INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It has often been reported that the majority of morbidity associated with acute pancreatitis is the result of the extrapancreatic complications of the disease. Major attention has been focused on the pulmonary system since 60% of deaths from acute pancreatitis occur within the first week of illness and has been associated with acute respiratory failure (1). A number of pulmonary complications has also been observed in patients with acute pancreatitis, including pleural effusion, atelectasis, pneumonia, and adult respiratory distress syndrome (ARDS).

In recent years, research has focused on the occurrence of ARDS, particularly during the initial phase of the disease. There is an improved awareness of the pathophysiology involved in this complication, and significant evidence has emerged that implicates proinflammatory cytokines as the mediators responsible for the escalation of localized pancreatic inflammation into a generalized systemic disease involving the lung (2, 3). Patients with high serum cytokines, for example, were found to have a significantly higher incidence of pulmonary, renal, and cardiovascular complications that were associated with longer hospital stays and overall higher mortality rates. On the other hand, basal atelectasis and cephalad shift of the diaphragm are frequently observed in patients with acute pancreatitis. This may be secondary to ascites raising the diaphragm and/or effusion in the pleural cavity causing secondary lung collapse. Alteration in diaphragmatic function in acute pancreatitis could also be a possible mechanism that may be responsible for its cephalad displacement and occurrence of atelectasis. Indeed, muscle function, including respiratory muscles, have been shown to be impaired during generalized sepsis (4, 5). Recently, diaphragmatic alteration has also been reported in an edematous pancreatitis rat model (6). However, it is not known if this dysfunction is related to a generalized process involving all the muscles (i.e., diaphragm and peripheral muscles) or to an abdominal process thus sparing peripheral muscle function. Therefore, the aim of the present study was to assess diaphragmatic and peripheral muscle function, i.e., Extensor Digitorum Longus (EDL) and Soleus (SOL), in a rat model of taurocholate-induced acute hemorrhagic necrotizing pancreatitis in order to determine if both these groups of muscles could simultaneously been altered in severe acute pancreatitis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Model

The experimental design was carried out according to the recommendations of the French Law (Ministère des Affaires Sociales et de la Solidarité Nationale et Ministère de l'Agriculture), and all experiments were approved by the local Animal Ethics Committee.

Sixty-eight male pathogen-free Sprague-Dawley rats weighing 180 to 200 g (Charles River Elbeuf, St. Aubin, France) were used in this experiment. All rats were housed individually and acclimatized to a 12-h light-dark cycle. They were maintained on Purina rat chow and tap water for a 3-d period before experimental setup and fasted overnight before the experiment with water ad libitum. The animals were laparotomized in the midline under intraperitoneal sodium pentobarbital (50 mg/kg; Sanofi, Gentilly, France) anesthesia. The duodenum and the common biliary-pancreatic duct were identified. The common biliary-pancreatic duct was catheterized transduodenally with a 24 gauge teflon catheter (Critikon, Issy-les-Moulineaux, France) and a 6.0 silk ligature was tightened around the catheter and the wall of the duct to prevent misdirected back flow to the duodenum during infusion. The animals were then placed in a proclive position (60 degrees) for 5 min. During the last 2 min of the surgical procedure, the hepatic duct was closed with a small bulldog clamp applied near the hilus of the liver. During intraductal infusion this clamp remained in place to prevent posterograde infusion into the liver.

The animals were randomly allocated into two groups. In the first group (n = 34), acute hemorrhagic pancreatitis was induced by infusion of 0.5% sodium taurocholate (0.2 ml/100 mg; Sigma Aldrich Chimie, St. Quentin, France) into the pancreatic duct. Perfusion was performed over a 10-min period at a pressure of 30 cm H₂O, which was controlled using a water manometer. In the second group (n = 34), isotonic saline was infused instead of sodium-taurocholate solution (control group). The catheter was removed and the duodenal wall was closed with a 6.0 prolene suture and the abdomen with a 4.0 nylon suture. Four milliliters of isotonic saline solution were injected intraperitoneally to compensate for the fluid loss associated with the surgical procedure. The animals were then placed back in cages with minimal environmental stimuli and allowed to recover from anesthesia. Animals were starved for 24 h with free water access.

Muscle Contractile Properties

Assessment of muscular function was performed 24 h after the surgical procedure. Each group (n = 30) was divided in three subgroups according to the muscle being studied: diaphragm, SOL, or EDL.

Diaphragm muscle preparation. The animals were anesthetized intraperitoneally with sodium pentobarbital (50 mg/kg). The thoracic and abdominal cavities were opened immediately, and the animals were exsanguinated via the abdominal aorta. The diaphragm, with its origins and insertion on the ribs and central tendon left intact, was quickly excised and transferred to a dissecting dish containing oxygenated Kreb's solution (composition in mmol: sodium chloride, 137; potassium chloride, 4; magnesium chloride, 1; potassium phosphate, 1; sodium bicarbonate, 12; calcium chloride, 2; and glucose, 6.5) where it was pinned to maintain a resting length during dissection. The Kreb's solution was frequently changed. Muscular strips from the lateral costal region of each hemidiaphragm were dissected. This region was chosen because it has previously been demonstrated to contain parallel fiber layers and to be composed by equally distributed fiber types (7). A strip from the middle part of the lateral costal region of the diaphragm (approximately 2 mm wide) was microscopically dissected with fibers attached to a portion of the ribs and distally to the central tendon. The end tendon flap was left intact, whereas the other end was cut free of the ribs and ligated with fine copper wire (20 µm diameter) for use in mounting the preparation.

Peripheral muscles. Soleus or EDL muscles were removed from the spontaneously breathing, intraperitoneally anesthetized animal (sodium pentobarbital 50 mg/kg) and transferred to a dissecting dish as described above. A strip (approximately 4 to 5 mm wide) was microscopically dissected and prepared in the same manner as for the diaphragmatic section.

Contractile function. The muscle bundles were mounted horizontally into an open-topped Plexiglas tissue chamber. They were immobilized by snaring the tendon with fine surgical silk into one end of the channel, and by connecting the other extremity of the bundle by means of a 20-µm copper wire, to the tip of the force transducer. The experimental chamber and force transducer were mounted on an antivibration table (Harvard Apparatus, South Natick, MA). The preparation was perfused with continually flowing Kreb's solution, which prevented alteration in salinity and permitted rapid elimination of toxic metabolites released from the muscles. The solution was bubbled with a mixture of 95% O₂, 5% CO₂. Temperature and pH were controlled (37° C and 7.40, respectively) and were monitored continuously during the experiment.

The preparation was electrically stimulated by means of rectangular platinium field electrodes mounted on both sides of the channel parallel to the tissue strip. Stimuli of 0.5 ms. duration were applied using a Grass S 48 stimulator (Grass Instruments, Quincy, MA). Stimulus intensity was increased until maximal twitch tension responses were obtained and set at 1.2 times the maximal to ensure supramaximal stimulation.

Isometric force was measured with a Grass FT 03 force transducer (Grass Instruments) mounted on a micrometer which allowed adjustment of the preparation to optimal length (L_o), i.e., the length at which peak twitch force was maximal. The signal from the force transducer was amplified and displayed simultaneously on a Gould TA 555 paper recorder (Gould Electronics, Cleveland, OH) and a Tektronix NS III storage oscilloscope (Tektronix, Beaverton, OR).

The force developed in response to twitch stimulation (Pt) along with the time to peak tension (TTP) and half relaxation time (1/2 RT) were calculated for a minimum of five twitches. The force-frequency relationship was studied using stimuli of 10, 20, 30, 50, and 100 Hz applied for 2 s. A 1-min rest period was allowed between each stimulus.

Endurance capacity of the muscle was assessed by recording the decrease in peak tension to repetitive (90/min) tetanic stimulation (30 Hz, 300 ms duration) for 3 min. The time until tension fell to 50% of the initial value (T50%) was used as an index of diaphragmatic endurance.

Breathing pattern and blood gas measurements. Four control animals and four with pancreatitis were studied before (T0) and 24 h (T24) after laparotomy. Respiratory rate (F), inspiration (TI) and expiration (TE) time were recorded using a noninvasive bias flow ventilated whole-body plethysmographic technique, and the TI/Ttot ratio was calculated (8). Precautions were taken to limit stress, which may alter breathing pattern. Awake unrestrained rats were therefore placed into the plethysmographic box several times before carrying out the experiment. Arterial blood gas analyses were performed through a silastic catheter (Plastimed, Saint-Leu-La-Forêt, France) implanted into the carotid under anesthesia 6 h before experiment, and washed out frequently with saline solution in order to keep it permeable. Arterial blood gas determinations for Pa_O2, Pa_CO2, and pH analysis was made in ambient air at T0 and at T24.

Histolopathologic examination. In 10 control animals and 10 animals with acute necrotizing pancreatitis the costal and the crural areas of the hemidiaphragm not used for contractile analysis were removed immediately after exsanguination. The muscle was covered with cryogel and frozen by immersion into isopentane cooled with liquid nitrogen and then stored at 80° C until examination. Serial sections 10 µm thick were cut on a cryostat after the samples that had been warmed to 18° C and subsequently stained with hematoxylin-eosin for histologic examination.

In the same animals, the entire pancreas was immediately removed and fixed in 10% formalin on anatomic orientation for histologic analysis. After dividing the organ into a duodenal and splenic part and embedding each in paraffin, one longitudinal section through each of the two specimens was stained with hematoxylin-eosin.

Bacteriologic examination. Bacteriologic examinations were performed in five animals in both groups undergoing peripheral muscle examination. Blood samples were obtained by aseptic abdominal aorta-puncture. Semiquantitative blood cultures were performed by placing 0.1 ml of fresh blood onto agar plates for bacterial counts.

Ten milliliters of cold phosphate-buffered saline (PBS; Gibco BRL, Cergy Pontoise, France) were injected intraperitoneally. A midline laparotomy was performed and peritoneal fluid was recovered. The peritoneal fluid was spread on agar plates for numeration of developed colonies.

Statistical Analysis

Values are given as mean ± SD. A repeated-measures analysis of variance with one repeated factor and one grouping factor was used to examine differences between groups in twitch, force frequency, and fatigue responses. Differences were considered significant at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pancreatitis Model

No deaths occurred before animals were killed (24 h). Animals that had received a taurocholate solution developed significant cephalad acute pancreatitis after histologic examination of the entire pancreas. In fact, the pancreas was swollen and had both smooth appearance and a gelatinous consistency. In addition, multiple petechial hemorrhages were subperitoneally observed. Histologic foci of massive necrosis, extensive interstitial edema, and, in some places, infiltrations with inflammatory cells were also detected. In contrast, saline control animals showed no macroscopic alteration of the pancreas. In a section from minimal focal edema, acinar architecture was entirely normal. No pathogen was cultured from blood or peritoneal samples from either the saline or the pancreatitis animals. With regard to ventilatory pattern, there was no significant difference in respiratory rate, inspiratory time and TI/Ttot ratio between the control and the pancreatitis groups at T0 and T24 (Table 1). Arterial blood gas values were in the normal range in the two groups.

Muscle Function

The weight, length at the study (at L_o), and cross-sectional area of the diaphragm and the peripheral muscle preparations were similar in both groups. However, diaphragmatic peak twitch force was significantly lower in the taurocholate group than in the saline group (p < 0.05). No significant change in time-to-peak tension or half relaxation time was observed (Table 2).

Typical diaphragmatic force-frequency curves and endurance tests from one taurocholate and one saline animal are depicted in Figures 1 and 2, respectively. Analysis of diaphragmatic force-frequency relationships showed that force generated in response to all stimulation frequencies was lower in the taurocholate group than in the saline group (Figure 3). In addition, decrease in endurance capacity was noticed in the pancreatitis group when compared with the saline group, as indicated by a significant reduction in the T50% (31.0 ± 10.5 versus 49.0 ± 10.3 s, respectively, p < 0.05).

View larger version (27K): [in this window] [in a new window]   Figure 1.   Typical diaphragmatic force-frequency curves from a control animal (left panel ) and a pancreatitis animal (right panel ).

View larger version (62K): [in this window] [in a new window]   Figure 2.   Typical tracing showing changes in diaphragmatic strength generation during an endurance test in a control animal (top panel ) and in a pancreatitis animal (bottom panel ). Endurance capacity was assessed by the time until tension fell to 50% of the initial value ( T50%).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Diaphragmatic force-frequency curves of the control group (open circles) (n = 10) and the acute necrotizing pancreatitis group (closed circles) (n = 10). Acute pancreatitis was associated with a significant reduction in diaphragmatic strength for all frequencies of stimulations. Values are mean ± SD. *p < 0.01 versus control.

Data regarding SOL and EDL muscle strength and endurance capacity from saline animals and animals with pancreatitis are shown in Table 3 and Figure 4. Twitch parameters, force-frequency curve, and endurance index were similar in both groups.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Force-frequency curves for Extensor Digitorum Longus (EDL) on the left panel and Soleus (SOL) on the right panel of the control group (open circles) and the pancreatitis group (closed circles). Values are mean ± SD. No significant change between the two groups was noticed for EDL and SOL.

Diaphragmatic Histology

The diaphragm in both groups exhibited a normal appearence on gross examination. No edema, infiltration, necrosis, or degenerating process was observed on histologic examination in the diaphragm of pancreatitis animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of our study demonstrate that acute necrotizing pancreatitis induces marked alteration in diaphragmatic function, whereas there was no change in peripheral muscle contractility.

Respiratory failure is a major complication of acute necrotizing pancreatitis and has been traditionally related to ARDS since the early 1970s (9). Recent experimental data suggest that ARDS is not the only cause of respiratory failure during Systemic Inflammatory Response Syndrome (SIRS) (10, 11). In fact, a large body of experimental evidence has demonstrated failure of the respiratory muscles during SIRS and recently during pancreatitis (6). Because maintenance of ventilation depends on the ability of the respiratory muscles to generate force, dysfunction of these muscles, and particularly of the diaphragm, during acute pancreatitis can lead to respiratory failure, which may be life-threatening.

Various experimental models have demonstrated that generalized sepsis induced either by Escherichia coli endotoxin or by Streptococcus pneumoniae injection was associated with deleterious effects on respiratory and peripheral muscles. Indeed, using a pneumococcal infection model in rats, Boczkowski and colleagues (4) and Ruff and Secrist (5) reported a reduction in diaphragmatic force and in limb muscle strength, respectively. Muscular impairment during sepsis is thought to be located at the area of the muscle contractile machinery, but the precise mechanisms responsible for the decreased muscle strength are still under investigation (12). Nevertheless, energetic impairment appears to be an important physiologic determinant of the muscle dysfunction. Hypotension, by reducing muscle metabolic substrate delivery, may play a major role as initial determinant of muscle dysfunction (10, 13, 14) and this mechanism should also be considered in severe acute pancreatitis. However, an energy substrate depletion and/or hypotension had been responsible for the decrease in diaphragmatic strength, and a decrease in the force of peripheral muscles should have also been observed. In the present study, acute necrotizing pancreatitis induced a marked alteration in diaphragmatic function, whereas no change in peripheral muscle function was observed. The discrepancy between these two groups of muscles strongly suggests that hemodynamic changes are not responsible for the severe diaphragmatic alteration. On the other hand, muscle fatigue may occur in pancreatitis animals if an overload or overactivity is imposed to the respiratory system because of acute lung injury development and reduction in pulmonary compliance. With regard to respiratory pattern, no significant difference was observed in either group. In addition, different conditions that may accentuate muscle dysfunction or increase the susceptibility to injury such as hypoxemia (15) and hypercapnia (16) were not observed in our model. Meanwhile, arterial blood gas values in both groups were in the normal range for animals in resting condition. These results suggest that an increase in the work of breathing induced by a severe alveolar intertitial damage could probably not account as a major factor contributing to the acute diaphragmatic dysfunction seen in this pancreatitis model.

Another pathway for muscular metabolic alteration during SIRS may also be triggered by systemic release of mediators such as prostaglandins, oxygen-reactive metabolites, and TNF, acting alone or in concert (17, 18). Indeed, acute pancreatitis is also a well-documented model of systemic inflammatory cytokine production (2, 3), which may cause alterations in muscle function. However, although a dramatic decrease in diaphragmatic function was observed in our model, limb muscle function remained unchanged. One explanation for the fact that peripheral muscles were spared in the present study is that a local, i.e., abdominal, rather than systemic process may be responsable for the functional difference. Several lines of evidence suggest that cytokines, mainly TNF and IL-1, could lead to skeletal muscle dysfunction during sepsis by increasing nitric oxide production, which has the ability to depress myofiber contractility directly (12). However, decreased muscular strength generation is achieved only when muscle is exposed at cytokine concentrations much higher than those reported in the serum in sepsis. TNF at concentration of 50 to 400 ng/ml did not impair muscle force generation (19), whereas a concentration of 10 µg/ml of the cytokine elicited a significant decrease in the force in the hamster (18). During acute necrotizing pancreatitis, because of the massive cytokine release from the necrotizing gland in the abdominal cavity, diaphragm is exposed to cytokine concentrations that reach severalfold higher than peripheral muscles. The high peritoneal cytokine concentration may explain why diaphragmatic function is severly impaired, whereas peripheral muscles function is spared. Recently, in a model of acute peritonitis and systemic sepsis induced by LPS, Lin and coworkers (20) reported a higher degree of sarcolemmal damage in the diaphragm than in the SOL. These investigators suggested that sarcolemmal lesions may be related to additive or even synergistic effects of contraction- induced mechanical stress (21) and sepsis-induced muscular injury (20). Although the precise role of sepsis and mechanical stress in muscular damage generation is not clearly established (20), it may be hypothesized that severe diaphragmatic sarcolemmal lesions may be involved, at least in part, in the dramatic pancreatitis-induced diaphragmatic dysfunction. Interestingly, sarcolemmal damage during sepsis is not associated with necrotic changes or inflammatory cell inflitration at the muscular level (20), a finding that supports the fact that major muscular impairment may develop in spite of normal microscopic histologic appearance, as observed in the present study.

We conclude that acute necrotizing pancreatitis in the rats is associated with an early and marked diaphragmatic dysfunction. The mechanisms by which this alteration occurs are not clearly known and may involve, at least in part, the pathway of increased nitric oxide production via mediators released by the pancreatic gland in the vicinity of the diaphragm and lesions induced by mechanical stress related to the ventilation process.

Our data support the hypothesis that alterations in the respiratory pump are contributing factors to the genesis of acute respiratory failure during acute necrotizing pancreatitis. Reduced diaphragmatic motion may also explain the cephalad ascension of the diaphragm and atelectasis that occur in the caudal-dependent parts of the lung in patients with severe acute pancreatitis.