In the study by Mahler and Harver, stepwise multiple regression revealed that the FEV,, age, and the baseline dyspnea index were the best predictors. Whereas Mahler and Harver did not find that the Dl added to the value of the predictive equation, they found in contrast to Carlson and coworkers that inclusion of a measurement of dyspnea added to the value of the prediction equation. For the maximal exercise ventilation (VEmax), the MW was again the best single predictor, and as with the Vo2max the correlation was highest in the most severely obstructed group. There was no real benefit in adding further flow variables in the multiple regression analysis. We noted however, that the addition of IC and RWTLC did make small contributions to the overall correlation. Our correlation compares favorably with that of Dillard et al who showed that a combination of FEV, and PIFR was better than the FEV, alone.
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Discussion (2)
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Discussion (1)
Although our study showed significant correlations between indices of expiratory airflow and exercise function in patients with COPD, the confidence intervals were wide, making individual predictions inaccurate. Moreover, stepwise multiple linear regression to include other predictive variables did not appreciably improve the estimates. Although the correlation coefficients were higher in the most severely obstructed group, the number of patients examined was small (n = 20) and a wide range of predicted values persisted. In patients with restrictive diseases, the correlation coefficients were generally poor. This wide scatter of the data detracts from the value of prediction equations that attempt to predict individual performance based on pulmonary function indices.
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Results (2)
Tables 4 and Table 5 show correlation coefficients for prediction of VEmax and Vo2max, respectively, in the three severity subsets of patients with airway obstruction. As seen previously, the measures of expiratory flow correlate most closely with measures of peak exercise capacity. Also apparent is the significantly higher r value for the relationship between MW and VEmax and both MW and FEV,, and Vo2max in the group with the most severe obstructive airways disease.
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Results (1)
Anthropometric and resting pulmonary function data for all obstructed, the three subsets of severity, and the patients with restrictive disease are shown in Table 1. The mean FEV,/FVC ratio ranged from 67 percent in the mild group to 35 percent in the severely obstructed group. As expected the FVC, maximum voluntary ventilation (MW), and DL decreased with increases in severity of the obstructive disease. Conversely, the TLC, RV, and RV/TLC ratio increased reflecting the greater degree of hyperinflation as the obstruction worsened. Table 2 shows the maximal exercise data in all patients with obstructive disease, the three subsets of severity, and the patients with restrictive disease. With increasing severity of obstruction, the peak Vo2 decreased. The mean peak Vo2 for the restricted patients was similar to the moderately obstructed group. Of note also was the presence of increasing degrees of hypoxemia in the moderate and severe groups, in addition to mild hypercapnia at peak work rates in the severely obstructed group.
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Methods (2)
Incremental Exercise Studies
Exercise studies were performed with the patient seated upright on an electrically braked cycle ergometer (Gould Goddart). All studies were performed with the patients breathing room air. The patient started with a 3-rnin period of unloaded cycling, after which the load was incremented by 10 to 20 W every 2 min until they reached their exercise fatigue limit. The volume of expired gas was measured by means of a turbine flow meter (VMM Alpha Technologies) and gas concentrations for carbon dioxide, and oxygen was sampled at the distal end of a 7-L mixing chamber by means of a mass spectrometer (model 1100 Perkins-Elmer). These values were used for calculation of minute ventilation (Ve), oxygen uptake (Vo,), and carlwn dioxide output (VcoJ. At the end of each 2-min exercise load, arterial blood gas samples were drawn and analyzed in a routine fashion in a blood gas analyzer (Corning 178, Ciba).
Prediction of Maximal Exercise Capacity in Obstructive and Restrictive Pulmonary Disease: Methods (1)
We retrospectively analyzed resting pulmonary function tests (PFTs) and incremental exercise studies collected over a 7-year period. We selected patients with obstructive disease based on an FEV/FVC of grater than one confidence interval (95 percent) below the normal predicted mean. These patients were further subdivided into mild, moderate, or severe obstructive airway disease based on the recommendations of the Interinountain Thoracic Society where mild= 1 to 2, moderate 2 to 4, and severe greater than 4 confidence intervals below the predicted level. Restrictive disease was defined as a reduction in the total lung capacity Inflow the 95 percent confidence interval of the predicted value. We found 149 patients with obstructive disease and 71 with restrictive disease in whom an exercise study had been performed within 1 month of the PFTs.
The role of exercise testing in chronic obstructive pulmonary disease (COPD) has expanded greatly to include assessment of exercise capacity, preoperative and disability evaluations, and exercise training prescription. Although viewed as the gold standard for determination of exercise capacity, expense and the need for trained personnel may preclude the use of exercise in the routine clinical setting. With these limitations in mind, considerable effort has been devoted to making estimates of peak exercise ventilation (VKmax) and oxygen uptake (Vo2max) in patients with COPD. Several investigators have suggested that estimates of peak exercise will enable physicians to evaluate the effort of the patient by comparing the actual level achieved with the predicted. These estimates have been based on resting measurements of flow rates, both inspiratory and expiratory, diffusion capacity, and inspiratory pressures.
None of the patients in this study presented data that are compatible with the AIDS diagnosis. They did not belong to any high-risk group, and results of serologic studies for HIV were normal throughout the evaluation (Table 2). On the other hand, the CD4:CD8 ratio was shown to be within normal limits, and CD4 subpopulations were always above 450 mm. A CD4 value of less than 250 mm in patients with HIV infection has been related to the risk of contracting PCP This pneumonia, however, can manifest in patients with a normal CD4 count, and in these cases, the PCP is probably associated with qualitative alterations of the cellular immune system. In support of this hypothesis, there are these findings in three patients with PCP who were previously healthy: absence of CD25+, peripheral T cells and low stimulation indexes in the presence of mitogens. We believe that the study of these parameters could serve as the basis for further investigation.
Although pleural effusion has been described in PCP, its presence is infrequent and is usually bilateral and low in quantity. Patients 1 and 4 presented unilateral pleural effusion, and in patient 4, the P carinii was identified in the pleural fluid specimen and in the pleural tissue obtained by thoracotomy. The P carinii has been isolated in multiple sites other than the lung, and only recently has it been isolated in the pleural tissue.
In two of our patients, microorganisms other than P carinii were isolated in respiratory specimens: P aeruginosa in patient 3, and multiple flora of Gram-negative germs in patient 5, the sample being taken with TPC in both cases. From a theoretic point of view, the possibility exists that patients of this report had suffered from pneumonia caused by microorganisms other than P carinii and the role of the latter was only a colonizer of the lung or airway. However, we consider this hypothesis most unlikely. Although there is serologic evidence of latent infection by P carinii,2 and the prevalence of this microorganism in adults with no predisposing illnesses is 0.31 percent according to necropsic studies, we have no knowledge of its identification in respiratory samples of subjects without pneumonia.