A 22-year-old male college student with prior diagnoses of mild obstructive sleep apnea, delayed sleep-wake phase disorder, and inadequate sleep hygiene presented to the sleep clinic for an evaluation of ongoing daytime hypersomnolence despite the use of mixed dextroamphetamine/amphetamine salts extended-release preparation (Adderall XR) 20 mg two times a day prescribed by his primary care provider for sleepiness.
He reported going to bed at 10:00 PM, falling asleep quickly, sleeping throughout the night until 8:00 AM, and not napping during the daytime but feeling extremely sleepy with an Epworth Sleepiness Scale score of 20/24. He described waking with a dry mouth but did not endorse snoring, snort arousals, or headaches upon awakening. There was no history suggestive of leg movements at night, cataplexy, hypnagogic/hypnopompic hallucinations or sleep paralysis. On examination, the patient had a patent oropharynx (Friedman grade I), neck circumference of 45 cm, body mass index of 36.63 kg/m2, and normal blood pressure. Download from his continuous positive airway pressure (CPAP) device set at 11 cm H2O pressure showed consistent usage for the entire reported sleep duration, minimal air leak, and a low residual apnea-hypopnea index. To evaluate his hypersomnolence further, the patient was tapered off his stimulant medication and two-week actigraphy followed by overnight polysomnography and Multiple Sleep Latency Testing (MSLT) was ordered.
QUESTION: What is the possible explanation for the significantly low oxyhemoglobin saturation (SpO2) recorded through the entire polysomnogram (Figure 1)?
Hypnogram from overnight polysomnography.
The red arrow indicates oxyhemoglobin saturation. Mean saturation was 73% with a range of 66% to 78%. Hypopneas were scored when there was a further drop in oxyhemoglobin saturation from the baseline by ≥ 4% accompanied by a reduction in the nasal pressure signal by ≥ 30% for ≥ 90% of the event, each lasting ≥ 10 seconds. No apneas were noted during this study. The arousal index was 14.9 events/h and sleep efficiency was 93.6%. Sleep onset latency was 4.5 minutes and initial rapid eye movement latency was 89.5 minutes with 21% of the total time spent in rapid eye movement sleep. N1 = stage N1 sleep, N2 = stage N2 sleep, N3 = stage N3 sleep, PLM = periodic limb movement, R = rapid eye movement sleep, RERA = respiratory-effort related arousal, W = wake.
Hypnogram from overnight polysomnography.
ANSWER: The patient had a low-oxygen affinity variant called hemoglobin Denver, which resulted in markedly low SpO2.
Pulse oximetry is an integral part of polysomnography and utilizes spectrophotometry to determine the proportion of hemoglobin that is saturated with oxygen in peripheral arterial blood.1 It has the advantage of being noninvasive and provides the opportunity to monitor SpO2 data continuously.
Low SpO2 on pulse oximetry may reflect true hypoxemia. Alternatively, this could result from conditions such as methemoglobinemia and sulfhemoglobinemia that are associated with reduced SpO2 in the context of a normal arterial blood oxygen tension (PaO2).2,3 Venous pulsations secondary to venous congestion may result in artefactual low SpO2.4 Improper application of the oximeter probe and certain nail paints can also result in decreased SpO2 readings.5
Inherited forms of abnormal hemoglobin, such as hemoglobin Denver in our case, can result in low SpO2 despite a high fraction of inspired oxygen. Although there are more than 1,000 known hemoglobin variants, low-oxygen affinity variants that are associated with low SpO2 are infrequent.6,7 Patients with low-oxygen affinity variant hemoglobin may be completely asymptomatic or present with cyanosis but may otherwise be clinically well.6,8
Our patient's mother carried a diagnosis of hemoglobin Denver. Our patient, who was asymptomatic from the hematologic standpoint with no cyanosis on examination, had undergone evaluation by a hematologist prior to his sleep clinic referral; a DNA sequence analysis revealed that he too had this hemoglobin variant. Arterial blood gas showed a pH of 7.41, PaO2 of 99.8 mmHg, PaCO2 of 38.6 mmHg, oxygen saturation of 79.3%, bicarbonate level of 24 mEq/L, an A-a gradient of 0, and normal carboxyhemoglogin and methemoglobin levels.
The patient's actigraphy results demonstrated a regular sleep schedule with sufficient time spent in bed. Overnight polysomnogram consisting of a total sleep time of 474 minutes showed an apnea-hypopnea index of 3 events/h on CPAP and a periodic limb movement index of 3 events/h as well. Current scoring criteria9 were used to determine disordered breathing events on polysomnography, although it should be borne in mind that in a case such as this, there may be a greater degree of desaturation during a given disordered breathing event to a right-shift in the oxyhemoglobin dissociation curve. The patient was found to have a mean sleep latency of 3.25 minutes (range of 2.0 to 4.0 minutes) on MSLT conducted the next day over four test nap opportunities while utilizing CPAP at the appropriate setting. He experienced two sleep-onset rapid eye movement periods, during naps 1 and 4. Results of urine toxicology for prescription and over-the-counter drugs performed during the MSLT were negative, including for stimulants. The patient's diagnosis was narcolepsy type II; the dose of his Adderall XR was increased to 60 mg in divided doses and a follow-up appointment scheduled in one month.
SLEEP MEDICINE PEARLS
Low saturation on pulse oximetry could result from a hemoglobin variant
Low SpO2 in asymptomatic patients should raise the suspicion of a hemoglobin variant
Arterial blood gas analyses may help clarify the clinical picture
DNA sequence analysis can confirm the diagnosis of a variant hemoglobin
The technologist performing the polysomnography should be informed prior to the study of the anticipated low SpO2 (if known) and decision whether to try CPAP or add supplemental oxygen should not be based on the SpO2 alone
This work was performed at Mayo Clinic, Rochester, Minnesota. All authors have seen and approved the manuscript. No financial support was received for this work. None of the authors have any relevant conflicts of interest and off-label or investigational product use to declare.
Mansukhani MP, Kolla BP, Altchuler SI. Unusually low oxyhemoglobin saturation on polysomnography. J Clin Sleep Med. 2017;13(4):629–631
Grace RF. Pulse oximetry. Gold standard or false sense of security? Med J Aust. 1994;160(10):638–644. [PubMed]
Aravindhan N, Chisholm DG. Sulfhemoglobinemia presenting as pulse oximetry desaturation. Anesthesiology. 2000;93(3):883–884. [PubMed]
Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology. 1989;70(1):112–117. [PubMed]
Secker C, Spiers P. Accuracy of pulse oximetry in patients with low systemic vascular resistance. Anaesthesia. 1997;52(2):127–130. [PubMed]
Chan MM, Chan MM, Chan ED. What is the effect of fingernail polish on pulse oximetry? Chest. 2003;123(6):2163–2164. [PubMed]
Stabler SP, Jones RT, Head C, Shih DT, Fairbanks VF. Hemoglobin Denver [alpha 2 beta 2(41) (C7) Phe-- > Ser]: a low-O2-affinity variant associated with chronic cyanosis and anemia. Mayo Clin Proc. 1994;69(3):237–243. [PubMed]
Thom CS, Dickson CF, Gell DA, Weiss MJ. Hemoglobin variants: biochemical properties and clinical correlates. Cold Spring Harb Perspect Med. 2013;3(3):a011858. [PubMed Central][PubMed]
Bruns CM, Thet LA, Woodson RD, Schultz J, Hla KM. Hemoglobinopathy case finding by pulse oximetry. Am J Hematol. 2003;74(2):142–143. [PubMed]
Berry RB, Brooks R, Gamaldo CE, et al; for the American Academy of Sleep Medicine. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. Version 2.3. Darien, IL: American Academy of Sleep Medicine; 2016.