INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Exercise testing before lung (LT) or heart-lung transplantation (HLT) has generally been limited to a 6-min walk test with oximetry or a modified Bruce protocol treadmill test (1- 3). Data obtained from patients awaiting transplantation show 6-min walk distances from 50 to 600 m (1, 2, 4, 5). In all studies using 6-min walk tests, large standard deviations and differences among the underlying diseases were observed; moreover, the 6-min walk test has limitations with regard to standardization, fluctuations in motivation, and interpretation of results (6). The 6-min walk test also does not provide any information about the factors limiting exercise performance.

Cardiopulmonary exercise tests are performed routinely in our transplant center, except with patients who have contraindications to these tests and/or are unable to bicycle. To our knowledge, except for three small-scale studies (7), no data have been published about exercise performance in lung transplant recipients before and shortly after transplantation.

Cardiopulmonary exercise testing in recipients of lung and heart-lung transplants demonstrates significant improvement of exercise tolerance in individuals severely disabled by their underlying cardiopulmonary disease. However, exercise capacity as measured by maximum oxygen uptake and work rate remains limited in most recipients despite substantial improvement and often normalization of cardiopulmonary function (4, 7, 9).

The purpose of the present investigation was to evaluate cardiovascular and ventilatory responses during incremental exercise before and shortly after LT and HLT.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Subjects

Exercise studies were done on 103 patients (63 women and 40 men) who had undergone LT or HLT for different underlying diseases. Of these patients, 46 (mean age: 32 yr) had undergone single lung transplantation (SLT), 32 (mean age: 31 yr) bilateral lung transplantation (DLT), and 25 (mean age: 29 yr) HLT. All LT patients had end-stage respiratory diseases (idiopathic pulmonary fibrosis [IPF] 32% [SLT: 86%; DLT: 14%], emphysema 28% [SLT: 84%; DLT: 16%], cystic fibrosis [CF] 23% [DLT: 100%], graft failure after LT 3% [SLT: 100%], and miscellaneous other diseases [lymphangioleiomyomatosis, radiation fibrosis, pneumoconiosis, hypersensitivity pneumonitis, congenital CF] 14% [SLT: 28%, DLT: 72%]). In the HLT group, the underlying diseases were primary pulmonary hypertension (44%), congenital cardiac diseases with Eisenmenger\Qs syndrome (36%), IPF (8%), CF (4%), and miscellaneous other diseases (lung hypoplasia, pulmonary atresia) (8%).

All patients were studied both before (median: 170 d; range: 11 to 330 d) and within 3 mo after transplantation (median: 55 d; range: 7 to 211 d). All patients were clinically stable at the time of evaluation before transplantation. All posttransplant studies were conducted after recovery from surgery and at times thereafter, when patients were without clinical complications and were physically well. No training sessions were conducted before study. LT and HLT were done according to standard procedures. The telescoping technique, supported by wrapping with a pericardial pedicle, was used for bronchial anastomosis. All transplant recipients received conventional immunosuppression with cyclosporine (CyA)/tacrolimus (FK) (CyA: 52%/FK: 48%), azathioprine/mycophenolate acid (azathioprine: 79%/mycophenolate acid: 21%), and prednisone (< 0.5 mg/kg/d). No individual was treated with -blockers. None had evidence of complications of bronchial anastomosis at the time of study. The hemoglobin concentration was measured in each patient. Written informed consent was obtained from the patients; the study had been approved by the local ethics committee.

Hemodynamic Measurements

Before transplantation, left- and right-heart catheterization was performed. A flow-directed pulmonary artery catheter (Baxter Healthcare Corp.) was inserted, using the right femoral vein approach. With this system, mean pulmonary artery wedge (), mean pulmonary artery (), and mean right atrial pressures () were recorded with digital readout monitors (Sirecust 404; Siemens, Erlangen, Germany). Cardiac output (CO) was measured with the thermodilution technique (SP 1435 Cardiac Index Computer; Gould Statham). Similarly, a pigtail catheter was inserted, using the right femoral artery approach. Left-ventricular end-diastolic (LVEDV) and end-systolic volumes were determined using the area-length method; left ventricular ejection fraction (LVEF) was calculated according to the standard formula (14).

Pulmonary Function Testing

Pulmonary function tests included FVC, FEV₁, and single-breath carbon monoxide lung transfer factor (TLCO) (Body Test, Jaeger, Würzberg, Germany). Quality control procedures and predicted values (%predicted) were applied according to the standards of the European Coal and Steel Community (ECSC) 1993 (15). Blood gases were analyzed at rest and at maximum work rate, using arterialized earlobe blood (ABL500; Radiometer, Copenhagen, Denmark).

Cardiopulmonary Exercise Testing

An incrementally progressive, symptom-limited cardiopulmonary exercise test was performed on all individuals. An electrically-braked cycle ergometer was used (Ergotest, Jaeger, Germany). Heart rate (HR) and rhythm were monitored with an electrocardiograph. The patients were connected to the respirometer via a two-way, low-resistance y-mouthpiece and a pneumotachograph, and breathed room air. The expired air was collected continously in a Douglas bag. O₂ and CO₂ were analyzed every 15 s (Ergopneumotest EOS, Jaeger, Germany). The Ergopneumotest instrument provided recordings of minute ventilation (E), respiratory frequency (f), oxygen consumption (VO₂), carbon dioxide output, and respiratory exchange ratio. After 5 min of adaptation to the mouthpiece, the workload was increased in 20-W steps every 3 min, to the point of exhaustion (inability to maintain a constant speed, and/or to reach 90% of predicted maximal HR and/or intolerable dyspnea) (16). When 90% of the predicted maximum HR was reached, exercise was stopped. Seven patients before and five patients after transplantation reached a an HR of  90% of the predicted maximum HR. Peak workload (P) was defined as the highest work level reached and maintained for at least 1 min. Similarly, peak HR (HR-E), peak oxygen uptake (O₂max) and peak ventilation (VE) were defined as the highest levels reached during the test. Simultaneously, physiologic dead space/tidal volume (Vdf/VT) was derived from the measured Pa_CO2. The alveolar to arterial oxygen gradient (A-a)PO₂ was determined with the simplified alveolar gas equation based on arterial blood gas analysis. The anaerobic threshold (AT) was determined by the V-slope method, and the ventilatory equivalent for carbon dioxide (VE/VCO₂) was measured at the anaerobic threshold. Standard predicted values for exercise testing were used (16).

Statistical Analysis

All data are presented as mean ± SD. A two sample t test was used to compare the data obtained for the different transplant groups (with Bonferroni's posttest correction where appropiate). Within-group analysis was done with a Mann-Whitney test (with Bonferroni's posttest correction where appropiate). A one-way analysis of variance (ANOVA) of the postexposure values was performed, using the preexposure values as covariates. A value of p < 0.05 was regarded as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Clinical Characteristics

The clinical characteristics of the transplant recipients are shown in Table 1. Patients undergoing SLT were significantly older and had a higher weight than patients undergoing DLT or HLT. Blood hemoglobin concentrations after transplantation were reduced in all groups (Hb: 10.9 ± 1.7 g/dl; range: 7.3 to 15.3 g/dl). The elapsed time from lung transplantation to the maximum exercise study was equivalent in the three groups (55 ± 9 d).

Hemodynamic Measurements

Before transplantation, as shown in Table 1, there were significant differences between the LT and HLT groups, particularly with regard to and PAR. Additionally, was slightly higher in the HLT group, whereas cardiac index (CI) was highest in the DLT group. There was no significant reduction of left ventriclular function in any of the groups.

Pulmonary Function Testing

Pulmonary function and blood gas data are shown in Tables 2 and 3. Before transplantation, spirometric results and TLCO were comparable in patients scheduled for SLT and DLT. In patients scheduled for HLT, we observed significantly higher values for the obtained lung function parameters.

As expected after transplantation, values of FEV₁ and TLCO in the SLT group were lower than in the two other groups. In all transplant recipients, we found a mild to moderate restrictive ventilatory defect, with reduced FVC and TLC. Patients undergoing DLT had TLCO levels that were close to normal with significantly higher Pa_O2 at rest and during exercise than did patients having SLT.

Cardiopulmonary Exercise Testing

Cardiovascular and ventilatory data from graded maximum exercise testing of the transplant recipients are shown in Tables 4 and 5. Before transplantation, we observed a similar limitation in all three groups. P and VO₂ were decreased; VO₂ had a mean value of 10.4 ml O₂/min/kg. The O₂ pulse, reflecting the capacity of the heart to deliver O₂ per heartbeat, was reduced to a similar extent in all groups as compared with healthy subjects (16). Before transplantation the HLT group had a lower HR at rest, the DLT group showed a higher AT than the SLT or HLT group. VE in liters per minute, and VT, were lower in patients scheduled for SLT and DLT than in those scheduled for HLT.

After transplantation, all three groups reached similar workloads at peak exercise (30% predicted), that were substantially lower than those achieved by normal control subjects. Also, the average VO₂ was significantly below that observed in controls, with patients undergoing DLT reaching significantly higher values than those undergoing SLT or HLT. These limitations did not appear to be due to HR (HR reserve: 57 beats/min), ventilation (breathing reserve: 41 L/min), or Pa_O2 during exercise (84 mm Hg), although the patients surpassed the AT (respiratory quotient [RQ] = 1.12). There were no significant differences between patients treated with CyA and patients treated with FK (e.g., the work rate in the CyA group was 29%, versus 30% in the FK group, and peak oxygen uptake in the CyA group was 13.2 ml/min/kg, versus 13.2 ml/ min/kg in the FK group).

Comparison of Pre- and Posttransplant Exercise Test Data

All recipients showed substantial improvement in workload and peak oxygen consumption. Also, slightly higher values of peak VE and AT were noted shortly after transplantation, without the differences reaching statistical significance. Significant improvements in O₂ pulse and VT were seen. Vdf/VT and (A-a)PO₂ during peak exercise showed the most prominent improvement following transplantation in all three groups (Table 6).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The goal of this study was to compare results of cardiopulmonary exercise testing before and after SLT, DLT, and HLT. As expected, exercise responses before transplantation were markedly limited in all three groups. All transplant recipients experienced improvement in their exercise tolerance after transplantation, with VO₂ averaging 43.9%, 40.2%, and 38.4%, predicted in patiens undergoing SLT, DLT, and HLT, respectively. Although patients with pulmonary fibrosis and chronic obstructive lung disease showed a reduced breathing reserve before transplantation, no evidence of ventilatory limitation was found. Furthermore, in all groups, AT was reduced both before and after transplantation. Following transplantation, all three transplant groups reached a similar maximum exercise peak (30% predicted), with the most pronounced changes seen in Vdf/VT and (A-a)PO₂ during peak exercise.

The pretransplant cardiopulmonary exercise test data showed severe exercise intolerance related to ventilatory and/ or circulatory limitations. Irrespective of patients' underlying disease, we did not find significant differences in cardiopulmonary exercise parameters, with the exception of a higher ventilatory reserve and VT in patients with pulmonary hypertension. Theodore and colleagues (9) investigated 10 patients with pulmonary hypertension, finding an overall aerobic capacity of 24% predicted. Ross and coworkers (8) examined the exercise capacity of another group of transplant candidates with a variety of pulmonary diseases. The aerobic capacity was dramatically reduced (29% predicted). In accord with our results (peak VO₂: 34% predicted), Ross and coworkers (8) observed not only ventilatory abnormalities but also evidence for circulatory impairment involving the heart, pulmonary vasculature, and systemic circulation in some of their patients. The reduced O₂ pulse and AT in our study group may have been related to a limitation of the cardiocirculatory system from pulmonary hypertension, musculoskeletal changes, or deconditioning. Although ventilation is excessive during exercise in patients before HLT, a sufficient ventilatory reserve is still available. Before transplantation, exercise limitation may be partly caused by hypoxemia and its effects on the circulation and/or exercising skeletal muscle. On the other hand, patients with relatively mild hypoxemia and pulmonary hypertension show the same degree of exercise limitation as their counterparts with severe hypoxemia. This suggests that in general, circulatory factors are more critical to exercise capacity than is arterial oxygenation (9).

The peak exercise capacities observed after transplantation in our study are similar to those previously reported for patients undergoing SLT (7, 8, 10, 17), DLT (7, 8, 10, 12, 20), and HLT (9, 13, 21). The extent of exercise limitation observed in patients undergoing SLT, DLT, and HLT is similar, and the pattern of limitation is somewhat stereotyped. This observation makes a primary disturbance in pulmonary or cardiac function rather unlikely, and may be indicative of peripheral deficiencies in transport and utilization of oxygen. Although the VO₂ levels achieved by the transplant recipients were substantially below those observed in controls, they allow at least moderate levels of work and exercise. Despite blunted chronotropic and inotropic responses during exercise (22), VO₂, reported in the majority of patients undergoing HLT in the same series, was better than in patients undergoing SLT. The immediate improvement in exercise after HLT is primarily the result of an improved circulation, as indicated by the differences in O₂ pulse. Although O₂ pulse is low in groups undergoing all three types of transplantation discussed here, this does not necessarily point to a primary disturbance in cardiac function, and may be more indicative of peripheral deficiencies in transport and utilization of oxygen. However, a contribution of exercise-induced right ventricular dysfunction cannot be excluded. The preferential perfusion of the transplanted lung after SLT implies that the upper limits of vascular compliance are reached when a single lung must accept the greater portion of the CO (20, 26). Miyoshi and associates (11) have proposed that a limitation in pulmonary vascular capacity after SLT may restrict CO, and thus restrict maximum oxygen delivery to exercising muscles.

There was no evidence of ventilatory limitation in either the LT or HLT group after transplantation. Therefore, peripheral factors may contribute substantially to exercise intolerance in transplant patients. Putative mechanisms for this include deconditioning, chronic anemia, loss of muscle mass, muscle atrophy, and interference by various drugs (11, 12, 17, 23, 24, 27). In particular, CyA may have a negative influence on exercise test results by its effects on muscle blood flow. Shortly after transplantation, there was no difference in cardiopulmonary exercise parameters between our patients who received CyA and those who were treated with FK. The muscle deconditioning and atrophy that are present before transplantation persist to some degree, at least in the early posttransplant period.

The most noteworthy findings in our study were the significant decrease in Vdf/VT and (A-a)PO₂ during exercise in all groups after transplantation. Exercise gas exchange, although somewhat better in DLT and HLT than in SLT patients, was more than adequate to meet the delivery requirements at the work levels achieved. The higher (A-a)PO₂ in SLT patients was apparently due to a ventilation-perfusion (VA/Q) imbalance and/or to a shortened red blood cell transit time through the reduced vascular bed in SLT patients (11). In agreement with the findings of Levy and colleagues (10), we found that Vdf/VT remained mildly elevated despite significant improvements in gas exchange. We suggest that VA/Q mismatch exists in both allograft and native lungs. Thus, reduction of the V/Q ratio in allograft and increase of the VA/Q ratio in native lungs may generate areas of shunt and wasted ventilation (31). This may explain the observed gas-exchange abnormalities that are more frequent after SLT than after DLT (12, 17). VA/Q mismatch seen in patients undergoing DLT and HLT, however, suggests that the pulmonary circulation or ventilation of transplanted lungs has subtle abnormalities caused by denervation of the lung, long-term sequelae of ischemia-reperfusion injury, prior episodes of acute rejection, or drugs (7, 10, 11, 32). This could lead to an increase in both anatomic and, through altered VA/Q matching, physiologic dead space. Whether the increased Vdf/VT following transplantation represents an abnormality of ventilation or of the pulmonary circulation cannot be determined from the available data. Vdf/VT measurements are also influenced by exercise breathing pattern, with the result that an abnormally high Vdf/VT in patients who have undergone SLT may result in part from an increase in respiratory frequency. On the other hand, we could demonstrate a similar breathing frequency in patients who had undergone SLT and HLT and who had significantly different values for Vdf/VT. Additionally, we did not see any difference in the partial pressure of carbon dioxide at maximum exercise in our three groups, suggesting that the increased Vdf/VT in LT and HLT patients is caused in part by a VA/Q mismatch.

In summary, our pretransplant cardiopulmonary exercise test data showed a severely decreased aerobic capacity. Abnormalities in ventilation played a more important role in exercise limitation in candidates for SLT or DLT than in candidates for HLT, although some displayed evidence of circulatory impairment involving the heart, pulmonary vasculature, and systemic circulation as well. This study shows that after transplantation, the exercise limitation observed in patients having SLT, DLT, and HLT is similar. The data suggest that the limitation is principally attributable to peripheral factors (peripheral circulation or peripheral muscle). No evidence of ventilatory limitation was found in either the LT or HLT group. Although inefficiencies in VA/Q matching, especially in SLT patients, are conspicious, their magnitude should not be limiting in the absence of a relevant hypoxemia.