INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Chronic lung hyperinflation that is pathognomonic for pulmonary emphysema displaces the diaphragm caudally to a mechanically disadvantaged position. This condition elevates the energetic demands placed upon the diaphragm and other respiratory muscles (1) and is associated with respiratory muscle fatigue and ultimately failure (2, 3).

In the widely accepted hamster model of emphysema, the diaphragm undergoes structural and functional adaptations which include chronic shortening consequent to loss of sarcomeres in series (4); possible diaphragm myocyte hypertrophy (6) but not in all instances (5, 10); capillary neogenesis and elevated capillary tortuosity (6, 8, 9); in certain instances, mitochondrial enzymes are elevated (increased [5, 8, 11]; unchanged [11, 12]); and elevated diaphragm blood flow during exercise but not at rest (11).

For diaphragm function to be sustained, sufficient substrate flux is requisite. The most logical and enduring mechanistic basis invoked to explain respiratory muscle fatigue and failure involves an O₂ supply-demand imbalance. It is therefore crucial to understand more about the relationship between diaphragm O₂ delivery (O₂) and O₂ utilization (O₂) in emphysema. It is certainly surprising that given the expected elevation of diaphragm metabolic demands (which are sufficiently powerful to cause profound diaphragmatic adaptations), diaphragm blood flow (and O₂) is not elevated at rest (11). Under these circumstances, it would be expected that fractional O₂ extraction must be elevated. However, our understanding of O₂ exchange within the diaphragm is hampered by the complexity of the diaphragmatic blood supply and its structural and functional heterogeneity.

We have demonstrated previously that phosphorescence quenching provides a tenable method for monitoring microvascular O₂ pressure (PO₂m) within the rodent diaphragm (13). PO₂m reflects the dynamic balance between O₂ and O₂ removal or O₂ within the microvascular space. The purpose of the present investigation was to utilize this technique to test the hypothesis that diaphragm PO₂m is decreased in emphysema. Specifically, measurements of diaphragm PO₂m were made across a range of inspired O₂ concentrations and blood pressures (hypovolemia induced by blood withdrawal) in control and emphysematous hamsters. Under each condition, PO₂m was reduced systematically in the diaphragm of emphysematous hamsters.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Animals

All procedures were conducted in accordance with the rules and regulations of the IACUC at Kansas State University. Male Syrian golden hamsters (Sasco, Omaha, NE), initially weighing ~ 120 g, were used in these studies. All animals were housed in individual cages (8 × 10 × 10 in.) and allowed free access to water and rodent chow.

Induction of Emphysema

Hamsters were assigned randomly to either a control (C) or emphysema (E) group and E was produced by intratracheal instillation of elastase (4, 14). Experiments were conducted 20-28 wk after elastase treatment to allow adequate time for adaptation within the diaphragm.

Experimental Preparation

Hamsters were anesthetized with pentobarbitol sodium (40 mg/kg intraperitoneally, to effect) and the right carotid artery was cannulated. Body temperature was maintained at 37-38° C and the diaphragm was accessed via laparotomy and superfused with a warmed (38° C) Krebs- Henseleit bicarbonate-buffered solution equilibrated with 95% N₂/5% CO₂ (22). During all experiments, the animals breathed spontaneously. Mean arterial pressure (MAP) was measured using a DigiMed blood pressure monitor (Micro-Med, Louisville, KY) and blood gases and pH using a NOVA Stat Profile (F103, Waltham, MA).

Experimental Protocol and Conditions

Two protocols were followed in each group of animals in the following order: (1) Using a purpose-built nosecone a series of six switches among 21, 10, and 100% inspired O₂ (balanced order) was performed. The inspired gas was switched rapidly < 1 s, and the new inspirate was sustained for 2-3 min at which time PO₂m had stabilized. Blood samples were taken at the respective inspired O₂s at the end of the protocol. (2) While breathing room air (21% O₂), MAP was lowered and raised by withdrawal and reinfusion of a series of ~ 0.3 ml aliquots of whole blood from the arterial cannula every 4-5 min.

Phosphorescence Quenching

Theory. The oxygen dependence of phosphorescence is described by the Stern-Volmer equation (17):

where T₀ and T are the phosphorescence lifetimes in the absence of oxygen and at an oxygen pressure PO₂m. The quenching constant k_Q is a second-order rate constant related to the frequency of collisions between O₂ and the excited triplet state of the porphyrin and the probability of energy transfer when collisions occur. PO₂m is calculated as

where T₀ and T are expressed in µs and k_Q in mm Hg¹ s¹. At 38° C and pH 7.4, k_Q = 409 and T₀ = 601 (18).

Phosphorescence quenching measurement of diaphragm Po₂m. Of the phosphorescent probe, palladium-meso-tetra[4-carboxyphenyl] porphyrin dendrimer (R2), 15-20 mg/kg was infused via the arterial cannula. Oxyphor R2 binds tightly to albumin as evidenced by the demonstration that R2 is essentially completely bound to albumin in solution at a concentration of 0.5% albumin (19). The concentration of albumin in rat serum is > 3 g/dl (i.e., > 6-fold that necessary for complete binding [20]). In addition, R2 at a pH of 7.4 possesses a net negative charge of approximately 14 mV. Both of these properties help to restrict R2 to the intravascular compartment. Diaphragm microvascular PO₂ (PO₂m) was determined using a PMOD 1000 Frequency Domain Phosphorimeter (Oxygen Enterprises, Ltd., Philadelphia, PA) with the common end of the bifurcated light guide placed ~ 2-4 mm caudal to the abdominal aspect of the medial costal region of the diaphragm about two-thirds of the radial distance between the central tendon and the costal margin (Figure 1). The excitation light (524 nm) is focused on an ~ 2 to 3-mm-diameter circle from a distance of 2-4 mm above the muscle surface and samples blood within the microvasculature up to 500 µm deep. The value of PO₂m reflects principally that of capillary blood as this compartment constitutes the majority of intramuscular blood volume (21). The phosphorescence signal (700 nm) was averaged for a 200-ms interval for each PO₂m measurement and the measurements were repeated at 2-s intervals. Within biological systems, the phosphorescent probe R2 is specific for O₂ and over the range of PO₂s found in the muscle microcirculation can resolve PO₂ within << 1 mm Hg.

View larger version (42K): [in this window] [in a new window]   Figure 1.   Schematic illustrating essential features of the preparation designed to measure diaphragm microvascular PO2 (PO2m) in the hamster.

Verification of Emphysema State

Immediately following euthanasia, lung volume was determined by saline displacement during complete immersion (15).

Statistical Analysis

Data are presented as mean ± SE. Two-way repeated measures ANOVA, one-way repeated measures ANOVA, and one-way ANOVA were used as appropriate. When differences were detected, a Student Newman-Keuls post hoc test was used to identify differences within groups and the Bonferroni post hoc test differences between groups. Comparisons between C and E groups were made by Student's t test. Significance was accepted at p  0.05 except in those instances where directional a priori hypotheses were established and a p  0.1 was used.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Data are presented for a total of 10 C (body weight, 135 ± 12 g) and 7 E (body weight, 127 ± 3 g, p > 0.05) animals. As expected, lung displacement weight was significantly higher in E than C animals (C, 1.4 ± 0.1; E, 2.9 ± 0.4 g, p < 0.05) and Pa_O2 breathing room air (21% O₂) lower in E than C animals (Table 1, C, 90 ± 8; E, 48 ± 7 mm Hg, p < 0.05).

Effects of Altered Inspired O₂

Arterial blood gas, acid-base, and cardiorespiratory responses. At each inspired O₂, Pa_O2 was reduced significantly in E compared with C animals (Table 1). In C only, Pa_CO2 was decreased at 10% O₂ and increased at 100% O₂ compared with the 21% O₂ condition. This altered Pa_CO2 in C elevated pH at 10% O₂ and decreased pH at 100% O₂ (both p < 0.05). The responses to altered inspired O₂ in E hamsters were extremely variable. Breathing frequency (f_b) increased with decreasing inspired O₂ and decreased with 100% O₂ in both C and E hamsters (all p < 0.05, Table 2). Hypoxia (10% O₂) reduced heart rate in E, but not C animals compared with 21% O₂ values (p < 0.05). MAP fell significantly in hypoxia compared with normoxic and hyperoxic values for both C and E hamsters but was not different between C and E animals at any inspired O₂. However, in C hamsters 100% O₂ increased MAP above 21% O₂ values (p < 0.05).

Diaphragm Po₂m. At each inspired O₂, diaphragm PO₂m was reduced significantly in E compared with C animals (Figure 2). However, the variability of this response was markedly greater under the 100% O₂ condition. Figure 3 demonstrates that diaphragm PO₂m was decreased in E animals independent of MAP alterations secondary to the inspired gas composition.

View larger version (27K): [in this window] [in a new window]   Figure 2.   Mean (± SE, n = 10 for control and 7 for emphysema) diaphragm microvascular PO2 (PO2m) in control and emphysematous hamsters at 10, 21, and 100% inspired O2. *p < 0.05 for 10 and 21% O2 and p < 0.1 for 100% O2.

View larger version (25K): [in this window] [in a new window]   Figure 3.   Relationship between mean arterial blood pressure and diaphragm microvascular PO2 (PO2m) at 10, 21, and 100% inspired O2. *p < 0.05 for 10, 21, and 100% O2 for control versus emphysema (n = 10 for control and 7 for emphysema).

Effects of Reduced MAP in Normoxia

Breathing normoxia, diaphragm PO₂m fell systematically with decreased MAP for C and E hamsters (Figure 4). Although there was considerable variability of diaphragm PO₂m for both E and C animals at any given MAP, diaphragm PO₂m was consistently lower in E than C animals. From the linear regression analysis, the diaphragm PO₂m difference between C and E was manifested in large part as a reduced intercept with relatively little difference in slope observed (Figure 4).

View larger version (29K): [in this window] [in a new window]   Figure 4.   Relationship between mean arterial blood pressure and diaphragm microvascular PO2 (PO2m) at 21% inspired O2. Y-intercept was significantly lower for emphysema condition.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present investigation has, for the first time, demonstrated that microvascular PO₂ (PO₂m) is lowered systematically in the diaphragm of emphysematous hamsters. This effect is present in normoxia, hypoxia, and hyperoxia and occurs across a broad range of arterial pressures that encompasses severe hypotension. Although it is intuitively obvious that arterial hypoxemia consequent to emphysema should lower diaphragm PO₂m, the calculated microvascular O₂ content in the diaphragm of emphysematous hamsters falls 2-fold more than arterial O₂ content in these animals (Figure 5). In the presence of an unchanged blood flow in the diaphragm of emphysematous compared with control hamsters at rest (11), this finding supports the contention that emphysema elevates diaphragmatic energy requirements (1) without an appropriate increase of O₂ delivery. The consequences of an emphysema-induced reduction of diaphragm PO₂m include a reduced O₂ extraction reserve that will mandate a greater blood flow increase to support elevated ventilatory and thus respiratory muscle demands (e.g., physical activity), and an exacerbated fall in intramyocyte PO₂ and associated metabolic sequellae (i.e., reduced [PCr], elevated [ADP] [23]) that are expected to enhance glycolytic activity and thereby predispose the diaphragm in emphysematous animals to fatigue.

View larger version (17K): [in this window] [in a new window]   Figure 5.   Mean arterial and diaphragm microvascular (PO2m) points for PO2 and hemoglobin (Hb) saturation in control (C) and emphysematous (E) hamsters. O2 contents estimated from Dill-Gomez relationship (35) are arterial: C, 23.6, E, 18.9; Microvascular O2 content: C, 13.9, E, 7.7 ml/dl.

Preparation Characteristics

It has been shown previously that the laparotomized preparation utilized herein does not perturb substantially the cardiovascular or respiratory condition of rats (13). The present data demonstrate that this is true also for the hamster preparation. Specifically, comparison with the data of Sexton and Poole (11) indicate that MAP, HR, Pa_O2, and Pa_CO2 are not altered by either the anesthesia or the open abdomen condition. It should be noted, however, that the arterial pH was somewhat lower (i.e., 7.28 [present data] versus 7.36 [11]) following surgery.

The hamster is adapted to a semifossorial environment (i.e., low inspired O₂, high CO₂) and consequently demonstrates certain structural and functional characteristics that differ from those of other small rodents and larger mammals. Relevant to the present investigation are increased hemoglobin-O₂ affinity compared with animals of similar size (i.e., P₅₀, hamster = 28, rat = 43 mm Hg, [24, 25]), and lower sensitivity to arterial blood gas perturbations (26). Thus, control hamsters at rest elicit mild hypoxemia as seen in the present and previous investigations (11, 26).

Severity of the Emphysema Condition

Elastase-induced emphysema in the hamster is considered to represent a close analog for human panacinar emphysema (8, 14, 16). With this model, the pathological increase in lung compliance peaks between 2 and 12 wk postelastase instillation (16) with all lung volumes augmented within 4 to 6 mo (4, 8, 14). Crucial to the present investigation, both structural and functional diaphragmatic adaptations are manifested within this period (4, 5, 8, 11, 14, 15). The reduced Pa_O2 and increased lung volume (saline displacement weight) compare closely with literature values for elastase-induced emphysema in the hamster and are consistent with moderate-to-severe emphysema (11, 19, 27). Although no formal morphometric measurements of the diaphragm were made in this investigation, the costal region appeared to be thicker and the muscle fiber(s) length from central tendon to costal margin was reduced. This is consistent with previous reports from our laboratory (6) and the work of others (4, 8, 11) in the hamster model of emphysema.

Responses to Hypoxia and Hyperoxia

Control, C. Hypoxia elevated f_b from 71 to 81 breaths/min, on average which reduced Pa_CO2 by 12 mm Hg (Tables 1 and 2). This response is mediated via peripheral chemoreceptor hypoxic sensitivity (29) and also the hypoxia-induced hypotension that is likely to potentiate the ventilatory response (30). Hyperoxia abolishes peripheral chemosensitivity (29), which acted to slow f_b and induced hypercapnia in the hamster (Table 1) as seen in the rat (13, 33, 34). In addition, MAP was increased significantly by the hyperoxic challenge (21% O₂, 95 ± 5, 100% O₂, 110 ± 7 mm Hg).

The responses detailed above help substantiate that the laparotomized hamster preparation necessary for monitoring diaphragm PO₂m preserves substantial physiological homeostasis as judged by regulation of arterial blood gases, MAP, HR, and hypoxic/hyperoxic responses.

Emphysema, E. The principal differences between C and E hamsters were as follows: (1) Hypoxia reduced HR in E but not C animals. (2) In response to hypoxia, f_b did not increase as much in E as in C and consequently, Pa_CO2 tended to be higher in the hypoxic E hamsters. (3) MAP was not elevated significantly by hyperoxia in E as it was in C hamsters. Taken together, these findings suggest that cardiorespiratory responsiveness was not as great in E as in C hamsters.

Diaphragm Po₂m. Under each inspired O₂ condition, diaphragm PO₂m was lower in E than in C hamsters. This indicates that the O₂-to-O₂ ratio was elevated in E. Radiolabeled microsphere measurements of diaphragm blood flow demonstrate that in the resting animal breathing room air, this variable was not altered significantly in E hamsters (11). Thus, the reduced diaphragm PO₂m found herein was due principally to a greater fractional O₂ extraction (i.e., increased O₂-to- O₂ ratio) with a modest contribution from the decreased arterial O₂ content (Figure 5). Assuming an arterial hemoglobin concentration of 18.3 g/dl (11) and using the Dill-Gomez table (35), arterial O₂ contents of 23.6 and 18.9 ml/dl are calculated for C and E hamsters, respectively. This difference widens substantially for microvascular O₂ content for which the respective values are 13.9 and 7.7 ml/dl. This is consistent with an increased diaphragmatic O₂ secondary to an elevated work of breathing (lung and airways resistances), possibly increased ventilation, and a mechanically disadvantaged diaphragm in E.

Hypoxia and hyperoxia are expected to affect diaphragm PO₂m via at least two mechanisms: altered diaphragmatic O₂ consequent to changes in ventilatory muscle work (i.e., increased in hypoxia, decreased in hyperoxia, see f_b in Table 2), and decreased (hypoxia) or increased (hyperoxia) arterial O₂ content. Either of these factors in separatum would lower (hypoxia) or raise (hyperoxia) Po₂m and this is precisely what was found with both factors operant (Figure 2).

It must be considered that the diaphragm exhibits considerable structural and functional heterogeneity both between the costal and crural diaphragms and also within the costal diaphragm (36). In the present investigation, as in many others (e.g., 6, 13, 15, 21, 36, 37), we chose to examine the medial costal diaphragm because this region undergoes the greatest absolute shortening and has the largest muscle mass within the costal diaphragm. Consequently, the medial costal diaphragm is likely to support a substantial portion of the inspiratory effort. Whether other diaphragmatic regions demonstrate similar behavior with respect to PO₂m could not be determined from the present investigation.

MAP and Diaphragm PO₂m

Diaphragm PO₂m in C and in E hamsters decreased in an approximately linear fashion with reductions in MAP (Figure 4) as demonstrated previously in the rat (13). However, at each MAP, diaphragm PO₂m was significantly lower in E versus C hamsters. Systemic hypotension is a potent ventilatory stimulus (31, 32) and will act to increase breathing and elevate diaphragm O₂. If diaphragm vascular conductance does not increase commensurate with this increased O₂, diaphragm PO₂m must fall and this is exactly what was found. At any given MAP, the lower diaphragm PO₂m in E versus C hamsters is again consistent with greater ventilatory muscle demands and elevated O₂.

Pattern of Ventilatory Muscle Recruitment in Emphysema and Chronic Obstructive Pulmonary Disease (COPD)

With progressive hyperinflation, the mechanical efficacy of the diaphragm is impaired which necessitates increased participation of the so-called accessory muscles to sustain the ventilatory effort. Thus, despite elevated neural drive to the diaphragm in patients with COPD (38), inspiratory discharge frequencies also increase for the parasternal intercostals and scalene muscles (39). In support of a greater recruitment and ventilatory participation of accessory muscles in emphysema, an elegant structural and functional analysis of the hamster medial scalene muscle has identified adaptations consistent with elevated recruitment. Specifically, Fournier and Lewis (40) documented elevated inspiratory electromyographic activity of the medial scalene muscle and an altered expression of myosin heavy chains (2A increased, 2X decreased) in concert with reciprocal changes in the proportion of Type IIa (increased) and IIx (decreased) fibers. In addition, oxidative enzyme activity (assessed via succinate dehydrogenase) was elevated 50-63% in all fibers of the medial scalene. From the above, it is apparent that the recruitment of the diaphragm in emphysema is increased, however, in the face of geometric impediment to the generation of useful inspiratory work by the diaphragm, greater recruitment and participation of accessory inspiratory muscles (i.e., parasternal intercostals, scalene) are necessary. The reduced diaphragm PO₂m reported herein is consistent with greater diaphragm recruitment and contractile activity in the absence of commensurately increased O₂ delivery (11).

Pathophysiological Implications of Lowered Diaphragm PO₂m in Emphysema

Blood-tissue O₂ transfer within the diaphragm may be described by Fick's law: O₂ = DO₂m (Pc_O2  Pi_O2), where DO₂m is the effective diffusing capacity for O₂ and Pc_O2 and Pi_O2 are mean capillary and intracellular PO₂s, respectively. The present investigation has demonstrated that diaphragm PO₂m, which is our estimate of Pc_O2, is ~ 35% lower in E compared with C hamsters. Consequently, to achieve a given O₂ (or more reasonably, an elevated O₂) in the E condition, either DO₂m must increase or Pi_O2 decrease or both. There is morphometric evidence that E stimulates capillary neogenesis (6, 19) thereby elevating capillary length and surface area per fiber volume ~ 14% (6), which would be expected to elevate DO₂m to the extent that the increased capillary surface area could be recruited for O₂ exchange. Because this increase is only a modest fraction (less than half) of the fall in PO₂m (and by extension Pc_O2), a fall in Pi_O2 would be expected to occur in order to facilitate the required O₂. At a given O₂, a decreased Pi_O2 would elevate perturbations of [ADP] and [PCr] (23). This, in turn, will act to accelerate glycolysis and utilization of very limited intramuscular glycogen reserves. We speculate that in E, the reduced diaphragm PO₂m may decrease Pi_O2 and exacerbate muscle glycogen depletion thereby contributing to or causing diaphragm fatigue and failure. If this is indeed the case, adaptations of oxidative capacity may prove essential to preserving diaphragm function in the patient with emphysema. Specifically, treatment modalities such as exercise training that elevate skeletal muscle oxidative capacity result in a reduced disturbance of intramyocyte [ADP] and [PCr] for any given O₂ (41). Consequently, the beneficial effects of exercise or specific respiratory muscle training undertaken by patients with COPD may arise in part from: (1) elevated diaphragm oxidative capacity-associated reductions in [ADP] and [PCr] perturbation and an associated decrease of glycogenolysis, and (2) potential improvement of O₂ delivery via alterations of diaphragm capillarity, which may elevate diaphragm O₂ diffusing capacity (DO₂m) and constrain the reduction of diaphragm PO₂m observed herein.