| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Allergy, Pulmonary and Critical Care Medicine, Department of Medicine, Vanderbilt University, Nashville, Tennessee
Correspondence and requests for reprints should be addressed to Lorraine B. Ware, M.D., T1218 MCN, 1161 21st Avenue S, Nashville, TN 37232-2650. E-mail: lorraine.ware{at}vanderbilt.edu
INTERSTITIAL LUNG DISEASE
Athol U. Wells
Royal Brompton Hospital
London, England
Pathogenesis of Idiopathic Pulmonary Fibrosis
Telomeres, regions of repetitive DNA at the end of chromosomes, shorten with cell division. Short telomeres limit tissue renewal capacity and are thought to play a role in diseases of aging. Telomerases are reverse transcriptase enzymes that add telomere repeats to chromosomes. Germ-line mutations in the components of telomerase (hTERT, hTR) have been associated with familial pulmonary fibrosis in some families (1). However, the role of telomerase mutations in idiopathic pulmonary fibrosis (IPF) is not known. Alder and colleagues (2) examined telomere lengths in leukocytes and alveolar epithelial cells in 64 patients with sporadic idiopathic interstitial pneumonia (IIP), also screening for mutations in hTERT and hTR, and reviewing cases for features of a telomere syndrome such as aplastic anemia or cryptogenic hepatic fibrosis. Telomere length was shorter in IIP cases (both IPF and non-IPF) than in age-matched control subjects, and this was mirrored by shortening of telomeres in leukocytes and alveolar epithelial cells that was of similar amplitude to that observed in familial IPF. A telomerase mutation was observed in only one case, although several individuals had a cluster of idiopathic pulmonary fibrosis and cryptogenic hepatic fibrosis, suggesting a telomere syndrome. These findings suggest that short telomeres are a signature of idiopathic interstitial pneumonia. The linkage between short telomeres and diseases related to aging supports the concept that IPF (and possibly other IIPs) can be viewed as a form of premature aging of the lung. Because known telomerase mutations did not account for short telomere length in this study, further studies are needed to identify the determinants of telomere shortening. The association of short telomere length with both IPF and other IIPs alike suggests the possibility of shared pathogenetic mechanisms between the IIPs.
Diagnosis: Nonspecific Interstitial Pneumonia
In the 2002 American Thoracic Society/European Respiratory Society (ATS/ERS) reclassification of the idiopathic interstitial pneumonias (3), uncertainties about the clinical phenotype of nonspecific interstitial pneumonia (NSIP) led to its designation as "provisional." Uncertainty has increased as differences have emerged between NSIP series in baseline high-resolution computed tomography (HRCT) features, bronchoalveolar lavage data, and outcome. Among patients with NSIP, clinical features of patient subgroups overlap with idiopathic pulmonary fibrosis (IPF), hypersensitivity pneumonitis (HP), or cryptogenic organizing pneumonia (COP). A workshop was convened by the American Thoracic Society to reach consensus on whether there is a discrete idiopathic NSIP entity and if so, to delineate its clinical, radiologic, and histologic definition and its distinction from other disorders (4). Investigators who had previously reported cases of idiopathic NSIP were invited to submit cases for review by the committee members (n = 305). After initial review, cases with complete clinical, radiologic, and pathologic information (n = 193) were reviewed in a series of workshops. Of 193 patients previously diagnosed as having NSIP, 67 cases were considered to have probable or definite NSIP, based upon a distinct phenotype, which included a histologic pattern of NSIP and an HRCT profile of predominantly lower zone disease, with volume loss, traction bronchiectasis, and reticular change, variably admixed with ground-glass attenuation. The remaining patients were reclassified to other diagnoses including IPF, COP, and HP. The report concludes that a genuine entity of idiopathic NSIP exists, with a typical clinical and HRCT profile, as distinct from variants of IPF, COP, and HP. Because only 67 of the 193 submitted cases were determined by the expert panel to have NSIP, it appears that the majority of patients currently diagnosed with NSIP would be reclassified as having an alternative disorder such as IPF, COP, or HP by a large panel of expert clinicians, radiologists, and histopathologists. On this basis, it appears prudent to consider a multidisciplinary evaluation of patients with NSIP, particularly if there are overlapping features with IPF, COP, or HP.
Staging of Interstitial Lung Disease in Systemic Sclerosis
In interstitial lung disease, a number of prognostic variables have been identified, including markers of disease severity. However, these predictors have largely been continuous variables without clear separation between patients with good and poor outcomes. Goh and colleagues (5) sought to develop a readily applicable prognostic algorithm for interstitial lung disease in systemic sclerosis with the goal of discriminating limited disease (good prognosis) from extensive disease (poor prognosis). The authors used a large retrospective cohort of patients with systemic sclerosis and interstitial lung disease in which baseline evaluation and follow-up was undertaken using a standardized protocol. After formal scoring and evaluation of threshold values for FVC and disease extent on HRCT against outcome, a simple staging system was developed in which rapid semi-quantitative HRCT assessment and an FVC threshold (70% of predicted normal) were integrated in a two-step algorithm. This scoring system identified large patient subgroups with limited and extensive disease; these subgroups had marked differences in outcome (both survival and progression-free survival). The staging system was equally accurate when tested by moderately experienced observers. Important advantages of this staging system are its rapidity, ease of application, reproducibility, and prognostic accuracy in moderately experienced hands. An immediate application of this staging system may be for selecting patients at higher risk of disease progression for enrolment in trials of new agents. Because this scoring system was developed in patients with systemic sclerosis, further work is needed to develop comparable prediction scores for other interstitial lung diseases.
Treatment of Idiopathic Pulmonary Fibrosis
IPF has a progressive and fatal course with median survival after biopsy confirmation of less than 3 years (6). No current treatment is effective in arresting progression of disease, despite a number of trials of new therapies in recent years. Endothelin-1 is a potent endogenous vasoconstrictor and profibrotic mediator that has been implicated in the pathogenesis of both pulmonary arterial hypertension and pulmonary fibrosis. The BUILD1 study (7) was a double-blind randomized placebo-controlled multicenter trial of the dual endothelin receptor antagonist bosentan compared with placebo in 158 patients with IPF (bosentan, n = 74; placebo, n = 84), diagnosed by biopsy or by ATS/ERS criteria. Patients were randomized to oral bosentan (62.5 mg twice daily increasing to 125 mg twice daily after 4 weeks) or placebo for a minimum of 12 months. The primary end-point was the six-minute-walk distance. Secondary end-points included time to death or disease progression, changes in individual pulmonary function variables, and changes in quality-of-life scales. There was no treatment effect of bosentan on the six-minute-walk distance, or on secondary end-points, in the whole population. In patients undergoing a diagnostic surgical biopsy (n = 99), there was a significant increase in time to death or disease progression, favoring bosentan. Despite skepticism about post hoc analyses, these subgroup analyses are essential in IPF studies to identify patient subgroups benefiting from treatment, with a view to refining criteria for future prospective studies. The apparent treatment effect of bosentan in the subgroup of patients with a biopsy confirmed diagnosis of IPF in the BUILD1 study justified a further study, with results expected within 12 months. The selective treatment effect in biopsied patients may reflect the fact that this subgroup tends to include a subset of patients with HRCT appearances less typical of IPF, with little or no honeycombing.
PULMONARY VASCULAR DISEASE
Harrison W. Farber
Boston University School of Medicine
Boston, Massachusetts
Pulmonary Thromboembolic Disease
Optimal management of acute pulmonary embolism (PE) requires medical expertise, diagnostic testing, and specialized therapies that may not be available consistently on weekends. Aujesky and colleagues (8) assessed whether associations exist between weekday or weekend admission and mortality or length of hospital stay for patients hospitalized with PE. Patients discharged from 186 acute care hospitals in Pennsylvania from January 2000 through November 2002 with a primary diagnosis of PE were analyzed. Random-effect logistic models were used to study the association between weekend admission and 30-day mortality. Discrete survival models were used to study the association between weekend admission and time to hospital discharge, adjusting for hospital (region, size, and teaching status) and patient factors (race, insurance, severity of illness, and use of thrombolytic therapy). Among 15,531 patient discharges with PE, 3,286 patients (21.2%) had been admitted on a weekend. Patients admitted on weekends had a higher unadjusted 30-day mortality rate (11.1% versus 8.8%) compared with patients admitted on weekdays, without any difference in length of stay. Patients admitted on weekends had significantly greater adjusted odds of dying (odds ratio, 1.17; 95% confidence interval [CI], 1.03–1.34) than patients admitted on weekdays. The higher mortality among patients hospitalized on weekends was driven by the increased mortality rate among the most severely ill patients. These data suggest inequality of care among PE patients dependent on the day of the week admitted, with the greatest differences observed for patients with the most severe disease. Although the underlying reasons for discrepancies in care cannot be discerned from this study, the findings suggest that efforts are needed to develop and ensure a consistent approach to the management of PE no matter the day of admission.
Chronic thromboembolic pulmonary hypertension (CTEPH) is optimally managed with surgical pulmonary endarterectomy (PTE). However, persistent pulmonary hypertension after surgical pulmonary endarterectomy is a major determinant of poor outcome. Skoro-Sajer and colleagues (9) sought to determine whether preoperative acute vasoreactivity can identify patients with CTEPH who will go on to develop persistent/recurrent pulmonary hypertension after endarterectomy, and whether the degree of acute vasoreactivity affects survival or freedom from lung transplantation. The authors analyzed right-sided heart catheterization data at baseline and after inhalation of 40 ppm nitric oxide for 20 minutes in 103 patients with CTEPH. Sixty-two patients subsequently underwent endarterectomy and were followed up for a median of 70.9 months. In this group, the change in mean pulmonary artery pressure (mPAP) during inhaled NO therapy predicted persistent/recurrent pulmonary hypertension after PTE. Patients in whom there was a reduction in mPAP of more than 10.4% with NO inhalation had a better postoperative outcome. The change in mPAP under iNO predicted death or lung transplantation in patients undergoing pulmonary endarterectomy with a sensitivity of 71% and a specificity of 90%, using a cutoff value of 10.4% with an estimated area under the receiver operating characteristic curve 0.87. A significant correlation was also found between change in mPAP and immediate postoperative pulmonary vascular resistance (r = 0.5, P < 0.0001). Given the morbidity and mortality associated with CTEPH, either treated medically or surgically, better selection of patients for medical therapy, endarterectomy, or lung transplant are necessary. This study suggests that the degree of acute pulmonary vasoreactivity as measured by the decrease in mPAP during inhaled NO therapy is a useful predictor of both immediate postoperative pulmonary vascular resistance and long term survival and less need for lung transplantation in adult patients with CTEPH.
Pulmonary Arterial Hypertension
The clinical relevance of exercise-induced pulmonary arterial hypertension (PAH), defined as a normal mPAP at rest but greater than 30 mm Hg with exercise, is not certain nor has its existence been well studied by direct measurements of central hemodynamics. Tolle and coauthors (10) hypothesized that exercise-induced PAH represents a mild but symptomatic stage of PAH, physiologically intermediate between resting PAH and normal. Using invasive cardiopulmonary exercise testing (radial and pulmonary arterial catheters and radionuclide ventriculography), they evaluated 406 consecutive clinically indicated cardiopulmonary exercise tests; the majority were ordered for evaluation of dyspnea or fatigue of unclear etiology or as part of an evaluation for cardiac or pulmonary transplantation. Subsequently, the invasive hemodynamic phenotype of exercise-induced PAH (n = 78) was compared with resting PAH (n = 15) and with that of normal subjects (n = 16). Patients with other or indeterminate diagnoses were excluded from the analysis. At maximum exercise, 
O2 was lowest in resting PAH, intermediate in exercise-induced PAH, and highest in normal subjects (55.8 ± 20.3% versus 66.5 ± 16.3% versus 91.7 ± 13.7% predicted). Mean pulmonary artery pressure (48.4 ± 11.1 versus 36.6 ± 5.7 versus 27.4 + 3.7 mm Hg) and pulmonary vascular resistance (294 ± 158 versus 161 ± 60 versus 62 ± 20 dyne · s · cm–5, P) followed an opposite pattern. An exercise-induced PAH plateau pattern (n = 32) was associated with lower O2max and maximum cardiac output and a higher resting pulmonary vascular resistance than an exercise-induced PAH takeoff pattern (n = 40). The plateau pattern was most common in resting PAH, and the takeoff pattern was present in nearly all normal subjects. The study confirms that exercise-induced PAH exists and is associated with exertional symptoms. The authors conclude that exercise-induced PAH may be an early, mild, and clinically relevant phase of the PAH spectrum. The clinical relevance of the entity and whether screening or treatment would be beneficial is not yet clear.
A second study relevant to PAH focused on genetic determinants of outcomes. Germline mutations in the gene encoding for bone morphogenetic protein receptor 2 (BMPR2) are associated with the development of both familial and sporadic pulmonary arterial hypertension (PAH). Sztrymf and colleagues (11) evaluated the influence of BMPR2 mutations on clinical outcomes in 223 patients with familial and idiopathic PAH from the French Network of Pulmonary Hypertension. Point mutation and large size rearrangements of BMPR2 were evaluated in all patients. Clinical, functional, and hemodynamic characteristics, as well as outcomes, were compared in BMPR2 mutation carriers and noncarriers. Sixty-eight BMPR2 mutation carriers (28 familial and 40 idiopathic PAH) were compared with 155 noncarriers (all displaying idiopathic PAH). Compared with noncarriers, BMPR2 mutation carriers were younger at diagnosis of PAH (36.5 ± 14.5 versus 46.0 ± 16.1 yr, P < 0.0001), had higher mean pulmonary artery pressure (64 ± 13 versus 56 ± 13 mm Hg, P < 0.0001), lower cardiac index (2.13 ± 0.68 versus 2.50 ± 0.73 L/min/m2, P = 0.0005), higher pulmonary vascular resistance (17.4 ± 6.1 versus 12.7 ± 6.6 mm Hg/L/min/m2, P < 0.0001), lower mixed venous oxygen saturation (59 ± 9% versus 63 ± 9%, P = 0.02), shorter time to death or lung transplantation and younger age at death, but similar overall survival. The authors conclude that BMPR2 mutation carriers with both familial and idiopathic PAH present approximately 10 years earlier than noncarriers, with more severe hemodynamic compromise at diagnosis, although overall survival after diagnosis is similar. The mechanisms by which BMPR2 mutations cause earlier onset of PAH remain unclear.
Pulmonary Venous Stenosis
Catheter ablation involving the pulmonary veins and the surrounding left atrial tissue is used increasingly to treat chronic atrial fibrillation. Despite the fact that catheter ablation may cure a substantial proportion of patients, severe complications have been observed. Pulmonary vein stenosis is a new clinical entity produced by radiofrequency energy delivery mainly within or at the orifice of the pulmonary veins. It is an under-recognized complication of radiofrequency ablation for atrial fibrillation and there is much uncertainty about the causative factors, incidence, diagnosis, treatment, and long- term sequelae. Pulmonary vein occlusion after radiofrequency ablation is probably underrecognized, with an estimated incidence ranging from 2% to 42% depending on ablative technique and investigative method (12, 13). Nehra and coworkers (14) present a case of pulmonary venous stenosis after catheter ablation for atrial fibrillation that culminated in pneumonectomy of the involved lung underscoring the fact that the functional loss of an entire lung due to pulmonary vein occlusion may mitigate any functional improvement derived from radiofrequency ablation. They further note that the usual diagnostic delay associated with pulmonary venous injury precludes early intervention to prevent these strictures. The authors make recommendations for the pre- and post-procedure evaluation of catheter ablation candidates, including the following. (1) Candidates with shortness of breath or exertional dyspnea should undergo pulmonary function studies with carbon monoxide diffusing capacity before their first radiofrequency ablation. Unrelated significant pulmonary disease may influence the decision to perform ablation. (2) Patients with new or progressive shortness of breath and dyspnea after radiofrequency ablation should be evaluated with pulmonary function studies, ventilation–perfusion scintigraphy, and computed tomographic angiography. (3) The diagnosis of stenosis of any severity in a single pulmonary vein should constitute a contraindication to radiofrequency ablation. It is important for clinicians to be aware of this entity and to include it in the differential diagnosis of unexplained dyspnea in patients who have undergone catheter ablation for atrial fibrillation.
Pulmonary Vasculopathy of Sickle Cell Disease
Pulmonary hypertension (PH) is emerging as a major complication and independent risk factor for death among adults with sickle cell disease (SCD). Using surface-enhanced laser desorption/ionization time of flight mass spectrometry (SELDI-TOF MS), Yuditskaya and colleagues (15) evaluated potential biomarkers of PH in plasma specimens from 27 homozygous sickle cell (HbSS) patients with PAH and 28 without PAH. They hypothesized that identification of new disease markers might lead to new mechanistic insights, prospective risk assessment, and potential risk reduction interventions. In patients with PAH, analysis consistently showed lower abundance of a 28.1-kD peak identified by high-resolution mass spectrometry as the oxidant-scavenging protein apolipoprotein A-I (apoA-I), which correlated with plasma levels by clinical assay of apoA-I and high-density lipoprotein (HDL) levels. HbSS patients with lower apoA-I levels also displayed impaired vasodilatory responses to acetylcholine (mean ± SEM, 189% ± 34% [n = 13] versus 339% ± 51% [n = 13], P < 0.001) suggesting that endothelial dysfunction may mediate this effect in PAH. As a group, patients with SCD demonstrated significantly lower apoA-I levels than African-American control subjects. The PAH cohort was further characterized by high levels of apolipoproteins A-II and B and serum amyloid A, and low levels of haptoglobin dimers and plasminogen. The authors suggest that these observations imply a relationship of apolipoproteins to the development of PAH vasculopathy in SCD. This is important, since these findings may implicate another pathway (in addition to nitric oxide and reactive oxygen species) in the development of PH in SCD. However, there are several limitations of this study. First, it is not clear whether all these patients had true PAH, since assessment of PH was based on tricuspid regurgitant jet velocity measured by echocardiography rather than by pulmonary artery catheterization. In addition, there are multiple different etiologies of PH in patients with sickle cell disease, and only PAH was investigated in this study.
PULMONARY INFECTIONS
Judd Shellito
Louisiana State University Health Sciences Center
New Orleans, Louisiana
Viral Infections
The emergence in 2003 of highly pathogenic H5N1 avian influenza has generated concern about a future influenza pandemic, similar to the 1918–19 influenza pandemic. To investigate the causes of death associated with the 1918–1919 epidemic, Morens and colleagues (16) re-cut lung tissue blocks from 58 military fatalities during 1918–19 and reviewed 109 published autopsy series from pandemic cases. Analysis of the re-cut lung tissue showed evidence of bacterial pneumonia in virtually all cases, with concomitant necrosis and desquamation of respiratory epithelium. Published autopsy series from 1918 cases reported that deaths were rarely caused by the influenza virus itself, but resulted from severe secondary pneumonia due to respiratory tract–colonizing bacteria. Lung, blood, and pleural fluid cultures from pandemic cases found bacterial superinfection in 93% of cases. Bacteria implicated in fatal cases were pneumococcus, streptococcus, staphylococcus, and bacillus influenzae (Haemophilus influenza) and mixed bacterial infections were seen in approximately 25% of cases. The length of time between symptom onset and death (10 days) was also more consistent with bacterial than viral pneumonia. The authors conclude that interactions between influenza virus and respiratory tract bacteria may be responsible for both the severity of the 1918–1919 pandemic and for the fatalities. This study should help to guide preparative efforts for future pandemics, a topic of considerable relevance in 2009 with the emergence of the epidemic H1N1 influenza A strain (17). Although it is unclear whether a future pandemic would follow the 1918–1919 pattern, the authors recommend that preparation for a future pandemic should include vaccination against bacterial pathogens and distribution of relevant antibiotics.
Many infants hospitalized for respiratory viral infections go on to develop asthma, but it is unclear whether this is a causal relationship or a marker of asthma susceptibility. To address this question, Wu and coinvestigators (18) studied a birth cohort of 95,310 children from five winter seasons in a Tennessee Medicaid population database. The relationship between birth date, the month of winter virus peak, and the development of asthma by age 5 was assessed for each infant. Infants who were approximately 4 months of age at the time of the winter virus peak were much more likely both to be hospitalized for bronchiolitis and to develop childhood asthma by age 5. This relationship was also true for high-risk asthma (hospitalization, steroid treatment). A history of maternal asthma further increased the risk for high-risk asthma by 82%. The parallel between seasonal variation in asthma risk and seasonal variation in bronchiolitis strongly supports a causal relationship between infant viral infection and early childhood asthma. These findings support efforts to prevent infant viral infection (vaccines, antiviral therapies) or to avoid exposure to virus at specific ages. Limitations of the study include lack of identification of specific viral pathogens (respiratory syncytial virus, rhinovirus, etc.) and the inability to consider atopic sensitization as a risk factor for asthma. Whether there is a link between winter virus infection and asthma in later childhood or as an adult is also not clear.
Mycobacterial Infections
Pulmonary infection with nontuberculous mycobacteria (NTM) in patients without obvious immunosuppression or pre-existing lung disease is increasing, but clinical data on this condition are sparse. There have been intriguing associations between pulmonary NTM infection and body habitus and pulmonary NTM infection and genetic abnormalities linked to bronchiectasis (e.g., CFTR). Kim and colleagues (19) report a prospective study of morphologic, immunologic, and genetic aspects of pulmonary NTM infection. The primary goal was to determine the factors that predispose to pulmonary NTM infection. Sixty-three patients referred to the NIH with pulmonary NTM from 2001–2005 with no prior history of HIV or CF were studied. Patients were compared with matched control subjects with regard to age, sex, body type, smoking status, place of residence, presenting symptoms, type of NTM infection, behavioral survey, menstrual history, prior respiratory history, aversion to coughing, routine labs, 
-1 AT levels, chest CT without contrast, CXR, 2D ECHO, PFTs, CFTR gene sequencing, sweat chloride test, and antibody titers. The patients were predominantly female (95%) with a distinct morphotype compared with control subjects—taller, thinner (even when controlling for NTM infection–related weight loss), and with excess scoliosis and pectus excavatum. 36.5% of patients had CFTR mutations compared with controls (16%); most were heterozygotes or compound heterozygotes, with normal sweat chloride levels. There was no aversion to coughing among the patients with NTM. The link between body habitus and NTM infection is intriguing, although it is unclear whether a specific body habitus predisposes to NTM infection directly or is simply a marker of a subtle host defense defect. The link between CFTR mutations and NTM suggests that the phenotype associated with CFTR mutations may extend well beyond classical cystic fibrosis, although based on the current study, CFTR mutations alone cannot explain the distinct clinical entity of pulmonary NTM.
Pneumonia
Ventilator-associated tracheobronchitis (VAT) has been described as the development of purulent respiratory secretions while undergoing mechanical ventilation (20). Palmer and colleagues (21) hypothesized that treating the proximal airway infection (VAT) with high concentrations of aerosolized antibiotics would reduce or prevent signs and symptoms of deep respiratory tract infection (ventilator-associated pneumonia, VAP) and also decrease bacterial resistance and the use of systemic antibiotics. They conducted a single-center, double- blind, randomized trial of aerosolized antibiotics (ABX, n = 19) compared with placebo (n = 24) to treat VAT. VAT was defined as greater than or equal to 2 ml of purulent secretions in 4 hours by suctioning plus organisms visible on Gram stain. Treatment in the ABX group was selected based on organisms visualized by Gram stain. Gram-positive organisms were treated with vancomycin 120 mg aerosolized every 8 hours and gram-negative organisms were treated with gentamicin 80 mg aerosolized every 8 hours. Treatment was given for 14 days or until extubation. The primary outcomes were indicators of VAP and clinical pulmonary infection score. Indicators of VAP declined in the ABX group from 74% to 36% at end of study, and were unchanged in the placebo group (75% to 79%). Aerosolized antibiotics were associated with a lower clinical pulmonary infection score, a lower blood WBC count at Day 14, reduced antibiotic resistance, less use of systemic antibiotics, and increased weaning. There was no effect on mortality. One of the major limitations of this study is that most patients with VAT also had VAP at the time of randomization. Thus, the effects of aerosolized antibiotics in this study might be best interpreted as adjunctive treatment for VAP rather than its prevention. Larger multicenter studies are needed for validation of these findings prior to any translation to routine clinical application. The lack of any increased antibiotic resistance in the aerosolized antibiotic group is encouraging, as this has been a confounder for prior studies of aerosolized antibiotics.
A second study in the area of pneumonia focused on prognosis in community-acquired pneumonia (CAP). Criteria to predict prognosis in patients with CAP have been validated with respect to 30-day mortality. Prediction tools such as CURB65 (22) are helpful for decision-making with regard to home versus hospital care. Less information is available to triage hospitalized patients with CAP for care in the ICU. As part of a larger Australian CAP study, Charles and colleagues (23) analyzed 882 episodes of CAP in 862 hospitalized patients to determine which clinical criteria were associated with the use of invasive respiratory (ventilator) or vasopressor support (IRVS). ICU admission occurred in 13.4% of cases and IRVS was required in 10.3%. Using clinical data, these investigators developed a prediction tool, SMART-COP: systolic BP < 90 (2 points), multi-lobar infiltrates on CXR (1 point), albumin < 3.5 gm/dl (1 point), respiratory rate age-adjusted increase (1 point), tachycardia > 125 bpm (1 point), confusion (1 point), oxygen age-adjusted decrease in PaO2 or SaO2 (2 points), pH (arterial) < 7.35 (2 points). With this tool, the point total can be used to predict the need for IRVS, and this predictive value exceeds that of CURB65 and the Pneumonia Severity Index (24) score. A SMART-COP score of 3 points or more identified 92% of patients who received IRVS. These findings were validated in five external CAP databases. Although there are some limitations of this study, the primary one being the inclusion of predominantly patients admitted to large urban hospitals, the findings suggest that the SMART-COP prediction rule may be useful for decision making regarding ICU admission for patients with CAP. Whether this clinical index is superior to or might be enriched by measurement of plasma biomarkers of severity of CAP is an important area for future investigation.
CARDIOPULMONARY EXERCISE TESTING AND PULMONARY REHABILITATION
Denis E. O'Donnell
Queens University
Kingston General Hospital
Kingston, Ontario, Canada
Exercise Testing in Idiopathic Pulmonary Fibrosis
IPF is characterized by progressive dyspnea, disability, and poor long-term survival. However, the rate of disease progression varies from patient to patient, complicating decisions regarding referral for lung transplantation and enrollment in experimental clinical trials. Fell and coinvestigators (25) hypothesized that symptom-limited peak O2 uptake (VO2), which is an integrated measure of respiratory, cardiovascular, locomotor muscle, and symptomatic impairment in IPF, would predict mortality in patients with IPF. They conducted a retrospective analysis of 117 patients with serial cardiopulmonary exercise tests and examined the relationship between peak VO2 and survival as calculated from the date of the first exercise test. After adjusting for age, sex, smoking status, baseline forced vital capacity, and diffusion capacity for carbon monoxide, the authors found that peak VO2 of less than 8.3 ml/kg/minute at baseline was associated with an increased risk of death (n = 8; hazard ratio, 3.24; 95% CI, 1.10–9.56; P = 0.03). The investigators concluded that the cut-off threshold of peak VO2 of 8.3 ml/kg/minute during incremental cardiopulmonary exercise test provides prognostic information for patients with IPF. This study adds to the growing body of evidence that peak VO2 measured during incremental cardiopulmonary exercise testing is a useful prognostic indicator of survival in patients with a variety of cardiopulmonary diseases. The study is limited by its retrospective design and the fact that no new physiologic insights were provided on mechanisms of reduced peak VO2 in IPF; therefore, future prospective population studies are required to further validate the clinical utility of this new VO2 threshold and to determine if it is responsive to therapeutic manipulation.
Pulmonary Rehabilitation in COPD
Respiratory mechanical abnormalities and associated respiratory discomfort limit participation in exercise training programs by some patients with COPD. Several interventions have been proposed as potential adjuncts to exercise training in symptomatic patients with COPD including noninvasive ventilation, hyperoxia, bronchodilator therapy, and specific inspiratory muscle training. The benefits of these interventions have not been conclusively demonstrated. Eves and colleagues (26) hypothesized that a combination of mechanical unloading and hyperoxia using helium hyperoxia (60:40, He:O2) would cause sufficient mechanical unloading of the respiratory muscles to facilitate incremental training, thereby increasing physiologic training effects in patients with COPD. The investigators undertook a randomized controlled study in 38 nonhypoxic patients with COPD (FEV1 47 ± 17% predicted). Patients received either helium hyperoxia (n = 19) or air (n = 19) during a supervised 6-week exercise training program. Cycle endurance time at a constant work rate after rehabilitation was the primary outcome measure. Secondary outcomes included health-related quality of life and change in exertional symptoms. The investigators found that adjunct helium hyperoxia therapy during pulmonary rehabilitation permitted patients to reach higher exercise intensity during training levels more quickly than in those randomized to air. Helium hyperoxia was associated with significantly greater improvements in cycle endurance time (HH versus air; 9.5 ± 9.1 versus 4.3 ± 6.3 minutes, respectively; P < 0.05). Improvements in St George's Respiratory Questionnaire scores were greater with helium hyperoxia (–7.6 ± –6.4) versus air (3.6 ± –5.6) (P < 0.05). The authors concluded that adjunct helium hyperoxia therapy facilitated the graduated exercise training regimen and resulted in greater improvements in cycle exercise endurance and perceived health status. It should be noted that patients who were enrolled in the study were nonhypoxemic with moderate disease severity; thus, it is unclear whether the benefits of helium hyperoxia adjunctive training are generalizable to hypoxic patients with more severe disease who constitute the bulk of individuals currently entering pulmonary rehabilitation programs. Furthermore, there was no improvement in VO2max in either group after six weeks of training. In addition, questions remain about the practicality and costs of helium hyperoxia-assisted training. The study provides a solid rationale for a larger, randomized, controlled trial to test the utility and cost effectiveness of helium hyperoxia as an adjunct to exercise training.
Exercise limitation in patients with COPD is multifactorial. Borghi-Silva and colleagues (27) took a different approach to unloading the respiratory muscles in COPD, hypothesizing that mechanical unloading of the respiratory muscles with noninvasive proportional assist ventilation (PAV) during exercise would improve oxygen delivery to the active peripheral muscle by increasing cardiac output and/or arterial oxygen content. They studied sixteen nonhypoxemic patients with moderate to severe COPD who were randomly assigned to PAV or sham unloading during constant work rate cycle exercise at 80% of predetermined peak work rate. Locomotor muscle oxygenation was measured by near-infrared spectroscopy (NIRS) during exercise. PAV was associated with improvements in perceived leg discomfort and end-exercise blood lactate. Leg muscle oxygenation was significantly enhanced with PAV compared with sham as indicated by an exercise-related decrease in the change in oxyhemoglobin and in the tissue oxygenation index (P < 0.01). There were no between-treatment differences in indirectly measured cardiac output or in arterial O2 saturation between PAV and sham. The results support the existence of a ventilatory "steal" phenomenon in patients with severe respiratory mechanical abnormalities during exercise whereby peripheral muscle oxygenation can be compromised. The corollary is that interventions that unload the overburdened inspiratory muscles may improve peripheral muscle oxygenation. This study is important because it is the first to explore the relationship between respiratory mechanical unloading and peripheral muscle oxygenation in patients with COPD during exercise and provides new insights into mechanisms of exercise limitation in such patients. Changes in peripheral muscle oxygenation may be one mechanism that underlies the common finding of intolerable leg discomfort with exercise in patients with COPD. The clinical relevance of these improvements in oxygen delivery to the peripheral muscles remains to be determined; however, these results suggest that this could very well be an important (and previously neglected) mechanism of exercise improvement after a number of therapeutic interventions in COPD.
Despite the proven clinical benefits of pulmonary rehabilitation in improving dyspnea, exercise capacity and health-related quality of life in symptomatic patients with COPD, access to hospital-based outpatient pulmonary rehabilitation is very limited world-wide. This unfortunate situation prompted Maltais and colleagues (28) to evaluate the effectiveness of a home-based self-monitored exercise training program as an alternative to traditional pulmonary rehabilitation. In a multicenter trial in 10 centers in Canada, 252 patients with moderate-to-severe COPD were randomized to either home-based or hospital-based PR. Dyspnea measured by the Chronic Respiratory Questionnaire (CRQ) at 1 year was the primary outcome measure. Both interventions were associated with significant improvements in the CRQ dyspnea subscale by an amount that exceeded the minimal clinically important difference (0.5). The difference in dyspnea scores between the interventions at 1 year were not clinically important and did not exceed the prespecified noninferiority margin of 0.5. Both programs had comparable and clinically meaningful improvements in secondary outcomes that included exercise endurance and health-related quality of life. In accordance with previous studies, the subjective and physiologic benefits eroded with time despite the institution of a partly supervised exercise maintenance program. The investigators concluded that self-monitored home-based pulmonary rehabilitation was equivalent in efficacy to supervised hospital-based pulmonary rehabilitation. Although there are some limitations, this is an important study and is one of the first to provide a solid scientific foundation for the future development of home-centered models of exercise reconditioning in patients with COPD.
Lung hyperinflation imposes significant mechanical constraints on lung volume expansion during exercise and is thought to contribute to intolerable dyspnea and exercise limitation. It has long been recognized that respiratory mechanical factors in COPD may compromise cardiac performance. Possible mechanisms include: (1) increased intrathoracic pressures, which reduce right ventricular preload and increase left ventricular afterload; (2) excessive expiratory muscle activation, which compromises venous return; and (3) acute pulmonary hypertension, secondary to dynamic pulmonary hyperinflation. Vassaux and colleagues (29) hypothesized that lung hyperinflation, as assessed by the inspiratory capacity/total lung capacity (IC/TLC) ratio, would be associated with reduced dynamic cardiac function as indirectly measured by the oxygen pulse during rest and exercise. Eighty-seven patients with severe COPD and 46 healthy control subjects underwent pulmonary function and cardiopulmonary exercise testing. Patients with less lung hyperinflation (IC/TLC > 25%) were compared with patients with more lung hyperinflation (IC/TLC 
25%). The IC/TLC ratio and the peak oxygen pulse were highly correlated (r = 0.95, P = 0.001). Patients with the most severe hyperinflation (IC/TLC ratio 25%; n = 45) had lower peak exercise capacity and oxygen pulse at a standardized work rate during exercise than less hyperinflated patients (IC/TLC > 25%). These results add to the growing body of evidence that respiratory mechanical abnormalities such as resting and dynamic hyperinflation compromise the rise in cardiac output during physical exertion in patients with severe COPD. This study is one of the first to explore this interaction in a large COPD cohort and sets the stage for further research on the clinical implications of abnormal cardiopulmonary interactions during activity in COPD.
FOOTNOTES
This is the first in a series of four executive summaries of the Clinical Year in Review Sessions presented at the American Thoracic Society International Conference in May, 2009. The main topics of each talk have been abstracted by the session chair based on the annotated bibliography provided by each presenter.
Conflict of Interest Statement: L.B.W. received grant support from Luminex $1,001–$5,000, Sirius Genomics $10,001–$50,000, and from the NIH $100,001-more.
(Received in original form July 3, 2009; accepted in final form July 6, 2009)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
