Nocturnal oxygen desaturation and spirometric parameters in adults with cystic fibrosis.

BACKGROUND Correction of nocturnal hypoxaemia in patients with cystic fibrosis may delay the development of pulmonary hypertension. Descriptive statistics used for nocturnal arterial oxygen saturation (SpO2) lack uniformity. The relationship between SpO2 and spirometric parameters has not previously been explored in a large number of exacerbations in adult patients with cystic fibrosis. METHODS Over a 21 month period overnight SpO2, forced expiratory volume in one second (FEV1), and forced vital capacity (FVC) were recorded on admission and discharge in 120 treatments of pulmonary exacerbations in 47 patients with cystic fibrosis who did not receive supplemental oxygen during recording. Nocturnal SpO2 was related to spirometric parameters for the whole group and individually in 11 patients, each of whom had at least five treatments. RESULTS There was a close linear relationship between the percentage of the recording spent with SpO2 < 90% and mean overnight SpO2. Mean SpO2 correlated moderately with percentage predicted FEV1(%FEV1), r = 0.6, and poorly with percentage predicted FVC (%FVC), r = 0.34. The relationship between mean SpO2 and % FEV1 was non-linear at mean SpO2 < 89%, but approximated to linearity above this value. After exclusion of treatments with mean SpO2 < 89% the regression relationship between mean SpO2 and %FEV1 was the same on admission and discharge. Individual correlation coefficients of mean SpO2 versus % FEV1 in the 11 patients with repeated treatments ranged from 0.57 to 0.77. The slopes of the regression lines did not differ, with a pooled slope of 0.116, but the intercepts varied widely. CONCLUSIONS In patients with cystic fibrosis mean overnight SpO2 can be substituted for percentage of recording < 90%. The relationship between mean SpO2 and percentage predicted FEV1 is non-linear at low values of SpO2 and is not influenced by treatment of pulmonary exacerbations. Patients with cystic fibrosis desaturate at a uniform rate compared with percentage predicted FEV1, but the value of FEV1 at which desaturation first occurs varies between patients. The spirometric values do not accurately predict nocturnal desaturation in a cystic fibrosis population, but FEV1 is a useful guide in individual patients with moderate desaturation.