SaO2 Nadir Research Articles

Sleep and breathing on the first night after adenotonsillectomy for obstructive sleep apnea

Adenotonsillectomy (T&A) has established effectiveness for the treatment of obstructive sleep apnea (OSA). However, more than 20% of children with OSA have respiratory compromise requiring medical intervention in the postoperative period. The reasons for this complication are not well-defined. We aimed to compare the nature and severity of sleep-disordered breathing in children with mild and severe OSA on the first night following adenotonsillectomy. Ten children were classified into groups of mild and severe OSA, based on preoperative testing. On the first night after T&A, they underwent polysomnography, including electroencephalograph, submental electromyography, bilateral electro-oculograms, monitoring of respiratory movements, heart rate, ECG, and oxygen saturation. Sleep-disordered breathing was assessed by the apnea-hypopnea index, the SaO(2) nadir, and the desaturation index, including dips in saturation below 90% (DI(90)). Sleep quality was assessed by sleep efficiency, time spent in each sleep state, and respiratory arousal index. Obstructive events occurred postoperatively in all children, but were more frequent in those with severe OSA preoperatively: the median (interquartile range) mixed/obstructive apnea/hypopnea indicies were 6.9 (2.2-9.8) events/hr and 21.5 (15.1-112.1) events/hr for the mild OSA group and the severe OSA group, respectively (P = 0.009). Obstructive events were the major cause of desaturation during sleep postoperatively. Sleep quality was severely disrupted in both groups, with reductions in both slow-wave sleep and rapid eye movement sleep. In conclusion, despite removal of obstructing lymphoid tissue, upper airway obstruction occurred on the first postoperative night in children with OSA. This study is the first to demonstrate the mechanism of respiratory compromise after adenotonsillectomy, a common postoperative complication in children with severe OSA.

Adult Identified with Congenital Central Hypoventilation Syndrome–Mutation in PHOX2b Gene and Late-Onset CHS

In light of our recent review of the genetic basis for congenital central hypoventilation syndrome (CCHS) (1), we bring to your attention a 35-year-old adult recently identified as having had late-onset CHS and the CCHS-mutation in PHOX2b. One year ago this patient presented with a history of snoring “for his whole life,” nocturnal gasping, and “stopping breathing and turning blue.” During polysomnography he had 46 obstructive events (38 hypopneas, 8 apneas), 16 central apneas, and an apnea–hypopnea index of 77 events/hour, with an SaO2 nadir of 65% in room air and 80% with supplemental oxygen. He temporarily responded to Bi-Pap with supplemental oxygen during sleep, but within 6 months was referred for uvulopalatopharyngoplasty. He was readmitted 48 hours after discharge with intermittent confusion, difficulty staying awake, and respiratory failure (awake arterial Pco2 of 73 mm Hg). His admission hematocrit was 73%, body temperature was 35.5 C, chest X-ray revealed cardiomegaly, and ECG revealed right axis deviation, enlargement, and hypertrophy. He required intubation and mechanical ventilation, then a tracheostomy; however, mechanical ventilation was discontinued. We were contacted by the family when an arterial blood gas assay upon awakening revealed a Pco2 of 129 mm Hg (pH 7.03). Because of his daughter’s diagnosis of late-onset CHS 3 years before, and documentation that she is heterozygous for the CCHS-related PHOX2b polyalanine expansion mutation, we conducted the same assay with the father’s DNA. Figure 1 demonstrates the identical mutation in the 4-year-old daughter and the father, as well as in the 2-yearold daughter, who also has alveolar hypoventilation (all three have a genotype of 20/25). In retrospect, we learned that the father uses Dexadrine to “stay on task,” has symptoms compatible with autonomic nervous system dysregulation (low body temperature, postural hypotension, pinpoint pupils minimally reactive to light, no shortness of breath despite markedly increased carbon dioxide levels), and has awakened with headaches and periods of disorientation for several years. On the basis of this remarkably late presentation of the alveolar hypoventilation component of CHS, we suggest that adult pulmonologists and intensivists consider the possibility of lateonset CHS, confirmable by PHOX2b assay, when caring for adults with hypercarbia of unknown etiology. From a scientific perspective, this case raises the possibility of a regulatory gene that modifies the expression or activity of PHOX2b, determining the severity of the autonomic dysregulation in late-onset CHS.

Recurrent hypoxemia in young children with obstructive sleep apnea is associated with reduced opioid requirement for analgesia.

Obstructive sleep apnea (OSA) in children is often associated with recurrent hypoxemia during sleep. In developing animals, central opioid neuropeptide content is high, and opioid receptors are up-regulated after recurrent hypoxia. The authors hypothesized that children with recurrent hypoxemia due to OSA might have altered central opioid functionality that could affect their responsiveness to opioid drugs. Using a retrospective database, we assessed the relation of age and preoperative oxygen saturation to the cumulative postoperative morphine dose administered for analgesia in children with OSA undergoing adenotonsillectomy. Inclusion criteria were (1) adenotonsillectomy for OSA; (2) no concomitant pathology; (3) intraoperative administration of short-acting opioid drugs; (4) endotracheal extubation on awakening in the operating room; and (5) morphine as the parenteral, postoperative analgesic. Forty-six children (16 girls) fulfilled the inclusion criteria. Age and preoperative arterial oxygen saturation (SaO2) nadir, either individually (P = 0.023, P = 0.0003, respectively) or in combination (P = 0.00009), exhibited a significant correlation to the morphine dose required for analgesia. Four of these children, aged 26.5 +/- 13.2 months, with a preoperative SaO2 nadir of 70.3 +/- 12.9%, did not require any postoperative morphine for analgesia at all. The authors speculate that the reduced morphine requirement for analgesia in children displaying oxygen desaturation associated with severe OSA may be related to their young age and to an up-regulation of central opioid receptors consequent to recurrent hypoxemia. In evaluating OSA in children, preoperative determination of the SaO2 nadir is important for predicting the postoperative opioid dosage required for analgesia.

Children with severe OSAS who have adenotonsillectomy in the morning are less likely to have postoperative desaturation than those operated in the afternoon.

To determine, in a subset of children previously reported, if the time of day when adenotonsillectomy for severe obstructive sleep apnea syndrome (OSAS) was performed affected the incidence of postoperative respiratory complications. Children having adenotonsillectomy were included if they had a polysomnographic diagnosis of severe OSAS within six months prior to operation. Patients who met the inclusion criteria were grouped by the occurrence of postoperative desaturation into a saturated (SAT) and desaturated (deSAT) group. The charts of children in group deSAT were reviewed. The clock time of the surgical procedure was recorded and categorized as morning (AM) or afternoon (PM). Eighty-eight patients met the inclusion criteria. There were 31 girls and 57 boys. The mean +/- SD age (yr) and weight (kg) were 4.6 +/- 2.9 yr and 20.8 +/- 14.5 kg respectively. There were 63 children in the SAT group and 25 in the deSAT group. Differences in age, weight and gender were not significant. The preoperative oxygen saturation (SaO2) nadir for the SAT and deSAT groups was 80.8 +/- 10.2% and 67.6 +/- 17.5% (P < 0.05) respectively. The preoperative obstructive apnea and hypopnea index was 15.8 +/- 10.2 and 35.7 +/- 34.6 events.hr(-1) (P < 0.05), respectively. Surgery in 63 (71.6%) children was performed in the AM. Univariate logistic regression identified PM surgery [odds ratio (OR) 4.6, 95% confidence interval (CI) 1.7 to 12.6, P = 0.002] and a preoperative SaO2 nadir < 80% (OR 3.6, 95% CI 1.4 to 9.4, P = 0.009) as risk factors predicting postadenotonsillectomy desaturation. Children with severe OSAS whose surgery is performed in the AM are less likely to desaturate following adenotonsillectomy than children whose surgery is performed in the PM.

Blood pressure in sleep disordered breathing

Aims: To investigate blood pressure (BP) in children with sleep disordered breathing (SDB). Methods: BP was measured during single night polysomnography in 23 suspected SDB child patients with adenotonsillar hypertrophy,...

Oxygen-conserving effects of apnea in exercising men.

We sought to determine whether apnea-induced cardiovascular responses resulted in a biologically significant temporary O(2) conservation during exercise. Nine healthy men performing steady-state leg exercise carried out repeated apnea (A) and rebreathing (R) maneuvers starting with residual volume +3.5 liters of air. Heart rate (HR), mean arterial pressure (MAP), and arterial O(2) saturation (Sa(O(2)); pulse oximetry) were recorded continuously. Responses (DeltaHR, DeltaMAP) were determined as differences between HR and MAP at baseline before the maneuver and the average of values recorded between 25 and 30 s into each maneuver. The rate of O(2) desaturation (DeltaSa(O(2))/Deltat) was determined during the same time interval. During apnea, DeltaSaO(2)/Deltat had a significant negative correlation to the amplitudes of DeltaHR and DeltaMAP (r(2) = 0.88, P < 0.001); i.e., individuals with the most prominent cardiovascular responses had the slowest DeltaSa(O(2))/Deltat. DeltaHR and DeltaMAP were much larger during A (-44 +/- 8 beats/min, +49 +/- 4 mmHg, respectively) than during R maneuver (+3 +/- 3 beats/min, +30 +/- 5 mmHg, respectively). DeltaSa(O(2))/Deltat during A and R maneuvers was -1.1 +/- 0.1 and -2.2 +/- 0.2% units/s, respectively, and nadir Sa(O(2)) values were 58 +/- 4 and 42 +/- 3% units, respectively. We conclude that bradycardia and hypertension during apnea are associated with a significant temporary O(2) conservation and that respiratory arrest, rather than the associated hypoxia, is essential for these responses.

Post–Apneic Inhalation Reverses Apnea-Induced Sympathoexcitation Before Restoration of Blood Oxygen Levels

Sleep apneas acutely increase sympathetic nerve activity (SNA) and thus arterial blood pressure. We hypothesized that after apnea, sympathoexcitation decreases before recovery of blood oxygen levels because of the predominant inhibitory effect respiratory factors exert over sympathetic nervous system activation. Seven healthy subjects were instrumented for arterial oxygen saturation (pulse oximetry, SaO2), leg muscle SNA (microneurography), and arterial pressure (Finapres). Supine subjects breathed 12% oxygen, 3% carbon dioxide, and 85% nitrogen for one min prior to apnea at the end of a normal tidal expiration. We accounted for circulatory delay in SaO2 measurement (5.4 +/- 0.4 s, mean +/- SE) as the time from the termination of apnea to the midpoint of the nadir of SaO2. SaO2 decreased to average 84 +/- 3% over the final 10 seconds of apnea, and recovered only partially to average 87 +/- 3% over the 10 seconds immediately following apnea. End-expiratory apnea increased SNA 14-fold from baseline levels of 217 +/- 37 units/10 seconds to 3063 +/- 442 units/10 seconds. However, SNA decreased to 93 +/- 32 units/10 s during the first 10 seconds after apnea. These findings indicate that sympathoinhibitory effects of respiratory signals, either lung inflation receptors or central respiratory inputs, predominate over sympathoexcitatory inputs from chemoreceptors to produce immediate and complete sympathoinhibition at the termination of a voluntary apnea. Arterial baroreflexes probably also contribute to sympathoinhibition after apnea.

The Efficacy of Oral Appliances in the Treatment of Persistent Sleep Apnea After Uvulopalatopharyngoplasty

Twenty-four patients who failed uvulopalatopharyngoplasty (UPPP) for obstructive sleep apnea (OSA) had an adjustable oral (Herbst) appliance made to treat the persistent apnea. Six patients discontinued the device prior to sleep evaluation. Eighteen patients had polysomnographic evaluations at baseline, post-UPPP, and with the Herbst appliance in place. The apnea-hypopnea index baseline (AHI) and arterial oxygen saturation (SaO2) nadir were 42.3+/-6.1 and 83.6+/-1.8%, respectively. There was no significant change in either parameter with surgery. With the oral appliance, the AHI fell to 15.3+/-4.4 (p < or = 0.01) and the SaO2 nadir increased to 87.9+/-1.2% (p < or = 0.05). Ten of the patients had control of the OSA with the Herbst appliance with a fall in the AHI to < 10. There were, in addition, two partial responders as defined by an AHI of <20 and a >50% fall in AHI compared with baseline and post-UPPP values. All but one of the responders and partial responders had complete resolution of subjective symptoms of daytime sleepiness with the appliance. An adjustable oral appliance appears to be an effective mode of therapy to control OSA after an unsuccessful UPPP.

Posthyperventilation Hypoxemia after Methacholine Inhalation.

The hypothesis of this study was that hypoxemia after methacholine (MTH) inhalation is related not only to ventilation/perfusion inhomogeneity, but also to posthyperventilation hypoxemia. To test the hypothesis, we paid special attention to changes in gas exchange and ventilation parameters after MTH inhalation. Six stable asthma patients were investigated, and SaO2, minute ventilation (V.E), oxygen uptake rate in the lung (V.O2), carbon dioxide output rate in the lung (V.CO2), and respiratory exchange ratio (R) were measured. The SaO2 level decreased from a baseline level (before MTH inhalation) of 96.8 +/- 1.0% (mean +/- SD) to the lowest level (the nadir SaO2) of 89.8 +/- 2.1% (p < 0.01) in 200 +/- 50 s after MTH inhalation and gradually increased toward the baseline level. V.CO2 increased just after MTH inhalation (post-MTH) with increased V.E, and decreased at the nadir SaO2 with baseline V.E and PaCO2, indicating a decrease in breath-by-breath V.A and an increase in dead space minute ventilation at the nadir SaO2, but V.O2 remained close to constant. R increased post-MTH, decreased at the nadir SaO2, and thereafter increased gradually toward the baseline level with a time constant of 5.6 min. The addition of CO2 to inspired air partially suppressed hypoxemia. The consensus is that hypoxemia after MTH is solely attributable to the ventilation/perfusion inhomogeneity, but posthyperventilation hypoxemia is another reasonable interpretation of the hypoxemia after MTH with decreased V.A, V.CO2, and R. It is speculated that posthyperventilation normoventilation in respect to V.CO2 with baseline PaCO2 after MTH inhalation resulted in posthyperventilation hypoxemia as a result of relative hypoventilation in respect to V.O2.

Children with Nocturnal Upper Airway Obstruction: Postoperative Orthodontic and Respiratory Improvement

Twenty children, aged 4-9 years, underwent adeno/tonsillectomy because of unequivocal anamnestic nocturnal obstructive breathing. Preoperatively, apnea-hypopnea index was >5 in 10 cases only, AI>1 in 17. Nineteen children had signs of increased respiratory labour in movement recordings and inspiratory EMG-activity. Oxygen desaturation index was 0 in 7 children, and nadir SaO2

Sleep stage influences the hemodynamic response to obstructive apneas.

Blood pressure (BP) rises at the termination of obstructive episodes in patients with sleep apnea. Although the relationship of these BP elevations to oxygen saturation (SaO2) and arousal has been explored, the influence of sleep stage is undefined. To examine the effects of sleep stage on the postapnea BP elevation, we enrolled 12 patients with obstructive sleep apnea (OSA), and successfully collected data from seven of these (all male). Subjects slept overnight in the sleep laboratory, with full sleep and respiratory monitoring. Arterial pressure was assessed continuously with a radial artery catheter (six patients) or with digital photoplethysmography (one patient). Apneas occurring in both rapid eye movement (REM) and non-rapid eye movement (NREM) sleep were matched for duration and degree of desaturation. When mean arterial pressure (MAP) at termination of apneas during NREM sleep associated with SaO2 nadirs 78 to 82% (NREM 80%) was compared with MAP following apneas in REM with the same SaO2 nadir (REM 80%), there was a significant difference (NREM 80% 122 +/- 15.3 mm Hg, REM 80% 132 +/- 11.0; p = 0.0109). We also analyzed the effect of oxygen desaturation on MAP during REM sleep, by comparing events with SaO2 nadirs of 78 to 82% with events with nadirs of < 75% (REM < 75%). In REM, further desaturation was associated with significant lengthening of the obstructive episodes and significantly higher postapnea BP increases (REM 75% 143 +/- 19.9 mm Hg, p = 0.0392). We conclude that sleep stage alters the hemodynamic response to obstructive apneas during sleep.

Body Fat Distribution and Sleep Apnea Severity in Women

The contribution of body fat distribution to sleep-disordered breathing in women has not been examined in detail (to our knowledge). Fifty women under 65 years of age were diagnosed as having obstructive sleep apnea (OSA) by all-night polysomnography in a 6-month period. Twenty-five women underwent body fat measurements of skin folds and circumferences. The 12 premenopausal and 13 postmenopausal women did not differ in regard to apnea hypopnea index (AHI), SaO2 nadir, body mass index (BMI), or anthropometric measurements. The AHI for these 25 patients was related to the severity of obesity assessed by triceps and subscapular skin folds, the sum of the skin folds, waist circumference, and BMI. The SaO2 nadir correlated with triceps and subscapular skin folds, the sum of the skin folds, and neck skin fold. Clinical features of this same group of 25 women were then compared with those of 45 men with OSA previously described by our laboratory. The women, despite similar age, had less severe OSA than the men (AHI of 34.4 +/- 5.4 vs 51.1 +/- 4.9, p < 0.05). Despite similar BMIs and waist circumference, the men had evidence of a greater degree of upper body obesity with a larger subscapular skin fold thickness, waist-hip ratio, and neck circumference. In addition, for a given degree of upper-body obesity, men had more severe sleep apnea. These findings may explain, at least in part, the greater severity of OSA in the men.

CPAP Therapy via Oronasal Mask for Obstructive Sleep Apnea

To determine the effectiveness of oronasal masks for positive pressure therapy in alleviating obstructive sleep apnea (OSA). Polysomnographic records of all 245 patients with OSA who underwent therapeutic trials of either continuous positive airway pressure (CPAP) or bilevel positive airway pressure between January 1991 and December 1992 were reviewed. Thirty patients who had been prescribed positive pressure therapy employing an oronasal mask were identified. Two patients known to be successfully treated with CPAP via oronasal mask underwent repeat polysomnography. The initial portion of the study was a diagnostic evaluation during which the patients were untreated. During the second portion of the study, both patients used CPAP via an oronasal mask while wearing a mouthpiece designed to maintain oral patency. The 30 patients with OSA who were identified in this study had significant amelioration of OSA while receiving positive pressure therapy via oronasal mask compared with the baseline, diagnostic polysomnogram (apnea index: 55.3 +/- 36.9-->1.6 +/- 3.7, p < 0.001; hypopnea index: 21.2 +/- 20-->2.7 +/- 4.9, p < 0.001; nadir of SaO2: 72.5 +/- 13.9-->87.1 +/- 4.3, p < 0.001, mean +/- SD). Improvement of OSA did not depend on maintenance of a closed mouth, as evidenced by elimination of sleep-disordered breathing in the two patients receiving positive pressure via oronasal mask while wearing a mouthpiece to keep the mouth open. Oronasal masks are a viable alternative interface for alleviating OSA with positive pressure therapy in those patients who are unwilling or unable to tolerate conventional nasal interfaces. Although there were no adverse consequences associated with the use of oronasal masks in our patients, appropriate safety precautions should be taken to minimize the possibility of aspiration of gastric contents and avoid untoward sequelae due to positive pressure device failure.

Ventilatory responses during wakefulness in children with obstructive sleep apnea.

The pathophysiology of the obstructive sleep apnea syndrome (OSAS) is not fully understood. In children, airway obstruction secondary to tonsilloadenoidal hypertrophy is the leading cause of OSAS. However, not all children with tonsilloadenoidal hypertrophy develop OSAS. Thus, other factors, including abnormalities in ventilatory control, may contribute to the etiology of OSAS. To test this, we performed polysomnography and hypercapnic and hypoxic ventilatory response testing in 20 children and adolescents with OSAS (mean age, 8 +/- 3 [SD] yr) and 19 control subjects. Only two children with OSAS were obese. Children with OSAS had an apnea index of 16 +/- 20, peak PETCO2 of 54 +/- 5 mm Hg, and SaO2 nadir of 84 +/- 13% during polysomnography. Ventilatory responses were performed by rebreathing techniques. The slope of the hypercapnic ventilatory responses, corrected for body surface area, was 1.74 +/- 0.79 L/min/m2/mm Hg PETCO2 in children with OSAS and 1.45 +/- 0.58 L/min/m2/mmHg PETCO2 in control subjects (NS). Hypoxic ventilatory responses, corrected for body surface area, were -0.94 +/- 0.49 L/min/m2/% SaO2 in children with OSAS and -0.95 +/- 0.45 L/min/m2/% SaO2 in control subjects (NS); however, the sample size was small. There was a weak inverse correlation between the slope of the hypercapnic ventilatory response and the duration of hypoventilation during polysomnography (r = -0.44, p < 0.05). We conclude that children with OSAS have normal ventilatory responses to hypercapnia, and they may have normal ventilatory responses to hypoxia. We speculate that abnormal central ventilatory drive plays little if any role in the pathogenesis of pediatric OSAS.(ABSTRACT TRUNCATED AT 250 WORDS)

Development of Nocturnal Respiratory Disturbance in Untreated Patients With Obstructive Sleep Apnea Syndrome

Patient histories suggest that obstructive sleep apnea syndrome (OSAS) is a progressive condition. To investigate whether this could be shown in data from sleep recordings, 42 patients with OSAS were retrospectively studied. All had undergone a screening recording of respiration movements and oximetry at least 6 months (average, 16 months) prior to a diagnostic polysomnogram including these parameters. No treatment was given in the meantime. In the first recording, mean oxygen desaturation index (ODI, average number of desaturations of > 4 percent per sleeping hour) was 10; periodic, obstructive respiration movements occurred during (average) 36 percent of total estimated sleeping time, and mean nadir SaO2 was 85 percent. In the second recording, mean ODI was 21 (significant change, p = 0.0002), periodic respiration time was 61 percent, and nadir SaO2 was 80 percent (p = 0.0001, respectively). In 26 of 42 patients (62 percent), ODI had increased by > 50 percent. Increases in ODIs and periodic breathing were significantly correlated to increases in body weight. There were, however, exceptional patients with considerable increases in respiratory disturbance despite weight loss. The greatest changes were found in the patients who had the highest apnea indices in the polysomnograms. Early treatment may therefore be justified, since a borderline case may change to severe OSAS in 1 to 2 years' time. Follow-up recordings of untreated patients are important.

Effects of repetitive airway obstruction on O2 saturation and systemic and pulmonary arterial pressure in anesthetized dogs.

We examined the effects of multiple repetitive airway obstruction (RAO) on arterial oxygen saturation (SaO2) and pulmonary and systemic arterial pressure in eight anesthetized spontaneously breathing dogs. SaO2 was monitored at the tongue with a pulse oximeter. RAO created by an electrical valve that was attached to a tracheal cannula was alternated with seven consecutive spontaneous breaths until the nadir SaO2 (nSaO2) became constant or decreased to less than 35%. Tracheal occlusion durations of 15, 30, 45 and 60 s were chosen arbitrarily. In each animal nSaO2 decreased with every trial number in an exponential fashion, and the rate of nSaO2 fall was greater for the longer occlusion duration. In each animal the increases in pulmonary arterial pressure (PAP) and systemic arterial pressure (SAP) were inversely related to the nSaO2 values, and the relationship between nSaO2 and PAP or SAP was identical for all occlusion durations. Moreover, when the animals breathed pure oxygen and SaO2 did not decrease, there were no significant increases in the PAP and SAP at similar levels of pleural pressure (Ppl). In another six dogs, the effects of RAO on PAP and SAP were compared with those of intermittent hypoxic exposure without apnea, which was achieved by the inhalation of hypoxic gas (4 to 6% O2, 5% CO2 in N2) instead of RAO, to examine the effects of interruption of ventilation. The relationships between nSaO2 and both pressures did not differ significantly from those during RAO.(ABSTRACT TRUNCATED AT 250 WORDS)

Normal polysomnographic values for children and adolescents.

Although polysomnography is routinely performed to evaluate children and adolescents with sleep-disordered breathing, normal polysomnographic values for the pediatric age group have not yet been established. We therefore performed overnight polysomnography in 50 normal children and adolescents (mean age 9.7 +/- 4.6 SD yr, range 1.1 to 17.4 yr). Of the children 56% were male. Chest wall motion, ECG, oronasal airflow, end-tidal PCO2 (PETCO2), arterial oxygen saturation (SaO2), and electrooculogram were monitored. Children had 0.1 +/- 0.5 (range 0 to 3.1) obstructive apneas per hour of total sleep time, with only 18% of children having any obstructive apneas. No child had obstructive apneas > 10 s in duration. Of the children 30% had central apneas > or = 10 s in duration, and one child had a central apnea associated with SaO2 < 90%. Peak PETCO2 was 46 +/- 4 mm Hg (range 38 to 53 mm Hg), and hypoventilation (PETCO2 > 45 mm Hg) occurred for 7 +/- 19% total sleep time (range 0 to 91%). The SaO2 nadir was 96 +/- 2% (range 89 to 98%), with only one child desaturating below 90% in association with a central apnea. We conclude that polysomnographic results in the pediatric age group differ from those in adults. Recommendations for normal polysomnographic criteria are given.

Transtracheal Oxygen, Nasal CPAP and Nasal Oxygen in Five Patients with Obstructive Sleep Apnea

The effect of transtracheal oxygen administration by means of a 9-French (2.7 mm) percutaneous catheter was assessed in five patients with severe obstructive sleep apnea. We hypothesized that the delivery of oxygen below the site of airway obstruction should reduce the arterial oxygen desaturation during apneas and hypopneas, thereby increasing respiratory stability. Standard sleep and respiratory measurements were recorded in these subjects with all-night polysomnography on nonconsecutive nights during four experimental conditions: room air (BL), nasal continuous positive airway pressure (CPAP), nasal O2 (NC O2), and transtracheal O2 (TT O2). In three of these subjects, room air was infused (TT RA) at flow rates comparable to TT O2. Compared with baseline room air measurements, TT O2 not only significantly increased the SaO2 nadir from 70.4 percent to 89.7 percent (p less than 0.01), but it also reduced the frequency of sleep apnea/hypopnea from 64.6 to 26.2/h sleep (p less than 0.01). NC O2 ameliorated desaturation during apnea/hypopnea (mean SaO2 nadir, 86.2 percent; p less than .01) but did not significantly alter frequency (59.0/h sleep). Nasal CPAP was the most effective means of reducing sleep apnea/hypopnea (13.8/h sleep) but did not abolish desaturations when apneas occurred (mean SaO2 nadir, 80.0 percent). Compared with oxygen, transtracheal infusion of room air appeared to be somewhat effective; however, the small number of studies with TT RA precluded statistical analysis. We believe that TT O2 is superior to NC O2 for some patients with obstructive sleep apnea because continuous oxygen flow below the site of airway obstruction more reliably prevents alveolar hypoxia and respiration is stabilized. Infusion of air or oxygen through the tracheal catheter flow may also increase mean airway pressure and reduce obstructive apnea similar to nasal CPAP. We conclude that TT O2 may be an effective alternative mode of therapy for some patients with severe sleep apnea/hypopnea when nasal CPAP is not tolerated or when combined oxygen and nasal CPAP are required.

Comparison of nap and overnight polysomnography in children.

Overnight polysomnography is the "gold standard" for diagnosing sleep-disordered breathing. However, the limited number of resources for pediatric polysomnography make the availability of a screening test for sleep-disordered breathing highly desirable. Therefore, we compared 1 hour daytime nap polysomnography to overnight polysomnography in 40 children [mean age, 5.4 +/- 0.8 (SE) years] with sleep-disordered breathing; 76% of children were sedated with chloral hydrate for nap polysomnography; none was sedated for overnight polysomnography. Studies were done 26 +/- 4 days apart. Chest wall motion, ECG, end-tidal PCO2 (PETCO2), arterial oxygen saturation (SaO2), and electrooculogram were monitored. Nap studies had a sensitivity of 74%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 17% in predicting sleep-disordered breathing. Significantly more children had obstructive apnea and desaturation (SaO2 less than 90%) during overnight polysomnography. The peak PETCO2 and the SaO2 nadir were significantly worse during overnight polysomnography. However, the percentage of time during which abnormalities were manifested did not differ between nap and overnight polysomnography. Despite the use of sedation, nap polysomnography underestimated sleep-disordered breathing. We conclude that sleep-disordered breathing detected by nap polysomnography is always confirmed by overnight polysomnography and speculate that nap polysomnography may be an effective screening method for sleep-disordered breathing. However, overnight polysomnography should be performed if nap polysomnography is inconclusive. Chloral hydrate may be used effectively to facilitate sleep for nap polysomnography in children.

Effect of Cardiac Output Reduction on Rate of Desaturation in Obstructive Apnea

The nadir of SaO2 during an obstructive apnea is dependent upon the apnea's duration and the rate of fall of saturation (dSaO2/dt). We postulated that a low Q, such as in patients with congestive heart failure with sleep apnea, or a reduction in Q, as seen in some humans during obstructive sleep apnea, might steepen dSaO2/dt. The mechanism postulated was lowering of SvO2 with increased pulmonary capillary blood oxygen uptake and faster depletion of alveolar oxygen. This study examines dSaO2/dt following the onset of apnea in eight spontaneously breathing adult baboons. Nonrepetitive obstructive apneas (30, 45, and 60 seconds) were created by clamping an indwelling cuffed endotracheal tube at the end of expiration. Following baseline measurements, the animals were given a bolus of a rapid-acting beta-adrenergic blocker followed by continuous infusion to reduce cardiac output and to limit the cardiovascular response to obstructive asphyxia. Fiberoptic catheters were used for continuous monitoring of SaO2, SvO2, and cardiac output. Esophageal pressure and relative thoracic gas volume (Respitrace) were monitored to insure equivalence of lung volume at the onset of apnea. Beta-adrenergic blockade reduced resting Q by a mean of 25 percent. The blocked vs unblocked dSaO2/dt was 0.73 vs 0.72 percent/s, 0.76 vs 0.73 percent/s, and 0.70 vs 0.71 percent/s for 30-second, 45-second, and 60-second apneas, respectively. Thus, mean dSaO2/dt for all durations of apneas was unaffected by beta-adrenergic blockade. We concluded that dSaO2/dt is not influenced by limited Q preceding or induced by obstructive asphyxia.