The study enrolled 75 patients, including 31 cases with OSA syndrome, 30 with COPD, and a control group including 14 subjects who were eligible and gave informed consent to participate in this prospective follow-up study. The diagnostic criteria for OSA and CPAP (continuous positive airway pressure) treatment were determined according to the standard criteria published by the American Academy of Sleep Medicine (AASM).18 The criteria are defined below. OSA patients displayed at least two of the following three symptoms: snoring, witnessed apnea, and excessive daytime sleepiness. Their AHI score was 5 or higher in the overnight polysomnography (PSG).18 CPAP treatment was given if AHI>30 or AHI between 5 and 30 with a cardiovascular co-morbidity. OSAS patients did not have COPD defined by spirometry.

Patients were routinely called after 3 months and 1 year periods to our outpatients clinic. For reassessment after 5 years patients were contacted by telephone interview (n=68, 58 alive, 10 died). 35 of them agreed to give uric acid samples (blood and urine) (Fig. 1). The compliance of CPAP use was evaluated after follow-up.

Figure 1.

Flowchart of the study procedures. PSG: polysomnography, PFT: pulmonary function tests, UA: uric acid, NIMV: non invasive mechanical ventilation, OSAS NN: obstructive sleep apnea syndrome nocturnal normoxemic, OSAS-NH: obstructive sleep apnea syndrome nocturnal hypoxemic, COPD-NN: chronic obstructive pulmonary disease nocturnal normoxemic, COPD-NH: chronic obstructive pulmonary disease nocturnal hypoxemic.

(0.37MB).

The recruitment criteria for COPD included a history of 10 or more pack-years of smoking or a history of biomass exposure, a forced expiratory volume in one second (FEV 1) of less than 80% of the predicted value after bronchodilator use and a ratio of FEV 1 to forced vital capacity (FVC) of 0.7 or less after bronchodilator use.19 The condition of the patients was graded according to the stages of disease defined by the Global initiative for chronic Obstructive Lung Disease (GOLD).20 COPD patients had moderate to severe COPD disease without daytime hypoxia.21 Patients who displayed any of the symptoms of snoring and/or witnessed apnea were not enrolled in the study. The AHI scores were below 5 in the overnight PSG. The subjects in the control group had neither OSA nor COPD. They were admitted to the sleep laboratory with a complaint of snoring. Patients with OSA and COPD, who spent >10% of sleep time with oxyhemoglobin saturation (SaO2) below 90% were classified as nocturnal hypoxemic (NH). Personal characteristics and disease-related factors were assessed based on patient records. Flowchart of the subjects and procedures are presented in Fig. 1.

Patients were not enrolled in the study if they had any disease or medication that might alter the metabolism or urinary excretion of uric acid, as in the following:

• 1.

  Excessive UA production: Physician-diagnosed gout, diabetes mellitus, hemolytic anemia, myelo-lymphoproliferative disease, psoriasis, Paget's disease, or glucose-6-phosphatase deficiency or glycogen storage disease.

• 2.

  Decreased UA excretion: Physician-diagnosed renal failure, acidosis, sarcoidosis, lead intoxication or berylliosis.

• 3.

  Medication: Salicylic acid (in excess of 2g/day), diuretic, cyclosporine, levodopa, phenylbutazone (in excess of 200mg/day), ethambutol, pyrazinamide, nicotinic acid, nitroglycerin (intravenous) theophylline, allopurinol.

• 4.

  Alcohol intake on the day of PSG and follow-up.

• 5.

  Exercise on the day of PSG.

Subjects were asked not to exercise or consume alcohol during the day before the PSG test, which could increase anaerobic metabolism and thus uric acid production. The study protocol was approved by the Institutional Review Board Ethics Committee of the research hospital. Informed consent was obtained from each patient.

Polysomnography was performed in the Sleep Laboratory with Embla polysomnograph running Somnologica software version 4.0 (Flaga hf. Medical Devices, Iceland) and included four electroencephalography (EEG) channels (C3 to A1, C4 to A2, O1 to A2, O2 to A1), right and left electrooculography (EOG) channels, one chin electromyography (EMG) channel and four tibialis anterior EMG channels, finger pulse oximeter, strain gauges for thoraco-abdominal movements, one electrocardiography (ECG) lead, a nasal airflow (pressure cannula), and a digital microphone to detect snoring. Polysomnography recordings were scored in 30-s epochs for sleep, breathing and oxygenation according to the standard criteria published by the American Academy of Sleep Medicine (AASM).18 Obstructive apneas were defined as complete cessation of oro-nasal airflow for at least 10s in the presence of chest-wall motion. Hypopneas were defined as 30% or higher reduction in respiratory airflow with 4% or higher arterial oxygen desaturation or 50% or higher reduction in respiratory airflow with 3% or higher arterial oxygen desaturation or an arousal with any of the above for at least 10s. AHI was calculated based on the total number of apneas and hypopneas per hour of sleep.22

Two urine samples were obtained from all patients to calculate UA excretion and the ΔUA/Cr ratio. The first urinary sample was obtained before PSG recording at night, and the second after the recording in the morning. The specimens obtained at night were kept at +4°C. Urine samples were analyzed in the morning for uric acid with uricose (enzymatic, calorimetric). Venous blood samples were obtained and analyzed for creatinine (calorimetric) with Jaffle method, BUN with urease (with kinetic method), and sodium with ion selective electrode (ISE) on a multi-parameter analyzer (Olympus AU 640; Olympus Optical, Tokyo, Japan). After 5 year UA excretion was calculated according to the formula in mg/dL unit, for the sample obtained in the morning23:

ΔUA/Cr was calculated with Hasday and Grum's method15:

Pulmonary Function Testing was performed in the standing position without nose clips (ZAN 100, Flowhandy, Germany).24 Inhaled salbutamol 400μg was administered and the post-bronchodilator measurements were performed after 15min. At least three, up to five forced expiratory maneuvers were performed to obtain three technically satisfactory blows. The highest FVC and FEV1 values were reported.

Data were expressed as frequencies for categorical variables; and as median (IQR) or as mean and standard deviation for continuous variables, where appropriate. Comparisons between groups were tested by chi-square testing for categorical values and One-way Analysis of Variance (ANOVA) test for continuous variables. Significant factors in the ANOVA test were analyzed in the pairwise comparisons after Bonferroni correction. Fisher's exact test was used for the comparison of categorical variables, when 25% or more of the expected cell counts were below 5. The univariate correlation between UA metabolites (UA excretion and ΔUA/Cr) and PSG findings were assessed with Spearman correlation coefficients. Multiple linear regression analysis was used to assess the independent correlation between UA metabolites and PSG findings, after the adjustment for age, gender and the factors related to UA metabolites in the univariate analysis. Regression coefficients (beta) and 95% CIs were reported. Normality distribution of UA metabolites was tested by Kolmogorov–Smirnov test. For ΔUA/Cr values, exclusion of an outlier provided the normal distribution in the linear regression analysis. Change in UA metabolites after 5 years was analyzed by Wilcoxon Signed Ranks test. The correlation between mortality and baseline UA metabolites was tested by using Mann–Whitney U test. A two-sided p value less than 0.05 was considered to be statistically significant in all comparisons. SPSS 11.0 program was used for the statistical analysis.

Patients were divided into four treatment groups and one control group: 15 patients with nocturnal hypoxemia (NH)-COPD, 15 patients with nocturnal normoxemic (NN)-COPD, 16 patients with NN-OSAS, 15 patients with NH-OSAS, and 14 control subjects. Of the 75 study subjects, 54 were male (72%). There were 21 current smokers and 21 ex-smokers. Characteristics of the disease groups and PSG findings are shown in Table 1. Differences according to gender, age, smoking status and BMI, AHI, T90%, UA excretion and ΔUA/Cr were found to be statistically significant among the patient groups.

UA excretion significantly correlated with age (correlation coefficient: 0.37, p<0.01), BMI (correlation coefficient: 0.28, p<0.05), T90% (correlation coefficient: 0.57, p<0.01), nocturnal mean SaO2 (correlation coefficient: −0.54, p<0.01) and nocturnal SaO2 nadir (correlation coefficient: −0.39, p<0.01). ΔUA/Cr correlated with AHI (correlation coefficient: 0.36, p<0.01), T90% (correlation coefficient: 0.52, p<0.01), nocturnal mean SaO2 (correlation coefficient: −0.39, p<0.01) and nocturnal SaO2 nadir (correlation coefficient: −0.42, p<0.01). Table 2 shows the correlation between UA metabolites and PSG variables.

Regression analysis revealed significant correlation between COPD-NH and UA excretion (beta coefficient: 0.28, 95%CI: 0.16–0.40). T90% was significantly correlated with UA excretion in the model that included AHI, age, gender and BMI (beta coefficient: 0.005, 95%CI: 0.003–0.007). AHI was not significantly associated with UA excretion in the model. The correlation between UA excretion and disease groups, and PSG findings adjusted for age, gender and BMI are shown in Table 3.

Regression analysis revealed that OSAS-NH and COPD-NH were significantly correlated with ΔUA/Cr (OSAS-NH, beta coefficient: 55.5, 95%CI: 33.7–77.4; COPD-NH, beta coefficient: 43.7, 95%CI: 21.6–65.9). In the model including AHI, age, gender and BMI, T90% displayed significant correlation with ΔUA/Cr (beta coefficient: 0.005, 95%CI: 0.003–0.007). AHI did not significantly correlate with ΔUA/Cr in the model. The correlation between ΔUA/Cr and disease groups, and PSG findings adjusted for age, gender and BMI are shown in Table 4.

After five year follow-up, 58 subjects could be contacted while 7 could not be contacted, and 10 had deceased (6 in COPD-NH, 3 in COPD-NN and 1 in OSAS-NH group). Of these 58 subjects, 35 provided blood and urine sample for UA measurements (5 each in control, COPD-NN and COPD-NH groups; 9 in OSAS-NN and 11 in OSAS-NH group). Of the 24 patients, who were offered NIMV, 11 were using NIMV (0/3 in COPD-NN, 3/3 in COPD-NH, 3/9 in OSAS-NN and 5/11 in OSAS-NH). There was no significant difference in terms of age, gender, BMI, AHI, T90%, UA excretion and ΔUA/Cr between the patients who could be contacted and those who could not be contacted. Eight patients in the baseline assessment and 11 patients in the second assessment had physician-diagnosed hypertension. Uric acid metabolites did not significantly differ between the patients with and without hypertension.

Among the 11 patients, who reported NIMV usage, UA excretion after NIMV was significantly lower than the baseline measurement. There was no significant difference after the 5-year period in the other patients, who did not use NIMV or were not offered NIMV. Table 5 shows the variation in UA metabolites between the baseline measurement and the measurement after 5 years.

Mortality significantly correlated with baseline UA excretion as shown in Fig. 2. Median and IQR range were 0.79 (0.51–0.89) and 0.41 (0.31–0.55) in the 10 deceased and 58 surviving patients, respectively (p: 0.001). However, mortality did not significantly correlate with ΔUA/Cr which was shown in Fig. 3.

Figure 2.

Association between the baseline UA excretion and mortality.

(0.06MB).

Figure 3.

Association between the baseline ΔUA/Cr and mortality.

(0.05MB).

Among the 10 deceased patients, 9 died from respiratory causes, and for 1 patient the cause of death could not be ascertained. Mortality was found to be correlated with age, smoking, T90% and nocturnal mean SaO2. After the adjustment for age, mortality displayed significant correlation with only nocturnal SaO2% mean (OR: 0.82, 95%CI: 0.69–0.97) and UA excretion (OR: 231.0, 95%CI: 3.5–1560.3) (Table 6).

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that they have followed the protocols of their work center on the publication of patient data.

The authors have obtained the written informed consent of the patients or subjects mentioned in the article. The corresponding author is in possession of this document.