INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WITH INCREASING AGE, ventilatory and circulatory responses to exercise should become less capable. For example, Shephard (25) reviewed that a longer time is required for ventilation, heart rate (HR), blood pressure (BP), and oxygen consumption (O₂) to attain equilibrium at any given workload with aging. Recently, some investigators maintained that the ventilatory and O₂ kinetics (time constant) during phase II, the exponential increase in these parameters from ~20 s after the start of exercise to the steady state, became longer in the elderly than in the young (1, 5). Nevertheless, little is known about the effect of aging on the "phase I" ventilatory response, which was the initial response to exercise within 20 s after exercise onset. Additionally, few investigators have examined HR and BP responses to exercise in the elderly immediately after the start of exercise. It appears that the initial ventilatory and circulatory responses to exercise should be very important in that these responses would affect the next phase of exercise. It is often the case that people become less active with advancing age; however, they sometimes have opportunities to move quickly for a short time (within a 20-s period), such as when crossing a street. If ventilation and HR do not increase during such brief exercise, then heavy strain may occur after the cessation of the movement. Accordingly, inquiring into the ventilatory and circulatory responses to brief and light exercise in the elderly is required.

A firm definition of phase I has not been determined; therefore, the methods for analyzing phase I and the interpretation of the results have been conflicting. As for the problem of time (the length of phase I), some investigators (14, 31) who were interested in O₂ kinetics assumed that phase I lasted until the end-tidal CO₂ partial pressure (PET_CO2) began to increase, and others (4, 26) considered that phase I lasted until 15 or 20 s after the onset of exercise. As for the extent of the phase I response, Sietsema et al. (26) defined the phase I value as the plateau value or the value attained 20 s after the start of exercise, whereas we previously determined that the mean of the first two breaths was the extent of phase I (22). On the contrary, when we investigate cardiorespiratory responses to exercise with a step increase in workload, we should take into account both the extent of change of the response from rest to a steady state (gain) and the increasing rate of the response [response time (RT), or time constant]. A considerable number of investigators have previously observed the phase I ventilatory response only from the point of the gain. However, it is possible that the increasing rate of ventilation and circulation within 20 s of the start of exercise may be slowed in elderly people, similar to the phase II response. Despite this, few investigators have studied the speed of the response to brief exercise in the elderly.

Ventilatory and circulatory responses to exercise have been thought to be caused by central command and peripheral neural reflex mechanisms (16). To examine the effects of the peripheral neural reflex mechanism, passive movements are widely employed because the effect of central command would be reduced to the minimum (9, 23). It is possible that passive movements could lead to defining the factors that cause the different cardiorespiratory responses to exercise that have been observed in the elderly.

The purpose of the present study was, therefore, to clarify the characteristics of ventilatory and circulatory responses to brief and light voluntary exercise in the elderly by comparing these responses to those observed in the young and additionally, if any differences exist, to examine the mechanism that causes the different responses between the elderly and the young, especially by means of passive movements. We hypothesized that ventilatory and circulatory responses at the onset of dynamic exercise would be attenuated or slowed in the elderly.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Thirteen healthy men {age: 61-73 yr [mean: 66.8 ± 4.3 (SD) yr], height: 163.3 ± 4.0 cm, body mass: 58.9 ± 3.6 kg} volunteered to participate in the present study and were designated the elderly group (ELD). Another group of 13 healthy younger [YNG; age: 18-28 yr (22.9 ± 3.0 yr)] men served as the control group. Their mean height and body mass were 173.6 ± 6.0 cm and 66.2 ± 8.1 kg, respectively. These parameters showed significant differences between ELD and YNG. The proportion of body fat, obtained by a bioelectrical impedance device (model TBF-101, Tanita), was almost equal between ELD and YNG (16.5 ± 4.82 and 17.0 ± 4.26%, respectively), whereas fat-free mass in ELD (49.1 ± 3.72 kg) was significantly lower than that in YNG (54.7 ± 4.78 kg). The subjects were strictly screened before the experiment, and those that had a history of cardiopulmonary diseases, received medications that seriously affect circulatory and respiratory responses, were obese, and/or were current smokers were excluded. All subjects in this study were leading healthy and active lives. The subjects were informed of the experimental protocol and the possible risks involved in this study before giving written consent. Approval for the study was given by the ethics committee of the Research Center of Health, Physical Fitness and Sports at Nagoya University.

Exercise. During the experiment, subjects sat with their backs against an experimental chair with an electrogoniometer attached to the right knee joint. Two weight belts, each equivalent to ~2.5% of the subject's body mass, were bound around both ankles. Subjects extended and relaxed both knee joints in an alternating pattern from ~110 to 20° in the anatomically flexed position. This voluntary exercise (VOL) began with the right limb, was signaled by the experimenter's voice just before the start of the inspiratory periods, and continued for ~20 s at the rate of one leg extension and relaxation per second for each leg. The starting point was determined by the flow curve monitored on the oscilloscope. Subjects were instructed to keep the upper body as still as possible. Passive movements (PAS) were achieved by having the experimenter alternately pull ropes that were connected to the subject's ankles, at the same pace, for the same length of time, and in the same posture as in VOL. All subjects were directed to relax and not resist the motion, and care was taken to immobilize the body as much as possible to avoid motion artifacts and muscle contractions.

Measurements. Tidal volume (VT), minute inspiratory ventilation (I), expiratory time, and inspiratory time (TI) were measured with the breath-by-breath technique with a hot-wire flowmeter (model RF-H, Minato Ika-gaku) attached to a respiratory face mask. The dead space of the mask was <100 ml. Respiratory frequency (f) was calculated from total respiratory time. Flow curve was monitored on a storage oscilloscope to ascertain the respiratory timing. End-tidal O₂ partial pressure (PET_O2) and PET_CO2 were obtained by a rapid O₂ and CO₂ analyzer (model MG-360, Minato Ika-gaku). To determine the circulatory response, HR was calculated beat to beat from R spike using an electrocardiogram through a bioamplifier (model AB-621G, Nihon Kohden). BP was also measured noninvasively beat to beat by using the arterial tonometric method (model JENTOW EX, Colin). In short, the principle of arterial tonometry is that BP at the radial artery can be obtained by measuring the reaction forces produced by flattening the radial. Recently, this method has become preferred over the conventional finger photoplethysmographic method (Finapres), and the accuracy and reliability of BP measured by tonometry can be confirmed when compared with the intra-arterial method (11, 24). A tonometric sensor was attached to the left wrist and placed on a padded platform at the level of the heart. The oscillometric calibrations were carried out for accurate tonometric measurement before, and sometimes during, the main experiment. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated by the tonometric signal, and mean blood pressure (MBP) was obtained from SBP and DBP. All signals were converted from analog to digital with an analog-to-digital converter board (model ADX-98H, Canopus) and a personal computer (model PC-9821Xa, NEC) at a sampling frequency of 100 Hz for 90 s, beginning and ending ~30 s before and after exercise. The digital data were stored in a hard disk unit and analyzed afterward by means of our own software on a personal computer.

Protocol. A preparatory experiment was carried out to familiarize the subjects with the experimental procedure and to check for abnormality of cardiorespiratory functions. The subjects practiced several movements so as not to respond suddenly to the motions and the experimenter's voice. The main experiment was performed on a different day at least 2 h after the last meal. Each mode of brief exercise or movement was repeated for six bouts at ~3-min intervals, in a random order, for each subject to avoid order effects. Thereafter, to compare the differences of the extent of the workload between the two groups, the same exercises and movements were continued for 3 min one time, and the mean of the last 30-s value (STD; 3-min steady state value) of the parameters was analyzed. All parameters were measured in the same way as the brief exercise.

Data analysis and statistics. Breath-by-breath and beat-to-beat data were aligned with the onset of exercise and linearly interpolated between each breath or beat to yield a data point at each 1-s interval. Ensemble averaging was done across all six repetitions. To confirm whether the parameters had actually changed as a result of exercise compared with at rest, ANOVA and Dunnett's t-test (post hoc) were carried out. Resting value (Rest) was the average of 20 s before brief exercise or movement. To compare differences between ELD and YNG, a Komogorov-Smirnov test was conducted to test for differences, and, if the difference was regular, then a two-sample t-test was done. If the difference was not regular, a nonparametric test (Mann-Whitney's U-test) was performed. The comparison between ELD and YNG consisted of three analyses: 1) comparing Rest, STD, and the relative change obtained by dividing STD by Rest; 2) comparing the relative change of I and HR during brief exercise normalized by Rest (0%) and STD (100%) in VOL and by Rest (0%) only in PAS, and comparing the relative change of MBP during brief exercise normalized by Rest (100%) in VOL and PAS; and 3) comparing the kinetics of I and HR responses in detail. For the third analysis, we regarded the value of the relative change at 15 s after the start of exercise and movement as a gain (Gain), and the time for reaching one-half of the Gain as RT. For MBP response, we considered the minimal value of the relative change during exercise and movement as Gain and the time for reaching the minimal value as RT. These statistical analyses were calculated by computer software (SPSS 8.0, SPSS), and the level of significance was set at P = 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Figure 1 shows the absolute means ± SE of ventilatory responses 20 s before, during, and after VOL and PAS for ELD and YNG. I in ELD increased slightly and gradually at the onset of VOL and reached a plateau within ~15 s. On the contrary, I in YNG increased promptly and extensively at the onset of exercise, stopped increasing at ~5 s, and remained constant until the end of exercise. A similar tendency was observed in PAS. A significant increase (P < 0.05) of I during exercise compared with Rest was detected in VOL and PAS from immediately after the start of exercise. Respiratory frequency increased quickly and greatly at the onset of exercise in both ELD and YNG. A significant increase (P < 0.05) of f during exercise compared with Rest was observed from the onset of exercise without delay. VT in ELD slightly decreased temporarily after the onset of exercise and returned to the resting level at ~10 s and thereafter gradually increased, whereas VT in YNG gradually increased throughout the exercise. A similar tendency was found in PAS. VT/TI showed the same change as I. PET_CO2 in ELD showed little change from Rest to 15 s after the onset of VOL (35.7 ± 3.7 vs. 35.6 ± 4.3 Torr) but began to increase gradually but slightly over the last 5 s (36.3 ± 3.4 Torr at 20 s). PET_CO2 in YNG decreased from Rest (36.8 ± 3.6 Torr) at ~5 s after VOL began (35.7 ± 3.8 Torr) and returned to the resting value at ~15 s (36.7 ± 3.9 Torr). PET_CO2 during PAS in both groups showed little change.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Ventilatory responses before, during, and after voluntary exercise (VOL; left) and passive movements (PAS; right) for 20 s in elderly (ELD) and young (YNG) men (n = 13 for both groups). I, minute inspiratory ventilation; VT, tidal volume; f, respiratory frequency. Values are means ± SE. Time 0 indicates exercise onset.

Figure 2 represents the cardiovascular responses in ELD and YNG. At rest, HR in ELD was lower than that seen in YNG, but there was no significant difference (Table 1). At the start of VOL, HR in ELD increased in an exponential fashion until ~10 s and thereafter showed a plateau. On the contrary, HR in YNG increased rapidly at the onset of VOL, increasing progressively until ~15 s and then slightly decreasing in an overshoot fashion. HR showed a significant increase (P < 0.05) during VOL compared with Rest from the beginning of exercise in both groups. HR during PAS demonstrated a similar but smaller change than VOL. HR during PAS was significantly (P < 0.05) larger than at Rest from ~3 s in both groups. After the exercise, HR in ELD decreased slower than that in YNG. Resting MBP, SBP, and DBP in ELD were significantly (P < 0.05) larger than those in YNG, but were within the normal ranges for each group. During VOL, these tendencies were retained. MBP, SBP, and DBP in both ELD and YNG began to decrease ~5 s after VOL began and reached a bottom value at ~10 s. After 10 s, all BP measurements increased gradually until the end of exercise. After exercise, they decreased and showed undershoot aspects before returning to the resting level. MBP, SBP, and DBP during PAS demonstrated similar, but smaller, changes.

View larger version (44K): [in this window] [in a new window]   Fig. 2.   Circulatory responses before, during, and after VOL (left) and PAS (right) for 20 s in ELD and YNG. HR, heart rate; MBP, mean blood pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure. Values are means ± SE.

View this table: [in this window] [in a new window]   Table 1.   Average of I, HR, and MBP at rest during brief exercise and at steady state during 3-min continuous exercise

Table 1 shows the means ± SD of I, HR, and MBP at Rest, during STD, and the relative change (STD/Rest). Although a few parameters of the absolute value showed significant difference between ELD and YNG, there was no significant difference in the relative change of these parameters.

Figure 3 indicates the relative changes of I and HR which were normalized by Rest (0%) and STD (100%) in VOL or by Rest (0%) only in PAS. Relative change of I in ELD was significantly (P < 0.05) lower than that in YNG at the first half of exercise during both VOL and PAS. On the contrary, relative change of HR in ELD during both VOL and PAS was significantly (P < 0.05) lower than that in YNG throughout the exercise.

View larger version (40K): [in this window] [in a new window]   Fig. 3.   Relative changes of I and HR during VOL (A and B, respectively) and I and HR during PAS (C and D, respectively) for 20 s in ELD and YNG. Relative change was obtained by normalizing by resting value during brief exercise (0%) and steady state value during 3-min continuous exercise (100%) in VOL and by resting value (0%) only in PAS. Values are means ± SE. * Significant difference between ELD and YNG at each time point (P < 0.05).

Figure 4 represents the kinetics of I, HR, and MBP responses. The Gain (15-s value) of I in ELD during VOL and PAS (41.3 ± 20.8 and 29.0 ± 21.7%, respectively) was slightly lower than that in YNG (52.8 ± 12.3% VOL and 43.2 ± 19.6% PAS), but there were no significant differences. Gain of HR in ELD during VOL and PAS (57.0 ± 19.5 and 5.6 ± 3.5%, respectively) was significantly (P < 0.05) lower than Gain of HR in YNG (107.1 ± 48.9% VOL, 14.1 ± 9.6% PAS). RT of I in ELD during VOL (4.6 ± 4.8 s) and PAS (5.7 ± 5.6 s) was significantly larger than RT in YNG (1.7 ± 1.4 s VOL and 1.2 ± 0.5 s PAS), although there were no significant differences in RT of HR between both groups. Gain of MBP, normalized by Rest (100%), in ELD during VOL (97.1 ± 3.6%) tended to be higher than that in YNG (94.6 ± 4.9%), and Gain of MBP in ELD during PAS (98.1 ± 2.7%) was significantly (P < 0.05) larger than that in YNG (95.1 ± 4.5%), indicating that ELD showed a smaller change than YNG. There were no significant differences in RT of MBP during VOL and PAS.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Comparison of kinetics of I, HR, and MBP during VOL (A and C) and PAS (B and D) for 20 s in ELD and YNG. Gain is calculated as the normalized relative change value at 15 s for I and HR or the minimum value for MBP during exercise or movement. RT, response time (when I and HR reach one-half of Gain or when MBP reaches minimum value). Values are means ± SE.* Significant difference between ELD and YNG (P < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we attempted to elucidate the ventilatory and circulatory responses at the onset of dynamic exercise in the elderly. The results are as follows. 1) Whereas I increased to a similar level at the latter part of exercise for 20 s in both ELD and YNG, the RT of I in ELD was larger than that in YNG. 2) RT of HR was similar in both groups, but Gain was attenuated in ELD. 3) MBP decreased temporarily at the start of VOL in both groups, although the extent of decline was reduced in ELD. PAS produced a similar change. Consequently, the elderly should have slowed ventilatory response and attenuated circulatory response at the start of exercise.

Methodology. When comparing cardiorespiratory responses among subjects or different groups, the setting of the workload should be a matter of serious debate. Unfortunately, we did not measure maximal muscular strength of leg extension and maximal O₂. However, the workload (2 weight belts, each equaling 2.5% of the subject's body mass) per unit of fat-free mass, which is assumed to be well correlated with the muscle mass, did not show any significant difference between ELD and YNG (2.89 ± 0.17 and 2.97 ± 0.16%, respectively); thus the exerted force, i.e., the load to thigh muscles, would be relatively equal between the two groups. Judging from the STD exercise exerted in the same way as the brief exercise, there was no significant difference in relative changes of I, HR, and MBP between ELD and YNG; therefore, the relative load to the respiratory and circulatory systems, compared with the resting values, would be almost identical between both groups. On the contrary, the relative demands estimated from peak value during maximal exercise may be greater for ELD because it is well accepted that maximal I, HR, and O₂ at maximal exercise decrease with aging (25). However, this should not weaken our findings, because the ELD did show slowed ventilatory and attenuated circulatory responses; nevertheless, the cardiorespiratory system must operate closer to the maximum in ELD than in YNG. Considering these points, it may safely be assumed that the workload used in the present study would not be a serious problem.

Ventilatory response. In the present study, there was no significant difference in Gain of I during exercise between ELD and YNG. This result indicates that the extent of I response at the start of exercise should be almost equal between the elderly and young. However, I in ELD showed a gradual increase and a longer RT than that of YNG. These observations provide strong evidence that the elderly should have a slower ventilatory response at the onset of exercise. This slowed I response was caused by the attenuated VT response in ELD, especially for the first 10 s, despite the rapid increase in f in both groups. In addition, VT/TI, which was assumed to be the index of the ventilatory drive (17), slowed more in ELD than in YNG and may indicate that ventilation in the elderly would be driven slowly at the onset of exercise.

HR response. In the present study, Gain in HR was lower in ELD than that in YNG, although RT showed no significant difference between them. This was different from the ventilatory response. With regard to the input side of the circulatory center, it is assumed that the rapid increase in HR at the onset of exercise is derived from the central command and peripheral neural reflex redundantly, similar to ventilation (13, 18), allowing neural inputs to the circulatory center to affect the response. On the contrary, the output mechanism is quite different between respiration and circulation. Many previous investigators (3, 19) have revealed that a rapid increase in HR at the start of exercise is caused by a rapid withdrawal of the parasympathetic nervous system. As we have already mentioned, the activity of the parasympathetic nerves in the elderly is already reduced at rest and is not further inhibited by the onset of exercise; thus HR in the elderly is only able to be increased slightly (30). Furthermore, it is supposed that the functions of the heart as an effective organ become attenuated with increasing age. For example, myocardial contraction time is prolonged, and the filling rate of the left ventricle is reduced in healthy older individuals (12).

BP response. MBP temporarily and significantly decreased ~10 s after the onset of VOL compared with at rest and thereafter increased in both groups. The relative drop of MBP in ELD was nonsignificant but rather small compared with that in YNG during VOL and significantly smaller during PAS. Previous investigators (3, 29, 32) reported the transient MBP fall at the onset of dynamic exercise in normal adult subjects and suggested that the causes were due to decreases in pressure reflex and total peripheral resistance. Moreover, peripheral mechanoreceptors may have some influence on the response because a similar MBP drop occurred during PAS in the present study, and Nóbrega et al. (23) suggested that mechanoreceptors activated by passive movements could cause the pressor response. The explanations for the attenuated MBP drop in ELD were attributed to the following: 1) lowered pressure reflex in the elderly, 2) difficulty in decreasing peripheral resistance because of reduced elasticity of blood vessels in the elderly (12, 25), and 3) attenuated neural inputs to the circulatory center, such as mechanoreceptors. More detailed investigation into BP response is necessary due to some of the limitations of our study; however, the initial fall of MBP in this study was not very important because the relative change of BP at the onset of exercise was much smaller than that of HR and ventilation.