New Predictive Equation for Optimal Continuous Positive Airway Pressure in Adult Patients with Obstructive Sleep Apnea

Objective: This study aimed to develop and validate a new continuous positive airway pressure (CPAP) prediction equation and compare it with other formulas.Material and Methods: We retrospectively included patients with obstructive sleep apnea who underwent a CPAP titration study between January 2012 and December 2016. All clinical and polysomnographic data were collected. The new prediction equation was developed using the first data set, and the predictability performance was validated using the second data set.Results: Among the 266 enrolled patients, 73.7% were male, and the mean body mass index (BMI) was 30.8±7.4 kg/m2 . Five variables, namely age, BMI, neck circumference (NC), apnea–hypopnea index (AHI), and minimum pulse oxygen saturation (Min SpO2 ), highly correlated with the optimal titration pressure, and were therefore included in the equation, as stated below:Predicted pressure (cm H2 O) = 2.26 + (0.02xAge) + (0.04xBMI) + (0.11xNC) + (0.04xAHI) - (0.04xMin SpO2 )This equation accounted for 54.4% of the variance in predicting the optimal titration pressure (R2 =0.544, p-value <0.001). Its optimal estimation was 62.0% in the validated group. The equation-derived predicted pressure correlated with good agreement with the laboratory-derived optimal titration pressure (r=0.70, 95% CI=0.6335–0.755, p-value<0.001) according to Bland–Altman analysis. Conclusion: Our equation is highly consistent with the CPAP titration study in predicting fixed CPAP pressure, and is thereby beneficial for sleep technicians in establishing a starting pressure for such studies at a sleep laboratory.

843 Paradoxical Waking Hypoxemia that Improves with Sleep

Introduction We present a case of paradoxically worsened hypoxia during wake phase of polysomnography while undergoing a CPAP titration study. Nighttime hypoxemia is a common feature in obstructive sleep apnea, due to obstructive events that manifest while sleeping. Excluding OSA, there remains an extensive differential for disease processes that cause hypoxemia while asleep; however, none of these processes can explain waking hypoxemia that improves upon sleeping. Report of case(s) A 70 year old male with severe OSA diagnosed by home sleep test (REI 46.5, nadir O2=76%) underwent polysomnography with PAP titration and demonstrated several hours of interrupted sleep without hypoxia and minimal obstructive events on CPAP 9–13 cmH2O. During the study, while awake at CPAP of 14 cmH2O, he developed hypoxia to mid-high 80s and supplemental oxygen bleed in was added starting at 3L and increased to 5L during a prolonged period of wakefulness. On CPAP 15 cmmH2O with 5L bleed-in, the patient fell asleep and oxygen saturation again increased to low 90s. He underwent an extensive workup for other cardiopulmonary causes of hypoxemia, with pulmonary function testing showing moderate obstructive ventilatory defect and mild DLCO impairment. An echocardiogram with saline contrast bubble study was relatively unremarkable, without evidence of right to left shunting. He underwent a chest CTA which was negative for pulmonary embolism, though it did reveal an enlarged pulmonary artery consistent with pulmonary hypertension. His chronic hypoxemia was treated with 2L supplemental oxygen during the day and bleed-in with CPAP at night. Conclusion Though nocturnal hypoxemia is common with OSA, polysomnography with paradoxical hypoxemia during wake phase has not been reported. Notably, the patient was without prolonged hypoxia during his sleep phase while on CPAP treatment with minimal apneic/hypopneic events. Pulmonary hypertension can also present as nocturnal hypoxemia, but it should worsen with sleep, rather than improve. There are case reports of right to left shunting worsened by PAP, though his hypoxemia persisted despite PAP. His paradoxical worsening hypoxemia with wakefulness is still unexplained. Support (if any):

838 Not All Hypercapnia is COPD

Introduction Sleep breathing disorder related to isolated unilateral or bilateral diaphragmatic dysfunction (DD), in the absence of a generalized neuromuscular disorder, is not well understood and often under-recognized. There have been only a few cases reported of apneas and hypoponeas during REM sleep due to diaphragmatic dysfunction. We present here, a case of an 62 year-old man who developed acute hypercapnic respiratory failure with presumed COPD exacerbation requiring invasive ventilation. Report of case(s) A 62-year-old man was found on the sidewalk extremely short of breath and was intubated in the emergency department. Initial post-intubation arterial blood gas showed pH 7.1, pCO2 82, pO2 263, CO3 25.5. CXR showed no infiltrates. Echocardiography showed EFof 55%-65%. Long-term tobacco use supported the picture of COPD exacerbation. However, PFT was within normal limits. HSAT one year prior which showed severe OSA with AHI 52.6 event per hour. Patient had not pursued positive airway pressure (PAP) titration study afterward. In ICU, he was treated for presumed COPD exacerbation and successfully weaned off invasive ventilation. Inpatient PAP titration study recommended IPAP 12 and EPAP 8 cm H20. A fluoroscopy of the diaphragm was performed and showed that the right diaphragm had limited mobility. Electromyogram did not show generalized myopathy. Conclusion In conclusion, this case report describes the presentation of sleep disordered breathing seen in patients with unilateral diaphragmatic palsy. In these patients, the respiratory events seen are mainly hypopneas and desaturations, worse in REM sleep and supine position. This was an unusual presentation of a patient with untreated OSA and unilateral diaphragmatic palsy. A characteristic finding in these patients is worsening of the OSA in supine position. This has been reported in several studies and was seen in our case as well. This case underscores the need for critical thinking and diagnostic reasoning in the evaluation of a patient with hypercapnic respiratory failure and consider a wide differential and not only COPD exacerbation as the cause. Unilaterally diaphragmatic palsy is a rare cause of hypercapnic respiratory failure but must be considered when seen with obstructive sleep apnea with predominantly hypopneas and hypoxemia out of proportion of the respiratory events. Support (if any):

572 Utility of Split Night Polysomnograms in Children

Introduction Overnight in-lab polysomnograms (PSG) are the gold standard for diagnosis of sleep disordered breathing in children. As the wait time for adenotonsillectomy (T & A) at our institution was several months, we implemented split night PSGs with positive airway pressure (PAP) during the initial diagnostic PSG if AHI&gt;30 (Emergency SNPSG). Planned SNPSG were performed on children who are undergoing PSG after T & A, eliminating follow-up titration PSG if the PSG is positive for OSA (residual AHI&gt;10/hour). We present data on the outcomes of the SNPSG. Methods Retrospective chart review of consecutive SNPSGs done over last 2 years at our institution was performed. Data on SNPSGs (planned or emergency), age, sex, diagnostic study duration, diagnostic study AHI, PAP pressure and subsequent PAP adherence were collected. Data on sleep efficiency, arousal, sleep architecture, REM sleep were compared between diagnostic and titration part of the SNPSG. Study was considered successful if patient was able to tolerate PAP during titration and also if adherent to PAP at follow up. Results 48 studies met the criteria for SNPSG, with 60% of SNPSG being emergency SNPSG with AHI&gt;30. Our cohort’s age ranged from 2–18 years (median age 8 years); 33 were males. Majority of the emergency SNPSG were in younger children (80% &lt; 5 years), 75% of them continued to use PAP (mostly CPAP) until T & A with wait time being more than 3 months. Average wait time for T & A was 4 months. 25% of patients continued to use PAP following T & A as residual AHI was &gt;10/hour. 98% of the patients were able to tolerate PAP during titration section with restoration of sleep architecture including REM with reduction in AHI, arousals and improved sleep efficiency. Bilevel PAP was used in 10% of patients in mostly planned SNPSG. Conclusion SNPSG can be implemented with fair degree of success during initial PSG with PAP used until T & A is performed. Planned SNPSG are also useful with residual severe OSA eliminating need for subsequent titration study. When indicated, 75% of our cohort continued to use PAP with fair adherence (&gt;70% nightly use) following SNPSG. Support (if any):

826 Central Sleep Apnea associated with Sodium Oxybate– A Case Series

Introduction Sodium oxybate (SO) is indicated to treat cataplexy and excessive daytime sleepiness (EDS) in patients with narcolepsy. Only a handful of cases have been reported of new-onset Central Sleep Apnea (CSA) in the setting of SO use. We present 3 patients who developed CSA in the setting of use of SO. Report of case(s) Patient 1: A 25-y/o man presented with hypersomnolence. His diagnostic polysomnogram (PSG) showed moderate Obstructive Sleep Apnea (OSA), and he was placed on Continuous Positive Airway Pressure (CPAP) therapy. Due to persistent hypersomnia in the setting of effectively treated OSA, he had a Multiple Sleep Latency Test (MSLT), which revealed pathological sleepiness with a mean latency of 3.8 minutes with a sleep-onset REM on the overnight polysomnogram. SO was started for clinical diagnosis of Narcolepsy after he failed other stimulant medications. Hypersomnolence improved though data from his PAP device, home sleep studies, re-titration studies performed when he was on SO demonstrated CSA following 1st or 2nd dose of SO. Patient 2: A 17-y/o man was diagnosed to have Narcolepsy with Cataplexy, based on PSG followed by MSLT. 20 years later, he was diagnosed with OSA based on a PSG and was treated with CPAP. A few years later, he was started on SO for fragmented sleep and EDS. A home sleep study performed when he was on SO, revealed CSA. Later, an in-lab titration study showed CSA with Cheyne-Stokes respiration (CSR), treated with Adaptive Servo-Ventilation (ASV) therapy. Patient 3: A 15-y/o man initially presented after several cataplectic episodes and was diagnosed with Narcolepsy with Cataplexy. His initial PSG showed no evidence of sleep-disordered breathing. A few years later, for persistent cataplectic events, he was started on SO with improvement in the episodes’ frequency. Several years later, a baseline PSG demonstrated OSA and CSA, with frequent CSA events soon after taking SO. The CPAP titration study, performed following the PSG, also revealed frequent CSA following the second dose of SO. Conclusion Close monitoring is warranted with SO use, given some narcolepsy patients’ predisposition to develop CSA. Follow-up studies are needed to address the pathogenesis and management strategies. Support (if any) None