Hypoxia

Study objective: Although the overnight increase in urinary uric acid/creatinine ratio ([DELTA]UA/Cr) is considered by some to be a marker of tissue hypoxia in patients with obstructive sleep apnea-hypopnea syndrome (OSAS), this index is not universally accepted. The purpose of this study was to confirm the validity of [DELTA]UA/Cr as a marker of tissue hypoxia by measuring the plasma level of adenosine during sleep, and also to test the hypothesis that the heart rate (HR) response to apnea is a determinant of tissue hypoxia.

Design: Intergroup comparative study.

Setting: A university hospital, Sapporo, Japan.

Patients: Eighteen patients with OSAS who had apnea-associated, moderate-to-severe arterial desaturation. The patients were classified into two groups: the [DELTA]UA/Cr-positive group, who were considered to have tissue hypoxia, and the [DELTA]UA/Cr-normal group, who were not.

Measurements and results: Although there were no significant differences between two groups of the patients in either arterial desaturation parameters or the apnea-hypopnea index, the plasma level of adenosine during sleep was significantly higher in the [DELTA]UA/Cr-positive group than in the [DELTA]UA/Cr-normal group. Successful treatment with nasal continuous positive airway pressure significantly decreased both [DELTA]UA/Cr and the plasma level of adenosine only in the [DELTA]UA/Cr-positive group. The magnitude of the HR increase after the termination of apnea was significantly smaller in the [DELTA]UA/Cr-positive group.

Conclusions: [DELTA]UA/Cr is a marker of tissue hypoxia, which does not necessarily parallel arterial desaturation indexes in OSAS. Intersubject variability in the HR response to apnea may explain the discrepancy between tissue hypoxia and arterial desaturation indexes.

(CHEST 2002; 122:1686-1694)

Key words: adenosine; heart rate; sleep apnea syndrome; tissue hypoxia; uric acid

Abbreviations: AHI = apnea hypopnea index; AI = apnea index; ATP = adenosine triphosphate; CPAP = continuous positive airway pressure; HR = heart rate; [DELTA]HRdecr = baseline heart rate subtracted from the minimum heart rate; [DELTA]HRincr = baseline heart rate subtracted from the maximum heart rate; OSAS = obstructive sleep apnea-hypopnea syndrome; Sa[O.sub.2] = arterial oxygen saturation; [DELTA]UA/Cr = urinary uric acid/creatinine ratio; [DELTA]UA/Cr = overnight increase in urinary uric acid/creatinine ratio

Obstructive sleep apnea syndrome (OSAS) is characterized by periodic apnea that is associated with cyclic changes in arterial oxygen saturation (Sa[O.sub.2]) and cardiovascular variables such as heart rate (HR) (1-4) and BP. (3,5,6) The level of Sa[O.sub.2] observed in this syndrome is often lower than that of any other disease except when patients are in the terminal stage. Thus, several investigators have attempted to examine tissue hypoxia in this syndrome in various ways, including near-infrared spectrometry (7) and nuclear magnetic resonance spectrometry. (8) Since a "gold standard" definition for tissue hypoxia does not exist, it has been used by some to simply denote the fall of oxygen pressure to a certain level or to represent organ dysfunction due to oxygen depletion. However, the term tissue hypoxia is more commonly defined as inadequate oxygen supply against oxygen demand in the integrity of cellular metabolic processes. When oxygen supplies are inadequate to meet the oxygen demands of the cells, formation of adenosine triphosphate (ATP) from adenosine diphosphate is impaired and a net degradation of ATP to adenosine diphosphate and adenosine monophosphate occurs. This leads to the release of purine intermediates (adenosine, inosine, hypoxanthine, and xanthine) and the purine catabolic end product, uric acid. Consequently, the elevated levels of ATP degradation products in body fluids have been suggested to be a good index of tissue hypoxia in neonates with infant respiratory distress syndrome, (9,10) exercising human subjects, (11-15) critically ill patients, (16-17) and, more recently, primary pulmonary hypertension. (18)

In OSAS, Hasday and Grum (19) were the first to measure the overnight change of urinary uric acid secretion, and to report that it could be a good index of tissue hypoxia. However, the validation of this index has not been confirmed by some subsequent studies, (20,21) because this index remained normal in some patients with OSAS who had severe nocturnal arterial oxygen desaturation, and because the indexes of overnight urinary uric acid excretion and of nocturnal arterial oxygen desaturation did not correlate in those subjects. Adenosine in plasma, as an intermediary of such an ATP degradation pathway, may be a more sensitive marker of tissue hypoxia than uric acid in urine, although the measurement of plasma adenosine is technically not easy because of its short half-life in plasma. (22,23) Indeed, we have recently provided evidence in healthy humans that the production of tissue adenosine increases even during 20 min of moderate arterial hypoxemia (Sa[O.sub.2] = 80%) by simultaneously measuring the plasma level of adenosine in an artery and vein with pretreatment using dipyridamole. (24) To our knowledge, there has been only one report (25) that measured the plasma concentration of adenosine in patients with OSAS. Although they demonstrated that it was significantly higher in those who had significant arterial oxygen desaturation during sleep compared with those who did not, blood sampling in their study was performed not during sleep, but the next morning when subjects were awake.

The aim of this study was to ascertain the validity of an increase in overnight urinary uric acid secretion as an index of ATP degradation associated with tissue hypoxia by simultaneously measuring the plasma level of adenosine during sleep in patients with OSAS. Additionally, we examined the hypothesis that the tissue hypoxia in OSAS is determined not only by arterial oxygen desaturation but also by poor cardiovascular compensation to apnea-induced hypoxemia, which might resolve the controversy that exists among previous reports. (19-21)

MATERIALS AND METHODS

Subjects

Eighteen adult male subjects with OSAS diagnosed by standard polysomnography were recruited for this study. Inclusion criteria of subjects for this study were as follows: the apneahypopnea index (AHI) was > 15/h, the minimum Sa[O.sub.2] was < 80%, and the percentage of time with Sao2 < 90% was > 5%. We excluded the subjects who had gout, diabetes mellitus requiring medication, and renal dysfunction assessed by the serum level of creatinine.

Of 18 subjects, 4 subjects had diabetes mellitus that was well controlled by diet therapy alone, and none had any signs of diabetic neuropathy, retinopathy, and nephropathy. Hypertension was diagnosed in two subjects before hospital admission; one patient was treated with a [beta]-antagonist, angiotensin-converting enzyme inhibitor, and Ca antagonist, and the other patient was treated with a Ca antagonist alone. Mild-to-moderate hypertension was diagnosed in six other subjects on hospital admission, but they did not receive any medication until the subsequent studies were completed. No subjects received medications such as thiazides, loop diuretics, allopurinol, or aspirin, which might affect the metabolism of purine nucleosides or the urinary excretion of uric acid. All the subjects gave their written informed consent. The study protocol was approved by the Ethics Committee of the Hokkaido University School of Medicine.

Experimental Protocol for the Measurement of Plasma Adenosine

On the study night, each subject refrained from consuming any beverages or food except water after dinner at 5 PM. The subjects underwent standard nocturnal polysomnography from 9 PM to 6 AM. We continuously recorded EEG (C3 to A2 and C4 to A1), right and left electro-oculogram, chin electromyogram, pulse oximetry (Biox 3740; Ohmeda; Louisville, CO), thoracic and abdominal movements (Respitrace; Ambulatory Monitoring; Ardsley, NY), and oronasal airflow by thermistor. The response mode of the pulse oximeter was set at fast speed, which updated display data every 1/3 s, and the pulse rate displayed was the average of the previous 5 s. (4)

Sa[O.sub.2] monitored by the pulse oximeter was recorded on-line into a computer, and the following oxygen desaturation parameters were analyzed: the minimum Sa[O.sub.2], the mean Sa[O.sub.2], and the ratio of oxygen desaturation time over total sleep time, such as time with Sa[O.sub.2] < 70%, time with Sa[O.sub.2] < 80%, and time with Sa[O.sub.2] < 90%. Apnea was defined as cessation of air flow for at least 10 s, and hypopnea as a reduction of air flow by > 50% of the mean amplitude of the subject at rest in the supine position, followed by a reduction in Sa[O.sub.2] of at least 4%.

In an attempt to repeatedly obtain blood samples, a 20-gauge catheter was placed in the forearm cubital vein, and physiologic saline solution, 40 mL/h, was drip infused throughout the night. For the measurement of plasma adenosine, we obtained blood samples with a stopping solution (24) before sleep at approximately 9 PM, during sleep, and after awakening at approximately 6 AM. Blood sampling before sleep and after awakening was done while the subject was confirmed to be awake in bed and in a supine position. Blood sampling during sleep was done 3 to 10 times (6.1 [+ or -] 1.8 times, mean [+ or -] SD) per night during the period when each subject demonstrated periodic apnea associated with Sa[O.sub.2] < 85% during non-rapid eye movement. The sampling time was approximately 20 s, and the sampling intervals were 40 min at minimum and 4 h at maximum.

We measured the plasma concentration of adenosine by high-pressure liquid chromatography fluorometric analysis following the protocol of Zhang et al. (26) The details were described in our previous publication. (24)

Measurement of Overnight Change in Urinary Uric Acid Secretion

Urine samples were collected at the study night when we measured the plasma concentration of adenosine. For measurement of the overnight change in urinary uric acid secretion, two urine specimens were collected per night from each subject: the first before going to sleep, and the second within 20 min after awakening the next morning. A sodium hydroxide solution, 2 N, 1:50 volume/volume, was added to avoid urate precipitation in each sample, which was stored at 4[degrees]C until measurement. Measurements of uric acid and creatinine were performed by the uricase-peroxidase method and the creatinase-peroxidase method, respectively. The overnight increase in urinary uric acid/creatinine ratio (UA/Cr) [[DELTA]UA/Cr] was calculated according to the equation proposed by Hasday and Grum (19):

[DELTA]UA/Cr = [(UA/Cr morning)/(UA/Cr evening) - 1] X 100%

In an attempt to confirm the normal range of [DELTA]UA/Cr reported in previous studies, we obtained 28 measurements of [DELTA]UA/Cr from 12 control subjects: 5 normal subjects who underwent polysomnography with suspicion of OSAS, and 7 patients with mild OSAS who did not show arterial oxygen desaturation of < 80% during sleep (Table 1).

Effect of Nasal Continuous Positive Airway Pressure Treatment

Of the 18 subjects enrolled in the above-mentioned study, 12 subjects were successfully treated by nasal continuous positive airway pressure (CPAP). For those subjects, we conducted the same overnight study 1 week after the introduction of nasal CPAP.

Heart Rate Response

In an attempt to examine the hypothesis that intersubject variability of the HR response to apnea is a determinant of tissue hypoxia in OSAS, we analyzed the changes in HR associated with apnea, using a polysomnographic chart on the study night. The HR response of each subject was assessed by the baseline HR subtracted from the maximum HR ([DELTA]HRincr) and the baseline HR subtracted from the minimum HR ([DELTA]HRdecr) [Fig 1]. The highest HR immediately after the termination of each episode of apnea was called the maximum HR, and the lowest HR observed during apnea was called the minimum HR. The average HR observed just before subjects fell into sleep on the study night was called the baseline HR.

[FIGURE 1 OMITTED]

For individual analysis, one author who was blind to any other data selected 30 to 40 apneic episodes throughout the night. Apnea episodes analyzed were selected regularly after a certain number of episodes of apnea to avoid any intentional selection bias. The average of 30 to 40 measurements was calculated and used as a representative value for each parameter in each subject. To ensure that the selected apneic episodes reflected the overnight data, we also calculated the average of the minimum Sa[O.sub.2] associated with the selected apneic episodes. Pearson correlation coefficient was calculated to examine the properties of the selection. The average of the minimum Sa[O.sub.2] for selected apneas in each subject was significantly correlated with the mean Sa[O.sub.2] (r = 0.95, p < 0.01) and time with Sa[O.sub.2] < 80% (r = 0.94, p < 0.01), which were obtained from the overnight data.

Data Analysis

Data are shown as means [+ or -] SEM unless otherwise specified. Comparison between two groups was done by unpaired t test when the equal variance was accepted by the F test, and by the Mann-Whitney U test when equal variance was denied. The statistical significance of the change in plasma adenosine or [DELTA]UA/Cr with nasal CPAP was examined by the Wilcoxon signed-rank test. To examine correlations between two variables, Pearson single correlation coefficient (r) or Spearman rank correlation coefficient was employed where appropriate, p < 0.05 was accepted as significant.

The mean [+ or -] SD value of [DELTA]UA/Cr in the control subjects was -32.0 [+ or -] 12.9%. We thus considered the normal range of AUA/Cr to be below zero for subsequent analysis, which was comparable to the range reported in some previous studies, (19,21) and slightly higher than the mean + 2 SD obtained from our control subjects. The subjects were classified into two groups by the value of zero: the [DELTA]UA/Cr-positive group and the [DELTA]UA/Cr-normal group. [DELTA]UA/Cr positive means that the urinary uric acid secretion during sleep was increased compared with that of the before-sleep condition.

RESULTS

The value of [DELTA]UA/Cr in the 18 subjects widely varied from - 60.8 to 58.3% (mean, - 6.1 [+ or -] 7.5%). Of them, nine subjects were classified into the [DELTA]UA/Cr-positive group, and the other nine subjects were classified into the normal group (Fig 2). The data obtained from 12 control subjects are also illustrated as individual mean values in Figure 2.

[FIGURE 2 OMITTED]

There were no significant differences between the two groups in age, body mass index, and the plasma level of uric acid. Unexpectedly, there were no significant differences either in polysomnographic data such as the apnea index (AI), AHI, or in any of the arterial oxygen desaturation indexes (Table 1). There was no significant correlation between these indexes and [DELTA]UA/Cr.

The intrasubject coefficient variation in the plasma concentration of adenosine during sleep was 24.2%. Since this intrasubject variation in the plasma concentration of adenosine was not apparently influenced by the timing of blood sampling with regard to apnea duration or apnea-associated arterial oxygen desaturation, the mean value of 3 to 10 measurements was considered to be a representative value for each subject for later analysis. The plasma concentrations of adenosine before sleep were 13.7 [+ or -] 2.0 nm for the [DELTA]UA/Cr-positive group and 9.2 [+ or -] 1.6 nm for the normal group, which were not significantly different (Fig 3). During sleep, its plasma concentration was slightly but significantly increased up to 15.2 [+ or -] 1.9 nm (p < 0.05) only in the [DELTA]UA/Cr-positive group, and was also significantly higher than that of the normal group during sleep (9.8 [+ or -] 1.7 nm, p < 0.05) [Fig 4]. Furthermore, there was a significant correlation (r = 0.55, p < 0.05) between [DELTA]UA/Cr and the plasma concentration of adenosine during sleep (Fig 5). The significant difference observed between the two groups in the plasma concentration of adenosine did not disappear even after awakening (13.6 [+ or -] 1.4 nm in the [DELTA]UA/ Cr-positive group and the 8.9 [+ or -] 1.6 nm in the normal group, p < 0.05) [Fig 3].

[FIGURES 3-5 OMITTED]

Treatment with nasal CPAP was performed for seven subjects (AI, 40.1 [+ or -] 5.4 episodes per hour) in the [DELTA]UA/Cr-positive group and in five subjects (AI, 56.2 [+ or -] 6.2 episodes per hour) in the normal group. The AHI decreased to five or less episodes per hour in all the subjects with this treatment. Approximately i week after the introduction of nasal CPAP, the value of [DELTA]UA/Cr significantly decreased from 17.5 [+ or -] 4.9 to - 12.0 [+ or -] 7.0% (p < 0.01) only in the [DELTA]UA/Cr-positive group. It did not significantly change in the normal group (from - 34.9 [+ or -] 7.1 to - 29.6 [+ or -] 7.8%) [Fig 6, top]. The plasma concentration of adenosine similarly decreased in all seven subjects from 14.8 [+ or -] 2.2 to 9.8 [+ or -] 1.0 nm (p < 0.05) again only in the [DELTA]UA/Cr-positive group. It did not significantly change in the normal group (from 8.0 [+ or -] 2.0 to 8.3 [+ or -] 1.5 nm, not significant) [Fig 3 and Fig 6, bottom]. The significant difference in the plasma concentration of adenosine observed during sleep between the two groups before the introduction of nasal CPAP disappeared after the treatment.

[FIGURE 6 OMITTED]

Concerning the HR analysis, the baseline HR levels were 67.4 [+ or -] 4.3 beats/min in the [DELTA]UA/Cr positive group and 62.7 [+ or -] 4.7 beats/min in the [DELTA]UA/ Cr-normal group. These values were not significantly different. The [DELTA]HRincr, but not [DELTA]HRdecr, was significantly different between the two groups; that is, the increase in HR from the baseline to the maximum was lower in the [DELTA]UA/Cr-positive group than in the [DELTA]UA/Cr-normal group (6.4 [+ or -] 1.4 beats/ rain vs 14.0 [+ or -] 1.6 beats/rain, respectively; p < 0.01) [Fig 7]. The average of the lowest Sa[O.sub.2] associated with analyzed apnea was 80 [+ or -] 3% in the [DELTA]UA/Cr-positive group, which was not significantly different from the 76 [+ or -] 4% in the [DELTA]UA/Cr-normal group. Thus, the observed difference in [DELTA]HRincr between the two groups could not be explained by the difference in the level of apnea-induced hypoxemia.

[FIGURE 7 OMITTED]

DISCUSSION

In this study, we demonstrated that the index of [DELTA]UA/Cr did not parallel the severity of the AI or arterial oxygen desaturation, but was significantly linked to the plasma level of adenosine, which is considered to be another marker of tissue hypoxia, in patients with OSAS. These data indicate that both indexes of [DELTA]UA/Cr and plasma adenosine during sleep reflect the same purine catabolic pathway associated with tissue hypoxia in these patients. We further demonstrated that the intersubject variability of the HR response to sleep apnea may be a determinant of the tissue hypoxia assessed by these parameters.

The first report by Hasday and Grum (19) claimed that the urinary uric acid/creatinine ratio was greater in the morning than before sleep, when subjects showed significant nocturnal arterial oxygen desaturation, and that the morning value was decreased to less than before-sleep values with successful treatment of sleep apnea by nasal CPAP. However, two subsequent studies (20,21) did not necessarily agree with the validity of this index as a marker of tissue hypoxia in this syndrome, because this index has poor sensitivity in detecting nocturnal hypoxemia and does not correlate with indexes of apnea-associated arterial oxygen desaturation. Consequently, this index is not recommended for routine use to assess the severity of OSAS at the present time despite the simplicity and noninvasiveness of the measurement.

In this study, we attempted to examine the validity of this index by measuring the plasma concentration of adenosine during sleep. We found that the [DELTA]UA/ Cr-positive group showed higher levels of plasma adenosine during sleep, and that both variables decreased to the range of the [DELTA]UA/Cr-normal group with successful treatment of sleep apnea by nasal CPAP. These data support the concept that these indexes reflect the ATP degradation possibly due to tissue hypoxia, which does not necessarily parallel arterial oxygen desaturation. Accordingly, the tissue hypoxia assessed by [DELTA]UA/Cr should be determined not only by arterial oxygen desaturation itself, but also by some other factors.

In general, tissue hypoxia is determined by the balance between arterial oxygen transport and oxygen demands in tissue. Oxygen transport is determined not only by Sa[O.sub.2] but also by the concentration of hemoglobin, the hemoglobin dissociation curve, cardiac output, the distribution of tissue blood flow, and others. (27) For instance, the degree to which a carbon dioxide increase (and a resultant pH decrease) causes a shift in the oxyhemoglobin dissociation curve may be a significant factor. This could give the appearance of a more significant hemoglobin desaturation without a concomitant reduction in tissue level oxygenation.

In this study, we demonstrated that [DELTA]HRincr could discriminate the subjects who had tissue hypoxia from those who did not, despite similar levels of respiratory disturbances and arterial oxygen desaturation in the two groups. In OSAS, there are cyclic fluctuations of HR and other cardiovascular responses associated with sleep apnea and resumption of ventilation, and such cyclic fluctuations are known to be highly variable among subjects. (3,4) The factors explaining those fluctuations include the bradycardiac response (28) due to stimulation of the carotid body by hypoxia, (4) changes in intrathoracic pressure due to attempted inspiration against an occluded pharynx, (29,30) resumption of ventilation itself, (31) disruption of sleep architecture (32) and primary or secondary periodic activation of the sympathetic nervous system. (33,34) In addition, the change in HR in response to hypoxia may be influenced by coexisting diseases such as diabetes mellitus or COPD. (35,36) The elevation of HR after the termination of apnea is considered to reflect the sympathetic nerve activation in response to hypoxia itself or as a result of the arousal response and/or the stretch receptor reflex associated with postapneic hyperventilation. Although the linkage of the poor HR response to apnea with tissue hypoxia we observed in this study does not indicate a causal relationship, it is possible that the poor cardiovascular compensation reflected in low [DELTA]HRincr may play a key role as a determinant of tissue hypoxia in OSAS.

In this study, we measured, for the first time, the plasma concentration of adenosine during sleep in patients with OSAS. We showed that the plasma concentration of adenosine during sleep was significantly higher than that of before-sleep levels in the [DELTA]UA/Cr-positive OSAS group, but not in the [DELTA]UA/ Cr-normal group. Adenosine accumulated in the extracellular space should enter peripheral circulation and then be immediately taken up by RBC or vascular endothelial cells. The process of re-uptake is so fast (22,23) that even a significant increase in the tissue production of adenosine hardly causes a visible increase in the plasma concentration of adenosine. (24) Accordingly, we speculate that the significant increase in the plasma concentration of adenosine we observed during sleep in the [DELTA]UA/Cr-positive OSAS group, although it was only to a small degree, should represent a much larger acceleration of adenosine production in tissue. Findley and colleagues (25) were the first to measure the venous concentration of adenosine in patients with OSAS, although blood was sampled in the morning when all the subjects were awake. Compared to normoxemic patients or normal volunteers, they reported that hypoxemic patients showed higher levels of plasma adenosine. Their observation agrees with our present findings that the significant difference in the plasma concentration of adenosine observed between the [DELTA]UA/Cr-positive group and the [DELTA]UA/Cr-normal group during sleep remained even the next morning when the subjects were awake. It is not clear, however, why and how long an increased level of adenosine as a result of sleep apnea stays after the awakening of patients.

Finally, mention should be made of the limitation of this study. Although both [DELTA]UA/Cr and the plasma concentration of adenosine should represent tissue hypoxia in some tissues or organs, neither of these parameters provides any information on potential variations of tissue hypoxia in a variety of tissue or organs. In vital organs such as the brain (37,38) or heart, (39-41) it may well be that a number of compensatory and protective mechanisms work in response to arterial hypoxemia before the breakdown of ATP occurs. Alternatively, these organs may readily produce adenosine so as to increase local blood flow and/or protect the integrity of cellular metabolic processes by the action of adenosine itself.

In conclusion, [DELTA]UMCr is a marker of tissue hypoxia because it reflects ATP degradation associated with sleep apnea. The reason why tissue hypoxia assessed by [DELTA]UA/Cr does not necessarily parallel arterial desaturation indexes in sleep apnea syndrome might be explained by intersubject variability ill the cardiovascular responses to apnea such as the HR response. Further studies are required to examine whether the presence of tissue hypoxia assessed by ATP degradation could be an independent risk factor for future cardiovascular events or even for mortality in patients with OSAS.

* From the First Department of Medicine (Drs. Saito, Nishimura, Shibuya, Makita, Tsujino, and Kawakami), Hokkaido University School of Medicine; and Department of Physical Therapy (Dr. Miyamoto), College of Medical Technology, Hokkaido University, Hokkaido, Japan.

REFERENCES

(1) Guilleminault C, Connolly S, Winkle B, et al. Cyclical variation of the heart rate in sleep apnoea syndrome: mechanisms, and usefulness of 24 h electrocardiography as a screening technique. Lancet 1984; 1:126-131

(2) George CF, Kryger MH. Sleep and control of heart rate. Clin Chest Med 1985; 6:595-601

(3) Bonsignore M, Marrone O, Insalaco G, et al. The cardiovascular effects of obstructive sleep apnoeas: analysis of pathogenic mechanisms. Eur Respir J 1994; 7:786-805

(4) Sato F, Nishimura M, Shinano H, et al. Heart rate during obstructive sleep apnea depends on individual hypoxic chemo-sensitivity of the carotid body. Circulation 1997; 96:274-281

(5) Ringler J, Basner R, Shanon R, et al. Hypoxemia alone does not explain blood pressure elevations after obstructive apneas. J Appl Physiol 1990; 69:2143-2148

(6) Shepard J. Gas exchange and hemodynamics during sleep. Med Clin North Am 1985; 69:1243-1264

(7) Hayakawa T, Terashima M, Kayukawa Y, et al. Changes in cerebral oxygenation and hemodynamics during obstructive sleep apneas. Chest 1996; 109:916-921

(8) Kamba M, Suto Y, Ohta Y, et al. Cerebral metabolism in sleep apnea: evaluation by magnetic resonance spectroscopy. Am J Respir Crit Care Med 1997; 156:296-298

(9) Jensen MH, Brinklov MM, Lillquist K. Urinary loss of oxypurines in hypoxic premature neonates. Biol Neonate 1980; 38:40-48

(10) Raivio KO. Neonatal hyperuricemia. J Pediatr 1976; 88: 625-630

(11) Sorlie D, Myhre K, Saugstad OD, et al. Release of hypoxanthine and phosphate from exercising human legs with and without arterial insufficiency. Acta Med Scand 1982; 211:281-286

(12) Harkness RA, Simmonds RJ, Coade SB. Purine transport and metabolism in man: the effect of exercise on concentrations of purine bases, nucleosides and nucleotides in plasma, urine, leukocytes and erythrocytes. Clin Sci 1983; 64:333-340

(13) Yamanaka H, Kawagoe Y, Taniguchi A, et al. Accelerated purine nucleotide degradation by anaerobic but not by aerobic ergometer muscle exercise. Metabolism 1992; 41:364-369

(14) Hellsten Y, Maclean D, Radegran G, et al. Adenosine concentration in the interstitium of resting contracting human skeletal muscle. Circulation 1998; 98:6-8

(15) MacLean DA, Sinoway LI, Leuenberger U. Systemic hypoxia elevates skeletal muscle interstitial adenosine levels in humans. Circulation 1998; 98:1990-1992

(16) Christensen EF, Jacobsen J, Anker-Moller E, et al. Increased urinary loss of uric acid in adults with acute respiratory failure requiring mechanical ventilation. Chest 1992; 102:556-559

(17) Grum CD, Simon RH, Dantzsker DR, et al. Evidence for adenosine triphosphate degradation in critically-ill patients. Chest 1985; 88:763-767

(18) Nagaya N, Uematsu M, Satoh T, et al. Serum uric acid levels correlate with the severity and the mortality of primary pulmonary hypertension. Am J Respir Crit Care Med 1999; 160:487-492

(19) Hasday JD, Grum CD. Nocturnal increase of urinary uric acid:creatinine ratio. Am Rev Respir Dis 1987; 135:534-538

(20) McKeon JL, Saunders NA, Murree-Allen K, et al. Urinary uric acid:creatinine ratio, serum erythropoietin, and blood 2,3-diphosphoglycerate in patients with obstructive sleep apnea. Am Rev Respir Dis 1990; 142:8-13

(21) Braghiroli A, Sacco C, Erbetta M, et al. Overnight urinary uric acid: creatinine ratio for detection of sleep hypoxemia: validation study in chronic obstructive pulmonary disease and obstructive sleep apnea before and after treatment with nasal continuous positive airway pressure. Am Rev Respir Dis 1993; 148:173-178

(22) Klabunde RE. Dipyridamole inhibition of adenosine metabolism in human blood. Eur J Pharmacol 1983; 93:21-26

(23) Moser GH, Schrader J, Deussen A. Turnover of adenosine in plasma of human and dog blood. Am J Physiol 1989; 256: C799-C806

(24) Saito H, Nishimura M, Shinano H, et al. Plasma concentration of adenosine during normoxia and moderate hypoxia in humans. Am J Respir Crit Care Med 1999; 159:1014-1018

(25) Findley LJ, Boykin M, Fallon T, et al. Plasma adenosine and hypoxemia in patients with sleep apnea. J Appl Physiol 1988; 64:556-561

(26) Zhang Y, Geiger JD, Lautt WW. Improved high-pressure liquid chromatographic-fluorometric assay for measurement of adenosine in plasma. Am J Physiol 1991; 260:G658-G664

(27) Collaborative Group on Intracellular Monitoring. Intracellular monitoring of experimental respiratory failure. Am Rev Respir Dis 1988; 138:484-487

(28) Zwillich C, Devlin T, White D, et al. Bradycardia during sleep apnea. J Clin Invest 1982; 69:1286-1292

(29) Hanly P, George C, Millar T, et al. Heart rate response to breath-hold: Valsalva and Mueller maneuvers in obstructive sleep apnea. Chest 1989; 95:735-739

(30) Andreas S, Hajak G, Breska B, et .al. Changes in heart rate during obstructive sleep apnoea. Eur Respir J 1992; 5:853-857

(31) James J, Daly M. Responses in apnoeic asphyxia: role of arterial chemoreceptors and the modification of their effects by a pulmonary vagal inflation reflex. J Physiol 1969; 201:87-104

(32) Ringler J, Garpestad E, Basner R, et al. Systemic blood pressure elevation after airway occlusion during NREM sleep. Am J Respir Crit Care Med 1994; 150:1062-1066

(33) Somers V, Phil D, Dyken M, et al. Sympathetic-nerve activity during sleep in normal subjects. N Engl J Med 1994; 328:303-307

(34) Morgan B, Denahan T, Ebert T. Neurocirculatory consequences of negative intrathoracic pressure vs asphyxia during voluntary apnea. J Appl Physiol 1993; 74:2969-2975

(35) Nishimura M, Miyamoto K, Suzuki A, et al. Ventilatory and heart rate response to hypoxia and hypercapnia in patients with diabetes mellitus. Thorax 1989; 44:251-257

(36) Miyamoto K, Nishimura M, Akiyama Y, et al. Augmented heart rate response to hypoxia in patients with COPD. Am Rev Respir Dis 1992; 145:1384-1388

(37) Phillis JW. Adenosine in the control of the cerebral circulation. Cerebrovasc Brain Metab Rev 1989; 1:26-54

(38) Ongini E, Adami M, Ferri C, et al. Adenosine A2A receptors and neuroprotection. Ann N Y Acad Sci 1997; 825:30-48

(39) Rubio R, Berne RM. Regulation of coronary blood flow. Prog Cardiovasc Dis 1975; 18:105-122

(40) Berne RM. The role of adenosine in the regulation of coronary blood flow. Circ Res 1980; 47:807-813

(41) Olsson RA, Pearson JD. Cardiovascular purinoceptors. Physiol Rev 1990; 70:761-845

Hiroshi Saito, MD, FCCP ([dagger]); Masaharu Nishimura, MD; Eiji Shibuya, MD; Hironi Makita, MD; Ichizo Tsujino, MD; Kenji Miyamoto, MD; and Yoshikazu Kawakami, MD, FCCP ([dagger])

([dagger]) Currently at Kohnan Hospital, Kohnan, Japan. This study was supported by Research Grant for the Intractable Diseases from the Ministry of Health and Welfare, Japan, and Science Research grant No. 04670451 from the Ministry of Education, Science, Sports and Culture, Japan. Dilazep was a gift of Kowa Co. Ltd.

Correspondence to: Masaharu Nishimura MD First Department of Medicine, Hokkaido University School of Medicine, North 15, West 7, Kita-ku, Sapporo, 060-8638, Japan; e-mail: ma-nishi@ med.hokudai.ac.jp

COPYRIGHT 2002 American College of Chest Physicians
COPYRIGHT 2003 Gale Group