We thank Dr. Kapoor for calling to attention potential discrepancy between the apnea-hypopnea index (AHI) on the continuous positive airway pressure (CPAP) compliance report and the oxygen desaturation index (ODI) on the overnight pulse oximetry (OPO).1
Most of the current CPAP machines detect the cessation or decrease in airflow for 10 seconds or longer and record apneas and hypopneas accordingly. It is important to bear in mind that residual AHI on compliance reports does not incorporate arterial oxygen desaturations or arousals. For example, respiratory events may be scored by CPAP but not by manual scoring of polysomnography (PSG) results because of the absence of desaturations or arousals.2 On the other hand, hypopneas associated with oxygen desaturation can be manually scored from PSG results with modest reduction in flow which may not meet the threshold for detection by a CPAP machine.2 Despite its limitations, an AHI < 10 events/h measured by a CPAP machine is usually associated with good treatment efficacy.2,3
The physiologic effects of decreased airflow on oxygen saturation depends on the baseline saturation. Episodes of airflow reductions that are not large enough to be detected by the CPAP machine may still result in a 3% or more oxygen desaturation if baseline saturation is close to the “slippery slope” of the hemoglobin oxygen dissociation curve as in the case described where mean saturation was 92%. The clinical consequences of such respiratory events may vary among patients; therefore, management is probably best individualized on a case-by-case basis. In a case of discrepancy between residual AHI and ODI, a repeat study for retitration may be warranted in a patient with uncontrolled signs or symptoms of sleep apnea, while observation may be appropriate in an asymptomatic patient. With respect to nocturnal hypoxemia (time spent below 90% saturation), oxygen therapy can be considered when hypoxemia is disproportionately low for the residual sleep apnea, possibly related to underlying chronic cardiopulmonary disorder.
Although ODI ≥ 10 events/h on an OPO is an acceptable cutoff for detecting AHI ≥ 15 events/h,4,5 further research is needed to clarify ODI cutoffs in those with lower mean saturation, as relatively smaller drops in arterial oxygen partial pressure are expected to result in greater changes in oxygen saturation. Future studies are needed to evaluate the diagnostic accuracy and clinical utility of OPO for monitoring the efficacy of sleep apnea treatment.
DISCLOSURE STATEMENT
Dr. Ayache reports no conflicts of interest. Dr. Strohl reports consulting for Inspire Medical Systems, Sommetrics, and Galvani Bioelectronics. He reports no conflicts of interest relevant to this letter.
CITATION
Ayache M, Strohl KP. Diagnostic accuracy and clinical utility of overnight pulse oximetry. J Clin Sleep Med. 2018;14(8):1439.
REFERENCES
- 1.Kapoor M. Discrepancy between oxygen desaturation index and apneahypopnea index: what do you do with the results? J Clin Sleep Med. 2018;14(8):1437. doi: 10.5664/jcsm.7296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Berry RB, Kushida CA, Kryger MH, Soto-Calderon H, Staley B, Kuna ST. Respiratory event detection by a positive airway pressure device. Sleep. 2012;35(3):361–367. doi: 10.5665/sleep.1696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li QY, Berry RB, Goetting MG, et al. Detection of upper airway status and respiratory events by a current generation positive airway pressure device. Sleep. 2015;38(4):597–605. doi: 10.5665/sleep.4578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Magalang UJ, Dmochowski J, Veeramachaneni S, Draw A, Mador MJ, El-Solh A, Grant BJ. Prediction of the apnea-hypopnea index from overnight pulse oximetry. Chest. 2003;124(5):1694–1701. doi: 10.1378/chest.124.5.1694. [DOI] [PubMed] [Google Scholar]
- 5.Chung F, Liao P, Elsaid H, Islam S, Shapiro CM, Sun Y. Oxygen desaturation index from nocturnal oximetry: a sensitive and specific tool to detect sleep-disordered breathing in surgical patients. Anesth Analg. 2012;114(5):993–1000. doi: 10.1213/ANE.0b013e318248f4f5. [DOI] [PubMed] [Google Scholar]