This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Letters to the Editor: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Letters to the Editor are posted
Right arrow Alert me if a correction is posted
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow Rights and Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by ELMARAGHY, A. W.
Right arrow Articles by SCHEMITSCH, E. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by ELMARAGHY, A. W.
Right arrow Articles by SCHEMITSCH, E. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Facebook   Add to Technorati   Add to Twitter  
What's this?
The Journal of Bone and Joint Surgery 81:1155-64 (1999)
© 1999 The Journal of Bone and Joint Surgery, Inc.

Pathophysiological Effect of Fat Embolism in a Canine Model of Pulmonary Contusion*

AMR W. ELMARAGHY, M.D.{dagger}, SERGEI AKSENOV, M.D.{dagger}, ROBERT J. BYRICK, M.D., F.R.C.P.(C){dagger}, ROBIN R. RICHARDS, M.D., F.R.C.S.(C){dagger} and EMIL H. SCHEMITSCH, M.D., F.R.C.S.(C){dagger}, TORONTO, ONTARIO, CANADA

Investigation performed at the Musculoskeletal Research Laboratory, Division of Orthopaedic Surgery, Department of Surgery, and the Department of Anaesthesia, St. Michael's Hospital, University of Toronto, Toronto


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: The objective of this study was to determine the individual and combined effects of pulmonary contusion and fat embolism on the hemodynamics and pulmonary pathophysiology in a canine model of acute traumatic pulmonary injury. Methods: After a thoracotomy, twenty-one skeletally mature dogs were randomly assigned to one of three groups. Unilateral pulmonary contusion alone was produced in Group 1 (seven dogs); pulmonary contusion and fat embolism, in Group 2 (seven dogs); and fat embolism alone, in Group 3 (seven dogs). Pulmonary contusion was produced by standardized compression of the left lung with a piezoelectric force transducer. Fat embolism was produced by femoral and tibial reaming followed by pressurization of the intramedullary canals. Cardiac output, systolic blood pressure, peak airway pressure, pulmonary arterial pressure, pulmonary capillary wedge pressure, partial pressure of arterial oxygen, and partial pressure of carbon dioxide were monitored for all groups. From these data, several outcome parameters were calculated: total thoracic compliance, alveolar-arterial oxygen gradient, and ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration. All of the dogs were killed after eight hours, and tissue samples were obtained from the brain, kidneys, and lungs for histological analysis. Lung samples were assigned scores for pulmonary edema (the presence of fluid in the alveoli) and inflammation (the presence of neutrophils or hyaline membranes, or both). The percentage of the total area occupied by fat was determined. Results: Pulmonary contusion alone caused a significant increase in the alveolar-arterial oxygen gradient but only after seven hours (p = 0.034). Fat embolism alone caused a significant transient decrease in systolic blood pressure (p = 0.001) and a significant transient increase in pulmonary arterial pressure (p = 0.01) and pulmonary capillary wedge pressure (p = 0.015). Fat embolism alone also caused a significant sustained decrease in the ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration (p = 0.0001) and a significant increase in the alveolar-arterial oxygen gradient (p = 0.0001). The combination of pulmonary contusion and fat embolism caused a significant transient increase in pulmonary capillary wedge pressure (p = 0.0013) as well as a significant sustained decrease in partial pressure of arterial oxygen (p = 0.0001) and a significant decrease in systolic blood pressure (p = 0.001) that lasted for an hour. Pulmonary contusion followed by fat embolism caused a significant increase in peak airway pressure (p = 0.015), alveolar-arterial oxygen gradient (p = 0.0001), and pulmonary arterial pressure (p = 0.01), and these effects persisted for five hours. Total thoracic compliance was decreased 6.4 percent by pulmonary contusion alone, 4.6 percent by fat embolism alone, and 23.5 percent by pulmonary contusion followed by fat embolism. The ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration was decreased 23.7 percent by pulmonary contusion alone, 52.3 percent by fat embolism alone, and 65.8 percent by pulmonary contusion followed by fat embolism. The mean pulmonary edema score was significantly higher with the combined injury than with either injury alone (p = 0.0001). None of the samples from the lungs demonstrated inflammation. Fat embolism combined with pulmonary contusion resulted in a significantly greater mean percentage of the area occupied by fat in the noncontused right lung than in the contused left lung (p = 0.001); however, no significant difference between the right and left lungs could be detected with fat embolism alone. The mean percentage of the glomerular and cerebral areas occupied by fat was greater with fat embolism combined with pulmonary contusion than with fat embolism alone (p = 0.0001 and p = 0.01, respectively). Conclusions: The combination of pulmonary contusion and fat embolism leads to more substantial pulmonary dysfunction than does either form of injury alone. The histological results suggest that the early effects seen following the combination of pulmonary contusion and fat embolism are mediated not by inflammatory changes but by redistribution of pulmonary perfusion by mechanical mechanisms. Clinical Relevance: The redistribution of pulmonary perfusion that occurs as a result of pulmonary contusion may potentiate both the pulmonary and the systemic complications associated with fat emboli if intramedullary nailing adds an embolic load to the venous circulation. The optimum timing and method of fracture fixation may depend not only on the time elapsed since the injury but also on the severity and stage of the associated pulmonary contusion.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Trauma is the leading cause of death in the first four decades of life and is exceeded only by cancer and atherosclerosis as the major cause of death in all age-groups1. A large proportion of traumatic deaths is attributable to thoracic trauma17. Furthermore, the rate of disability from injury exceeds that of mortality by a ratio of three to one, and since the victims of trauma are primarily young and in their most productive years, trauma exacts a great toll in terms of societal cost and human suffering1,4.

Victims of trauma may sustain a pulmonary contusion or fat embolism, or both; each has a detrimental effect on pulmonary pathophysiology33. Pulmonary contusion is a common finding among individuals who have multiple injuries32. Although hypoxia is the most commonly observed early consequence8, pulmonary contusion can produce long-term sequelae, such as pulmonary fibrosis and decreased functional residual capacity16. Pulmonary contusion also places the patient at increased risk for sepsis and for adult respiratory distress syndrome, which has a reported mortality rate of as high as 51 percent (forty-seven of ninety-two patients)27,28.

Victims of multiple trauma often also have fractures of the long bones of the extremities, which can lead to fat embolism as a result of the fracture itself or of its treatment with intramedullary nailing18,34. In addition to immediate pulmonary dysfunction from hypoxia, these patients are at risk for fat-embolism syndrome, which is potentially life-threatening15,19,25,26,30. The optimum time to fix fractures of long bones so that it does not aggravate pulmonary dysfunction in patients who have concomitant pulmonary injury remains controversial3,13,18,22. Some authors have suggested that, in patients who have associated pulmonary contusion, it may be preferable to avoid traditional intramedullary fixation with reaming until pulmonary function improves18. In order to critically evaluate the method and timing of intervention aimed at preventing or ameliorating severe pulmonary dysfunction, it is crucial to understand how fat embolism after pulmonary injury affects pulmonary function. The purpose of the present study was to determine the individual and combined effects of pulmonary contusion and fat embolism on the hemodynamics and pulmonary pathophysiology in a canine model of acute traumatic pulmonary injury.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Twenty-one skeletally mature mongrel dogs (twelve female and nine male) that weighed a mean of 27.4 kilograms (range, 23.8 to 32.0 kilograms) were used in this study. All procedures were approved by our institutional Animal Care Committee and were performed in an operating room with aseptic technique. The animals were premedicated with 0.05 milligram of acepromazine per kilogram of body weight and 0.6 milligram of atropine per kilogram of body weight. The left side of the neck and chest and the left hindlimb were shaved and were cleaned sequentially with 4 percent chlorhexidine gluconate soap (Hibitane; Ayerst Laboratories, Montreal, Quebec, Canada), 70 percent isopropyl alcohol, and 10 percent povidone-iodine (Betadine; Purdue Frederick, Pickering, Ontario, Canada). Anesthesia was induced with an intravenous injection of twenty-five milligrams of thiopental sodium per kilogram of body weight and was maintained after endotracheal intubation on a Ventimeter (Air-Shields, Halboro, Pennsylvania) with a mixture of nitrous oxide (48.5 percent), halothane (1.5 percent), and oxygen (50.0 percent), at a respiratory rate of twenty breaths per minute. For each dog, the ventilator was set to deliver a tidal volume of fifty milliliters (volume of the system) plus ten milliliters per kilogram of body weight. One and a half milligrams of oxymorphone was administered after induction as a narcotic analgesic. At this point, anteroposterior and lateral radiographs of the chest were made to ensure acceptable placement of the endotracheal tube and to serve as a reference for subsequent radiographs.

The operative procedure was performed with the animal in right lateral recumbency. Before the operation, an 18-gauge catheter (Arrow, Reading, Pennsylvania) was inserted in the right femoral artery and a 9.0-French catheter (Arrow) was inserted in the left internal jugular vein. A 7.5-French Swan-Ganz catheter (Baxter, Irvine, California) was introduced into the internal jugular vein and was advanced into the pulmonary artery. Its position was confirmed with image intensification. The heart rate as well as systemic and pulmonary pressures were monitored and recorded on Perisoft software (Perimed, Jarfalla, Sweden) and a Samsung personal computer (Seoul, South Korea). A urinary catheter was inserted and was connected to a closed gravity-drainage system. An arterial blood-gas sample from each dog was analyzed before the operation, and the respiratory rate was modified as necessary to ensure a partial pressure of carbon dioxide that was between twenty-five and thirty-five millimeters of mercury (3.33 and 4.67 kilopascals). Once the respiratory rate had been established with acceptable values for the partial pressure of carbon dioxide, baseline measurements were made and the experiment timer began at zero minutes.

As they were acquired, the twenty-one dogs were sequentially assigned a number of one, two, or three, and the operative protocol corresponding to that number was performed. This randomized the dogs effectively into three groups of seven each while preventing the results for any group from being altered by an operative learning curve. All animals had a thoracotomy starting at zero minutes. Starting at thirty minutes, pulmonary contusion was produced in the dogs in Groups 1 and 2, and the thoracotomy was closed by sixty minutes in all groups. Pulmonary fat embolism was produced in the dogs in Groups 2 and 3 starting at 120 minutes, with completion by 180 minutes. All dogs were monitored from 181 minutes until they were killed at 480 minutes, at which time a postmortem examination was performed. In order to rule out substantial pneumothorax or hemothorax, radiographs of the chest were made for all animals at 120 minutes after the thoracotomy and at 480 minutes, just before the animals were killed. The dogs in all three groups were treated similarly with respect to intravascular volume replacement and the administration of anesthetics.

Core body temperature and urine output were measured at zero, sixty, 120, 185, 240, and 480 minutes. Cardiac output, systolic blood pressure, peak airway pressure, pulmonary arterial pressure, pulmonary capillary wedge pressure, and partial pressures of arterial oxygen and carbon dioxide were determined at each time-point (zero, thirty, sixty, 120, 181, 185, 195, 210, 240, and 480 minutes). A radiograph of the chest was made at zero, 120, and 480 minutes. A gauge on the ventilator displayed the peak airway pressure, and a model-5120 oxygen monitor (Ohmeda, Madison, Wisconsin) in the ventilation circuit continuously measured the fraction of inspired oxygen. Arterial blood samples were analyzed with the automated 287 Blood Gas System (CIBA-Corning, Medfield, Massachusetts) to determine the partial pressures of arterial oxygen and carbon dioxide. Systolic blood pressure was monitored through the catheter in the femoral artery. Core body temperature, pulmonary arterial pressure, pulmonary capillary wedge pressure, and cardiac output (measured by thermodilution with the COM-2 cardiac output computer [Baxter]) were measured through the catheter in the pulmonary artery.

From these measured values, several outcome parameters were calculated at each time-point for each dog. These included total thoracic compliance (tidal volume divided by peak airway pressure), ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration, and alveolar-arterial oxygen gradient (the difference between the calculated alveolar oxygen tension and the measured arterial oxygen tension). Alveolar oxygen tension was determined according to the formula: FiO2 x (atmospheric pressure - vapor pressure of water) - PaCO2/R (where FiO2 is the fractional inspired oxygen concentration, PaCO2 is the partial pressure of arterial carbon dioxide, and R is the respiratory quotient with a value of 0.8).

The thoracotomy was performed through the sixth intercostal space on the left with resection of the seventh rib. The parietal pleura was entered, and rib spreaders were used to expose the left lung. After approximation of the ribs with a single heavy suture, the thoracotomy was closed by repair of the parietal pleura along with adjacent intercostal muscle in a continuous fashion. The final purse-string suture in the pleura was tightened only after evacuation of the pleural space with a large (twice the tidal volume) ventilated breath. The superficial muscle layer was then approximated, and the skin was closed with running sutures. A chest tube was not needed for any animal because there was no evidence of substantial pneumothorax and positive pressure ventilation was maintained.

Pulmonary contusion was produced after the thoracotomy in Groups 1 and 2. Five standardized locations in the left lung were inserted between two aluminum discs twenty square centimeters in area and were compressed over a piezoelectric force transducer (Kistler, Amherst, New York) with a C-clamp. The force transducer was connected to an amplifier (model 5004; Kistler) and then to the computer through an A/D converter (Perimed). The transducer had been previously calibrated at 1.0 to 1.5 kilograms per square centimeter, which was determined to require a force range of 200 to 300 newtons. This level of force was maintained for twenty seconds.

In Groups 2 and 3, fat embolism was produced by femoral and tibial reaming followed by pressurization of the intramedullary canals29. A lateral parapatellar approach to the left knee joint was made. A starting hole was made with a 9.0-millimeter drill-bit approximately four millimeters anterior to the insertion of the medial cruciate ligament, and the femur was reamed in a retrograde fashion with sequential use of cannulated flexible reamers (Precision; Howmedica, Rutherford, New Jersey) from 6.0 to 9.0 millimeters in size. After a 6.0-millimeter starting hole was made, the tibia was reamed in an antegrade fashion with a 6.0-millimeter reamer. Blood and reaming debris were cleared from the joint with gentle lavage with saline solution, and a custom-fit resin canal restrictor was inserted into the femur and one was inserted into the tibia. Methylmethacrylate cement (Simplex P; Howmedica) was introduced into each medullary canal with a twenty-milliliter syringe, ensuring that the tip was press-fit into the starting hole. Immediately after its introduction, the cement was pressurized by the insertion of a contoured Steinmann pin, five millimeters wide and 150 millimeters long, into the canal. The knee joint and the skin were then closed with running sutures.

At 480 minutes, all animals were killed with an overdose of pentobarbital sodium (340 milligrams per milliliter solution) and a postmortem examination was performed. The lungs, heart, kidneys, and brain were removed. The heart was examined for evidence of a patent foramen ovale. The lungs were inflated with 100 percent oxygen to a pressure of fifteen centimeters of water and were stored with the kidneys and brain in 10 percent neutral buffered formalin. To maintain inflation of the lungs, an umbilical clamp was applied to the trachea. Ten samples were taken from the lungs of each dog: two each, from the right superior lobe, the right middle lobe, the right inferior lobe, the left superior lobe, and the left inferior lobe. Eight samples were taken from the brain of each dog: two each, from the basal ganglia, the brain stem, and the frontal and occipital lobes. Ten samples were taken from the kidneys of each dog: five each, from the right and the left kidney. Histological sections of all samples were stained for fat with osmium tetroxide (BCN Chemicals, Beaconsfield, Quebec, Canada). Five-micrometer sections were also stained with hematoxylin and eosin.

Histomorphometric analysis was performed with a computer-aided image analyzer (model 3001; Leco, Mississauga, Ontario, Canada). For each dog, two fields per section of lung (for a total of twenty) at a magnification of 100 times were analyzed for pulmonary edema and were given a score of 0 points (no fluid in any alveolus), 1 point (fluid in less than 50 percent of the alveoli), or 2 points (fluid in more than 50 percent of the alveoli)29. The same number of lung fields per animal were analyzed, at a magnification of 500 times, for neutrophils or hyaline membranes, or both, and were given a similar composite score of 0, 1, or 2 points. For each dog, twenty fields per section of lung (for a total of 200) at a magnification of 100 times were used to calculate the percentage of the area that was occupied by fat relative to the area that was occupied by lung parenchyma29. Similarly, twenty fields per section of brain (for a total of 160) at a magnification of 500 times were used to calculate the percentage of the area that was occupied by fat relative to the area that was occupied by brain parenchyma, and twenty fields per section of kidney (for a total of 200) at a magnification of 200 times were used to calculate the percentage of the area that was occupied by fat relative to the area that was occupied by the glomerulus29.

Statistical Analysis
Data are reported as the mean and the standard error of the mean. Analysis of variance was performed to determine whether there were any differences among the means at the different time-points within each group or among the groups at each time-point. If appropriate, post-testing was performed with the Tukey-Kramer multiple comparisons test. All statistical analysis of data was performed with Instat2 software (GraphPad, London, Ontario, Canada).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 

Complications and Postmortem Observations
No dog had evidence of pneumothorax on the radiographs of the chest or of a patent foramen ovale at the postmortem examination25.

There were three intraoperative deaths; two were in Group 2 and one was in Group 3. All three deaths occurred within minutes after the pressurization of the intramedullary canals and the production of fat embolism and were associated with profound systemic hypotension, pulmonary hypertension, desaturation, and finally cardiac arrest. Post mortem, there were substantial gross fatty changes in the kidneys, lungs, and coronary circulation. Increased right ventricular afterload resulting from increased pulmonary vascular resistance, combined with coronary fat occlusion, probably produced an immediate fatal cardiac event in these dogs. The data available for these dogs were examined, and the direction and magnitude of change in the outcome parameters were found to be similar to those of the rest of the dogs in their groups. Given that these results were not aberrations or statistical outliers, we included the available data for the dogs that died in the statistical analysis of their respective groups.

Outcome Parameters

Hemodynamic Findings
Temperature and urine output: Over the eight-hour course of this experiment, the temperature of the dogs decreased a mean of 1.9 degrees Celsius. The mean urine output was seventy-two milliliters per hour. We detected no significant difference among the groups with regard to temperature or urine output, with the numbers available.

Cardiac output: Thoracotomy, pulmonary contusion, and fat embolism alone or in combination were not found to have a significant effect on cardiac output compared with the baseline values. We also did not detect a significant difference in cardiac output over time or among the groups at any time-point in the experiment.

Systolic blood pressure: Neither thoracotomy nor pulmonary contusion alone (Group 1) was found to significantly affect systolic blood pressure (Fig. 1), with the numbers available. Fat embolism alone (Group 3) caused a significant decrease in systolic blood pressure at 181 minutes (p = 0.001), with the blood pressure returning to baseline values within five minutes (Fig. 1). The combination of pulmonary contusion and fat embolism (Group 2) caused a significant decrease in systolic blood pressure at 181 minutes (p = 0.001). This hypotensive effect persisted for one hour (until 240 minutes), after which the blood pressure returned to baseline values (Fig. 1). At 181 minutes, the mean blood pressures in Groups 2 and 3 were significantly lower than that in Group 1 (p = 0.0002) (Fig. 1). The mean blood pressure in Group 2 was significantly lower than that in either Group 1 or Group 3 at 185, 195, and 210 minutes (p = 0.0035, 0.0004, and 0.0009, respectively). At 240 minutes, the mean blood pressure in Group 2 was significantly lower than that in Group 1 (p = 0.0095).



View larger version (23K):
[in this window]
[in a new window]
  		Figs. 1 through 4: Graphs showing the mean (and standard error of the mean) for several outcome parameters for the three groups. T = 0 indicates the start of the experiment; T = 30, immediately before pulmonary contusion was created; T = 60, immediately after pulmonary contusion was created; T = 120, sixty minutes after pulmonary contusion was created and immediately before fat embolism was produced; T = 181, one minute after fat embolism was produced; T = 185, five minutes after fat embolism was produced; T = 195, fifteen minutes after fat embolism was produced; T = 210, thirty minutes after fat embolism was produced; T = 240, sixty minutes after fat embolism was produced; and T = 480, the end of the experiment. One millimeter of mercury = 0.1333 kilopascal.* = Groups 2 and 3 were significantly different from Group 1, $ = Group 2 was significantly different from Groups 1 and 3, and # = Group 2 was significantly different from Group 1. Fig. 1: Systolic blood pressure.

 

Pulmonary Physiology
Peak airway pressure: Thoracotomy, pulmonary contusion (Group 1), and fat embolism alone (Group 3) were not found to have a significant effect on peak airway pressure, with the numbers available. The combination of pulmonary contusion and fat embolism (Group 2) caused a significant increase in peak airway pressure at 181 minutes compared with the baseline pressure (p = 0.015). This effect continued until 480 minutes (p = 0.045) (Table I). Despite this increase compared with the baseline values in Group 2, no significant difference in peak airway pressure was detected among the groups at any specific time-point, even at the end of the experiment.


View this table:
[in this window]
[in a new window]
  		TABLE I PULMONARY PHYSIOLOGY AND GAS EXCHANGE DATA*

 
Total thoracic compliance: The mean total thoracic compliance was increased, compared with the baseline values, after the thoracotomy, at thirty minutes, in all groups (Table I). This effect was significant in Group 2 (p = 0.041) and Group 3 (p = 0.038), but it disappeared after the chest wall was closed, at sixty minutes. Neither pulmonary contusion alone (Group 1) nor fat embolism alone (Group 3) was found to significantly decrease the mean total thoracic compliance, with the numbers available. The combination of pulmonary contusion and fat embolism (Group 2) caused a significant decrease in total thoracic compliance at 181 minutes (p = 0.0002) compared with the baseline values, and this effect persisted until the end of the experiment. At 480 minutes, the mean total thoracic compliance had decreased 6.4 percent in Group 1, 4.6 percent in Group 3, and 23.5 percent in Group 2 compared with the baseline values (Table I). The mean total thoracic compliance in Group 2 at 480 minutes was significantly lower than that in Group 1 and Group 3 (p = 0.049).

Pulmonary arterial pressure: Neither thoracotomy nor pulmonary contusion alone (Group 1) caused a change in pulmonary arterial pressure compared with the baseline values (Fig. 2). Fat embolism alone (Group 3) caused a significant increase in pulmonary arterial pressure at 181 minutes (p = 0.01), but this effect was transient and the pressure returned to baseline after fifteen minutes (Fig. 2). The combination of pulmonary contusion and fat embolism (Group 2) also significantly increased pulmonary arterial pressure at 181 minutes (p = 0.01). This pulmonary hypertensive effect persisted until the end of the experiment (Fig. 2). At 181 and 185 minutes, the mean pulmonary arterial pressure in both Group 2 (p = 0.0069) and Group 3 (p = 0.023) was significantly greater than that in Group 1. From 195 to 480 minutes, only the mean pulmonary arterial pressure in Group 2 was significantly greater than that in Group 1 (p = 0.028 at 195 minutes, p = 0.048 at 210 minutes, p = 0.043 at 240 minutes, and p = 0.04 at 480 minutes) (Fig. 2).



View larger version (23K):
[in this window]
[in a new window]
  		Pulmonary arterial pressure.

 
Pulmonary capillary wedge pressure: Neither thoracotomy nor pulmonary contusion alone (Group 1) was found to significantly affect pulmonary capillary wedge pressure (Fig. 3), with the numbers available. Fat embolism alone (Group 3) caused a significant increase in pulmonary capillary wedge pressure at 181 minutes (p = 0.015), but the pressure returned to baseline within five minutes (Fig. 3). Pulmonary contusion and fat embolism (Group 2) caused a significant increase in pulmonary capillary wedge pressure at 181 minutes (p = 0.0013); the pressure also returned to baseline within five minutes. The mean pulmonary capillary wedge pressure in Group 2 was significantly greater than that in Group 1 at 181 minutes (p = 0.0079) (Fig. 3).



View larger version (26K):
[in this window]
[in a new window]
  		Pulmonary capillary wedge pressure.

 

Gas Exchange
Partial pressure of arterial oxygen: Thoracotomy and pulmonary contusion alone (Group 1) were not found to have a significant effect on partial pressure of arterial oxygen, with the numbers available. Fat embolism alone (Group 3) caused a significant decrease in partial pressure of arterial oxygen at 181 minutes compared with the baseline values (p = 0.0001), and this effect persisted until 480 minutes (Table I). The combination of pulmonary contusion and fat embolism (Group 2) also caused a significant decrease in partial pressure of arterial oxygen at 181 minutes (p = 0.0001); this effect also persisted until 480 minutes (Table I). At 181 minutes, the mean partial pressure of arterial oxygen in both Group 2 and Group 3 was significantly lower than that in Group 1 (p = 0.0092). At 480 minutes, only the mean partial pressure of arterial oxygen in Group 2 was significantly lower than that in Group 1 (p = 0.042) (Table I).

Partial pressure of carbon dioxide: Partial pressure of arterial carbon dioxide was not found to be significantly affected by thoracotomy or pulmonary contusion alone (Group 1), with the numbers available. Fat embolism alone (Group 3) produced a significant increase in partial pressure of arterial carbon dioxide at 480 minutes compared with the baseline values (p = 0.01). The combination of pulmonary contusion and fat embolism (Group 2) produced an immediate significant increase in partial pressure of arterial carbon dioxide at 181 minutes (p = 0.045), which persisted until 480 minutes. There was no significant difference among the groups at any time-point with regard to the mean partial pressure of arterial carbon dioxide (Table I).

Ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration: We did not find that thoracotomy and pulmonary contusion alone (Group 1) had an effect on the ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration, with the numbers available. Fat embolism alone (Group 3) caused a significant decrease in the ratio at 181 minutes compared with the baseline values (p = 0.0001), and this effect persisted until 480 minutes (Table I). Pulmonary contusion alone in Group 2 caused a significant decrease in the ratio at sixty minutes (p = 0.002). When pulmonary contusion was combined with fat embolism in Group 2, the magnitude of the decrease nearly quadrupled at 181 minutes (p = 0.0001); this effect persisted until 480 minutes. The mean ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration at 480 minutes was decreased 23.7 percent in Group 1, 52.3 percent in Group 3, and 65.8 percent in Group 2 compared with the baseline values (Table I). From 181 to 480 minutes, the mean ratios in both Group 2 and Group 3 were significantly lower than those in Group 1 (p = 0.011 at 181 minutes and p = 0.046 at 480 minutes) (Fig. 4). However, we did not detect a significant difference between the mean ratios in Group 2 and those in Group 3.



View larger version (22K):
[in this window]
[in a new window]
  		Ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration.

 
Alveolar-arterial oxygen gradient: Thoracotomy was not found to have a significant effect on the alveolar-arterial oxygen gradient, with the numbers available. Pulmonary contusion alone (Group 1) caused an increase at sixty minutes compared with the baseline values, but this trend was not found to be significant until seven hours later, at 480 minutes (p = 0.034). Fat embolism alone (Group 3) caused an immediate significant increase in the alveolar-arterial oxygen gradient at 181 minutes (p = 0.0001), and this effect persisted until 480 minutes. Pulmonary contusion alone in Group 2 caused a significant increase in the alveolar-arterial oxygen gradient at sixty minutes (p = 0.003). When pulmonary contusion was combined with fat embolism in Group 2, the alveolar-arterial oxygen gradient nearly doubled at 181 minutes (p = 0.0001); this effect persisted until 480 minutes. At 181 minutes, the mean alveolar-arterial oxygen gradient in both Group 2 and Group 3 was significantly greater than that in Group 1 (p = 0.019) (Table I).

Histological Findings
None of the lung samples had evidence of neutrophils or hyaline membrane (an inflammatory score of 0 points). The mean pulmonary edema score was significantly greater (p = 0.0001) in Group 2 (0.83 ± 0.04 point) than in Group 1 (0.51 ± 0.04 point) or Group 3 (0.44 ± 0.04 point). We detected no significant difference in the mean pulmonary edema score between Groups 1 and 3. We also detected no significant difference in the overall mean percentage of the area of the lung occupied by fat between Group 2 (ratio, 0.76 ± 0.04) and Group 3 (ratio, 0.88 ± 0.06). Within Group 3, we detected no significant difference between the right and left lungs with regard to the mean percentage of the area occupied by fat. In Group 2, the right lung had a significantly greater mean percentage of the area occupied by fat (ratio, 0.89 ± 0.05) than the left lung (ratio, 0.64 ± 0.05) (p = 0.001). The mean percentage of the glomerular area occupied by fat was significantly greater in Group 2 (ratio, 0.27 ± 0.06) than in Group 3 (ratio, 0.12 ± 0.02) (p = 0.0001). The mean percentage of the cerebral area occupied by fat was significantly greater in Group 2 (ratio, 0.43 ± 0.05) than in Group 3 (ratio, 0.25 ± 0.03) (p = 0.01).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Pulmonary dysfunction in patients who have sustained a traumatic injury may lead to respiratory failure, substantial morbidity, and death6. Among the most common causes of pulmonary dysfunction after multiple injuries are pulmonary contusion and fat embolism13,31. These two modes of damage are not mutually exclusive; pulmonary contusion frequently coexists with multiple fractures of long bones and resultant fat embolism32.

Much is already known about the pathophysiology of pulmonary contusion as a result of numerous experimental animal models. Previous investigators have dropped weights from a fixed height onto the closed chest of animals7,8,21, while Shepard et al.31 standardized a method of pulmonary contusion that involves firing a blank from a .38-caliber revolver onto the plated chest of anesthetized dogs. Gross and histological studies have also shown that, when untreated, a lesion occurs immediately after energy dispersal and then regresses over ten days and that mechanical ventilation has a protective effect on the contused lung32. The specific histopathological changes of pulmonary contusion were studied previously by Oppenheimer et al.21 in a dog model. Those authors determined that pulmonary contusion causes a leak of blood and plasma that floods the air spaces, which reduces ventilation and compliance, decreases the regional ventilation-to-perfusion ratio, and produces hypoxemia.

In the present study, compression damage was produced under direct visualization at the time of the open thoracotomy. This method allowed better standardization of the contusion and associated injury of the chest wall among the three groups. Support for our method of pulmonary contusion is found in the experimental work of Craven et al.8, who reported that, in the setting of positive pressure ventilation, neither flail chest nor a thoracotomy had a significant effect on the size of the contusion or the decrease in pulmonary function. With the use of positive pressure ventilation in the current study, the animals that had pulmonary contusion alone (Group 1) were not found to have a significant decrease in the total thoracic compliance or a noticeable impairment of gas exchange.

Many previous experimental studies of pulmonary contusion failed to reproducibly demonstrate evidence of clinical pulmonary dysfunction, such as hypoxemia. The preservation of oxygenation was most likely the result of a well established mechanism whereby pulmonary vasculature responds to alveolar hypoxia with vasoconstriction2. This phenomenon of regional vasoconstriction causes a physiological redistribution of blood to noncontused areas of the lung14, thus maintaining pulmonary gas exchange despite a demonstrable unilateral contusion.

The results of our present study are in agreement with the current understanding of pulmonary dysfunction following pulmonary contusion alone. Pulmonary contusion alone caused a small increase in the alveolar-arterial oxygen gradient, but this effect was not found to be significant until seven hours after the contusion. Partial pressure of arterial oxygen and the ratio of partial pressure of arterial oxygen to fractional inspired oxygen concentration were not found to be affected by pulmonary contusion alone. These findings are readily explained by hypoxic pulmonary vasoconstriction in the contused left lung, which would shift a relatively greater proportion of blood flow to the noncontused right lung, thus minimizing the mismatch of ventilation and perfusion2. Additional support for the regulatory mechanism of hypoxic pulmonary vasoconstriction in our experiment was found with histological analysis. A greater percentage of the area was occupied by fat in the noncontused right lung than in the contused left lung in the dogs in Group 2 (pulmonary contusion and fat embolism), whereas the percentage of the area that was occupied by fat was the same for both lungs in Group 3 (fat embolism alone). Thus, more fat emboli went to the normal lung than to the contused lung in the dogs in Group 2. Because of the regulatory mechanism of hypoxic pulmonary vasoconstriction, partial pressure of arterial oxygen alone is not a good indicator of the extent of pulmonary contusion. With regard to the other outcome parameters, our results indicate that, with mechanisms of compensation and ventilatory support, the degree of unilateral pulmonary contusion had a minimum adverse effect on hemodynamic values (systolic blood pressure and cardiac output), pulmonary physiology (peak airway pressure, total thoracic compliance, pulmonary arterial pressure, and pulmonary capillary wedge pressure), and gas exchange (partial pressure of carbon dioxide).

Fractures of long bones and their repair are associated with pulmonary fat embolism10,12, which often compromises pulmonary function24,35. In one canine model of fat embolism, the alveolar-arterial oxygen gradient was increased after intramedullary nailing, both with and without reaming; this effect was not seen after fixation with a plate29. Because the lungs receive the entire cardiac output, they have a major role in filtering emboli that could threaten the systemic circulation, and thus they protect the coronary, renal, and cerebral circulation20. It was demonstrated experimentally, in a canine model of fat embolism, that deformable fat globules can pass through the lung vasculature under high pulmonary arterial pressures5,29, as was consistently observed with fat embolism in the present study. In our experiment, three dogs (two in Group 2 and one in Group 3) died within minutes after pressurization of the medullary canal with cement, and there was gross evidence of systemic fat embolization. Histological evidence of systemic fat embolism was seen in the rest of the animals in Groups 2 and 3.

Our experimental model of fat embolism alone (Group 3) caused a significant transient decrease in systolic blood pressure and a significant transient increase in pulmonary arterial pressure and pulmonary capillary wedge pressure. Fat embolism alone resulted in a greater acute decrease in partial pressure of arterial oxygen than did pulmonary contusion alone, and this decrease was accompanied by a significant increase in the alveolar-arterial oxygen gradient. The results of the present study indicate that the degree of fat embolism produced by unilateral femoral and tibial pressurization had a significantly greater effect on pulmonary physiology and gas exchange than did pulmonary contusion alone.

In Group 2, in which fat embolism was superimposed on pulmonary contusion, there was a significant decrease in total thoracic compliance, a significant increase in peak airway pressure and pulmonary capillary wedge pressure, and a persistent elevation in pulmonary arterial pressure. Profound and sustained hypotension, which lasted an hour, was noted only in Group 2. The Group-2 dogs also had an immediate significant increase in partial pressure of carbon dioxide. Histological analysis of lung specimens from Group 2 demonstrated the greatest mechanical damage, as evidenced by significantly greater pulmonary edema scores, compared with the damage caused by either injury alone (Groups 1 and 3). It is clear that the combination of pulmonary contusion and fat embolism in our model produced greater impairment of hemodynamic and pulmonary function than did either injury alone.

As far as we know, the effects of combined pulmonary contusion and fat embolism were investigated in only one previous study9. That study, of a sheep model of fat embolism with intramedullary reaming alone or after blunt injury of the chest, demonstrated no hypoxemia and only a small transient increase in pulmonary vascular resistance despite evidence of embolism on histological study and on intravascular ultrasound. Those authors did not observe the most important clinical features of pulmonary dysfunction and concluded that pulmonary contusion does not prime the lung for damage by fat embolism. The results of that model are not consistent with the observations of Pape et al.24, who reported substantial sustained pulmonary dysfunction, even after intramedullary reaming alone, in patients who had sustained multiple traumatic injuries. Our data clearly demonstrate a significant deterioration in pulmonary function with fat embolism alone and an even greater dysfunction when fat embolism is preceded by pulmonary contusion. It is possible that the first study9 was unable to demonstrate dysfunction because the volume of embolic fat was lower or the pulmonary contusion was less extensive, resulting in less ventilation and perfusion mismatch, compared with these parameters in our study.

There is clinical evidence that, in the presence of pulmonary contusion, early intramedullary nailing can have a detrimental effect on pulmonary function. In a prospective study of patients who had multiple injuries, Pape et al.23 found that early primary intramedullary nailing with reaming and resultant embolism of medullary fat was associated with an increase in pulmonary dysfunction and adult respiratory distress syndrome. In previous clinical studies, investigators selected patients on the basis of an assessment of the overall severity of the injury3,11,13, without specifically addressing the effect of concomitant injury of the lung. The results of our study suggest that the greater the extent of preexisting pulmonary contusion, the greater the detrimental effects of fat emboli.

In summary, our experimental canine model of trauma demonstrated that partial pressure of arterial oxygen is not an accurate indicator of the severity of pulmonary contusion. We observed that the combination of pulmonary contusion and fat embolism may potentiate both pulmonary dysfunction and systemic embolization after trauma. Furthermore, our results suggest that the optimum time for fixation of fractures, and additional embolic load, may depend on the severity and stage of the underlying pulmonary contusion rather than simply on the time after the injury.


   Footnotes
 
*No benefits in any form have been received or will be received form a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were Grant MA-11292 of the Medical Research Council of Canada and The Physicians' Services Incorporated of Ontario.

{dagger}Musculoskeletal Research Laboratory, Division of Orthopaedic Surgery, Department of Surgery (A. W. E., S. A., R. R. R., and E. H. S.), and Department of Anaesthesia (R. J. B.), St. Michael's Hospital, University of Toronto, 55 Queen Street East, Suite 800, Toronto, Ontario M5C 1R6, Canada. The e-mail address for Dr. Schemitsch is schemitsche@the-wire.com.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Alexander, R. H., and Proctor, H. J.: Advanced Trauma Life Support; Course for Physicians—Student Manual, pp. 11-12. Chicago, American College of Surgeons, 1993.
  2. Benumof, J. L., and Wahrenbrock, E. A.: Blunted hypoxic pulmonary vasoconstriction by increased lung vascular pressures. J. Appl. Physiol., 38: 846-850, 1975.[Abstract/Free Full Text]
  3. Bone, L. B.; Johnson, K. D.; Weigelt, J.; and Scheinberg, R.: Early versus delayed stabilization of femoral fractures. A prospective randomized study. J. Bone and Joint Surg., 71-A: 336-340, March 1989.[Abstract/Free Full Text]
  4. Bone, L. B.: Emergency treatment of the injured patient. In Skeletal Trauma, pp. 127-145. Edited by B. D. Browner, J. B. Jupiter, A. M. Levine, and P. G. Trafton. Philadelphia, W. B. Saunders, 1992.
  5. Byrick, R. J.; Mullen, J. B.; Mazer, C. D.; and Guest, C. B.: Transpulmonary systemic fat embolism. Studies in mongrel dogs after cemented arthroplasty. Am. J. Respir. and Crit. Care Med., 150: 1416-1422, 1994.[Abstract]
  6. Clark, G. C.; Schecter, W. P.; and Trunkey, D. D.: Variables affecting outcome in blunt chest trauma: flail chest vs. pulmonary contusion. J. Trauma, 28: 298-304, 1988.[Medline]
  7. Cohn, S. M., and Zieg, P. M.: Experimental pulmonary contusion: review of the literature and description of a new porcine model. J. Trauma, 41: 565-571, 1996.[Medline]
  8. Craven, K. D.; Oppenheimer, L.; and Wood, L. D.: Effects of contusion and flail chest on pulmonary perfusion and oxygen exchange. J. Appl. Physiol., 47: 729-737, 1979.[Abstract/Free Full Text]
  9. Duwelius, P. J.; Huckfeldt, R.; Mullins, R. J.; Shiota, T.; Woll, T. S.; Lindsey, K. H.; and Wheeler, D.: The effects of femoral intramedullary reaming on pulmonary function in a sheep lung model. J. Bone and Joint Surg., 79-A: 194-202, Feb. 1997.[Abstract/Free Full Text]
  10. Fabian, T. C.: Unraveling the fat embolism syndrome. New England J. Med., 329: 961-963, 1993.[Free Full Text]
  11. Goris, R. J.; Gimbrere, J. S.; van Niekerk, J. L.; Schoots, F. L.; and Booy, L. H.: Early osteosynthesis and prophylactic mechanical ventilation in the multitrauma patient. J. Trauma, 22: 895-903, 1982.[Medline]
  12. Hulman, G.: The pathogenesis of fat embolism. J. Pathol., 176: 3-9, 1995.[Medline]
  13. Johnson, K. D.; Cadambi, A.; and Seibert, G. B.: Incidence of adult respiratory distress syndrome in patients with multiple musculoskeletal injuries: effect of early operative stabilization of fractures. J. Trauma, 25: 375-384, 1985.[Medline]
  14. Kadiri, Y. Z.; Kay, J. C.; Kovacs, K.; and Noble, W. H.: Pulmonary embolism distribution to ventilated and unventilated lungs in the dogs: a cause of hypoxaemia. Canadian Anaesth. Soc. J., 27: 216-222, 1980.
  15. Kerstell, J.: Pathogenesis of post-traumatic fat embolism. Am. J. Surg., 121: 712-715, 1971.[Medline]
  16. Kishikawa, M.; Yoshioka, T.; Shimazu, T.; Sugimoto, H.; Yoshioka, T.; and Sugimoto, T.: Pulmonary contusion causes long-term respiratory dysfunction with decreased functional residual capacity. J. Trauma, 31: 1203-1210, 1991.[Medline]
  17. Lewis, F. R.: Thoracic trauma. Surg. Clin. North America, 62: 97-104, 1982.[Medline]
  18. Lhowe, D. W.: Open fractures of the femoral shaft. Orthop. Clin. North America, 25: 573-580, 1994.[Medline]
  19. Lindeque, B. G. P.; Schoeman, H. S.; Dommisse, G. F.; Boeyens, M. C.; and Vlok, A. L.: Fat embolism and the fat embolism syndrome. A double-blind therapeutic study. J. Bone and Joint Surg., 69-B(1): 128-131, 1987.
  20. Malik, A. B.: Pulmonary microembolism. Physiol. Rev., 63: 1114-1207, 1983.[Free Full Text]
  21. Oppenheimer, L.; Craven, K. D.; Forkert, L.; and Wood, L. D.: Pathophysiology of pulmonary contusion in dogs. J. Appl. Physiol., 47: 718-728, 1979.[Abstract/Free Full Text]
  22. Pahud, B., and Vasey, H.: Delayed internal fixation of femoral shaft fractures—is there an advantage? A review of 320 fractures. J. Bone and Joint Surg., 69-B(3): 391-394, 1987.
  23. Pape, H. C.; Auf'm'Kolk, M.; Paffrath, T.; Regel, G.; Sturm, J. A.; and Tscherne, H.: Primary intramedullary femur fixation in multiple trauma patients with associated lung contusion—a cause of posttraumatic ARDS?. J. Trauma, 34: 540-548, 1993.[Medline]
  24. Pape, H. C.; Regel, G.; Dwenger, A.; Krumm, K.; Schweitzer, G.; Krettek, C.; Sturm, J. A.; and Tscherne, H.: Influences of different methods of intramedullary femoral nailing on lung function in patients with multiple trauma. J. Trauma, 35: 709-716, 1993.[Medline]
  25. Pell, A. C. H.; Hughes, D.; Keating, J.; Christie, J.; Busuttil, A.; and Sutherland, G. R.: Brief report: fulminating fat embolism syndrome caused by paradoxical embolism through a patent foramen ovale. New England J. Med., 329: 926-929, 1993.[Free Full Text]
  26. Peltier, L. F.: Fat embolism. A perspective. . Clin. Orthop., 232: 263-270, 1988.
  27. Pepe, P. E.; Potkin, R. T.; Reus, D. H.; Hudson, L. D.; and Carrico, C. J.: Clinical predictors of the adult respiratory distress syndrome. Am. J. Surg., 144: 124-130, 1982.[Medline]
  28. Petty, T. L., and Newman, J. H.: Adult respiratory distress syndrome. Western J. Med., 128: 399-407, 1978.[Medline]
  29. Schemitsch, E. H.; Jain, R.; Turchin, D. C.; Mullen, J. B.; Byrick, R. J.; Anderson, G. I.; and Richards, R. R.: Pulmonary effects of fixation of a fracture with a plate compared with intramedullary nailing. A canine model of fat embolism and fracture fixation. J. Bone and Joint Surg., 79-A: 984-996, July 1997.[Abstract/Free Full Text]
  30. Scopa, M.; Magatti, M.; and Rossitto, P.: Neurologic symptoms in fat embolism syndrome: case report. J. Trauma, 36: 906-908, 1994.[Medline]
  31. Shepard, G. H.; Ferguson, J. L.; and Foster, J. H.: Pulmonary contusion. Ann. Thoracic Surg., 7: 110-119, 1969.[Medline]
  32. Trinkle, J. K.; Furman, R. W.; Hinshaw, M. A.; Bryant, L. R.; and Griffen, W. O.: Pulmonary contusion. Pathogenesis and effect of various resuscitative measures. Ann. Thoracic Surg., 16: 568-573, 1973.[Medline]
  33. Weigelt, J. A.: The adult respiratory distress syndrome. In The Multiply Injured Patient with Complex Fractures, pp. 42-54. Edited by M. H. Myers. Philadelphia, Lea and Febiger, 1984.
  34. Wozasek, G. E.; Simon, P.; Redl, H.; and Schlag, G.: Intramedullary pressure changes and fat intravasation during intramedullary nailing: an experimental study in sheep. J. Trauma, 36: 202-207, 1994.[Medline]
  35. Wozasek, G. E.; Thurnher, M.; Redl, H.; and Schlag, G.: Pulmonary reaction during intramedullary fracture management in traumatic shock: an experimental study. J. Trauma, 37: 249-254, 1994.[Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Facebook Facebook   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 JBJSHome page
K. Mohanty, J.N. Powell, D. Musso, M. Traboulsi, I. Belenkie, J.B.M. Mullen, and J.V. Tyberg
The Effect of a Venous Filter on the Embolic Load During Medullary Canal Pressurization: A Canine Study
J. Bone Joint Surg. Am., June 1, 2005; 87(6): 1332 - 1337.
[Abstract] [Full Text] [PDF]

Home page
 JBJSHome page
R. V. Patel, Y.-H. Kim, S.-W. Oh, and J.-S. Kim
Prevalence of Fat Embolism Following Bilateral Simultaneous and Unilateral Total Hip Arthroplasty
J. Bone Joint Surg. Am., May 28, 2003; 85(6): 1164 - 1165.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Letters to the Editor: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Letters to the Editor are posted
Right arrow Alert me if a correction is posted
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow Rights and Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by ELMARAGHY, A. W.
Right arrow Articles by SCHEMITSCH, E. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by ELMARAGHY, A. W.
Right arrow Articles by SCHEMITSCH, E. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Facebook   Add to Technorati   Add to Twitter  
What's this?