INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary surfactant is a complex mixture of phospholipids and proteins excreted into the fluid lining the alveoli of the lung (1). Its main function is to lower the surface tension of the fluid layer. A low surface tension prevents alveoli collapse at end-expiration and also prevents the formation of alveolar edema. In addition, surfactant plays a role in host defense against invading micro-organisms (2, 3).

Immediately after lung transplantation in animals, pulmonary surfactant is affected in its composition and biophysical function (4). While the total amount of surfactant phospholipids remains constant after lung transplantation, its composition changes unfavorably in at least three aspects: the proportion of surface-active heavy subtype surfactant, the percentage of phosphatidylcholine (PC) in the phospholipids, and the amount of surfactant protein A (SP-A) decrease. The function of this altered surfactant has been shown to be impaired when tested in vitro for its capacity to reduce the surface tension at an air/fluid interface (4). In the transplanted lung, this poor biophysical function of surfactant is inhibited additionally by serum proteins leaked into the alveolar space (5).

The changes of surfactant result from the acute injury inflicted upon the lungs during the transplantation procedure. Among other factors, ischemia-reperfusion is generally considered to be most damaging for transplants (7). Prolonging the ischemia time leads to more severe lung injury (8), with progressive effects on surfactant and lung function (4). In clinical lung transplantation, the injury is recognized as diffuse roentgenographic infiltrates, protein-rich lung edema, and decreased gas exchange and decreased lung compliance. This syndrome is referred to as the reimplantation response (9).

Treatment with exogenous surfactant, instilled before reperfusion into lung transplants, has been shown to improve pulmonary surfactant as investigated for a few hours after reperfusion (6, 10). The surfactant treatment increased the total amount of phospholipids (6), the proportion of surface-active heavy subtype (6, 10), and the percentage of PC to normal or supranormal values, and it preserved the amount of SP-A (6). As a result of these improvements in composition, the in vitro biophysical function of surfactant was improved (6). Early surfactant treatment also improved the function of the lung immediately after transplantation.

All studies on pulmonary surfactant in lung transplants have focused on the first few hours after transplantation but recovery of surfactant composition and function on a longer term is expected to be important for both lung transplant function and host defense. It is yet unknown whether altered surfactant recovers from the transplantation injury during the first post-transplantation week, in parallel with the symptoms of the reimplantation response (9), and whether this recovery is affected by early surfactant treatment.

These two questions were investigated in this study by transplanting left lungs syngeneically in rats. The severity of transplantation injury was varied by using a 6- and 20-h lung ischemia period. For comparison, a sham-operated group without transplantation or ischemia-reperfusion but with injury caused by the surgical procedures in the recipient was included. Surfactant was instilled into half of the transplanted lungs in each ischemia group. During 1 wk, lung transplants were investigated for recovery from injury, and, at the end of this week, for lung function and for amount, composition, and function of its pulmonary surfactant.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Groups

Inbred rats (LEW strain, specific-pathogen free, male, weighing 240- 300 g) were used for donors, recipients, sham-operated, and normal animals. All animals received humane care in compliance with the Dutch regulations and law. Left lungs were transplanted syngeneically to avoid influence of rejection or immunosuppressive treatment. Lungs from six experimental groups were investigated for recovery from transplantation injury and for properties of pulmonary surfactant: (1) transplanted after 6-h ischemia (untreated 6-h, n = 6); (2) transplanted after 20-h ischemia (untreated 20-h, n = 10); (3) transplanted after 6-h ischemia plus surfactant treatment (surfactant-treated 6-h, n = 6); (4) transplanted after 20-h ischemia plus surfactant treatment (surfactant-treated 20-h, n = 9); and as controls, (5) sham-operated rats without ischemia-reperfusion but with the effects of left sided thoracotomy, preparation of pulmonary artery and mechanical ventilation (sham, n  = 4); and (6) normal, non-operated rats (normal, n = 5).

Lung Transplantation and Surfactant Treatment

Left lung transplantation was performed as previously described (5). For harvesting the left lung from the donor, the left pulmonary artery was flushed with 25 ml cold (8° C) saline at 40 cm water pressure while the lung was ventilated with a positive end expiratory pressure (PEEP) of 4 cm H₂O. After dissection of the left main bronchus the lung was allowed to deflate. Then the lung was submersed in 8° C saline and placed in the refrigerator at 8° C for 5 or 19 h. One hour before the planned start of reperfusion the lung was taken out of the refrigerator and covered by a sheet of thin plastic.

The recipient rat was anesthetized with a mixture of fluothane (1- 2%), N₂O, and oxygen and ventilated with peak inspiratory pressures (PIP) of 16 cm H₂O, PEEP of 2 cm H₂O at 40 breaths/min and infused with saline containing 0.3 mM bicarbonate (± 3 ml/h). The left lung transplant was implanted through a left sided thoracotomy. The pulmonary artery and vein of the lung transplant and recipient were anastomosed first. In the left pulmonary vein a small canula, filled with heparin solution, was placed just above the anastomose site (secured with a purse string suture) to facilitate blood sampling for blood gas analysis. Finally, the left main bronchus was sutured and surfactant or a bolus of air was instilled.

For surfactant treatment we used Alveofact^®, an organic solvent extracted surfactant from bovine lung (Thomae GmbH, Biberach, Germany). Just before reperfusion of the left lung transplant the surfactant (25 mg phospholipids per ml) in a dose of 100 mg/kg body weight (BW) was instilled directly into the main bronchus of the transplant, as described previously (6). The untreated groups received a bolus of air instead of surfactant. Air instead of saline was used as control because instillation of 1 ml saline caused the death of all five animals (6-h cold ischemia) on ventilator in an earlier performed experiment. Reperfusion of the lung transplant was allowed by releasing the clip on the pulmonary vein first and then that on the pulmonary artery. Warm ischemia time (= ischemia time during implantation) was 1 h ± 10 min; total lung ischemia times were 6 h ± 15 min or 20 h ± 30 min. After the lung function was measured, the canula in the pulmonary vein was removed, the chest was closed and the rat was weaned from the ventilator.

Sham Operation

The sham operation was similar to the recipient operation including limited dissection of tissue around the hilus but without transplantation of the left lung. The sham-operated rats were used as controls for the recovery of surfactant from surgical injury without ischemia-reperfusion.

Symptoms of Reimplantation Response

Immediate lung function. The immediate effect of transplantation injury after 25 min of reperfusion was assessed by the AaO₂ gradient in the left lung, the blood flow in the left pulmonary artery and by the peri-operative mortality. Five minutes before blood was taken from the left pulmonary vein for gas analysis, we suctioned the tube and trachea to remove excessive lung fluids and ventilated the lung four times with an increased maximum pressure of 30 cm H₂O to fully expand the lung. The PO₂ and PCO₂ were measured with a blood gas analyzer (ABL 330; Radiometer, Copenhagen, Denmark). The alveolar-arterial O₂ gradient (AaO₂ = PA_O2 Pa_O2) was calculated using the alveolar gas equation to determine the alveolar PO₂ (PA_O2) · (PA_O2 = FI_O2 · (Pb  PH₂O)  (PA_CO2/R); FI_O2 = fraction inspired oxygen; Pb = barometric air pressure; PH₂O = partial pressure of water vapor at 37° C; PA_CO2 = alveolar CO₂ pressure (assumed to be equal to Pa_CO2); Pa_O2 = arterial O₂ pressure; R = respiratory quotient (assumed to be equal to 0.8 [11]). The blood flow in the pulmonary artery was measured with an ultrasound probe (1 mm [RB]; Transonic Systems Inc., Ithaca, NY) placed round the artery.

Mortality

To wean the rat from the ventilator, the PIP, PEEP, and frequency were gradually decreased and the FI_O2 lowered to 0.3. In some cases the lung transplantation procedure led to the death of the animal during the weaning procedure or shortly thereafter. The mortality was used as an immediate measure of the severity of the transplantation injury.

Chest Roentgenograms

To assess recovery from transplantation injury we used a ventilation score (9) of chest roentgenograms taken at Days 2, 5 and 7 after transplantation. To ascertain a good contrast, a short exposure time (0.06 s) and a mammography film was used. Two investigators blinded for the experimental conditions but not for the day after transplantation or sham operation scored the chest roentgenograms. The score, indicating the degree of ventilation, was the summation of 2 scores and ranged from 0 to 6: one score for aerated surface area (3 = totally -2/3 aerated, 2 = 2/3-1/3, 1 = 1/3 - minimal, 0 = not aerated) and a second score for lung density (3 = normal, 2 = slightly or patchily increased, 1 = moderately increased, and 0 = opaque).

Lung Function at One Week

At 1 wk gas exchange, dynamic compliance and pulmonary perfusion were measured. For these measurements the animals were anesthetized with chloralhydrate intraperitoneally, intubated and ventilated with FI_O2 = 1, at 24 cm H₂O PIP, 5 cm H₂O PEEP, and a frequency of 55 breaths/min. After 10 min of ventilation the dynamic compliance of both lungs was determined and blood was drawn from the left ventricle to measure the PO₂ and PCO₂. The dynamic compliance of both lungs was calculated as: Vmax/24 cm H₂O/ (kg BW donor + kg BW recipient)/ 2, where Vmax is the volume at 24 cm water pressure. Due to adhesions it was impossible to dissect the left lung for separate function measurements without severely damaging the lung; therefore the measured dynamic compliance is the total of left and right lungs. The left lung transplant perfusion was measured by intravenously injected Tc191 labeled macro-aggregates and expressed as percentage counts of total counts (9).

Injury Variables in BALF

After measurement of the lung function at 1 wk, heart and lungs were taken out en bloc and left and right lungs were lavaged separately for five times with ice-cold saline as described previously (5). The bronchoalveolar fluid (BALF) was collected on ice and the total recovered volume noted. Most untreated 20-h ischemia left lung transplants had to be lavaged with smaller volumes than normal to prevent rupture of these lungs. This resulted in a recovered mean volume of 40.8  ± 11.1 ml /kg for the untreated 20-h ischemia group and 54.4 ± 5.8 ml/kg for all other groups; recovery of the BALF was > 89% for all groups. Right lungs of all groups were lavaged with a mean volume of 101 ± 9.2 ml/kg. The percentage of PMNs and the amount of alveolar proteins in BALF were used as indicators of transplantation injury. Alveolar cells were isolated from BALF by centrifugation at 150 g for 10 min at 4° C. PMNs were differentiated from alveolar macrophages and lymphocytes on stained cytospin preparations. In the remaining cell-free BALF the total amount of alveolar proteins was measured according to Lowry modified by adding SDS (12). This cell-free BALF was also used for surfactant determinations.

Pulmonary Surfactant at 1 wk: Surfactant in BALF

To analyze the properties of pulmonary surfactant, we assessed its composition in four ways: the total amount of surfactant phospholipids, the percentage of heavy subtype surfactant, the percentage PC of total phospholipids, and the amount of SP-A; in addition we measured the surface tension lowering properties of surfactant in vitro. The amount of phospholipids was measured with a perchloric acid method (13) after lipid extraction (14). The amount of total phospholipids was calculated by summation of the amounts of phospholipids of the heavy and light subtype fractions of a 1 ml BALF sample after these fractions were separated by centrifugation at 40,000 g for 15 min as described previously (6). To determine the composition of surfactant phospholipids, the remainder of BALF was centrifuged overnight at 100,000 g, 4° C, to concentrate all the surfactant in a pellet. The phospholipids in the pellet were separated by thin layer chromatography according to Touchstone with solvent E (15). The amount of SP-A in a sample of the BALF was assessed with a sandwich ELISA method (16). The measured amounts of SP-A represent endogenous SP-A of the lung transplant since the instilled surfactant contains no SP-A. For in vitro surface tension measurements the pelleted surfactant was extracted with chloroform-methanol and resuspended in saline with 20 mM CaCl₂ to a final concentration of 2 mg phospholipids per ml as described previously (6). Using this purification protocol the final suspension contains the surfactant phospholipids and the hydrophobic surfactant proteins SP-B and SP-C. The surface tension of the surfactant suspension was analyzed in a pulsating bubble surfactometer (Electronetics Corporation, Amherst, NY) according to Enhorning (17). We used  at RMIN, the minimum surface tension, at 20 pulsations and 37° C as the variable of in vitro surfactant function.

Statistical Analysis

For comparison between ventilation scores of the different groups the Mann-Whitney U Sum Rank test was performed. All the other data of the different groups were assessed by single factor ANOVA. Differences with a p value of < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung Transplant Injury and Recovery

Immediately after transplantation, acute lung injury was seen as severe lung dysfunction. At 25 min after reperfusion, the AaO₂ gradient and blood flow were abnormal in both untreated groups compared with the sham group (Table 1). Surfactant treatment improved both the AaO₂ gradient (p = 0.06) and blood flow in the 6-h ischemia group, but not in the 20-h ischemia group (Table 1). The total mortality was 36% without a significant influence of ischemia time or surfactant treatment (Table 1). The mortality occurred exclusively during the first post-operative hour, and in animals that showed severe injury after 25 min of reperfusion as measured by an AaO₂ gradient over 70 kPa.

The chest roentgenograms of the transplanted and sham-operated rats showed initial injury and then a gradual recovery of the lungs during the week (Figure 1). On the first roentgenogram, taken 2 d after operation, all lungs were graded lower than the normal ventilation score of 6. In particular the untreated transplants scored poorly, but even sham-operated lungs were abnormal. On Day 5, lungs in all groups scored slightly better and continued to improve on the roentgenograms at 1 wk. Then, most transplants only showed small radiographic infiltrates, except for lungs in the untreated ischemia groups, of which the 20-h ischemia lungs scored consistently lower than the sham-operated lungs. Importantly, surfactant treatment had significantly increased the ventilation score in the 20-h ischemia group.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Ventilation scores of left transplanted and sham-operated lungs. Ventilation scores improved during 1 wk after lung transplantation, except in the untreated 20-h ischemia group. In the untreated 20-h ischemia group the score was lower than in the sham control group throughout the week (*p < 0.05) and also lower than the surfactant treated 20-h ischemia group (p < 0.05). Data are given as medians; error bars indicate minimum or maximum scores given in a group.

In most groups the lung function recovered to sham values by one week (Table 2). Only in the untreated 20-h group the AaO₂ gradient and dynamic compliance of both lungs remained lower than in the sham-operated group. In the 20-h ischemia group surfactant treatment resulted in a significantly higher dynamic compliance of both lungs, in parallel with the higher ventilation score of the lung transplants. The perfusion was reduced in all transplants, without obvious influence of ischemia or surfactant treatment. Symptoms of persistent lung injury in the transplants as measured by amounts of alveolar protein and % PMN in BALF were absent; the values were as low as in normal rats (Table 2).

Pulmonary Surfactant

We found that, without treatment, surfactant in the transplanted left lungs was still abnormal at 1 wk after operation (Table 3). Generally, surfactant changes were more severe in the 20-h ischemia group than in the 6-h group and least severe in the sham-operated group. The amount of surfactant phospholipids was significantly lower in these untreated transplanted groups; the proportion heavy subtype was low but within normal range, and the percentage PC was normal (Table 3). The amount of SP-A was significantly lower in the transplanted groups but not in the sham-operated group. The changes in surfactant after transplantation did not impair the in vitro biophysical function of surfactant extracted from BALF; the minimum surface tension of surfactant was in the normal range (Table 3).

Surfactant treatment at the time of lung transplantation appeared to improve or even normalize the surfactant changes at 1 wk. The amount of phospholipids was improved to normal values (Table 3). Surfactant treatment resulted in a slightly higher SP-A amount in both the 6- and 20-h ischemia groups, but the large variation in lung injury within each transplantation group prevented a significant improvement (Table 3). When analyzing this variation, it appeared that the amount of SP-A at 1 wk correlated strongly with the initial acute injury at the time of reperfusion expressed by the AaO₂ gradient (Figure 2). This correlation, including surfactant-treated and untreated transplants, was highly significant (p < 0.000001). The in vitro biophysical function of surfactant remained normal (Table 3).

View larger version (22K): [in this window] [in a new window]   Figure 2.   Amount of SP-A in BALF at 1 wk after transplantation as a function of the AaO2 gradient in the transplant at 25 min of reperfusion. Each symbol represents this relation for a single lung transplanted rat. Regression analysis revealed a highly significant correlation indicating that the severity of the immediate lung transplant injury after reperfusion determines the amount of SP-A 1 wk after transplantation.

Surfactant phospholipids from the right native lungs of the transplanted and sham-operated animals were virtually normal with a slight decrease of proportion of heavy subtype surfactant in all operated groups (data not shown). In both the untreated and surfactant-treated 20-h ischemia group, however, an isolated significant lower amount of SP-A in right lungs to 40% of normal values was measured (82.3 ± 18.4, and 78.4 ± 7.8 versus 197.6 ± 51.3 µg/kg, respectively, p < 0.01).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previous studies have shown that pulmonary surfactant is affected early during the reimplantation response after lung transplantation (4, 10). The surface-active surfactant components, such as its heavy subtype, PC, and SP-A were decreased, which impaired its in vitro biophysical function (4, 10). This impairment is aggravated in vivo by leaked serum proteins. The present study shows that pulmonary surfactant at one week after transplantation is still deficient. In the untreated 6-h ischemia group the total amount of surfactant phospholipids was only two-thirds of normal and the amount of SP-A only half of normal; the relative amounts of heavy subtype and PC were normal. Strikingly, the function of lung transplants in the untreated 6-h ischemia group had recovered almost completely from the reimplantation response. In the 20-h ischemia group the total amount of surfactant phospholipids was further decreased to a third and the amount of SP-A to a quarter of normal as an effect of the more severe transplantation injury. In this group the lung transplants still showed features of the reimplantation response: abnormal chest X-ray, impaired gas exchange, and a low dynamic compliance of both lungs. In sham-operated rats, ventilation without ischemia-reperfusion had no significant effect on pulmonary surfactant. Thus, pulmonary surfactant has not fully recovered from the transplantation injury at one week after syngeneic lung transplantation even in lungs in which the reimplantation response apparently is resolved.

The deficient amounts of surfactant phospholipids and SP-A at 1 wk might be caused by the immediate transplantation injury of surfactant producing type II pneumocytes after transplantation. Semik and colleagues found that the type II cells in rat lung transplants were depleted from surfactant-containing lamellar bodies up to one week after transplantation (18). These authors hypothesized that this depletion was caused by a deficient synthesis of surfactant after transplantation, which hypothesis would also explain the low alveolar amounts of surfactant in our present study. How the function of type II cells is affected by transplantation injury might be explained in two ways.

First, as shown in other lung injury models, type II cells start to proliferate in response to injury of type I cells. These proliferating type II cells have a decreased lamellar body pool size (19), which is consistent with a decreased synthesis of surfactant. Second, production or secretion of phospholipids and SP-A by type II cells can be impaired by reactive oxygen metabolites and cytokines (20), which are generated after lung transplantation (24, 25). Both these mechanisms may have induced the changes of surfactant that we measured at 1 wk after the initial transplantation injury in the present study.

The present study also shows that treatment with surfactant, instilled as a single dose before reperfusion, increased the surfactant phospholipids to normal amounts at one week after lung transplantation. This normal amount of surfactant might consist merely of instilled surfactant, however, in other acute lung injury models a significant amount of instilled surfactant is quickly metabolized by type II cells or cleared by macrophages (3). Another explanation is that the instillation of surfactant preserves the production of surfactant. This explanation is supported by the study of Semik and coworkers showing that depletion of lamellar bodies in type II cells during the transplant-reperfusion period is prevented by surfactant treatment (18). In conclusion, the persistent normal amount of surfactant phospholipids after surfactant treatment in our lung transplants might be the result of both suppletion and a preserved surfactant metabolism.

Interestingly, surfactant treatment did not only improve surfactant amounts, but also enhanced the recovery from transplantation injury; the ventilation score, the AaO₂ gradient and the dynamic compliance improved to sham control values. One can think of two ways in which surfactant treatment improved lung function at 1 wk. First, the higher amount of alveolar surfactant might have resulted in a more widespread stabilization of alveoli, thereby improving lung function. Second, the high amount of exogenous surfactant may have suppressed the inflammation process that is provoked by ischemia-reperfusion. Such an anti-inflammatory capacity of surfactant has been shown in vitro and in in vivo (lung) injury models (26). Apparently, the effect of surfactant instillation remained locally, that is in the alveolar compartment of the lung, as lung perfusion was not improved. We did not assess whether this low perfusion was due to abnormalities in the pulmonary artery or in the arterioles. Despite the absent effect on lung perfusion, one dose of surfactant before reperfusion clearly improves the recovery of lung transplants from serious transplantation injury.

It seems not so surprising that our surfactant treatment had no effect on SP-A because the instilled surfactant contained no SP-A. Yet, in a previous study instilled surfactant preserved the amount of endogenous SP-A as measured one hour after reperfusion. This preservative effect was attributed to the instilled surfactant phospholipids which were thought to minimize the breakdown of SP-A by products of activated PMNs (6). The low amounts of SP-A at one week after transplantation seems not to correlate with ongoing SP-A breakdown caused by PMNs because few PMNs were present in BALF (Table 3). Instead, the observed correlation between increased severity of immediate transplantation injury and low amounts of SP-A at 1 wk after injury (Figure 2) might indicate that besides initial SP-A breakdown the SP-A producing type II cells (and Clara cells) are also damaged. This theory of an impaired SP-A production is in agreement with the clinical finding that, for more than a year after transplantation, the amount of SP-A is still low in lung transplant patients (27). This persistently low SP-A level might be of relevance for the local antibacterial defense (3) of transplanted lungs.

Few patients have been treated with surfactant after lung transplantation. One report supports the experimental data that it is useful treatment in early graft failure (28). Especially prevention of graft failure after prolonged ischemia times might prove to be successful clinically. A promising way of surfactant treatment is to treat first the donor and then again the lung transplant just before reperfusion (29).

In conclusion the present study shows, first, that surfactant is not fully recovered at one week after transplantation even in lungs which have apparently recovered from transplantation injury. Second, this study shows that early surfactant treatment in lung transplantation enhances recovery from transplantation injury and has persistent beneficial effects on pulmonary surfactant.