Author + information
- Received April 26, 1996
- Revision received July 12, 1996
- Accepted September 20, 1996
- Published online January 1, 1997.
- Tuan Peng Chua, BSc, MD, MRCPA,*,
- Derek Harrington, BSc, MRCPA,
- Piotr Ponikowski, MDA,
- Katharine Webb-Peploe, MA, MRCPA,
- Philip A Poole-Wilson, MD, FRCP, FESC, FACCA and
- Andrew J.S Coats, DM, FRCP, FESC, FACCA
- ↵*Dr. Tuan Peng Chua, Department of Cardiac Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY, United Kingdom.
Objectives. We sought to test the hypothesis that suppression of chemosensitivity (respiratory response to arterial blood gases) with dihydrocodeine may improve dyspnea and exercise tolerance in patients with chronic heart failure.
Background. Exertional dyspnea is a common limiting symptom in patients with chronic heart failure. The mechanisms underlying this symptom are not fully understood but may be related to increased ventilation caused, in part, by the augmentation of chemosensitivity. Suppression of chemosensitivity with mild opiates may thus improve this symptom as well as exercise tolerance.
Methods. Twelve men with chronic heart failure (mean [± SE] age 65.5 ± 1.5 years, range 58 to 75; left ventricular ejection fraction 21.3 ± 3.0%, range 8 to 39) received placebo or dihydrocodeine (1 mg/kg body weight) on two separate days in a randomized, double-blind design. One hour later, hypoxic and hypercapnic chemosensitivities were assessed using the transient inhalations of pure nitrogen and the rebreathing of 7% carbon dioxide in 93% oxygen, followed by treadmill cardiopulmonary exercise testing. The symptoms of dyspnea and fatigue during the exercise test were assessed using a modified Borg scale from 0 to 10.
Results. There was a significant fall in hypoxic and hypercapnic chemosensitivities with dihydrocodeine administration compared with placebo (0.447 ± 0.096 vs. 0.746 ± 0.104 liter/min per percent arterial oxygen saturation, p = 0.005; 2.480 ± 0.234 vs. 2.966 ± 0.283 liter/min per mm Hg, p = 0.01, respectively). Exercise duration was prolonged from 455 ± 27 s on placebo to 512 ± 27 s (p = 0.001) with dihydrocodeine, and peak oxygen consumption increased from 18.0 ± 0.6 to 19.7 ± 0.6 ml/kg per min (p = 0.002). The ventilatory response to exercise, characterized by the regression slope relating minute ventilation to carbon dioxide output, decreased from 34.19 ± 2.35 to 30.85 ± 1.91 (p = 0.01). With dihydrocodeine administration, the change in the modified Borg score for dyspnea was −0.80 (p = 0.003) at 6 min and −0.33 (p = 0.52) at peak exercise, whereas that for fatigue did not change significantly. Arterial oxygen saturation was maintained during exercise despite dihydrocodeine administration (99.3% at rest vs. 98.9% at peak exercise, p = 0.21).
Conclusions. Augmented chemosensitivity is important in the pathophysiology of chronic heart failure. Its suppression with dihydrocodeine was associated with a reduction of exercise ventilation, an improvement in exercise tolerance and a decrease in breathlessness. Pharmacologic modulation of chemosensitivity may benefit patients with chronic heart failure and merits further investigation.
(J Am Coll Cardiol 1997;29:147–52)>
Chronic heart failure causes profound physiologic alterations in the body, including changes in neurohormonal, skeletal muscle and respiratory function. Although the survival of patients with chronic heart failure has been improved with neurohormonal blocking agents such as angiotensin-converting enzyme inhibitors ([1–4]) and beta-blockers (), the symptoms remain debilitating. Patients are often limited by exertional dyspnea. The origin of dyspnea is multifactorial ([6, 7]), but the increased ventilation seen during exercise in these patients may play a role (). Other than increased exercise ventilation, patients with chronic heart failure also have an abnormal breathing pattern such that at a given minute ventilation, respiratory rate is increased while the changes in tidal volume are less significant compared with normal subjects ([9, 10]). The onset of dyspnea during exercise may indeed be related to the disproportionate increase in respiratory rate relative to tidal volume during exercise (). We have recently demonstrated that the respiratory response to arterial blood gases, or chemosensitivity, in patients with chronic heart failure is augmented and correlated significantly with the ventilatory response to exercise (). This suggests that upregulated arterial chemoreceptors may play a role in mediating the increased exercise ventilation in these patients and contribute, in part, to their symptoms.
Opiate drugs have a respiratory depressant effect and are used to relieve breathlessness—morphine being used in the treatment of acute pulmonary edema and dihydrocodeine in the palliation of breathlessness in chronic obstructive pulmonary disease () and malignancies (). Accordingly, dihydrocodeine has also been shown to increase the exercise tolerance of patients with chronic obstructive pulmonary disease (). Respiratory depression may be due to the reduced responsiveness of chemoreceptors to arterial blood gases, as was demonstrated with the short-term administration of morphine (). Thus far, the effects of opiates on the exertional dyspnea and exercise tolerance in patients with chronic heart failure have not been studied. We have therefore assessed the effects of dihydrocodeine on chemosensitivity and the exercise tolerance in these patients, on the premise that the pharmacologic suppression of an augmented chemosensitivity may confirm its pathophysiologic importance in chronic heart failure and that it may also produce symptomatic improvement, just as neurohormonal blocking agents benefit patients with chronic heart failure.
1.1 Patient characteristics and intervention.
Twelve men with stable chronic heart failure of New York Heart Association functional class II and III were studied (). The mean (±SE) age was 65.5 ± 1.5 years (range 58 to 75), and all patients had a multigated acquisition radionuclide left ventricular ejection fraction <40% (mean 21.3 ± 3.0%, range 8% to 39%). The etiology of chronic heart failure is shown in Table 1. All patients were taking diuretics and angiotensin-converting enzyme inhibitors and were nonedematous. None of them had chest pain or inducible ischemia during previous exercise testing or a history of pulmonary disease. Hypoxic and hypercapnic chemosensitivities were assessed in these patients 1 h after the patients received placebo or dihydrocodeine (1 mg/kg body weight) in a randomized, double-blind design on two separate days, followed by treadmill exercise testing on each occasion. The placebo and dihydrocodeine were given in the form of a drink made up to 200 ml with bitter lemon, prepared by the Department of Pharmacy, Royal Brompton Hospital. The study was approved by the local ethics committee and all patients gave written informed consent.
1.2 Hypoxic chemosensitivity assessment.
Hypoxic chemosensitivity was assessed using the transient hypoxic ventilatory response method ([11, 16]). This method was chosen for practical and safety reasons because patients were not subjected to prolonged episodes of hypoxia. Briefly, after a period of quiet breathing in room air, the patients were given transient inhalations of pure nitrogen, without being aware of the timing of these inhalations, for two to eight breaths. This was repeated 10 times to provide a wide range of arterial oxygen saturations from 75% to 100%, with appropriate intervals of air breathing between exposures to allow arterial oxygen saturation and end-tidal carbon dioxide concentration to return to baseline. The maximal minute ventilation after each period of nitrogen inhalation was obtained by averaging the largest two consecutive breaths using breath by breath analysis with a heated pneumotachograph (Amis 2000, Innovision, Odense, Denmark), calibrated before each test. Continuous monitoring of oxygen and carbon dioxide was done at the mouth by respiratory mass spectrometry (Amis 2000, Innovision, Odense, Denmark), also calibrated before each test. Arterial oxygen saturation was measured using a pulse oximeter (model N-200E, Nellcor) set at the fast mode with a response time of 2 to 3 s and a lightweight probe clipped gently on the patient’s right earlobe. Minute ventilation was plotted against the lowest arterial oxygen saturation reached for each period of nitrogen inhalation. The hypoxic chemosensitivity was obtained as the slope of the regression line relating minute ventilation to arterial oxygen saturation and expressed in terms of liters per minute per percent oxygen saturation (liter/min per %Sao2).
1.3 Hypercapnic chemosensitivity assessment.
Hypercapnic chemosensitivity was assessed using the rebreathing of 7% carbon dioxide in 93% oxygen through a 6-liter bag for 4 min ([11, 17]). The test was stopped sooner if patients were too breathless to continue or if end-tidal carbon dioxide concentration exceeded 10%. As before, minute ventilation was measured breath by breath using a heated pneumotachograph, and continuous monitoring of oxygen and carbon dioxide was done at the mouth by respiratory mass spectrometry. Hypercapnic chemosensitivity was obtained as the slope of the regression line relating minute ventilation to end-tidal carbon dioxide concentration and expressed in terms of liters per minute per millimeters of mercury (liters/min per mm Hg).
1.4 Cardiopulmonary exercise testing.
Cardiopulmonary exercise testing was performed using a modified Bruce protocol (), beginning with “stage 0” at 1.0 mph and 5% gradient. Respiratory gas exchange analysis (oxygen consumption, carbon dioxide output and minute ventilation) was carried out breath by breath by means of respiratory mass spectrometry and a heated pneumotachograph using the same system as previously described, calibrated before each exercise testing (). Heart rate, assessed electrocardiographically, and blood pressure, measured manually using a mercury sphygmomanometer, were recorded before exercise, at the end of each 3-min exercise stage and at peak exercise. Arterial oxygen saturation was measured using a pulse oximeter (Model N-200E, Nellcor) with a probe placed on the right supraorbital artery. At the end of each exercise stage and at peak exercise, the patient was also asked to score the level of dyspnea and fatigue using a modified Borg scale (). The scale rates the level of perceived symptoms from 0 (none) to 10 (maximum). Patients were also asked the major symptom that stopped them from continuing after each exercise test.
1.5 Statistical analysis.
Statistical analysis was performed using the paired Student ttest or repeated measures analysis of variance and Scheffe’s F multiple comparisons test, as appropriate. A value p < 0.05 was considered significant. The results are expressed as mean value ± SE.
All patients completed the study. One patient felt nauseated after dihydrocodeine, but this manifested after he completed the exercise testing.
2.1 Chemosensitivity and cardiopulmonary exercise testing.
Table 2shows the significant reduction in hypoxic and hypercapnic chemosensitivities after the administration of dihydrocodeine. This reduction amounted to 40% for the hypoxic chemosensitivity and 16% for hypercapnic chemosensitivity when compared with placebo. The results of the cardiopulmonary exercise tests are also shown in Table 2. Peak oxygen consumption increased from 18.0 ml/kg per min to 19.7 ml/kg per min after dihydrocodeine (p = 0.002), which was associated with a significant improvement in exercise time. Oxygen consumption at equivalent stages during submaximal exercise, however, did not vary significantly between placebo and dihydrocodeine (before exercise [placebo vs. dihydrocodeine]: 5.2 vs. 3.5 ml/kg per min, p = 0.13; at 3 min: 10.5 vs. 11.4 ml/kg per min, p = 0.41; at 6 min: 13.9 vs. 14.4 ml/kg per min, p = 0.41). The ventilatory response to exercise, characterized by the slope of the regression line relating minute ventilation to carbon dioxide output during exercise, was reduced by ∼10% with dihydrocodeine compared with placebo. The respiratory exchange ratio (R) at peak exercise was increased with dihydrocodeine, but this was not statistically significant. The predominant limiting symptom at peak exercise on placebo was dyspnea in 10 patients and fatigue in 2. Of these 10 patients, the major limiting symptom at peak exercise with dihydrocodeine remained dyspnea for eight and fatigue for two.
2.2 Exercise heart rate, blood pressure and arterial oxygen saturation.
Table 3shows the hemodynamic data, end-tidal carbon dioxide concentration and arterial oxygen saturation at rest, at 3 min of exercise, at 6 min of exercise and at peak exercise. There was no difference in heart rate or blood pressure with dihydrocodeine administration versus placebo.
End-tidal carbon dioxide concentration was significantly higher with dihydrocodeine than with placebo at peak exercise. There was a nonsignificant fall in arterial oxygen saturation at peak exercise when comparing dihydrocodeine with placebo. Focusing on the end-tidal carbon dioxide concentration when patients received placebo, there was a trend toward a decrease at peak exercise compared with the rest value. In contrast, when patients received dihydrocodeine, there was a trend toward an increase in the end-tidal carbon dioxide concentration during exercise. A nonsignificant fall in arterial oxygen saturation during exercise was also seen, compared with the rest value.
2.3 Exercise ventilation and Borg score.
Fig. 1compares the minute ventilation, respiratory rate and tidal volume during exercise with dihydrocodeine and placebo. There was a significant reduction in minute ventilation during submaximal exercise with dihydrocodeine, which was accounted for principally by the reduction in respiratory rate. The minute ventilation at peak exercise was similar with both treatments, bearing in mind that patients receiving dihydrocodeine exercised for a longer time. Table 4shows the modified Borg score for dyspnea and for fatigue. It demonstrates a reduction in breathlessness at submaximal exercise with the administration of dihydrocodeine, but, in contrast, there was no significant change in the perception of fatigue.
We have shown that dihydrocodeine reduced breathlessness, at least during submaximal exercise, and also improved exercise tolerance. There are several possible ways that dihydrocodeine may do this.
3.1 Chemosensitivity and dyspnea.
The first and most likely way that dihydrocodeine reduced breathlessness and improved exercise tolerance is through a reduction of chemosensitivity. The mean hypoxic and hypercapnic chemosensitivities of the chronic heart failure patients who participated in this study while receiving placebo were significantly higher than the normal control values in our laboratory (0.293 ± 0.056 liter/min per %Sao2and 2.02 ± 0.24 liter/min per mm Hg, respectively) (). We have demonstrated that the augmented chemosensitivity was reduced with dihydrocodeine administration. There was a concomitant reduction in minute ventilation at all stages, except at peak exercise, accounted for mainly by a decrease in respiratory rate. As pointed out earlier, the genesis of dyspnea may, in part, be due to an excessive increase in exercise ventilation as well as the disproportionate rise in respiratory rate relative to tidal volume (). A higher respiratory rate may conceivably cause respiratory muscle fatigue, contributing to the sensation of dyspnea (). In addition, such a characteristic mode of ventilation is inefficient because the proporation of dead space (hence wasted) ventilation in increased ([8, 9]). That minute ventilation was similar at peak exercise with placebo and dihydrocodeine administration also lends credence to the notion that the predominant mechanism by which dihydrocodeine causes respiratory depression is through a reduction of chemoreceptor-mediated respiratory drive rather than by the generalized depression of the central nervous system.
Of relevance is the recent finding that there is an association between blunted hypoxic chemosensitivity and the lack of perception of dyspnea in asthmatic patients with a history of a near-fatal status asthmaticus (). Other studies have shown a reduction in the perception of dyspnea in asthmatic patients who had both carotid bodies removed as part of the treatment of asthma, suggesting a close association between hypoxic chemosensitivity and the perception of this symptom ([23, 24]). It may be that afferent signals from chemoreceptors not only act on the medullary respiratory centers, but are also directly perceived as breathlessness. Although a greater reduction in hypoxic chemosensitivity with dihydrocodeine administration vis-à-vis hypercapnic chemosensitivity (40% vs. 16%) was noted, our study does not provide further information on the specific contribution of each toward the generation of dyspnea. Despite a reduction in chemosensitivity after dihydrocodeine administration, the major limiting symptom at peak exercise did not change in the majority of patients, probably because the patients exercised for a longer time and reached a similar level of minute ventilation at peak exercise, compared with placebo.
3.2 Alteration in hemodynamic data and sympathetic drive.
The second possible way that dihydrocodeine improved exercise tolerance may be through the beneficial hemodynamic effects of dihydrocodeine. Morphine has been shown to have both venodilatory () and systemic arteriolar dilatory effects () secondary to a reduction in sympathetic activity. This reduction may again be partly due to the suppression of chemoreceptors, as they have an excitatory action on the brain stem and can increase sympathetic outflow ([27, 28]). Indeed, increased sympathetic outflow caused by an augmented chemoreflex is thought to contribute to the neurohormonal abnormality in chronic heart failure (). However, previous studies have shown that improving hemodynamic data with vasodilator drugs in patients with chronic heart failure did not necessarily improve exercise tolerance or symptoms in the short term (). The hemodynamic effects of dihydrocodeine are also probably less than those of morphine considering that it is less potent (). Although we did not measure the central venous pressure or the systemic vascular resistance in these patients, there were no evident changes in blood pressure or heart rate to suggest a significant alteration in hemodynamic effects or sympathetic drive. Thus, it is less likely that dihydrocodeine improved exercise tolerance by altering hemodynamic data or sympathetic activity.
3.3 Alteration in central perception.
The third possible way that dihydrocodeine reduced breathlessness and improved exercise tolerance is through the alteration of central perception of discomfort, both of dyspnea and fatigue. However, it is unlikely that this is the main explanation because there was an absence of significant improvement in the perceived symptom of fatigue with dihydrocodeine, as would be expected if there was a general alteration in central perception.
3.4 Maintenance of arterial oxygen saturation.
As expected with the reduction in chemosensitivity, there was a significant increase in end-tidal carbon dioxide concentration at peak exercise with dihydrocodeine compared with placebo. In contrast, the fall in arterial oxygen saturation at peak exercise with dihydrocodeine administration was not significant compared with the pre-exercise value. This suggests that oxygen saturation was satisfactorily maintained during exercise, despite dihydrocodeine administration, and did not cause significant hypoxia.
3.5 Increased peak oxygen consumption.
We were surprised to see that there was an improvement in peak oxygen consumption after dihydrocodeine administration. The most likely reason for this is that with the reduction in the perceived sensation of breathlessness, these patients were able to improve their exercise duration. Because of the delayed oxygen kinetics in patients with chronic heart failure (), the peak oxygen consumption may not reflect maximal oxygen consumption, and the slow upward drift with exercise prolongation may account for the improvement in peak exercise consumption with dihydrocodeine. As discussed earlier, it is unlikely that there was a significant alteration in hemodynamic data or sympathetic drive with dihydrocodeine to affect cardiac work.
It is also pertinent to note that oxygen consumption at rest and during submaximal exercise with dihydrocodeine administration was not significantly different from that with placebo. This is in agreement with a previous report with codeine administration (), although in contrast to another with morphine (). It is probably due to the lesser potency of dihydrocodeine in comparison to morphine. It is thought that morphine may reduce the metabolic rate for, as yet, unknown physiologic reasons (). The observation in our study that oxygen consumption during submaximal exercise with dihydrocodeine was similar to that with placebo goes against the reduction of metabolic rate as a factor in causing increased exercise tolerance, as has been suggested in studies of chronic obstructive pulmonary disease ().
3.6 Study limitations.
All patients in this study were men. This was not a deliberate study design; patient recruitment was random and voluntary. It is, however, very unlikely that the general findings of the study would be altered if there were female patients in our study. Chemosensitivity is also not known to be affected by gender ().
Our study was designed to demonstrate the contribution of chemosensitivity to increased exercise ventilation and breathlessness in chronic heart failure due to left ventricular systolic dysfunction. We therefore did not recruit patients with mitral stenosis in our study. Furthermore, patients with mitral stenosis have much restrictive lung disease (more so than chronic heart failure patients), and minute ventilation may underestimate the true output of the respiratory center ([36, 37]). The interpretation of data in these patients may be difficult and is confounded by other pulmonary factors.
This study has several implications. Dihydrocodeine appears to be beneficial in the short term under exercise laboratory conditions. This is likely to be related to the suppression of the augmented chemosensitivity seen in patients with chronic heart failure. The potential benefits of longer term administration of dihydrocodeine in chronic heart failure patients who are limited by breathlessness despite optimal conventional therapy merit further investigation. This, however, has to be considered against the background of possible drug tolerance and the small risk of dependence. Other pharmacologic means of chemoreceptor modulation (such as oxygen therapy []) are also likely to produce symptomatic benefit, just as neurohormonal blocking agents benefit patients with chronic heart failure, and they also merit further research. Second, this study demonstrates that augmented chemosensitivity contributes to the abnormal respiratory function of patients with chronic heart failure and has pathophysiologic significance. Finally, the chemoreflex is not only important in influencing the increased exercise ventilation and the perception of breathlessness in chronic heart failure, but may also contribute to the neurohormonal abnormality by increasing sympathetic outflow through its excitatory action on the brain stem, as discussed ([27, 28]).
☆ Dr. Chua and Dr. Webb-Peploe are recipients of British Heart Foundation junior research fellowships. Dr. Harrington is supported by the Robert Luff Foundation, London, England, United Kingdom; Dr. Ponikowski by the European Society of Cardiology, Sophia Antipolis, France; and Dr. Coats by the British Heart Foundation and Viscount Royston Trust, London.
- Received April 26, 1996.
- Revision received July 12, 1996.
- Accepted September 20, 1996.
- The American College of Cardiology
- Sullivan MJ,
- Higginbotham MB,
- Cobb FR
- Yokoyama H,
- Sato H,
- Hori M,
- Takeda H,
- Kamada T
- Chua TP,
- Clark AL,
- Amadi AA,
- Coats AJS
- The Criteria Committee of the New York Heart Association.,
- Dolgin M
- Read DJC
- Bruce RA,
- Blackman JR,
- Jones JW
- Beaver WL,
- Wasserman K,
- Whipp BJ
- El-Manshawi A,
- Killian KJ,
- Summers E,
- Jones NL
- Mancini DM,
- Henson D,
- LaManca J,
- Levine S
- Vismara LA,
- Leaman DM,
- Zelis R
- Shepherd JT
- Rubin SA,
- Chatterjee K,
- Parmley WW
- Marks P,
- Ashraf H,
- Root TR
- Santiago TV,
- Johnson J,
- Riley DJ,
- Edelman NH
- Moore DP,
- Weston AR,
- Hughes JMB,
- Oakley CM,
- Cleland JGF
- Hirshman CA,
- McCullough RE,
- Weil JV
- Reed JW,
- Ablett M,
- Cotes JE
- Andreas S,
- Morguet AJ,
- Werner GS,
- Kreuzer H