Interleukin-6 as an early marker for fat embolism
© Raj et al; licensee BioMed Central Ltd. 2009
Received: 02 February 2009
Accepted: 13 June 2009
Published: 13 June 2009
Fat Embolism is a complication of long bone fractures, intramedullary fixation and joint arthroplasty. It may progress to fat embolism syndrome, which is rare but involves significant morbidity and can occasionally be fatal. Fat Embolism can be detected at the time of embolization by transoesophageal echocardiography or atrial blood sampling. Later, a combination of clinical signs and symptoms will point towards fat embolism but there is no specific test to confirm the diagnosis. We investigated serum Interleukin-6 (IL-6) as a possible early marker for fat embolism.
An animal study was conducted to simulate a hip replacement in 31 adult male Sprague Dawley rats. The procedure was performed under general anesthesia and the animals divided into 3 groups: control, uncemented and cemented. Following surgery and recovery from anaesthesia, the rats allowed to freely mobilize in their cages. Blood was taken before surgery and at 6 hours, 12 hours and 24 hours to measure serum IL-6 levels. The rats were euthanized at 24 hours and lungs removed and stained for fat. The amount of fat seen was then correlated with serum IL-6 levels.
No rats in the control group had fat emboli. Numerous fat emboli were seen in both the uncemented and cemented implant groups. The interleukin levels were raised in all groups reaching a peak at 12 hours after surgery reaching 100 pg/ml in the control group and around 250 pg/ml in the uncemented and cemented implant groups. The IL-6 levels in the control group were significantly lower than any of the implant groups at 12 and 24 hours. At these time points, the serum IL-6 correlated with the amount of fat seen on lung histology.
Serum IL-6 is a possible early marker of fat embolism.
Fat embolism occurs when fat and marrow content from bone gains access to the circulation and becomes trapped in the capillaries of the lungs or other organs . Bone marrow in the long bones of mature adults contains up to 92% of fat . The fat is released following skeletal trauma [3, 4] or from manipulation of the medullary canal during hip and knee arthroplasty . Fat emboli in the lungs trigger an inflammatory process leading to hemorrhage and leakage of proteins into the alveoli .
Inflammatory mediators have also been implicated in the lung changes following fat embolism [7, 8] such as IL-6, CD-11b expression, elastase and s-E-selectin . One study showed that IL-6 and TNF-alpha were elevated in cases of multiple fractures . Syrbu concluded that following fat embolism broncho-alveolar lavage contained inflammatory mediators .
IL-6 is one of the most important cytokine in the acute inflammatory phase . It is also one of the mediator that is released very early in an injury process . At the moment, there is no specific blood test that can be used to detect fat embolism. Diagnosis is mainly by clinical features and supported by a few laboratory tests. This often results in delayed treatment which can be detrimental to the patient. Having a reliable marker for fat embolism would be of great clinical benefit and the aim of our study was to determine whether IL-6 was such a marker.
An animal model using Sprague Dawley rats was used for this study. Many studies have utilized rats as a model for fat embolism [13–17]. Most of them used external infusion of fatty acids to simulate fat embolism. However bone marrow cells are a significant part of the material embolized and in order to reproduce the clinical situation more closely we chose to embolize the animal's own medullary content rather than injecting fatty acids. The medullary cavity of Sprague Dawley rats has been shown to contain around 5% of fat cells . Our animal model for fat embolism is described in the methodology section. It has been shown that rats have a heart circulation that closely resembles that of humans making them suitable to study the effects of embolism on the cardio repiratory system . The rat animal model was chosen as it allowed us to carry out a surgical procedure generating fat embolislm, measure serum levels of IL-6 and provided lung biopsy samples to confirm that embolism had actually occurred and also to quantify the amount of fat deposited in the lungs. For those reasons the Sprague Dawley rat was used as an animal model in this study.
The objective of this study was to assess whether the serum concentration of an inflammatory marker could be correlated to the extent of inflammation secondary from fat embolism. The ultimate aim would be to use the marker as an early diagnostic tool for a condition (FES) notorious for late detection. Although there are many inflammatory markers, we selected IL-6 as a marker for a few reasons. Firstly the ELISA kit for measuring IL-6 required 0.5 mls of blood for each measurement. As inflammatory cytokines are elevated in hemorrhagic shock [20, 21] we wanted to reduce the amount of blood taken by limiting our study to IL-6. Removal of up to 7.5% (1.5 mls in our 500 gm rats) of the total circulating blood volume is considered to be safe  and in order to avoid stressing the animals and falsely raising the inflammatory cytokines we selected to study only one inflammatory marker. Secondly, IL-6 is an inflammatory marker that is activated in the early phase of the acute inflammatory response and therefore would allow early detection and treatment of fat embolism. Finally, we had to limit ourselves to one marker as the rat model did not allow us to draw enough blood to study multiple markers.
Thirty one adult male Sprague Dawley rats were used. They were divided into 3 groups: control (n = 9), uncemented implant (n = 10) and cemented implant (n = 12). Prior approval from the institutional animal ethics committee was obtained. These rats were housed in standard solid floor cages with wood shaving litter. The animals were induced using inhalational halothane anesthesia with a mixture of oxygen. Depth of anesthesia was assessed by pedal reflex. After each animal was anaesthetized, it was placed in a supine position and the skin over both knee joints was shaved and disinfected. A skin incision was made over the flexed knee joint. A medial parapatella incision allowed retraction of the patella laterally, exposing the distal femur.
There were 3 different groups:
A 2 mm drill perforated the articular cartilage at the superior end of the inter-condylar notch, taking great care not to breach the medullary canal. The medial parapatella wound was closed using Vicryl sutures and the skin with staples. This was repeated on the contralateral side.
ii) uncemented implant group
iii) cemented implant group
The same surgeon performed all surgery. The three groups were alternated sequentially until all the rats were operated. Blood (0.5 mls) was taken from a tail vein before surgery, at 6 hours and at 12 hours while the animals were briefly anaesthetised. At 24 hours, all the rats were euthanized. Intracardiac blood was then taken immediately after death. A midline chest incision was made to expose the lungs and heart. The trachea was transected and the lungs expanded using formalin saline injected via the cut trachea. The right upper lobe and the left lobe were processed for histological examination. Osmium Tetroxide was used to stain fat within the tissues [23, 24]. The lungs were wax-impregnated and sectioned. Finally they were counterstained with Haematoxilin and Eosin.
Slides were viewed using light microscopy with a help of a pathologist. In order not to under or over estimate the amount of embolised material, all fat globules seen in the slide were counted at 10× magnification. Subsequently, the surface area for the lung section was estimated using a 10 mm by 10 mm square grid (the grid was made up of 1 mm2 boxes). The crossectional areas of the right upper lobe and the left lobs sections were determined. The density of fat emboli per 100 mm2 was then calculated.
The blood collected was initially centrifuged at 4000 revolutions per minute for 4 minutes. The serum was then isolated and stored in a minus 80 degrees fridge until all the samples had been collected. On the day of ELISA analysis, the samples were thawed to room temperature. The serum was then processed following the recommendations of the IL-6 kit manufacturer (R&D Systems).
All the reagents (which were stored at minus 20 degrees Celsius) were brought to room temperature. The 96 well ELISA plate was filled with 50 μl of assay diluents in each well. Each sample was processed in duplicate with the final result calculated as the mean of both values. The first 2 rows (8 wells in each row) were filled up with 50 μl of either the standard or control samples. There were 7 standards of known IL-6 values, which ranged from the lowest to the highest measurable IL-6 concentration that was provided by the manufacturer (62.5 pg/l to 4000 pg/l) and the control was a single sample of known concentration (given by the manufacturer) that was used to ensure quality control. The subsequent 10 rows were filled with 50 μl of test serum to each well and these too were processed in duplicate. All plates were incubated for 2 hours. They were then aspirated and washed 5 times using a plate washer. Subsequently 100 μl of conjugate was added to each well and incubated for 2 hours. It was then aspirated and washed with a buffer solution 5 times. Finally 100 μl of substrate solution was added and protected from light for 30 minutes. Then 100 μl of stop solution was applied to each well. The optical density was then read at 450 nm on an Omron spectrophotometer. The reading was printed out and the duplicates averaged out. A standard curve was drawn using the values obtained from the 7 standards. The control sample ensured that the quality readings obtained was within the set limits. Finally, the values for the test samples were worked out from the standard curve.
We postulated that more extensive fat embolism occurred in the uncemented and cemented implant groups. We used the Analysis of Variance (ANOVA) to see if there were any statistical differences between the three groups with regards to the amount of fat. Subsequently, we applied Fisher's Least Significant Difference (LSD) as a post hoc test to study if there were any differences between these three groups. We used p < 0.05 as the confidence interval. We also postulated that the IL-6 levels would be higher if there was more fat embolism seen, therefore a comparison was calculated with the IL-6 levels at the various time points. ANOVA was used with LSD as a post hoc procedure to see if there was any difference in the 3 groups studied in the four different time points.
Finally, we wanted to see if we could correlate the fat embolism to the measured IL-6 levels. For this, we initially combined all the three different groups and compared the IL-6 levels at the three different time points (6 hours, 12 hours and 24 hours) with the amount of fat embolism seen. Correlation of the number of fat emboli and IL-6 levels at 6 hours, 12 hours and 24 hours irrespective of the three groups was carried out using the Spearmans Rank Correlation test. This test gives a Spearman Correlation Coefficient between -1 and +1. A value closer to +1 will indicate a direct positive correlation. All the tests completed were using the Statistical Package for Social Sciences (SPSS) Version 11 for Windows.
Out of the 31 rats in this study 3 rats died during surgery. All of them belonged to the cemented implant group and they were excluded from the study.
The most likely cause of the raised IL-6 levels in the implant groups is fat embolism as the only difference between them and the control group was the fat seen on lung histology. The correlation coefficient at 6 hours between IL-6 and the number of emboli seen was 0.292. At 12 and 24 hours the correlation coefficient was 0.494 and 0.405 respectively. This indicates that there was a positive correlation between fat embolism and IL-6. This correlation was stronger at 12 and 24 hours.
The receptor for IL-6 is found on many cell surfaces including resting normal T-cells, activated normal B-cells, myeloid cell lines and others. IL-6 stimulates acute phase reaction that is responsible for fever, increased erythrocyte sedimentation rate and activation of clotting cascades, all of which are seen in established fat embolism syndrome.
IL-6 is sensitive to tissue damage  and local levels of IL-6 may be higher than systemic levels [26–29]. In our study only systemic levels of IL-6 were measured and they could be an overspill of the local IL-6  and we therefore assumed that systemic IL-6 levels correlate with local IL-6 levels. Inflammation in the lungs probably is thought to be due to the toxic effects of the free fatty acids released from the breakdown of fat. Histology of the lungs in this study confirmed the presence of inflammation at the time of death. (Figure 3).
One study claimed that the inflammatory changes in the lungs were immune mediated  whereas another concluded that fat emboli are not the only cause of FES . It is interesting that although fat embolism will occur in almost all cases of long bone fractures, only a few cases will develop fat embolism syndrome. One possible explanation is the production of inflammatory mediators. Evidence has shown that IL-6 is also responsible for adipose tissue metabolism, lipoprotein lipase activity and hepatic triglyceride secretion . There is a polymorphism of the IL-6 gene that will result in abnormalities in IL-6 transcription rate. Experimental studies in humans comparing the activity of this gene have shown that patients with polymorphism of the IL-6 gene are prone to lipid abnormalities . Although this was not demonstrated by these authors, this abnormal IL-6 transcription could well lead to an abnormal reaction to fat emboli in the lungs and the development of a full blown fat embolism syndrome.
Various types of orthopaedic operations lead to an increase in IL-6 serum levels [28, 33]. This explains why in the present study there was a rise in IL-6 levels in all 3 groups as the surgery involved some degree of muscle injury to get access to the knee joint. However, although the surgical approach was the same in the 3 groups, the variation in IL-6 was significantly different and we have shown that fat embolism demonstrated histologically must be a major factor.
There was significant variability in IL-6 levels post surgery within each group of animals. This variability has also been shown by Minetto in his study if IL-6 levels after hip surgery . It is possible that different individuals release different quantities of IL-6 when presented with the same stimuli. Apart from that, he noted that the upward slope of the IL-6 curve was related to the duration of surgery. He also found that higher IL-6 levels were associated with higher postoperative fever but without evidence of differences in postoperative problems, time to mobilize, or duration of stay in hospital. Postoperative infection was also associated with higher IL-6 levels . In our study, all animals were operated on in a small animal operating theatre using clean surgical techniques. The surgical time varied between 10 to 15 minutes and the same surgical protocol was used in all cases by one single operator.
Inflammatory cells are responsible for the release of IL-6. If it was released solely as a result of tissue trauma, it should have increased straight after surgery and peaked at 6 hours. However in our study the peak was at 12 hours that makes tissue trauma from the surgical procedure unlikely. We have shown that release of IL-6 was due to fat embolism and secondary lung tissue damage. This effect was noted up to 24 hours. It is likely that emboli in the lung trigger an inflammatory reaction, which takes time to become established before IL-6 levels increase. This may explain why there was no correlation between IL-6 rise and the amount of fat emboli at 6 hours.
The significant correlation shown at 12 hours and 24 hours demonstrates that IL-6 rise is related to fat embolism. In this study, the correlation was 0.494 and 0.405 at 12 and 24 hours respectively
This is an animal model and although we tried to reproduce a surgical procedure as closely as possible, the findings cannot be automatically transferred to clinical practice. IL-6 is only one of the inflammatory markers and there are surely others involved in the pathogenesis of fat embolism.
We have shown, using an animal fat embolism model, that there are lung changes which correlate with IL-6 serum levels which makes us believe that IL-6 may be one of the early markers of fat embolism. Further research is required to validate the use of IL-6 as a reliable marker to detect fat embolism in a clinical setting.
- Parisi DM, Koval K, Egol K: Fat embolism syndrome. Am J Orthop. 2002, 31 (9): 507-12.PubMedGoogle Scholar
- Goodsitt MM, Hoover P, Veldee MS, Hsueh SL: The composition of bone marrow for a dual-energy quantitative computed tomography technique. A cadaver and computer simulation study. Invest Radiol. 1994, 29 (7): 695-704. 10.1097/00004424-199407000-00006.View ArticlePubMedGoogle Scholar
- Batra P: The fat embolism syndrome. J Thorac Imaging. 1987, 2 (3): 12-7. 10.1097/00005382-198707000-00005.View ArticlePubMedGoogle Scholar
- Kontakis GM, Tossounidis T, Weiss K, Pape HC, Giannoudis PV: Fat embolism: special situations bilateral femoral fractures and pathologic femoral fractures. Injury. 2006, 37 (Suppl 4): S19-24. 10.1016/j.injury.2006.08.037.View ArticlePubMedGoogle Scholar
- Koessler MJ, Pitto RP: Fat and bone marrow embolism in total hip arthroplasty. Acta Orthop Belg. 2001, 67 (2): 97-109.PubMedGoogle Scholar
- Szabo G, Magyar Z, Reffy A: The role of free fatty acids in pulmonary fat embolism. Injury. 1977, 8 (4): 278-83. 10.1016/0020-1383(77)90102-4.View ArticlePubMedGoogle Scholar
- Giannoudis PV, Pape HC, Cohen AP, Krettek C, Smith RM: Review: systemic effects of femoral nailing: from Kuntscher to the immune reactivity era. Clin Orthop Relat Res. 2002, 378-86. 10.1097/00003086-200211000-00056. 404Google Scholar
- Sun TS, Chen XB, Liu Z, Ma ZY, Zhang JZ: [Relationship between the operation time of multiple fractures with system inflammation changes and clinical outcomes]. Zhonghua Wai Ke Za Zhi. 2008, 46 (13): 961-5.PubMedGoogle Scholar
- Giannoudis PV, Abbott C, Stone M, Bellamy MC, Smith RM: Fatal systemic inflammatory response syndrome following early bilateral femoral nailing. Intensive Care Med. 1998, 24 (6): 641-2. 10.1007/s001340050631.View ArticlePubMedGoogle Scholar
- Syrbu S, Thrall RS, Smilowitz HM: Sequential appearance of inflammatory mediators in rat bronchoalveolar lavage fluid after oleic acid-induced lung injury. Exp Lung Res. 1996, 22 (1): 33-49. 10.3109/01902149609074016.View ArticlePubMedGoogle Scholar
- Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F: Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J. 2003, 374 (Pt 1): 1-20. 10.1042/BJ20030407.PubMed CentralView ArticlePubMedGoogle Scholar
- Murtaugh MP, Baarsch MJ, Zhou Y, Scamurra RW, Lin G: Inflammatory cytokines in animal health and disease. Vet Immunol Immunopathol. 1996, 54 (1–4): 45-55. 10.1016/S0165-2427(96)05698-X.View ArticlePubMedGoogle Scholar
- Liu DD, Hsieh NK, Chen HI: Histopathological and biochemical changes following fat embolism with administration of corn oil micelles: a new animal model for fat embolism syndrome. J Bone Joint Surg Br. 2008, 90 (11): 1517-21. 10.1302/0301-620X.90B11.20761.View ArticlePubMedGoogle Scholar
- Kao SJ, Chen HI: Nitric oxide mediates acute lung injury caused by fat embolism in isolated rat's lungs. J Trauma. 2008, 64 (2): 462-9. 10.1097/TA.0b013e318058aa2e.View ArticlePubMedGoogle Scholar
- El-Ali KM, Gourlay T: Assessment of the risk of systemic fat mobilization and fat embolism as a consequence of liposuction: ex vivo study. Plast Reconstr Surg. 2006, 117 (7): 2269-76. 10.1097/01.prs.0000218715.58016.71.View ArticlePubMedGoogle Scholar
- Agnantis N, Gyras M, Tserkezoglou A, Apostolikas N, Papacharalampous N: Therapeutic effect of bovine albumin in the experimental fat embolism syndrome. Respiration. 1998, 53 (1): 50-7. 10.1159/000195396.View ArticleGoogle Scholar
- Liu DD, Kao SJ, Chen HI: N-acetylcysteine attenuates acute lung injury induced by fat embolism. Crit Care Med. 2008, 36 (2): 565-71. 10.1097/01.CCM.0000299737.24338.5C.View ArticlePubMedGoogle Scholar
- Martin RB, Zissimos SL: Relationships between marrow fat and bone turnover in ovariectomized and intact rats. Bone. 1991, 12 (2): 123-31. 10.1016/8756-3282(91)90011-7.View ArticlePubMedGoogle Scholar
- Ballaux PK, Gourlay T, Ratnatunga CP, Taylor KM: A literature review of cardiopulmonary bypass models for rats. Perfusion. 1999, 14 (6): 411-7.View ArticlePubMedGoogle Scholar
- Ayala A, Wang P, Ba ZF, Perrin MM, Ertel W, Chaudry IH: Differential alterations in plasma IL-6 and TNF levels after trauma and hemorrhage. Am J Physiol. 1991, 260 (1 Pt 2): R167-71.PubMedGoogle Scholar
- Hierholzer C, Kalff JC, Omert L, Tsukada K, Loeffert JE, Watkins SC, Billiar TR, Tweardy DJ: Interleukin-6 production in hemorrhagic shock is accompanied by neutrophil recruitment and lung injury. Am J Physiol. 1998, 275 (3 Pt 1): L611-21.PubMedGoogle Scholar
- Nahas K, Ph Banex JPP, Rabemampianina Y: Effects of acute blood removal via sublingual vein on haematological and clinical parameters in Sprague-Dawley rats. Laboratory Animals. 2000, 34: 362-71. 10.1258/002367700780387804.View ArticlePubMedGoogle Scholar
- Abramowsky CR, Pickett JP, Goodfellow BC, Bradford WD: Comparative demonstration of pulmonary fat emboli by "en bloc" osmium tetroxide and oil red O methods. Hum Pathol. 1981, 12 (8): 753-5. 10.1016/S0046-8177(81)80179-7.View ArticlePubMedGoogle Scholar
- Davison PR, Cohle SD: Histologic detection of fat emboli. J Forensic Sci. 1987, 32 (5): 1426-30.View ArticlePubMedGoogle Scholar
- Hong X, Ye TH, Zhang XH, Ren HZ, Huang YG, Bu YF: Changes of interleukin-6 and related factors as well as gastric intramucosal pH during colorectal and orthopaedic surgical procedures. Chin Med Sci J. 2006, 21 (1): 57-61.PubMedGoogle Scholar
- Krohn CD, Reikeras O, Mollnes TE, Aasen AO: Complement activation and release of interleukin-6 and tumour necrosis factor-alpha in drained and systemic blood after major orthopaedic surgery. Eur J Surg. 1998, 164 (2): 103-8. 10.1080/110241598750004742.View ArticlePubMedGoogle Scholar
- Arnold JP, Haeger M, Bengtson JP, Bengtsson A, Lisander B: Release of inflammatory mediators in association with collection of wound drainage blood during orthopaedic surgery. Anaesth Intensive Care. 1995, 23 (6): 683-6.PubMedGoogle Scholar
- Clementsen T, Krohn CD, Reikeras O: Systemic and local cytokine patterns during total hip surgery. Scand J Clin Lab Invest. 2006, 66 (6): 535-42. 10.1080/00365510600889635.View ArticlePubMedGoogle Scholar
- Perl M, Gebhard F, Knoferl MW, Bachem M, Gross HJ, Kinzl L, Wolf Strecker: The pattern of preformed cytokines in tissues frequently affected by blunt trauma. Shock. 2003, 19 (4): 299-304. 10.1097/00024382-200304000-00001.View ArticlePubMedGoogle Scholar
- Schemitsch EH, Jain R, Turchin DC, Mullen JB, Byrick RJ, Anderson GI, Richards RR: 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 Joint Surg Am. 1997, 79 (7): 984-96.PubMedGoogle Scholar
- Fernandez-Real JM, Broch M, Vendrell J, Gutierrez C, Casamitjana R, Pugeat M, Richart C, Ricart W: Interleukin-6 gene polymorphism and insulin sensitivity. Diabetes. 2000, 49 (3): 517-20. 10.2337/diabetes.49.3.517.View ArticlePubMedGoogle Scholar
- Fernandez-Real JM, Broch M, Vendrell J, Richart C, Ricart W: Interleukin-6 gene polymorphism and lipid abnormalities in healthy subjects. J Clin Endocrinol Metab. 2000, 85 (3): 1334-9. 10.1210/jc.85.3.1334.View ArticlePubMedGoogle Scholar
- Hashimoto T, Hiruta H, Yamada Y, Yamanaka M, Koh J, Miyata M: [Changes in cytokines during perioperative period of hip arthroplasty in patients older than 80 years of age]. Masui. 2003, 52 (11): 1214-7.PubMedGoogle Scholar
- Minetto MA, Oprandi G, Saba L, Mussino S, Aprato A, Masse A, Angeli A, Gallinaro P: Serum interleukin-6 response to elective total hip replacement surgery. Int Orthop. 2006, 30 (3): 172-6. 10.1007/s00264-006-0073-3.PubMed CentralView ArticlePubMedGoogle Scholar
- Di Cesare PE, Chang E, Preston CF, Liu CJ: Serum interleukin-6 as a marker of periprosthetic infection following total hip and knee arthroplasty. J Bone Joint Surg Am. 2005, 87 (9): 1921-7. 10.2106/JBJS.D.01803.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.