Exercise Testing in Pulmonary Sarcoidosis*

Transcription

1 Exercise Testing in Pulmonary Sarcoidosis* Joseph I. Matthews, Lt Col, MC, F.C.C.P;t and RobertG. Hooper, M.D., F.C.C.P.t The variable natural history of sarcoidosis and the toxicity of corticosteroids result in many clinical situations where there is controversy concerning the need for treatment. Progressive incremental testing is an excellent method to identify physiologic mechanisms responsible for exercise limitation. It is therefore ideal to determine if subjective symptoms such as dyspnea are due to cardiac abnonnalities, pulmonary abnormalities, or poor physical conditioning. Thirtyone patients with sarcoidosis underwent progressive incremental exercise testing. Four of 14 asymptomatic patients and eight of 17 symptomatic patients demonstrated pulmonary abnormalities which potentially limited exercise tolerance. These consisted of an abnormal respiratory pattern or gas exchange abnormalities, or both. Patients with corn- pletely normal routine pulmonary function studies almost always performed normally with exercise. Symptomati#{243} patients with multiple abnormalities on routine pulmonary function studies invariably demonstrated a pulmonary limitation on exercise testing. Patients with one or two abnormalities on routine pulmonary function studies, regardless of.the presence or absence of parenchymal infiltrates, required exercise testing to determine if symptoms were due to physiologically significant abnormalities of the respiratory system. The important variables necessary to be measured, arterial desaturation and an abnormal respiratory pattern, can be measured noninvasively with a minimum of equipment. he natural history of pulmonary sarcoidosis is known to be variable; nevertheless, treatment with corticosteroids has become standard practice for symptomatic patients with abnormal roentgenograms and abnormal pulmonary function. The toxicity of corticosteroids plus the variable natural history result in many clinical situations where there is controversy regarding the need for therapy, and difficult therapeutic decisions must be made. Examples of such situations include the symptomatic patient with normal ventilation as measured on routine pulmonary function studies, the patient with pulmonary parenchymal abnormalities by chest roentgenogram and normal ventilation as measured on routine pulmonary function studies, and the asymptomatic patient with abnormal ventilation as measured on routine pulmonary function studies. We, and others,67 have noted the frequent disparity between subjective symptoms and the objective measurement of pulmonary function studies. Exercise testing, specifically hypoxemia with exercise,78 has been touted as correlating better with overall lung pathology than routine pulmonary function studies. However, to date, no one has systematically exercised a large population of patients with sarcoidosis and evaluated the clinical usefulness of exercise testing. *Fmm the Pulmonary Disease Service, Walter Reed Army Medical Center, Washington, D.C. The opinions or assertions contained herein are the private views of the authors and are not to be construed as reflecting the views of the Department of the Army or the Department of Defense. tpresently Chief; Pulmonary Disease Service, Brooke Army Medical Center, Fort Sam Houston, Texas. Wirector, Pulmonary Medicine, St. Luke s Hospital Medical Center, Phoenix. Manuscript received April 22; revision accepted June 25 Reprint requests: Lt. Col. Matthews, Box446, BrookeArmy Medical Center, Fort Sam Houston, Texas Exercise requires an integrated physiologic response from the cardiopulmonary system which can be measured and quantified. This response has been well studied.9 #{176} Normally, there is a linear increase in cardiac output (heart rate) and oxygen consumption with increasing levels of work. This relationship between the heart rate, oxygen consumption, and work load remains linear until the maximum predicted heart rate is reached. Minute ventilation and carbon dioxide production also increase linearly at low levels of work. However, under conditions of rapidly increasing work load, a crisp anaerobic threshold is reached at approximately 6 percent of the maximum cardiac output. At this point, lactic acid accumulates, carbon dioxide production is increased, and expired carbon dioxide concentration plus minute ventilation increase out of proportion to oxygen consumption and cardiac output. The response of a normal 37-year-old man tested in our laboratory is shown in Figure 1. The shaded areas show the normal range. #{176} Minute ventilation normally increases with a small increase in respiratory rate and a large increase in tidal volume (TV). The response of a 37-year-old man tested in our laboratory is shown in Figure 2. The solid line depicts the normal range. #{176} Under normal conditions, exercise is limited by cardiac output and physical conditioning. Unless the fbrced expiratory volume at one second (FEV,) is reduced significantly, maximum predicted ventilation (FEy1 x 35)11 is not attained, even in highly trained athletes. Potential causes of ventilatory limitation include (a) gas exchange abnormalities as evidenced by an increased alveolar-arterial gradient for oxygen P(A-a)O, and arterial oxygen desaturation; (b) an abnor- CHESTI83/1/JANUARY,

2 C E. I- eo I.., N V2 (stpd) (I/nUn) 3 3. I2(STPD) (I/mm) 2 16 C ISO #{163} 14 I U 12 1 eo V2(STPD) (I/mm) NORMAL CONTRL - PATIENT 7#{149}--- PATIENT z I- _- I.- 4 U, eo PATIENT 31 B BI MIPSJTES EXERCISE FIGURE 1. Graph plotting (clockwise from lower left) heart rate (f,) versus oxygen consumption (Vo5), minute ventilation (V5) versus Vo,, carbon dioxide production (Vco2) versus Vo2, and oxygen saturation versus minutes of exercise. Shaded areas depict normal range #{176}for men. (Patient 8 is a woman; normal range for women is shifted to left.) A normal control and three representative patients with sarcoidosis are shown. The only abnormality is the decrease in arterial saturation from 97 percent to 92 percent in patient 7. mal inefficient ventilatory pattern manifested by a rapid respiratory rate, low TV, increased dead space fraction (V/T), and increased work of breathing; (c) reduced ventilatable lung to the point where the maximum predicted ventilation is attained before the maximum predicted heart rate. When arterial oxygen desaturation or an inefficient ventilatory pattern occurs, the individual can partially compensate by increasing his cardiac output (heart rate), and thus, the individual may still attain his maximum predicted heart rate. Individuals with reduced ventilatable lung will not reach their maximum predicted heart rate. Individuals in poor physical condition may not reach a cardiac or pulmonary limitation. Progressive incremental exercise testing, with measurement of appropriate cardiac and pulmonary variables, would thereibre be an excellent way to identify physiologic mechanisms responsible Ibr exercise limitation. It would be an ideal method to determine if subjective symptoms such as dyspnea are due to cardiac abnormalities, pulmonary abnormalities, or poor physical conditioning. This report summarizes the experience at Walter Reed Army Medical Center in the use of progressive incremental exercise testing as a technique to aid in the evaluation of patients with sarcoidosis. Using a cycle ergometer, 31 patients with sarcoidosis were tested. The objectives of this study were as fbllows: (1) to evaluate the relationship between chest symptoms, roentgenographic abnormalities, and physiologic abnormalities as measured by exercise; (2) to establish if and when such testing was of value in patients with pulmonary sarcoidosis; and (3) to determine whether exercise testing was helpful in making a decision to utilize corticosteroid therapy. METHODS The patients included 21 men and 1 women with an age range of 21 to 45 years. All patients had a complete medical history taken and a complete physical examination performed. Routine laboratory studies included a complete blood count, a urinalysis, and an automated blood chemistry profile which included calcium, phosphorus, bilirubin, alkaline phosphatase, serum glutarnic-oxaloacetic transaminase, lactic dehydrogenase, urea nitrogen, and creatinine. Routine pulmonary function studies included a forced vital capacity (FVC), an FEy1, FEV1/FVC ratio, residual volume (RV) by nitrogen washout, a total lung capacity ( rlc), and a diffusion capacity for carbon monoxide (Dco), measured on an SRL automated laboratory. Histologic confirmation of sarcoidosis was obtained in 3 of the patients (transbmnchial lung biopsy specimen, 19; mediastinoscopy, five; open lung biopsy specimen, three; supraclavicular lymph node biopsy specimen, two; and liver biopsy specimen, one); only one patient was without tissue diagnosis (case 15). This patient presented with anterior uveitis and bilateral hilar adenopathy and refused an invasive procedure. The roentgenographic stages of disease included nine patients with stage 1 disease or hilar adenopathy alone, 18 patients with stage 2 disease or hilar adenopathy and parenchymal lung disease, and four patients with stage 3 disease or parenchymal lung disease without hilar adenopathy. The parenchymal abnormalities consisted of one patient with a multinodular alveolar infiltrate resembling metastatic cancer, and one patient with bilateral upper lobe apical infiltrates. The remaining 2 patients had a diffuse bilateral interstitial pattern. 76 Exercise Testing hi Pulmonary Sarcoldosis (Matthews, Hooper)

3 Progressive incremental work rate exercise testing was performed using a cycle ergometer. During exercise, the ECG; heart rate, expiratory flow, minute ventilation, tidal volume, and mixed expired concentrations of carbon dioxide and oxygen were continuously monitored and recorded on an eight-channel thermal tipped recording system (Hewlett Packard). Carbon dioxide concentrations were measured from a mixing chamber with an LB-2 Beckman analyzer, while oxygen concentrations were measured on an OM-U Beckman analyzer. Arterial oxygen saturation was continuously estimated with an ear oximeter (Hewlett Packard) which had been calibrated per the manufacturer s instructions. Work was begun at zero work load and progressively increased by 25 W per minute until exercise was limited by exhaustion. If a definite (5 percent) or questionable (3 percent) decrease in arterial oxygen saturation occurred, or a questionable abnormality of any kind occurred, a steady state study was performed at 75 percent of the maximum work load achieved. An arterial line was placed in the radial or brachial artery, and blood gas levels were obtained at rest and after five minutes of exercise. In addition to the variables already noted, the P(A-a)2, the volume of dead space corrected for the breathing valve, the dead space fraction, and the alveolar ventilation were calculated. Ventilation was recorded at BTPS; carbon dioxide production and oxygen consumption were recorded at STPD. Symptoms were considered to be significant if they included a cough which interfered with daily activity and could not be attributed to other causes, or shortness of breath which resulted in a reduction of the patient s normal activities. Spirometric standards utilized were from Moms et al, 2 lung volume standards were from Boren et al, 2 and Dco standards were from Burrows et al. 4 Arbitrarily, pulmonary function studies were considered abnormal if the Table 1-Roentgenographw Stage 7, Ii, and III Sarcoidosis* Case No. Age/Sex Symptoms FEV,/ Pa2 FVC (mm Ratio DLco GE (Est Hg) TLC (% Pred) (%) (% Pred) BR % Sat) STAGE I 1. 3F Cough 2. 26M None 3. 2SF Cough; DOE 4. 23F Cough; DOE 5. 3M None 6. 34M None 7. 31M None 8. 24F DOE 9. 27F DOE STAGE II 1. 23M DOE M None M None M None M None M None M DOE 17. 3F DOE F None F Cough; DOE 2. 3F Cough; DOE 21. 2M None M DOE M DOE M None M Cough; DOE M Cough; DOE M None STAGE III M DOE M DOE 3. 43M None M DOE (128) (95) (7) (74) (85) (77) (96) (124) (88) (74) (89) (87) (98) (91) (11) (115) (14) (122) (14) (87) (63) (81) (75) (88) (7) (95) (7) (62) (111) (86) (86) (19) (79) (112) (55) (6) (68) (84) (94) (68) (48) (5) (84) (57) (63) (74) (81) (85) (12) (12) (76) (68) (68) (1) (12) (75) (96) (95) (85) (78) (81) (8) t t t t t t *DOE indicates dyspnea on exertion; RR, respiratory rate; and GE, gas exchange at beginning and ending of exercise. tthe mm Hg at beginning and ending of steady state exercise. CHEST/83/1/JAMJARV

4 TLC was less than 8 percent of predicted, the Dco was less than 8 percent of predicted, the FEV,/FVC ratio was less than 75 percent, or the arterial PaO2was less than 8 mm Hg. These values are somewhat arbitrary and may result in a small number of normal subjects to be labeled abnormal and a few abnormal individuals to be labeled normal. However, the dividing line between normal and abnormal is not distinct, and these cutoffs are consistent with recommendations in the literature. 6 On progressive incremental exercise testing, gas exchange was considered limiting if estimated arterial oxygen desaturation of greater than 5 percent occurred, an estimated saturation of less than 9 percent occurred, or ifa patient s Pa2 decreased by 1 mm Hg or more on the steady state exercise test. Lung mechanics were considered potentially limiting if respiratory rate of greater than 45 breaths per minute occurred and was associated with a small TV (less than 5 percent of VC). #{176} Ventilation was considered limiting if greater than 9 percent of ventilatory reserve (based on FEy, x 35) was used at maximum exercise. RESULTS Data on the 31 subjects symptomatology, routine resting pulmonary function studies, and exercise tests were tabulated and are summarized in Table 1. None of the patients demonstrated either cardiac arrhythmias or ECG changes. Graphs comparing heart rate to oxygen consumption, minute ventilation to oxygen consumption, oxygen consumption to carbon dioxide production, and minute ventilation versus respiratory rate and percent vital capacity were prepared on all patients. The relationship between heart rate and oxygen consumption was appropriate for the work load in all instances. None of the patients utilized greater than 9 percent of their ventilatory reserve. Table 1 summarizes the results for the stage I patients. Five of nine stage I patients were symptomatic. Three stage I patients had a decreased TLC, and four had an abnormally low DCO. The exercise studies in this group of patients demonstrated a pulmonary limitation in only one patient. This patient was limited by both arterial desaturation and an elevated respiratory rate associated with a small TV. This patient was asymptomatic, and the only abnormality fbund on routine pulmonary function study was an abnormally low DCO. Table 1 then summarizes the results of the 18 roentgenographic stage II patients. Nine of the 18 stage II patients had significant symptoms. Routine pulmonary function abnormalities were noted in 13 of the 18 patients and included five with an abnormal resting PaO,, six with a decreased TLC, five with a decreased FEV,/FVC ratio, and nine with a decreased DCO. Eight of the nine symptomatic patients had at least one abnormality on routine pulmonary function studies. Nine stage II patients demonstrated pulmonary abnormalities which were potentially limiting on their progressive incremental exercise tests. These included two patients with significant gas exchange abnormalities alone, three patients with significant gas exchange abnormalities and an abnormal ventilatory pattern, and four patients with an abnormal ventilatory pattern alone. The abnormalities were noted in six of the nine symptomatic patients and three of the nine asymptomatic patients. Four of five patients with significant gas exchange abnormalities had symptoms while one was completely asymptomatic. The remaining two asymptomatic patients had a markedly elevated respiratory rate in association with a small TV as a sole abnormality on exercise testing. Interestingly, four of five patients with an abnormality in gas exchange had an abnormally low Dco. The only exception to this finding was one patient with a DCO of 81 percent of predicted who demonstrated significant arterial desaturation from 97 percent to 93 percent by ear oximetry, and arterial oxygenation drop from a PaO, of 88 to a PaO, of 72 on steady state testing. Finally, Table 1 summarizes the results for the stage III patients. Three of the four patients were symptomatic. The same three patients demonstrated abnormalities in routine pulmonary function studies. The one asymptomatic individual had a normal pulmonary function study and a completely normal incremental and steady state exercise test. Two patients had potential pulmonary limitation on progressive incremental exercise testing, and both demonstrated a reduced Dco. When all patients are considered, 28 percent (4/14) of the asymptomatic patients and 46 percent (8/17) of the symptomatic patients had a potential pulmonary limitation with progressive incremental exercise testing. Of the symptomatic patients, 54 percent (9/17) did not have a ventilatory abnormality which could be identified as a cause of their dyspnea, or a limiting factor in their ability to perlbrm exercise. Table 2 demonstrates the relationship between the presence of symptoms, functional abnormalities, and the frequency of an abnormal exercise test. It can be seen that the presence of symptoms in association with multiple pulmonary function test abnormalities is highly indicative of abnormal findings with exercise. In addition, it should be noted that the absence of pulmonary function abnormalities on routine testing would seem to be a strong indicator that normal findings with exercise would be recorded. The one abnormality recorded in a patient with normal functions was a stage II individual who was also asymptomatic. The exercise abnormality was limited to a rapid ventilatory rate with a small TV at maximum exercise. Of interest is the fact that all six patients with desaturation below 9 percent had an abnormal diffusing capacity. The additional patient with abnormal gas exchange, who did not drop his PaO, below 7 with exercise, had a borderline diffusing capacity of 81 percent of predicted. While a normal Dco was predictive of a normal exercise test, an abnormal DCO was not specific. Seven patients with an abnormal Dco had a 78 ExercIse Testing In Pulmonary Sarcoldosls (Matthews, Hooper)

5 normal exercise study. 6 DISCUSSION Corticosteroids have become accepted therapy for significant pulmonary sarcoidosis despite the absence of a long-term randomized clinical trial of steroid treatment. - Unfortunately, the definition of significant pulmonary sarcoidosis is hard to establish. The association of symptoms, resting pulmonary function abnormalities, and parenchymal infiltrates is generally accepted as significant disease. Patients demonstrating only one or two of these factors can present confusing clinical pictures. The recommendations for therapy often are conflicting. Mitchell and Scadding4 have strongly recommended that only patients with significant pulmonary symptoms should be subjected to the dangers of corticosteroid therapy. Israel et al 1 and Young et al 8 have postulated that the natural history of sarcoidosis is unaltered by steroid therapy. If one accepts these premises, then patients with symptoms unrelated to physiologic ventilatory abnormalities would not benefit from corticosteroid therapy. The dangers of corticosteroid therapy dictate caution in prescribing them in situations where a clear-cut benefit cannot be demonstrated. Some examples will emphasize how exercise testing aided in the decisionmaking process in controversial situations. CASE REPORTS 5 E 4 3 C., > 2 I VE (BTPS)(I/min) I#{212}O VE (BTPS) (I/mln) NORMAL CONTROLS- PATIENT 7s PATIENT 8-. PATIENT 31 Patient 4 is a 23-year-old black woman referred for evaluation of shortness of breath and hilar adenopathy (stage I). Evaluation was normal or negative except for a lymph node biopsy specimen on mediastinoscopy which revealed noncaseating granulomata. Pulmonary function studies showed an FVC of 2.87 L (73 percent of predicted, an FEy, of 2.26 L (65 percent of predicted), a TIC of 4.47 L (77 percent of predicted), and a Dco of 17 mi/mm/mm Hg (68 percent of predicted). Exercise testing was entirely normal, and she was not treated with steroids. Patient 7 is a 31-year-old asymptomatic black man referred for evaluation of hilar adenopathy (stage I). Evaluation was normal or negative except for noncaseating granulomata on transbronchial lung biopsy specimen, and a Dco of 19.6 mi/mm/mm Hg (75 percent of predicted). Incremental exercise testing (Fig 1 and 2) revealed a decrease in estimated arterial saturation from 97 percent to 92 percent. He also demonstrated a respiratory rate of 53 breaths per minute with a TV of 16 ml, which was 47 percent of VC. freatment with steroids resulted in a marked improvement in exercise tolerance which, in retrospect, he attributed to his lack of conditioning. Patient 8 is a 24-year-old black woman referred for evaluation of shortness of breath, bilateral hilar adenopathy, and abnormal pulmonary function studies. Evaluation was normal or negative except for a transbronchial lung biopsy specimen which showed noncaseating granulomata, an FVC of 2.49 L (74 percent of predicted), a TIC of 3.56 L (74 percent of predicted), and a Dco of 27 mi/mm/mm Hg (96 percent of predicted). Exercise testing results were entirely within normal limits (Fig 1 and 2), and recommendation was that steroid therapy was not indicated. Her attending physician did not agree with our evaluation and started her on prednisone 6 mg/day. She became cushingoid but did not have any change in symptoms or in her pulmonary function studies. Patient 1 is a 23-year-old black man referred for evaluation of hilar Ficuai 2. Graph plotting respiratory frequency (f,.) versus minute ventilation (VE) and tidal volume as percent vital capacity (VC) versus V5. The asterisks depict expected curves. A normal control and three representative patients are shown. The only definitely abnormal curve is patient 7, who developed an f8 of 53 while his tidal volume was 47 percent of his VC. adenopathy, interstitial lung disease (stage II), and shortness of breath. Evaluation was normal or negative except for a traitsbronchial lung biopsy specimen which showed noncaseating granulomata, a resting PaO, of77mm Hg, and normal FVC, FEy,, TIC, and Dco. Exercise testing was entirely normal. Treatment with steroids was not recommended. Patient 16 is a 45-year-old black man referred for evaluation of shortness of breath, hilar adenopathy, and interstitial lung disease. A diagnosis of sarcoidosis was made on the basis of mediastinoscopy. Pulmonary function studies revealed an FVC of 4.8 L (97 percent of predicted), an FEy, of 2.57 L (61 percent of predicted), and FEVJFVC ratio of 53 percent. The TIC was 8.96 L (122 percent of predicted), and Dco was 16.1 mi/mm/mm Hg (55 percent of predicted). No improvement was noted with bronchodilators, and his pulmonary function abnormalities were thought to be secondary to chronic obstructive pulmonary disease secondary to his use of tobacco. Incremental exercise testing showed a decrease in his estimated arterial saturation from 86 percent to 75 percent at 75 W work. A steady state exercise test was performed at 5 W work. Resting arterial PaO, was 6mm Hg and decreased to 39 mm Hg in the fifth minute of exercise. Resting VS/rD was 45 percent and decreased to 3 percent with exercise. This response was interpreted as more typical of what would be expected from sarcoidosis than from emphysema. Steroid therapy was instituted. The patient had a marked symptomatic improvement and return of his FEy, to normal. CHEST/83/1/JANUARY,

6 Patient 19 is a 34-year-old black woman referred for evaluation of shortness of breath, bilateral hilar adenopathy, and interstitial lung disease. Evaluation was normal or negative except for a traitsbronchial lung biopsy specimen which showed noncaseating granulomata, an FVC of 2.14 L (56 percent of predicted), an FEy, of 1.86 L (57 percent of predicted), atlc of3.78 (63 percent of predicted), and a Dco of 18.1 mi/miss/mm Hg (84 percent of predicted). Exercise testing was normal except for a respiratory rate of 5 beats per minute with a TV of 7 ml (32.7 percent VC). Corticosteroid therapy was recommended. She responded symptomatically. A repeat exercise test was planned, but she did not return for the study. Patient 31 is a 36-year-old black man admitted for evaluation of shortness of breath and bilateral interstitial lung infiltrates (stage 3). A diagnosis of sarcoidosis was made by open lung biopsy specimen which revealed noncaseating granulomata and no other abnormalities. The FVC was 4.37 L (76 percent of predicted); TLC was 6.8 L (76 percent of predicted); and Dco was 26.2 mm/mm/mm Hg (76 percent of predicted). Results of exercise test (Fig 1 and 2) were totally normal. He did not receive corticostemid therapy. He continued to have minimal dyspnea on exertion but was not significantly limited in any way. Follow-up pulmonary function studies three years later revealed an FVC of 4.28 L (76 percent of predicted), a TIC of 6.11 L (81 percent of predicted), and a Dco of 31.5 mi/mm/mm Hg (96 percent of predicted). COMMENTS Our study (Table 2) clearly demonstrates the lack of correlation between symptoms, resting pulmonary function studies, and physiologic limitation of exercise. This is consistent for all three stages of roentgenographic disease. This discrepancy between symptoms and the results of progressive incremental exercise testing is disturbing but not necessarily surprising. Symptoms arise from an individual s subjective interpretation of somatic events and are tempered by personality and the reason for seeking medical assistance. Most certainly, the disease process may be related to and producing the symptoms in our patients without a physiologic limitation to exercise; however, the subjective nature of symptoms leads one to question how important the symptoms are as indicators of the extent of disease and the need for therapy. Progressive incremental exercise testing, when performed with an ear oximeter, can provide an easy noninvasive technique to evaluate the response to exercise. The technique can demonstrate physiologic Table 2-Relationship Between Symptoms, Pulmonary Functions, and Exercise Testing Clinical Exercise Test Normal Abnormal Total No symptoms Normal PFTs Symptoms Normal PF Fs 2 2 No symptoms Abnormal PFrs Symptoms Abnormal PVrs Total mechanisms that are utilized maximally or respond abnormally with exercise. Some variables are of individual importance, such as heart rate, maximum minute ventilation, and the status of oxygenation. Others require evaluation of the interrelationships between variables that are important with exercise testing and these include the heart rate versus the oxygen consumption, and the relationship between the minute ventilation as compared to the TV, respiratory rate, oxygen consumption, and the carbon dioxide production. In all our patients, the majority of these variables and relationships were surprisingly normal. Abnormal findings were limited to arterial oxygen desaturation and high respiratory rates associated with small TV. These abnormalities occurred in 12 of 31 patients tested. Primarily, patients in whom it was felt that the exercise test would be of benefit in their evaluation were studied, since the number of studies performed was limited. Thus, patients with stage II or stage III disease, and symptomatic patients with stage I disease were preferentially studied, while asymptomatic patients with stage I disease or patients with obvious systemic disease were studied only if practical. It is likely that the frequency of abnormal exercise studies would be lower if all individuals seen with sarcoid were studied. Ideally, there are three indications lbr exercise testing in patients with sarcoidosis: (1) to document the physiology of exercise symptoms; (2) to document the relationship between abnormal routine function tests and physiologic response to exercise; and (3) to quantify exercise tolerance while following therapy or evaluating disability. Exercise testing is too expensive and too time-consuming to be employed as a routine procedure for all patients with sarcoidosis. To use the procedure in a cost-effective manner, criteria are required to identify people who are likely to benefit from the data generated. This experience has been very helpful in this regard. First, patients who have normal pulmonary function studies generally have normal exercise tests regardless of the roentgenographic stage of the disease. Only one abnormality was noted in II such patients (patient 12 with a high respiratory rate). Secondly, three or more abnormalities on routine pulmonary function studies were invariably associated with a pulmonary limitation to exercise. On the other hand, significant information not otherwise available can be obtained by exercise testing of patients with one or two abnormalities on pulmonary function studies. Physiologic pulmonary limitations cannot be predicted in this group, regardless of the presence or absence of symptoms and regardless of the roentgenographic stage of disease. Last, and probably most important, the variable of clinical interest to most physicians, severe hypoxemia with exercise (PaO, of less than 7 mm Hg), occurred 8 Exercise Testing In Pulmonary Sarcoldosls (Matthews, Hooper)

7 almost exclusively in patients who had an abnormal single breath diffusion capacity. The only exception was one patient with a borderline low Dco. Potential pulmonary limitations occurring during exercise were restricted to (1) increased frequency of the respiratory rate with maximum exercise in association with small TV, and (2) hypoxemia. While the significance of the latter is obvious, the importance of high respiratory rates during exercise is really unknown. Normal values of maximum respiratory rates and TV are poorly defined and will require more studies to define accurately their limits.9 Our criteria, if anything, are probably over-stringent. We (unreported observations) and others9 #{176}have observed that in the absence of underlying lung disease, a respiratory rate of 4 or greater is uncommon until the TV exceeds 5 percent of the vital capacity. Using a respiratory rate of 4, three additional patients (two symptomatic and one asymptomatic) would have had an abnormal exercise test. The results of our study would not be altered whether these patients were called normal or abnormal, but our approach to these three individual patients was modified by our uncertainty in interpreting their respiratory pattern. The implication for the clinician who sees a limited number of patients is that exercise testing can be performed with the monitoring of just a limited number of variables, The essential variables would have to include work load, heart rate, ECG, oxygenation, and respiratory rate. With the use of an ear oximeter, the testing can be performed noninvasively in an easy, relatively rapid manner. The value of progressive incremental exercise testing in patients with pulmonary sarcoidosis is demonstrated by our experience utilizing this procedure in these patients. Four major conclusions can be drawn. First, progressive incremental exercise testing is useful in the evaluation of patients with confusing clinical pictures. Second, clinical factors can be used to identify patients who will benefit from employing the procedure, therefore improving its cost effectiveness. Third, the relationship between dyspnea and exercise limitations is not always a close one. And finally, adequate exercise studies can be performed noninvasively on patients with sarcoidosis while monitoring a limited number of physiologic variables. ACKNOWLEDGMENT: The authors gratefully acknowledge the secretarial assistance of Mrs. Carolyn Norton in the preparation of the manuscript. REFERENCES 1 Smellie H, Hoyle C. The natural history of pulmonary sarcoidosis. Quart J Med 196; 29: Sones M, Israel HL. Course and prognosis of sarcoidosis. Am J Med 196; 29: Johns CJ, Zachary JB, Ball WC Jr. A ten year study of corticosteroid treatment of pulmonary sarcoidosis. J Hopkins Med J 1974; 134: Mitchell DN, Scadding JG. Sarcoidosis. Am Rev Respir Dis 1974; 11: DeRemee BA. The present status of treatment of pulmonary sarcoidosis: a house divided. Chest 1977; 71: Sharma OP, Colp C, Williams MH. Course of pulmonary sarcoidosis with and without corticosteroid therapy as determined by pulmonary function studies. Am J Med 1966; 41: Young RL, Loudon RE, Krumholz LE, et al. Pulmonary sarcoidosis: 1. Pathophysiologic correlations. Am Rev Respir Dis 1968; 97: Huang CT, Heurich AE, Rosen Y, et al. Pulmonary sarcoidosis: roentgenographic, functional, and pathologic correlations. Respiration 1979; 37: Wasserman K, Whipp BJ. Exercise physiology in health and disease. Am Rev Respir Dis 1975; 122: Jones NL, Campbell EJM, Edwards RHT, Robertson DG. Clinical exercise testing. Philadelphia: WB Saunders Co, 1975: chap 2 11 Clark TJH, Freedman S. Campbell EJM, et al. The ventilatory capacity of patients with chronic airway obstruction. Clin Sd 1969; 36: Morris JF, Koski A, Johnson LC. Spimmetric standards for healthy nonsmoking adults. Am Rev Respir Dis 1971; 13: Boren HG, Kory RC, Syner JC. The Veterans Administration- Army cooperative study of pulmonary function: II. The lung volume and its subdivision in normal men. Am J Med 1966; 41: Burrows B, Kasik JE, et al. Clinical usefulness of the single breath pulmonary diffusing capacity test. Am Rev Respir Dis 1961; 84: Bates DV, Macklem PT, Christie RV. Respiratory function in disease, 2nd ed. Philadelphia: WB Saunders, 1971: Intermountain Thoracic Society. Clinical pulmonary function testing, 1st ed. Kanner RE, Morris AH, edo Israel HL, Fouts DW, Beggs BA. A controlled trial of prednisone treatment of sarcoidosis. Am Rev Respir Dis 1973; 17: Young RL, Harkleroad LE, Loudon RE, Weg JG. Pulmonary sarcoidosis: a prospective evaluation of glucocorticoid therapy. Ann Intern Med 197; 73:27-12 CHESTI83/1/JANUARy,