Journal of Critical Care (2006) 21, 156 – 162
Experience with a step-down respiratory care center at a tertiary referral medical center in Taiwan Jian Su MD, Chang-Yi Lin MD*, Pei-Jan Chen MD, Fung J. Lin MD, Shiow-Kwan Chen RRT, Hsu-Tah Kuo MD Chest Division, Department of Internal Medicine, Mackay Memorial Hospital, Taipei 104, Taiwan Keywords: Weaning; Prolonged mechanical ventilation; Respiratory care center
Abstract Objective: The aim of the study was to describe the outcome of patients after 1 year’s implementation of an integrated delivery system for respiratory care mandated by the National Health Insurance Bureau in Taiwan. Design: A retrospective observational study was conducted in a step-down respiratory care center (RCC). Patients: Patients included adults receiving prolonged mechanical ventilation (z21 days). Measurements and Main Results: A total of 224 cases were available for review; 108 (48.2%) patients were successfully weaned. Those who failed weaning had a longer stay in the intensive care unit and RCC (25.1 vs 20.9 and 31.4 vs 18.6 days, P b .05), but there were no differences in the patients’ ages (74.3 vs 70.4 years, P = .17) or the Simplified Acute Physiology Score II (52 vs 46.9, P = .18) before admission to the RCC. After discharge from the RCC, only 4.9% of the patients still on a ventilator were weaned within 1 year. Patients who failed weaning in the RCC had a shorter overall survival (5.2 vs 10.4 months, P b .05) and a lower 1-year survival (23.6% vs 44.6%, P b .05). Conclusion: Patients admitted to the RCC were still critically ill. Patients who failed weaning in the RCC had had a longer intensive care unit and RCC stay and a worse outcome after leaving the RCC. D 2006 Elsevier Inc. All rights reserved.
1. Introduction Intensive care units (ICUs) were developed in the 1960s to provide specialized care for critically ill patients . Acute respiratory distress necessitating temporary ventilatory support is a very common reason for admission to an ICU. However, approximately 3% to 6% of patients admitted to an ICU require prolonged mechanical ventilation (PMV) * Corresponding author. Tel.: + 886 2 25433535x2257. E-mail address: [email protected]
(C.-Y. Lin). 0883-9441/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcrc.2005.10.004
beyond the acute stage [2-5]. Most patients on PMV are otherwise stable and no longer need sophisticated ICU monitoring. However, they do require specialized respiratory care to manage their PMV. Keeping these patients in the ICU for such care reduces available ICU beds, increases medical expenses, and leads to suboptimal quality of care [6 -8]. Therefore, attempts have been made to develop weaning protocols and respiratory care teams to manage such patients in a specialized intermediate unit [9- 15]. In Taiwan, an integrated delivery system has been developed in which patients on ventilatory support for more than 14 days are
Experience with a step-down RCC
enrolled. The profiles of all patients enrolled in the integrated delivery system are to be reported to the National Health Insurance Bureau (NHIB), with this information available online to other hospitals as needed. If the overall condition is stable, the patient is transferred to a respiratory care center (RCC) within 21 days after starting mechanical ventilation (MV). The NHIB reimburses for treatment in an RCC for a maximum of 42 days. If MV is still required at that point, patients are transferred to an extended care facility, unless their condition has otherwise deteriorated, in which case they return to an acute care facility. Extended care facilities are to report to the NHIB and the original hospital monthly on the condition of such patients. Mackay Memorial Hospital is a tertiary referral center with 1200 beds in 2 facilities in the greater Taipei area. There are a total of 8 adult ICUs including 1 coronary care unit, 1 cardiovascular surgical unit, 3 medical intensive care units, 1 burn unit, and 2 surgical intensive care units. A 21-bed RCC was established in October 2000 according to NHIB regulations to accommodate patients from all the adult ICUs. It is staffed by an integrated team of chest and critical care physicians, nurses, respiratory therapists, dietitians, and social workers. In this article, we report the results after 1 year of operation of the RCC.
2. Methods The medical records of all patients on PMVadmitted to the RCC from October 2000 to September 2001 were reviewed. Criteria for admission to the unit included being older than 17 years, MV for more than 14 days, arterial oxygenation of more than 60 mm Hg with the fraction of inspired oxygen of 0.55 or less, and positive end-expiratory pressure of 10 cm H2O or less. Patients were not admitted to the RCC if they were in shock; had massive gastrointestinal bleeding, acute myocardial infarction, or severe liver failure; or if they were postoperative and additional surgery was planned. If patients were diagnosed with contagious tuberculosis, they were transferred out of the RCC to a tuberculosis isolation ward for further treatment. Age, sex, history of ventilatory dependency, duration of ventilation, underlying diseases, weaning attempts, and outcome were analyzed. Successful weaning was defined as complete independence from MV for at least
Table 1 Characteristics of 224 patients on mechanical ventilation Characteristic
Mean F SD or %
Age (y) Male SAPS Duration of MV before transfer to RCC Length of RCC stay Comorbidity
73.9 F 12.5 54% 49.6 F 11.6 23.1 F 11.7 d 25.3 F 14.7 d 1.8 F 0.9
Table 2 Characteristics of patients who succeeded at and failed weaning Age (y) Pre-RCC ICU stay (d) RCC stay (d) Glasgow Coma Scale Albumin SAPS a
70.4 F 14.5 20.9 F 8.4
74.3 F 12.8 25.1 F 13.7
18.6 F 10.8 7.7 F 2.9
31.4 F 15.1 7.4 F 3.1
2.6 F 0.5 46.9 F 10.4
2.4 F 0.5 52.0 F 12.2
72 hours. Patients who still required nocturnal ventilation were considered to have failed weaning. On admission to the RCC, a pulmonary intensive care specialist who directed the unit evaluated the patient’s general condition and reviewed records of previous weaning trials. An individualized weaning plan was then implemented and tailored according to the patient’s condition. For the purpose of analysis, patients were classified according to why they required PMV rather than by the reason for initiation of ventilatory support in the first place. All patients were on invasive MV (Puritan-Bennett 7200, Nellcor Puritan Bennett Inc, Hayward, CA) and continuous cardiovascular monitoring. Weaning started with a trial of partial ventilatory support with pressure support or by trying spontaneous breathing on a T-piece as suggested by Esteban and Alia . For a T-piece trial, spring-load high-flow continuous positive airway pressure was sometimes used to match patients’ demand. Patients who were alert, able to cooperate, and who could expectorate very well were extubated if they could tolerate 2 hours of partial pressure support at 8 cm H2O or a T-piece trial. Noninvasive ventilation was sometimes used if extubation failed. In patients with a tracheotomy, sometimes, an oxygen collar was used instead of a T-piece during a spontaneous trial in order to keep up with patient’s comfort. Catheters, central lines, parenteral nutrition, urinary catheters, and feeding tubes were removed as soon as possible. During the day, 2 residents and 1 pulmonary critical care specialist were on duty to care for the patients. Physicians in other subspecialties were available for consultation as necessary. At night, 1 resident was on duty to deal with emergencies. During the day, the unit director evaluated each patient in the unit for suitability for weaning. Two respiratory therapists were on duty for each 8-hour shift. Nurses were specifically trained in caring for patients on PMV. One nurse cared for a maximum of 3 patients during each 8-hour shift. A dietitian visited each patient at least thrice weekly to evaluate his or her nutritional status and to give advice on nutritional support. Psychological support included visits by social workers, encouragement to exercise by the family, and watching television. Rehabili-
J. Su et al.
Table 3 Reasons for ventilatory dependency in patients who succeeded or failed weaning in the RCC Disease category
Total no. of cases
Pulmonary disease COPD Pneumonia Lung cancer Pneumoconiosis Neurologic disease Intra-cranial hemorrhage Hypoxic encephalopathy Stroke C-spine injury Brain tumor Cardiovascular disease Congestive heart failure Myocardial infarction Gastrointestinal disease Colon cancer postoperative Cholelithiasis postoperative Liver cirrhosis Upper gastrointestinal bleeding Other gastrointestinal cancer Ileus postoperative Sepsis Renal failure Others
36 20 12 3 1 33 17 10 5 1 0 10 9 1 8 4
45 24 14 5 2 19 13 4 0 1 1 20 20 0 11 2
0 9 6 6
1 5 10 6
14 16 12
tation included passive range-of-motion exercises, tilt table to standing, ergometry (upper and lower extremities), inspiratory muscle training, and ambulation with portable ventilators if possible. Patients with terminal cancer were equipped with patient-controlled analgesia for pain control as needed and received hospice care from specialists. Patient outcome, defined as weaning success or failure, was assessed by chart review. After patients left the RCC, we followed them by telephone contact and review of outpatient records. The hospital’s financial department provided data on medical expenses in the RCC and ICU.
3. Results From October 2000 to September 2001, 272 patients were admitted to our RCC. Forty-eight patients were excluded from the analysis, including 4 who had already been on long-term MV, 5 with tuberculosis, 10 who left against medical advice as soon as they were admitted, and 29 with incomplete data, leaving a total of 224 patients available for the study. The mean age of the patients was 73.9 years, and they had a mean Simplified Acute Physiology Score II (SAPS) of 49.6 (Table 1). A total of 108 patients (48.2%) were successfully weaned, whereas 116 failed weaning within the 42-day stay. The latter group included 55 (24.6% of the total) who died. Those who failed weaning had significantly longer stays in the ICU (25.1 vs 20.9 days, P = .003) and RCC (31.4 vs 18.6 days, P b .001). The average age (74.3 vs 70.4 years, P = .17) and simplified acute physiology score II (52 vs 46.9, P = .18) were slightly higher in the failure group, but were not significantly different (Table 2). Patients had been admitted to the ICU primarily with pulmonary, neurologic, or cardiovascular disease. However, pulmonary disease accounted for the largest group of patients on PMV admitted to the RCC. Only 36 (44.4%) of 81 patients with pulmonary disease were successfully weaned, with chronic obstructive pulmonary disease (COPD) being the main problem accounting for weaning failure in this group. Only 20 (45.5%) of 44 patients with COPD were successfully weaned. Pneumonia was the second main cause of weaning failure, with only 12 (46.2%) of 26 patients successfully weaned. Among patients with cardiovascular disorders, most had serious heart disease such as critical aortic valve stenosis or cardiomyopathy. Only 10 (33.3%) of 30 patients in this category were weaned. Gastrointestinal malignancy was present in 13 patients, of whom 6 had colon cancer and
2.1. Statistical analysis The Kaplan-Meier method was used to estimate the cumulative probability of survival as a function of the number of months after RCC discharge for patients who succeeded at or failed weaning. Differences were tested for significance with the log-rank test. All continuous data are expressed as the mean F SD and were analyzed by using an unpaired Student t test with a 2-tailed P value. A P value of less than .05 was considered to be statistically significant. All data were analyzed using SPSS 11 software for Windows (SPSS, Chicago, Ill).
Fig. 1 Cumulative survival after RCC discharge for patients who succeeded at or failed weaning.
Experience with a step-down RCC required MV postoperatively. Four of these patients were weaned in the RCC. Four other patients with cancer (2 cases of esophageal cancer, 1 of hepatoma, and 1 of pancreatic cancer) died in the RCC. Sixteen patients had renal failure requiring regular hemodialysis, and only 6 (37.5%) were successfully weaned (Table 3). During the study period, 169 patients survived to leave the RCC, 108 of whom were successfully weaned and 61 failed and were still on MV at discharge. All of the latter were transferred to other facilities, as insurance does not cover home ventilation services. Of the 169 patients discharged, 19 were lost to follow-up (13 who had been weaned and 6 still on MV). Of the 55 patients who had failed weaning and for whom follow-up information was available, 47 died within 1 year (Fig. 1). Compared with those who were successfully weaned, the patients who had failed had a lower 1-year survival (23.6% vs 44.6%) and shorter survival (5.2 vs 10.4 months, P b .01). After 1 year, only 3 of the surviving patients were off MV, with 5 (8.2%) still needing ventilatory support. In 1999, before the RCC was established, the average ICU stay for patients on MV was 39.5 days. This decreased to 24.8 days in 2001. Over this same period, there was no change in the percentage of patients on MV who were successfully weaned in the ICU (46% vs 43%) nor in the total duration of MV (59.7 vs 57.3 days). The average daily cost of care for patients on MV dropped from $411.1 in the ICU to $234.9 in the RCC.
4. Discussion In our heterogeneous population, we found that although patients had been in an ICU for more than 14 days, they were still critically ill as reflected by the their high illness severity score. Scheinhorn et al  found that patients in a regional weaning center were as critically ill as patients commonly admitted to the ICU. Most of our patients were elderly and had significant comorbidity, particularly renal insufficiency and coronary artery disease. Seneff  also noted that comorbidity complicated the management of patients on PMV. This is one of the reasons RCC is necessary as a stepdown unit because most patients requiring prolonged PMV are not well enough to be transferred directly from the ICU to an ordinary medical ward, nursing home, or home care. We successfully weaned 48.2% of the patients admitted to our RCC, whereas 24.6% died while in the RCC. Although several studies on weaning from PMV have been published, comparison is difficult because of differences in patient populations and in the definitions of PMV and of successful weaning. Reported success in weaning varies widely from 32% to 87% of patients [17 - 22]. For example, Schfnhofer et al  reported successful weaning in 68% and a mortality of 24.3%. They included patients who required noninvasive nocturnal ventilation as successfully weaned, and they excluded patients with terminal diseases,
159 which differed from our criteria. Our RCC admits patients from both the medical and the surgical ICU intensive units, more like an earlier study by Scheinhorn et al . They weaned 53% of 421 patients and had a mortality of 27.6%. They also found, as did we, that the mean duration of ventilator dependency before transfer from the ICU and the length of stay in the RCC were shorter in patients who were successfully weaned. This suggests that patients who are successfully weaned may be in better condition to begin with, allowing more rapid stabilization in the ICU. The slightly lower SAPS in our patients who succeeded might reflect this as well. Another factor that might influence outcome is age. Previous studies have varied in terms of results for elderly patients [23,24]. Two thirds of our patients were older than 70 years. However, we could not demonstrate a statistical difference in age between those who failed and those who succeeded at weaning. Perhaps if we had had a larger proportion of younger patients, we might have found a significant difference. Reported weaning rates for patients with pulmonary disease range from 25% to 41% [17,18,25], and our results (44.4% of patients with pulmonary disease successfully weaned) are in line with these other studies. Noninvasive nocturnal ventilation has been investigated as a means to improve results in patients with COPD, but it is only beneficial in patients with less severe disease . Patients with decompensated cardiovascular disease performed poorly, with only one third being weaned. Schfnhofer et al  reported a mortality of 56.3% for similar patients. Thoner  found an in-hospital mortality of 66.7% in patients with a previous history of cardiopulmonary compromise. Using continuous electrocardiographic monitoring, Chatila et al  detected a 6.4% incidence of cardiac ischemia during weaning in patients at high risk for coronary artery disease. This might explain the poor outcome for these patients. Renal failure also predicted poor results, with only 37.5% of our patients on hemodialysis being weaned. Chao et al  reported that only 13% of patients with renal failure could be weaned compared with 58% of those who had a serum creatinine level of less than 2.5 mg/dL. Their patients with renal failure also had a higher mortality rate (84%). Dujic et al  found abnormalities in lung diffusing capacity and decreased expiratory flow and vital capacity in patients with uremia, which were partially reversed by hemodialysis. Most of their patients had several comorbid diseases such as diabetes and cardiovascular disorders. Our results were similar to those reported by Noble et al  who studied patients in a Scottish ICU and found a high mortality in patients with both respiratory and renal failure (64.8%). Only 33% (4/11) of those with preexisting chronic renal failure survived. The differences in reported outcome for patients with renal failure may again be because of the heterogeneous populations. Chao and colleagues reported on patients who had been on ventilation for nearly 50 days, whereas
160 Noble et al investigated patients in the acute stage of illness. Our patients were transferred directly from an acute setting, which might explain the similarity of our results to those of Noble’s group. Patients who failed weaning had a poor outcome after leaving the RCC, with a low 1-year survival and a poor prognosis for subsequent weaning. Scheinhorn et al  also found a low survival in patients on PMV in a long-term care facility. Swinburne et al  showed that the duration of MV was related to the survival for patients older than 80 years. By contrast, 65% of patients who had been successfully weaned in the RCC were alive 1 year after discharge. Comorbid disease was also health-related to outcome, with a lower survival for patients with liver, renal, heart, or lung disease. Spicher and White  also found that patients intubated postoperatively or with neurologic disease had a higher survival than did those with cardiac and pulmonary diseases, similar to our results. Most patients who failed weaning had renal, cardiac, or pulmonary disorders, and they also had the poorest 1-year outcome. A strong motivation for establishing an RCC is the growing cost of medical care, a major public health concern [34,35]. Both patients and health care providers are troubled by increasing restraints on health care resources. Arguments have been advanced to deny life support to the very aged as a strategy to conserve limited sources . Our experience demonstrates the economic benefits of an RCC, where care was considerably less costly than in an ICU. Part of this is because of a higher patient-to-staff ratio, 3 to 1 in the RCC, compared with 2 to 1 in our ICUs. In addition, we managed patients with terminal disease in the RCC, individuals for whom it may be difficult to justify the expense of a prolonged ICU stay. Scheinhorn et al  found that, although the total duration of MV may not be shortened by use of an RCC, the length of stay in the ICU is definitely shortened when a step-down unit is available. From an economic viewpoint, an RCC then makes good sense.
5. Conclusions An RCC provides good management for patients on PMV at a lower cost than maintaining them in an ICU. A substantial proportion of patients admitted to the RCC can be successfully weaned, but the long-term outlook for patients who fail weaning in the RCC remains grim. In an era of increasing concern about limited resources, this type of specialized step-down unit is a good alternative to the ICU.
References  Colice GL. Historical perspective on the development of mechanical ventilation Principles and practice of mechanical ventilation. 1st ed. Boston7 Little, Brown and Co; 1995. p. 1 - 35.  Stauffer JL, Fayter NA, Graves B, Cromb M, Lynch JC, Goebel P. Survival following mechanical ventilation for acute respiratory failure in adult men. Chest 1993;104:1222 - 9.
J. Su et al.  Daly BJ, Rudy EB, Thompson KS, Happ MB. Development of a special care unit for chronically critically ill patients. Heart Lung 1991;20:45 - 52.  Lemaire F. Difficult weaning. Intensive Care Med 1993;19:S69-S73.  Carson SS, Bach PB, Brzozowski L, Leff A. Outcomes after longterm acute care. An analysis of 133 mechanically ventilated patients. Am J Respir Crit Care Med 1999;159:1568 - 73.  Bach PB, Carson SS, Leff A. Outcome and resource utilization for patients with prolonged critical illness managed by university-based or community-based subspecialists. Am J Respir Crit Care Med 1998;158:1410 - 5.  Cohen IL, Lambrinos J, Fein A. Mechanical ventilation for the elderly patient in intensive care: incremental charges and benefits. JAMA 1993;269:1025 - 9.  Wagner DP. Economics of prolonged mechanical ventilation. Am Rev Respir Med 1989;140(Suppl):S14 - 8.  Cohen IL, Bari N, Strosberg MA, Weinberg PF, Wacksman RM, Millstein BH, et al. Reduction of duration and cost of mechanical ventilation in an intensive care unit by use of a ventilatory management team. Crit Care Med 1991;19:1278 - 84.  Zibrak JD, O’Donnell CR, Munroe S. Weaning patients from mechanical ventilatory support: results of weaning team managed by respiratory care specialists. Am Rev Respir Med 1992;145:A516.  Scheinhorn DJ, Chao DC, Stearn-Hassenpflug M, Wallace WA. Outcomes in post-ICU mechanical ventilation: a therapist-implemented weaning protocol. Chest 2001;119:236 - 42.  Petty TL, Lakshminaryan S, Sahn SA, Zwillich CW, Nett LM. Intensive respiratory care unit. Review of ten years’ experience. JAMA 1975;233:34 - 7.  Gracy DR, Viggiano RW, Naessens JM, Hubmayr RD, Silverstein MD, Koenig GE. Outcomes of patients admitted to a chronic ventilatordependent unit in an acute-care hospital. Mayo Clin Proc 1992; 67:131 - 6.  Scheinhorn DJ, Chao DC, Stearn-Hassenpflug M, LaBree LD, Heltsley DJ. Post-ICU mechanical ventilation: treatment of 1123 patients at a regional weaning center. Chest 1997;111:1654 - 9.  Seneff MG, Zimmerman JE, Knaus WA, Wagner DP, Draper EA. Predicting the duration of mechanical ventilation. The importance of disease and patient characteristics. Chest 1996;110:469 - 79.  Esteban A, Alia I. Clinical management of weaning from mechanical ventilation. Intensive Care Med 1998;24:999 - 1008.  Schfnhofer B, Euteneuer S, Nava S, Suchi S, Kohler D. Survival of mechanical ventilated patients admitted to a specialised weaning centre. Intensive Care Med 2002;28:908 - 16.  Scheinhorn DJ, Artinian BM, Catlin JL. Weaning from prolonged mechanical ventilation: the experience at a regional weaning center. Chest 1994;105:534 - 9.  Confalonieri M, Gorini M, Ambrosino N, Mollica C, Corrado A. Scientific Group on Respiratory Intensive Care of the Italian Association of Hospital Pneumonologists. Respiratory intensive care unit in Italy: a national census and prospective cohort study. Thorax 2001;56:373 - 8.  Gracy DR, Hardy DC, Naessens JM, Silverstein MD, Hubmayr RD. The Mayo ventilator-dependent rehabilitation unit: a 5-year experience. Mayo Clin Proc 1997;72:13 - 9.  Bagley PH, Cooney E. A community-based regional ventilator weaning unit: development and outcomes. Chest 1997;111:1024 - 9.  Dasgupta A, Rice R, Mascha E, Litaker D, Stoller JK. Fouryear experience with a unit for long-term ventilation (respiratory special care unit) at the Cleveland Clinic Foundation. Chest 1999; 116:447 - 55.  Goldberg AI. Life-sustaining technology and the elderly: prolonged mechanical ventilation factors influencing the treatment decision. Chest 1988;94:1277 - 82.  Nicolas F, Le Gall JR, Alperovitch A, Loirat P, Villers D. Influence of patients age on survival, level of therapy and length of stay in intensive care units. Intensive Care Med 1987;13:9 - 13.
Experience with a step-down RCC  Hill AT, Hopkinson RB, Stableforth DE. Ventilation in a Birmingham intensive care unit 1993-95: outcome for patients with chronic obstructive pulmonary disease. Respir Med 1998;92:156 - 61.  Plant PK, Owen JL, Elliott MW. Early use of non-invasive ventilation for acute exacerbation of chronic obstructive pulmonary disease on a general respiratory ward: a multicentre randomized trial. Lancet 2000;335:1931 - 5.  Thoner J. Outcome and costs of intensive care. A follow-up study on patients requiring prolonged mechanical ventilation. Acta Anaesthesiol Scand 1987;31:693 - 8.  Chatila W, Ani S, Guaglianone D, Jacob B. Cardiac ischemia during weaning from mechanical ventilation. Chest 1996;209:1577 - 83.  Chao DC, Scheinhorn DJ, Stearn-Hassenpflug M. Impact of renal dysfunction on weaning from prolonged mechanical ventilation. Crit Care 1997;1:101 - 4.  Dujic Z, Tocilj J, Ljutic D, Eterovic D. Effects of hemodialysis and anemia on pulmonary diffusing capacity, membrane diffusing capacity and capillary volume in uremic patients. Respiration 1991;58:277 - 81.  Noble JS, Mackirdy FN, Donaldson SI, Howie JC. Renal and respiratory failure in Scottish ICUs. Anaesthesia 2001;56:124 - 9.  Swinburne AJ, Fedullo AJ, Bixbi K, Lee DK, Wahl GW. Respiratory failure in the elderly: analysis of outcome after treatment with mechanical ventilation. Arch Intern Med 1993;153:1657 - 62.  Spicher JE, White DP. Outcome and function following prolonged mechanical ventilation. Arch Intern Med 1987;147:421 - 5.  Gracey DR, Gillespie D, Nobrega F, Naessens JM, Krishan I. Financial implications of prolonged ventilator care of Medicare patients under the prospective payment system: a multicenter study. Chest 1987;91:424 - 7.  Kurek CJ, Cohen IL, Lambrinos J, Minatoya K, Booth FV. Clinical and economic outcome of patients undergoing tracheostomy for prolonged mechanical ventilation in New York State during 1993: analysis of 6353 cases under diagnosis-related group 483. Crit Care Med 1997;25:983 - 8.  Veatch RM. Justice and the economics of terminal illness. Hastings Cent Rep 1988;34 - 40.
Commentary The b rights Q for patients with prolonged respiratory failure The delivery of efficient and effective health care is embodied in the paradigm of providing the right care for the right patient in the right place at the right time using the right resources. Achieving these brightsQ for patients with prolonged respiratory failure and dependency on mechanical ventilation can prove challenging. One approach to accomplish these goals is the use of weaning centers that specialize in the care of patients with this condition. In this issue, Su et al report their experience with a respiratory care center in Taiwan, how well did they provide patient brights,Q and if this was an approach that should be generally adopted. The first report is defining the right care. An evidencebased review in 2001 provides guidance regarding weaning strategies in general . Unfortunately, there are no good randomized trials that address the difficulty to wean patient. Observational studies, tabulated in that same review, identify a relatively consistent approach regarding ventilation strategies. But a holistic approach needs to consider all aspects of patient care. In particular, nutrition support 
161 and general conditioning  are important in this population because muscle weakness is likely a major contributor to the respiratory failure, as well as a long-term problem even after successful weaning . Selecting the right patients for this type of prolonged critical care requires consideration of medical facts together with patient and societal values and preferences. As evidenced by the 48% success rate in this study, which is similar to other reports, it is not possible to reliably determine in advance those patients who will actually wean from mechanical ventilation. In the current study, there appear to be some inconsistencies with patient selection. For example, 10 patients are reported to have left the center against medical advice, which is difficult to understand if they were ventilator dependent. Also, reasons for ventilator dependency are generally limited to respiratory dysfunction, which can be categorized as problems with respiratory drive, lung disease, or respiratory muscle function. Cardiovascular dysfunction may be a contributing factor, but this can be considered as a muscle problem (inability to provide adequate oxygen delivery to the respiratory muscles) [5,6]). Sepsis or gastrointestinal causes, as listed in this paper, do not typically reflect patients with isolated respiratory failure who are otherwise in a stable medical condition. Interestingly, adult respiratory distress syndrome does not appear in the list of causes. Better tools to predict those patients who will wean successfully are needed. Validating that the right patients are selected also requires the correct outcome measures. In this study, success was defined as 72 hours free of mechanical ventilation. Discharge from acute care might be a more appropriate measure. In addition, long-term follow-up that includes quality of life measures, functional status, and patient preferences are essential pieces of information to judge the true value of this type of care . Finally, let us consider the right place, timing, and resources together. In the current report, these three brightsQ appear to be dictated by the reimbursement methods. Patients were admitted for only 21 days because this was the maximum that is covered by the local health care system. However, experience reported by other centers indicates that this may be approximately the median time for success, and failure to wean is not declared until at least 3 months of continuing attempts . And although attention is focused on patients with ventilation dependency, our own analysis of patients with prolonged intensive care unit (ICU) stay found that respiratory failure accounts only for one third of this population. Equal numbers of patients with prolonged ICU stay have multiple organ failure, which clearly requires acute care, and other single system failures. The latter constitute patients with neurologic, renal, or cardiac disease who require more intense nursing or monitoring than can be provided on a general ward. If a goal of regional weaning centers is to improve access to ICU beds for other patients, then this goal can be enhanced by evaluating alternatives to the traditional ICU. These