Copyright ©ERS Journals Ltd 2007 Pharyngeal narrowing in end-stage renal disease: implications for obstructive sleep apnoea1 Dept of Medicine, University of Calgary, Calgary, AB, 2 Dept of Medicine, University of Toronto, and 3 Dept of Medicine, Humber River Regional Hospital, Toronto, ON, Canada. CORRESPONDENCE: P. J. Hanly, 1421 Health Sciences Centre, 3330 Hospital Drive NW, Calgary, Alberta, Canada T2N 4N1. Fax: 1 4032836151. E-mail: phanly{at}ucalgary.ca Keywords: Dialysis, kidney failure, pharyngometry, sleep apnoea, upper airway
Received: December 11, 2006
Sleep apnoea is common in patients with end-stage renal disease (ESRD). It was hypothesised that this is related to a narrower upper airway. Upper airway dimensions in patients with and without ESRD and sleep apnoea were compared, in order to determine whether upper airway changes associated with ESRD could contribute to the development of sleep apnoea.
An acoustic reflection technique was used to estimate pharyngeal cross-sectional area. Sleep apnoea was assessed by overnight polysomnography. A total of 44 patients with ESRD receiving conventional haemodialysis and 41 subjects with normal renal function were studied. ESRD and control groups were further categorised by the presence or absence of sleep apnoea (apnoea/hypopnoea index The pharyngeal area was smaller in patients with ESRD compared with subjects with normal renal function: 3.04±0.84 versus 3.46±0.80 cm2 for the functional residual capacity and 1.99±0.51 versus 2.14±0.58 cm2 for the residual volume. The pharynx is narrower in patients with ESRD than in subjects with normal renal function. In conclusion, since a narrower upper airway predisposes to upper airway occlusion during sleep, it is suggested that this factor contributes to the pathogenesis of sleep apnoea in dialysis-dependent patients. Sleep apnoea has been reported in up to 50–70% of patients with end-stage renal disease (ESRD) 1, a value at least ten times higher than the prevalence reported in the general population 2. The pathogenesis of sleep apnoea in patients with ESRD remains unclear. Although sleep apnoea is not corrected by conventional haemodialysis or peritoneal dialysis 3, 4, it has been reversed both by nocturnal haemodialysis and kidney transplantation 5–7, indicating that its pathophysiology is uniquely associated with the development of chronic renal failure. Previous investigators have observed features of both central and obstructive sleep apnoea (OSA) in patients with ESRD 3–7, which suggests that its pathogenesis is related both to destabilisation of central respiratory control and upper airway occlusion. In patients without renal failure, the pathogenesis of OSA is associated with anatomical or dynamic narrowing of the upper airway 8, 9. Individuals with a narrower pharynx are predisposed to upper airway occlusion during sleep and the development of chronic renal failure may create or enhance this in several ways. First, reduced lung volume associated with respiratory muscle weakness or pulmonary oedema can decrease upper airway size 10. Secondly, both fluid overload and systemic inflammation could cause upper airway oedema and thereby narrow the airway. Thirdly, uremic myopathy or neuropathy involving the upper airway dilator muscles may reduce airway size. Although there is evidence of both sensory and motor neuropathy in the upper airway in patients with OSA and normal renal function 11, 12, this has not been assessed in patients with ESRD. The objective of the present study was to compare the dimensions of the pharynx in a large group of patients with and without ESRD, further subdivided into individuals with and without sleep apnoea, in order to determine whether ESRD is associated with a narrowed upper airway, which could contribute to the development of sleep apnoea.
Patient recruitment All patients receiving conventional haemodialysis (in-centre, 4 h, 3 days·week–1) at the Humber River Regional Hospital, St. Michael's Hospital and the Toronto General Hospital (all Toronto, ON, Canada), who were referred to the sleep laboratory at St. Michael's Hospital for suspected sleep apnoea, were invited to participate in the present study. A detailed medical history was obtained from each patient, including the cause of renal failure, duration of dialysis treatment, dialysis schedule and medications. A control group, matched for body mass index (BMI) with the ESRD group, was recruited from subjects referred for polysomnography who had no history of kidney disease, cardiovascular dysfunction or upper airway surgery and from healthy volunteers (departmental staff and university students) who had no history of snoring or clinical features of sleep apnoea and were not taking medications that might influence sleep apnoea. The study protocol was reviewed and approved by the research ethics board at St. Michael's Hospital, and all patients gave written informed consent to participate in the study.
Polysomnography
Dialysis adequacy PRU = (pre-BUN–post-BUN)/(pre-BUN)x100 (1) where pre- and post-BUN are pre- and post-dialysis BUN measurements, and were obtained from the dialysis clinics at the time of polysomnography.
Pharyngometry
Measurements were obtained at the end of a normal tidal breath (functional residual capacity; FRC) and at the end of a forced expiration (residual volume; RV). These measurements were performed during oral breathing; nasal breathing was prevented by nose clips. The pharyngeal cross-sectional area was estimated between the oropharyngeal junction and the glottis (fig. 1
Analysis Mean data and SDs were analysed using ANOVA, regression analysis and unpaired t-tests. Nominal data were analysed using Chi-squared analysis. All p-values <0.05 were considered statistically significant.
Patient demographics and dialysis adequacy A total of 85 patients (51 males and 34 females) were recruited, aged 18–77 yrs (table 1 10 events·h–1. There were 44 patients with ESRD and 41 controls. Sex distribution was similar between the groups. Although there were no differences in age between ESRD and control groups, apnoeic patients were significantly older than nonapnoeic patients. By study design, the groups were matched for BMI in order to control for the potential impact of obesity on the upper airway.
The most common cause of ESRD was chronic glomerulonephritis, followed by diabetes mellitus, hypertension, polycystic kidney disease, haemolytic uraemic syndrome and pyelonephritis. In eight patients, the cause of ESRD was unknown. The duration of conventional haemodialysis treatment, as well as the effectiveness of dialysis, reflected by PRU, BUN and serum creatinine, was similar between apnoeic and nonapnoeic patients. Five patients with ESRD were taking benzodiazepines and were instructed to use their medication as usual.
Polysomnography
Total sleep time and sleep efficiency were greater in nonapnoeic than apnoeic patients within the ESRD group in contrast to the control group, wherein both total sleep time and sleep efficiency were greater in apnoeic than nonapnoeic patients (table 2
Pharyngometry
Previous studies in the nonrenal failure population have found a narrower upper airway in patients with OSA compared with healthy subjects 8, 9, 17, 18. Narrowing of the upper airway increases the likelihood of upper airway occlusion during sleep, when diminished dilator muscle tone and gravitational forces associated with the supine position combine to narrow the airway further. Previous investigators 17, using the acoustic reflection technique, reported that the pharyngeal cross-sectional area was 26% smaller in patients with OSA, than in those without OSA. It was found that the pharyngeal cross-sectional area in patients with ESRD was 12% smaller than in non-ESRD control subjects. Recent work using the same methodology as in the present study 19 has shown that increasing the pharyngeal cross-sectional area as little as 6%, concurrent with weight loss, was sufficient to reduce the severity of sleep apnoea, as reflected by the AHI, by 73%. Consequently, it is believed that the difference in cross-sectional area found between patients with ESRD and subjects with normal renal function is clinically significant, and may contribute to the development of sleep apnoea in this patient population. Pharyngeal narrowing in nonrenal failure patients with OSA may be associated with increased para-pharyngeal fat due to obesity, and thickened lateral pharyngeal walls possibly due to genetic inheritance or trauma associated with recurrent intraluminal negative pressure swings and snoring vibration. Upper airway calibre is also influenced by sex and age 20, 21. However, these mechanisms do not explain the present findings, since there were no significant differences in BMI, age, sex distribution or severity of sleep apnoea between patients with ESRD and with normal renal function (tables 1
Upper airway size is significantly altered by changes in lung volume: widening as lung volume increases and narrowing as lung volume falls 10. Consequently, pharyngeal narrowing in ESRD patients may be related to reduced lung volume, as reflected by the smaller observed ERV (table 4 Increased vascular distension in the upper airway due to fluid overload could contribute to pharyngeal narrowing in ESRD. In healthy subjects, decreasing central venous pressure by impeding venous return to the heart using leg cuff inflation increased upper airway dimensions 25. Fluid overload could also lead to interstitial oedema of the pharyngeal wall or para-pharyngeal tissues, which can narrow the airway. This suggestion is supported by the recent report that fluid displaced rostrally from the legs increases pharyngeal resistance in healthy subjects 26. Alternatively, upper airway oedema and pharyngeal narrowing could be caused by systemic inflammation. Pharyngeal narrowing has been noted in pre-eclampsia 27, which may be related to widespread systemic inflammation and oedema. ESRD is a chronic inflammatory state and similar mechanisms may contribute to pharyngeal narrowing in this patient population. Another possibility is upper airway dilator muscle dysfunction due to neuropathy or myopathy associated with chronic uraemia, or to the underlying cause of ESRD, such as diabetes mellitus. Sensory neuropathy has been demonstrated in the upper airway of OSA patients with normal renal function 12 and may exacerbate the disease process: topical anaesthesia of the upper airway increases apnoea duration in patients with OSA 28. Muscle denervation has been described in nonrenal failure patients with OSA 11, and may contribute to upper airway narrowing. Although it was found that the pharyngeal cross-sectional area was smaller in patients with ESRD than in control subjects, it was not significantly different between those with and without OSA. These findings are not unique. Stauffer et al. 29 measured the pharyngeal cross-sectional area using computerised tomography and found no difference between males with and without OSA matched for age and BMI. Similarly to the present study, the authors included both snoring and nonsnoring subjects in the control group. Inclusion of snorers in the present control group may account for the similarities observed in pharyngeal size between apnoeic and nonapnoeic patients, as there is evidence that the pharyngeal cross-sectional area is similar between nonapnoeic snorers and patients with OSA 17. Conflicting results between studies may also be related to differences in the measurement technique. Acoustic reflection measures the cross-sectional area from the oropharyngeal junction to the glottis, but it does not measure the dimensions of the velopharynx or the shape or configuration of pharyngeal structures. More recent studies 9, 18 have noted differences in the velopharyngeal cross-sectional area between patients with and without OSA, but have failed to find differences in the size of the oropharynx. Differences in the configuration of the velopharynx (lateral narrowing), but not the oropharynx, also distinguished apnoeic from nonapnoeic patients 18. The relevance of these findings with regard to the present results is highlighted by the observation that the velopharynx has been identified as the primary site of occlusion in patients with OSA 9. Notwithstanding these potential explanations for the absence of significant differences between apnoeic and nonapnoeic patients, the present findings suggest that pharyngeal narrowing alone does not account for the development of sleep apnoea in patients with ESRD, and that interaction with another pathogenic factor is required. It is not believed that the presence of sleep apnoea in ESRD patients in the present study was due to the timing of dialysis, since polysomnographic and respiratory assessments in ESRD patients were performed in a standardised fashion (within 24 h of the last haemodialysis session). Furthermore, the development of sleep apnoea was not related to the variability in the efficiency of haemodialysis, since no difference was found in PRU, BUN or serum creatinine between ESRD patients with and without sleep apnoea. It is possible that the development of sleep apnoea in the present patient population depends on the interaction between upper airway narrowing and the stability of ventilatory control 30. There is evidence that instability in central control of respiration can be associated with upper airway occlusion during sleep. In an experimental model of central control instability induced by transient hypoxia during sleep 31, the likelihood that central instability was accompanied by upper airway closure was greater if the airway was narrow. More recently, it has been reported that increased loop gain, which reflects ventilatory instability, is associated with the development of OSA in patients whose upper airway closes during sleep at a near-zero luminal pressure 32. Increased respiratory chemoreflex responsiveness in patients with ESRD and OSA has been previously observed 33, a factor which is known to destabilise central respiratory control. It is possible that the combination of central destabilisation and upper airway narrowing contributes to the development of OSA in patients with ESRD. The present study has some limitations. First, upper airway measurements were performed during wakefulness while other respiratory measurements were performed during sleep. The current authors acknowledge that sleep onset induces changes in the upper airway that are pivotal to the development of OSA. However, it is believed that the pharyngeal narrowing observed during wakefulness continued during sleep and contributed to the pathogenesis of sleep apnoea in patients with ESRD. Secondly, acoustic reflection measurements in the upper airway vary significantly between individuals, which can make it difficult to find significant differences between groups of subjects. The comparison between groups may be limited by a lack of overall power, attributed in part to this inherent variability and also to the small sample size. Nevertheless, the fact that significant differences were found between patients with and without ESRD, despite the inherent variability in the measurement, makes the present findings more robust. Finally, the present study design was cross-sectional and consequently cannot infer causality between pharyngeal changes and sleep apnoea in patients with ESRD. Nevertheless, it is believed that pharyngeal narrowing is likely to contribute to upper airway occlusion during sleep, particularly when it is combined with other factors, such as instability in central respiratory motor output. In summary, the pharyngeal cross-sectional area is reduced in patients with end-stage renal disease. Since a narrower upper airway predisposes patients to upper airway occlusion during sleep, it is suggested that this factor contributes to the pathogenesis of sleep apnoea in end-stage renal disease. Further studies are required to understand how upper airway narrowing develops and to determine what additional mechanisms contribute to the development of sleep apnoea in patient populations such as the present one.
The authors would like to thank J. Gabor for his technical assistance, and S. Crombach, J. Zaltzman and R. Prasad (all University of Toronto, Toronto, ON, Canada) for their assistance with patient recruitment.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||