Effects of Starvation and Refeeding on Lung Mechanics and Morphometry3 HAMID SAHEBJAMI and CHARLES L. VASSALLO
SUMMARY Rats receiving one fifth of their usual daily food consumption for 10 days showed a significant increase in static recoil pressure of the lung (Pst[L]) due to surface forces at low lung volumes during inflation; their tissue Pst (L) decreased significantly over the entire volume-pressure loop compared to that of control rats. After 1 week of refeeding, the surface Pst (L) returned almost completely to normal, but tissue Pst (L) remained abnormally low. In starved rats, air-space enlargement with minimal loss of interalveolar septa was associated with a significant increase in mean linear intercept and volume fraction of air spaces, and with a significant decrease in corrected internal surface area and surface fraction of air space. These alterations returned partially toward normal in the refed group. We conclude that starvation increases surface elastic forces, decreases tissue elasticity of lung, and leads to air-space enlargement; refeeding leads to restoration of surface forces without the return of tissue elasticity to normal and to less severe air-space enlargement.
Introduction W e have previously shown that prolonged starvation is associated with significant alterations in the mechanical a n d ultrastructural characteristics of excised rat lungs (1). Rats allowed o n e fifth of their usual daily food c o n s u m p t i o n for 3 weeks showed a significant increase in surface elastic forces of the l u n g a n d a significant decrease in l u n g tissue elasticity (1). T h e s e mechanical changes were associated with significant decreases in the n u m b e r of lamellar bodies in g r a n u l a r pneumonocytes as well as in the vol-
(Received in original form October 10, 1978 and in revised form December 11,1978) 1 From the Pulmonary Section, Veterans Administration Medical Center, Cincinnati, Ohio 45220, and the Department of Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio. 2 This study was supported by grant 7690-01 from the Veterans Administration. 3 Requests for reprints should be addressed to Hamid Sahebjami, M.D., VA Medical Center, 3200 Vine St., Cincinnati, Ohio 45220.
u m e density of cytoplasm, mitochondria, and lamellar bodies of these cells in starved lungs (1). I n our previous communication, we pointed out the p o t e n t i a l significance of food deprivation in relation to h u m a n l u n g diseases, particularly p u l m o n a r y insufficiency a n d chronic obstructive p u l m o n a r y disease. Because short-term starvation leads to certain biochemical changes in the l u n g in the absence of significant alterations in l u n g mechanics ( 2 5), a n d because we were n o t aware of any previous studies of the influence of p r o l o n g e d starvation on l u n g function, in our initial experiments we elected to subject rats to r a t h e r extreme degrees of caloric restriction. T h e purpose of the present study was to e x a m i n e the influence of lesser degrees of starvation o n the static deflation volume-pressure (V-P) relationships a n d l u n g m o r p h o m e t r i c features. I n this m a n n e r , the results would be m o r e applicable to clinical medicine. W e also studied the process of recovery from starvation by refeeding. Materials and Methods Animals. Specific pathogen-free male rats of Long-
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 119, 1979
443
444
SAHEBJAMI AND VASSALLO
Evans descent (Blue-Spruce Farms, Inc., Altamont, N. Y.) were used. Some rats were initially placed in metabolic cages for 3 days, and their average daily food consumption (Purina rat chow composed of 23 per cent protein, 4.5 per cent fat, 6 per cent fiber, and 10.5 per cent ash and minerals) was measured. T h e animals were then divided into 3 groups. Rats in the control group were given food and water ad libitum. Rats in the starved group were allowed one fifth of their usual daily food consumption for 10 days, but were allowed water ad libitum. The refed group was initially treated like the starved group, but after the tenth day of starvation was allowed food and water ad libitum for 1 week. Animals were housed separately throughout the experimental period to reduce the potential for spread of infection, and in cages to prevent coprophagy. Volume-pressure measurements. Rats were killed by exsanguination after an intraperitoneal injection of 5 mg of sodium pentobarbital per 100 g of body weight. A midsternotomy incision was made; the lungs were exposed using a small retractor and were freed inside the thoracic cage, leaving them connected only to the trachea. An endotracheal tube was inserted, and the entire animal was placed in a vacuum jar for 3 min. After the lungs were degassed, the tracheal tube was connected by a Y-connector to a calibrated syringe attached to one limb and a Statham PM-5 pressure transducer (Statham Laboratories, Hato Rey, P.R.) connected to the other limb. Maximal lung volume at a transpulmonary pressure of 30 cm H 2 0 (MLV 30 ) was predicted for each animal from the regression on body weight (BW), in g: MLV 3 0 = B W X 0.0368 + 4.573 ±1.67. From air V-P curves studied in 136 normal male rats weighing 103 to 604 g, we have shown that MLV 3 0
has a high linear relationship with body weight (r = 0.93) as the independent variable (6). To determine directly the static recoil pressure of the lung (Pst[L]) during inflation and deflation, we inflated and deflated the lung stepwise with air to a known volume (10, 20, 30 . . . 100 per cent of MLV 30 ). T h e Pst (L) was recorded at each point after pressure had remained stable for at least 15 sec. If the pressure did not remain constant, this was taken as evidence of an air leak, and the lung was discarded. T h e volume of air was read directly from the syringe. Pressure was recorded on a Hewlett-Packard Model 17401A recorder (Hewlett-Packard, Dayton, Ohio). At MLV 3 0 , the lungs appeared to be completely inflated. The inflation and deflation processes were repeated once, and values obtained in the final procedure were taken for statistical analysis. After completion of air V-P measurements, the lungs were excised, weighed, and again degassed in a jar for 3 min. Saline V-P measurements were then performed, inflating and deflating the lungs stepwise in the same manner as described for air. The saline V-P apparatus and technique were those described by Mead and associates (7). The volume of saline was read directly from the injecting syringe, and the pressure was recorded from a water manometer after being stable for at least 15 sec at each value. If the pressure did not remain constant, this was taken as evidence of a leak, and the lung was discarded. Lungs were then removed from the saline bath, blotted dry, and reweighed. The initial weight was subtracted from the final weight, and the difference was taken as a measure of the amount of saline retained in the lungs. This volume, plus the volume removed during the deflation, was compared to the total volume injected to estimate the small leakage that could have
TABLE 1 INFLUENCE OF S T A R V A T I O N A N D R E F E E D I N G ON BODY W E I G H T A N D LUNG W E I G H T OF RATS
Body weight, g Initial Final Refeeding Lung weight, g Wet Dry
Lung dry weight/wet weight X 100 Lung wet weight/final body weight X 1 00
Control Rats
v •
sl
' N-
4
* s-* * / ™ - > • ^xii. »•:
%
^ T-> A % ¿¿Ai •* k\
Fig. 5. The lung of a control rat fixed by inflation with formalin at a pressure of 20 cm H 2 0 for 48 hours, showing normal terminal bronchiole and parenchyma (hematoxylin and eosin stain; original magnification: x 40). sue elastic forces dominated and shifted the air V-P curves upward and to the left of the control. This observation emphasizes the importance of both air and saline studies in the interpretation of V-P relationships. In the present study, the dual influence of food deprivation on the mechanical characteristics of the lung was such that an increase in surface elastic forces tended to shift the V-P curves downward and to the right, whereas a decrease in tissue elasticity tended to shift the curves upward and to the left; therefore, at any point over the inflation and deflation limbs of air V-P curves, the position of the curve was determined by the balance between these 2 forces. Because it appears that starvation has a more profound and lasting influence on tissue elasticity, in the absence of saline V-P studr ies one is apt not only to miss a substantial amount of information, but also to misinterpret the results. The alterations in surface elastic forces shown in this study are not related to interstitial or intra-alveolar fluid accumulation, because the ratios of lung dry weight to wet weight were not significantly different between the starved and control rats. Also, light microscopic studies done on pathologic specimens from these lungs did not reveal any evidence of fluid or cell accumulation in the lung parenchyma.
Although we have not performed biochemical studies on the lungs of starved animals, the indirect evidence presented here and elsewhere (l) suggests that changes in surface elastic forces are most probably due to a decrease in available surface-active material. Food deprivation decreases the activity of enzymes required for phospholipid synthesis (14). It could also deplete the amount of available substrate essential for surfactant synthesis. We have previously shown a 40 per cent decrease in the volume density of granular pneumonocyte mitochondria in prolonged starvation, suggesting inadequate energy production necessary for the secretion of surfactant into the air spaces as a likely factor (1). Starvation, according to this study, had a profound influence on tissue elasticity of the lung. The significant decrease in tissue Pst (L) during both inflation and deflation saline V-P curves could only be explained by the depletion of connective tissue components. Starved rats lost approximately 20 per cent of their initial body weight in 10 days. The decrease in dry weight of the lung was also substantial in starved animals. Studies on fasting for 72 hours have failed to show significant alterations in total protein content of the lung or its tissue elasticity (3, 4). This is most probably related to the fact that a substantial amount of lung protein was con-
449
EFFECTS OF STARVATION AND REFEEDING ON LUNG
tained in the connective tissue with a slow turnover and, thus, was unaffected by short-term food deprivation. However, prolonged starvation would be likely to lead to loss of connective tissue components and concomitant changes in tissue elasticity. Effects of refeeding on lung mechanics, according to our observations, were interesting. From V-P relationships with air and saline it appeared that after 1 week of refeeding, whereas surface Pst (L) returned toward normal, tissue Pst(L) remained significantly decreased and essentially unchanged compared to starved lungs. This again could be explained by the fact that surfactant belonged to a more labile pool with a half-life of 14 hours (15), whereas connective tissue had a much slower turnover rate. Therefore, surfactant was more readily replenishable than connective tissue. It follows that in early stages of recovery from starvation, tissue elastic forces are still diminished and require longer periods of refeeding to recover completely. Indeed, the possibility exists that, due to irreversible damage to air spaces, a complete return of tissue forces to normal with refeeding may never occur. The presence of enlarged air spaces in the lungs of starved animals demonstrated in pathologic specimens was very interesting and unex-
¿
-J& ff
. ....
¡J-s
%
"x
*"*
< fi
pected. The Lm, which is a measure of distance between the alveolar septa, and the Vva, which is an index of the lung tissue occupied by air spaces, were both significantly increased in starved rat lung. At the same time, the ISA4 and the Sva were significantly decreased. These morphometric data indicated an increase in the average interalveolar distance that can result from overinflation of the lungs, loss of interalveolar septa, or both. Distortion of lung architecture was very unlikely with an inflation pressure of 20 cm H 2 0 , even when there was a decrease in the tensile strength of lung tissue (16). Therefore, the observed enlargement of terminal air spaces in starved lungs was probably real and was probably chiefly related to alveolar dilatation with minimal loss of interalveolar septa. These morphometric measures confirmed results of saline V-P curves and represented alterations in tissue characteristics of terminal air spaces. Considering the much smaller body and lung sizes of starved rats, these changes in lung architecture became even more significant. After refeeding, all morphometric and histologic alterations were still present, but to a lesser extent. We would like to emphasize that the mechanical and morphologic alterations shown in this study are significant by statistical definition and not necessarily from a physiologic point of view.
'
>
. \ ,
^C
i
......
Fig. 6. The lung of a starved rat prepared the same way as a control lung, showing enlarged air spaces with minimal loss of interalveolar septa. The interstitium appears normal (hematoxylin and eosin stain; original magnification: X 40).
450
SAHEBJAMI AND VASSALLO
•
SUMMARY Rats receiving one fifth of their usual daily food consumption for 10 days showed a significant increase in static recoil pressure of the lung (Pst[L]) due to surface forces at low lung volumes during inflation; their tissue Pst (L) decreased significantly over the entire volume-pressure loop compared to that of control rats. After 1 week of refeeding, the surface Pst (L) returned almost completely to normal, but tissue Pst (L) remained abnormally low. In starved rats, air-space enlargement with minimal loss of interalveolar septa was associated with a significant increase in mean linear intercept and volume fraction of air spaces, and with a significant decrease in corrected internal surface area and surface fraction of air space. These alterations returned partially toward normal in the refed group. We conclude that starvation increases surface elastic forces, decreases tissue elasticity of lung, and leads to air-space enlargement; refeeding leads to restoration of surface forces without the return of tissue elasticity to normal and to less severe air-space enlargement.
Introduction W e have previously shown that prolonged starvation is associated with significant alterations in the mechanical a n d ultrastructural characteristics of excised rat lungs (1). Rats allowed o n e fifth of their usual daily food c o n s u m p t i o n for 3 weeks showed a significant increase in surface elastic forces of the l u n g a n d a significant decrease in l u n g tissue elasticity (1). T h e s e mechanical changes were associated with significant decreases in the n u m b e r of lamellar bodies in g r a n u l a r pneumonocytes as well as in the vol-
(Received in original form October 10, 1978 and in revised form December 11,1978) 1 From the Pulmonary Section, Veterans Administration Medical Center, Cincinnati, Ohio 45220, and the Department of Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio. 2 This study was supported by grant 7690-01 from the Veterans Administration. 3 Requests for reprints should be addressed to Hamid Sahebjami, M.D., VA Medical Center, 3200 Vine St., Cincinnati, Ohio 45220.
u m e density of cytoplasm, mitochondria, and lamellar bodies of these cells in starved lungs (1). I n our previous communication, we pointed out the p o t e n t i a l significance of food deprivation in relation to h u m a n l u n g diseases, particularly p u l m o n a r y insufficiency a n d chronic obstructive p u l m o n a r y disease. Because short-term starvation leads to certain biochemical changes in the l u n g in the absence of significant alterations in l u n g mechanics ( 2 5), a n d because we were n o t aware of any previous studies of the influence of p r o l o n g e d starvation on l u n g function, in our initial experiments we elected to subject rats to r a t h e r extreme degrees of caloric restriction. T h e purpose of the present study was to e x a m i n e the influence of lesser degrees of starvation o n the static deflation volume-pressure (V-P) relationships a n d l u n g m o r p h o m e t r i c features. I n this m a n n e r , the results would be m o r e applicable to clinical medicine. W e also studied the process of recovery from starvation by refeeding. Materials and Methods Animals. Specific pathogen-free male rats of Long-
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 119, 1979
443
444
SAHEBJAMI AND VASSALLO
Evans descent (Blue-Spruce Farms, Inc., Altamont, N. Y.) were used. Some rats were initially placed in metabolic cages for 3 days, and their average daily food consumption (Purina rat chow composed of 23 per cent protein, 4.5 per cent fat, 6 per cent fiber, and 10.5 per cent ash and minerals) was measured. T h e animals were then divided into 3 groups. Rats in the control group were given food and water ad libitum. Rats in the starved group were allowed one fifth of their usual daily food consumption for 10 days, but were allowed water ad libitum. The refed group was initially treated like the starved group, but after the tenth day of starvation was allowed food and water ad libitum for 1 week. Animals were housed separately throughout the experimental period to reduce the potential for spread of infection, and in cages to prevent coprophagy. Volume-pressure measurements. Rats were killed by exsanguination after an intraperitoneal injection of 5 mg of sodium pentobarbital per 100 g of body weight. A midsternotomy incision was made; the lungs were exposed using a small retractor and were freed inside the thoracic cage, leaving them connected only to the trachea. An endotracheal tube was inserted, and the entire animal was placed in a vacuum jar for 3 min. After the lungs were degassed, the tracheal tube was connected by a Y-connector to a calibrated syringe attached to one limb and a Statham PM-5 pressure transducer (Statham Laboratories, Hato Rey, P.R.) connected to the other limb. Maximal lung volume at a transpulmonary pressure of 30 cm H 2 0 (MLV 30 ) was predicted for each animal from the regression on body weight (BW), in g: MLV 3 0 = B W X 0.0368 + 4.573 ±1.67. From air V-P curves studied in 136 normal male rats weighing 103 to 604 g, we have shown that MLV 3 0
has a high linear relationship with body weight (r = 0.93) as the independent variable (6). To determine directly the static recoil pressure of the lung (Pst[L]) during inflation and deflation, we inflated and deflated the lung stepwise with air to a known volume (10, 20, 30 . . . 100 per cent of MLV 30 ). T h e Pst (L) was recorded at each point after pressure had remained stable for at least 15 sec. If the pressure did not remain constant, this was taken as evidence of an air leak, and the lung was discarded. T h e volume of air was read directly from the syringe. Pressure was recorded on a Hewlett-Packard Model 17401A recorder (Hewlett-Packard, Dayton, Ohio). At MLV 3 0 , the lungs appeared to be completely inflated. The inflation and deflation processes were repeated once, and values obtained in the final procedure were taken for statistical analysis. After completion of air V-P measurements, the lungs were excised, weighed, and again degassed in a jar for 3 min. Saline V-P measurements were then performed, inflating and deflating the lungs stepwise in the same manner as described for air. The saline V-P apparatus and technique were those described by Mead and associates (7). The volume of saline was read directly from the injecting syringe, and the pressure was recorded from a water manometer after being stable for at least 15 sec at each value. If the pressure did not remain constant, this was taken as evidence of a leak, and the lung was discarded. Lungs were then removed from the saline bath, blotted dry, and reweighed. The initial weight was subtracted from the final weight, and the difference was taken as a measure of the amount of saline retained in the lungs. This volume, plus the volume removed during the deflation, was compared to the total volume injected to estimate the small leakage that could have
TABLE 1 INFLUENCE OF S T A R V A T I O N A N D R E F E E D I N G ON BODY W E I G H T A N D LUNG W E I G H T OF RATS
Body weight, g Initial Final Refeeding Lung weight, g Wet Dry
Lung dry weight/wet weight X 100 Lung wet weight/final body weight X 1 00
Control Rats
v •
sl
' N-
4
* s-* * / ™ - > • ^xii. »•:
%
^ T-> A % ¿¿Ai •* k\
Fig. 5. The lung of a control rat fixed by inflation with formalin at a pressure of 20 cm H 2 0 for 48 hours, showing normal terminal bronchiole and parenchyma (hematoxylin and eosin stain; original magnification: x 40). sue elastic forces dominated and shifted the air V-P curves upward and to the left of the control. This observation emphasizes the importance of both air and saline studies in the interpretation of V-P relationships. In the present study, the dual influence of food deprivation on the mechanical characteristics of the lung was such that an increase in surface elastic forces tended to shift the V-P curves downward and to the right, whereas a decrease in tissue elasticity tended to shift the curves upward and to the left; therefore, at any point over the inflation and deflation limbs of air V-P curves, the position of the curve was determined by the balance between these 2 forces. Because it appears that starvation has a more profound and lasting influence on tissue elasticity, in the absence of saline V-P studr ies one is apt not only to miss a substantial amount of information, but also to misinterpret the results. The alterations in surface elastic forces shown in this study are not related to interstitial or intra-alveolar fluid accumulation, because the ratios of lung dry weight to wet weight were not significantly different between the starved and control rats. Also, light microscopic studies done on pathologic specimens from these lungs did not reveal any evidence of fluid or cell accumulation in the lung parenchyma.
Although we have not performed biochemical studies on the lungs of starved animals, the indirect evidence presented here and elsewhere (l) suggests that changes in surface elastic forces are most probably due to a decrease in available surface-active material. Food deprivation decreases the activity of enzymes required for phospholipid synthesis (14). It could also deplete the amount of available substrate essential for surfactant synthesis. We have previously shown a 40 per cent decrease in the volume density of granular pneumonocyte mitochondria in prolonged starvation, suggesting inadequate energy production necessary for the secretion of surfactant into the air spaces as a likely factor (1). Starvation, according to this study, had a profound influence on tissue elasticity of the lung. The significant decrease in tissue Pst (L) during both inflation and deflation saline V-P curves could only be explained by the depletion of connective tissue components. Starved rats lost approximately 20 per cent of their initial body weight in 10 days. The decrease in dry weight of the lung was also substantial in starved animals. Studies on fasting for 72 hours have failed to show significant alterations in total protein content of the lung or its tissue elasticity (3, 4). This is most probably related to the fact that a substantial amount of lung protein was con-
449
EFFECTS OF STARVATION AND REFEEDING ON LUNG
tained in the connective tissue with a slow turnover and, thus, was unaffected by short-term food deprivation. However, prolonged starvation would be likely to lead to loss of connective tissue components and concomitant changes in tissue elasticity. Effects of refeeding on lung mechanics, according to our observations, were interesting. From V-P relationships with air and saline it appeared that after 1 week of refeeding, whereas surface Pst (L) returned toward normal, tissue Pst(L) remained significantly decreased and essentially unchanged compared to starved lungs. This again could be explained by the fact that surfactant belonged to a more labile pool with a half-life of 14 hours (15), whereas connective tissue had a much slower turnover rate. Therefore, surfactant was more readily replenishable than connective tissue. It follows that in early stages of recovery from starvation, tissue elastic forces are still diminished and require longer periods of refeeding to recover completely. Indeed, the possibility exists that, due to irreversible damage to air spaces, a complete return of tissue forces to normal with refeeding may never occur. The presence of enlarged air spaces in the lungs of starved animals demonstrated in pathologic specimens was very interesting and unex-
¿
-J& ff
. ....
¡J-s
%
"x
*"*
< fi
pected. The Lm, which is a measure of distance between the alveolar septa, and the Vva, which is an index of the lung tissue occupied by air spaces, were both significantly increased in starved rat lung. At the same time, the ISA4 and the Sva were significantly decreased. These morphometric data indicated an increase in the average interalveolar distance that can result from overinflation of the lungs, loss of interalveolar septa, or both. Distortion of lung architecture was very unlikely with an inflation pressure of 20 cm H 2 0 , even when there was a decrease in the tensile strength of lung tissue (16). Therefore, the observed enlargement of terminal air spaces in starved lungs was probably real and was probably chiefly related to alveolar dilatation with minimal loss of interalveolar septa. These morphometric measures confirmed results of saline V-P curves and represented alterations in tissue characteristics of terminal air spaces. Considering the much smaller body and lung sizes of starved rats, these changes in lung architecture became even more significant. After refeeding, all morphometric and histologic alterations were still present, but to a lesser extent. We would like to emphasize that the mechanical and morphologic alterations shown in this study are significant by statistical definition and not necessarily from a physiologic point of view.
'
>
. \ ,
^C
i
......
Fig. 6. The lung of a starved rat prepared the same way as a control lung, showing enlarged air spaces with minimal loss of interalveolar septa. The interstitium appears normal (hematoxylin and eosin stain; original magnification: X 40).
450
SAHEBJAMI AND VASSALLO
•
