| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Intensive Care (Dr. Vincent), Erasme Hospital, Free University of Brussels, Belgium; Department of Critical Care Medicine (Drs. Fink and Pinsky), University of Pittsburgh School of Medicine, Pittsburgh PA; Department of Pulmonary and Critical Care Medicine (Dr. Marini), Regions Hospital, University of Minnesota, St Paul MN; Department of Medicine (Dr. Sibbald), Sunnybrook & Women’s College Health Sciences Center, Toronto, Canada; Department of Intensive Care Medicine (Dr. Singer), University College, London, UK; Department of Surgical Intensive Care (Dr. Suter), Hôpital Cantonal Universitaire, Geneva, Switzerland; Department of Critical Care Medicine (Dr. Cook), McMaster University Medical Center, Hamilton, Canada; Emergency Services Department (Dr. Pepe), Parkland Health & Hospital System and the University of Texas Southwestern Medical Center, Dallas, TX; and Department of Intensive Care Medicine (Dr. Evans), Royal Brompton Hospital, London, UK.
Correspondence to: Jean-Louis Vincent, MD, PhD, Department of Intensive Care Medicine, Erasme University Hospital, Route de Lennik 808, 1070 Brussels, Belgium; e-mail: jlvincen{at}ulb.ac.be
Abstract
Over the last quarter of a century, intensive care medicine has developed into an established hospital specialty with its own unique identity and characteristics. Significant advances have occurred, mostly in a succession of small steps rather than any dramatic leap, with many being linked to advances in health care across other disciplines. In addition, many changes have resulted from the scientific identification of the detrimental effects of certain traditional practices once thought to be therapeutic. Here, in an attempt to learn from the past and offer guidance for future progress, we detail some of the key changes in various aspects of intensive care medicine including respiratory, cardiovascular, metabolic, and nutritional care, as well as sepsis, polytrauma, organization, and management.
Key Words: ARDS • cardiopulmonary resuscitation • cardiovascular care • critical care • intensive care • invasive monitoring • mechanical ventilation • multiorgan failure • polytrauma • sepsis
Imagine a patient with sepsis-related ARDS in your ICU.... and now imagine how that same patient would have been managed 25 years ago. Looking back on how patients were managed in the past demonstrates clearly just how much intensive care medicine has changed! However, do these changes in intensive care medicine represent progress, and has this progress kept pace with other disciplines?
Respiratory Interventions
Considerable progress has been made in the development of modern respirators. The bulky giants of 25 years ago have been replaced by slim-line, portable models. The elements that impact the work of breathing during assisted ventilation are now better defined, and the performance characteristics of ventilators—valve resistance, flow delivery, and monitoring—have also been improved. These developments have increased patient comfort and have allowed earlier detection of adverse patient/ventilator interactions.1 Recognition of the presence and consequences of auto-positive end-expiratory pressure (PEEP) has enabled reductions in its adverse effects on hemodynamics and in work of breathing. The use of pressure support has dramatically improved the management of respiratory failure, primarily by its easy adaptation to patient needs.
The development of noninvasive mechanical ventilation has resulted in successful avoidance of endotracheal intubation and decreased mortality in hypercapnic respiratory failure,2 although the benefit in other forms of respiratory failure is less well established. Continuous positive airway pressure and bi-level positive airway pressure can help to improve oxygenation and ventilation in patients with cardiogenic pulmonary edema, and when applied noninvasively to patients without contraindications may prevent the need for intubation.
Respiratory monitoring has also advanced considerably. The needle gauge has been replaced by simultaneous displays of airway pressure and flow, and useful information yielded by those variables is now effortlessly and continuously available. Devices such as pulse oximetry to monitor hemoglobin oxygen saturation noninvasively have improved our ability to detect and remedy any deterioration in a more timely fashion. Better pain management—including the use of regional anesthesia—together with more appropriate sedation regimens has enabled us to shorten the duration of mechanical ventilation.
In acute lung injury and ARDS, the development of CT scanning has allowed a better understanding of the anatomic alterations, delineating zones of complete collapse, recruitable regions, and well-aerated regions. The conditions under which recruiting maneuvers are an effective adjunct to ventilation to improve gas exchange and when they are most likely to cause hemodynamic stress are better understood. Nevertheless, the concept of whether and how to recruit alveoli—including how to adjust PEEP levels3—remains contentious.
Some studies have suggested that mortality from ARDS may have decreased,4 and a number of factors may be involved in these changes, including the realization that ventilating with lower tidal volumes reduces the inflammatory response and decreases mortality.5 The ideal ranges for tidal volume and plateau airway pressure have yet to be defined but are clearly lower than those used in the past. Experimental studies demonstrate overwhelmingly the potential for high transpulmonary pressures to injure even previously normal lungs, especially without sufficient PEEP application. Given the potential for injury related to aggressive ventilation, hypercapnia has been accepted as a legitimate and usually well-tolerated consequence of lung-protective ventilation.6
Interventions such as surfactant, inhaled nitric oxide, or systemic pharmacologic agents have not consistently reduced mortality from ARDS, and steroid therapy has also yielded equivocal results. Prone positioning may improve gas exchange but has not been shown to definitely improve outcome in patients with respiratory failure.7 Our inability to demonstrate a survival advantage of these particular interventions used to improve oxygenation is perhaps not surprising considering that mortality from refractory hypoxia is rare, and mortality in ARDS patients is most commonly due to sepsis and multiple organ failure.
Infections
Despite improvement in some hygiene measures, nosocomial infections remain an important and widespread problem.8 In particular, the emergence of multiresistant organisms has even led to the temporary closure of some units. Belatedly, more rapid screening tools and diagnostic techniques based on molecular probes are being developed, which should facilitate early identification of pathogens, resulting perhaps in more effective isolation of patients and earlier institution of targeted antibiotics. Development in imaging techniques has considerably improved identification of the source of infection, and the need for "empirical laparotomies" has almost disappeared.
Guidelines are being increasingly developed to limit the extended use of antibiotics9 and to narrow the spectrum of activity or stop antibiotics after 3 to 4 days depending on the identification (or not) of pathogens, sensitivity patterns, and clinical response. Selective decontamination of the digestive tract has been shown to decrease pneumonia and probably mortality10 but is not used widely, primarily because of costs and the fears of long-term bacterial resistance. Nursing patients with the head of the bed elevated at 45° can decrease the incidence of gastroesophageal reflux in patients receiving mechanical ventilation, and may reduce the development of nosocomial pneumonia.11 The use of antibiotic-coated catheters has decreased the incidence of catheter-related infections.
Sepsis
We now understand much more about the pathophysiology of sepsis and sepsis-related syndromes, although there are still many gaps in our knowledge. We recognize that sepsis represents an exaggerated systemic inflammatory response triggered by infection rather than a direct attack on the tissues by microorganisms that antibiotics may or may not kill. We now realize that the development of multiple organ failure involves multiple pathways including inflammatory, immune, microvascular, hormonal, bioenergetic, and metabolic systems. We see that there is a marked temporal change with early up-regulation of these systems, often followed in prolonged sepsis by a down-regulation that may be equally, if not more, injurious. The degree of perturbation of each of these systems has been shown to correlate with poor outcomes, yet we still do not fully understand the additive or mitigating interactions among them.
Better understanding of the complex network of mediators has led to the development of new strategies based on an immunomodulatory therapeutic approach, to be used in conjunction with the still essential hemodynamic resuscitation and eradication of infection. There have been numerous "false dawns," as seen with anti-endotoxin antibodies, nitric oxide synthesis inhibition, and cytokine antagonists.12 Only activated protein C has produced a statistically significant mortality reduction in patients with severe sepsis and septic shock,13 leading to the marketing and licensing of the first antisepsis product, drotrecogin alfa (activated). The mode of action of this agent has been only partly uncovered and involves effects other than just its anticoagulation properties.14 Many discussions have focused on the proper indications of the drug, largely because of the concern about bleeding, the high costs involved, and the lack of benefit in patients with less severe sepsis.
The recognition that high doses of steroids were harmful has been replaced by the realization that lower doses of glucocorticoids may be beneficial in patients with septic shock, especially in those with abnormal adrenal function.15 The importance of aggressive early and complete fluid and vasoactive drug resuscitation has been shown in severe sepsis/septic shock,16 although the exact reasons for this observed benefit from multifaceted resuscitation have not yet been determined.
Cardiovascular Management
The need for invasive hemodynamic monitoring has been increasingly challenged,17 and its use has decreased over the last decade.18 Other less invasive techniques have been developed, but whether this results in improved outcomes in all patients remains to be seen. Increasingly, continuous, noninvasive, and metabolic monitors are becoming available that may supplant more invasive monitoring devices in resuscitation algorithms. Using functional hemodynamic monitoring to define responsiveness in the optimization of blood flow has been shown to improve outcomes in cardiac surgery patients.19
There is still debate regarding superiority of the available vasopressor agents. We have learned that the maintenance of cardiac output or oxygen delivery at predetermined, supranormal levels does not universally result in better outcomes,20 although so-called "preoptimization" may be beneficial in high-risk surgical patients, especially when it involves the correction of underlying hypovolemia.21
Acute Coronary Syndromes
Thrombolytic therapy has been simplified for patients with acute myocardial infarction (AMI), and invasive percutaneous interventions are more commonly applied in acute coronary syndromes in many settings with suitable facilities.22 For some patients with AMI, prehospital thrombolysis can save lives in appropriate circumstances, particularly when invasive procedures would not be readily available. In critically ill patients, the diagnosis of non–ST-segment elevation myocardial infarction is difficult because ECG changes are common and often nonspecific, and increased troponin levels are also nonspecific, occurring among patients without flow-limiting coronary artery disease. Finally, eliciting classic ischemic symptoms from ICU patients is rarely possible due to decreased level of consciousness from their underlying condition or drug infusions. The utilization of life-saving medication for AMI such as ß-blockade, angiotensin-converting enzyme inhibitors, anticoagulants, and inhibitors of platelet adhesiveness has not been universal in AMI patients outside the ICU setting; however, in the ICU, their utilization is often precluded by dependence on ß-agonists, hypotension, and coagulopathy. Importantly, a too-liberal administration of antiarrhythmic agents in such patients may be more harmful than beneficial.23 New approaches to AMI in critically ill patients have not been developed, but the management of cardiogenic shock secondary to myocardial infarction has improved considerably.
Cardiopulmonary Resuscitation
There have been few systematic studies in human cardiopulmonary resuscitation (CPR); hence, almost all the changes in recommendations over the last 25 years result from low-level evidence. Innovative technology has led to the increased availability and ease of use of defibrillators, which, even when used by untrained bystanders, can improve outcomes from out-of-hospital cardiac arrest due to ventricular fibrillation, the most common reason for cardiac arrest in the community.24 We have learned that we used excessively frequent ventilatory rates, and that frequent or lengthy interruptions of cardiac compressions can be deleterious. Traditional therapies, for example, the routine use of bicarbonate, have been challenged. Likewise, amiodarone has replaced lidocaine in the management of life-threatening ventricular arrhythmias, although long-term outcome data are lacking, particularly in out-of-hospital ventricular fibrillation. Vasopressin has been introduced as a potent and reliable vasoconstrictor during CPR, especially in combination with epinephrine and particularly in cases presenting with asystole.25 Therapeutically induced mild hypothermia for 24 h after resuscitation has also improved neurologic outcomes for some patients.26
Polytrauma
Trauma care has improved considerably over the past 25 years, largely from combined improvements in assessment, triage, resuscitation, and emergency and intensive care. The development of trauma centers has continued to evolve in North America with data supporting their efficacy. However, a study27 in patients with penetrating torso injuries rapidly transferred to a level 1 trauma center has shown that IV fluid resuscitation should be limited in the absence of significant head trauma, as massive fluid administration early after hemorrhage may increase bleeding by raising intravascular pressures, disrupting clot formation, and diluting coagulation factors.
Cerebral Resuscitation
Monitoring the brain remains difficult, although intracranial pressure monitoring is now widely used. Tissue oxygen monitoring or microdialysis techniques have become available, but how they contribute to decreased complications is hard to define. We still lack good markers of cerebral damage, so that the evaluation of cerebral lesions remains largely based on the Glasgow coma score and neurologic status. Evaluation of cerebral blood flow and oxygen availability at the bedside remains difficult.
Induced hypothermia can protect the neurons in hypoxic encephalopathy,26 but the beneficial effects in traumatic states have not been established.28 Thrombolytic therapy with tissue plasminogen activator improves outcomes when given early (within 3 h of onset) to patients with ischemic stroke.29 Hyperglycemia may aggravate cerebral lesions, another reason to closely monitor blood sugar levels (see below).
Metabolic
Stricter control of blood sugar levels has been shown to decrease mortality in a mixed group of ICU patients, primarily surgical, many being admitted after cardiac surgery.30 Trials are underway to try to better target the most beneficial glucose level in medical-surgical ICU patients. The administration of steroids in septic shock has evolved over time, with the use of massive doses of methylprednisolone to limit the inflammatory response being replaced by the concept of relative adrenal insufficiency, leading to the administration of low doses of hydrocortisone in septic shock.15 Vasopressin administration in septic shock may also be based on relative deficiency in these circumstances. Trials are underway to define the potential benefit of administration of low doses of vasopressin in septic shock.
Nutrition
The concept 25 years ago was to administer relatively high caloric intake to cope with the catabolic state, and that we could administer this support parenterally just as easily as enterally. Since then, we have learned that gut-associated lymphatic tissue represents a major immune barrier and that enteral feeding promotes its activation, while total parenteral nutrition without enteral feeding induces gut-associated lymphatic tissue atrophy and increased risk of infection.31 Evidence has shown that overfeeding is not only useless but also potentially deleterious and that the enteral route is superior to the parenteral route; either the former route is actually better (to preserve gut structure and integrity), and/or the latter route is worse (for increasing infectious complications in particular). We now start enteral nutrition early, without waiting for bowel sounds to be present, even in conditions such as pancreatitis or in postoperative states. We often use prokinetic agents to minimize gastroesophageal aspiration in patients with impaired gut motility, although the evidence in support of this practice is controversial. Despite the theoretical advantages of immunonutrition, such feeds have not been demonstrated consistently to be safe or effective.32
Liver Failure
Increased use of liver transplantation has reduced mortality from acute liver disease.33 The application of extracorporeal systems such as the molecular adsorbent recirculating system has not been shown to decrease mortality.
Renal Failure
The development of continuous techniques of blood purification and fluid regulation enables us to avoid the "ups and downs" of intermittent dialysis on fluid, electrolytes, and osmotic shifts/balances. Initially developed as a simple arteriovenous circuit, the technique has evolved into a continuous venovenous system with relatively complex instruments. We have learned to use more biocompatible membranes in dialysis filters and not to aggressively promote ultrafiltration if it induces hemodynamically significant intravascular hypovolemia. We are not yet sure about the ideal intensity or dose of extracorporeal epuration techniques.34
Diuretic use has decreased in recent years, as it may worsen renal function,35 especially in the presence of hypovolemia. The so-called renal-dose dopamine has not been demonstrated to produce outcome benefit, and has been largely abandoned.36
Fluid Administration and Transfusion
The use of blood transfusions has decreased, especially after an important prospective, randomized Canadian study37 showed that a conservative approach to transfusion may result in lower mortality rates. This may be related to immune effects as the same investigators have shown improved outcomes with leukodepleted blood (against historical controls).38 Improved separation and preservation technology now facilitates selective and timely administration of appropriate blood components. Erythropoietin reduces the need for transfusion in long-term patients but has not been associated with changes in outcome.39
Albumin administration has been controversial for decades, as it is hard to demonstrate beneficial effects and the costs are high. It had been suggested that albumin administration results in higher mortality rates,40 but a large study41 in Australia and New Zealand has demonstrated that albumin and normal saline solution result in equivalent outcomes when used for fluid resuscitation.
Process of Care
Critical care medicine is driven by the organization of health-care delivery, and this is the area where we have seen the most obvious improvements. Observational studies42 have shown that a closed ICU with an ICU physician in charge of a multidisciplinary team doing regular rounds is associated with a shorter length of ICU stay and lower mortality. Daily rounds at the bedside can reduce the rate of complications.43 Prophylactic therapies against GI bleeding in high-risk patients and against deep venous thrombosis can improve outcomes, although the most effective methods of prophylaxis are still unclear.44
Markedly improved outcomes have been realized when interventions shown to reduce morbidity and mortality in randomized trials are used consistently. Protocolized care and continuous quality improvement have all resulted in markedly better outcomes, reduced costs, and minimized medical errors. Recent focus on the frequent incidence of preventable medical errors has lead to attempts to improve safety of hospital systems.45
Critical Care Without walls
Pioneered in Australia, now widely adopted in the United Kingdom and being introduced elsewhere in the world, is the development of "outreach" services or medical emergency teams to the general wards to both identify and treat at an earlier stage patients who are deteriorating and at risk of becoming critically ill, and to follow up patients discharged from the ICU to ensure that adequate care is being provided in this "step-down" environment. However, a recent study46 suggests this approach may not affect the incidence of cardiac arrest, unplanned ICU admissions, or unexpected death.
Health Sciences Research
One of the major advances in intensive care medicine has been the advent of the large, multicenter study not directly related to a new commercial product but organized by critical or intensive care societies or a government-funded body to address an important management question. Leading the way have been the Canadians (eg, sucralfate vs ranitidine, Transfusion Requirements in Critical Care); the Italians (eg, prone ventilation, supranormal oxygen delivery); the Australians (renal dose dopamine, albumin vs normal saline solution); the US ARDS-NET Group; European initiatives (eg, European Prevalance of Infection in Intensive Care, Sepsis Occurrence in Acutely Ill Patients, the CORTICUS project, Pulmonary Artery Catheters in Patient Management), and, most recently, the US National Institutes of Health Resuscitation Outcomes Consortium. Most of these trials have been completely altruistic, with intensivists collaborating together to answer clinically important questions. As a consequence, our knowledge base has expanded considerably, leading to identification of helpful as well as harmful practices and, eventually, to appropriate changes in our management of patients.
To complement and augment these studies, scoring systems (Acute Physiology and Chronic Health Evaluation, Simplified Acute Physiology Score, Sequential Organ Failure Assessment, and Therapeutic Intervention Scoring System) have been developed and have provided a level of sophistication and detail not available to any other speciality. In the United Kingdom, an ongoing audit has provided the necessary proof to convince the government into funding more critical care beds by demonstrating the impact of refused/delayed admissions, premature discharges, and nighttime discharges.
Our Failures
Many recent advances have occurred as a result of bench-to-bedside testing of hypotheses, and large collaborative clinical research networks. These efforts have had globally positive effects, yet we also acknowledge that many of our interventions have been deleterious (such as mechanical ventilation with high tidal volumes, RBC transfusions, and excessive sedation) [Table 1 ]. As we reflect on our achievements in intensive care medicine over the past 25 years, we should perhaps also briefly mention some of our failures. The path of progress has not been smooth and, not infrequently, hopes have been dashed as promising approaches have been shown in randomized controlled trials to have no effect or even to worsen outcomes. Examples of such studies include the excessive use of inotropic agents to increase oxygen delivery20; growth hormone administration in long-stay critically ill patients47; infusion of the tumor necrosis factor receptor Fc fusion protein48 and the nitric oxide synthase inhibitor, NG-monomethyl L-arginine,49 to patients with septic shock; use of intratracheal surfactant in patients with ARDS50; and infusion of diaspirin cross-linked hemoglobin in the treatment of severe traumatic hemorrhagic shock.51 Although not providing the hoped-for therapeutic panacea, these "failed" studies have nevertheless advanced the field of intensive care medicine.
|
Progress in intensive care medicine has been considerable over the past quarter century, accomplished usually by a succession of small steps rather than by any one dramatic change, and linked to advances in health care across other disciplines. Importantly, we have learned that not everything that improves physiologic measures improves survival; thus, physiologic surrogates need to be tested in large-scale trials whenever feasible. Indeed, prospective randomized clinical trials are ongoing to clarify many of the areas of confusion and debate highlighted in this document, and further studies will be needed to evaluate supportive care modalities as well as new pharmacologic interventions.
Intensive care medicine has become a recognizable identity, with a unique knowledge base and sophisticated skill set, which requires considerable health-care resources to be delivered optimally. Until recently, words such as "performance measurement," "accountability," "management," and "leadership" were not part of our traditional vocabulary. However, today we are beginning to measure what we do, to set goals, and to make plans to achieve them within our ICUs and in the context of the hospitals and broader health-care systems in which we work. The need for ICU beds has increased substantially over time and will continue to do so as a result of our aging population. Reflecting on our past and building on our strengths, the next 25 years promise to be exciting and challenging for all involved in intensive care medicine.
Footnotes
Abbreviations: AMI = acute myocardial infarction; CPR = cardiopulmonary resuscitation; PEEP = positive end-expiratory pressure
Learning objectives: 1. Distinguish the significant advances in intensive care medicine that have occurred, largely as a result of a succession of small steps in many aspects of intensive care medicine. 2. Evaluate the many advances that have occurred as a result of scientific identification of the detrimental effects of certain traditional practices once thought to be therapeutic.
This article has been prepared on the occasion of the 25th International Symposium on Intensive Care and Emergency Medicine.
The following authors have indicated to the American College of Chest Physicians that no significant relationships exist with any company/organization whose products or services may be discussed in this article submission: Michael R. Pinsky, MD; Jean-Louis Vincent, MD; Timothy W. Evans, MD; John J. Marini, MD; Peter Suter, MD; Deborah Cook, MD; and Paul Pepe, MD. The following authors have disclosed financial relationships with a commercial party (grant information and company names appear as provided): Mervyn Singer, MD: UK Department of Health: Grant Deltex: Unrestricted Educational Grant. Mitchell Fink, MD: Defense Advanced Research Projects Agency (DARPA): Grant National Institutes of Health (NIH): Grant Critical Therapeutics, Inc.: Consultant; NovoNordisk: Consultant; Inotek Pharmaceuticals, Inc.: Fiduciary Position; Midway Pharmaceuticals, Inc.: Fiduciary Position. William Sibbald, MD: Canadian Institute of Health Research: Grant; Heart and Stroke Foundation of Canada: Grant.
Received for publication July 22, 2005. Accepted for publication November 30, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
