American Journal of Respiratory and Critical Care Medicine

Preeclampsia is the predominant cause of admissions to neonatal intensive care. The diurnal blood pressure pattern is flattened or reversed in preeclampsia. We hypothesized that snoring and partial upper airway obstruction contribute to nocturnal rises in blood pressure. We tested this hypothesis by controlling sleep- induced upper airway flow limitation and snoring with nasal positive pressure. Eleven women with preeclampsia underwent two consecutive polygraphic sleep studies with simultaneous beat-to-beat blood pressure monitoring. Average blood pressure for the night overall and in each sleep stage was calculated. Sleep architecture was similar on the two study nights. Sleep-induced partial upper airway flow limitation occurred in all patients in the initial study. Autosetting nasal continuous positive airway pressure (CPAP) applied at a mean maximal pressure of 6 ± 1 cm H2O eliminated flow limitation throughout sleep on the treatment night. Blood pressure was markedly reduced on the treatment night [(128 ± 3)/(73 ± 3)] when compared with the initial nontreatment study night [(146 ± 6)/(92 ± 4)], p = (0.007)/(0.002). We conclude that partial upper airway obstruction during sleep in women with preeclampsia is associated with increments in blood pressure, which can be eliminated with the use of nasal CPAP.

Preeclampsia is a disease that occurs spontaneously only in humans. It is currently the predominant cause of maternal and fetal morbidity and mortality, occurring in 7 to 9% of the pregnant population (1, 2). Preeclampsia is a multisystem disease induced by a primary placental abnormality (3-7) whose first clinical sign is usually hypertension developing after the twentieth week of pregnancy (8). Other signs occurring in preeclampsia include proteinuria, elevated liver and renal enzymes, and peripheral and cerebral edema.

Most current modes of treatment for preeclampsia focus on blood pressure reduction, and early delivery with removal of the placenta, which initializes resolution of the disease. The treatment of hypertension associated with preeclampsia ranges from bed rest in mild cases, to multidrug administration. In most centers, the first-line pharmacological treatment of choice is a centrally acting antihypertensive agent. These include clonidine, a centrally acting α-agonist, and methyldopa, a centrally acting sympatholytic agent. For acute hypertensive crises, intravenous hydralazine is usually the drug of choice (9). As with the use of all other drugs during pregnancy, there is a certain amount of risk to the growth and well-being of the fetus associated with the use of antihypertensive medications. The aim of any antihypertensive therapy in preeclampsia is to minimize this risk to both mother and fetus.

Hypertension associated with preeclampsia usually has a characteristic diurnal “flattening” or reversal such that nocturnal blood pressure is increased relatively more than daytime blood pressure (10, 11). Previous experience has shown that specific nocturnal hypertension or “flattening” of the diurnal blood pressure rhythm is associated with snoring and obstructive sleep apnea (12-15). Snoring occurs in at least 27% of women in the third trimester of pregnancy (16). We hypothesized that there is some pathology, including snoring, which is occurring during sleep in these women, contributing to the nocturnal rise in blood pressure. Furthermore, we also suggested that correction of any pathological changes in ventilation, which occurred during sleep, would result in a normalization of the blood pressure.


We studied 11 women with severe preeclampsia (classified according to the Australian Society for the Study of Hypertension in Pregnancy [ASSHP] criteria [17]) admitted consecutively to the antenatal ward at King George V (KGV), Royal Prince Alfred Hospital, Sydney, Australia. The only entry criterion into the study was the presence of hypertension documented according to ASSHP, and proteinuria. Mean age of the patients was 34 ± 2 yr, mean gestational age was 35 ± 1 wk. Demographically, patients were within normal limits for prepregnancy body mass indices (BMI 27 ± 1 kg · m−2; range, 24 to 30 kg · m−2) and weight gain during pregnancy (21 ± 2 kg; range, 13 to 26 kg).

Two patients had preexisting essential hypertension with superimposed preeclampsia. One of the patients had asthma controlled with β-agonist on an “as-needed” basis. All other patients were well before presenting with preeclampsia. Because of the severity of their disease, all were in-patients on the antenatal ward of KGV. Written, informed consent was obtained from all patients included in the study protocol, and the protocol was approved by the Central Sydney Area Health Service Ethics Review Committee.


The severity of disease in all patients required that they were medicated with at least one antihypertensive agent. The first-line treatment for hypertension in preeclampsia at this hospital is clonidine (Catapres), and the mean daily dose taken by our patients was 525 ± 110 μg (range, 300 to 1,200 μg). The two patients in the group with preeclampsia superimposed on essential hypertension were also taking 200 mg and 100 mg, respectively, of hydralazine to supplement the clonidine dosage. However, there was no change in medications for any of the patients between the nontreatment and treatment nights. The last dose of medication before the onset of the sleep study was between 10:00 p.m. and 11:00 p.m.


All 11 patients included in the protocol had beat-to-beat blood pressure (Finapres, Ohmeda, Englewood, CO), nasal flow, and SaO2 measured on two consecutive nights. In addition to these three parameters, seven of the 11 patients also had full polysomnography on these same consecutive nights. The two study nights were not randomized for treatment and control, because we needed to diagnose obstruction before treatment could be instituted. Polysomnography was performed using the portable Compumedics Sleepwatch System (Compumedics p-series; Compumedics, Melbourne, Australia). Polysomnograpic recordings included two electroencephalography (EEG) channels (C3/A2 and O2/A1), two electro-oculogram (EOG) channels, submental electromyogram (EMG), SaO2 , and heart rate were measured via a finger oximeter attached to the Compumedics unit. We included a noncalibrated, highly sensitive measure of instantaneous nasal airflow so that a broad spectrum of sleep-induced upper airway dysfunction could be identified. This was achieved by the use of a pair of nasal prongs connected to a high-fidelity pressure transducer (incorporated in the Compumedics p-series). Nasal flow was recorded at a sampling rate of 125 Hz.

Sleep Staging

Sleep staging was performed according to recognized criteria (18), based on 30-s epochs. Percentages of each sleep stage were calculated according to the total time spent in all sleep stages. Arousal analysis was performed according to standard criteria, which includes alpha intrusion into the EEG for at least 3 s, accompanied by activation of submental EMG.

Analysis of Ventilation

To assess ventilation during sleep, nasal airflow was carefully analyzed. This was achieved using two different methods. A total of 10 random samples of 30 s (epochs) for each sleep stage in each patient were selected. Epochs in which the nasal prongs had clearly become partially or completely dislodged from the nose were discarded. Epochs in which the patient was clearly mouth breathing (as indicated by a sustained large reduction in nasal tidal volume [Vt] with no change in SaO2 , and no obvious signs of partial flow limitation such as low- or high-frequency oscillations in the nasal flow signal) were also discarded. In the first instance, a visual analysis of the nasal flow gave us information on the pattern of flow limitation as it occurred. In the second instance, the area under the curve was calculated for each breath to give a measure of Vt. Vt for each of the flow-limited breaths was expressed as a percentage of an average of the Vt of 5 breaths measured during a stable period of wakefulness.

Respiratory events (apneas and hypopneas) and arousals from sleep were also scored according to standard criteria. These criteria are as follows:

Apnea. Complete cessation of nasal airflow for at least 10 s, ending with arousal and/or a decrease in SaO2 of at least 4%.

Hypopnea. At least a 50% reduction in nasal airflow for at least 10 s, terminated by arousal and/or accompanied by a decrease in SaO2 of at least 4%.

Experimental Treatment with Nasal Continuous Positive Airway Pressure (CPAP)

We used an autosetting CPAP device (Autoset; Resmed Ltd., North Ryde, Australia) to undertake the experimental treatment arm of the study. This device identifies the presence of upper airway dysfunction by identifying inspiratory flow limitation and automatically increases nasal airway positive pressure to unload the upper airway and normalize the inspiratory airflow pattern (19-22).

Cardiovascular Measurement and Analysis

Continuous noninvasive arterial blood pressure measurement was performed using the Penaz method (23) (Finapres device). Blood pressure measurement with the Finapres device was compared at the beginning of each study using sphygmomanometry. Mean beat-to-beat blood pressure and heart rate were calculated for the whole night in all patients, and for individual sleep stages in patients who also had full polysomnography.

Measurement of Uric Acid

Serum uric acid has been demonstrated to be an important predictor of severity and outcome in preeclampsia (24). This parameter is measured as a part of the routine assessment of preeclamptic patients when admitted to KGV. We therefore sought to measure this parameter both before and after nasal CPAP treatment. Serum uric acid was measured within the 2 d preceding the CPAP treatment study and again within 2 d after the CPAP study in eight of the 11 patients. In these patients, 10 ml of venous blood was collected in clotting tubes and serum uric acid measured. The method used was uricase amino-phenazone reaction with peroxidase, measured at 505 nm (25).

Statistical Analysis

Changes in blood pressure, which are reported according to sleep stage, are calculated from the seven patients who, in addition to blood pressure and respiratory parameters, also had sleep stage measured. Values reported independently of sleep stage are calculated from all 11 patients. Values are reported as mean ± SEM. Comparison between the night off CPAP and the night on CPAP was performed using paired t tests, and the resultant p values are reported. Pearson correlation coefficients were calculated using regression analysis. p Values of less than 0.05 were considered significant, however, all p values are reported.

Blood pressure according to the Finapres device at the beginning of the study was within 10 mm Hg of the recording obtained using sphygmomanometry.


Sleep architecture was unaffected by the administration of nasal CPAP (Table 1), with mean total sleep times of 259 ± 22 min and 223 ± 48 min during the non-CPAP treatment night and the CPAP treatment night, respectively (p = 0.44). The percentage of time spent in rapid eye movement (REM) sleep did not change between the nontreatment night (9 ± 3%) and treatment nights (10 ± 5%, p = 0.82). The number of arousals from sleep did not change between nontreatment (35 ± 9 per hour) and treatment (28 ± 10 per hour of sleep, p = 0.732) nights.


Time (± SEM) in min
Study TimeWakeStages 1/2 NREM SleepStages 3/4 NREM SleepREM Sleep
Initial study401 ± 19142 ± 6166 ± 1967 ± 1327 ± 10
CPAP372 ± 55149 ± 46125 ± 3368 ± 730 ± 17
p Value0.5880.900.180.940.771

Definitions of abbreviations: NREM = non–rapid eye movement; REM = rapid eye movement.

Ventilation during Sleep

Table 2 summarizes changes in ventilation during sleep on the first (nontreatment) study. None of the patients had sleep apnea hypopnea syndrome according to standard American Sleep Disorders Association (ASDA) criteria, all having apnea–hypopnea indices (AHI) of less than 10. Similarly we observed only minor degrees of audible snoring. However, all patients demonstrated a characteristic form of sleep-induced upper airway flow limitation. The feature of this flow limitation, which occurred only during inspiration, was the presence of low-frequency oscillations in flow of 1 to 5 Hz (Figure 1). The low-frequency oscillations characteristic of these patients were in contrast to the high frequencies commonly occurring when audible snoring is present (26, 27). This pattern of flow limitation was, however, associated with a mild decrease in Vt of 17 ± 4%. This characteristic pattern of low-frequency inspiratory flow limitation (LFIFL) occurred in 72 ± 4% of the breaths, which were analyzed according to “Analysis of Ventilation.”


StageVt Decrease (%)No. of Breaths (%)RDI
NREM11 ± 272 ± 56 ± 2
REM18 ± 370 ± 32 ± 1
All sleep13 ± 272 ± 45 ± 1

Definition of abbreviations: NREM = non–rapid eye movement; REM = rapid eye movement; RDI = respiratory disturbance index.

Treatment Pressures

Maximal nasal CPAP pressure that was required to eliminate airflow limitation (according to the Autoset device) was relatively low in most patients. The mean pressure required was 6 ± 1 cm H2O (5 to 10 cm H2O). Once maximal pressure was applied, nasal airflow in all patients was unobstructed for the remainder of the study. Of the 11 patients included in the study, seven continued the use of nocturnal nasal CPAP until delivery (for a mean of 5 d after the initial treatment study), three delivered the following day, and one decided not to use the device (with continuation of pregnancy for another 2 d).

Blood Pressure

Table 3 displays the blood pressure according to sleep stage on the initial nontreatment night in comparison to that during treatment with nocturnal nasal CPAP. In all patients there was a cumulative increase in blood pressure throughout the night (Figure 2). The one exception to this was a patient who spent 3 h awake in the middle of the night, and who subsequently had a reduction in blood pressure until sleep resumed, at which time blood pressure again increased cumulatively. Treatment with the Autoset device, and consequent reversal of sleep-related inspiratory flow limitation, resulted in an absence of blood pressure increments over the course of the night.


Stages 1/2 NREM SleepStages 3/4 NREM SleepREM SleepAll Sleep
Nontreatment137 ± 481 ± 2144 ± 488 ± 2140 ± 684 ± 5139 ± 483 ± 2
Treatment128 ± 374 ± 3130 ± 576 ± 3131 ± 476 ± 3129 ± 473 ± 3
p Value  0.005 0.017  0.015 0.039  0.019 0.032  0.006 0.004

Blood pressures during the initial 5 min of the nontreatment and treatment studies were similar (Figure 3) [(125 ± 7)/(74 ± 7)] and [(129 ± 4)/(73 ± 4)] during nontreatment and treatment nights, respectively, p = (0.536/0.892). However, during the last 5 min of the nontreatment study, blood pressure was markedly elevated above the initial baseline [(158 ± 9)/(100 ± 9), p = 0.005/0.023]. In contrast, blood pressure during the last 5 min of the treatment study was not significantly elevated above the initial baseline recording [(131 ± 2)/(78 ± 3) (p = 0.65/0.16)] (Figure 3). Mean overnight blood pressure was markedly reduced during the night of treatment with nasal CPAP [(129 ± 4)/ (73 ± 3)] when compared with the nontreatment night [(149 ± 6)/(93 ± 5)], p = 0.012/0.007.

Heart Rate

Although there were marked changes in blood pressure for all stages of sleep during CPAP treatment, heart rate did not change significantly during any of the periods measured.

Uric Acid

After treatment with nasal CPAP, uric acid was reduced in all patients studied when compared with measures before treatment was introduced. The mean reduction in uric acid was 0.44 ± 0.02 mmol/L to 0.41 ± 0.02 mmol/L (p = 0.006).

The major finding of this study is that nocturnal blood pressure increments, which occur in women with preeclampsia, are effectively eliminated by the application of nasal CPAP. Furthermore, rises in blood pressure noted to occur during sleep in women with preeclampsia were associated with a very clear pattern of sleep-induced upper airway inspiratory flow limitation. This flow limitation was characterized by very-low-frequency oscillations (typically 1 to 5 Hz) occurring on the inspiratory portion of the flow curve. This LFIFL was associated with a reduction in Vt (typically 10 to 20%). This pattern of flow limitation was eliminated by the application of nasal CPAP.

While all patients were taking a variety of antihypertensive medications, these were not altered between the initial nontreatment, and the following treatment nights. We avoided bias in choosing the patients to be included in the study by approaching all appropriate patients (that is, those with both hypertension and proteinuria) admitted to the antenatal ward of KGV for preeclampsia. The women involved in the study were therefore generally indicative of a wide selection of patients with preeclampsia. The weight gain during pregnancy was within normal range; however, prepregnancy weight was slightly outside the recommended healthy range (BMI 27 ± 1 kg · m−2). While this was true, prepregnancy obesity is known to be a risk factor for developing preeclampsia (28).

In this study it was necessary that the two studies be performed in order, with a simple diagnostic test first, followed by a CPAP treatment night if indicated (as it was in all 11 studies). Potentially this may have resulted in an order effect. However, a number of factors suggest that this was not the case. First, at the beginning of each of the nontreatment and treatment studies, while patients were awake and relaxed, blood pressure was similar. Second, the most significant reduction in blood pressure resulting from the application of nasal CPAP was seen during slow-wave sleep (systolic blood pressure reduced from 144 ± 4 mm Hg to 130 ± 5 mm Hg, and diastolic blood pressure reduced from 88 ± 3 mm Hg to 76 ± 3 mm Hg). Furthermore, blood pressure on the initial treatment night was at its peak during slow-wave sleep (average systolic and diastolic blood pressures, respectively, during stage 4 sleep 147 ± 4 mm Hg and 89 ± 3 mm Hg). Slow wave sleep is the stage of sleep during which parasympathetic function predominates. Thus, one would expect that if emotional factors played a predominant role in increasing blood pressure on the first night, the nadir would be in slow-wave sleep. We therefore feel it valid to discount an order effect being the predominant cause of a decrease in blood pressure during CPAP treatment.

The Finapres device was used to measure continuous beat-to-beat blood pressure noninvasively. This device has been validated in a number of studies under a number of different conditions, including hypertension treated with medication (29-32). These studies indicate that the Finapres reliably provides a true measure of both systolic and diastolic blood pressure.


As previously described, these women demonstrated a characteristic pattern of oscillatory flow during inspiration. For want of a more appropriately descriptive term, we have labeled this pattern LFIFL, and we believe that this has not been described before. The pattern was associated with a marginal decrease in Vt of 10%. However, this level of reduction in Vt was insufficient to cause a measurable reduction in SaO2 , although it most likely was associated with a marginal reduction in PaO2 and concomitant marginal increase in PaCO2 .

The reason for the inspiratory flow limitation with characteristic low-frequency flow oscillations in these women with preeclampsia is unknown. It is possible that it occurs commonly in normal pregnancy and is the result of circulating tissue relaxant factors. Alternatively, it could be part of the syndrome of preeclampsia. This is a generalized disorder of capillary function, tissue edema being a key clinical feature. Organ edema, including cerebral, hepatic, and lung, forms a portion of the basis of mortality in this disease. The upper airway dysfunction may thus be the result of edema in the upper airway. Tissue edema has the potential to induce flow limitation by narrowing the upper airway. Tissue edema over a long segment of the upper airway will result in a lower frequency of oscillation than is seen when the site of narrowing is an isolated segment. This is because if a longer segment of narrowing is involved, it will have a greater mass and the dynamics of airflow through a tube are such that a greater mass will result in a lower frequency of oscillation.

According to results described here, the mechanism of nocturnal blood pressure increments is related to sleep-induced changes in upper airway patency. Associated with the LFIFL, we observed a remarkably large percentage of the nontreatment night during which our patients demonstrated limited reduction in the nasal flow. This reduction was in the order of a 10 to 20% decrease in Vt, with no apparent change in respiratory rate. This was clearly not sufficient to produce measurable decreases in SaO2 in these women. However, this limited amount of reduction in nasal airflow is likely to have resulted in some modest increase in PaCO2 tension. The cardiovascular effects of relative hypercapnia during preeclamptic pregnancy or indeed, during normal pregnancy, are currently unclear. We suggest that in light of other data on hypercapnia in the nonpregnant individual that clearly show a marked hypertensive response to even modest increases in PaCO2 (33-36), this may be contributing significantly to the marked increase in blood pressure noted to occur in these women. Furthermore, whereas normal pregnancy is associated with marked decreases in peripheral vascular reactivity, in preeclamptic pregnancies, the opposite is true, with a marked increase in peripheral vascular reactivity (37, 38). Thus, the increase in CO2 which we propose occurs as a result of partial flow limitation during sleep, in combination with the phenomenon of increased peripheral vascular reactivity associated with preeclampsia, results in a dangerous continual increase in blood pressure over the course of the night. Eliminating partial obstruction using the Autoset device, and thereby allowing for normal Vt may avoid increases in CO2, with a consequent improved control of nocturnal blood pressure.

Although we suggest that elimination of nocturnal increments in PaCO2 is likely to be the predominant factor contributing to the success of nasal CPAP in reducing nocturnal blood pressure, other mechanisms may also be involved. A simple mechanical effect on intrathoracic pressure may be implicated. With increased positive pressure in the chest cavity, one would expect to also have a decreased venous return, and thereby, a decreased cardiac output, leading to decreased blood pressure. However, the very low CPAP pressures required to correct upper airway flow limitation in this group (6.5 ± 0.5 cm H2O, range 4 to 10 cm H2O) are unlikely to have been the predominant cause of such large changes in blood pressure.

Finally, we reported that in this group of patients, circulating levels of uric acid have also been reduced after the application of nocturnal nasal CPAP. There could be a number of reasons for this. It may simply be a result of the improved blood pressure control, with an improved blood flow through the kidneys. Or, it may be independent of changes in blood pressure and result from improved nocturnal perfusion of the kidneys after the reversal of massive peripheral vasoconstriction for similar reasons to those discussed in relation to the blood pressure changes.

Nasally applied CPAP as a therapy for obstructive sleep apnea (OSA) was first described in 1981 (39) and has become the standard treatment for severe OSA. While these patients did not have OSA, they did have some upper airway resistance syndrome, as previously described by Guilleminault and colleagues (40). Furthermore, with the application of nocturnal nasal CPAP to eliminate this upper airway resistance syndrome, we have demonstrated a significant improvement in two important indicators of severity of disease in preeclampsia. We do not suggest that we are able to treat the underlying cause of the disease of preeclampsia, which is still unknown. However, with the application of nocturnal nasal CPAP, we may be able to extend the time in utero for these already compromised fetuses by improving maternal blood pressure control, and we therefore suggest that the possibility of its routine use as a therapy in this condition should be considered.

1. Pietrantoni M., O'Brien W. F.The current impact of the hypertensive disorders of pregnancy. Clin. Exp. Hypertens.161994479492
2. Long P. A., Abell D. A., Beischer N. A.Parity and pre-eclampsia. Aust. N.Z. J. Obstet. Gynaecol.191979203206
3. Roberts J. M., Redman C. W.Pre-eclampsia: more than pregnancy-induced hypertension. Lancet341199314471451
4. Redman C. W.Immunological aspects of pre-eclampsia. Baillieres Clin. Obstet. Gynaecol.61992601615
5. Lyall F., Greer I. A., Boswell F., Young A., Macara L. M., Jeffers M. D.Expression of cell adhesion molecules in placentae from pregnancies complicated by pre-eclampsia and intrauterine growth retardation. Placenta161995579587
6. Conrad K. P., Benyo D. F.Placental cytokines and the pathogenesis of preeclampsia. Am. J. Reprod. Immunol.371997240249
7. Marzi M., Vigano A., Trabattoni D., Villa M. L., Salvaggio A., Clerici E., Clerici M.Characterization of type 1 and type 2 cytokine production profile in physiologic and pathologic human pregnancy. Clin. Exp. Immunol.1061996127133
8. Worley R. J.Pathophysiology of pregnancy-induced hypertension. Clin. Obstet. Gynecol.271984821834
9. Nissen J. C.Treatment of hypertensive emergencies of pregnancy. Clin. Pharmacol.11982334343
10. Redman C. W., Beilin L. J., Bonnar J.Reversed diurnal blood pressure rhythm in hypertensive pregnancies. Clin. Sci. Mol. Med. Suppl.31976687s689s
11. Beilin L. J., Deacon J., Michael C. A., Vandongen R., Lalor C. M., Barden A., Davidson L., Rouse I.Diurnal rhythms of blood pressure, plasma renin activity, angiotensin II and catecholamines in normotensive and hypertensive pregnancies. Clin. Exp. Hypertens. B21983271293
12. Imai Y., Abe K., Munakata M., Sakuma H., Hashimoto J., Imai K., Sekino H., Yoshinaga K.Circadian blood pressure variations under different pathophysiological conditions. J. Hypertens. Suppl.81990S12532
13. Sforza E., Parchi P., Contin M., Cortelli P., Lugaresi E.Do autonomic cardiovascular reflexes predict the nocturnal rise in blood pressure in obstructive sleep apnea syndrome? Blood Press.31994295302
14. Lapinski, M., T. Przybylowski, J. Lewandowski, A. Januszewicz, B. Wocial, J. Kowalski, W. Droszcz, and W. Januszewicz. 1993. Diurnal blood pressure rhythm and urinary catecholamine excretion in obstructive sleep apnoea and essential hypertension. J. Hypertens. Suppl. 11(Suppl. 5):1995.
15. Bayliss D. A., Millhorn D. E.Central neural mechanisms of progesterone action: application to the respiratory system [Review]. J. Appl. Physiol.731992393404
16. Loube D. I., Poceta J. S., Morales M. C., Peacock M. D., Mitler M. M.Self-reported snoring in pregnancy: association with fetal outcome [see comments]. Chest1091996885889
17. Brown M. A., Buddle M. L.What's in a name? Problems with the classification of hypertension in pregnancy. J. Hypertens.15199710491054
18. Rechtschaffen, A., and A. Kales. 1968. A Manual of Standardized Terminology, Techniques and Scoring Systems for Sleep Stages of Human Subjects. National Institutes of Health, Washington, DC.
19. Berthon Jones, M., S. Lawrence, C.E. Sullivan, and R. GrunsteinNasal continuous positive airway pressure treatment: current realities and future. Sleep191996S131135
20. Kiely J. L., Delahunty C., Matthews S., McNicholas W. T.Comparison of a limited computerized diagnostic system (ResCare Autoset) with polysomnography in the diagnosis of obstructive sleep apnoea syndrome. Eur. Respir. J.9199623602364
21. Teschler H., Farhat A. A., Exner V., Konietzko N., Berthon M., JonesAutoSet nasal CPAP titration: constancy of pressure, compliance and effectiveness at 8 month follow-up. Eur. Respir. J.10199720732078
22. Gugger M., Mathis J., Bassetti C.Accuracy of an intelligent CPAP machine with in-built diagnostic abilities in detecting apnoeas: a comparison with polysomnography. Thorax50199511991201
23. Penaz J.Criteria for set point estimation in the volume clamp method of blood pressure measurement [Review]. Physiol. Res.411992510
24. Voto L. S., Illia R., Darbon H. A., Grosso, Imaz F. U., Margulies M.Uric acid levels: a useful index of the severity of preeclampsia and perinatal prognosis. J. Perinat. Med.161988123126
25. Poulos, V., and A. Piesse. 1995. Handbook of Laboratory Medicine. Royal Prince Alfred Hospital, Camperdown.
26. Fiz J. A., Abad J., Jane R., Riera M., Mananas M. A., Caminal P., Rodenstein D., Morera J.Acoustic analysis of snoring sound in patients with simple snoring and obstructive sleep apnoea. Eur. Respir. J.9199623652370
27. Dalmasso F., Prota R.Snoring: analysis, measurement, clinical implications and applications. Eur. Respir. J.91996146159
28. Sibai B. M., Gordon T., Thom E., Caritis S. N., Klebanoff M., McNellis D., Paul R. H.Risk factors for preeclampsia in healthy nulliparous women: a prospective multicenter study. The National Institute of Child Health and Human Development Network of Maternal–Fetal Medicine Units. Am. J. Obstet. Gynecol.1721995642648
29. Gabriel A., Linblad L. E., Angleryd C.Noninvasive versus invasive beat-to-beat monitoring of blood pressure. Clin. Physiol.121992229235
30. Virolainen J.Use of non-invasive finger blood pressure monitoring in the estimation of aortic pressure at rest and during the Mueller manoeuvre. Clin. Physiol.121992619628
31. Penzel T., Mayer J., Peter J. H., Podszus T., Voigt K. H., von Wichert P.Continuous non-invasive blood pressure monitoring in patients with sleep disorders. Physiol. Res.4119921117
32. Lindqvist A.Beat-to-beat agreement of non-invasive finger artery and invasive radial artery blood pressure in hypertensive patients taking cardiovascular medication. Clin. Physiol.151995219229
33. Lin Y. C., Shida K. K., Hong S. K.Effects of hypercapnia, hypoxia, and rebreathing on circulatory response to apnea. J. Appl. Physiol. Respir. Environ. Exerc. Physiol.541983172177
34. Rothe C. F., Maass-Moreno R., Flanagan A. D.Effects of hypercapnia and hypoxia on the cardiovascular system: vascular capacitance and aortic arch chemoreceptors. Am. J. Physiol.2591990932939
35. Serebrovskaya T. V.Comparison of respiratory and circulatory human responses to progressive hypoxia and hypercapnia. Resp.5919923541
36. Somers V. K., Mark A. L., Zavala D. C., Abboud F. M.Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J. Appl. Physiol.67198921012106
37. Ramsay M., Broughton F., Pipkin, Rubin P.Comparative study of pressor and heart rate responses to angiotensin II and noradrenaline in pregnant and non-pregnant women. Clin. Sci.821992157162
38. Ito M., Nakamura T., Yoshimura T., Koyama H., Okamura H.The blood pressure response to infusions of angiotensin II during normal pregnancy: relation to plasma angiotensin II concentration, serum progesterone level, and mean platelet volume. Am. J. Obstet. Gynecol.166199212491253
39. Sullivan C. E., Issa F. G., Berthon M., Jones, Eves L.Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet11981862865
40. Guilleminault C., Stoohs R., Clerk A., Cetel M., Maistros P.A cause of excessive daytime sleepiness. The upper airway resistance syndrome. Chest1041993781787
Correspondence and requests for reprints should be addressed to Dr. Natalie Edwards, David Read Laboratory, Rm. 450, Dept. Medicine (D06), Sydney University, NSW 2006, Australia. E-mail:

Dr. Edwards was supported by a project grant from the National Health and Medical Research Council.

Professor Sullivan is a consultant to, and minor shareholder in, Resmed, the manufacturer of the Autoset.


No related items
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record