© 2007 American Thoracic Society doi: 10.1164/rccm.200612-1762ED
Critical Illness NeuromyopathyFrom Risk Factors to PreventionCentre Hospitalier de Poissy-Saint-Germain, Poissy, France
Royal North Shore Hospital and Northern Clinical School, University of Sydney, Sydney, Australia Critical illness neuromyopathy (CINM) can significantly delay weaning from mechanical ventilation and impair physical recovery (1, 2). CINM is often unrecognized and, depending on the population studied, may affect between 25 and 75% of longer-stay intensive care unit (ICU) patients (3, 4). In this issue of the Journal (pp. 480–489), Hermans and colleagues report that intensive insulin therapy prevented CINM in a subset of patients enrolled in their medical ICU trial (5). That intensive insulin therapy may prevent CINM is exciting, but the report highlights the difficulty of diagnosing CINM in a research context and raises important issues of clinical trial methodology. Hermans and colleagues diagnosed CINM solely on the presence of spontaneous electrical activity in muscle on electroneuromyography. They did not identify altered muscle membrane excitability, which is increasingly recognized as a cause of muscle weakness in critically ill patients (6). Had they done so, they may have reported a different result. In addition, they only report outcomes in patients treated in the ICU for 7 days or more and do not present an intention-to-treat analysis. The incidence of CINM in patients who received the study and control treatments for up to 7 days goes unreported and this significantly reduces the certainty of the findings (7). Furthermore, because the subgroup studied could not be identified at the start of treatment, clinicians do not know to which of their patients the findings apply (8). Although selecting appropriate subgroups and reporting intention-to-treat analyses are essential components of clinical trial design, studying CINM presents particular difficulties. CINM can be detected by clinical examination or electroneuromyography but usually not until patients have been treated with mechanical ventilation for several days. Clinical examination is focused on detecting muscle weakness, which is a uniform manifestation of CINM, regardless of whether the axon, muscle contractile proteins, or muscle membrane excitability are involved alone or in combination (4, 9). As the patient must be cooperative, daily assessment from the beginning of mechanical ventilation is rarely possible. Electroneuromyography is the alternative diagnostic approach. Electroneuromyography allows earlier diagnosis but its use requires considerable resources and can cause discomfort. It is neither practical nor ethical to conduct repeat testing on a daily basis for a clinical trial. To date, only patients who remain in an ICU for 5 to 7 days have been assessed for CINM, and because many patients will have been discharged or will have died by this time, an intention-to-treat analysis of a clinical trial may appear impractical. However, the methodologic difficulties can be overcome by asking the right questions in the right population. We should ask whether a treatment reduces the incidence of CINM and does that result in better outcomes for patients. The study population should be patients likely to receive at least 5 days of mechanical ventilation and in whom the clinical diagnosis of CINM is feasible. This will likely mean excluding patients with primary neurologic diseases or brain injury. To determine if CINM is prevented, the outcome measure could be the number of patients discharged alive from ICU who were never diagnosed with CINM. To determine if patients' outcomes are improved, time to recovery from CINM and long-term functional recovery of survivors should be reported. In our opinion, the diagnosis of CINM should be made by clinical examination conducted by a blinded observer using a recognized scoring system. For example, the Medical Research Council scale is a simple and robust score that quantifies muscle weakness (10). It has satisfactory interobserver reproducibility in ventilated patients with Guillain-Barré syndrome (10), and in patients with acute stroke (11). Although not yet validated in the general ICU population, it appears to be an ideal tool for the diagnosis of CINM. The timing of clinical testing may be difficult because patients must be sufficiently recovered to cooperate; the earliest this is likely is at the start of weaning from mechanical ventilation. Serial testing should be conducted from the start of weaning until the time of discharge from the ICU. Having recognized the methodologic difficulties inherent in such research, what are the implications of Hermans and colleagues' report for clinicians? Enthusiasm for intensive insulin therapy arose from the reduction in mortality in Van den Berghe's first trial (12). This enthusiasm has been tempered somewhat by the findings of the second trial (13), and because the Efficacy of Volume Substitution and Insulin Therapy in Severe Sepsis (VISEP) and Glucontrol studies have both stopped early due to lack of efficacy and concerns about the safety of intensive insulin therapy (14–16). A fourth, multicenter trial in Australasia and North America (the NICE-SUGAR [Normoglycaemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation] study) plans to recruit 6,100 patients and report its findings in 2008 (17). If NICE-SUGAR shows a clear mortality benefit, then intensive insulin therapy will likely become a standard of care. If not, then effects on secondary outcomes, such as the incidence of CINM and duration of mechanical ventilation, will assume considerable importance. Data from the two European trials may confirm or refute Hermans and colleagues' findings as the incidence of CINM was assessed in the VISEP trial using electrophysiology and in Glucontrol using clinical examination (F. Brunkhorst and J.-C. Preiser, personal communications). If intensive insulin therapy does prevent CINM, then the benefit to patients should be a reduction in the duration of mechanical ventilation and more rapid recovery from critical illness. From the patient's perspective, these outcomes are valuable regardless of why they occur. Van den Berghe and colleagues have reported that intensive insulin therapy significantly reduced the time to weaning in their medical ICU trial (13), but not the median duration of ventilation in their surgical ICU trial (12). The VISEP, Glucontrol, and NICE-SUGAR trials will report the impact of intensive insulin therapy on duration of mechanical ventilation in a large population of patients treated in different ICUs. The collective findings should be broadly applicable in ICUs around the world. Hermans and colleagues' report that a metabolic therapy may prevent CINM is exciting. From the clinicians' viewpoint, the report presents those who have adopted intensive insulin therapy with no reason to change their practice. Others may be concerned about the methodologic issues and that intensive insulin therapy has not yet been proven beneficial in multicenter trials. Those who remain skeptical may prefer to wait for the full reports of the VISEP and Glucontrol trials and possibly for the results of the NICE-SUGAR study. It is hoped that other strategies to prevent CINM will be developed and tested in randomized controlled trials. The difficulties of conducting such research should not be underestimated and Hermans and colleagues are to be congratulated on their work. The results of future trials will be regarded with the greatest certainty if the outcomes most relevant to patients are reported in the intention-to-treat population. FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
