Article Text


Improving delivery of care in cystic fibrosis
P237 Does desaturation on exercise predict nocturnal hypoxia in children with cystic fibrosis?
  1. J Keating1,
  2. N Collins1,
  3. M Rosenthal1,
  4. A McKee1,
  5. D Urquhart2
  1. 1Royal Brompton Hospital, London, UK
  2. 2Royal Hospital for Sick Children, Edinburgh, UK


Introduction Exercise induced arterial hypoxia (EIAH) is well documented in cystic fibrosis (CF); though its relationship with nocturnal hypoxia remains unclear. In adult CF, hypoxia occurs more frequently during sleep than exercise,1 2 but this has not been investigated in children.

Aim To identify whether EIAH is a reliable predictor of nocturnal hypoxia in CF children.

Methods Retrospective case note review of 34 children with CF. All performed the modified shuttle test,3 overnight pulse oximetry and spirometry in the second week of an elective admission for intravenous antibiotics. EIAH was defined as drop in SpO2 ≥4% from baseline.4 There is no agreed definition of nocturnal hypoxia, we assessed numerous overnight variables, mean and lowest SpO2 and % time with SpO2 <92% in sleep. Data were analysed using SPSS software (SPSS).

Results 34 subjects were assessed, median age 12 (range 7–16) years, and FEV1 72 (25–103) % predicted. Nine demonstrated EIAH and five had nocturnal desaturation (SpO2 <92% >10% sleep time). All children with SpO2 <92% for >10% of sleep also had EIAH, this was significant (p=0.0086). The relationship between nocturnal hypoxia and lowest exercise SpO2 (p=0.0076) was also significant. However, 6 of those with EIAH were not hypoxic overnight. The sensitivity of EIAH detecting sleep hypoxia was 75% and specificity 80%. Positive predictive value (PPV) was 33% and negative predictive value 96%.

Conclusion The PPV of EIAH for detecting nocturnal hypoxia is poor, but the definition of EIAH must be questioned. SpO2 varying from 90% to 87% does not meet EIAH criteria, yet a fall from 97% to 93% does. We must also note that although the relationship between nocturnal hypoxia and EIAH appears significant the number of subjects in this group is extremely small. We conclude EIAH is not a reliable predictor of nocturnal desaturation and advocate a sleep study to confidently detect sleep hypoxia.

Abstract P237 Table 1

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.