Abstract
The chronically critically ill (CCI) comprise a rapidly growing population of patients who have survived acute critical illness, only to be left with ongoing organ dysfunctions requiring high levels of specialized care for months or years. In many ways, CCI is an “iatrogenic” process, reflecting the ability of modern life support technologies to keep patients alive for prolonged periods of time despite ongoing life threatening illness. Venues of care for the CCI patient include acute care hospitals (both ICU and step-down facilities), specialized long term acute care hospitals, and, less commonly, skilled nursing facilities, or even the home. Importantly, CCI patients transition among these venues frequently, reflecting the nature of CCI to be punctuated with episodes of acute critical illness. Management of the CCI population requires a special combination of intensive care and rehabilitative skills.
- chronic critical illness
- mechanical ventilation
- critical illness
- acute care
- intensive care
- ICU
- step-down
- long-term acute care
- skilled nursing facilities.
Introduction
The chronically critically ill (CCI) comprise a rapidly growing population of patients who have survived acute critical illness, only to be left with ongoing organ dysfunctions requiring high levels of specialized care for months or years. This is a relatively new group of patients who are a direct by-product of the fact that modern technology can keep patients alive over long periods of time despite serious multi-organ dysfunction. The chronically critically ill and patients requiring prolonged mechanical ventilation (PMV) often overlap. However it is possible to have CCI without requiring ventilator support, and, conversely, to require PMV without being chronically critically ill (eg, long-term neuromuscular dysfunction, such as amyotrophic lateral sclerosis or muscular dystrophy). Having said that, the vast majorities of CCI patients also do require PMV, and the costs and outcomes of these 2 groups of patients often are lumped together in databases and outcome studies. Because CCI patients require high resource consumption, payers have become particularly concerned about cost control and assuring cost-effective care in this population. As a consequence, decisions on the care of these patients are often driven by payers and regulators, and these decisions may not always be in the best interest of the patient.
It is important to remember that the CCI population requires a different “mindset” in terms of care delivery, a concept often missed by both clinicians and administrators. Unlike acute critical illness patients, the clinical status of these patients is usually characterized by slow fluctuations in function and care needs; progress (or deterioration) occurs slowly, often over weeks or months. Importantly, however, this slowly changing baseline can frequently be interrupted by acute events such as sepsis or heart failure, necessitating an escalation of care. Because of this erratic clinical trajectory, caregivers with unique skills, encompassing both rehabilitation and critical care interest/expertise are required. Importantly, because outcomes can be poor, a culture of care that has a heavy palliative care influence is critical. Finally, the organ failures in these patients, because they are chronic, often require a somewhat different approach to management than similar organ failures in the acute care environment. All of these concepts support the argument for specialized centers to care for CCI patients.
Because of the growing numbers, growing cost, and growing awareness of the special needs of these patients (especially respiratory support), the American Respiratory Care Foundation and the journal, Respiratory Care, convened the 49th Journal Conference to address a number of aspects in the management of the CCI patient. As with previous Journal Conferences, 14 invited experts were brought together to formally address various assigned topics. Lively discussions among the participants followed each of these presentations. Both the content of the presentations and the discussions that followed are being published in this issue of Respiratory Care. Below is a summary of these presentations/discussions and my take on the messages being delivered.
Definitions, Prevalence, and Outcomes
Although the concept of CCI is most likely understood by many who care for critically ill patients, applying an exact definition has proven problematic. In broad terms, CCI refers to a population of patients who have survived acute critical illness but who have persistent organ dysfunction after the initial insult, requiring weeks, months, or even years of specialized care.1
As noted above, there is considerable overlap of CCI and PMV patients, and, indeed, because the CCI population is difficult to define from readily available clinical/administrative databases, the epidemiology, demographics, healthcare utilization, and outcomes of the CCI population are often estimated from data extracted from PMV databases.1 Unfortunately, however, there is also no universally accepted definition of PMV and, thus, available databases do not always provide consistent information. A common threshold for defining PMV is 21 days of mechanical ventilation, although some argue for 14 or 28 days.2 The placement of a tracheostomy for long-term care has also been used as an indicator of the need for PMV and, indeed, a separate diagnosis-related group (DRG) for a tracheostomy after 96 hours of ventilation (not due to airway surgical procedures) exists.3 Other definitions use administrative composites (eg, 4 ventilator days plus a tracheostomy plus 21 ICU days) or the need for mechanical ventilation in a subacute setting as a definition.3,4 Regardless of the definition used, it is apparent that the PMV/CCI population is rapidly growing. Indeed, there are estimates as high as 250,000 PMV/CCI patients currently in the United States.5 More importantly, it is estimated that there will be 50–100% growth in this population with every upcoming decade.5,6
The ultimate outcomes in this population are depressing (although there can be bright spots).1 In general, yearly mortality is 40–50%, a figure that has been a remarkably consistent observation over many studies over the last few decades.4,7,8 A recently described score (ProVent score) that is calculated at 21 days of mechanical ventilation was remarkably accurate at predicting mortality.9 This score is based upon 4 parameters: age > 50, platelets < 150,000/μL,3 the need for vasopressors, and the need for dialysis. If none of these factors was present, survival was over 80%. In contrast if all 4 were present, there was virtually 100% mortality at one year.
Unfortunately, even if a patient does survive a year, functional status is often quite poor.1 In most studies, the vast majority of survivors still required substantial caregiver support, with only a small fraction actually being discharged home.10 Interestingly, although physical function is often quite poor, some patients do seem to adapt with decent social functioning and mental health. Indeed, physical disability may be acceptable to many if cognition is reasonably intact.
Mechanism of Chronic Critical Illness
In essence, CCI is a persistent ongoing inflammatory state following an initial inflammatory insult.11,12 CCI is characterized by persistent elevations of many pro-inflammatory cytokines and a failure of anti-inflammatory processes to modulate and repair.13 The reasons for this imbalance are multifactorial. The initial insult may have been overwhelming, with the patient's defenses unable to adequately cope, especially if there is underlying poor baseline function and/or chronic disease. Death is averted only because of medical technology.
Organ dysfunction from the original inflammatory process may have left the patient with heart failure, liver failure, adrenal failure, neuromyopathies, impaired cognition, hormonal dysregulation, and renal failure.11 Underlying all of these is a fundamental cellular dysfunction characterized by impaired glucose metabolism and abnormal cellular energetics.14 We should also not underestimate the effects of inappropriate clinician responses on the development/perpetuation of CCI. These include inadequate antibiotics, inappropriate ventilator settings producing ventilator-induced lung injury, fluid overload, electrolyte mismanagement, malnutrition, excessive sedation, and exposure of the patient to nosocomial infection risks.15
It is important to realize that the body's response to stress in acute critical illness and CCI are fundamentally different.12,16 Acute critical illness is characterized by a Darwinian stress response of “fight or flight” in the setting of an acute insult such as sepsis or trauma. There is an outpouring of hypothalamic-pituitary-adrenal stress hormones, and catabolism occurs to provide immediate glucose and amino acids. There is a down-regulation of non-life-support functions, such as gut perfusion and reproductive function, while inflammatory cytokines convert liver function to the production of acute phase reactants. A Darwinian response does not exist in CCI. Evolution has not prepared humans for a protracted inflammatory state that requires a variety of artificial life support technologies to persist. Indeed, prior to the modern era, a patient either recovered from acute critical illness or died. Without modern life support technologies, CCI would not exist.
There are no “standard” serum markers of CCI. Although C-reactive protein and pro-calcitonin measurements are clinically available indicators of an inflammatory state, their role in defining CCI or driving decision making in CCI is largely unknown.
Venues of Care for the Patient With CCI
Most CCI patients are initially treated in acute care hospitals for the inciting illness.2,17,18 However, when the acute illness fails to completely resolve, the inflammatory state transitions to the chronic phase. Thereafter, many of these patients may remain in traditional short-term acute care (STAC) facilities, either in ICUs or less intense (and less costly) step-down units.2,17,18 Importantly, the real advantage to a step-down unit may be that it opens up a full ICU bed for use by another patient. This fact is rarely included in cost estimates. STAC hospitals are paid under the conventional Medicare DRG system. However, even using DRGs focused on patients requiring PMV, these CCI-PMV patients are often outliers whose care is almost never reimbursed enough to cover costs.18
Another common site for CCI care is the long-term acute care hospital (LTAC).2,17–20 LTACs have experienced tremendous growth over the last 2 decades, but this has slowed recently, due to regulatory and payer restrictions. Proponents of the long-term acute care model argue that the culture of care is more rehabilitation oriented, with an emphasis on physical therapy, occupational therapy, and respiratory therapy that is better suited for the prolonged recovery of patients with CCI-PMV. LTACs are also paid through a DRG system, but the weights are different, to reflect the long-term needs of these patients. Importantly, an LTAC must have a 25 day average stay to qualify for these DRGs. Again, outliers often occur, with outlier payments rarely covering the full cost. Unlike STAC facilities, short-term outliers in an LTAC also come with a financial penalty to the hospital.
There have been conflicting data as to whether LTACs are more cost-effective than STACs in caring for these patients.18–20 Regardless of venue, however, it is clear that costs for the care of these patients are substantial, and, because of the often poor outcomes, cost effectiveness analyses often show unfavorable results.
One last point about the venue of care is that patients commonly are switched between different venues of care. This was dramatically illustrated in one study that looked at 5 venues of care (home, LTACs, STACs, skilled nursing facilities, and in-patient rehabilitation facilities) and found that on average a patient experienced almost 5 transitions among these 5 sites in the first year.21 This speaks volumes to the fact that, although these patients are thought of as chronically critically ill, their course is frequently punctuated by episodes of deterioration and acute critical illness.
Pediatric Chronic Critical Illness
CCI has been less well characterized or studied in pediatric patients, but it may be a growing population for the same reasons it is growing in the adult population.22,23 In contrast to adults, however, many pediatric PMV patients have congenital abnormalities involving the cardiorespiratory and neuromuscular system that result in a long-term need for mechanical ventilation.23,24
In managing CCI-PMV pediatric patients, it is once again important to remember that pediatric patients are not simply little adults.23 They have different anatomy, physiology, psychosocial, and economic situations than adults. The pediatric respiratory system has obviously smaller airways and weaker cough, and these pose important challenges to caregivers. There is also less cardiac reserve and less nutritional reserve. Importantly, comorbidities are often congenital and can have substantial impact on overall function. With the pediatric patient, parents are often called in to be primary caregivers in the home. This can lead to caregiver burnout and psychological and financial disasters.
Managing the Patient with Chronic Critical Illness
Mechanical Ventilatory Support
Ventilator dependence is multifactorial in this population.2,25 The injured lung has abnormal mechanics, abnormal dead space, and impaired gas exchange that can overload the neuromuscular capabilities of the patient with CCI. Malnutrition, drugs, and critical care neuromyopathies all contribute to impaired respiratory drive and poor respiratory muscle function that add to the need for mechanical ventilator support.
Managing the ventilator is often more art than evidence-based.2,25 Clearly settings should be “lung-protective” (ie, tidal volumes in the 6–8 mL/kg [ideal body weight] range and plateau pressures < 30 cm H2O),26 but beyond this, a strong evidence base to guide care is lacking. There is no clear consensus on the “best” way to reduce/remove ventilator support in this population. Indeed, all efforts are going to be futile until many of the underlying causes of ventilator dependence are reversed.27
Extrapolating from the clear benefit of daily spontaneous breathing trials in acute critical illness can be problematic in PMV patients, who often have had many failed spontaneous breathing trials.2,27 Because of this, many authorities argue that spontaneous breathing trials should not be attempted until the patient has had the level of respiratory support reduced to some modest level (eg, pressure support in the 10–15 cm H2O range).2,25 Along with this should be a reduction in PEEP and FIO2 to < 8 cm H2O and 0.5, respectively. When all these criteria are met, then short attempts at unassisted spontaneous breathing can be attempted. These periods are then lengthened as tolerated. In general, prolonging unassisted spontaneous breathing through the night is done more cautiously.
Most patients in this situation have tracheostomies, and tracheostomy care is obviously important.28 Speaking valves can be very helpful to enhancing the patient's ability to communicate and improving general attitude. Decannulation can be considered when the capability to do unassisted spontaneous breathing permanently is established. However, this should be done with caution in patients with secretion clearance issues or obstructive apneas. As will be discussed subsequently, mobility and sedation reduction are important in optimizing ventilatory support reduction.29 To this end it is critically important to have a well trained and enthusiastic respiratory therapy and physical therapy team, along with appropriate portable monitors and ventilators to assure that patient activity is maximized.
It is essential to note that some patients will never be weaned from the ventilator. In some studies this is as many as 25% of survivors at the end of the first year.2 This raises the issue of when attempts at weaning should be abandoned. While this is certainly an individual patient decision, the literature indicates that 90% or more of those patients who eventually are weaned have the weaning complete by 90 days of mechanical ventilatory support.2
The role of noninvasive ventilation (NIV) in the CCI population is not clear. There is compelling evidence that NIV with a mask may avoid the need for intubation in acute respiratory failure, especially patients with exacerbation of COPD.30,31 NIV also delays the need for tracheostomy and invasive mechanical ventilation in a number of chronic progressive neuromuscular diseases, including amyotrophic lateral sclerosis and muscular dystrophy.30,31 This population also appears to benefit from cough assist devices using positive pressure.32 There are also a number of small observational and randomized trials of using nocturnal NIV, particularly in stable COPD patients with hypercarbia.30 Nocturnal NIV in restrictive diseases appears to have a more consistently positive database, and it may help in airway clearance in cystic fibrosis.30 Another application is in the obesity hypoventilation patient. Clearly there is a role for both CPAP and NIV in patients with obstructive sleep apnea.30
Unfortunately, supporting evidence for NIV effectiveness in patients with CCI is lacking.31 The use of NIV in this population therefore must be based upon extrapolation from clinical data in other patients, as noted above. In general NIV is most likely going to be an option primarily for those CCI patients who require only nocturnal support.
In the unusual CCI/PMV patient whose ventilatory support can be managed in the home, there are a number of challenges.33,34 First, the device to be used is driven not only by medical necessity but also by regulatory/reimbursement considerations.34 In general, a patient requiring only nocturnal support with NIV often only qualifies for a “respiratory assist device” or RAD. A RAD is characterized by the use of a single-circuit setup without an exhalation valve and utilizing a controlled leak gas delivery system. In contrast, tracheostomy patients requiring ventilatory support for most hours of the days require a true ventilator, a device capable of completely controlling minute ventilation through a closed system with an exhalation valve.
Traditionally, RADs are usually not considered “life support,” while ventilators usually are (ie, loss of support would result in serious harm or death). In recent years, however, these operational definitions have blurred: RAD devices are now approved for tracheostomy patients, ICU ventilators have been adapted for mask use, and tracheostomy patients can often be supported with only intermittent (ie, non-life-support) technology. As a consequence, the clinical application and the technical capabilities of a device are often not aligned. Not surprisingly, reimbursement for a ventilator is considerably higher than for a RAD. More importantly, clinical support is also included in ventilator reimbursement, but not for a RAD, a situation that places the RAD user in a suboptimal situation. Future regulatory/reimbursement schemes need to address this evolving technology. Patients need technology and clinical support based on their clinical needs, not on outdated/confusing technical considerations.
A second important challenge in managing home ventilatory support is the substantial drain on family resources of time, energy, and money.24,34,35 Continuous (ie, 24 h/d) attendants are necessary for home invasive mechanical ventilatory support, and to manage this, at least 2 well trained family members (or friends) supplemented by professional help are required. Under these conditions, caregiver burnout is common. In addition to caregivers, a backup ventilator is usually required, and devices need battery power for use in transport and in case of power failures.
Metabolic/Nutritional Support
Many of the features of the CCI patient involve persistent catabolism, malnutrition, and neuro-endocrine imbalances.12,14,36 Hyperglycemia is a common problem seen in the CCI, as is bone resorption, vitamin D insufficiency, and anasarca.12,36 Managing these multiple abnormalities requires careful nutritional support to avoid under- and over-feeding, and avoidance of the refeeding syndrome. Tight glucose control seems important, but this requires careful nutritional management and monitoring.37 Bone dysfunction requires multiple strategies, including calcium replacement, vitamin D replacement, and bisphosphonates. Nutritional supplements that should not be forgotten include glutamine, zinc, vitamin A, vitamin C, and arginine. There is much controversy about the role of manipulation of the hypothalamic-pituitary-adrenal axis with such things as growth hormones, androgens, thyroid, and hypothalamic releasing factors.38 Evidence supporting this is controversial and clearly needs more study. It is important to always monitor for hypoadrenalism, especially in those with tapering glucocorticoid treatments.
Infection Issues
The chronically critically ill are clearly at increased risk from infection, for multiple reasons.39,40 These include the presence of invasive devices, the phenomenon of immune “exhaustion” or “paralysis,” malnutrition, hyperglycemia, and drugs (especially inappropriate antibiotics). These infection issues are magnified by the presence of “bad bugs” in the various care venues, which include vancomycin-resistant enterococcus, methicillin-resistant Staphylococcus aureus, and resistant Gram-negative organisms.
Minimizing infection risks involves adhering to Centers for Disease Control care bundles, which include chlorhexidine baths, environmental cleaning, surveillance programs, isolation of those who are infected, and education on infection control (especially hand hygiene).40,41 There are also some new cleaning strategies, including ultraviolet light and hydrogen peroxide vapor, that are under study.
Device-associated infections can also be limited by using appropriate care bundles.40,41 For ventilator-associated pneumonia, these include head of bed elevation, minimized sedation, oral care, subglottic suctioning, peptic ulcer disease prophylaxis, and deep venous thrombophlebitis prophylaxis. Along with this, patients should receive the flu and pneumococcal vaccine. The role of silver-impregnated tubes is quite controversial, and because of the expense it is unclear if these are cost-effective. Blood stream infection care bundles include hand hygiene, complete barriers for central line insertion, chlorhexidine usage, proper site selection, and daily assessment for continued need of the catheter. Finally, catheter-associated urinary tract infection care bundles also focus on similar strategies and programs limiting urinary catheter use. As with the endotracheal tube, there is controversy regarding the role of silver-impregnated catheters. Importantly for all these care bundles, there are national benchmarks for the number of devices that should be used in patients, and national benchmarks on what the infection rate should be.39 These should be part of every care venue's quality assurance plan.
Neuromyopathies
Neuromyopathies commonly occur in the chronically critically ill, and the median prevalence across a number of studies is almost 60%.42,43 The neuromyopathies of CCI involve several processes. First, there is critical illness myopathy, which is a direct muscle injury in the setting of a system inflammatory response, impaired oxygen delivery/uptake, protein catabolism, and impaired repair mechanisms. This is often complicated by disuse atrophy, comorbidities, and age. There is also a critical illness polyneuropathy that is a diffuse axonal injury in the setting of the system inflammatory response and impaired oxygen delivery/uptake. Importantly, myopathies and neuropathies may coexist. However, outcomes are better in pure myopathies.
Approaches to prevention and management of critical illness neuropathies involve several things.43 First, there is good evidence supporting tight glucose control in limiting the development of neuropathies.37 It is also important to reduce exposure to possible toxins such as neuromuscular blockers and steroids (possibly through hyperglycemic effects). Electrolytes need to be optimized, especially potassium, phosphate, calcium, and magnesium. And, finally, there is a growing evidence base supporting the role of early mobilization, usually in conjunction with sedation minimization.29 Indeed, the costs of a mobility team seem easily balanced by reductions in hospital and ICU stay.
Neuro-cognitive Impairment
Delirium (both agitated and quiet) is the most common neuro-cognitive abnormality in the CCI population, and has been documented to occur in as many as 70% of this population.44,45 The etiology is multifactorial and includes systemic inflammation, hypotension, electrolyte shifts, sleep deprivation, hypoxemia, and drugs (especially benzodiazepine).
Managing these dysfunctions is complex.44,45 Clearly, sedation should be minimized and every attempt should be made to restore the sleep cycle. Careful monitoring of metabolic functions is also important. Caregivers should never underestimate the importance of human contact in helping patients stay oriented and in touch with their surroundings. Finally, the mobility programs noted for the neuromyopathies in helping patients also help avoid delirium.29
Palliative Care
Palliative care focuses on symptom relief, extensive communication to align treatment with patient preferences, transitional planning, and both patient and family support.46–48 This can be provided by focused palliative care teams, but the principles should be included in the care plan of every caregiver. Pain and depression are common symptoms that need treatment. It is also important to remember that friends/family can also experience tremendous physical, financial, and emotional burdens (especially if they too are caregivers) and must always be included in a plan of palliative care.
Palliative care should not be a sequel to failed care.48 These principles should be integrated into the overall care plan right from the start. Indeed, it is difficult to imagine a group of patients with greater needs for high quality palliative care than the CCI population.
Summary
The chronically critically ill comprise a rapidly growing population of patients who have survived acute critical illness only to be left with ongoing organ dysfunctions requiring high levels of specialized care for months or years. In many ways, CCI is an “iatrogenic” process reflecting the ability of modern life support technologies to keep patients alive for prolonged periods of time despite ongoing life threatening illness. Venues of care for the CCI patient include acute care hospitals (both ICU and step-down facilities), specialized long-term acute care hospitals, and, less commonly, skilled nursing facilities or even the home. Importantly, CCI patients transition among these venues frequently, reflecting the nature of CCI to be punctuated with episodes of acute critical illness.
It is important to remember that the CCI population requires a different “mindset” in terms of care delivery, a concept often missed by both clinicians and administrators. Unlike acute critical illness patients, the clinical status of these patients is usually characterized by slow fluctuations in function and care needs with progress (or deterioration) occurring slowly, often over weeks or months. Importantly, however, this slowly changing baseline can frequently be interrupted by acute events such as sepsis or heart failure necessitating an escalation of care. Because of this erratic clinical trajectory, caregivers with unique skills encompassing both rehabilitation and critical care interest and expertise are required. Managing the various organ failures also require strategies that, while similar to management in the acute ICU, often have special considerations. Importantly, because outcomes can be poor, a culture of care that has a heavy palliated care influence is important.
Footnotes
- Correspondence: Neil R MacIntyre MD FAARC, Division of Pulmonary and Critical Care Medicine, Duke University Hospital, Box 3911, Durham NC 27710. E-mail: neil.macintyre{at}duke.edu.
This paper was derived from presentations at the 49th Respiratory Care Journal Conference, “The Chronically Critically Ill Patient,” held September 9–10, 2011, in St Petersburg, Florida.
Dr MacIntyre has disclosed a relationship with CareFusion and Trudell Medical.
- Copyright © 2012 by Daedalus Enterprises Inc.
