Nocturnal Hypoxemia in Patients with Cancer

Article Outline

• References
• Copyright

To the Editor:

In a survey of 100 palliative care inpatients with a variety of cancers, we found that 35 experienced nocturnal hypoxemia, as defined by an oxygen saturation (SaO₂) <90% for ≥2% of the monitored nighttime, together with higher levels of mental fatigue.¹ When considering possible etiologic factors, we had excluded patients with a daytime SaO₂ <90% or known obstructive sleep apnea. Those with nocturnal hypoxemia were more likely to have pulmonary disease and lower values of percent predicted forced expiratory volume in 1 second and day SaO₂. There was no difference in performance status, body mass index, sniff nasal inspiratory pressure, and opioid or other sedative drug use between those with and without nocturnal hypoxemia.¹ However, sleep per se can reduce ventilation.² This does not normally lead to a significant change in the oxygen content of blood because of the horizontal portion of the oxygen dissociation curve. Nonetheless, patients with a lower day SaO₂ will lie closer to the steep part of the oxygen dissociation curve and a reduction in ventilation with sleep could then lead to significant hypoxemia. To explore this, we have further examined our data.

A wrist oximeter (Pulsox 3i, Minolta, NJ) recorded daytime SaO₂ after 20 minutes resting in both a sitting and a lying position, and the lower value used in the analysis. Mean nocturnal SaO₂ was assessed over a single night, using a minimum period of five hours of sleep as judged by an absence of movement artefact. For those patients exhibiting a fall between day and mean night SaO_2, the difference was plotted against day SaO₂. SaO₂ values were also converted to an estimated partial pressure of oxygen in arterial blood (PaO₂) using an equation that assumes a standard oxygen dissociation curve.³ Spearman's rank correlation coefficient was used to examine relationships.

There was a fall between day and mean night SaO₂ in 47 patients, representing 24 (69%) and 23 (35%) of those categorized with and without nocturnal hypoxemia, respectively. The mean (standard deviation) day SaO₂ was lower in those with nocturnal hypoxemia compared with those without (94 [2] vs. 96 [1], P<0.001). There was no significant correlation between fall at night and day SaO₂ (R²=0.001, P=0.84) (Fig. 1a). This most likely reflects the narrow range we studied; others finding a relationship included patients with day SaO₂ levels as low as 70% and night falls as high as 20%.² Nonetheless, the relationship between the night fall in PaO₂ and night fall in SaO₂ was such that, across a range of values of PaO₂, the change in SaO₂ was generally greater for the group with nocturnal hypoxemia, which had lower day SaO₂ levels (Fig. 1b).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  Mean night fall in SaO₂ vs. a) day SaO₂ and b) fall in night PaO₂ (estimated) for patients with (open circles) and without (closed circles) nocturnal hypoxemia.

In conclusion, a sleep-related reduction in ventilation could be sufficient to explain the falls in night PaO₂ and SaO₂ seen in patients with cancer, with those with a lower day SaO₂ level more likely to experience a degree of nocturnal hypoxemia associated with impaired mental functioning. Further work, including more detailed sleep studies, is required to confirm our findings.

Back to Article Outline

References 

1. Wilcock A, England R, El Khoury B, et al. The prevalence of nocturnal hypoxemia in advanced cancer. J Pain Symptom Manage. 2008;36:351–357
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (160 KB)
   • CrossRef
2. Stradling JR, Lane DJ. Nocturnal hypoxaemia in chronic obstructive pulmonary disease. Clin Sci. 1983;64:213–222
   • View In Article
3. Kelman GR. Digital computer subroutine for the conversion of oxygen tension into saturation. J Appl Physiol. 1966;21:1375–1376
   • View In Article