Hospital outcomes and long-term survival after referral to a specialized weaning unit

Transcription

1 British Journal of Anaesthesia, 118 (4): (2017) doi: /bja/aex031 Critical Care CRITICAL CARE Hospital outcomes and long-term survival after referral to a specialized weaning unit M. G. Davies*, T. G. Quinnell, N. S. Oscroft, S. P. Clutterbuck, J. M. Shneerson and I. E. Smith Respiratory Support and Sleep Centre, Papworth Hospital NHS Foundation Trust, Cambridge CB23 3RE, UK *Corresponding author. michael.davies10@nhs.net Abstract Background. Prolonged invasive mechanical ventilation (IMV) is a frequent challenge, and an increasing number of patients are transferred from intensive care units to long-term acute care hospitals or specialized weaning units. There are few published data for discharge home rates, use of noninvasive ventilation (NIV), or long-term survival. Methods. A case-note and database review was conducted of patients admitted to a UK national specialized weaning unit for weaning from IMV between 1992 and Patients were grouped into diagnostic categories according to the predominant cause of weaning failure. Weaning outcomes and long-term survival were assessed according to diagnostic group and mode of ventilation on discharge. Results. Four hundred and fifty-eight patients were transferred for weaning from IMV. Four hundred and seventeen (91%) survived to hospital discharge, of whom at least 343 (82%) were ultimately discharged to their own home. Three hundred and thirty (72%) weaned from IMV, of whom 142 weaned from all ventilation and 188 weaned to nocturnal NIV. Weaning success was highest for patients with chronic obstructive pulmonary disease and chest wall disorders. Median survival from unit discharge was 25 months (interquartile range 5 74), with the longest survival seen for patients discharged with nocturnal NIV [37 (12 81) months]. Conclusions. These results confirm successful weaning outcomes for patients transferred to a specialized weaning and long-term ventilation service. In contrast to other service models, most patients achieved discharge to their own home. Key words: noninvasive ventilation; prolonged mechanical ventilation: Respiration, Artificial; ventilator weaning Prolonged dependence on invasive mechanical ventilation (IMV) is an important clinical challenge. A UK intensive care unit (ICU) point-prevalence survey 1 showed that 11% of ICU patients were clinically stable, but had required IMV for 21 or more consecutive days. Such patients have higher mortality and occupy a disproportionate number of ICU bed days, leading to increased health-care costs. 23 In the USA, the annual cost of prolonged dependence on IMV, a key feature of chronic critical illness, 4 is estimated to be $35 billion. 5 An international consensus document concluded that standard ICUs may lack the structure and focus to manage patients with weaning failure. 6 A range of organizational models have emerged. In the USA, long-term acute care hospitals 7 represent the predominant model. These manage a range of persisting organ failure(s), including prolonged IMV dependence. In contrast, specialized weaning units, which focus on weaning alone, are the recommended model of care within the UK. 8 Both models offer an economic advantage compared Editorial decision: December 1, 2016; Accepted: January 18, 2017 VC The Author Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please journals.permissions@oup.com 563

2 564 Davies et al. Editor s key points There are few data on long-term outcomes in intensive care unit patients who require prolonged mechanical ventilation. In this study of patients referred to a specialized weaning unit throughout a 20 yr period, a high proportion survived to hospital discharge, and 75% returned home. Almost one-third of patients were weaned completely from ventilatory support, and a further 41% required nocturnal ventilation only. These results show better long-term outcomes than previously reported. comprises individuals who are considered unweanable in their referring ICU; they have survived the acute pathology but remain in single-organ respiratory failure. For example, we are unable to accept patients with haemodynamic instability or a need for daily renal replacement therapy until these issues have resolved. However, patients with progressive neuromuscular conditions are accepted, whereas these were excluded from analysis by Damuth and colleagues 10 on the grounds of expected poor prognosis. In the UK, there is a separate spinal cord injury service, so patients with high-spinal pathologies are not routinely referred to our service. Patients were grouped according to the primary diagnostic cause of weaning failure, specifically neuromuscular disorders (NMD), chronic obstructive pulmonary disease (COPD), postsurgical (PS), non-chronic obstructive pulmonary disease (COPD) respiratory disorders (RESP), chest wall disorders (CWD), and others. with continued ICU care because of their lower staff-to-patient ratios. 29 However, the results of a recent meta-analysis of 124 studies by Damuth and colleagues 10 showed concerning variation in patient outcomes; hospital mortality was 26%, and only 22% of patients were discharged home. Worse outcomes were seen for patients treated in the USA than in the UK; better outcomes were seen for patients treated in specialized weaning units. 10 The authors concluded that patient selection was likely to account for such variation, proposing that US patients with a poor prognosis may be more likely to be intubated, proceed to tracheostomy, and continue IMV than patients treated elsewhere. The implication is that variation in outcome could be reduced by better patient selection for intubation and via earlier withdrawal of life support for patients considered to be less likely to survive or achieve discharge home. However, the meta-analysis did not provide patient-level data to confirm the assumption around patient selection, nor did it assess outcomes according to the methods of weaning used. Here, we present the results of a consistent approach to weaning that overcomes this aspect of variation. The patients are well documented and have long follow-up, with accurate discharge and survival data. The data show referral trends and outcomes throughout the 20 yr in which the organizational model has been in place. Methods Service infrastructure The Respiratory Support and Sleep Centre (RSSC) at Papworth Hospital, Cambridge, UK has provided a national specialist weaningservicesince1992andaimstodischargepatientshomewith the maximal possible independence from ventilatory support. 11 The service infrastructure is consistent with current National Health Service (NHS) specialist weaning specifications. 12 It comprises a team of respiratory physicians, nurses (usually 1:2, with variation according to patient need), and physiotherapists with a specialist interest and skills in weaning and home ventilation. It is also supported by hospital-wide services, including dietetics, speech therapy, and occupational therapy. Patients All patients who have required prolonged IMV are accepted for transfer, providing that they are otherwise clinically stable at the time of transfer. Referrals are not declined because of any perceived inability to wean. The patient group therefore Weaning from IMV On transfer, the medical problems leading to weaning failure are reviewed and management is optimized where possible. Invasive mechanical ventilation is continued, and increased if necessary, to normalize arterial carbon dioxide concentrations (Pa CO2 ) before weaning is attempted. Noninvasive ventilation (NIV) is considered in each patient as a bridge to ventilatory independence and as a possible destination treatment. Weaning consists of daytime periods of unsupported spontaneous breathing. 11 When possible, the tracheostomy cuff is deflated and a speaking valve used to encourage communication. This weaning method is applied by a specialist nursing and physiotherapy team, skilled in tracheostomy management and in the application of NIV to enable the gradual withdrawal of IMV. There is a daily consultant ward round. Weaning success is defined as unit survival plus liberation from IMV to self-ventilation (SV) or nocturnal NIV, in line with international consensus. 13 After successful weaning, continued rehabilitation is often needed before discharge home. This is delivered within our unit, although external transfer is sometimes required. Transfer to ICU is occasionally necessary if a temporary escalation of care is required or to await a local community care package in the event of weaning failure. Data collection After institutional approval, data were obtained from an electronic database, cross-referenced to a retrospective review of Papworth Hospital medical case notes. All patients transferred for weaning between January 1, 1992 and December 31, 2011 were included. Survival status as of December 31, 2014 was recorded using the NHS Summary Care Record application. Thus, confirmed survival data were available for a minimum of 3 yr after weaning unit admission. Statistical analysis Data were analysed using IBM SPSS Statistics version 23.0 (Chicago, IL, USA). Categorical data are presented as a number and percentage. Continuous data are presented as the mean (SD) or median and interquartile range (IQR) when not normally distributed. Normal distribution of the data was tested using the Shapiro Wilk test. Differences between groups were tested using the Wilcoxon rank sum test for non-normally distributed data. Logistic regression was used to investigate the relationship between factors. Kaplan Meier survival curves were calculated from the date of weaning unit discharge, with censoring

3 Outcomes from a specialized weaning unit 565 Table 1 Patient characteristics expressed as a number (percentage) unless otherwise stated. *Of the 165 patients with neuromuscular conditions, at least 82 had a diagnosis associated with progressive disease. These diseases were as follows: motor neurone disease (n¼33), muscular dystrophies (n¼18), non-icu myopathies (n¼17), and myotonic dystrophy (n¼14). Patients with a clinical diagnosis of COPD were included. Where available (n¼69, sometimes post-discharge), spirometry data showed median forced expiratory volume in 1 s of 0.6 litre (22%) and functional vital capacity of 2 litres (58%). Previous operative procedures included cardiothoracic surgery (n¼32), laparotomy (n¼23), neurosurgery (n¼8), and others (n¼3). Bronchiectasis (n¼10), post-pneumonia (n¼10), interstitial lung disease (n¼6), chronic severe asthma (n¼6), cystic fibrosis (n¼6), bronchial carcinoma (n¼4), and other respiratory disorders (n¼8). Scoliosis post-poliomyelitis (n¼12), congenital/ other scoliosis (n¼19), and previous thoracoplasty (n¼4). COPD, chronic obstructive pulmonary disease; ICU, intensive care unit; IMV, invasive mechanical ventilation; IQR, interquartile range; Pa CO2, arterial partial pressure of carbon dioxide; RSSC, Respiratory Support and Sleep Centre Patient characteristics Patients admitted 458 Age [yr; median (IQR)] 61 (50 71) Males 259 (57) Patients transferred from external ICUs 417 (91) ICU length of stay [days; median (IQR)] 33 (18 55) RSSC length of stay [days; median (IQR)] 27 (16 49) IMV requirements on admission to weaning unit Full 422 (92) Nocturnal 36 (8) Pa CO2 on admission to weaning unit 9.9 ( ) [kpa; median (IQR)] Diagnostic group Neuromuscular* 165 (36) COPD 104 (23) Postsurgical 66 (14) Non-COPD respiratory disorders 50 (11) Chest wall disorders 35 (8) Others 38 (8) on December 31, The log-rank test was used to compare survival between categorical groups. A P-value of <0.05 was considered statistically significant. Results Patient characteristics Four hundred and fifty-eight patients were transferred, of whom patient characteristics data were available for 457 (Table 1). On weaning unit admission, 422 (92%) of all patients were 24 h IMV dependent, with the remaining 36 (8%) requiring at least nocturnal IMV but already achieving some daytime periods of SV. Median Pa CO2 was 9.9 ( ) kpa on weaning unit admission. Most patients were transferred from an ICU external to Papworth Hospital (n¼417; 91%). Weaning unit outcomes Weaning success and survival Four hundred and seventeen patients (91%) survived to unit discharge, and outcome data were available for 416. Weaning from IMV was achieved in 330 patients (72%); 188 (41%) weaned to nocturnal NIV, and 142 (31%) weaned to SV on discharge. Eighty-six (19%) patients failed to to wean from IMV, and 40 (9%) patients died within the unit (online-only supplementary Fig. S1). Median Pa CO2 for all patients on discharge was 6.1 ( ) kpa (P<0.001 compared with unit admission). Noninvasive ventilation was used during the weaning process in 72% of patients alive at discharge. Data with respect to mode of ventilation on discharge are shown in Table 2. Length of stay and discharge location Median weaning unit length of stay was 27 (16 49) days, although it was longer for patients who failed to wean (P¼0.002; Table 2). Of the 418 patients who survived to unit discharge, 343 (82%) ultimately achieved discharge to their own home; this represents 75% of all patients transferred for weaning (Table 3). A further 48 patients (12%) required long-term institutional care, and accurate discharge location could not be verified for 25 (6%) patients. Factors associated with weaning outcomes Weaning outcomes varied according to diagnosis (Fig. 1). Weaning success (nocturnal NIV or SV) ranked as follows: CWD (88.5%), COPD (88%), PS (68%), RESP (64%), and NMD (63%). Patients with NMD experienced low unit mortality rates (4%) but the highest rate of weaning failure (33%), typically as a result of severe bulbar dysfunction. Controlling for age, patients with COPD had significantly lower odds of continued IMV or death compared with all other diagnostic groups except CWD (P¼0.001). There was a higher rate of weaning success (nocturnal NIV or SV) for females (79%) than males (67%, P¼0.007), but no association was seen for age or ICU length of stay. For patients with ICU stays >40 days, weaning success was 78%, in comparison to 66% for patients transferred after 40 days. Long-term survival Survival from weaning unit discharge at 1, 3, and 5 yr was 65, 41, and 29%, respectively, and varied according to diagnosis (Fig. 2) and mode of ventilation on discharge (Fig. 3). Median survival for all patients was 25 (5 74) months. Patients who weaned to nocturnal NIV survived longer [37 (12 81) months] than those who weaned to SV [20 (6 78) months; P<0.03]. Survival was shorter for patients who failed to wean from IMV [6 (1 32) months] compared with patients who weaned (P<0.001; Fig. 3). Within the continued IMV group, survival was longer for patients with NMD [n¼55, 17 (2 34) months] compared with non- NMD conditions [n¼31, 2 (0 18) months; P¼0.02]. Patients who were ultimately discharged home lived longer [32 (IQR 12 83) months] than survivors who were not [1 (IQR 0 5) months; P<0.001]. Patients discharged directly home lived longer [n¼205, 40 (IQR 15 92) months] than those discharged home after further care elsewhere [n¼138, 24 (IQR 6 68) months; P¼0.003]. Of patients who were transferred to an external ICU (n¼58), 21 continued to require institutional care (hospital or care setting). For this subgroup (5% of all patients transferred), post-weaning unit survival was only 0.7 (0.2 3) months and therefore most in keeping with in-hospital mortality. Long-term survival was inversely associated on univariate analysis with age, but was not associated with sex or weaning unit length of stay. However, a weaning unit length of stay >40 days was associated with a reduction in long-term survival (P¼0.04).

4 566 Davies et al. Table 2 Outcomes with respect to mode of ventilation on discharge from weaning unit. Values are expressed as the median (interquartile range) unless otherwise stated. ICU, intensive care unit; IMV, invasive mechanical ventilation; NIV, nocturnal noninvasive ventilation; Pa CO2, arterial partial pressure of carbon dioxide; SV, self-ventilation Status at discharge SV NIV IMV Died Patients [n (%)] 142 (31) 188 (41) 86 (19) 40 (9) Age (yr) 63 (53 72) 61 (49 70) 58 (45 68) 70 (53 77) Males [n (%)] 77 (54) 97 (52) 58 (67) 26 (65) ICU length of stay (days) 32 (18 54) 31 (17 54) 37 (23 84) 34 (18 51) Weaning unit length of stay (days) 24 (15 40) 25 (16 43) 37 (18 77) 48 (9 61) Admission Pa CO2 (kpa) 8.3 ( ) 11.3 ( ) 10.1 ( ) 9.8 ( ) Discharge Pa CO2 (kpa) 5.7 ( ) 6.4 ( ) 6.1 ( ) Ultimately discharged home [n (%)] 116 (82) 174 (93) 53 (62) Table 3 Initial and ultimate discharge destination of patients who survived to discharge (n¼418) Initial discharge location Ultimate discharge location [n (%) of patients alive at unit discharge] Home Institutional care Unknown Home 205 (49.0) External ward 100 (23.9) 27 (6.5) 12 (2.8) External intensive care unit 28 (6.7) 21 (5.0) 9 (2.2) Hospice/nursing home/other 10 (2.4) 0 2 (0.5) Unknown (1.0) Total 343 (82) 48 (11.5) 27(5.5) Weaning unit outcome (%) Total NMD COPD PS RESP CWD Others SV NIV Continued IMV Died Fig 1 Weaning outcomes for all patients (data available for n¼456) and according to primary diagnostic group.

5 Outcomes from a specialized weaning unit year 3-year 5-year 88 Survival post-discharge (%) Total (416) NMD (159) COPD (97) PS (57) Resp (37) CWD (34) Other (32) Fig 2 The 1, 3, and 5 yr survival for all patients transferred and with respect to diagnostic group. CWD, chest wall disorders (e.g. scoliosis); NMD, neuromuscular disorder; PS, prolonged invasive mechanical ventilation after major surgery; RESP, non-chronic obstructive pulmonary disease respiratory disease. Cumulative proportion surviving Nocturnal NIV Self-ventilation Continued IMV Months from discharge Fig 3 Survival according to mode of ventilation on discharge, as estimated by the Kaplan Meier method. Nocturnal noninvasive ventilation (NIV) vs continued invasive mechanical ventilation (IMV), P<0.001; NIV vs self-ventilation (SV), P<0.03. Five year trends Outcomes were assessed in four 5 yr cohorts between 1992 and Age, sex, and location of ICU care (external vs internal) did not differ over time. However, the case-mix changed; patients with NMD or COPD represent 59% of all transfers (range 56 60% across cohorts) and proportions remained similar. From 1992 to 2011, the postsurgical group increased (from 7 to 21% of all transfers) and the chest wall group decreased (from 16 to 3%). Although weaning unit length of stay did not change, the median prior ICU length of stay increased from 21 (12 41) days in to 41 (28 65) days in (P<0.001). Weaning success and unit survival did not differ when assessed in 5 yr cohorts. Discussion Patients who require prolonged IMV consume significant ICU resources, and many die in hospital. 2 3 We show that external transfer to a specialized weaning unit can achieve high levels of weaning success (72%) and satisfactory long-term outcomes. At 1 yr, 65% of our patients were alive compared with 27 41% for other service models. 10 In line with the inclusion criteria of Damuth and colleagues, 10 all patients had required continuous IMV for >14 days, a tracheostomy for >96 h, or both. We received 91% of all patients from external units. At the time of transfer, 92% of patients required continuous IMV. We optimized gas exchange before weaning, often by initially increasing ventilation. Median Pa CO2 improved from 9.9 kpa on admission to 6.1 kpa on discharge, suggesting that suboptimal ventilation might have contributed to weaning failure before transfer. With resolution of the acute episode, most ICU patients experience rapid and complete liberation from all ventilatory support. In contrast, weaning success after prolonged IMV includes continued nocturnal NIV 13 to reflect the prevalence of chronic ventilatory co-morbidities in this patient group. Within our service, we use NIV to facilitate decannulation for patients with limited respiratory reserve. In the event of continued ventilatory failure after decannulation, we continue nocturnal NIV on discharge in keeping with evidence of benefit for selected patients with chronic ventilatory failure attributable to neuromuscular, 14 COPD, 15 or chest wall disorders, 16 or after weaning from prolonged IMV Compared with the complex and expensive

6 568 Davies et al. community care requirements for domiciliary tracheostomy ventilation, the relative simplicity of NIV may enable more patients to be discharged home. In our series, 41% of all patients were discharged on nocturnal NIV, of whom 93% ultimately achieved discharge home. Long-term NIV was also associated with improved survival compared with other modes of ventilation on discharge. Although we recognize and emphasize the retrospective nature of these data that are uncontrolled for case-mix, we believe that NIV is an integral part of successful weaning and post-discharge care for selected patients who have required prolonged IMV. For patients who have required prolonged IMV, estimating prognosis is a crucial part of clinical decision-making, including weaning unit referral. Such prognostication is challenging 20 in the absence of long-term data. Using discharge home rates as an outcome measure to assess the likely benefit of continued active care has been proposed In the recent meta-analysis, only 22% of patients transferred for weaning achieved discharge home, 10 suggesting that continuation of active care, including transfer for weaning, may be inappropriate for some patients. In our series, 49% of patients were discharged direct to their own home, and an additional 33% achieved discharge home after a period of rehabilitation or other institutional care elsewhere. The weaning unit is only part of the continuum of care for patients who have required prolonged IMV. Underestimating final discharge home rates could lead to unnecessarily pessimistic decision-making within ICU. However, our results show that caution is also required when interpreting published unit survival rates. Unit mortality was 9% for patients transferred to our service. However, transfer back to the ICU is sometimes necessary. In our series, an additional 5% of patients did not achieve discharge home after requiring transfer back to ICU. Median survival was <1 month in this group. Combining these two groups (14%) is a more realistic reflection of the total mortality in our series and reflects the frailty of the patient cohort, but still compares favourably with the 29% mortality rate described via meta-analysis. 10 Nevertheless, our data show the importance of quantifying overall hospital survival rather than weaning unit survival alone. The use of existing studies to compare different care models is limited by differing referral patterns, difficulties in adjusting for co-morbidity and case-mix, and retrospective data collection. However, it is still informative to consider our results alongside those of the recent meta-analysis. 10 Increasing age and co-morbidity are associated with a worse prognosis, 22 and our population was slightly younger [61 (50 71) vs 64 (63 66) yr]. However, the meta-analysis excluded patients with chronic, progressive neuromuscular conditions on the grounds of expected poorer prognosis. We included all patients with neuromuscular conditions, including those with progressive conditions such as motor neurone disease. 23 Our data show that patients with progressive neuromuscular conditions can benefit from transfer to a specialized weaning unit; although weaning rates were lower than for other diagnostic groups, discharge home rates and long-term survival were similar. Throughout the last two decades, advances in ICU care may have impacted on weaning unit referral patterns. Assessed in 5 yr cohorts, patients with NMD and COPD were consistently the most common reasons for referral. Over time, fewer patients with chest wall disorders, such as scoliosis, were referred for weaning, reflecting improved respiratory surveillance and use of long-term NIV. 24 In contrast, transfers after postsurgical problems have increased, consistent with improved ICU survival after complex surgery, but at the expense of increasing chronic critical illness. 25 We found no change in overall weaning rates, weaning unit length of stay, or survival when assessed in 5 yr cohorts. However, between 1992 and 2011 the median duration of prior ICU care for patients in our series has almost doubled to 41 days. Although this may reflect increasing patient complexity and bed availability issues, it could represent missed opportunities for earlier transfer. We hope that the patient-level data provided in this series will help intensivists to identify patients who are most likely to benefit from early referral to a specialized weaning service, although we conclude that further studies are required to clarify the optimal timing for transfer. In the UK, this is especially important for those with COPD, in whom prognostic pessimism often limits intubation rates. 26 Our weaning rate for patients with COPD was higher than that for all the other main diagnostic groups. Timely identification and transfer of such patients could also increase ICU bed availability and reduce costs. In conclusion, we show that a multidisciplinary approach by a specialized weaning unit can provide a successful service model for patients who require liberation from prolonged IMV. In contrast to the results from other service models, we show that most patients transferred to our unit achieved discharge home, often with the continued use of nocturnal NIV. Median survival exceeded 2 yr, and more than a quarter of our patients survived longer than 5 yr. Authors contributions Study design/planning: M.G.D., T.G.Q., N.S.O., J.M.S., I.E.S. Study conduct: M.G.D., S.P.C. Data analysis: S.P.C., M.G.D. Writing paper: M.G.D. Revising paper: all authors Supplementary material Supplementary material is available at British Journal of Anaesthesia online. Acknowledgements The authors acknowledge the contribution of Mrs Danielle Horton, Ms Rebecca Chadwick, and Dr Martina Mason for their input into data collection and initial analysis. Declaration of interest None declared. References 1. Rose L, Fowler RA, Fan E, et al. Prolonged mechanical ventilation in Canadian intensive care units: a national survey. J Crit Care 2015; 30: Wagner DP. Economics of prolonged mechanical ventilation. Am Rev Respir Dis 1989; 140: S Lone NI, Walsh TS. Prolonged mechanical ventilation in critically ill patients: epidemiology, outcomes and modelling the potential cost consequences of establishing a regional weaning unit. Crit Care 2011; 15: R Nelson JE, Cox CE, Hope AA, Carson SS. Chronic critical illness. Am J Respir Crit Care Med 2010; 182:

7 Outcomes from a specialized weaning unit Kahn JM, Le T, Angus DC, et al. The epidemiology of chronic critical illness in the United States. Crit Care Med 2015; 43: Boles JM, Bion J, Connors A, et al. Weaning from mechanical ventilation. Eur Respir J 2007; 29: Kahn JM, Benson NM, Appleby D, Carson SS, Iwashyna TJ. Long-term acute care hospital utilization after critical illness. JAMA 2010; 303: NHS Modernisation Agency Report. Critical Care Programme. Weaning and Long Term Ventilation. London: NHS Modernisation Agency, Pilcher DV, Bailey MJ, Treacher DF, Hamid S, Williams AJ, Davidson AC. Outcomes, cost and long term survival of patients referred to a regional weaning centre. Thorax 2005; 60: Damuth E, Mitchell JA, Bartock JL, Roberts BW, Trzeciak S. Long-term survival of critically ill patients treated with prolonged mechanical ventilation: a systematic review and meta-analysis. Lancet Respir Med 2015; 3: Smith IE, Shneerson JM. A progressive care programme for prolonged ventilatory failure: analysis of outcome. Br J Anaesth 1995; 75: A14/S/a NHS standard contract for respiratory: complex home ventilation (adult) (2013/14). Available from spiratory-comp-home-vent.pdf (last accessed November 1, 2016). 13. MacIntyre NR, Epstein SK, Carson S, Scheinhorn D, Christopher K, Muldoon S. Management of patients requiring prolonged mechanical ventilation: report of a NAMDRC consensus conference. Chest 2005; 128: Bourke SC, Tomlinson M, Williams TL, Bullock RE, Shaw PJ, Gibson GJ. Effects of non-invasive ventilation on survival and quality of life in patients with amyotrophic lateral sclerosis: a randomised controlled trial. Lancet Neurol 2006; 5: Köhnlein T, Windisch W, Köhler D, et al. Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial. Lancet Respir Med 2014; 2: Buyse B, Meersseman W, Demedts M. Treatment of chronic respiratory failure in kyphoscoliosis: oxygen or ventilation? Eur Respir J 2003; 22: Quinnell TG, Pilsworth S, Shneerson JM, Smith IE. Prolonged invasive ventilation following acute ventilatory failure in COPD: weaning results, survival, and the role of noninvasive ventilation. Chest 2006; 129: Heinemann F, Budweiser S, Jörres RA, et al. The role of noninvasive home mechanical ventilation in patients with chronic obstructive pulmonary disease requiring prolonged weaning. Respirology 2011; 16: Schönhofer B, Euteneuer S, Nava S, Suchi S, Köhler D. Survival of mechanically ventilated patients admitted to a specialised weaning centre. Intensive Care Med 2002; 28: Cox CE, Martinu T, Sathy SJ, et al. Expectations and outcomes of prolonged mechanical ventilation. Crit Care Med 2009; 37: Kahn JM. Improving outcomes in prolonged mechanical ventilation: a road map. Lancet Respir Med 2015; 3: Dermot Frengley J, Sansone GR, Shakya K, Kaner RJ. Prolonged mechanical ventilation in 540 seriously ill older adults: effects of increasing age on clinical outcomes and survival. J Am Geriatr Soc 2014; 62: Chadwick R, Nadig V, Oscroft NS, Shneerson JM, Smith IE. Weaning from prolonged invasive ventilation in motor neuron disease: analysis of outcomes and survival. J Neurol Neurosurg Psychiatry 2011; 82: Ward S, Chatwin M, Heather S, Simonds AK. Randomised controlled trial of non-invasive ventilation (NIV) for nocturnal hypoventilation in neuromuscular and chest wall disease patients with daytime normocapnia. Thorax 2005; 60: Zilberberg MD, de Wit M, Pirone JR, Shorr AF. Growth in adult prolonged acute mechanical ventilation: implications for healthcare delivery. Crit Care Med 2008; 36: Wildman MJ, Sanderson CF, Groves J, et al. Survival and quality of life for patients with COPD or asthma admitted to intensive care in a UK multicentre cohort: the COPD and Asthma Outcome Study (CAOS). Thorax 2009; 64: Handling editor: Jonathan Thompson