| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Charlotte, NC
Dr. Raymond is Director of Occupational and Environmental Medicine, Carolinas Healthcare System, and Professor of Family Medicine, University of North Carolina at Chapel Hill.
Correspondence to: Lawrence W. Raymond, MD, ScM, FCCP, PO Box 32861, Charlotte, NC 28232-2861; e-mail: larryr{at}med unc.edu.
At the end of maximal exercise, heart rate drops exponentially in healthy humans, as vagal tone returns from its profound exertional withdrawal and sympathetic activation wanes.1234 After Bruce protocol exercise, for example, the healthy heart slows by > 10 to 12 beats/min, 60 s after end-exercise (18 beats/min for stress echocardiography).5 A delayed heart rate recovery (HRR) has been associated with increased all-cause mortality in most studies,67891011 but not all.12 Cole et al6 reported that the 6-year mortality rate from all causes was 19% in patients with an HRR of < 13 beats/min, at 1 min after end-exercise (relative risk, 4.0; 95% confidence interval [CI], 3.0 to 5.2). Later, patients studied by Nishime et al8 who had an abnormal HRR experienced a 5-year all-cause mortality rate of 8% (risk ratio, 4.16; 95% CI, 3.33 to 5.19), and a mortality rate of 9% in patients whose abnormal HRR was coupled with chronotropic incompetence. When both of these abnormalities occurred in persons with Duke treadmill scores of 4 or worse, the 5-year all-cause mortality rate was 18%.13 The report of Seshadri et al in this issue of CHEST (see page 1287) raises the possibility that some of this excess mortality is due to underlying pulmonary disease rather than to cardiovascular causes. Their retrospective analysis of 627 treadmill patients identified 229 (36.5%) with an abnormal HRR, while 188 (30%) had an abnormal chronotropic index. These HRR and chronotropic index abnormalities appear to be closely correlated (r = 0.86 for the respective means, as percentages of abnormal values, when analyzed by quartiles of FEV1 percent predicted [Table 2 of Sheshadri et al]). One wonders how many patients had abnormal Duke treadmill scores.1415 Perhaps future publications will let us know whether the prognostic implications of these observations in patients with lung disease are as gloomy as they are in patients suspected of having cardiovascular conditions.
The present study is intriguing for several other reasons. First is the finding that impaired HRR was present across a biological gradient (ie, that the lower the FEV1, the higher the likelihood of abnormal HRR, in both smokers and nonsmokers). The frequency of abnormal chronotropic indexes displayed the same inverse biological gradient as did the abnormal HRR. Second, while the abnormal HRR and chronotropic index were seen in patients with COPD, they were most marked in patients with a restrictive spirometric pattern. Third, the results have neuroanatomic appeal. Many authors attribute an abnormal HRR to a loss of normal parasympathetic function (see below), the efferent limb of which involves cardiac vagal motoneurones, the cell bodies of which reside in both the dorsal motor nucleus of the vagus and the nucleus ambiguus,161718 which are sensitive to both baroreceptor and respiratory activity. Thus, afferent activity from diseased lungs (or cardiovascular structures) may be integrated within these medullary vasomotor and respiratory oscillators, which then can mediate autonomic dysfunction (AD), as manifested by abnormal HRR, chronotropic index, and other more elaborate criteria.192021
How did this study arise? Were the authors merely "panning for gold" in a mother lode of archived treadmill test results? Hardly. Their prior work includes attention to AD as an independent risk factor for cardiovascular and all-cause mortality in patients with diabetes22 and in persons with impaired fasting glucose levels.23 Others24 have since shown an association between higher fasting insulin levels and AD, manifested as impaired HRR, in nondiabetic subjects with normal fasting glucose levels. Since AD also has been reported in COPD patients,19202125 it seems natural that patients who had undergone both exercise testing and spirometry would be an opportune group with which to perform a study like the present one. It was not designed to identify hypoxemia as a necessary factor in the abnormal HRR seen in these patients. Thus, the extent to which hypoxemia was a determinant of this abnormality is unclear. Most studies of AD in COPD patients have involved such (hypoxemic) COPD patients.19202125 Indeed, in some patients the severity of the AD correlated with the degree of hypoxemia.21 However, AD also was identified by Stein et al26 in patients with 
1-antitrypsin deficiency, in whom no hypoxemia had been described (nor was oxygen listed in the table of their therapies). They found increased mean and minimum heart rates, and decreased heart rate variability in the PiZ patients who had COPD, but not in five PiZ patients with normal FEV1. The impaired vagal modulation of heart rate was accentuated with decreasing FEV1. Commenting on the opposite findings of increased vagotonia reported by Volterrani et al,27 Stein et al26 wondered whether the difference might be due to reflex changes from the acute withdrawal of therapeutic agents in the latter study, or to a lesser severity of the COPD (mean [± SD] FEV1, 29.6 ± 10.6% predicted26 vs 52 ± 8.3% predicted27). It is also noteworthy that patients in the latter study had only mild hypoxemia (mean PO2, 71 ± 14 mm Hg; mean PCO2, 40 ± 10 mm Hg). Stewart et al21 reported AD in COPD patients who had more severe obstruction, with mean FEV1 values of < 42% predicted (range, 27 to 41.6% predicted), with or without hypoxemia or edema. All eight patients with both of the latter abnormalities also had AD, whereas only three of eight nonedematous, hypoxemic patients did. Perhaps, with regard to AD, hypoxemia serves as an indicator of the severity of distorted lung mechanics in patients with COPD and restrictive disease.
Not all studies of the mechanism of HRR in healthy subjects have come to the same conclusion. Savin et al28 found that HRR in healthy 31-year-old men was exponential under control conditions, as well as during either sympathetic or parasympathetic (or double) blockade. This led them to state that "the postexercise exponential decline of heart rate is not dependent on autonomic control but rather is an intrinsic property of the intact circulation" (perhaps related to reduced venous return, atrial stretch receptor activity, or circulating catecholamine levels). Their schematic of autonomic changes after peak exercise, however, portrayed sympathetic withdrawal as playing a dominant role and parasympathetic reactivation a minimal one in cardiodeceleration immediately after peak exercise. This graphic display is a mirror image of the one used by Robinson et al1 to account for cardiac acceleration during increasing levels of exertion. Rosenwinkel et al29 cast doubt on the sympathetic withdrawal construct depicted by Savin et al,28 pointing out that circulating norepinephrine levels do not drop with the end of vigorous exertion. More recently, Imai et al3 found that early HRR was unaffected by ß-blockade, but was abolished by atropine. They inferred that excessive sympathetic nervous system activity does not play a dominant role in immediate HRR. Some of the apparent discord in the above viewpoints is explained by the different time frames implied by the term "postexercise," Savin et al28 having used 5-min analyses, while Imai et al3 presented time constants for 30 s (and 120 s) postexercise, these being the most relevant to the present discussion.
What is the mechanism of the impaired HRR in patients with abnormal spirometry? It could reflect a lack of efferent vagal discharge or excessive sympathoadrenomedullary activity, a combination of the two, or nonautonomic factors. In the last category, an increase in central body temperature due to brief treadmill exercise is most unlikely as a possible factor in driving the cardiac rate at end-exercise.30 However, one is reminded of the prolonged half-time for the recovery of gas-exchange variables in COPD patients, due to impaired alveolar ventilation.31 With regard to sympathetic influences, patients with chronic heart failure were excluded by Seshadri et al, hence that source of increased sympathetic activity did not play a role, but other authors32333435 have reported evidence of excessive resting sympathetic activity in patients with COPD. Corresponding information for the period during exercise appears to be limited to the recent study of Bartels et al.36 They used heart rate variability to assess cardiac autonomic modulation in 53 COPD patients whose mean FEV1 was 35% predicted, and who exercised maximally but reached only 70% of their age-predicted maximal heart rate. These patients showed evidence of increased parasympathetic modulation during exercise, with a decrease in the balance of sympathetic to parasympathetic influences. As expected, the opposite result was found in age-matched control subjects. No postexercise observations were included, however. (One hesitates to extrapolate the above findings to HRR kinetics, since parasympathetic influences are often quite transitory due to the rapid hydrolysis of acetylcholine in the sinus node.) No other indexes of cardiac autonomic function were measured by Bartels et al,36 nor have corresponding observations been published for restrictive lung disease. More direct observations in patients with lung disease are needed to determine the mechanisms underlying the present HRR findings.
In addition to the issues addressed above, a number of research opportunities are suggested by the findings of Seshadri et al. Do agents that reduce sympathetic tone such as angiotensin-converting enzyme inhibitors or ß-blockers improve HRR in patients with COPD who receive these agents, even though HRR did not improve in patients with congestive heart failure who received carvedilol or metoprolol?37 Might resting heart rate identify the patients who are most likely to benefit? One also wonders how common AD is in patients with restrictive lung disease, whether hypoxemia is important in its genesis in such patients, and whether a reversal of hypoxemia (either in the short term or the long term) also reverses AD in patients with hypoxemic lung disease. Finally, does pulmonary rehabilitation38 improve HRR, as it does in cardiac patients?39404142 In all of these questions, clinicians will look for answers concerning quality of life and survival.
How else will clinicians find the present results of Seshadri et al interesting and useful? For some, the findings may sound a note of caution concerning the use of short-acting anticholinergic agents (eg, ipratropium and oxitropium) and long-acting anticholinergic agents (eg, tiotropium). While these agents have much to recommend them43 (beneficial effects include improvements in exercise tolerance, quality of life, and spirometry, and reduced exacerbations in COPD patients), there is no published evidence that they improve survival. And one study of 2,382 hospitalized patients found that those treated with ipratropium experienced higher rates of death from asthma (odds ratio [OR], 4.04; 95% CI, 1.47 to 11.13), COPD (OR, 7.75; 95% CI, 2.21 to 27.14), and cardiovascular diseases (OR, 3.55; 95% CI, 1.05 to 11.94) over the 3 years after hospital discharge.44 Autopsy results from seven patients with asthma or COPD who had received ipratropium showed that all had mucus plugging of airways, which was widespread in five. In contrast, of eight similar patients who did not receive ipratropium, only two had extensive mucus plugging. On a more positive note, the present findings will encourage some clinicians to make greater use of graded exercise testing in patients with COPD or restrictive disease to gain insight into their prognosis.45 Such testing, with particular attention to HRR and other measures of vagal tone, might help to tailor pulmonary (and cardiovascular) interventions such as the use of pulmonary rehabilitation and optimization of medications.42 The results presented by Seshadri et al could provide a new (as yet unproven) incentive for the greater use of pulmonary rehabilitation3843 to improve vagal tone. Such was the finding of La Rovere et al.46 In their study of 95 survivors of myocardial infarction, half were randomized to 4 weeks of endurance training, which improved baroreceptor sensitivity by 26%. Those patients who demonstrated this improvement after such training had a lower 10-year cardiac mortality rate (p < 0.04) compared to the nonresponders.
As with many careful studies, the present one raises a number of important questions borne of heart-lung interactions in the laboratory setting. One hopes the answers to such questions will inform us of treatment steps that will improve the functional pas de deux of both organs in those other settings where our patients work, play, and repose.
References
This article has been cited by other articles:
|
|
 |
|
  D. R. Ricardo, M. B. de Almeida, B. A. Franklin, and C. G. S. Araujo Initial and Final Exercise Heart Rate Transients: Influence of Gender, Aerobic Fitness, and Clinical Status Chest, January 1, 2005; 127(1): 318 - 327. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
