Addition of anabolic steroids to pulmonary rehabilitation lation. Reduction in fat-free mass is a better predictor of resulted in increases in fat-free mass but no ...
Editorials
recruitment maneuvers, the potential for shear stresses and therefore lung injury should be reduced further by protocols that incorporate systematic effective recruitment maneuvers. The criteria for weaning from HFOV to CV also need to be re-evaluated. In the initial trial of Fort and coworkers, levels of mean airway pressure were maintained until the FiO2 decreased to less than 0.4 (2). Only then was the mean airway pressure weaned and the patient was transferred to CV when the mean airway pressure was 20 to 22 cm H2O. In this trial, mean airway pressure was increased until the FiO2 decreased to 0.6. Then both FiO2 and mean airway pressure were weaned alternately between FiO2’s of 0.6 to 0.5. CV was reinstituted at an FiO2 of 0.5 and mean airway pressure of less than or equal to 24 cm H2O. Patients therefore exited this trial of HFOV with gas-exchange criteria that serve as the entry criteria of many other ventilator trials. No justification is given for this design. Neonatal trials demonstrate clearly that a lung-protective effect with HFOV requires you to start early before the lung is damaged and continue until it is no longer vulnerable to ventilator-induced lung injury (9). This study initiated HFOV earlier than prior trials but also discontinued it sooner. This is the most puzzling design feature. A patient with a saturation of 88% on 50% oxygen still has a 40 to 50% venous admixture fraction, presumably from ongoing atelectasis or consolidation. The mean airway pressure of 24 cm H2O also reflects a lung that is still requiring a rather high mean pressure to maintain this aeration. Patients being ventilated postoperatively following cardiopulmonary bypass, long surgeries, or massive transfusion, without primary lung pathology, generally have levels of mean airway pressure of 8 to 12 cm H2O on an FiO2 of 0.4 to 0.5. Such patients can be ventilated indefinitely without detectable lung injury. There must be some definable level of mean airway pressure and venous admixture at which the lung becomes vulnerable to ventilator-induced lung injury. It is more likely to be a range than a threshold value, but I doubt it is 24 cm H2O. In neonatal ventilation, entire comparative trials of HFOV versus CV have been executed below the mean airway pressure at which HFOV was discontinued in this trial (10). This trial definitely takes us one step further in the incremental reintroduction of lung-protective HFOV. It represents a tremendous amount of careful work within the limitations of both our knowledge of ventilator-induced lung injury and the technology available at the time of study design. As the authors stated, the next step in this saga will be to match a better HFOV protocol (i.e., with volume-recruitment ma-
787
neuvers and higher frequencies) against whatever is the best lower frequency alternative at the time. At least such studies can now be designed with much more extensive evidence for the safety and possible benefit of such approaches. Alison Barbara Froese, M.D. Departments of Anesthesiology, Physiology, and Pediatrics Queen’s University Kingston, Ontario, Canada References 1. Derdak S, Mehta S, Stewart TE, Smith T, Rogers M, Buchman TG, Carlin B, Lowson S, Granton J, and the Multicenter Oscillatory Ventilation for Acute Respiratory Distress Syndrome Trial (MOAT) Study Investigators. High-frequency oscillatory ventilation for acute respiratory distress syndrome in adults: a randomized, controlled trial. Am J Respir Crit Care Med 2002;166:801–808. 2. Fort P, Farmer C, Westerman J, Johannigman J, Beninati W, Dolan S, Derdak S. High-frequency oscillatory ventilation for adult respiratory distress syndrome: a pilot study. Crit Care Med 1997;25:937–947. 3. Mehta S, Lapinsky SE, Hallet DC, Merker D, Groll RJ, Cooper AB, MacDonald RJ, Stewart TE. A prospective trial of high-frequency oscillation in adults with acute respiratory distress syndrome. Crit Care Med 2001;29:1360–1369. 4. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome: The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301–1308. 5. Venegas JG, Fredberg JJ. Understanding the pressure cost of ventilation: why does high-frequency ventilation work? Crit Care Med 1994;22: S49–S57. 6. Kolton M, Cattran CB, Kent G, Volgyesi G, Froese AB, Bryan AC. Oxygenation during high-frequency ventilation compared with conventional mechanical ventilation in two models of lung injury. Anesth Analg 1982;61:323–332. 7. McCulloch PR, Forkert PG, Froese AB. Lung volume maintenance prevents lung injury during high frequency oscillatory ventilation in surfactant deficient rabbits. Am Rev Respir Dis 1988;137:1185–1192. 8. Lapinsky SE, Aubin M, Mehta S, Boiteau P, Slutsky AS. Safety and efficacy of a sustained inflation for alveolar recruitment in adults with respiratory failure. Intensive Care Med 1999;25:1297–1301. 9. Clark RH, Gertsmann DR, Null DM. Prospective randomized comparison of high frequency oscillatory and conventional ventilation in respiratory distress syndrome. Pediatrics 1992;89:5–12. 10. Courtney SE, Durand DJ, Asselin JM, the Neonatal Ventilation Study Group. Early high frequency oscillatory ventilation (HFOV) vs. synchronized intermittent mandatory ventilation (SIMV) in very low birth weight (VLBW) infants: The Neonatal Ventilation Study Group [abstract]. Pediatr Res 2001;49:387A.
DOI: 10.1164/rccm.2206005
Muscle Mass, Not Body Weight, Predicts Outcome in Patients with Chronic Obstructive Pulmonary Disease It has been known for many years that weight loss and low body weight are common in patients with advanced chronic obstructive pulmonary disease (COPD). Moreover, it has been shown that low body weight is associated with increased mortality independent of lung function in patients with COPD (1, 2). However, unlike starvation in which the predominant body compartment affected is fat, the weight loss of COPD is similar to that of other chronic diseases and preferentially involves the loss of muscle mass. It is this loss of muscle mass, rather than body weight per se, that is likely to be primarily responsible for the observed negative conse-
quences. To the extent that body weight reflects lean body mass, body weight will be a good surrogate marker. But under conditions, such as obesity, in which body weight does not accurately reflect lean body mass, its discriminatory power will diminish markedly. Obesity is extremely common in the western world and is increasing in incidence. Thus, body weight may not be the ideal measure to reflect nutritional status in patients with COPD. For example, body weight and body composition have been measured in 255 patients with COPD (3). Low body weight implying nutritional depletion was defined as a weight
788
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 166 2002
less than 90% of ideal body weight and was seen in 89 of 255 patients (35%). A reduction in fat-free mass without a reduction in body weight was seen in 24 of 255 patients (9.4%). A reduction in body weight without a reduction in fat-free mass was seen in 23 of 255 patients (9.0%). Thus, discrepancies between measures of body weight and measures of fat-free mass are not uncommon in this patient population. Reduction in fat-free mass is a better predictor of peak exercise performance than the body mass index (4). Patients with a reduction in fat-free mass have worse healthrelated quality of life than underweight patients with preserved fat-free mass (5). These results again suggest that it is a reduction in fat-free mass rather than weight per se that affects patients functionally. In this issue of AJRCCM (pp. 809–813), Marquis and colleagues (6) evaluate whether a measure of muscle mass would be a better predictor of mortality than measures of body weight. They measured mid-thigh muscle cross-sectional area by computerized tomography scan and body mass index in 142 patients with COPD enrolled in their pulmonary rehabilitation program. Patients were followed for 41 ⫾ 18 months. During this period, 25 patients died representing a mortality rate of 17.6%. Mean body mass index was essentially normal for this group of patients, whereas mean midthigh muscle cross-sectional area was 72% of the normal area. This observation in itself points out the limitations of using body weight for assessing depletion of lean body mass. With multivariate analysis, only mid-thigh muscle cross-sectional area and the FEV1 were significant predictors of mortality. Thus, it is muscle mass and not body weight that is the important physiologic variable. In this study, muscle mass was measured directly by computerized tomography scan. The muscles that were assessed, the proximal muscles of the lower extremity, are particularly affected in patients with COPD (7, 8). Whether similar findings would have been obtained if the investigators had made a global assessment of fat-free mass remains to be determined. This distinction is important. Some methods for assessing fat-free mass are relatively simple and less cumbersome and expensive than a computerized tomography scan and could be more easily used in the clinical arena. The investigators tried to estimate mid-thigh muscle cross-sectional area from simple anthropomorphic measurements. Unfortunately, although this approach has been used successfully in normal subjects, it proved unreliable in this patient population. Patients entering a pulmonary rehabilitation program represented the study population in this study. Whether such patients differ from patients with COPD who do not choose to enter a rehabilitation program needs to be determined. The smoking history of the patients is not clearly stated. Many rehabilitation programs require that patients have successfully quit smoking for a certain period of time before entry in the rehabilitation program. Clearly, continued smoking would be expected to have a significant effect on progression of COPD and mortality, and how this influences muscle mass needs to be determined. In this study, the authors showed a relationship between muscle mass and mortality. Correlation, however, does not prove causation. Thus, we cannot say whether a reduction in muscle mass causes an increase in mortality or whether a reduction in muscle mass is merely a reflection of severity of disease. If a reduction in muscle mass is responsible for the increased mortality, then interventions that successfully increase muscle mass should lead to improvement in mortality. No study addressing this issue has been performed. Several short-term studies have addressed whether interventions
that increase muscle mass can lead to improvements in exercise capacity and quality of life. Addition of strength training to a pulmonary rehabilitation program led to significant increases in muscle strength and mass but to no additional increase in exercise capacity or quality of life as compared with that achieved by an endurance exercise program (9). Addition of anabolic steroids to pulmonary rehabilitation resulted in increases in fat-free mass but no additional increase in exercise capacity (10, 11). The mechanisms of muscle atrophy remain to be fully elucidated. In particular, is deconditioning largely or solely responsible for the muscle atrophy or does the disease itself induce systemic effects that promote muscle atrophy? In other clinical models of cachexia, such as cancer, acquired immunodeficiency syndrome, or chronic heart failure, hormonal changes and release of proinflammatory cytokine mediators are believed to be important mechanistically (12). Therapies that address these underlying mechanisms could potentially result in more favorable outcomes than simple reconditioning programs. In summary, this study highlights the significance of changes in body composition in patients with COPD and addresses the most important outcome parameter of all: mortality. It further points out the limitations of using simple measures of body weight to address nutritional status and muscle mass in patients with COPD. M. Jeffery Mador, M.D. Division of Pulmonary, Critical Care and Sleep Medicine State University of New York at Buffalo Veterans Administration Medical Center Buffalo, New York References 1. Wilson DO, Rogers RM, Wright EC, Anthonisen NR. Body weight in chronic obstructive pulmonary disease: the National Institutes of Health intermittent positive–pressure breathing trial. Am Rev Respir Dis 1989;139:1435–1438. 2. Gray-Donald K, Gibbons L, Shapiro SH, Macklem PT, Martin JG. Nutritional status and mortality in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1996;153:961–966. 3. Schols AMWJ, Soeters PB, Dingemans AMC, Mostert R, Fratzen PJ, Wouters EFM. Prevalence and characteristics of nutritional depletion in patients with stable COPD eligible for pulmonary rehabilitation. Am Rev Respir Dis 1993;147:1151–1156. 4. Baarends EM, Schols AM, Mostert R, Wouters EF. Peak exercise response in relation to tissue depletion in patients with chronic obstructive pulmonary disease. Eur Respir J 1997;10:2807–2813. 5. Mostert R, Goris A, Weling-Scheepers C, Wouters EF, Schols AM. Tissue depletion and health related quality of life in patients with chronic obstructive pulmonary disease. Respir Med 2000;94:859–867. 6. Marquis K, Debigare R, Lacasse Y, LeBlanc P, Jobin J, Carrier G, Maltais F. Midthigh muscle cross-sectional area is a better predictor of mortality than body mass index in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2002;166:809–813. 7. Bernard S, LeBlanc P, Whittom F, Carrier G, Jobin J, Belleau R, Maltais F. Peripheral muscle weakness in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998;158:629–634. 8. Gosselink R, Troosters T, Decramer M. Distribution of muscle weakness in patients with stable chronic obstructive pulmonary disease. J Cardiopulm Rehabil 2000;20:353–360. 9. Bernard S, Whittom F, LeBlanc P, Jobin J, Belleau R, Berube C, Carrier G, Maltais F. Aerobic and strength training in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999;159: 896–901. 10. Schols AMWJ, Soeters PB, Mostert R, Pluymers RJ, Wouters EFM. Physiologic effects of nutritional support and anabolic steroids in patients with chronic obstructive pulmonary disease: a placebo-con-
Editorials trolled randomized trial. Am J Respir Crit Care Med 1995;152:1268– 1274. 11. Ferreira IM, Verreschi IT, Nery LE, Goldstein RS, Zamel N, Brooks D, Jardim JR. The influence of 6 months of oral anabolic steroids on body mass and respiratory muscles in undernourished COPD patients. Chest 1998;114:19–28.
789 12. Kotler DP. Cachexia. Ann Intern Med 2000;133:622–634.
DOI: 10.1164/rccm.2206003
CD8 Cytotoxic T Cells and the Development of New Tuberculosis Vaccines The majority of immunocompetent people exposed to Mycobacterium tuberculosis (MTB) acquire lifelong protective immunity against active tuberculosis (TB). An understanding of the components of this naturally acquired protective immune response and its antigenic targets is a prerequisite for the rational design of an improved TB vaccine. MTB infection induces a potent human leukocyte antigen (HLA) class II– restricted Th1-type CD4 T cell response, and the human immunodeficiency virus epidemic has taught us that this limb of the cellular immune system is an essential part of the protective immune response against TB. The role of HLA class I–restricted CD8 cytotoxic T cells (CTLs) in host defense against MTB is less clear. In murine models, CD8 CTLs are essential for protection against MTB (1) and specifically seem to play a role in long-term immune control of the bacillus (2), but it was not until 1998 that the first HLA class I–restricted antigen-specific CD8 CTLs were identified in MTB-infected humans (3). Since then, our understanding of this important T cell population in the host response to MTB infection in humans has expanded apace. A total of 10 secreted and somatic MTB protein antigens have now been identified as targets of CD8 CTL in humans. In general, these T cells secrete interferon-␥, have cytolytic activity, recognize MTB-infected macrophages in vitro (4), and in some reports, actually suppress the growth of intracellular bacilli (5). These are attributes that are consistent with a protective role for these cells in vivo. Recently, novel vaccination strategies using naked DNA and recombinant viral vectors have made it possible to induce HLA class I–restricted CD8 CTLs, as well as HLA class II– restricted CD4 T cells in humans. Hence, the characterization of new MTB antigens as targets of both CD4 and CD8 T cells is an important part of the global effort to develop an effective vaccine against TB. If we could generate an array of defined peptide epitopes covering a broad enough range of common HLA types, HLA restriction of immune responsiveness would no longer be a barrier to the use of a multisubunit vaccine (based on multiple epitopes or the antigens that contain them) in genetically heterogeneous human populations (6). Delineating immunodominant epitopes will of course also help to monitor the immunogenicity of novel vaccines in clinical trials (6). It is in this context that the report by Lewinsohn and colleagues (7) in this issue of AJRCCM (pp. 843–848) represents another step forward on the long path toward an effective TB vaccine. This article demonstrates that Mtb39, already known to be a potent CD4 T cell antigen (8), is also a target of CD8 CTL in MTB-infected people. In an elegant series of experiments, Lewinsohn and coworkers (7) have exploited the highly sensitive interferon-␥ enzyme-linked immunospot (ELISPOT) assay (9) to identify circulating Mtb39-specific CD8 T cells that secrete interferon-␥, display cytolytic activ-
ity, and recognize MTB-infected target cells in vitro. These CD8 T cells were identified directly ex vivo in the ELISPOT assay, using autologous dendritic cells infected with an adenovirus recombinant for the Mtb39 gene (adenoMtb39). Each spot in the ex vivo ELISPOT assay represents the “footprint” of an interferon-␥–secreting CD8 T cell specific for an HLA class I–restricted epitope expressed at the surface of the adenoMtb39-infected dendritic cell. Lewinsohn and coworkers (7) went on to delineate the minimal peptide epitope in Mtb39 by testing cloned CTL against a panel of overlapping peptides representative of the entire Mtb39 protein. The minimal epitope is 10 amino acids long and is restricted through the HLA class I molecule HLA-B44. Interestingly, this immunodominant epitope might well have been overlooked if computerized peptide motif algorithms (for predicting which peptides are likely to bind HLA class I molecules) had been used to select candidate epitopes for screening. This study thus adds to the growing body of evidence that highlights overlapping peptides, with or without recombinant viral vectors (10), as the most efficient strategy for new epitope discovery. Another advantage of the ex vivo ELISPOT assay is that it gives the actual frequency of circulating epitope-specific interferon-␥–secreting CD8 T cells in peripheral blood (9). In the three donors in this report, this frequency was quite high, ranging from 1:3,000 to 1:15,000 of all circulating CD8 T cells, almost as high as the frequencies of CD8 CTLs specific for two other MTB antigens, ESAT-6 (3, 11) and CFP10 (12), noted previously. These CTL frequencies are of a similar level to those directed against viral epitopes in people with latent EBV or CMV infection, in whom CD8 CTL are believed to mediate long-term immune control of the virus. Of note, the donors in the article by Lewinsohn and coworkers (7), as well as the previously described donors with a potent CD8 CTL response to ESAT-6 (11) and CFP10 (12), had latent TB infection. We know little about the biology of the pivotal host–pathogen relationship that underlies latent TB infection, but the bacillary burden (and hence antigen load) in this state is considered to be very low, and the vast majority of people with latent TB infection have mounted a successful, long-lived, and protective immune response. Since it is antigen load that drives antigen-specific T cell frequencies in vivo, the high frequency of MTB antigen-specific CD8 CTLs in healthy people with latent TB infection is all the more remarkable. Moreover, in some donors, longitudinal followup over 2 years has shown that this potent, highly focused CTL response can also be very durable (11). Thus, careful immunologic studies of MTB-exposed healthy donors with latent TB infection, of which the article by Lewinsohn and coworkers (7) is the latest example, suggest that potent, long-lived CD8 CTL responses to certain MTB antigens may play a role in long-term control of MTB infection.
