Prolonged Weaning from Mechanical Ventilation: Results from Specialized Weaning Centers
A registry-based study from the WeanNet Initiative
; ; ; ; ; ;
Background: To accommodate the increasing number of patients requiring prolonged weaning from mechanical ventilation, specialized weaning centers have been established for patients in whom weaning on the intensive care unit (ICU) was unsuccessful.
Methods: This study aimed to determine both the outcome of treatment and the factors associated with prolonged weaning in patients who were transferred from the ICU to specialized weaning centers in Germany during the period 2011 to 2015, based on a nationwide registry covering all specialized weaning centers currently going through the process of accreditation by the German Respiratory Society.
Results: Of 11 424 patients, 7346 (64.3%) were successfully weaned, of whom 2236 were switched to long-term non-invasive ventilation; 1658 (14.5%) died in the weaning unit; and 2420 (21.2%) could not be weaned. The duration of weaning decreased significantly from 22 to 18 days between 2011 and 2015 (p <0.0001). Multivariate analysis revealed that the factor most strongly associated with in-hospital mortality was advanced age (odds ratio [OR] 11.07, 95% confidence interval [6.51; 18.82], p <0.0001). The need to continue with invasive ventilation was most strongly associated with the duration mechanical ventilation prior to transfer from the ICU (OR 4.73 [3.25; 6.89]), followed by a low body mass index (OR 0.38 [0.26; 0.58]), pre-existing neuromuscular disorders (OR 2.98 [1.88; 4.73]), and advanced age (OR 2.96 [1.87; 4.69]) (each p <0.0001).
Conclusions: Weaning duration has decreased over time, but prolonged weaning is still unsuccessful in one third of patients. Overall, the results warrant the establishment of specialized weaning centers. Variables associated with death and weaning failure can be integrated into ICU decision-making processes.
Invasive mechanical ventilation via an endotracheal or tracheal tube is a frequently performed procedure in the intensive care unit (ICU). When the acute condition that has required invasive mechanical ventilation is resolved, the ventilation must be discontinued. This process, known as weaning (Box), constitutes a significant part of intensive care (1, 2). A delay in weaning has been demonstrated to be associated with higher complication rates, longer hospital stays, and a lower ICU survival rate (3, 4). In 2005, following the International Consensus Conference (ICC), the process of weaning was categorized, according to expert opinion, into three groups based on the number, timing, and success of spontaneous breathing trials (SBT): simple, difficult, and prolonged weaning (Box). Subsequent studies showed that prolonged weaning is indeed associated with the worst prognosis (5, 6, 7, 8, 9, 10). However, the factors associated with weaning failure are not included in the ICC criteria.
Importantly, the ICC classification (2) does not address patients who have been transferred to specialized weaning centers because weaning on the ICU has failed. These patients may eventually be weaned successfully after several weeks of invasive mechanical ventilation, although some require long-term non-invasive ventilation (NIV) (11). By contrast, patients with definitive weaning failure are transferred to long-term care facilities or even discharged home to continue long-term invasive mechanical ventilation there (12, 13, 14). Failure of weaning, however, is associated with substantially greater morbidity and mortality and also represents an economic burden on the health care system (Box).
To improve the success of weaning, specialized weaning centers have been established in many countries, including Germany (15, 16, 17, 18, 19). In addition, the German Respiratory Society (DGP) has compiled detailed guidelines on the topic of prolonged weaning (16). The specialized weaning centers are accredited in accordance with defined criteria laid down by the DGP’s WeanNet initiative (20). This initiative may well be highly important in light of the steadily increasing number of patients with ventilation outside the hospital setting; an exponential increase was recently reported in the number of patients with established home mechanical ventilation who are subsequently readmitted to the hospital, from 24 845 in 2006 to 86 117 in 2016 (21), although detailed information on the development of prolonged weaning is still lacking.
As a condition for accreditation, specialized weaning centers are required to register their patients in a predefined database. This German registry formed the basis of this study, whose goal was to assess the success of treatment in weaning centers. Moreover, the study also aimed to assess changes in weaning outcome over a period of 5 years and to identify factors associated with successful weaning.
Ethics committee approval
WeanNet registry data from the period between 1 January 2011 and 1 January 2016 were analyzed. Although data collection was carried out prospectively for the purpose of the accreditation of German weaning centers, all analyses were performed exploratively without predefined hypotheses. The study was approved by the ethics committee of Witten/Herdecke University (eMethods).
Additional detailed information on the WeanNet initiative and the accreditation criteria is provided in the (eMethods).
Anonymous raw data were exported from the weaning registry and imported into the Statistica (version 10) software tool. Original variables were recoded for analysis where necessary and further time periods and additional control variables were calculated (eMethods). A plausibility analysis was performed according to predefined criteria. Source data verification was initiated in all cases of inconsistency or missing data with regard to weaning classification. Missing values were added and incorrect data were corrected in all cases of successful source data verification by November 2016.
All cases with persistently inconsistent values for treatment phases and weaning classification were subsequently excluded from the final analysis, as were all data sets entered before 1 January 2011. Only patients with prolonged weaning according to the ICC classification were analyzed (Figure 1). In the remaining cases, individual implausible parameters were set to “missing” according to predefined criteria before commencing the analysis. Weaning success was defined as discharge from the weaning unit without the need for continued invasive ventilation.
Statistical analysis was performed by data-quest GmbH in close collaboration with the DGP. The results were expressed as median and quartiles. A two-tailed p-value less than 0.05 was considered to be significant if not otherwise stated. All data were analyzed using Statistica (version 10) software. Univariate and multivariate analyses of the variables associated with the risk of mortality, invasive community mechanical ventilation, and long-term non-invasive community mechanical ventilation were performed, respectively. The local α level was adjusted for multiple tests according to the Bonferroni method (eMethods).
Data were analyzed from a total of 11 424 patients with prolonged weaning who were treated in 85 specialized weaning centers (Figure 1, eMethods, eTable 1).
Demographic data are provided in Table 1. Further data on comorbidities at admission and on mechanical ventilation are given in eTables 2 and 3, respectively.
The outcome of weaning changed significantly over the 5-year observation period (chi-squared = 47.4; df = 12; p <0.0001), with a decrease in weaning failure (Figure 2). Accordingly, the proportion of patients discharged alive and without invasive community ventilation improved from 60.0% in 2011 to 66.2% in 2015. Furthermore, over the same period there was a significant reduction in weaning duration from 22 days (interquartile range 12–36) to 18 days (interquartile range 11–31) (Kruskal–Wallis H = 67.7; p <0.0001) (eFigure 1).
Factors associated with in-hospital mortality in specialized weaning centers
Univariate analysis (total study cohort: N = 11 424) revealed that compared with average in-hospital mortality, patients with postoperative ventilation had a significantly higher mortality rate, while those on mechanical ventilation due to chronic obstructive pulmonary disease (COPD) had a significantly lower mortality rate (configural frequency analysis [CFA]: p <0.05 (eFigure 2, eMethods). Other factors associated with increased in-hospital mortality were advanced age, higher number of comorbidities, and low tidal volume (eTable 4, eMethods).
According to multivariate analysis, the factor most strongly associated with in-hospital mortality was advanced age, followed by low tidal volume and other factors (Table 2a).
Factors associated with weaning failure
Univariate analysis (hospital survivors: N = 9766) revealed that both COPD patients and those with neuromuscular disorders had significantly higher than average rates of weaning failure, while postoperative patients and those with acute respiratory distress syndrome (ARDS) had significantly lower than average weaning failure rates (CFA: p <0.05 (eFigure 3, eMethods). Further analysis demonstrated that the duration of mechanical ventilation before transfer to the weaning unit and high Eastern Cooperative Oncology Group (ECOG) performance status were the variables most strongly associated with unsuccessful weaning (eTable 4, eMethods).
According to multivariate analysis, the factor most strongly associated with unsuccessful weaning was a longer period of mechanical ventilation prior to transfer to the weaning unit, followed by a low body mass index (BMI), pre-existing neuromuscular disorders, advanced age, and others factors (Table 2b).
Factors associated with the need for long-term non-invasive ventilation
Univariate analysis (successful weaning: (16) N = 7346) revealed that patients with COPD, neuromuscular disorders, obesity hypoventilation syndrome, or restrictive thoracic disorders were significantly more likely to need long-term non-invasive ventilation (CFA: p <0.05 (eFigure 4, eMethods). Further analysis showed that the factors most strongly associated with the initiation/continuation of long-term non-invasive ventilation were identified as pre-existing home mechanical ventilation, high BMI, and young age (eTable 4, eMethods).
According to multivariate analysis, the factors most strongly associated with the need for continuation of non-invasive ventilation were pre-existing non-invasive ventilation, high BMI, and young age, followed by neuromuscular disorders and other factors (Table 2c).
This is the first large, national multicenter study to investigate weaning outcome and the associated determining factors in patients with prolonged weaning. Thus, our cohort comprised severely ill patients requiring weaning, a group that had not been sufficiently addressed and characterized in previous trials. The most important findings are as follows:
Prolonged weaning is a clinically significant and prevalent phenomenon in Germany. However, the true incidence of prolonged weaning, cannot be assessed on the basis of the existing data, since in Germany many patients with prolonged weaning are treated in facilities other than specialized weaning centers. Nevertheless, it should be noted that this cohort of patients with prolonged weaning has been markedly under-represented in previous trials, which focused solely on ICU patients. In the most recent epidemiological study from France, only 37 (1.4%) of 2729 patients were tracheostomized and classed as non-weanable (11), against 2420 (21.2%) of 11 424 patients in our study.
Furthermore, about two thirds of patients with prolonged weaning who could not be weaned on the ICU were successfully weaned after transfer to a specialized weaning unit. We also found that the weaning success rate improved over the 5-year observation period. This may represent a learning curve, although changes in patient selection cannot be excluded with certainty. Thus, the establishment of specialized weaning centers based on clearly defined requirements seems warranted, although this needs to be further verified by future studies using a prospective approach.
This study defined factors associated with weaning outcome: The factor most strongly associated with in-hospital mortality on the weaning unit was advanced age of the patient. Other factors, including the duration of mechanical ventilation on the transferring ICU, low BMI, pre-existing neuromuscular disorders, and advanced age, were also strongly associated with weaning failure in patients who survived their stay in hospital. Finally, several factors were found to be associated with the need to continue non-invasive ventilation in the long term after hospital discharge, the most important being pre-existing long-term non-invasive ventilation, obesity hypoventilation syndrome, neuromuscular disorders, and younger age.
This analysis has important clinical implications. First, in patients requiring intensive care it has to be anticipated that weaning from mechanical ventilation may be considerably delayed or even impossible in the case of advanced age or the presence of conditions that predispose to chronic respiratory insufficiency. Second, the results of this study argue in favor of the establishment of weaning centers devoted to this specific subset of patients, since it has been postulated that there will be a steady increase in the complexity and severity of the underlying conditions in ICU patients, potentiating the likelihood of unsuccessful weaning (22). This is extremely important in view of the steadily increasing number of patients with community ventilation in Germany (21). This study has shown that the time spent on the ventilator in the ICU is negatively associated with weaning success. This association was also found in a recent small single-center study (23). Early transfer of the patient to a specialized weaning unit thus seems justifiable, although this aspect was not specifically addressed in our study.
Despite the successful results achieved in specialized weaning centers, 25% of patients who did not die in the weaning unit could not be weaned and were discharged to long-term invasive ventilation in the community. However, the continuation of invasive ventilation in a community setting subsequent to weaning failure is associated with severe impairments in health-related quality of life, especially in patients with COPD; this may well raise ethical concerns with regard to current ICU treatment practices (24,25). Moreover, the present study has identified clinical parameters that are associated with weaning failure. This may facilitate the decision-making process on the ICU, but further prospective investigations are needed in this regard. Undoubtedly this would initiate an ethical debate on the exact circumstances under which mechanical ventilation should be implemented. Self-evidently, the precise circumstances of the individual patient must always be taken into account.
This study has some limitations, mostly related to patient selection. First, the data refer primarily to patients with prolonged weaning who were transferred to specialized weaning centers. The findings therefore cannot be extrapolated to patients with prolonged weaning who remain on the ICU that provided the initial treatment. In this regard, it should be noted that there were no predefined criteria for or against transfer to a weaning center. This partially limits the generalizability of the findings. Nevertheless, it should be emphasized that the high numbers of patients and centers broadly reflect clinical reality. Second, although patients were treated at weaning centers, it cannot be ruled out that some patient data were not registered. Specifically, patients had to provide informed written consent for inclusion of their data in the registry; it thus seems unlikely that the data of the most severely ill patients, who died early in the course of their treatment, are all included in the weaning registry. Moreover, although plausibility analyses was performed according to predefined criteria, source data verification was not possible in all cases. The limits of plausibility were set arbitrarily, and data correctness for values within these limits could not be guaranteed. Third, the plausibility analyses led to exclusion of some patients from analysis due to inconsistencies in dates and classification, but the absence of these cases (with the consequence of incomplete data sets) may have reduced statistical strength. Fourth, data from the transferring ICU were only sparsely available, which may have prevented more detailed analysis. For example, the true weaning duration in the transferring hospital could not be documented. Finally, we did not address the status of patients who were transferred from the weaning center to another hospital.
In conclusion, patients with prolonged weaning who are transferred to a specialized weaning center following several weeks of invasive ventilation on the ICU represent an epidemiologically important subset of weaning patients that has not been sufficiently characterized by previous studies. In this setting, although the outcome of weaning is still significantly restricted, it has continued to improve over time. The establishment of variables associated with hospital mortality and weaning failure may be helpful in designing future studies on decision-making processes in intensive care medicine.
The study was supported by the German Respiratory Society (DGP).
The authors would like to thank S. Dieni for editorial assistance.
Conflict of interest statement
Prof Windisch’s study group has received research grants from Weinmann, Vivisol, Heinen und Löwenstein, VitalAire (all Germany), and Philips Respironics (USA). Prof Windisch has received speaking fees from Löwenstein and ResMed.
Dr. Geiseler has received speaking fees from Löwenstein, Philips, and Berlin-Chemie.
Prof. Pfeifer has received speaking fees from ResMed.
Herr Suchi has received research grants from the German Respiratory Society (DGP) for statistical analyses.
Prof. Schönhofer is first author of the German consensus-based guideline “Prolonged Weaning.”
The remaining authors declare that no conflict of interest exists.
Manuscript received on 31 May 2019, revised version accepted on 17 December 2019
Prof. Dr. med. Wolfram Windisch
Lungenklinik, Kliniken Köln gGmbH
Lehrstuhl für Pneumologie, Universität Witten/Herdecke
Fakultät für Gesundheit/Department für Humanmedizin
Ostmerheimer Str. 200, 51109 Köln, Germany
Cite this as:
Windisch W, Dellweg D, Geiseler J, Westhoff M, Pfeifer M, Suchi S, Schönhofer B: Prolonged weaning from mechanical ventilation: results from specialized weaning centers—a registry-based study from the WeanNet Initiative.
Dtsch Arztebl Int 2020; 117: 197–204. DOI: 10.3238/arztebl.2020.0197
eMethods, eTables, eFigures
Herdecke University: Prof. Dr. med. Wolfram Windisch
Kloster Grafschaft Hospital GmbH, Academic Teaching Hospital of Marburg University, Schmallenberg-
Grafschaft: PD Dr. med. Dominic Dellweg
Vest Hospital, Department of Internal Medicine IV: Respiratory, Ventilation, and Sleep Medicine, Academic Teaching Hospital of Bochum University, Marl: Dr. med. Jens Geiseler
Department of Respiratory Medicine, Hemer Lung Hospital, Witten/Herdecke University: Dr. med. Michael Westhoff
Department of Respiratory Medicine, Donaustauf Hospital, Regensburg University: Prof. Dr. med. Michael Pfeifer
Data-quest GmbH – Statistics and Data Management, Göttingen: Stefan Suchi
Department of Respiratory, Intensive Care, and Sleep Medicine, Siloah Hospital, Hanover: Prof. Dr. med. Bernd Schönhofer
|1.||McConville JF, Kress JP: Weaning patients from the ventilator. N Engl J Med 2012; 367: 2233–9 CrossRef MEDLINE|
|2.||Boles JM, Bion J, Connors A, et al: Weaning from mechanical ventilation. Eur Respir J 2007; 29: 1033–56 CrossRef MEDLINE|
|3.||Epstein SK, Ciubotaru RL: Independent effects of etiology of failure and time to reintubation on outcome for patients failing extubation. Am J Respir Crit Care Med 1998; 158: 489–93 CrossRef MEDLINE|
|4.||Thille AW, Richard J-CM, Brochard L: The decision to extubate in the intensive care unit. Am J Respir Crit Care Med 2013; 187: 1294–302 CrossRef MEDLINE|
|5.||Peñuelas O, Frutos-Vivar F, Fernández C, et al: Characteristics and outcomes of ventilated patients according to time to liberation from mechanical ventilation. Am J Respir Crit Care Med 2011; 184: 430–7 CrossRef MEDLINE|
|6.||Tonnelier A, Tonnelier J-M, Nowak E, et al: Clinical relevance of classification according to weaning difficulty. Respir Care 2011; 56: 583–90 CrossRef MEDLINE|
|7.||Sellares J, Ferrer M, Cano E, et al: Predictors of prolonged weaning and survival during ventilator weaning in a respiratory ICU. Intensive Care Med 2011; 37: 775–84 CrossRef MEDLINE|
|8.||Funk GC, Anders S, Breyer MK, et al: Incidence and outcome of weaning from mechanical ventilation according to new categories. Eur Respir J 2010; 35: 88–94 CrossRef MEDLINE|
|9.||Pu L, Zhu B, Jiang L, et al: Weaning critically ill patients from mechanical ventilation: a prospective cohort study. J Crit Care 2015; 30: 862. e7–13 CrossRef MEDLINE|
|10.||Jeong BH, Ko MG, Nam J, Yoo H, et al: Differences in clinical outcomes according to weaning classifications in medical intensive care units. Plos One 2015; 10: e0122810 CrossRef MEDLINE PubMed Central|
|11.||Schönhofer B, Euteneuer S, Nava S, et al: Survival of mechanically ventilated patients admitted to a specialised weaning centre. Intensive Care Med 2002; 28: 908–16 CrossRef MEDLINE|
|12.||Béduneau G, Pham T, Schortgen F, et al: Epidemiology of weaning outcome according to a new definition. The WIND Study. Am J Respir Crit Care Med 2017; 195: 772–83 CrossRef MEDLINE|
|13.||Windisch W, Geiseler J, Simon K, et al: German national guideline for treating chronic respiratory failure with invasive and non-invasive ventilation: Revised edition 2017 – Part 1. Respiration 2018; 96: 66–97 CrossRef MEDLINE|
|14.||Windisch W, Geiseler J, Simon K, et al: German national guideline for treating chronic respiratory failure with invasive and non-Invasive ventilation. Revised edition 2017: Part 2. Respiration 2018; 96: 171–203 CrossRef MEDLINE|
|15.||Hannan LM, Tan S, Hopkinson K, et al: Inpatient and long-term outcomes of individuals admitted for weaning from mechanical ventilation at a specialized ventilation weaning unit. Respirology 2013; 18: 154–60 CrossRef MEDLINE|
|16.||Schönhofer B, Geiseler J, Dellweg D, et al.: Prolongiertes Weaning – S2k-Leitlinie Deutschen Gesellschaft für Pneumologie und Beatmungsmedizin (eds.). Pneumologie 2019; 73: 723–814 CrossRef MEDLINE|
|17.||Davies MG, Quinnell TG, Oscroft NS, et al: Hospital outcomes and long-term survival after referral to a specialized weaning unit. Br J Anaesth 2017; 118: 563–9 CrossRef MEDLINE|
|18.||Scheinhorn DJ, Chao DC, Stearn-Hassenpflug M, et al: Post-ICU mechanical ventilation: treatment of 1,123 patients at a regional weaning center. Chest 1997; 111: 1654–9 CrossRef MEDLINE|
|19.||Scheinhorn DJ, Hassenpflug MS, Votto JJ, et al: Post-ICU mechanical ventilation at 23 long-term care hospitals: a multicenter outcomes study. Chest 2007; 131: 85–93 CrossRef MEDLINE|
|20.||Schönhofer B, Geiseler J, Pfeifer M, et al: WeanNet: a network of weaning units headed by pneumologists. Pneumologie 2014; 68: 737–42 CrossRef MEDLINE|
|21.||Karagiannidis C, Strassmann S, Callegari J, et al: Epidemiologische Entwicklung der außerklinischen Beatmung: Eine rasant zunehmende Herausforderung für die ambulante und stationäre Patientenversorgung. Dtsch Med Wochenschr 2019; 144: e58–e63 CrossRef MEDLINE|
|22.||Polverino E, Nava S, Ferrer M, et al: Patients‘ characterization, hospital course and clinical outcomes in five Italian respiratory intensive care units. Intensive Care Med 2010; 36: 137–42 CrossRef MEDLINE|
|23.||Magnet MS, Bleichroth H, Huttmann SE, et al: Clinical evidence for respiratory pump insufficiency predicts weaning failure in long-term ventilated, tracheotomised patients: a retrospective analysis. J Intensive Care 2018; 16; 6: 67 CrossRef MEDLINE PubMed Central|
|24.||Huttmann SE, Windisch W, Storre JH: Invasive home mechanical ventilation: living conditions and health-related quality of life. Respiration 2015; 89: 312–21 CrossRef MEDLINE|
|25.||Huttmann SE, Magnet FS, Karagiannidis C, et al: Quality of life and life satisfaction are severely impaired in patients with long-term invasive ventilation following ICU treatment and unsuccessful weaning. Ann Intensive Care 2018; 8: 38 CrossRef MEDLINE PubMed Central|
|26.||Deutsche Gesellschaft für Pneumologie und Beatmungsmedizin e. V.: Erhebungsbogen zur Zertifizierung von Weaning-Zentren. https://pneumologie.de/fileadmin/user_upload/Erhebungsbogen_zur_Zertifizierung_Weaning-Zentren_Version_06.pdf (last accessed on 10 February 2020).|