S75 FACTORS CONTRIBUTING TO RESIDUAL DAYTIME SLEEPINESS IN PATIENTS WITH OBSTRUCTIVE SLEEP APNOEA SYNDROME TREATED BY CONTINUOUS POSITIVE AIRWAYS PRESSURE

SJ Doe, N Green, M Tomlinson, V Jeebun, H Myers, P Close, GJ Gibson. Department of Respiratory Medicine, Freeman Hospital, Newcastle upon Tyne, UK

Background: Although continuous positive airways pressure (CPAP) is a highly effective treatment for obstructive sleep apnoea syndrome (OSAS) some patients complain of persistent daytime sleepiness, the reasons for which are not well understood.

Aim: To investigate the factors contributing to residual daytime sleepiness in a cohort of patients with OSAS compliant with CPAP treatment.

Method: Prospective data collection from patients established on previously titrated, fixed pressure CPAP for at least 12 months. Entry criteria: OSAS with an apnoea/hypopnoea index (AHI) of 15/h or greater; CPAP compliance more than 4 h/night for at least the preceding 12 months. Patients were categorised by their current Epworth sleepiness score (ESS): non-sleepy group (ESS ⩽10); sleepy group (ESS >10). Patients in group 2 underwent re-titration studies using an autotitrating machine (RespMed Autoset) to ensure optimal CPAP usage and identify central apnoeas.

Results: 59 patients were included, 19 sleepy and 40 non-sleepy. 48 (81%) were men. Mean age was 49.4 years (range 23–67). The groups are compared in the table. No significant differences were seen in: age; smoking status; frequency of relevant co-morbidities and previous or current shift working. 16 patients with an ESS greater than 10 have completed re-titration studies with the following results: CPAP set too low (three); CPAP too high (one); significant mask leak (one); central apnoeas noted (one); CPAP pressure to overcome apnoeas intolerable (three); no issue noted (seven; 44%). In addition, three had co-morbidity, which may have contributed to sleepiness (severe chronic obstructive pulmonary disease; multisystem atrophy; ischaemic heart disease and diabetes).

Conclusions: Patients with residual daytime sleepiness were sleepier before treatment with a trend to more severe disease. They were more likely to be current snorers using higher CPAP pressures and to have had empirical pressure increases. The use of medications that cause drowsiness was more prevalent. Re-titration enabled the identification and management of specific issues in most patients. This represents and is a useful tool for assessing persistent sleepiness in patients compliant with therapy. There remains a significant proportion in whom no cause of residual daytime somnolence is identifiable.

S76 ARTERIAL STIFFNESS IN THE OBSTRUCTIVE SLEEP APNOEA/HYPOPNOEA SYNDROME

¹M Vennelle, ¹A Jones, ²TW Mackay, ¹DE Newby, ¹NJ Douglas, ²RL Riha. ¹University of Edinburgh, Edinburgh, UK, ²Department of Sleep Medicine, Royal Infirmary of Edinburgh, Edinburgh, UK

Introduction and Objectives: The obstructive sleep apnoea/hypopnoea syndrome (OSAHS) is an independent risk factor for cardiovascular disease. One marker of cardiovascular dysfunction is increased arterial stiffness. Arterial stiffness can be measured non-invasively using aortic pulse wave velocity (PWV) and the augmentation index (AIx), which is a measure of the augmentation of central aortic pressure due to pulse wave reflection. We aimed to determine whether there was a relationship between OSAHS severity and arterial stiffness in patients without known cardiovascular disease.

Methods: We recruited consecutive patients presenting to the sleep clinic with symptoms suggestive of OSAHS and an AHI of 15 or greater on polysomnography. All patients were continuous positive airways pressure (CPAP) naive and had no history of cardiovascular disease, hypertension or diabetes. Radial pulse wave analysis and aortic (carotid–femoral) PWV were carried out by a single operator using applanation tonometry (Sphygmocor). The AIx, corrected to a heart rate of 75 (AIx 75) was derived from radial pulse wave analysis. Analysis was undertaken using standard methods (SPSS, version 14). Pearson or Spearman’s rank correlations were also performed. A p value of 0.05 or less was considered statistically significant.

Results: 48 patients (mean age 46 years (SD 8.6), 65% male) were recruited. Baseline characteristics were: body mass index 30.7 kg/m² (SD 4.3), fasting cholesterol 5.2 mmol/l (SD 0.9), blood pressure systolic 129 mm Hg (SD 13), diastolic 76 mm Hg (SD 9), mean arterial pressure 94 mm Hg (SD 9), AHI 35 h/sleep (SD 21), ESS 11 (SD 5). Mean aortic PWV was 7.6 m/s (SD 1.4) and AIx75 18% (SD 10). Aortic PWV was positively correlated with the number of 2% oxygen desaturations per hour of sleep (r  =  0.32; p = 0.032), age (r  =  0.48; p = 0.001), fasting cholesterol (r  =  0.35; p = 0.021), systolic blood pressure (SBP) (r  =  0.35; p = 0.019) and diastolic blood pressure (r  =  0.41; p = 0.006). AIx75 correlated strongly with age (r  =  0.68; p⩽0.001) and SBP (r  =  0.29; p = 0.045) only.

Conclusion: In this population of OSAHS patients without known cardiovascular disease, increasing arterial stiffness correlated most strongly with increasing age. Aortic PWV correlated significantly with the number of 2% oxygen desaturations. A randomised, controlled trial is ongoing to determine if these measures of vascular stiffness change with CPAP therapy.

S77 THE EFFECT OF BARIATRIC SURGERY ON OBSTRUCTIVE SLEEP APNOEA

N Sopher, J New, PM Turkington. Salford Royal Hospital, Salford, UK

Introduction: Bariatric surgery is increasingly being explored as an option to aid weight loss in patients with significant obesity. However, the relationship between weight loss and improvement in obstructive sleep apnoea (OSA) in this group is not well established.

Aims: To establish whether obese patients with OSA show improvement in their OSA after bariatric surgery.

Methods: All patients being considered for bariatric surgery who were referred to the Salford Royal Hospital sleep service with possible OSA between September 2006 and June 2007 were reviewed. Patients who had a positive sleep study (Stardust, Respironics UK) were asked to have a repeat study at least 3 months after their surgery.

Results: 47 patients were screened for possible OSA prior to bariatric surgery. 22 (47%) patients were found to have OSA. The median (interquartile range; IQR) preoperative Epworth score was 8 (3–10), despite significant OSA on the sleep study (mean (SD) RDI 35.6 (23.8)). However, all patients reported non-specific feelings of lethargy/malaise. Of these 22 patients, 13 underwent sleep studies following their bariatric surgery. 11 of these 13 patients settled on to continuous positive airways pressure (CPAP) therapy and reported themselves as better or much better with CPAP. The median (IQR) Epworth score fell significantly to 2 (0–3) (p<0.05) on CPAP. The table shows the pre and post-surgery data for the 13 patients who underwent a follow-up study. Only three of the 11 patients who started CPAP continued to use it 12 months post-surgery. None of the patients who discontinued CPAP reported a return of their symptoms.

Conclusions: OSA is common in patients being considered for bariatric surgery. Many report non-specific symptoms but few specifically report daytime sleepiness. Despite this the majority of the patients in our group used CPAP before and for a period after the surgery with subjective symptomatic improvement. Following bariatric surgery significant improvements in OSA were seen. Although part of these changes may represent the phenomenon of regression to the mean, subjective improvements in symptoms were noted by the patients. Many were eventually able to discontinue their CPAP therapy after weight loss without a return of their symptoms.

S78 A PROSPECTIVE, RANDOMISED, SINGLE-BLIND, CROSSOVER STUDY COMPARING STANDARD BI-LEVEL WITH VOLUME ASSURED NON-INVASIVE POSITIVE PRESSURE VENTILATION IN CHRONIC OBSTRUCTIVE PULMONARY DISEASE PATIENTS WITH CHRONIC VENTILATORY FAILURE

NS Oscroft, TG Quinnell, JM Shneerson, IE Smith. Papworth Hospital NHS Foundation Trust, Cambridge, UK

Introduction: Persisting ventilatory failure associated with chronic obstructive pulmonary disease (COPD) is increasingly managed with domiciliary non-invasive positive pressure ventilation (NIPPV). Volume assured NIPPV (va-NIPPV) has been proposed as a means of optimising ventilation and improving treatment compliance by continually adjusting inspiratory pressures to match ventilatory requirements. However, few data for efficacy have been reported.

Methods: Twenty-five stable chronic obstructive pulmonary disease (COPD) patients, with ventilatory failure managed with nocturnal NIPPV (NIPPY 2; B & D Electromedical), were recruited and randomly assigned to standard NIPPV (VPAP III STA, Resmed) or va-NIPPV (AutoVPAP, Resmed) for 8 weeks, then crossed over. Diurnal arterial blood gases self-ventilating, overnight ventilation parameters on NIPPV, pulmonary function, exercise tolerance and treatment compliance were assessed at baseline and after each trial arm.

Results: Twenty-four subjects (13 male) completed the trial. The mean age was 65 years (9.2), body mass index 31.5 kg/m² (9.4), FEV₁ 0.6 litres (0.3) and FVC 2.1 litres (1). At enrolment pH was 7.39 (0.04), PaO₂ 8.1 kPa (1.1) and PaCO₂ 6.7 kPa (1.1) and overnight mean O₂ saturation (mSpO₂) was 90.2% (2.7) with a mean transcutaneous CO₂ (mPtcCO₂) 6.1 kPa (1.6). Improvements in daytime PaCO₂ (6.7 vs 6.3 kPa, p = 0.04) and overnight mSpO₂ (90.2 vs 91.3%, p = 0.03) were seen with standard NIPPV from baseline and compared with va-NIPPV (PaCO₂ 6.3 vs 6.7 kPa, p = 0.05, mSpO₂ 91.3 vs 89.8%, p = 0.02). Following va-NIPPV improvements in spirometry were seen from baseline (FEV₁ 0.6 vs 0.7 litres, p = 0.05, FVC 2.1 vs 2.3 litres, p = 0.02) but not when compared with standard NIPPV. There were no significant differences in compliance but 60% preferred va-NIPPV.

Discussion: Small and probably clinically insignificant differences were apparent following treatment with different ventilators and ventilatory modes but a majority of patients preferred va-NIPPV. Since effectively treated, compliant subjects were selected and increased use was not expected. However, having shown clinical equivalence it is now appropriate to study whether this new mode of ventilation can increase initial acceptance and compliance in ventilator-naive subjects.

S79 SEX RATIOS FOR VENTILATORY INTERVENTION IN MOTOR NEURONE DISEASE: BIOLOGY OR BIAS?

R Chadwick, T Quinnell, J Shneerson, I Smith. Respiratory Support and Sleep Centre, Papworth Hospital, Cambridge, UK

Introduction and Objectives: Motor neurone disease (MND) is more common in men, with reported sex ratios (male : female) varying between 1.3 : 1 and 2 : 1.1 One published report shows an excess of men in a population referred for weaning from invasive ventilation.2 The aim of this study was to examine the sex ratios in our cohort of MND patients requiring ventilatory support.

Methods: A retrospective review of MND patients referred between 1992 and 2007 inclusive was performed. Data extracted included: patients’ sex, age at presentation and use of assisted ventilation, either elective non-invasive ventilation (NIV) or unplanned intubation. All intubated patients were referred from ITU to our progressive care programme (PCP) for weaning following acute respiratory failure. Sex ratios were calculated by dividing the number of men by that of women. Data were analysed using Pearson’s χ² and odds ratios (OR) with a p value of 0.05 considered statistically significant.

Results: We identified 362 MND patients with a mean age (SD) of 63.6 years (10.9). The sex ratio of the total MND cohort was 1.8 : 1 (table). Comparisons sought between groups identified a significant predominance of men in patients who accepted long-term NIV compared with those not offered NIV (odds ratio (OR) 1.59, 95% CI 0.99 to 2.58) and in the PCP group compared with the whole cohort (excluding PCP patients; OR 2.94, 95% CI 1.10 to 7.89).

Conclusions: Our results demonstrate an overall sex ratio within the quoted ranges for the MND population. However, there is a preponderance of men electively accepting NIV. The imbalance is more marked for patients referred for weaning. The sex ratio of our MND PCP group is identical to that reported in a published case series.2 Differences may be related to the clinical variant of MND (chest wall muscle weakness or bulbar onset). Further work is required to determine to what extent these sex ratios reflect differences in the biology of the condition as opposed to differences in attitudes between patients, families and physicians according to the patients’ sex.

S80 NON-INVASIVE VENTILATION ALLOWS SAFE PERCUTANEOUS ENDOSCOPIC GASTROSTOMY INSERTION IN PATIENTS WITH NEUROMUSCULAR DISEASE AT RISK OF RESPIRATORY FAILURE

MAI Khan, K Davies, S Johnson, D Martin, A Bentley. University Hospital of South Manchester, Manchester, UK

Introduction: Percutaneous endoscopic gastrostomy (PEG) is a key prognostic factor in neuromuscular disorders including motor neurone disease (MND) and Duchenne muscular dystrophy (DMD). The timing of feeding gastrostomies is crucial in patients at risk of respiratory failure as evidenced by reduced vital capacity, reduced muscle pressures, evidence of alveolar hypoventilation or even mild daytime hypercapnia (PaCO₂ >6.5 kPa). We report the safe use of non-invasive ventilation (NIV) during PEG placement in these high-risk patients.

Methods: 21 patients participated. An experienced endoscopist performed PEG placement under conscious sedation. NIV was supervised by a specialist nurse and a respiratory/critical care trained physician. Earlobe blood gases (ELG) were recorded before and after the procedure. Transcutaneous CO₂ (TcCO₂) and oxygen saturations (SpO₂) were monitored throughout. Inspiratory pressure was increased as indicated by rising TcCO₂; this usually occurred at oesophageal intubation, due to increased mouth leak. Supplemental oxygen was entrained to maintain SpO₂ greater than 92%.

Results: 21 patients (11 male, 10 female; average age 58 years) underwent PEG placement. Seventeen had MND and one each had DMD, myasthenia gravis, mitochondrial cytopathy and minicore myopathy. An average of 2 mg midazolam and 25 μg fentanyl was administered, guided by bispectral index monitoring. An increase in inspiratory pressure was required (mean initial IPAP 16; mean maximum IPAP 22) to maintain ventilation. A rise in TcCO₂ from a mean of 4.6 to a mean of 5.9 kPa was observed during the procedure (p<0.001; paired t test). All patients were safely oxygenated and ELG post-procedure confirmed adequate ventilation (table). There were no immediate complications. Fifteen patients were discharged on home NIV. 17 patients are still alive, a mean of 201 days after the procedure. Four MND patients have died having survived a mean of 173 days.

Conclusion: We have shown that NIV can be used to allow safe PEG insertion in patients with neuromuscular disease in or at high risk of respiratory failure. Previously, these patients may have been denied PEG placement. This work supports the view that the respiratory management of neuromuscular patients is best undertaken in a specialist centre with appropriate expertise.