. Maximal WR, peak V˙ o2, GET V˙ o2, ⌬V˙ o2/⌬WR, and BP at peak exercise were significantly lower in the patients than in normal subjects. The patients with valvular disease had a lower heart rate at peak exercise and a higher slope of ⌬V˙ e/⌬V˙ co2 than normal subjects. ⌬O2Hb at the foreClinical Investigations

Figure 1. Plots of NIRS at the forehead and Spo2 during exercise in a normal subject (left, A; a 61-year-old male subject) and a patient with valvular heart disease (right, B; patient 32 in Table 1). Data of NIRS were collected every 2 s and expressed as a 5-point moving average. O2Hb ⫽ oxyhemoglobin; HHb ⫽ deoxyhemoglobin.

head during exercise was significantly lower in the patients with valvular disease than in normal subjects (0.97 ⫾ 3.96 ␮mol/L vs 4.12 ⫾ 4.83 ␮mol/L, p ⫽ 0.005). ⌬TOI also tended to be lower in the patients. Relation Between ⌬O2Hb During Exercise and Cardiopulmonary Indexes Figure 3 shows the relation between ⌬O2Hb at the forehead and the parameters of cardiopulmonary exer-

cise testing for all subjects (33 patients and 33 normal subjects). ⌬O2Hb was lower in subjects with lower peak V˙ o2 values, with a significant positive correlation between the two variables (r ⫽ 0.61, p ⬍ 0.0001). ⌬O2Hb at the forehead was also significantly correlated with GET V˙ o2 (r ⫽ 0.46, p ⬍ 0.0001) and ⌬V˙ o2/⌬WR (r ⫽ 0.57, p ⬍ 0.0001). ⌬O2Hb at the forehead showed a significant negative correlation with ⌬V˙ e/⌬V˙ co2 (r ⫽ ⫺ 0.45, p ⫽ 0.0001). When the analysis was performed only on the patients with valvular disease,

Figure 2. Graphs showing the reproducibility of the ⌬O2Hb at the forehead (left, A) and the Bland and Altman method20 with horizontal lines corresponding to the mean and the 2 SD above and below the mean of differences between ⌬O2Hb measured in the two tests (right, B). The line of identity is shown (left, A). www.chestjournal.org

CHEST / 125 / 1 / JANUARY, 2004

185

Table 2—Hemodynamic and Respiratory Gas Variables in Normal Subjects and Patients With Valvular Disease* Variables Male/female gender, No. Age, yr Height, cm Weight, kg Rest Left ventricular ejection fraction, % Heart rate, beats/min Systolic BP, mm Hg Diastolic BP, mm Hg Gas exchange ratio Peto2, mm Hg Petco2, mm Hg Spo2, % Peak exercise Heart rate, beats/min Systolic BP, mm Hg Diastolic BP, mm Hg Gas exchange ratio Peto2, mm Hg Petco2, mm Hg Spo2, % ⌬O2Hb, ␮mol/L ⌬HHb, ␮mol/L ⌬TOI, % Exercise parameters Maximum WR, W Peak V˙ o2, mL/min/kg ⌬V˙ o2/⌬WR, mL/min/W ⌬V˙ e/⌬V˙ co2 GET WR, W GET V˙ o2, mL/min/kg

Normal Subjects (n ⫽ 33)

Patients With Valvular Disease (n ⫽ 33)

p Value

19/14 60.3 ⫾ 8.3 162.3 ⫾ 8.7 61.9 ⫾ 10.9

18/15 64.7 ⫾ 12.2 161.0 ⫾ 10.2 57.6 ⫾ 11.3

1.00 0.09 0.58 0.12

70.8 ⫾ 7.1 73.1 ⫾ 13.8 132.8 ⫾ 18.7 77.8 ⫾ 10.0 0.87 ⫾ 0.06 111.2 ⫾ 3.3 33.5 ⫾ 2.6 97.9 ⫾ 1.2

58.8 ⫾ 13.4 79.5 ⫾ 14.8 129.3 ⫾ 19.4 70.0 ⫾ 14.0 0.90 ⫾ 0.05 112.8 ⫾ 3.2 33.1 ⫾ 2.9 97.8 ⫾ 1.6

⬍ 0.0001 0.07 0.46 0.01 0.03 0.048 0.57 0.74

146.5 ⫾ 17.0 217.1 ⫾ 23.3 103.1 ⫾ 13.2 1.12 ⫾ 0.07 112.3 ⫾ 4.6 39.8 ⫾ 3.9 96.5 ⫾ 2.1 4.12 ⫾ 4.83 0.07 ⫾ 1.45 2.61 ⫾ 5.24

130.7 ⫾ 29.2 181.0 ⫾ 36.8 90.2 ⫾ 20.0 1.09 ⫾ 0.08 113.1 ⫾ 6.2 38.1 ⫾ 6.5 96.3 ⫾ 2.4 0.97 ⫾ 3.96 0.33 ⫾ 1.62 0.42 ⫾ 4.00

0.009 ⬍ 0.0001 0.003 0.09 0.54 0.19 0.77 0.005 0.51 0.06

109.4 ⫾ 25.6 22.4 ⫾ 4.0 10.1 ⫾ 1.4 29.2 ⫾ 3.4 67.2 ⫾ 15.0 16.2 ⫾ 3.3

77.8 ⫾ 30.6 16.4 ⫾ 5.6 7.6 ⫾ 2.5 34.7 ⫾ 9.6 49.6 ⫾ 21.1 12.8 ⫾ 3.8

⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001 0.003 0.0003 0.0002

*Data presented are mean value ⫾ SD unless otherwise indicated. HHb ⫽ deoxyhemoglobin.

⌬O2Hb at the forehead was still significantly correlated with peak V˙ o2 (r ⫽ 0.64, p ⬍ 0.0001), GET V˙ o2 (r ⫽ 0.47, p ⫽ 0.005), ⌬V˙ o2/⌬WR (r ⫽ 0.55, ˙ ˙ p ⫽ 0.0008), and ⌬Ve/⌬Vco2 (r ⫽ ⫺ 0.43, p ⫽ 0.013). Even in the analysis of the normal subjects, ⌬O2Hb was significantly correlated with peak V˙ o2 (r ⫽ 0.45, p ⫽ 0.008), ⌬V˙ o2/⌬WR (r ⫽ 0.51, p ⫽ 0.002), and ⌬V˙ e/⌬V˙ co2 (r ⫽ ⫺ 0.47, p ⫽ 0.005). ⌬O2Hb at the forehead was significantly lower in the patients with peak V˙ o2 ⬍ 15 mL/min/kg (n ⫽ 16) than in those with peak V˙ o2 ⱖ 15 mL/min/kg (n ⫽ 17): ⫺ 1.04 ⫾ 3.01 ␮mol/L vs 2.86 ⫾ 3.89 ␮mol/L (p ⫽ 0.003). ⌬O2Hb During Exercise in Patients With Valvular Heart Disease Table 3 presents a comparison of measured variables between the patients who showed an increase in oxyhemoglobin (⌬O2Hb ⬎ O, n ⫽ 18) and those who showed a decrease in the same variable (⌬O2Hb ⬍ 0, n ⫽ 15). There were no differences between the two groups in the prescribed medica186

tions, or in the gender, age, or height. However, the patients with ⌬O2Hb ⬍ 0 had a lower body weight, and there were several differences in the cardiopulmonary variables measured. Specifically, peak V˙ o2 was lower (12.6 ⫾ 3.8 mL/min/kg vs 19.6 ⫾ 4.8 mL/min/kg, p ⬍ 0.0001) and left ventricular ejection fraction tended to be lower (54.3 ⫾ 16.5% vs 62.5 ⫾ 9.0%, p ⫽ 0.08) in the patients with ⌬O2Hb ⬍ 0 than in those with ⌬O2Hb ⬎ 0. The patients with ⌬O2Hb ⬍ 0 had a lower GET V˙ o2 and ⌬V˙ o2/⌬WR than those with ⌬O2Hb ⬎ 0, as well as a higher ⌬V˙ e/⌬V˙ co2. While there was no difference in Spo2 at peak exercise between the two groups, the patients with ⌬O2Hb ⬍ 0 had slightly but significantly higher Peto2 values at peak exercise. The patients with ⌬O2Hb ⬍ 0 had significantly lower Petco2 at peak exercise (34.8 ⫾ 7.0 mm Hg vs 40.8 ⫾ 4.6 mm Hg, p ⫽ 0.006). Discussion In a recent investigation of the cerebral metabolism of heart failure patients using proton magClinical Investigations

Figure 3. Scatterplots showing the relationships of ⌬O2Hb with the peak V˙ o2 (top left, A), GET V˙ o2 (top right, B), ⌬V˙ o2/⌬WR (bottom left, C), and ⌬V˙ e/⌬V˙ co2 (bottom right, D) in 33 patients with valvular heart disease and 33 normal subjects.

netic resonance spectroscopy, Lee et al3 discovered abnormalities in the cerebral metabolism of patients with advanced heart failure. They speculated that this abnormality was chiefly attributable to cerebral hypoperfusion.3 In the present study, we found that the cerebral oxygenation during exercise was strongly related to the indexes of cardiopulmonary variables in patients with valvular heart disease. Cerebral oxyhemoglobin was found to decrease during maximal exercise in nearly half of the patients with valvular heart disease enrolled in the study. Parameters of Cardiopulmonary Exercise Testing ⌬O2Hb at the forehead during exercise was found to be significantly related to peak V˙ o2, GET, ⌬V˙ o2/⌬WR, and ⌬V˙ e/⌬V˙ co2 obtained from cardiopulmonary exercise testing. Peak V˙ o2, an index normally determined by maximum cardiac output during exercise, correlates well with the degree of hemodynamic abnormality in patients with cardiovascular disease.21–23 GET (anaerobic threshold), a threshold of the exercise intensity above which lactic acidosis develops,24,25 also reflects the degree of cardiovascular impairment.23,26 The slope of ⌬V˙ o2/⌬WR is determined www.chestjournal.org

by the increasing cardiac output and increasing difference between arterial and mixed venous oxygen content during incremental exercise. ⌬V˙ o2/⌬WR is approximately 10 mL/min/W in healthy subjects,27 and falls to progressively lower levels in patients with heart disease as the disease worsens.27–29 The slope of ⌬V˙ e/ ⌬V˙ co2 ranges from approximately 24 to 34 in normal subjects,19,30 –32 and rises at a progressively steeper rate in cardiac patients as their heart failure grows more severe.30,32,33 A steeper slope of ⌬V˙ e/⌬V˙ co2 is assumed to reflect an increase in the ratio of pulmonary dead space to tidal volume or a decrease in the regulatory set point for Paco2. The increase in the ratio of pulmonary dead space to tidal volume in cardiac patients is probably due to ventilation/perfusion mismatching, ie, reduced or absent perfusion in the wellventilated lung.33,34 Thus, the correlations of cerebral ⌬O2Hb with these indexes strongly suggest that the change in cerebral oxyhemoglobin during exercise is related to cardiopulmonary function during exercise. Mechanisms of the Decrease in Cerebral Oxyhemoglobin During Exercise In the present study, the increase in cerebral oxyhemoglobin during exercise was smaller in the patients with valvular heart disease than in normal CHEST / 125 / 1 / JANUARY, 2004

187

Table 3—Hemodynamic and Respiratory Gas Variables in Patients With Increased O2Hb (⌬O2Hb> 0) and Decreased O2Hb (⌬O2Hb < 0) During Exercise* Variables Male/female gender, No. Age, yr Height, cm Weight, kg Rest Left ventricular ejection fraction, % Heart rate, beats/min Systolic BP, mm Hg Diastolic BP, mm Hg Gas exchange ratio Peto2, mm Hg Petco2, mm Hg Spo2, % Peak exercise Heart rate, beats/min Systolic BP, mm Hg Diastolic BP, mm Hg Gas exchange ratio Peto2, mm Hg Petco2, mm Hg Spo2, % ⌬O2Hb, ␮mol/L ⌬HHb, ␮mol/L ⌬TOI, % Exercise parameters Maximum WR, W Peak V˙ o2, mL/min/kg ⌬V˙ o2/⌬WR, mL/min/W ⌬V˙ e/⌬V˙ co2 GET WR, W GET V˙ o2, mL/min/kg

Patients With ⌬O2Hb ⬎ 0 (n ⫽ 18)

Patients With ⌬O2Hb⬍ 0 (n ⫽ 15)

p Value

11/7 62.6 ⫾ 13.2 162.5 ⫾ 10.2 61.3 ⫾ 10.8

7/8 67.2 ⫾ 10.8 159.2 ⫾ 10.3 53.3 ⫾ 10.7

0.49 0.28 0.37 0.04

62.5 ⫾ 9.0 78.1 ⫾ 12.2 129.7 ⫾ 16.4 68.8 ⫾ 15.5 0.89 ⫾ 0.06 111.9 ⫾ 2.9 33.9 ⫾ 1.8 97.9 ⫾ 1.3

54.3 ⫾ 16.5 81.2 ⫾ 17.7 128.8 ⫾ 23.1 71.5 ⫾ 12.4 0.90 ⫾ 0.04 114.0 ⫾ 3.3 32.2 ⫾ 3.7 97.6 ⫾ 1.9

0.08 0.56 0.89 0.60 0.91 0.07 0.10 0.56

139.4 ⫾ 25.9 199.8 ⫾ 33.7 94.6 ⫾ 21.3 1.11 ⫾ 0.07 111.1 ⫾ 4.9 40.8 ⫾ 4.6 96.9 ⫾ 1.6 3.49 ⫾ 3.57 ⫺ 0.07 ⫾ 1.64 1.71 ⫾ 4.51

120.1 ⫾ 30.3 158.5 ⫾ 26.9 85.1 ⫾ 17.6 1.07 ⫾ 0.09 115.4 ⫾ 6.9 34.8 ⫾ 7.0 95.7 ⫾ 3.0 ⫺ 2.06 ⫾ 1.55 0.84 ⫾ 1.50 ⫺ 1.13 ⫾ 2.68

0.06 0.0006 0.18 0.13 0.04 0.006 0.15 ⬍ 0.0001 0.12 0.04

94.5 ⫾ 28.9 19.6 ⫾ 4.8 9.0 ⫾ 1.9 30.4 ⫾ 4.5 57.2 ⫾ 17.8 14.4 ⫾ 3.4

57.7 ⫾ 18.4 12.6 ⫾ 3.8 5.9 ⫾ 2.2 39.7 ⫾ 11.7 40.5 ⫾ 21.8 10.9 ⫾ 3.4

0.0002 ⬍ 0.0001 0.0001 0.004 0.02 0.007

*Data presented are mean value ⫾ SD unless otherwise indicated. See Table 2 for expansion of abbreviation.

subjects. In fact, the cerebral oxyhemoglobin even decreased during exercise in 15 of the 33 patients, indicating the presence of cerebral hypoperfusion. In a comparison between the patients with decreased oxyhemoglobin (⌬O2Hb ⬍ 0) and increased oxyhemoglobin (⌬O2Hb ⬎ 0), the former had a lower peak V˙ o2, a lower slope of ⌬V˙ o2/⌬WR, and a lower systolic BP at peak exercise. These findings confirm that the decrease in cerebral oxyhemoglobin must be at least partially attributable to the impaired increase in cardiac output during exercise. Another possible factor influencing cerebral oxygenation might be the level of Paco2 during exercise. Cerebral blood flow is known to positively correlate with Paco2: a decrease in Paco2 leads to cerebral hypoperfusion. The patients with ⌬O2Hb ⬍ 0 had significantly lower Petco2 at peak exercise than those with ⌬O2Hb ⬎ 0. In a normal lung, Petco2 is known to exceed Paco2 during exercise and drop slightly lower than Paco2 while at rest.19 Thus, the average Paco2 at peak exercise in the patients with ⌬O2Hb ⬍ 0 might be even lower than 34.8 mm Hg, 188

the value of Petco2 noted at peak exercise in the present study. The decline in Paco2 during exercise in these patients might be attributable to hyperventilation. The patients with ⌬O2Hb ⬍ 0 had lower GET WR than those with ⌬O2Hb ⬎ 0 (40.5 ⫾ 21.8 W vs 57.2 ⫾ 17.8 W), in addition to a higher slope of ⌬V˙ e/⌬V˙ co2. Accordingly, we speculate that these patients soon entered a state of lactic acidosis that elicited the hyperventilation indicated by their higher ⌬V˙ e/⌬V˙ co2 values, and subsequently lowered the level of Paco2.31 There was an overlap in cerebral ⌬O2Hb values between the patients and normal subjects. Moreover, apparent decreases in oxyhemoglobin were noted during exercise in some of the normal subjects (Fig 3). ⌬O2Hb was negatively correlated with age, and our subjects were relatively advanced in years. For these reasons, we speculated that the decrease in oxyhemoglobin during exercise might also be attributable to the subjects’ hidden cerebrovascular disease. This possibility will have to be evaluated in a future study. Clinical Investigations

Methodology of Measuring Tissue Oxygenation by NIRS Transcranial Doppler ultrasound, which provides continuous measurements of blood flow velocity, has been used to evaluate cerebral hemodynamics. However, it has been suggested that the measurement of cerebral blood flow velocity with this technique does not accurately reflect the actual blood flow during dynamic exercise.35,36 The major advantage of NIRS is its potential for noninvasive measurements of tissue oxygenation. Since its invention by Jo¨ bsis4 in 1977, NIRS has been greatly improved and adopted as an established tool for monitoring cell metabolism, cerebral hemodynamics, and oxygen transport to tissue.6,9 NIRS uses nondamaging doses of nearinfrared radiation in the wavelength range from 700 to 1,000 nm.6 Hemoglobin displays oxygen-dependent absorption characteristics in this region, and thus can noninvasively be detected.6 When NIRS is attached at the forehead, the emitted laser light passes through the skull and is dispersed through the brain tissue.7 NIRS attached at the forehead measures brain tissue oxygenation at a depth of approximately 1 cm from the brain surface.8 NIRO-300 also measures a TOI by applying a photon-diffusion theory.6,9 Al-Rawi et al11 reported the usefulness of measuring TOI for the detection of intracranial oxygenation changes. In the present study, the pattern of the change in TOI during exercise was similar to that of oxyhemoglobin, as shown in Figure 1 for representative subjects. However, the difference in ⌬TOI during exercise between the patients with valvular heart disease and normal subjects did not reach a statistical significance. This might be attributable to the characteristic of TOI, which is determined by both oxyhemoglobin and deoxyhemoglobin in the tissue. Study Limitations In evaluating cerebral oxygenation using NIRS attached at the forehead, we have to consider the possibility of contamination from extracranial tissues. In the present study, there was no difference in Spo2 between the patients with ⌬O2Hb ⬍ 0 and those with ⌬O2Hb ⬎ 0 when the measurement was taken at the left earlobe near the NIRS probes. As in the case of the scalp and skull, an earlobe is perfused by an external carotid artery. Hence, we believe that the influence of extracranial information was negligible for the observed difference in ⌬O2Hb between the patients and normal subjects and between the subgroups of the patients. We selected patients with valvular heart disease for this investigation on cerebral oxygenation, as they are likely to be burdened with a valvular stenosis or regurgitation that impairs www.chestjournal.org

the increase in forward cardiac output during exercise. Although the left ventricular ejection fraction in our patients was relatively preserved (58.8% on average), we believe that the present findings are applicable to patients with any type of cardiac disease involving a left ventricular systolic dysfunction. In a future study, however, the factors influencing cerebral oxygenation during exercise, which might be partly related to the etiology of heart disease, will have to be further clarified. A future study will have to be conducted to determine whether the noninvasive measurements of cerebral oxyhemoglobin during exercise can be used for evaluating the presence of cerebrovascular disease. Another issue to determine will be the threshold level of cerebral oxyhemoglobin influencing the brain function. Conclusions The present findings strongly suggest that cerebral oxygenation during exercise is a function of the cardiovascular and pulmonary systems. It was also found that cerebral hypoperfusion arises during exercise in some cardiac patients in whom the cardiac output fails to increase normally. ACKNOWLEDGMENT: We thank Osamu Nagayama, BS, Kaori Inagawa, BS, Tomoko Maeda, BS, Takuro Kubozono, MD, Keiko Oikawa, MD, and Hiroyuki Iinuma, MD, of the Cardiovascular Institute, and Toshimitsu Momose, MD, of the University of Tokyo.

References 1 Zelis R, Sinoway LI, Musch TI, et al. Regional blood flow in congestive heart failure: concept of compensatory mechanisms with short and long time constants. Am J Cardiol 1988; 62:2E– 8E 2 Saxena PR, Schoemaker RG. Organ blood flow protection in hypertension and congestive heart failure. Am J Med 1993; 94 (Suppl 4A):4S–12S 3 Lee CW, Lee JH, Kim JJ, et al. Cerebral metabolic abnormalities in congestive heart failure detected by proton magnetic resonance spectroscopy. J Am Coll Cardiol 1999; 33: 1196 –1202 4 Jo¨ bsis FF. Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 1977; 198:1264 –1267 5 van der Zee P, Cope M, Arridge SR, et al. Experimentally measured optical pathlengths for the adult head, calf and forearm and the head of the newborn infant as a function of inter optode spacing. Adv Exp Med Biol 1992; 316:143–153 6 Matcher SJ, Kirkpatrick P, Nahid K, et al. Absolute quantification methods in tissue near infrared spectroscopy. Proc SPIE 1993; 2389:486 – 495 7 Nollert G, Mo¨ hnle P, Tassani-Prell P, et al. Determinants of cerebral oxygenation during cardiac surgery. Circulation 1995; 92(Suppl):II327–II333 8 Villringer K, Minoshima S, Hock C, et al. Assessment of local brain activation: a simultaneous PET and near-infrared spectroscopy study. Adv Exp Med Biol 1997; 413:149 –153 CHEST / 125 / 1 / JANUARY, 2004

189

9 Suzuki S, Takasaki S, Ozaki T, et al. A tissue oxygenation monitor using NIR spatially resolved spectroscopy. Proc SPIE 1999; 3597:582–592 10 Quaresima V, Sacco S, Totaro R, et al. Noninvasive measurement of cerebral hemoglobin oxygen saturation using two near infrared spectroscopy approaches. J Biomed Opt 2000; 5:201–205 11 Al-Rawi PG, Smielewski P, Kirkpatrick PJ. Evaluation of a near-infrared spectrometer (NIRO 300) for the detection of intracranial oxygenation changes in the adult head. Stroke 2001; 32:2492–2500 12 Lightfoot JT, Tankersley C, Rowe SA, et al. Automated blood pressure measurements during exercise. Med Sci Sports Exerc 1989; 21:698 –707 13 Koike A, Hiroe M, Adachi H, et al. Oxygen uptake kinetics are determined by cardiac function at onset of exercise rather than peak exercise in patients with prior myocardial infarction. Circulation 1994; 90:2324 –2332 14 Koike A, Yajima T, Adachi H, et al. Evaluation of exercise capacity using submaximal exercise at a constant work rate in patients with cardiovascular disease. Circulation 1995; 91: 1719 –1724 15 Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986; 60:2020 –2027 16 Sue DY, Wasserman K, Moricca RB, et al. Metabolic acidosis during exercise in patients with chronic obstructive pulmonary disease. Chest 1988; 94:931–938 17 Koike A, Weiler-Ravell D, McKenzie DK, et al. Evidence that the metabolic acidosis threshold is the anaerobic threshold. J Appl Physiol 1990; 68:2521–2526 18 Koike A, Itoh H, Kato M, et al. Prognostic power of ventilatory responses during submaximal exercise in patients with chronic heart disease. Chest 2002; 121:1581–1588 19 Wasserman K, Hansen JE, Sue DY, et al. Principles of exercise testing and interpretation. Baltimore, MD: Lippincott, Williams & Wilkins, 1999; 10 – 82 20 Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310 21 Weber KT, Kinasewitz GT, Janicki JS, et al. Oxygen utilization and ventilation during exercise in patients with chronic cardiac failure. Circulation 1982; 65:1213–1223 22 Weber KT, Janicki JS. Cardiopulmonary exercise testing for evaluation of chronic cardiac failure. Am J Cardiol 1985; 55:22A–31A

190

23 Itoh H, Koike A, Taniguchi K, et al. Severity and pathophysiology of heart failure on the basis of anaerobic threshold (AT) and related parameters. Jpn Circ J 1989; 53:146 –154 24 Wasserman K, McIlroy MB. Detecting the threshold of anaerobic metabolism in cardiac patients during exercise. Am J Cardiol 1964; 14:844 – 852 25 Wasserman K. New concepts in assessing cardiovascular function. Circulation 1988; 78:1060 –1071 26 Koike A, Itoh H, Taniguchi K, et al. Detecting abnormalities in left ventricular function during exercise by respiratory measurement. Circulation 1989; 80:1737–1746 27 Hansen JE, Sue DY, Oren A, et al. Relation of oxygen uptake to work rate in normal men and men with circulatory disorders. Am J Cardiol 1987; 59:669 – 674 28 Cohen-Solal A, Chabernaud JM, Gourgon R. Comparison of oxygen uptake during bicycle exercise in patients with chronic heart failure and in normal subjects. J Am Coll Cardiol 1990; 16:80 – 85 29 Koike A, Hiroe M, Adachi H, et al. Anaerobic metabolism as an indicator of aerobic function during exercise in cardiac patients. J Am Coll Cardiol 1992; 20:120 –126 30 Metra M, Dei Cas L, Panina G, et al. Exercise hyperventilation in chronic congestive heart failure, and its relation to functional capacity and hemodynamics. Am J Cardiol 1992; 70:622– 628 31 Koike A, Hiroe M, Taniguchi K, et al. Respiratory control during exercise in patients with cardiovascular disease. Am Rev Respir Dis 1993; 147:425– 429 32 Chua TP, Ponikowski P, Harrington D, et al. Clinical correlates and prognostic significance of the ventilatory response to exercise in chronic heart failure. J Am Coll Cardiol 1997; 29:1585–1590 33 Buller NP, Poole-Wilson PA. Mechanism of the increased ventilatory response to exercise in patients with chronic heart failure. Br Heart J 1990; 63:281–283 34 Wasserman K, Zhang YY, Gitt A, et al. Lung function and exercise gas exchange in chronic heart failure. Circulation 1997; 96:2221–2227 35 Madsen PL, Sperling BK, Warming T, et al. Middle cerebral artery blood velocity and cerebral blood flow and O2 uptake during dynamic exercise. J Appl Physiol 1993; 74:245–250 36 Giller CA, Giller AM, Cooper CR, et al. Evaluation of the cerebral hemodynamic response to rhythmic handgrip. J Appl Physiol 2000; 88:2205–2213

Clinical Investigations