INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the early experimental and clinical experience with single-lung transplantation (SLT) for emphysema, concern was expressed that the native hyperinflated lung may displace the mediastinum toward the transplanted side and cause crowding and dysfunction of the graft (1, 2). These initial concerns did not materialize and, over the last decade, SLT has become an established therapeutic option for the treatment of end-stage emphysema (3). Yet, there have been reports of recipients showing progressive hyperinflation of the native lung associated with severe hemodynamic and/or respiratory compromise who required differential lung ventilation (6), volume reduction surgery (10), bullectomy (13), lobectomy (14, 15), or pneumonectomy (16) of the native lung, or retransplantation (15).

Although the factors predisposing to this complication are still debated (15), recent studies have suggested that, in most patients, the presence of the native hyperinflated lung is not disadvantageous to the function of the graft (17, 18). A similar conclusion was reached by Margulies and colleagues (19) who, using a canine model of unilateral papain-induced emphysema, showed that the volume of the nonemphysematous lung did not fall significantly after induction of emphysema in the contralateral lung, but these results cannot be extrapolated without caution to humans because of interspecies differences in rib cage and mediastinal compliance. So it appears that the mechanisms of interaction between the native and the transplanted lung and the functional consequences of the mediastinal shift toward the graft are not fully understood. To gain new insight into this question, we used computed tomography (CT) to measure the separate volumes of the native and transplanted lungs in a group of patients with SLT for emphysema and we determined (1) how these volumes change over time after transplantation, and (2) how the presence of the native hyperinflated lung influences the volume of the graft.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Between 1990 and 1997, SLT for emphysema was performed 14 times in 14 patients at our institution. Of these, 10 patients were alive at the time of the present investigation and form the basis of this report. Details of the patients and their donors are given in Table 1. At the time of surgery they were (mean ± SE) 52.7 ± 2.2 yr of age and were severely obstructed and hyperinflated; compared with predicted normal values (20), their plethysmographic TLC and FRC averaged 160 ± 9% and 254 ± 14%, respectively, and FEV₁ was 20 ± 2%. They all had emphysema as indicated by severe hyperinflation, low diffusing capacity, and extended areas of low attenuation on chest CT. Eight patients received a left SLT and two patients a right SLT. They were all receiving triple immunosuppressive therapy with ciclosporine, azathioprine, and methylprednisolone. Two studies were performed.

Study 1

We first analyzed CT data that had been obtained before and after SLT in nine patients (Patients 1 to 9 in Table 1). Patient 10 was not included in this study because her follow-up was less than 1 yr. The acquisitions were made during breathholding at TLC, the subjects lying supine with arms at the sides. The subjects were asked to keep the glottis closed and to relax the respiratory muscles. High resolution CT scans were performed on a Somatom Plus 4A (Siemens, Erlangen, Germany) set to the following conditions: 1-mm slice thickness with 10-mm interval, 140 kV, 171 mA, scanning time 1 s. Acquisitions were performed from the lung apex to the lung base. The Pulmo CT option from Siemens (21) was used to trace the lung contours on each scan and measure lung area (Figure 1); TLC was then calculated using values of lung area and slice interval. In this study, we included in the data analysis the scan that had been obtained in the pretransplant assessment and one scan performed during each of the first three postoperative years as part of the patients' routine follow-up. At the time of the postoperative scans, the patients were in a stable clinical condition with no evidence of infection or acute rejection.

View larger version (119K): [in this window] [in a new window]   Figure 1.   Axial CT slice of the chest illustrating the tracing of lung contours in one representative patient with a left SLT. The arrow indicates location of the anterior mediastinal line. Note the shift of the mediastinum toward the graft; lines A and B were used to quantify the degree of mediastinal shift (see text for details).

Study 2

We measured on a single occasion the TLC and FRC of the native and transplanted lungs in the 10 patients of the study. These values were compared with those obtained in 10 nonsmoking normal subjects matched for sex, age, height, and weight with the recipients (Table 1). In each subject, the scans were obtained using the same equipment as in Study 1, but using the spiral mode (10-mm slice thickness, 12 mm/s table speed, 140 Kv, 200 mA, 1-s rotation time); one spiral acquisition was thus performed at TLC and at FRC from the lung apex to the lung base. The duration of the acquisition varied from 15 to 25 s according to the height of the subject and the volume studied. Images were reconstructed every 10 mm and lung volume was calculated as described above. At each volume, we determined the degree of mediastinal shift by measuring the angle between the midsagittal line (line A) and the line (line B) joining the anterior aspect of the vertebral body and the anterior mediastinal line (Figure 1). In normal subjects lines A and B are almost superimposed, but in the patients the anterior mediastinal line was displaced toward the graft, with the angle between lines A and B increasing in magnitude with the degree of mediastinal shift (Figure 1).

Statistical analysis was performed using a paired t test and ANOVA when appropriate. All data are reported as mean ± SE. The level of statistical significance was taken as p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study 1

Data were available for all patients but one before surgery. On average, the preoperative scan was obtained 326 ± 50 d before surgery, and postoperative scans were obtained 239 ± 52, 588 ± 89, and 932 ± 130 d after surgery; data were available for all patients during the first and second postoperative year and for eight patients during the third postoperative year. At the time of the first and second postoperative scans, no patient had developed a bronchiolitis obliterans syndrome (BOS), but at the time of the third scan, one patient was in BOS Stage 1 and one patient in BOS Stage 2 (22). It can be seen in Figure 2 that SLT markedly reduced TLC on the transplanted side from 3.42 ± 0.28 L before to 2.06 ± 0.12 L after surgery (p < 0.0002), but it did not affect TLC on the native side (3.57 ± 0.28 L before versus 3.72 ± 0.26 after SLT; NS). The volume of the native lung and of the graft did not change significantly from the first to the third postoperative evaluation.

View larger version (13K): [in this window] [in a new window]   Figure 2.   Average (± SE) values of TLC on the native (closed circles) and the transplanted (open circles) side before and after SLT for emphysema. Transplantation was performed at Day 0 (vertical dashed line). Values were available in eight patients before surgery, nine patients for the first two postoperative measurements, and eight patients for the third postoperative measurement.

Study 2

At the time of this study (on average, 697 ± 197 d after SLT) the patients were all in a clinically stable state, although one patient had developed a BOS Stage 1. Results of pulmonary function tests are given in Table 2. All patients had a displacement of the mediastinum toward the graft (Figure 1); the angle between lines A and B averaged 20.4 degrees at TLC and 24.5 degrees at FRC. The TLC of the graft was markedly reduced when compared with that of the ipsilateral lung in the control subjects, as shown in Figure 3; for the 10 patients studied, it averaged 79 ± 5% of the control value (p < 0.005). On the other hand, the FRC of the graft was normal, averaging 100 ± 9% of the FRC of the control lung. The figure also shows that, as expected, the FRC and TLC of the native emphysematous lung were much larger than that of either the graft or the ipsilateral lung in the control subjects (p < 0.002).

View larger version (28K): [in this window] [in a new window]   Figure 3.   Average (± SE) values of TLC and FRC of the native lung and of the graft (Tx) in 10 patients with SLT for emphysema (open bars). Values are compared with those obtained on the ipsilateral side in 10 matched normal control subjects (hatched bars). Statistical differences are shown only for adjacent bars: **p < 0.005; ***p < 0.0005.

In this study, the control subjects were matched with the recipients who were slightly smaller than the donors (Table 1). As a result, predicted lung volumes were larger in the donors than in the control subjects and the recipients, but the differences were small and statistically nonsignificant, i.e., on average, 0.54 L for TLC and 0.05 L for FRC. Consequently, matching the control subjects with the donors rather than the recipients would not have changed the conclusion of the study.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although no patient in this study developed hemodynamic or respiratory dysfunction associated with hyperinflation of the native lung, all of them showed a displacement of the mediastinum toward the transplanted side (Figure 1). Yet, the volume of the native lung remained very stable after compared with before SLT, and the volumes of both the native and the transplanted lungs at full inflation did not change significantly over a 3-yr period. In this group of SLT recipients, we could thus not demonstrate progressive hyperinflation of the native lung with crowding of the graft. This is in agreement with the data of Cheriyan and colleagues (23) obtained before and up to 6 mo after SLT in seven patients with chronic obstructive pulmonary disease. Using planimetry on chest radiographs, they observed that although the TLC of the native lung slightly increased after compared with before surgery, the volume of the two lungs did not change significantly in the postoperative period (23).

In keeping with these observations, we found that the volume of the native lung was much larger than that of the graft. This finding could be anticipated since SLT does not reduce hyperinflation on the native side. Therefore, the following discussion will focus on the comparison between the volume of the graft and the volume of the ipsilateral lung in the control subjects.

We observed that the TLC of the graft was significantly reduced compared with that of the control lung. This is unlikely to reflect a stiff noncompliant lung because the studies were performed while the patients were free of infection and rejection and the lung parenchyma had a normal appearance on the CT; in addition, stable heart-lung and double-lung transplant recipients disclose normal or nearly normal lung pressure-volume curves (24). Alternatively, weakness of the inspiratory muscles may contribute to decrease the volume of the transplanted lung at full inflation, as suggested by the observation that in SLT recipients esophageal pressure at TLC is less negative than normal (23). Such weakness may be due to a number of factors, including chronic use of steroids before and/or after surgery (27) and suboptimal nutritional status. Moreover, because of the interdependence between the two sides of the chest wall (see below), the hyperinflation on the side of the native lung might make the inspiratory muscles on the side of the graft operate at shorter than optimal lengths (28, 29), and hence reduce their pressure-generating capacity. Finally, the decreased TLC of the graft may also be related to the shift of the mediastinum, which would reduce lung volume in much the same way as a "space-occupying lesion" would.

The latter mechanism should also decrease the volume of the graft at end-expiration. However, one of the most conspicuous findings of this study has been that the FRC of the graft was similar to that of the control lung. This observation suggests that the loss of volume produced by displacement of the mediastinum toward the graft was counterbalanced by expansion of the chest wall, i.e., by a caudal displacement of the diaphragm and/or an increase in rib cage dimensions on the transplanted side. In this study, we tested the latter possibility by measuring the cross-sectional area of the rib cage in nine patients and nine matched control subjects. The measurements were made at FRC on a CT slice located midway between the manubrium sterni and the xyphoid process. On average the cross-sectional area of the rib cage on the side of the graft was 175 ± 10 cm² in the patients and 154 ± 7 cm² in the size-matched normal subjects (p < 0.003). This result indicates that the normal FRC of the graft was accounted for, at least in part, by expansion of the rib cage, which offset the effect of the mediastinal shift.

Two mechanisms may produce expansion of the rib cage on the side of the graft. First, the increase in rib cage anteroposterior diameter, which occurs in patients with emphysema (28) should be present on the side of the native lung after SLT and produce some increase in the dimensions of the cage on the side of the graft. Second, the chest wall in patients with long-standing hyperinflation may undergo irreversible changes in its passive elastic properties and relaxation volume. Previous studies have shown that such changes lead to a persistent increase in rib cage dimensions at FRC after heart-lung or double-lung transplantation in patients with preoperative thoracic hyperinflation (25, 26).

Because the patients were severely obstructed before surgery, it is likely that the high FRC of the native lung was dynamically determined and that this lung did not reach a static equilibrium over the time period of the CT acquisition. With more time for deflation, the native lung end-expiratory volume might have decreased further, allowing the mediastinum to be less shifted and the FRC of the graft to be even greater than measured in this study.

To further explore the factors that influenced the volume of the graft, we computed the relations between the actual FRC and TLC of the graft and the values predicted for the donor. No significant correlation was found, thus suggesting that the volume of the donor lung did not have a major role in determining the volume of the transplanted lung. On the other hand, we found significant correlations between the volume of the native and transplanted lung. These relations are shown for FRC and TLC in Figure 4, with the volume of each lung being expressed as a percent of the volume of the ipsilateral lung in the control subjects. The relations had a positive slope, which indicates that the volume of the graft tended to be greater in patients with a greater degree of hyperinflation in the native lung. This observation thus contradicts the belief that, by amplifying the displacement of the mediastinum, a larger native lung would tend to reduce the volume of the graft.

View larger version (13K): [in this window] [in a new window]   Figure 4.   Relationship between the volume of the native lung and the volume of the graft in 10 patients with SLT for emphysema. Volume is expressed as a percent of the volume of the ipsilateral lung in 10 matched normal control subjects.

In summary, the present studies have shown that the TLC of the graft and of the native lung do not change over time after SLT for emphysema. Although all patients show some degree of mediastinal shift toward the graft, expansion of the rib cage allows preservation of a normal FRC of the transplanted lung. On the other hand, the TLC of the graft is substantially reduced.