- Email: [email protected]
Intravasation of bone marrow content. Can its magnitude and effects be modulated by low pressure reaming in a porcine model?
Injury, Int. J. Care Injured 41 (2010) S2, S9–S15
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Intravasation of bone marrow content. Can its magnitude and effects be modulated by low pressure reaming in a porcine model? Elisabeth Ellingsen Husebyea, *, Torstein Lybergb , Helge Opdahlc , Olav Røisea,d a Department
of Orthopaedics, Oslo University Hospital, Ullevaal, Norway for Clinical Research, Oslo University Hospital, Ullevaal, Norway c The Norwegian National Center for NBC Medicine, Oslo University Hospital, Ullevaal, Norway d Division of Critical Care, Oslo University Hospital and Faculty of Medicine University of Oslo, Norway b Center
article info
abstract
Keywords: Suction irrigation reaming Intramedullary reaming Femoral shaft fracture Animal study Porcine model Bone marrow embolization Inflammation and coagulation response Intramedullary pressure RIA
Introduction: Intramedullary orthopaedic procedures may increase the intramedullary pressure (IMP) and thereby cause intravasation of bone marrow contents. In recent studies by the authors the reamer–irrigator–aspirator (RIA) has been demonstrated to reduce IMP and coagulation-, fibrinolysis- and cytokine responses, but did not prove any significant difference in cardiopulmonary function parameters or numbers of emboli when compared to a traditional reaming (TR) system. The correlations between IMP increase, regardless type of reamer, and inflammatory- and coagulation responses, pulmonary embolization, and cardiopulmonary alterations have, however, not previously been analyzed in this material. Our hypothesis was that a lower IMP would result in reduced occurrence of pulmonary embolization, reduced inflammatoryand coagulation responses, as well as reduced cardiopulmonary alterations. Materials and Methods: Twenty-eight young Norwegian landrace pigs were exposed to femoral intramedullary reaming, with either the TR (n = 10) or the RIA (n = 10) system, or used as controls (n = 8). IMP was recorded during reaming and nailing. Serial blood samples for demonstration of coagulation-, fibrinolysis-, and cytokine activation were withdrawn peroperatively and until 72 hours post nail insertion. Circulatory and pulmonary effects were monitored peroperatively and until two hours postoperatively. The animals were sacrificed 72 hours post nail insertion and lung tissue biopsies were harvested and examined for lung emboli. Results and Conclusions: A strong correlation between increased IMP and increased coagulationand cytokine responses was found. The number of emboli was not significantly correlated to IMP, but was strongly correlated to changes in the coagulation- and cytokine responses. No clinical relevant correlations were observed between increased IMP or numbers of lung emboli and changes in hemodynamic- or pulmonary function parameters. A correlation between coagulation activation and cytokine activation was observed. This study confirms the connection between increased IMP, increased coagulation activation and the magnitude of pulmonary emboli in a model evaluating the effects of intramedullary reaming of intact pig femora. In this model, the lowering of IMP during reaming, as obtained with RIA, reduced the magnitude of and the effects of bone marrow extravasation. © 2010 Elsevier Ltd. All rights reserved.
Introduction Early stabilization of femoral shaft fractures is associated with reduced pulmonary complication and mortality.3,4,12,24,32 On the other hand, early intramedullary nailing (IMN) has also been associated with increased inflammation and morbidity, and increased incidence of multi organ failure (MOF).11,30,31,36,39 Intramedullary * Corresponding author. Elisabeth Ellingsen Husebye, MD. Department of Orthopaedics, Oslo University Hospital, Ullevaal, Kirkeveien 166, 0407 Oslo, Norway. Tel.: +47 22118080/+47 23027884; fax: +47 23027410. E-mail addresses: [email protected] (E.E. Husebye); [email protected] (T. Lyberg); [email protected] (H. Opdahl); [email protected] (O. Røise). 0020-1383/ $ – see front matter © 2010 Elsevier Ltd. All rights reserved.
procedures may increase the intramedullary pressure (IMP) and thereby cause intravasation of bone marrow content.19,27,40,41 The resulting clinical manifestation probably depends on the quantity of bone marrow elements entering the circulation and the magnitude of clogging of the pulmonary circulation.6,16,23,26,37,42 An association between the magnitude of a surgical procedure and the extent of inflammatory response to the operative trauma has been demonstrated.1,13,28,29 A reamer–irrigator–aspirator (RIA) system has previously been demonstrated by the present authors to reduce IMP,14 reduce coagulation-, fibrinolytic- and cytokine responses,17 but demonstrated no significant differences in cardiopulmonary function parameters,15 neither significant lowering of the number of pulmonary emboli15 when compared to a traditional reaming
S10
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
(TR) system. Possible correlations between IMP, pulmonary embolization, cardiopulmonary function parameters and inflammatory responses have previously not been explored. Our hypothesis was that keeping the IMP low during reaming of the femoral canal would result in reduced occurrence of pulmonary embolization, reduced inflammatory- and coagulation responses, and decreased cardiopulmonary alterations. Data from animals, where reaming was performed with either the RIA or the TR system was therefore pooled for this analysis of the impact of high and low IMPs. Material and methods The study was conducted in accordance with “Regulations on Animal Experimentation” under The Norwegian Animal Welfare Act and approved by the Norwegian Animal Research Authority. Twenty-eight Norwegian landrace pigs (15 males and 13 females), age 3–3.5 months, weighing 27.9–38.0 kg, were used. The reaming groups (TR and RIA group) consisted of 10 animals each. Eight animals were used as controls. Six animals were excluded due to anaesthesia-related events prior to operation, exsanguinations or a shorter than protocol stabilization time. Before surgery the animals were housed one or two animals per stall and returned to individual stalls after the operation. The size of the cages was 200 cm × 300 cm. Room temperature was 20±2ºC, with 40% ± 5% humidity and there was a 12 hour light/dark cycle. The pigs were fed a standard diet, prior to surgery they fasted overnight, but had free access to water. Choice of experimental model As one of the goals of the study was to evaluate the effects of low pressure reaming, represented by RIA reaming, and the RIA-reamer heads were available with the smallest diameter of 12 mm, the experimental models were limited to large animal models, however, the animals should be handable. The young pigs chosen for the study had open epiphysis. To determine the localization of the epiphysis and the shape and dimensions of the femur, radiographs of the femur of a pig used for other research purposes were taken. The anatomy of the femur showed a tight isthmus (in the TR group: 2.14±0.03 and in the RIA group: 2.15±0.04 cm8 ) and a much wider condylar area. The femoral neck was short compared to humans. Furthermore, the femora from two other pigs were divided longitudinally for visual evaluation of the epiphysis (Fig. 1). The preplanning included also a surgical approach, and the evaluation of the femur after reaming with the RIA system (Fig. 1).
Pre-reaming procedures The animals were orally intubated and mechanically ventilated. The anaesthesia was induced and maintained with intravenous propofol (Diprivan® , 20 mg/ml, AstraZeneca, Caponago, Italy) and fentanyl (Fentanyl® , 50 mg/ml, Alpharma, Linz, Austria). 10–15 ml/kg/h Ringer-acetate solution was administrated intravenously during the anaesthesia period. The animals were sacrificed on the third postoperative day. A detailed procedure is described by Husebye et al..15 A flow-directed pulmonary artery (PA) catheter (744HF75, SwanGanz CCOmbo CCO/SvO2 catheter 7.5F, Edwards Critical-Care Division, Irvine, CA, USA) was inserted in the left internal jugulary vein with the tip in the pulmonary artery. An 18-gauge catheter was inserted into the carotid artery. An 18-gauge catheter was inserted in the left femoral vein. These catheters were used for arterial, mixed venous and femoral vein blood sampling during the operation and until two hours after the nail was inserted. The catheter in the carotid artery was two hours post nail insertion connected to a vascular access port (VAP) (Celsite® Access Port ST301 SIL (2.8 × 1.1), B-Braun Medical, Chasseneuil, France), placed in the posterolateral aspect of the neck, for arterial blood sampling on awake animals six hours to 72 hours post nail insertion. Two hours after the insertion of the nail the femoral vein catheter and the pulmonary artery catheters were removed. Peroperatively all catheters were flushed with a solution of heparin (Heparin Leo® , 5000 U/ml, LEO Pharma A/S, Ballerup, Denmark) and Macrodex (Macrodex® , 60 mg/ml with sodium chloride, Pharmalink AB, Upplands Vasby, ¨ Sweden) as described by Husebye et al..15 The total dose of heparin given during the study period was approximately 2000 U. Mean pulmonary artery pressure (MPAP), central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), heart rate (HR), mean arterial pressure (MAP), cardiac output (CO) and mixed venous oxygen saturation (SvO2 ) were monitored. Systemic (SVR) and pulmonary vascular resistance (PVR) and the alveolo-arterial oxygen (PA O2 − Pa O2 ) differences were calculated. Indexed values were not used due to uniform size of the animals. The registration of IMP was performed with a transducer tipped pressure monitoring catheter with a diameter of 1.35 mm (Integra NeuroScienceTM Camino1, Model 110–4B, San Diego, USA), placed just cranial to the distal femoral epiphysis (Fig. 1). The pressure catheter was removed 10 minutes after insertion of the nail. The procedure is described in detail by Husebye et al..14 The time schedule for blood sampling and cardiopulmonary and IMP registrations is given in Table 1. Surgical technique
Fig. 1. A longitudinally divided pig femur post RIA reaming with distal femoral epiphysis (white arrow) and drill bit located proximal to the epiphysis, where the intramedulary pressure catheter was placed.
Through a longitudinal incision the access to the proximal femur was made. The femoral marrow cavity was opened with an awl, a guide-wire was introduced, and antegrade reaming and nailing was performed. For traditional reaming (Stryker, Kiel, Germany, Model: Bixcut, 0226/3095/3100/3105/3110/3115/3120) a sequential reaming, using increasing steps of 0.5 mm in reamer diameter (9.5 mm – 12 mm) was used. A 12 mm reamer head was used for the one-step RIA (Synthes, Paoli, PA, USA, 325.250S) technique. The RIA equipment was connected to a flushing tube and a suction tube with an internal lumen of 5.5 mm and 30 mm H2 Osuction force. A standard Große-Kempf nail (Stryker Trauma GmbH, Schonkirchen, ¨ Germany), cut at a length of 10 cm, with a diameter of 10 mm was used and left inside. The animals in the control group were exposed to a sham operation with no reaming or nailing of the femoral cavity.
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
S11
Table 1 Time schedule for data related to surgical procedures a Time
Surgical procedure
Blood samples
Cardiopulmonary registrations
IMP registration
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
Blood sample location a m
v
Coagulation, fibrinolysis, cytokines, blood gases
x
x
A
Baseline
B
(A+ 15 min)
After reaming
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
C
(B+ 7 min)
After nail insertion
Coagulation, fibrinolysis, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
D
(C+ 30 min)
30 min after nail insertion
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
E
(D+ 90 min)
2 h after nail insertion
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
F
(E+ 4 h)
6 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
G
(F+ 18 h)
24 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
H
(G+ 24 h)
48 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
I
(H+ 24 h)
72 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
a
The table shows the time schedule for blood sampling at different sampling locations; arterial (a), mixed venous (m) and femoral vein (v) blood and cardiopulmonary and intramedullary pressure (IMP) registrations and calculations; 75 minutes after completion of soft tissue preparation (baseline, A), after reaming (B), immediately after insertion of the nail (C), and at 30 minutes (D), two (E), six (F), 24 (G), 48 (H) and 72 (I) hours after insertion of the nail.
Blood sampling and assays Blood sampling was performed as described in Table 1. In the control group the time after reaming (B) and after the nail was inserted (C) were set by the mean levels in the TR and RIA groups. The following cytokines were analyzed using commercially available ELISA kits; TNF-a, IL-1b, IL-6, IL-8 and IL-10. For evaluation of the coagulation and fibrinolysis systems, thrombin-antithrombin complex (TAT), tissue plasminogen activator (t-PA) activity and plasminogen activator inhibitor-1 (PAI-1) activity were analyzed. The specifications of the analyzing kits, blood sample handling and storage is described by Husebye et al..17 Blood gases were analyzed (ABL735, 2002, Radiometer Medical ApS, Copenhagen, Denmark) at time points described in Table 1. Lung specimens The lungs were excised immediately after the death, and the hearts examined to exclude the presence of intra-cardiac shunts. Two lung tissue samples were harvested from the periphery of each lung lobe. The total area of the lung histology preparations was calculated using Auto Sketch 6.0 (1998) planimetry program. The incidence of emboli was expressed as numbers of emboli per square centimetre lung area. The preparation of the lung biopsies has previously been described in detail by Husebye et al..15 Statistics Statistical analyses were performed using the Statistical Package for Social Science (SPSS) software, version 16.0 (SPSS Inc, Chicago, IL, USA). The relationship between IMP, lung emboli, inflammatoryand coagulation- and fibrinolytic responses, and cardiopulmonary response were analyzed using Pearson product-moment correlation coefficient (r). For IMP and amount of lung emboli absolute values were used, and for cardiopulmonary alterations and inflammatory-, fibrinolytic- and coagulation responses the difference (D values)
from baseline levels (A) were used. Correlations were considered significant at P levels ≤0.05 and considered strong when r ≤ −0.500 or r > 0.500. Main results The statistics and results are based on the same material used for previous separately published papers.14,15,17 IMP levels were measured in 15 animals. One animal in the RIA group was excluded due to a perioperatively inflicted supracondylar femoral fracture, and three animals in the TR group due to perioperative perforation of the distal medial femoral cortex. The highest IMP in the TR group was demonstrated at the first or the second reaming step. In both the TR and the RIA groups the peak IMP levels were used for the correlation analyses. IMP was not measured in the control group. Lung histology evaluation was performed in 22 study animals. All observed lung emboli were found in small arteries with a diameter of 0.08–2.0 mm.15 In the control group one single lung embolus was found. Coagulation-, fibrinolysis- and cytokine activation were studied in 22 animals. We used the peak levels of arterial blood D TAT, D -PA, D PAI-1, and arterial D IL-6 and femoral vein blood D IL-6 for the statistics. Arterial blood D TAT and D t-PA levels demonstrated a biphasic curve with peak levels at reaming completed (B) and at six hours post nail insertion (F). Arterial blood D PAI-1 levels showed peak levels at six hours post nail insertion (F). The cytokine response was mainly represented by increased procedure-related D IL-6 levels for which a peak was present in arterial blood six hours post nail insertion (F) and in femoral vein blood after nail insertion (C). Hemodynamic and respiratory function was evaluated in 22 study animals. For correlation analyses peak D values and/or D values at those points of registration demonstrating group differences between the TR and RIA reaming. For D CO, D PVR, D SVR, D MPAP, D SvO2 , and D Aa difference the analyses were performed after nail
S12
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
Fig. 2. Scatter plots demonstrating significant, strong positive correlations between peak IMP levels and arterial blood D TAT (2A) and femoral vein blood D TAT levels (2B) at completed reaming, numbers of lung emboli (PE) and arterial blood D TAT (2C) and femoral vein blood D TAT levels (2D) at completed reaming, numbers of lung emboli (PE) and arterial blood D IL-6 at six hours post nail insertion (2E), and arterial blood D TAT at completed reaming and arterial blood D IL-6 at six hours post nail insertion (2F).
insertion (C) and at two hours post nail insertion (E), and for D CVP and D PCWP after nail insertion (C). Intramedullary pressure, coagulation-, fibrinolysis and cytokine responses, and hemodynamic and pulmonary function parameters A strong correlation was demonstrated between peak IMP and arterial blood D TAT levels at reaming (B) (r = 0.516, p = 0.049)
(Fig. 2A) and an even stronger correlation was present between peak IMP and femoral vein blood D TAT levels at reaming (B) (r = 0.759, p = 0.001) (Fig. 2B). No significant correlations were present between peak IMP and peak arterial and femoral vein blood D t-PA and D PAI-1 levels (data not shown). The local activation of IL-6 (femoral vein blood) at nail insertion (C) increased significantly related to increasing peak IMP values (r = 0.554, p = 0.032). We found no correlation between IMP and numbers of lung emboli. No clinical
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
relevant significant correlations were present between increased IMP and hemodynamic- or pulmonary function parameters (data not shown). Numbers of lung emboli, coagulation-, fibrinolysis- and cytokine responses and hemodynamic and pulmonary function parameters A strong correlation was found between numbers of lung emboli and arterial D TAT levels (r = 0.646, p = 0.001) (Fig. 2C) and femoral vein blood D TAT levels (r = 0.547, p = 0.008) (Fig. 2D) at completed reaming (B). No significant correlations were present between numbers of lung emboli and peak arterial and femoral vein blood D t-PA and D PAI-1 levels (data not shown). Increased numbers of lung emboli correlated positively to increased peak arterial D IL-6 levels (r = 0.569, p = 0.006) (Fig. 2E). No clinical relevant correlations were present between increased numbers of lung emboli and hemodynamic- or pulmonary function parameters (data not shown). Coagulation, fibrinolysis and cytokine activation A strong positive correlation was present between arterial D TAT levels at completed reaming and arterial D IL-6 levels at six hours post nail insertion (r = 0.532, p = 0.012) (Fig. 2F). Discussion The reamer–irrigator–aspirator (RIA) system has previously been compared to traditional reaming (TR) by the present authors. These studies demonstrated reduced peak IMP14 (mean 188 mmHg in the TR and mean 33 mmHg in the RIA group), reduced coagulation-, fibrinolytic- and cytokine responses,17 lower numbers of lung emboli (mean 0.31 in the TR and mean 0.21 lung emboli per square centimetre lung area in the RIA group [ns]),15 and no significant differences in cardiopulmonary function parameters,15 when compared to the TR system. The RIA reamer represents a low pressure reaming system. The correlations between IMP increase and pulmonary embolization, inflammatory- and coagulation responses, and the cardiopulmonary alterations were, however, not previously explored in this material. Such analysis disclosed a strong relationship between increased IMP and increased coagulation- and cytokine responses, as well as between increased numbers of lung emboli and coagulation- and cytokine responses. Additionally, a strong relationship between activation of coagulation- and cytokine responses was demonstrated. However, no correlation between IMP and numbers of lung emboli was observed, neither were clinical relevant correlations found between hemodynamic- or pulmonary function parameters and increased IMP numbers of lung emboli. Reamed IMN of the femur represents a surgical trauma. The systemic trauma response is multifactorial and represents a unique interaction between coagulation, fibrinolysis, the complement cascade, vascular endothelial cells, circulating white blood cells, reactive oxygen species, and pro- and anti-inflammatory cytokines. The magnitude of the inflammatory response seems to correlate with the extent of injury (operation or trauma),1,13,28,28 and therefore; the burden of the operation could be evaluated through the response it creates. In this study the cytokine response was mainly represented by increased IL-6, demonstrating peak arterial levels six hours post nail insertion, which is in accordance with other studies10,11,21,29 and peak femoral vein blood IL-6 levels at nail insertion, which is quite early post procedure. Increased IL-6 levels are generally associated with an inflammatory response.2 A strong correlation between peak arterial blood IL-6 levels and numbers of lung emboli, as well as peak femoral vein blood IL-6 levels and IMP were demonstrated, which may indicate a link between
S13
IMP, local inflammatory effects and embolization. Additionally, a strong positive correlation was present between peak arterial TAT at completed reaming and peak arterial IL-6 levels. Both peak arterial and femoral vein blood TAT levels correlated strongly to both increased IMP and increased numbers of lung emboli. These results are in accordance with the literature, where increased coagulation response correlated with the intraoperative embolic response as well as the stimulation of other markers of inflammation.6,7,11,20,33 The evaluation of the consequences of bone marrow embolization is complex; both the estimation of volume of bone marrow material entering the circulation, and the evaluation of the potential harmful effects. The basic requirement is to prove the actual intravasation, then the amount of bone marrow intravasation has to be estimated, and finally the effect of the bone marrow intravasation on the test individuals has to be explored. Wenda et al. found increased IMP during intramedullary reaming and nailing of the femur in experimental studies, and that the elevated IMP resulted in intravasation of bone marrow contents evaluated echocardiographically.40,41 Already at an IMP of 50 mmHg, snow-flurry in the vena cava was demonstrated.41 In that study the degree of embolization seemed to depend directly on the pressure level. In our experimental model, comparing IMP increase related to traditional reaming and reaming with the RIA device, the mean IMP levels were 188 and 33 mmHg, respectively.14 The IMP related to RIA was below the IMP level at which circulating embolic material could be echocardiographically detected.41 We could, however, not demonstrate a significant correlation between increasing IMP levels and numbers of lung emboli in lung tissue. Echocardiographical verification of intravascular bone marrow content and even examination of the fat emboli material withdrawn from venous blood during IMN, gives the opportunity to detect the actual presence and estimate the volume of bone marrow content in the blood stream in a semiquantitative manner. As our experimental model was already complex, such investigations were not included. Retrospectively, obtaining an estimate of the volume of embolic material and analyzing its correlations to IMP, numbers of lung emboli, coagulation-, fibrinolysis- and cytokine responses and hemodynamic or pulmonary function would have been interesting. Histological evaluation of lung tissue after IMN in experimental models has demonstrated the presence of bone marrow emboli.18,35 The highest concentration of emboli was found within the first postoperative day, with equally distribution to all lung lobes.5,20 In our model the animals were sacrificed on the third postoperative day, which according to the literature does not represent the time where most pulmonary emboli are expected to be found.18,35 Several different methods for analyzing the amount of fat in the lung parenchyma, have, however, been applied,9,18,26,34 which make the comparison of the amount of pulmonary emboli difficult. The magnitude of hemodynamic and pulmonary changes after pulmonary embolism has been demonstrated to be a function of size and numbers of the emboli and the underlying state of cardiopulmonary function.6,42 The exact amount of emboli needed to cause cardiopulmonary dysfunction has not been established. The vessels in a normal pulmonary circulation are easily dilated when the pressure increases, therefore both pulmonary arterial pressures and vascular resistance may remain almost unchanged despite occlusion. In older literature it has been estimated that 30–50% of a previous normal pulmonary circulation has to be occluded before hemodynamic alterations can be detected.23 The hemodynamic alterations related to microscopic pulmonary embolism and microthrombosis associated with conditions like acute lung injury (ALI) and adult respiratory distress syndrome (ARDS) could be circulatory failure through the effect of increased mean pulmonary artery pressure (MPAP).26 The right ventricle can be expected to fail when 75% of the pulmonary vasculature
S14
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
becomes obstructed.38 Such failure leads to end-diastolic dilatation and displacement of the septum wall, which results in reduced end-diastolic filling of the left ventricle37,42 and further aggravates the circulatory failure. In our experimental model the animals were young and healthy, and the embolic load was probably not sufficiently massive to cause marked cardiopulmonary alterations, and therefore we found no clinical relevant correlations between numbers of lung emboli or parameters of cardiopulmonary function. The length of the pig femur relative to body size is about 50% shorter than in humans (approximately 12–13% of the body length versus 25% in humans) resulting in a smaller potential volume of bone marrow content that can be extruded into the vascular system of the pigs. Additionally, the trabecular bone is most probably denser, the cortical bone less dense, and the venous circulation better when compared to human adult bone.25 Both a wider and shorter bone and denser trabecular bone are conditions that probably lower the IMP increase during intramedullary reaming. In the present model we used intact femora. A fracture may represent a “first-hit” in the pulmonary circulation, which might have made a procedure-related increase in PVR easier to detect. On the other hand, the presence of a superposed fracture would probably have lowered the increases of IMP, as the fracture may serve as a vent and allow bone marrow escape via the fracture diastases during reaming. The length of the pig femora were approximately 15 cm, and the distance from fossa piriformis to the distal femoral epiphysis was 11–12 cm.8 The short femur and additionally the location of the femur in the pig make a fracture model complicated and unreliable. An osteotomy of the femur as a fracture substitute, however, is possible, but with the length of the porcine femur, the created IMP would probably be too low to test the study hypothesis. In an experimental study of femoral IMN on intact and fractured canine femora the release of triglycerides was greater in a non-fracture model,22 suggesting more bone marrow content intravasation in intact femora. As the effect of bone marrow embolization during reaming and nailing was one goal of our study, we are not certain that inducing a femoral fracture prior to surgery would have increased the pathophysiological changes. In this experimental model we demonstrated that femoral low pressure reaming had beneficial effects on IMP and coagulationand cytokine responses. We could, however, not demonstrate that this reduced IMP had a direct and significant effect on numbers of lung emboli. In future studies echocardiography should be included for the evaluation of magnitude of bone marrow content intravasation during reaming.
Conclusion In this model on intramedullary reaming and nailing of intact pig femora, there was a strong correlation between the numbers of lung emboli and increased coagulation- and cytokine activation. The coagulation- and cytokine activation correlated also strongly to the procedure-related IMP increase. The numbers of lung emboli were, however, not directly correlated to IMP increase or changes in the circulation or pulmonary gas exchange. Overall, in this model, low pressure reaming seems to reduce the magnitude and the effect of the bone marrow intravasation.
Competing interests Each of the authors certifies that he or she has no commercial associations that might pose a conflict of interest in connection with the submitted manuscript. The manuscript has not been submitted or published elsewhere.
Acknowledgements The pig study was supported by grants from the AO Research Fund of the AO Foundation (Project no. 04-R71) and Synthes. Operating tools were delivered by Stryker and Synthes. The authors wish to express their gratitude to Ingeborg Løstegaard Goverud at the Department of Pathology, Oslo University Hospital, Ullevaal, Norway, for excellent histological preparations and to Trude Aspelin and Lisbeth Sætre, Center for Clinical Research, Oslo University Hospital, Ullevaal, Norway, for carefully analyzing the blood samples. We also acknowledge the staff, and in particular Morten Eriksen, at the Institute of Experimental Medical Research, University of Oslo, for their support. References 1. Baigrie RJ, Lamont PM, Kwiatkowski D, et al. Systemic cytokine response after major surgery. Br J Surg 1992;79:757–60. 2. Biffl WL, Moore EE, Moore FA, Peterson VM. Interleukin-6 in the injured patient. Marker of injury or mediator of inflammation? Ann Surg 1996;224:647–64. 3. Bone LB, Johnson KD, Weigelt J, Scheinberg R. Early versus delayed stabilization of femoral fractures. A prospective randomized study. J Bone Joint Surg Am 1989;71:336–40. 4. Brundage SI, McGhan R, Jurkovich GJ, et al. Timing of femur fracture fixation: effect on outcome in patients with thoracic and head injuries. J Trauma 2002; 52:299–307. 5. Cheung NM CW, Leung KS. A study of the pathway of marrow fat intravasation during intramedullary reaming. Osteo Trauma Care 2004;12:140–4. 6. Christie J, Robinson CM, Pell AC, et al. Transcardiac echocardiography during invasive intramedullary procedures. J Bone Joint Surg Br 1995;77:450–5. 7. Christie J, Robinson CM, Singer B, Ray DC. Medullary lavage reduces embolic phenomena and cardiopulmonary changes during cemented hemiarthroplasty. J Bone Joint Surg Br 1995;77:456–9. 8. Ellingsen Husebye E, Lyberg T, Madsen JE, et al. The early effects of intramedullary reaming of the femur on bone mineral density; an experimental study in pigs. Scand J Surg 2009;98:189–94. 9. Elmaraghy AW, Aksenov S, Byrick RJ, et al. Pathophysiological effect of fat embolism in a canine model of pulmonary contusion. J Bone Joint Surg Am 1999;81:1155–64. 10. Gebhard F, Pfetsch H, Steinbach G, et al. Is interleukin 6 an early marker of injury severity following major trauma in humans? Arch Surg 2000;135:291–5. 11. Giannoudis PV, Smith RM, Bellamy MC, et al. Stimulation of the inflammatory system by reamed and unreamed nailing of femoral fractures. An analysis of the second hit. J Bone Joint Surg Br 1999;81:356–61. 12. Goris RJ, Gimbrere JS, van Niekerk JL, et al. Early osteosynthesis and prophylactic mechanical ventilation in the multitrauma patient. J Trauma 1982;22:895–903. 13. Harwood PJ, Giannoudis PV, van Griensven M, et al. Alterations in the systemic inflammatory response after early total care and damage control procedures for femoral shaft fracture in severely injured patients. J Trauma 2005;58:446–52; discussion: 52–4. 14. Husebye EE, Lyberg T, Madsen JE, et al. The influence of a one-step reamer– irrigator–aspirator technique on the intramedullary pressure in the pig femur. Injury 2006;37:935–40. 15. Husebye EE, Lyberg T, Opdahl H, Laurvik H, Røise O. Cardiopulmonary response to reamed intramedullary nailing of the femur comparing traditional reaming with a one-step reamer–irrigator–aspirator reaming system: an experimental study in pigs. J Trauma 2010;69:E6–14. 16. Husebye EE, Lyberg T, Roise O. Bone marrow fat in the circulation: clinical entities and pathophysiological mechanisms. Injury 2006;37(Suppl 4):S8–18. 17. Husebye EE, Opdahl H, Røise O, et al. Coagulation, fibrinolysis, and cytokine responses to intramedullary nailing of the femur. An experimental study in pigs comparing traditional reaming and reaming with a one-step reamer–irrigator– aspirator system. Injury 2010 Jul 20. 18. Jacobovitz-Derks D, Derks CM. Pulmonary neutral fat embolism in dogs. Am J Pathol 1979;95:29–42. 19. Kim YH, Kim JS, Hong KS, et al. Prevalence of fat embolism after total knee arthroplasty performed with or without computer navigation. J Bone Joint Surg Am 2008;90:123–8. 20. Krebs J, Ferguson SJ, Hoerstrup SP, et al. Influence of bone marrow fat embolism on coagulation activation in an ovine model of vertebroplasty. J Bone Joint Surg Am 2008;90:349–56. 21. Lin E, Calvano SE, Lowry SF. Inflammatory cytokines and cell response in surgery. Surgery 2000;127:117–26. 22. Manning JB, Bach AW, Herman CM, Carrico CJ. Fat release after femur nailing in the dog. J Trauma 1983;23:322–6. 23. McIntyre KM, Sasahara AA. The hemodynamic response to pulmonary embolism in patients without prior cardiopulmonary disease. Am J Cardiol 1971;28:288–94.
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
24. Nau T, Aldrian S, Koenig F, Vecsei V. Fixation of femoral fractures in multipleinjury patients with combined chest and head injuries. ANZ J Surg 2003;73: 1018–21. 25. Nurmi JT, Jarvinen TL, Kannus P, et al. Compaction versus extraction drilling for fixation of the hamstring tendon graft in anterior cruciate ligament reconstruction. Am J Sports Med 2002;30:167–73. 26. Orsini EC, Byrick RJ, Mullen JB, et al. Cardiopulmonary function and pulmonary microemboli during arthroplasty using cemented or non-cemented components. The role of intramedullary pressure. J Bone Joint Surg Am 1987;69:822–32. 27. Papathanasopoulos A, Nikolaou V, Petsatodis G, Giannoudis PV. Multiple trauma: an ongoing evolution of treatment modalities? Injury 2009;40:115–9. 28. Pape HC, van Griensven M, Hildebrand FF, et al. Systemic inflammatory response after extremity or truncal fracture operations. J Trauma 2008;65:1379–84. 29. Pape HC, Grimme K, van Griensven M, et al. Impact of intramedullary instrumentation versus damage control for femoral fractures on immunoinflammatory parameters: prospective randomized analysis by the EPOFF Study Group. J Trauma 2003;55:7–13. 30. Pape HC, Regel G, Dwenger A, et al. Influence of thoracic trauma and primary femoral intramedullary nailing on the incidence of ARDS in multiple trauma patients. Injury 1993;24(Suppl 3):S82–103. 31. Pape HC, Schmidt RE, Rice J, et al. Biochemical changes after trauma and skeletal surgery of the lower extremity: quantification of the operative burden. Crit Care Med 2000;28:3441–8. 32. Riska EB, von Bonsdorff H, Hakkinen S, et al. Prevention of fat embolism by early internal fixation of fractures in patients with multiple injuries. Injury 1976;8: 110–6.
S15
33. Robinson CM, Ludlam CA, Ray DC, et al. The coagulative and cardiorespiratory responses to reamed intramedullary nailing of isolated fractures. J Bone Joint Surg Br 2001;83:963–73. 34. Schult M, Kuchle R, Hofmann A, et al. Pathophysiological advantages of rinsing– suction–reaming (RSR) in a pig model for intramedullary nailing. J Orthop Res 2006;24:1186–92. 35. Soloway HB, Robinson EF, Sleeman HK, et al. Resolution of experimental fat embolism. Arch Pathol 1970;90:230–4. 36. Stavlas P, Giannoudis PV. Bilateral femoral fractures: does intramedullary nailing increase systemic complications and mortality rates? Injury 2009;40:1125–8. 37. Tapson VF. Acute pulmonary embolism. N Engl J Med 2008;358:1037–52. 38. Tapson VF. Acute pulmonary embolism. Cardiol Clin 2004;22:353–65, v. 39. Waydhas C, Nast-Kolb D, Trupka A, et al. Posttraumatic inflammatory response, secondary operations, and late multiple organ failure. J Trauma 1996;40:624–30; discussion: 30–1. 40. Wenda K, Ritter G, Ahlers J, von Issendorff WD. [Detection and effects of bone marrow intravasations in operations in the area of the femoral marrow cavity]. Unfallchirurg 1990;93:56–61. 41. Wenda K, Runkel M, Degreif J, Ritter G. Pathogenesis and clinical relevance of bone marrow embolism in medullary nailing – demonstrated by intraoperative echocardiography. Injury 1993;24(Suppl 3):S73–81. 42. Wood KE. Major pulmonary embolism: review of a pathophysiologic approach to the golden hour of hemodynamically significant pulmonary embolism. Chest 2002;121:877–905.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Intravasation of bone marrow content. Can its magnitude and effects be modulated by low pressure reaming in a porcine model? Elisabeth Ellingsen Husebyea, *, Torstein Lybergb , Helge Opdahlc , Olav Røisea,d a Department
of Orthopaedics, Oslo University Hospital, Ullevaal, Norway for Clinical Research, Oslo University Hospital, Ullevaal, Norway c The Norwegian National Center for NBC Medicine, Oslo University Hospital, Ullevaal, Norway d Division of Critical Care, Oslo University Hospital and Faculty of Medicine University of Oslo, Norway b Center
article info
abstract
Keywords: Suction irrigation reaming Intramedullary reaming Femoral shaft fracture Animal study Porcine model Bone marrow embolization Inflammation and coagulation response Intramedullary pressure RIA
Introduction: Intramedullary orthopaedic procedures may increase the intramedullary pressure (IMP) and thereby cause intravasation of bone marrow contents. In recent studies by the authors the reamer–irrigator–aspirator (RIA) has been demonstrated to reduce IMP and coagulation-, fibrinolysis- and cytokine responses, but did not prove any significant difference in cardiopulmonary function parameters or numbers of emboli when compared to a traditional reaming (TR) system. The correlations between IMP increase, regardless type of reamer, and inflammatory- and coagulation responses, pulmonary embolization, and cardiopulmonary alterations have, however, not previously been analyzed in this material. Our hypothesis was that a lower IMP would result in reduced occurrence of pulmonary embolization, reduced inflammatoryand coagulation responses, as well as reduced cardiopulmonary alterations. Materials and Methods: Twenty-eight young Norwegian landrace pigs were exposed to femoral intramedullary reaming, with either the TR (n = 10) or the RIA (n = 10) system, or used as controls (n = 8). IMP was recorded during reaming and nailing. Serial blood samples for demonstration of coagulation-, fibrinolysis-, and cytokine activation were withdrawn peroperatively and until 72 hours post nail insertion. Circulatory and pulmonary effects were monitored peroperatively and until two hours postoperatively. The animals were sacrificed 72 hours post nail insertion and lung tissue biopsies were harvested and examined for lung emboli. Results and Conclusions: A strong correlation between increased IMP and increased coagulationand cytokine responses was found. The number of emboli was not significantly correlated to IMP, but was strongly correlated to changes in the coagulation- and cytokine responses. No clinical relevant correlations were observed between increased IMP or numbers of lung emboli and changes in hemodynamic- or pulmonary function parameters. A correlation between coagulation activation and cytokine activation was observed. This study confirms the connection between increased IMP, increased coagulation activation and the magnitude of pulmonary emboli in a model evaluating the effects of intramedullary reaming of intact pig femora. In this model, the lowering of IMP during reaming, as obtained with RIA, reduced the magnitude of and the effects of bone marrow extravasation. © 2010 Elsevier Ltd. All rights reserved.
Introduction Early stabilization of femoral shaft fractures is associated with reduced pulmonary complication and mortality.3,4,12,24,32 On the other hand, early intramedullary nailing (IMN) has also been associated with increased inflammation and morbidity, and increased incidence of multi organ failure (MOF).11,30,31,36,39 Intramedullary * Corresponding author. Elisabeth Ellingsen Husebye, MD. Department of Orthopaedics, Oslo University Hospital, Ullevaal, Kirkeveien 166, 0407 Oslo, Norway. Tel.: +47 22118080/+47 23027884; fax: +47 23027410. E-mail addresses: [email protected] (E.E. Husebye); [email protected] (T. Lyberg); [email protected] (H. Opdahl); [email protected] (O. Røise). 0020-1383/ $ – see front matter © 2010 Elsevier Ltd. All rights reserved.
procedures may increase the intramedullary pressure (IMP) and thereby cause intravasation of bone marrow content.19,27,40,41 The resulting clinical manifestation probably depends on the quantity of bone marrow elements entering the circulation and the magnitude of clogging of the pulmonary circulation.6,16,23,26,37,42 An association between the magnitude of a surgical procedure and the extent of inflammatory response to the operative trauma has been demonstrated.1,13,28,29 A reamer–irrigator–aspirator (RIA) system has previously been demonstrated by the present authors to reduce IMP,14 reduce coagulation-, fibrinolytic- and cytokine responses,17 but demonstrated no significant differences in cardiopulmonary function parameters,15 neither significant lowering of the number of pulmonary emboli15 when compared to a traditional reaming
S10
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
(TR) system. Possible correlations between IMP, pulmonary embolization, cardiopulmonary function parameters and inflammatory responses have previously not been explored. Our hypothesis was that keeping the IMP low during reaming of the femoral canal would result in reduced occurrence of pulmonary embolization, reduced inflammatory- and coagulation responses, and decreased cardiopulmonary alterations. Data from animals, where reaming was performed with either the RIA or the TR system was therefore pooled for this analysis of the impact of high and low IMPs. Material and methods The study was conducted in accordance with “Regulations on Animal Experimentation” under The Norwegian Animal Welfare Act and approved by the Norwegian Animal Research Authority. Twenty-eight Norwegian landrace pigs (15 males and 13 females), age 3–3.5 months, weighing 27.9–38.0 kg, were used. The reaming groups (TR and RIA group) consisted of 10 animals each. Eight animals were used as controls. Six animals were excluded due to anaesthesia-related events prior to operation, exsanguinations or a shorter than protocol stabilization time. Before surgery the animals were housed one or two animals per stall and returned to individual stalls after the operation. The size of the cages was 200 cm × 300 cm. Room temperature was 20±2ºC, with 40% ± 5% humidity and there was a 12 hour light/dark cycle. The pigs were fed a standard diet, prior to surgery they fasted overnight, but had free access to water. Choice of experimental model As one of the goals of the study was to evaluate the effects of low pressure reaming, represented by RIA reaming, and the RIA-reamer heads were available with the smallest diameter of 12 mm, the experimental models were limited to large animal models, however, the animals should be handable. The young pigs chosen for the study had open epiphysis. To determine the localization of the epiphysis and the shape and dimensions of the femur, radiographs of the femur of a pig used for other research purposes were taken. The anatomy of the femur showed a tight isthmus (in the TR group: 2.14±0.03 and in the RIA group: 2.15±0.04 cm8 ) and a much wider condylar area. The femoral neck was short compared to humans. Furthermore, the femora from two other pigs were divided longitudinally for visual evaluation of the epiphysis (Fig. 1). The preplanning included also a surgical approach, and the evaluation of the femur after reaming with the RIA system (Fig. 1).
Pre-reaming procedures The animals were orally intubated and mechanically ventilated. The anaesthesia was induced and maintained with intravenous propofol (Diprivan® , 20 mg/ml, AstraZeneca, Caponago, Italy) and fentanyl (Fentanyl® , 50 mg/ml, Alpharma, Linz, Austria). 10–15 ml/kg/h Ringer-acetate solution was administrated intravenously during the anaesthesia period. The animals were sacrificed on the third postoperative day. A detailed procedure is described by Husebye et al..15 A flow-directed pulmonary artery (PA) catheter (744HF75, SwanGanz CCOmbo CCO/SvO2 catheter 7.5F, Edwards Critical-Care Division, Irvine, CA, USA) was inserted in the left internal jugulary vein with the tip in the pulmonary artery. An 18-gauge catheter was inserted into the carotid artery. An 18-gauge catheter was inserted in the left femoral vein. These catheters were used for arterial, mixed venous and femoral vein blood sampling during the operation and until two hours after the nail was inserted. The catheter in the carotid artery was two hours post nail insertion connected to a vascular access port (VAP) (Celsite® Access Port ST301 SIL (2.8 × 1.1), B-Braun Medical, Chasseneuil, France), placed in the posterolateral aspect of the neck, for arterial blood sampling on awake animals six hours to 72 hours post nail insertion. Two hours after the insertion of the nail the femoral vein catheter and the pulmonary artery catheters were removed. Peroperatively all catheters were flushed with a solution of heparin (Heparin Leo® , 5000 U/ml, LEO Pharma A/S, Ballerup, Denmark) and Macrodex (Macrodex® , 60 mg/ml with sodium chloride, Pharmalink AB, Upplands Vasby, ¨ Sweden) as described by Husebye et al..15 The total dose of heparin given during the study period was approximately 2000 U. Mean pulmonary artery pressure (MPAP), central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), heart rate (HR), mean arterial pressure (MAP), cardiac output (CO) and mixed venous oxygen saturation (SvO2 ) were monitored. Systemic (SVR) and pulmonary vascular resistance (PVR) and the alveolo-arterial oxygen (PA O2 − Pa O2 ) differences were calculated. Indexed values were not used due to uniform size of the animals. The registration of IMP was performed with a transducer tipped pressure monitoring catheter with a diameter of 1.35 mm (Integra NeuroScienceTM Camino1, Model 110–4B, San Diego, USA), placed just cranial to the distal femoral epiphysis (Fig. 1). The pressure catheter was removed 10 minutes after insertion of the nail. The procedure is described in detail by Husebye et al..14 The time schedule for blood sampling and cardiopulmonary and IMP registrations is given in Table 1. Surgical technique
Fig. 1. A longitudinally divided pig femur post RIA reaming with distal femoral epiphysis (white arrow) and drill bit located proximal to the epiphysis, where the intramedulary pressure catheter was placed.
Through a longitudinal incision the access to the proximal femur was made. The femoral marrow cavity was opened with an awl, a guide-wire was introduced, and antegrade reaming and nailing was performed. For traditional reaming (Stryker, Kiel, Germany, Model: Bixcut, 0226/3095/3100/3105/3110/3115/3120) a sequential reaming, using increasing steps of 0.5 mm in reamer diameter (9.5 mm – 12 mm) was used. A 12 mm reamer head was used for the one-step RIA (Synthes, Paoli, PA, USA, 325.250S) technique. The RIA equipment was connected to a flushing tube and a suction tube with an internal lumen of 5.5 mm and 30 mm H2 Osuction force. A standard Große-Kempf nail (Stryker Trauma GmbH, Schonkirchen, ¨ Germany), cut at a length of 10 cm, with a diameter of 10 mm was used and left inside. The animals in the control group were exposed to a sham operation with no reaming or nailing of the femoral cavity.
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
S11
Table 1 Time schedule for data related to surgical procedures a Time
Surgical procedure
Blood samples
Cardiopulmonary registrations
IMP registration
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
Blood sample location a m
v
Coagulation, fibrinolysis, cytokines, blood gases
x
x
A
Baseline
B
(A+ 15 min)
After reaming
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
C
(B+ 7 min)
After nail insertion
Coagulation, fibrinolysis, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
x
D
(C+ 30 min)
30 min after nail insertion
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
E
(D+ 90 min)
2 h after nail insertion
Coagulation, fibrinolysis, cytokines, blood gases
x
x
x
CO, SVR, PVR, HR, MAP, CVP, MPAP, PCWP, SvO2 , PA O2 –Pa O2 difference
F
(E+ 4 h)
6 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
G
(F+ 18 h)
24 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
H
(G+ 24 h)
48 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
I
(H+ 24 h)
72 h after nail insertion
Coagulation, fibrinolysis, cytokines
x
a
The table shows the time schedule for blood sampling at different sampling locations; arterial (a), mixed venous (m) and femoral vein (v) blood and cardiopulmonary and intramedullary pressure (IMP) registrations and calculations; 75 minutes after completion of soft tissue preparation (baseline, A), after reaming (B), immediately after insertion of the nail (C), and at 30 minutes (D), two (E), six (F), 24 (G), 48 (H) and 72 (I) hours after insertion of the nail.
Blood sampling and assays Blood sampling was performed as described in Table 1. In the control group the time after reaming (B) and after the nail was inserted (C) were set by the mean levels in the TR and RIA groups. The following cytokines were analyzed using commercially available ELISA kits; TNF-a, IL-1b, IL-6, IL-8 and IL-10. For evaluation of the coagulation and fibrinolysis systems, thrombin-antithrombin complex (TAT), tissue plasminogen activator (t-PA) activity and plasminogen activator inhibitor-1 (PAI-1) activity were analyzed. The specifications of the analyzing kits, blood sample handling and storage is described by Husebye et al..17 Blood gases were analyzed (ABL735, 2002, Radiometer Medical ApS, Copenhagen, Denmark) at time points described in Table 1. Lung specimens The lungs were excised immediately after the death, and the hearts examined to exclude the presence of intra-cardiac shunts. Two lung tissue samples were harvested from the periphery of each lung lobe. The total area of the lung histology preparations was calculated using Auto Sketch 6.0 (1998) planimetry program. The incidence of emboli was expressed as numbers of emboli per square centimetre lung area. The preparation of the lung biopsies has previously been described in detail by Husebye et al..15 Statistics Statistical analyses were performed using the Statistical Package for Social Science (SPSS) software, version 16.0 (SPSS Inc, Chicago, IL, USA). The relationship between IMP, lung emboli, inflammatoryand coagulation- and fibrinolytic responses, and cardiopulmonary response were analyzed using Pearson product-moment correlation coefficient (r). For IMP and amount of lung emboli absolute values were used, and for cardiopulmonary alterations and inflammatory-, fibrinolytic- and coagulation responses the difference (D values)
from baseline levels (A) were used. Correlations were considered significant at P levels ≤0.05 and considered strong when r ≤ −0.500 or r > 0.500. Main results The statistics and results are based on the same material used for previous separately published papers.14,15,17 IMP levels were measured in 15 animals. One animal in the RIA group was excluded due to a perioperatively inflicted supracondylar femoral fracture, and three animals in the TR group due to perioperative perforation of the distal medial femoral cortex. The highest IMP in the TR group was demonstrated at the first or the second reaming step. In both the TR and the RIA groups the peak IMP levels were used for the correlation analyses. IMP was not measured in the control group. Lung histology evaluation was performed in 22 study animals. All observed lung emboli were found in small arteries with a diameter of 0.08–2.0 mm.15 In the control group one single lung embolus was found. Coagulation-, fibrinolysis- and cytokine activation were studied in 22 animals. We used the peak levels of arterial blood D TAT, D -PA, D PAI-1, and arterial D IL-6 and femoral vein blood D IL-6 for the statistics. Arterial blood D TAT and D t-PA levels demonstrated a biphasic curve with peak levels at reaming completed (B) and at six hours post nail insertion (F). Arterial blood D PAI-1 levels showed peak levels at six hours post nail insertion (F). The cytokine response was mainly represented by increased procedure-related D IL-6 levels for which a peak was present in arterial blood six hours post nail insertion (F) and in femoral vein blood after nail insertion (C). Hemodynamic and respiratory function was evaluated in 22 study animals. For correlation analyses peak D values and/or D values at those points of registration demonstrating group differences between the TR and RIA reaming. For D CO, D PVR, D SVR, D MPAP, D SvO2 , and D Aa difference the analyses were performed after nail
S12
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
Fig. 2. Scatter plots demonstrating significant, strong positive correlations between peak IMP levels and arterial blood D TAT (2A) and femoral vein blood D TAT levels (2B) at completed reaming, numbers of lung emboli (PE) and arterial blood D TAT (2C) and femoral vein blood D TAT levels (2D) at completed reaming, numbers of lung emboli (PE) and arterial blood D IL-6 at six hours post nail insertion (2E), and arterial blood D TAT at completed reaming and arterial blood D IL-6 at six hours post nail insertion (2F).
insertion (C) and at two hours post nail insertion (E), and for D CVP and D PCWP after nail insertion (C). Intramedullary pressure, coagulation-, fibrinolysis and cytokine responses, and hemodynamic and pulmonary function parameters A strong correlation was demonstrated between peak IMP and arterial blood D TAT levels at reaming (B) (r = 0.516, p = 0.049)
(Fig. 2A) and an even stronger correlation was present between peak IMP and femoral vein blood D TAT levels at reaming (B) (r = 0.759, p = 0.001) (Fig. 2B). No significant correlations were present between peak IMP and peak arterial and femoral vein blood D t-PA and D PAI-1 levels (data not shown). The local activation of IL-6 (femoral vein blood) at nail insertion (C) increased significantly related to increasing peak IMP values (r = 0.554, p = 0.032). We found no correlation between IMP and numbers of lung emboli. No clinical
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
relevant significant correlations were present between increased IMP and hemodynamic- or pulmonary function parameters (data not shown). Numbers of lung emboli, coagulation-, fibrinolysis- and cytokine responses and hemodynamic and pulmonary function parameters A strong correlation was found between numbers of lung emboli and arterial D TAT levels (r = 0.646, p = 0.001) (Fig. 2C) and femoral vein blood D TAT levels (r = 0.547, p = 0.008) (Fig. 2D) at completed reaming (B). No significant correlations were present between numbers of lung emboli and peak arterial and femoral vein blood D t-PA and D PAI-1 levels (data not shown). Increased numbers of lung emboli correlated positively to increased peak arterial D IL-6 levels (r = 0.569, p = 0.006) (Fig. 2E). No clinical relevant correlations were present between increased numbers of lung emboli and hemodynamic- or pulmonary function parameters (data not shown). Coagulation, fibrinolysis and cytokine activation A strong positive correlation was present between arterial D TAT levels at completed reaming and arterial D IL-6 levels at six hours post nail insertion (r = 0.532, p = 0.012) (Fig. 2F). Discussion The reamer–irrigator–aspirator (RIA) system has previously been compared to traditional reaming (TR) by the present authors. These studies demonstrated reduced peak IMP14 (mean 188 mmHg in the TR and mean 33 mmHg in the RIA group), reduced coagulation-, fibrinolytic- and cytokine responses,17 lower numbers of lung emboli (mean 0.31 in the TR and mean 0.21 lung emboli per square centimetre lung area in the RIA group [ns]),15 and no significant differences in cardiopulmonary function parameters,15 when compared to the TR system. The RIA reamer represents a low pressure reaming system. The correlations between IMP increase and pulmonary embolization, inflammatory- and coagulation responses, and the cardiopulmonary alterations were, however, not previously explored in this material. Such analysis disclosed a strong relationship between increased IMP and increased coagulation- and cytokine responses, as well as between increased numbers of lung emboli and coagulation- and cytokine responses. Additionally, a strong relationship between activation of coagulation- and cytokine responses was demonstrated. However, no correlation between IMP and numbers of lung emboli was observed, neither were clinical relevant correlations found between hemodynamic- or pulmonary function parameters and increased IMP numbers of lung emboli. Reamed IMN of the femur represents a surgical trauma. The systemic trauma response is multifactorial and represents a unique interaction between coagulation, fibrinolysis, the complement cascade, vascular endothelial cells, circulating white blood cells, reactive oxygen species, and pro- and anti-inflammatory cytokines. The magnitude of the inflammatory response seems to correlate with the extent of injury (operation or trauma),1,13,28,28 and therefore; the burden of the operation could be evaluated through the response it creates. In this study the cytokine response was mainly represented by increased IL-6, demonstrating peak arterial levels six hours post nail insertion, which is in accordance with other studies10,11,21,29 and peak femoral vein blood IL-6 levels at nail insertion, which is quite early post procedure. Increased IL-6 levels are generally associated with an inflammatory response.2 A strong correlation between peak arterial blood IL-6 levels and numbers of lung emboli, as well as peak femoral vein blood IL-6 levels and IMP were demonstrated, which may indicate a link between
S13
IMP, local inflammatory effects and embolization. Additionally, a strong positive correlation was present between peak arterial TAT at completed reaming and peak arterial IL-6 levels. Both peak arterial and femoral vein blood TAT levels correlated strongly to both increased IMP and increased numbers of lung emboli. These results are in accordance with the literature, where increased coagulation response correlated with the intraoperative embolic response as well as the stimulation of other markers of inflammation.6,7,11,20,33 The evaluation of the consequences of bone marrow embolization is complex; both the estimation of volume of bone marrow material entering the circulation, and the evaluation of the potential harmful effects. The basic requirement is to prove the actual intravasation, then the amount of bone marrow intravasation has to be estimated, and finally the effect of the bone marrow intravasation on the test individuals has to be explored. Wenda et al. found increased IMP during intramedullary reaming and nailing of the femur in experimental studies, and that the elevated IMP resulted in intravasation of bone marrow contents evaluated echocardiographically.40,41 Already at an IMP of 50 mmHg, snow-flurry in the vena cava was demonstrated.41 In that study the degree of embolization seemed to depend directly on the pressure level. In our experimental model, comparing IMP increase related to traditional reaming and reaming with the RIA device, the mean IMP levels were 188 and 33 mmHg, respectively.14 The IMP related to RIA was below the IMP level at which circulating embolic material could be echocardiographically detected.41 We could, however, not demonstrate a significant correlation between increasing IMP levels and numbers of lung emboli in lung tissue. Echocardiographical verification of intravascular bone marrow content and even examination of the fat emboli material withdrawn from venous blood during IMN, gives the opportunity to detect the actual presence and estimate the volume of bone marrow content in the blood stream in a semiquantitative manner. As our experimental model was already complex, such investigations were not included. Retrospectively, obtaining an estimate of the volume of embolic material and analyzing its correlations to IMP, numbers of lung emboli, coagulation-, fibrinolysis- and cytokine responses and hemodynamic or pulmonary function would have been interesting. Histological evaluation of lung tissue after IMN in experimental models has demonstrated the presence of bone marrow emboli.18,35 The highest concentration of emboli was found within the first postoperative day, with equally distribution to all lung lobes.5,20 In our model the animals were sacrificed on the third postoperative day, which according to the literature does not represent the time where most pulmonary emboli are expected to be found.18,35 Several different methods for analyzing the amount of fat in the lung parenchyma, have, however, been applied,9,18,26,34 which make the comparison of the amount of pulmonary emboli difficult. The magnitude of hemodynamic and pulmonary changes after pulmonary embolism has been demonstrated to be a function of size and numbers of the emboli and the underlying state of cardiopulmonary function.6,42 The exact amount of emboli needed to cause cardiopulmonary dysfunction has not been established. The vessels in a normal pulmonary circulation are easily dilated when the pressure increases, therefore both pulmonary arterial pressures and vascular resistance may remain almost unchanged despite occlusion. In older literature it has been estimated that 30–50% of a previous normal pulmonary circulation has to be occluded before hemodynamic alterations can be detected.23 The hemodynamic alterations related to microscopic pulmonary embolism and microthrombosis associated with conditions like acute lung injury (ALI) and adult respiratory distress syndrome (ARDS) could be circulatory failure through the effect of increased mean pulmonary artery pressure (MPAP).26 The right ventricle can be expected to fail when 75% of the pulmonary vasculature
S14
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
becomes obstructed.38 Such failure leads to end-diastolic dilatation and displacement of the septum wall, which results in reduced end-diastolic filling of the left ventricle37,42 and further aggravates the circulatory failure. In our experimental model the animals were young and healthy, and the embolic load was probably not sufficiently massive to cause marked cardiopulmonary alterations, and therefore we found no clinical relevant correlations between numbers of lung emboli or parameters of cardiopulmonary function. The length of the pig femur relative to body size is about 50% shorter than in humans (approximately 12–13% of the body length versus 25% in humans) resulting in a smaller potential volume of bone marrow content that can be extruded into the vascular system of the pigs. Additionally, the trabecular bone is most probably denser, the cortical bone less dense, and the venous circulation better when compared to human adult bone.25 Both a wider and shorter bone and denser trabecular bone are conditions that probably lower the IMP increase during intramedullary reaming. In the present model we used intact femora. A fracture may represent a “first-hit” in the pulmonary circulation, which might have made a procedure-related increase in PVR easier to detect. On the other hand, the presence of a superposed fracture would probably have lowered the increases of IMP, as the fracture may serve as a vent and allow bone marrow escape via the fracture diastases during reaming. The length of the pig femora were approximately 15 cm, and the distance from fossa piriformis to the distal femoral epiphysis was 11–12 cm.8 The short femur and additionally the location of the femur in the pig make a fracture model complicated and unreliable. An osteotomy of the femur as a fracture substitute, however, is possible, but with the length of the porcine femur, the created IMP would probably be too low to test the study hypothesis. In an experimental study of femoral IMN on intact and fractured canine femora the release of triglycerides was greater in a non-fracture model,22 suggesting more bone marrow content intravasation in intact femora. As the effect of bone marrow embolization during reaming and nailing was one goal of our study, we are not certain that inducing a femoral fracture prior to surgery would have increased the pathophysiological changes. In this experimental model we demonstrated that femoral low pressure reaming had beneficial effects on IMP and coagulationand cytokine responses. We could, however, not demonstrate that this reduced IMP had a direct and significant effect on numbers of lung emboli. In future studies echocardiography should be included for the evaluation of magnitude of bone marrow content intravasation during reaming.
Conclusion In this model on intramedullary reaming and nailing of intact pig femora, there was a strong correlation between the numbers of lung emboli and increased coagulation- and cytokine activation. The coagulation- and cytokine activation correlated also strongly to the procedure-related IMP increase. The numbers of lung emboli were, however, not directly correlated to IMP increase or changes in the circulation or pulmonary gas exchange. Overall, in this model, low pressure reaming seems to reduce the magnitude and the effect of the bone marrow intravasation.
Competing interests Each of the authors certifies that he or she has no commercial associations that might pose a conflict of interest in connection with the submitted manuscript. The manuscript has not been submitted or published elsewhere.
Acknowledgements The pig study was supported by grants from the AO Research Fund of the AO Foundation (Project no. 04-R71) and Synthes. Operating tools were delivered by Stryker and Synthes. The authors wish to express their gratitude to Ingeborg Løstegaard Goverud at the Department of Pathology, Oslo University Hospital, Ullevaal, Norway, for excellent histological preparations and to Trude Aspelin and Lisbeth Sætre, Center for Clinical Research, Oslo University Hospital, Ullevaal, Norway, for carefully analyzing the blood samples. We also acknowledge the staff, and in particular Morten Eriksen, at the Institute of Experimental Medical Research, University of Oslo, for their support. References 1. Baigrie RJ, Lamont PM, Kwiatkowski D, et al. Systemic cytokine response after major surgery. Br J Surg 1992;79:757–60. 2. Biffl WL, Moore EE, Moore FA, Peterson VM. Interleukin-6 in the injured patient. Marker of injury or mediator of inflammation? Ann Surg 1996;224:647–64. 3. Bone LB, Johnson KD, Weigelt J, Scheinberg R. Early versus delayed stabilization of femoral fractures. A prospective randomized study. J Bone Joint Surg Am 1989;71:336–40. 4. Brundage SI, McGhan R, Jurkovich GJ, et al. Timing of femur fracture fixation: effect on outcome in patients with thoracic and head injuries. J Trauma 2002; 52:299–307. 5. Cheung NM CW, Leung KS. A study of the pathway of marrow fat intravasation during intramedullary reaming. Osteo Trauma Care 2004;12:140–4. 6. Christie J, Robinson CM, Pell AC, et al. Transcardiac echocardiography during invasive intramedullary procedures. J Bone Joint Surg Br 1995;77:450–5. 7. Christie J, Robinson CM, Singer B, Ray DC. Medullary lavage reduces embolic phenomena and cardiopulmonary changes during cemented hemiarthroplasty. J Bone Joint Surg Br 1995;77:456–9. 8. Ellingsen Husebye E, Lyberg T, Madsen JE, et al. The early effects of intramedullary reaming of the femur on bone mineral density; an experimental study in pigs. Scand J Surg 2009;98:189–94. 9. Elmaraghy AW, Aksenov S, Byrick RJ, et al. Pathophysiological effect of fat embolism in a canine model of pulmonary contusion. J Bone Joint Surg Am 1999;81:1155–64. 10. Gebhard F, Pfetsch H, Steinbach G, et al. Is interleukin 6 an early marker of injury severity following major trauma in humans? Arch Surg 2000;135:291–5. 11. Giannoudis PV, Smith RM, Bellamy MC, et al. Stimulation of the inflammatory system by reamed and unreamed nailing of femoral fractures. An analysis of the second hit. J Bone Joint Surg Br 1999;81:356–61. 12. Goris RJ, Gimbrere JS, van Niekerk JL, et al. Early osteosynthesis and prophylactic mechanical ventilation in the multitrauma patient. J Trauma 1982;22:895–903. 13. Harwood PJ, Giannoudis PV, van Griensven M, et al. Alterations in the systemic inflammatory response after early total care and damage control procedures for femoral shaft fracture in severely injured patients. J Trauma 2005;58:446–52; discussion: 52–4. 14. Husebye EE, Lyberg T, Madsen JE, et al. The influence of a one-step reamer– irrigator–aspirator technique on the intramedullary pressure in the pig femur. Injury 2006;37:935–40. 15. Husebye EE, Lyberg T, Opdahl H, Laurvik H, Røise O. Cardiopulmonary response to reamed intramedullary nailing of the femur comparing traditional reaming with a one-step reamer–irrigator–aspirator reaming system: an experimental study in pigs. J Trauma 2010;69:E6–14. 16. Husebye EE, Lyberg T, Roise O. Bone marrow fat in the circulation: clinical entities and pathophysiological mechanisms. Injury 2006;37(Suppl 4):S8–18. 17. Husebye EE, Opdahl H, Røise O, et al. Coagulation, fibrinolysis, and cytokine responses to intramedullary nailing of the femur. An experimental study in pigs comparing traditional reaming and reaming with a one-step reamer–irrigator– aspirator system. Injury 2010 Jul 20. 18. Jacobovitz-Derks D, Derks CM. Pulmonary neutral fat embolism in dogs. Am J Pathol 1979;95:29–42. 19. Kim YH, Kim JS, Hong KS, et al. Prevalence of fat embolism after total knee arthroplasty performed with or without computer navigation. J Bone Joint Surg Am 2008;90:123–8. 20. Krebs J, Ferguson SJ, Hoerstrup SP, et al. Influence of bone marrow fat embolism on coagulation activation in an ovine model of vertebroplasty. J Bone Joint Surg Am 2008;90:349–56. 21. Lin E, Calvano SE, Lowry SF. Inflammatory cytokines and cell response in surgery. Surgery 2000;127:117–26. 22. Manning JB, Bach AW, Herman CM, Carrico CJ. Fat release after femur nailing in the dog. J Trauma 1983;23:322–6. 23. McIntyre KM, Sasahara AA. The hemodynamic response to pulmonary embolism in patients without prior cardiopulmonary disease. Am J Cardiol 1971;28:288–94.
E.E. Husebye et al. / Injury, Int. J. Care Injured 41 (2010) S9–S15
24. Nau T, Aldrian S, Koenig F, Vecsei V. Fixation of femoral fractures in multipleinjury patients with combined chest and head injuries. ANZ J Surg 2003;73: 1018–21. 25. Nurmi JT, Jarvinen TL, Kannus P, et al. Compaction versus extraction drilling for fixation of the hamstring tendon graft in anterior cruciate ligament reconstruction. Am J Sports Med 2002;30:167–73. 26. Orsini EC, Byrick RJ, Mullen JB, et al. Cardiopulmonary function and pulmonary microemboli during arthroplasty using cemented or non-cemented components. The role of intramedullary pressure. J Bone Joint Surg Am 1987;69:822–32. 27. Papathanasopoulos A, Nikolaou V, Petsatodis G, Giannoudis PV. Multiple trauma: an ongoing evolution of treatment modalities? Injury 2009;40:115–9. 28. Pape HC, van Griensven M, Hildebrand FF, et al. Systemic inflammatory response after extremity or truncal fracture operations. J Trauma 2008;65:1379–84. 29. Pape HC, Grimme K, van Griensven M, et al. Impact of intramedullary instrumentation versus damage control for femoral fractures on immunoinflammatory parameters: prospective randomized analysis by the EPOFF Study Group. J Trauma 2003;55:7–13. 30. Pape HC, Regel G, Dwenger A, et al. Influence of thoracic trauma and primary femoral intramedullary nailing on the incidence of ARDS in multiple trauma patients. Injury 1993;24(Suppl 3):S82–103. 31. Pape HC, Schmidt RE, Rice J, et al. Biochemical changes after trauma and skeletal surgery of the lower extremity: quantification of the operative burden. Crit Care Med 2000;28:3441–8. 32. Riska EB, von Bonsdorff H, Hakkinen S, et al. Prevention of fat embolism by early internal fixation of fractures in patients with multiple injuries. Injury 1976;8: 110–6.
S15
33. Robinson CM, Ludlam CA, Ray DC, et al. The coagulative and cardiorespiratory responses to reamed intramedullary nailing of isolated fractures. J Bone Joint Surg Br 2001;83:963–73. 34. Schult M, Kuchle R, Hofmann A, et al. Pathophysiological advantages of rinsing– suction–reaming (RSR) in a pig model for intramedullary nailing. J Orthop Res 2006;24:1186–92. 35. Soloway HB, Robinson EF, Sleeman HK, et al. Resolution of experimental fat embolism. Arch Pathol 1970;90:230–4. 36. Stavlas P, Giannoudis PV. Bilateral femoral fractures: does intramedullary nailing increase systemic complications and mortality rates? Injury 2009;40:1125–8. 37. Tapson VF. Acute pulmonary embolism. N Engl J Med 2008;358:1037–52. 38. Tapson VF. Acute pulmonary embolism. Cardiol Clin 2004;22:353–65, v. 39. Waydhas C, Nast-Kolb D, Trupka A, et al. Posttraumatic inflammatory response, secondary operations, and late multiple organ failure. J Trauma 1996;40:624–30; discussion: 30–1. 40. Wenda K, Ritter G, Ahlers J, von Issendorff WD. [Detection and effects of bone marrow intravasations in operations in the area of the femoral marrow cavity]. Unfallchirurg 1990;93:56–61. 41. Wenda K, Runkel M, Degreif J, Ritter G. Pathogenesis and clinical relevance of bone marrow embolism in medullary nailing – demonstrated by intraoperative echocardiography. Injury 1993;24(Suppl 3):S73–81. 42. Wood KE. Major pulmonary embolism: review of a pathophysiologic approach to the golden hour of hemodynamically significant pulmonary embolism. Chest 2002;121:877–905.