Sleep, 15(3):261-263
© 1992 American Sleep Disorders Association and Sleep Research Society
Preliminary Report: Validity of Symptom Analysis and Daytime Polysomnography in Diagnosis of Sleep Apnea
*Department of Otorhinolaryngology, and t Department of Clinical Neurophysiology, Karolinska Hospital, Stockholm. Sweden
Summary: The aim of this study was twofold: first, to see if the prevalence of the sleep apnea syndrome (SAS) in a given population could be fairly estimated by our patient questionnaire, mainly based upon the 1979 American Sleep Association definition of SAS; and second, to investigate whether the severity of SAS could be similarly accurately measured by daytime polysomnography (DPSG), as an alternative to the more demanding all-night polysomnography (NPSG). Of 42 patients consecutively examined due to rhonchopathy, 18 had the clinical diagnosis ofSAS, which was based on the three symptoms-snoring, sleep disturbances and diurnal hypersomnia-if reported to occur habitually. In 11 patients the diagnosis was established by NPSG [apnea index (AI) > 10]. However, in only 10 of the 18 cases NPSG indicated the diagnosis giving a positive predictive value of 56%. When comparing DPSG versus NPSG in 36 patients, the AI ranged from -23 to +65, and the mean AI value was found to be twice as high in the former (mean difference 9.0 ± 18.4; p < 0.01). The positive predictive value of DPSG was 63% (10/ 16). Both the self-report and DPSG were burdened with some 25% false-positive results, and DPSG gave far too variable AI values to be reliable in staging the disease. On the other hand, the negative predictive values were high, 96% (23/24) and 100% (20120), respectively, indicating their usefulness for screening purposes. Key Words: Sleep apnea syndrome- Validity-Daytime polysomnography-Self-report.
Rhonchopathy is caused by intermittent upper air- (6,7). However, cutoff scores of this whole range are way obstruction while sleeping, and its most serious still used in parallel in the contemporary literature, outcome is responsible for the obstructive sleep apnea although the most frequently used seems to be AI (ARI) syndrome (SAS). The American Sleep Association (1) > 10. defines SAS as a potentially lethal condition characFor cost-effective diagnosis of sleep disturbances, a terized by heavy snoring, multiple mixed or obstnic- daytime nap polysomnography (DPSG) has been sugtive sleep apneas and diurnal hypersomnia. For con- gested (8). Most sleep laboratories, however, use the firmation of sleep apnea and for staging severity of the more time-consuming all-night polysomnography disease, polysomnographic investigations are required (NPSG), which provides a better physiological picture (1,2). The borderline between "health" and "disor- of natural sleep behavior and therefore a more reliable der", based on number of apneas and hypopneas per AI. There are few published studies on DPSG valihour (AI, ARI), remains, however, ambiguous (3), and dation. In one study based on six patients, a decrease the suggested cutoff scores for significant sleep apnea of rapid eye movement (REM) sleep and an underesvary from 5 to 30 (2,4). The higher the cutoff score, timation of AI were found in DPSG as compared to the less night-to-night variability (5), and the better the NPSG (9). correlation to the clinical features of the syndrome The present study attempts to answer two questions, using NPSG as a reference throughout. First, is a thorough self-report reliable enough as a screening method Accepted for publication November 1991. for SAS, and second, is the AI obtained by DPSG as Address correspondence and reprint requests to Dr. Haraldsson, Department of Otorhinolaryngology, Karolinska Hospital, Stock- reliable for staging the disease as that obtained by NPSG? holm, Sweden. 261
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
*Per-Olle Raraldsson, *Christer Carenfelt, tEvert Knutsson, tRans E. Persson and *Johan Rinder
262
P.-O. HARALDSSON ET AL.
PATIENTS AND METHODS
Self-report questionnaire The questions used for clinical diagnosis ofSAS were Do you suffer from snoring? Have you been told of repeated breath cessations during sleep? Do you suffer from mid-sleep awakenings? Do you have difficulty initially falling asleep, or falling asleep after mid-sleep awakenings? Are you so tired that you suffer from involuntary daytime sleep spells? The ENT examination included a fiber-rhinolaryngoscopy with Mueller maneuver, but in the present study the clinical diagnosis was based on a self-report. The replies were coded 0-4 [0, do not know; 1, hardly ever or never; 2, occasionally; 3, often (several times a week); and 4, incessantly]. Unless otherwise specifif:d, responses of often or incessantly constituted a yes jor a constant symptom, whereas never or occasionally constituted a no. Polysomnography Investigations with DPSG and NPSG included registration of electroencephalogram, electrooculogram and submental electromyogram for assessment of sleep. Oral and nasal thermistors together with a strain gauge transducer were used to indicate apnea. Transcutaneous oxymetry was done in NPSG using an Ohmeda Biox 3740 Pulse Oxymeter. DPSG, supposed to last at least 90 minutes, was performed in the morning following 24 hours of sleep deprivation. For polysomnographic confirmation of diagnosis the total number of apneas lasting 2:: 10 seconds was registered. Two arbitrary limits for disease were used, AI 2:: 5 and AI > 10. Sleep, Vol. 15, No.3, 1992
Wilcoxon's test for matched pairs was used to compare the AI in D PSG as opposed to NPSG. By using DPSG results and patient self-reports as observed classification and NPSG as correct classification, the numbers of true positives (TP), false positives (FP), false negatives (FN) and true negatives (TN) were determined. For each test, sensitivity can be calculated as TP/(TP + FN), specificity as TN/(TN + FP), predictive value of a positive test as TP/(TP + FP) and predictive valult of a negative test as TN/(TN + FN). NPSG was related to a clinical diagnosis obtained from the self-reports given at the time of the investigation. RESULTS The distribution of clinical diagnoses and the corresponding AI findings in NPSG among the 42 patients examined were: 18 suffered from SAS (AI, mean 15, range 1-42), 14 from incomplete SAS (AI, mean 4.6, range 0-33) and 10 patients (AI, range 0-3) were simple snorers. The sensitivity of the clinical screening method was 80% (12/15) when the AI of 2: 5 was chosen as the cutoff score and 91% (10/11) when a limit of > 10 was used. Of the 24 cases who did not fulfill the criteria used for the clinical diagnosis ofSAS, only one patient had the syndrome according to NPSG (AI> 10) (false nltgative 1124). None ofthose 10 patients classified as simple snorers had apneas indicating SAS. Of 13 patients with AI > 20, all had reported breath cessations, and all but one also had DMS. Among 36 patients evaluated by DPSG, 16 were considered to suffer from SAS due to the finding of AI > 10. All cases had AI > 10 in NPSG, indicating a high sensitivity of the nap study for screening. When AI 2:: 5 was used as the arbitrary limit for disease the sensitivity was 80% (12/15). The mean sleeping time for NPSG was 421 ± 40 and for DPSG 66 ± 20 minutes. The mean AI, estimated by DPSG, was on average twice as high as that obtained from NPSG (mean difference being 9.0 ± 18.4; P < 0.01), but AI ranged from -23 to +65, indicating that the nap study is not worthwhile for grading disease. In Table 1 the results are summarized showing that the predictive values for negative outcome are high for ·both screening methods tested. However, clinical screening and DPSG are handicapped by their high rate offalse-positive outcomes, ifthe results ofNPSG are true. DISCUSSION Symptom analysis from a self-report, DPSG and NPSG all look at the disease from different, not fully
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
All patients were adults of both genders that had visited the Department of Otorhinolaryngology bI~ cause of rhonchopathy, defined as habitual snoring causing social, mental or physical handicaps. Rhonchopathy was graded from chronic snoring with social handicap to the clinical diagnosis of SAS, indicated by the triad of habitual symptoms: snoring, sleep disturbances [reported sleep apneas or difficulty maintaining sleep (DMS), without difficulty in initiating sleep] and diurnal hypersomnia. Incomplete SAS was used to designate patients without a full triad of symptoms, but who complained of habitual snoring. Forty-two consecutively selected patients were tested, using NPSG, and 36 of these also underwent DPSG.
Statistics
VALIDITY OF SYMPTOM ANALYSIS AND DAYTIME PSG IN SAS
263
TABLE 1. SAS- validity ofclinical diagnosis and ofdaytime polysomnography (DPSG, AI > 10) vs. all- night polysomnography (NPSG)a
Sleep. Vol. 15. No.3. 1992
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
of a new group of patients (7). Our results are in good agreement with theirs. Thus, by anamnesis very few cases of moderate to severe sleep apnea go unrecognized, but at the expense of many false-positive cases. Arbitrary limits for disease (NPSG) AI 2: 5 AI> 10 By definition false positive means nothing but the lack of an arbitrary number of apneas (of an arbitrary Clinical diagnosis, no. 42 Predictive value, positive 12/18 (67%) 10/18 (56%) chosen minimum duration) per hour of sleep. The clin6127 (22%) 8/31 (26%) False positive ical response of a uvulopalatopharyngoplasty in sePredictive value, negative 21/24 (88%) 23124 (96%) lected cases is convincing, even when the patients do False negative 1/11 (9%) 3/15 (20%) not fulfill any AI criterium, as the majority consider DPSG, no. 36 themselves cured of excessive day-time somnolence Predictive value, positive 12/19 (63%) 10/16 (63%) 7121 (33%) False positive 6126 (23%) (12). Is it explained by placebo effect? Probably not. 14/17 (82%) Predictive value, negative 20120 (100%) Recently, the diagnosis "upper airway resistance synFalse negative 3/15 (20%) OlIO (0%) drome" was coined for these cases (13). In this sense a Clinical criteria: snoring, sleep disturbances and diurnal hyperthere are no false positives. Thus, NPSG is the most somnia habitually. useful tool for separating the two syndromes and, moreover, the best method for staging sleep apnea. comparable aspects. However, using NPSG as the reference, the present report shows that both DPSG and REFERENCES the self-report are sensitive tools for diagnosis of SAS but lead to a high incidence of false-positive interpre- I. Roffwarg H, et al. Association of Sleep Disorders Centers. Ditations. Heavy snoring alone may cause sleep disturagnostic classification of sleep and arousal disorders. Sleep 1979; 2(1):1-137. bances and diurnal sleepiness (10), but when combined 2. Guilleminault C, Cumminskey J, Dement We. Sleep apnea synwith reports of breath cessations or frequent mid-sleep drome: recent advances. Adv Intern Med 1980;26:347-72. awakenings, SAS should be suspected. 3. Berry D, Webb W, Block J. Sleep apnea syndrome-a critical review of the apnea index as a diagnostic criterion. Chest 1984; Although DPSG was found to be very sensitive for detecting sleep apnea, it tended to overestimate the AI. 4. 86:529-31. Lugaresi E, Cirignotta F, Gerardi R, Montagna P. Snoring and Associated with the use of sleep deprivation are REMsleep apnea: natural history of snorers disease. In: Guilliminault C, Partinen M, eds. Obstructive sleep apnea syndrome: clinical sleep rebound, attenuated ventilatory chemosensitivresearch and treatment. New York: Raven Press, 1990:25-36. ity and delayed neuromuscular reflexes, which may 5. Wittig R, Romaker A, Zorik F, Roehrs T, Conway W, Roth T. have caused an increase in the AI (2). Also, the short Night-to-night consistency of apneas during sleep. Am Rev Respir Dis 1984; 129:244-6. registration time used probably contributed to the high 6. Crocker B, Olson L, Saunders N, Hensley M, McKeon J, et al. level of apneas, due to a predominance of periodic Estimation of the probability of disturbed breathing during sleep breathing and central apneas during initial sleep (11). before a sleep study. Am Rev Respir Dis 1990;142:14-8. Although DPSG might save money, the sleep depri- 7. Kapuniai L, Andrew D, Crowell D, Pearce J. Identifying sleep from self-reports. Sleep 1988;11(5):430-6. vation is inconvenient for the patient, and the method 8. apnea Goode B, Slyter H. Daytime polysomnogram diagnosis of sleep per se is less suitable for grading sleep apnea. disorders. J Neural Neurosurg Psychiatry 1983;46: 159-61. Not surprisingly, the symptom that best predicts sleep 9. Fourre J, Scoles Y, Nahmias J, Karetzky MS. The use of nap in diagnosing sleep apnea. Sleep Res 1986;15:122. apnea is the family report of breath cessations (6), 10. studies Hoffstein Y, Mateika JH, Mateika S. Snoring and sleep archiwhereas other single symptoms associated with SAS tecture. Am Rev Resp Dis 1991;143:92-6. seem to lack predictive value (7). However, when com- II. Waldhorn R. Cardiopulmonary consequences of obstructive sleep apnea. In: Fairbanks D, et ai., eds. Snoring and obstructive sleep binations of such symptoms are used for screening of apnea. New York: Raven Press, 1987:39-64. SAS the sensitivity is high, especially in cases of severe 12. Haraldsson P-O, Carenfelt C, Persson HE, Sachs C, Tornros J. Simulated long-term driving performance before and after uvusleep apnea (7). In the present study, we took advantage lopalatopharyngoplasty. ORL 1991;53:106-10. of the three characteristics given by the American Sleep 13. Guilleminault C, Stoohs R. Upper airway resistance syndrome. Association (1) when selecting screening questions. Two III World Congress on Sleep Apnea and Rhonchopathy, Tokyo, 1991:36 (abstract). questions, snoring and breath cessations, were tested by Kapuniai et al. and found to be valid for retesting
© 1992 American Sleep Disorders Association and Sleep Research Society
Preliminary Report: Validity of Symptom Analysis and Daytime Polysomnography in Diagnosis of Sleep Apnea
*Department of Otorhinolaryngology, and t Department of Clinical Neurophysiology, Karolinska Hospital, Stockholm. Sweden
Summary: The aim of this study was twofold: first, to see if the prevalence of the sleep apnea syndrome (SAS) in a given population could be fairly estimated by our patient questionnaire, mainly based upon the 1979 American Sleep Association definition of SAS; and second, to investigate whether the severity of SAS could be similarly accurately measured by daytime polysomnography (DPSG), as an alternative to the more demanding all-night polysomnography (NPSG). Of 42 patients consecutively examined due to rhonchopathy, 18 had the clinical diagnosis ofSAS, which was based on the three symptoms-snoring, sleep disturbances and diurnal hypersomnia-if reported to occur habitually. In 11 patients the diagnosis was established by NPSG [apnea index (AI) > 10]. However, in only 10 of the 18 cases NPSG indicated the diagnosis giving a positive predictive value of 56%. When comparing DPSG versus NPSG in 36 patients, the AI ranged from -23 to +65, and the mean AI value was found to be twice as high in the former (mean difference 9.0 ± 18.4; p < 0.01). The positive predictive value of DPSG was 63% (10/ 16). Both the self-report and DPSG were burdened with some 25% false-positive results, and DPSG gave far too variable AI values to be reliable in staging the disease. On the other hand, the negative predictive values were high, 96% (23/24) and 100% (20120), respectively, indicating their usefulness for screening purposes. Key Words: Sleep apnea syndrome- Validity-Daytime polysomnography-Self-report.
Rhonchopathy is caused by intermittent upper air- (6,7). However, cutoff scores of this whole range are way obstruction while sleeping, and its most serious still used in parallel in the contemporary literature, outcome is responsible for the obstructive sleep apnea although the most frequently used seems to be AI (ARI) syndrome (SAS). The American Sleep Association (1) > 10. defines SAS as a potentially lethal condition characFor cost-effective diagnosis of sleep disturbances, a terized by heavy snoring, multiple mixed or obstnic- daytime nap polysomnography (DPSG) has been sugtive sleep apneas and diurnal hypersomnia. For con- gested (8). Most sleep laboratories, however, use the firmation of sleep apnea and for staging severity of the more time-consuming all-night polysomnography disease, polysomnographic investigations are required (NPSG), which provides a better physiological picture (1,2). The borderline between "health" and "disor- of natural sleep behavior and therefore a more reliable der", based on number of apneas and hypopneas per AI. There are few published studies on DPSG valihour (AI, ARI), remains, however, ambiguous (3), and dation. In one study based on six patients, a decrease the suggested cutoff scores for significant sleep apnea of rapid eye movement (REM) sleep and an underesvary from 5 to 30 (2,4). The higher the cutoff score, timation of AI were found in DPSG as compared to the less night-to-night variability (5), and the better the NPSG (9). correlation to the clinical features of the syndrome The present study attempts to answer two questions, using NPSG as a reference throughout. First, is a thorough self-report reliable enough as a screening method Accepted for publication November 1991. for SAS, and second, is the AI obtained by DPSG as Address correspondence and reprint requests to Dr. Haraldsson, Department of Otorhinolaryngology, Karolinska Hospital, Stock- reliable for staging the disease as that obtained by NPSG? holm, Sweden. 261
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
*Per-Olle Raraldsson, *Christer Carenfelt, tEvert Knutsson, tRans E. Persson and *Johan Rinder
262
P.-O. HARALDSSON ET AL.
PATIENTS AND METHODS
Self-report questionnaire The questions used for clinical diagnosis ofSAS were Do you suffer from snoring? Have you been told of repeated breath cessations during sleep? Do you suffer from mid-sleep awakenings? Do you have difficulty initially falling asleep, or falling asleep after mid-sleep awakenings? Are you so tired that you suffer from involuntary daytime sleep spells? The ENT examination included a fiber-rhinolaryngoscopy with Mueller maneuver, but in the present study the clinical diagnosis was based on a self-report. The replies were coded 0-4 [0, do not know; 1, hardly ever or never; 2, occasionally; 3, often (several times a week); and 4, incessantly]. Unless otherwise specifif:d, responses of often or incessantly constituted a yes jor a constant symptom, whereas never or occasionally constituted a no. Polysomnography Investigations with DPSG and NPSG included registration of electroencephalogram, electrooculogram and submental electromyogram for assessment of sleep. Oral and nasal thermistors together with a strain gauge transducer were used to indicate apnea. Transcutaneous oxymetry was done in NPSG using an Ohmeda Biox 3740 Pulse Oxymeter. DPSG, supposed to last at least 90 minutes, was performed in the morning following 24 hours of sleep deprivation. For polysomnographic confirmation of diagnosis the total number of apneas lasting 2:: 10 seconds was registered. Two arbitrary limits for disease were used, AI 2:: 5 and AI > 10. Sleep, Vol. 15, No.3, 1992
Wilcoxon's test for matched pairs was used to compare the AI in D PSG as opposed to NPSG. By using DPSG results and patient self-reports as observed classification and NPSG as correct classification, the numbers of true positives (TP), false positives (FP), false negatives (FN) and true negatives (TN) were determined. For each test, sensitivity can be calculated as TP/(TP + FN), specificity as TN/(TN + FP), predictive value of a positive test as TP/(TP + FP) and predictive valult of a negative test as TN/(TN + FN). NPSG was related to a clinical diagnosis obtained from the self-reports given at the time of the investigation. RESULTS The distribution of clinical diagnoses and the corresponding AI findings in NPSG among the 42 patients examined were: 18 suffered from SAS (AI, mean 15, range 1-42), 14 from incomplete SAS (AI, mean 4.6, range 0-33) and 10 patients (AI, range 0-3) were simple snorers. The sensitivity of the clinical screening method was 80% (12/15) when the AI of 2: 5 was chosen as the cutoff score and 91% (10/11) when a limit of > 10 was used. Of the 24 cases who did not fulfill the criteria used for the clinical diagnosis ofSAS, only one patient had the syndrome according to NPSG (AI> 10) (false nltgative 1124). None ofthose 10 patients classified as simple snorers had apneas indicating SAS. Of 13 patients with AI > 20, all had reported breath cessations, and all but one also had DMS. Among 36 patients evaluated by DPSG, 16 were considered to suffer from SAS due to the finding of AI > 10. All cases had AI > 10 in NPSG, indicating a high sensitivity of the nap study for screening. When AI 2:: 5 was used as the arbitrary limit for disease the sensitivity was 80% (12/15). The mean sleeping time for NPSG was 421 ± 40 and for DPSG 66 ± 20 minutes. The mean AI, estimated by DPSG, was on average twice as high as that obtained from NPSG (mean difference being 9.0 ± 18.4; P < 0.01), but AI ranged from -23 to +65, indicating that the nap study is not worthwhile for grading disease. In Table 1 the results are summarized showing that the predictive values for negative outcome are high for ·both screening methods tested. However, clinical screening and DPSG are handicapped by their high rate offalse-positive outcomes, ifthe results ofNPSG are true. DISCUSSION Symptom analysis from a self-report, DPSG and NPSG all look at the disease from different, not fully
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
All patients were adults of both genders that had visited the Department of Otorhinolaryngology bI~ cause of rhonchopathy, defined as habitual snoring causing social, mental or physical handicaps. Rhonchopathy was graded from chronic snoring with social handicap to the clinical diagnosis of SAS, indicated by the triad of habitual symptoms: snoring, sleep disturbances [reported sleep apneas or difficulty maintaining sleep (DMS), without difficulty in initiating sleep] and diurnal hypersomnia. Incomplete SAS was used to designate patients without a full triad of symptoms, but who complained of habitual snoring. Forty-two consecutively selected patients were tested, using NPSG, and 36 of these also underwent DPSG.
Statistics
VALIDITY OF SYMPTOM ANALYSIS AND DAYTIME PSG IN SAS
263
TABLE 1. SAS- validity ofclinical diagnosis and ofdaytime polysomnography (DPSG, AI > 10) vs. all- night polysomnography (NPSG)a
Sleep. Vol. 15. No.3. 1992
Downloaded from https://academic.oup.com/sleep/article-abstract/15/3/261/2749282 by guest on 20 January 2019
of a new group of patients (7). Our results are in good agreement with theirs. Thus, by anamnesis very few cases of moderate to severe sleep apnea go unrecognized, but at the expense of many false-positive cases. Arbitrary limits for disease (NPSG) AI 2: 5 AI> 10 By definition false positive means nothing but the lack of an arbitrary number of apneas (of an arbitrary Clinical diagnosis, no. 42 Predictive value, positive 12/18 (67%) 10/18 (56%) chosen minimum duration) per hour of sleep. The clin6127 (22%) 8/31 (26%) False positive ical response of a uvulopalatopharyngoplasty in sePredictive value, negative 21/24 (88%) 23124 (96%) lected cases is convincing, even when the patients do False negative 1/11 (9%) 3/15 (20%) not fulfill any AI criterium, as the majority consider DPSG, no. 36 themselves cured of excessive day-time somnolence Predictive value, positive 12/19 (63%) 10/16 (63%) 7121 (33%) False positive 6126 (23%) (12). Is it explained by placebo effect? Probably not. 14/17 (82%) Predictive value, negative 20120 (100%) Recently, the diagnosis "upper airway resistance synFalse negative 3/15 (20%) OlIO (0%) drome" was coined for these cases (13). In this sense a Clinical criteria: snoring, sleep disturbances and diurnal hyperthere are no false positives. Thus, NPSG is the most somnia habitually. useful tool for separating the two syndromes and, moreover, the best method for staging sleep apnea. comparable aspects. However, using NPSG as the reference, the present report shows that both DPSG and REFERENCES the self-report are sensitive tools for diagnosis of SAS but lead to a high incidence of false-positive interpre- I. Roffwarg H, et al. Association of Sleep Disorders Centers. Ditations. Heavy snoring alone may cause sleep disturagnostic classification of sleep and arousal disorders. Sleep 1979; 2(1):1-137. bances and diurnal sleepiness (10), but when combined 2. Guilleminault C, Cumminskey J, Dement We. Sleep apnea synwith reports of breath cessations or frequent mid-sleep drome: recent advances. Adv Intern Med 1980;26:347-72. awakenings, SAS should be suspected. 3. Berry D, Webb W, Block J. Sleep apnea syndrome-a critical review of the apnea index as a diagnostic criterion. Chest 1984; Although DPSG was found to be very sensitive for detecting sleep apnea, it tended to overestimate the AI. 4. 86:529-31. Lugaresi E, Cirignotta F, Gerardi R, Montagna P. Snoring and Associated with the use of sleep deprivation are REMsleep apnea: natural history of snorers disease. In: Guilliminault C, Partinen M, eds. Obstructive sleep apnea syndrome: clinical sleep rebound, attenuated ventilatory chemosensitivresearch and treatment. New York: Raven Press, 1990:25-36. ity and delayed neuromuscular reflexes, which may 5. Wittig R, Romaker A, Zorik F, Roehrs T, Conway W, Roth T. have caused an increase in the AI (2). Also, the short Night-to-night consistency of apneas during sleep. Am Rev Respir Dis 1984; 129:244-6. registration time used probably contributed to the high 6. Crocker B, Olson L, Saunders N, Hensley M, McKeon J, et al. level of apneas, due to a predominance of periodic Estimation of the probability of disturbed breathing during sleep breathing and central apneas during initial sleep (11). before a sleep study. Am Rev Respir Dis 1990;142:14-8. Although DPSG might save money, the sleep depri- 7. Kapuniai L, Andrew D, Crowell D, Pearce J. Identifying sleep from self-reports. Sleep 1988;11(5):430-6. vation is inconvenient for the patient, and the method 8. apnea Goode B, Slyter H. Daytime polysomnogram diagnosis of sleep per se is less suitable for grading sleep apnea. disorders. J Neural Neurosurg Psychiatry 1983;46: 159-61. Not surprisingly, the symptom that best predicts sleep 9. Fourre J, Scoles Y, Nahmias J, Karetzky MS. The use of nap in diagnosing sleep apnea. Sleep Res 1986;15:122. apnea is the family report of breath cessations (6), 10. studies Hoffstein Y, Mateika JH, Mateika S. Snoring and sleep archiwhereas other single symptoms associated with SAS tecture. Am Rev Resp Dis 1991;143:92-6. seem to lack predictive value (7). However, when com- II. Waldhorn R. Cardiopulmonary consequences of obstructive sleep apnea. In: Fairbanks D, et ai., eds. Snoring and obstructive sleep binations of such symptoms are used for screening of apnea. New York: Raven Press, 1987:39-64. SAS the sensitivity is high, especially in cases of severe 12. Haraldsson P-O, Carenfelt C, Persson HE, Sachs C, Tornros J. Simulated long-term driving performance before and after uvusleep apnea (7). In the present study, we took advantage lopalatopharyngoplasty. ORL 1991;53:106-10. of the three characteristics given by the American Sleep 13. Guilleminault C, Stoohs R. Upper airway resistance syndrome. Association (1) when selecting screening questions. Two III World Congress on Sleep Apnea and Rhonchopathy, Tokyo, 1991:36 (abstract). questions, snoring and breath cessations, were tested by Kapuniai et al. and found to be valid for retesting
