- Email: [email protected]
The effect of irrigation time and type of irrigation fluid on cartilage surface friction
Author’s Accepted Manuscript The effect of irrigation time and type of irrigation fluid on cartilage surface friction F. Stärke, F. Awiszus, C.H. Lohmann, C. Stärke
www.elsevier.com/locate/jmbbm
PII: DOI: Reference:
S1751-6161(17)30391-0 http://dx.doi.org/10.1016/j.jmbbm.2017.09.008 JMBBM2492
To appear in: Journal of the Mechanical Behavior of Biomedical Materials Received date: 5 July 2017 Revised date: 29 August 2017 Accepted date: 4 September 2017 Cite this article as: F. Stärke, F. Awiszus, C.H. Lohmann and C. Stärke, The effect of irrigation time and type of irrigation fluid on cartilage surface friction, Journal of the Mechanical Behavior of Biomedical Materials, http://dx.doi.org/10.1016/j.jmbbm.2017.09.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The effect of irrigation time and type of irrigation fluid on cartilage surface friction F. Stärke, F. Awiszus, C.H. Lohmann, C. Stärke Orthopedics Department, University Hospital Magdeburg, Germany
Corresponding author: Dr. med. Christian Stärke, MD, PhD Orthopedics Department University Hospital Magdeburg, Germany Leipziger Straße 44 39120 Magdeburg E-Mail: [email protected] Tel.: 0049 391 6714022
Abstract: It is known that fluid irrigation used during arthroscopic procedures causes a wash-out of lubricating substances from the articular cartilage surface and leads to increased friction. It was the goal of this study to investigate whether this effect depends on the time of irrigation and type of fluid used. Rabbit hind legs were used for the tests. The knees were dissected and the friction coefficient of the femoral cartilage measured against glass in a boundary lubrication state. To determine the influence of irrigation time and fluid, groups of 12 knees received either no irrigation (control), 15, 60 or 120 min of irrigation with lactated Ringer’s solution or 60 minutes of irrigation with normal saline or a sorbitol/mannitol solution. The time of irrigation had a significant effect on the static and kinetic coefficient of friction (CoF), as had the type of fluid. Longer irrigation time with Ringer’s solution was associated with increased friction coefficients (relative increase of the kinetic CoF compared to the control after 15, 60 and 120 min: 16%, 76% and 88% respectively). The sorbitol/mannitol solution affected the static and kinetic CoF significantly less than either Ringer’s or normal saline. The washout of lubricating glycoproteins from the cartilage surface and the associated increase of friction can be effectively influenced by controlling the time of irrigation and type of fluid used. The time of exposure to the irrigation fluid should be as short as possible and monosaccharide solutions might offer a benefit compared to salt solutions in terms of the resultant friction.
Keywords: knee cartilage friction arthroscopy lubrication
Introduction
Arthroscopic techniques are commonly considered to be less invasive and less harmful than open procedures. During arthroscopy an irrigation fluid, e.g. normal saline, lactated Ringer’s solution, or a carbohydrate solution, is commonly used for the distension of the joint and to allow for a sufficient vision. While certain adverse effects of irrigation fluids have been demonstrated repeatedly in the past(Jurvelin et al., 1994; Reagan et al., 1983), these insights have little impact on the clinical practice.
However, recently it was also demonstrated in a bovine model that irrigation with Ringer’s solution causes a wash-out of Lubricin (aka PRG4 or Superficial Zone Protein, SZP), which is associated with an increased (CoF)(Teeple et al., 2016). Lubricin has been recognized as an important surface proteoglycan, that is mandatory to maintain the ultra-low friction properties of cartilage and joint homeostasis(Jay and Waller, 2014; Peng et al., 2015; Waller et al., 2013). A depletion of surface proteoglycans would increase friction, which again increases shear forces in the cartilage(Wong et al., 2008). Increased shear has been linked to chondrocyte apoptosis(Lane Smith et al., 2000; Smith et al., 2004). Thus, a depletion of surface proteoglycans and altered friction properties could make the joint cartilage more vulnerable to external loads and be an underrecognized adverse effect of standard arthroscopic procedures. While the effects described above could be clearly shown for a bovine knee model using a fixed amount of Ringer’s solution as irrigation fluid(Teeple et al., 2016), it remains unclear whether other types of irrigation fluids would affect friction properties of cartilage in the same way and whether the duration of irrigation is a factor. Both questions could have an impact on clinical practice as they can potentially be addressed during surgery.
The goal of the study was therefore to investigate in a rabbit model whether the CoF of cartilage is affected by the type of the irrigation fluid used and by the time of exposure. It was our hypothesis that irrigation causes an increased friction, which is time dependent and related to the type of fluid.
Methods
Commercially available whole rabbit legs, which had been deep frozen immediately after death of the animals, were used for the study. Prior to the tests, the legs were thawed at room temperature for 6 -8 hours. The legs were then carefully dissected to expose the knee joint. All knee ligaments were sequentially cut until the tibia and femur were separated. Particular care was taken at this point not to touch or wipe the articulating surfaces to avoid mechanical removal of surface proteoglycans. The knees were then temporarily stored at 6° C in boxes with saturated moist air. Generally, friction tests were done within 6 – 8 hours of specimen preparation. Prior to the tests the knees were warmed to ambient temperature, which was 23° C during all experiments.
There were 72 knees used in the study with N=12 per group. Following scenarios were investigated:
-control group, i.e. no irrigation (t0)
-15 min irrigation with lactated Ringer’s solution (t15_ring) -60 min irrigation with lactated Ringer’s solution (t60_ring) -120 min irrigation with lactated Ringer’s solution (t120_ring)
-60 min irrigation with Normal saline (t60_nacl) -60 min irrigation with a sorbitol/mannitol solution (Purisole SM, Fresenius Kabi, Germany; 27g Sorbitol + 5,4g Mannitol /l; t60_puri).
For the process of irrigation, groups of 6 knees were placed in 1000 ml of fresh irrigation fluid of the respective type. The fluids were sterile, standard hospital preparations. An electrical pump kept the fluid moving continuously to simulate intraoperative conditions. The knees were stored in saturated moist air at room temperature until the friction test, which was accomplished generally within one hour after the irrigation.
Friction testing A linear reciprocating tribometer was custom-designed to meet the specific requirements of the experiments (Fig. 1). Interchangeable glass slides were used as the friction partner for the cartilage. The rabbit femora were directly attached to the friction force sensor using custom-made 3D-printed fixtures (Fig. 1).
A gauged weight, born on the friction force sensor, was used to apply a defined normal force to the specimen. As variations of the individual bone anatomy could affect the true normal force, a load cell was placed below the tribometer, allowing to measure the effective normal force accurately for each experiment. The function of this custom-made tribometer was validated with a number of pilot experiments, e.g. testing 5 mm PTFE spheres against glass (12 tests, speed and normal load as in the actual study) yielded highly repeatable measurements with a kinetic CoF of 0.04 (SD 0.004). This value fits published data, gained under comparable conditions, very well(Pooley and Tabor, 1972).
For the main experiments, the rabbit femora were fixed to the friction force sensor at an angle of 45 degrees in its long axis and with an internal or external rotation of approximately 20 degrees. Thereby, only one condyle was in contact with the glass slide, which served as the friction partner (Fig. 1). For each test a fresh glass slide was used. The average (measured) normal force was 0.82 Newton (SD
0,048). Although not necessary for the calculation of the CoF, the average contact area was determined in a pilot study in 12 specimens. For this test the knees were dissected and fixed in the tribometer as described earlier. The condyle in contact with the glass slide was dried with a paper towel and stained with black ink. A sheet of scaled paper was placed between the condyle and the glass slide, thus leaving the footprint of the contact zone on the paper. A photo copy of the paper was imported into ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997-2016 ) and the area of the contact zone exactly determined. An average contact area of 4.22 (SD 0.45) mm^2 was found, giving an average contact pressure of 0.19 MPa. The test protocol was adopted from earlier studies published elsewhere(Gleghorn et al., 2007; Peng et al., 2015): First, the respective femur was fixed to the tribometer. Four drops of phosphate buffered saline (PBS, ~ 200µl) were applied to the glass slide, covering the contact area with the femoral condyle and thus preventing drying out. The condyle was then brought in contact with the glass slide, loaded with the normal force and allowed to reach equilibrium for 2 minutes. The sliding motion was then started with a speed of 0.5 mm/s and a way of 10 mm in alternating directions for 3 minutes. The slow speed was chosen to allow the measurements in a boundary lubrication state and to minimize the effect of hydrodynamic lubrication(Hills, 2000; Jahn et al., 2016). Data was directly recorded to a file using a standard PC and MATLAB/DAQtoolbox software (Mathworks, Natick, MA; USA). The friction force was determined from the recorded files. The values for the kinetic CoF were extracted from the data in the middle of the second plateau. The friction force for the static CoF was determined from the second peak, at which point the direction of motion changed from one direction to the other. The static and kinetic CoF(µ) for each knee was calculated from the normal force and the friction force with µ=Ff/Fn.
Statistics Each series of tests (effect of irrigation time and the type of irrigation fluid) was analyzed using an ANOVA, followed by a post hoc comparison with a Scheffé adjustment for multiple comparisons. The SPSS software suite ( v. 23, IBM Corp., Armonk, NY, USA) was used for the analysis.
Results Both, static and kinetic CoF were highly significantly dependent on the irrigation time (p<0.001) and on the type of irrigation fluid (p<0.001).
Irrigation with lactated Ringer’s solution increased the kinetic friction coefficient by 16%, 76% and 88% after 15, 60 and 120 min minutes of exposure respectively (Fig. 2, 3). Post hoc comparison showed
significant differences between 60 min and 120 min of treatment and the control but not between 15 min and control and 60 min and 120 min of exposure. The same effects were shown for the static coefficient of friction.
Concerning the type of irrigation fluid, a highly significant effect was found for the treatment with Purisole, which increased kinetic and static CoF less than either normal saline or lactated Ringer’s solution. The latter both fluids instead did not differ from each other (Fig. 4).
Discussion Exposing knee joint cartilage to arthroscopic irrigation fluids increased generally the coefficient of friction in our model. We assume that this effect is caused by a depletion of surface-glycoproteins like Lubricin(Teeple et al., 2016). Both, irrigation time and type of the irrigation fluid, were related to the magnitude of the effect. While the difference between the non-irrigated controls and 15 min of irrigation was not statistically significant, there is probably not a clear-cut threshold beyond which cartilage friction would be affected. It is reasonable to assume a continuous relationship between the time of exposure and the depletion of surface glycoproteins and thus the resultant friction, although this relationship is probably not linear. This has potential implications for the clinical practice. As increased cartilage friction is associated with several adverse effects on a cellular level(Lane Smith et al., 2000; Wong et al., 2008), limiting irrigation time to the necessary amount seems advisable. Dry arthroscopy might be considered for steps in the arthroscopic procedure not relying on continuous irrigation. Furthermore, immediate weight bearing and motion exercises, that have been proposed in the context of accelerated rehabilitation after ligament surgery for example, might be re-considered in that light. It seems rather advisable to allow the cartilage for a certain recovery time after the procedure, although it is yet unknown how long it takes to restore normal concentrations of lubricin or other surface proteoglycans after an arthroscopy. Quick procedures like a partial meniscectomy, that can often be accomplished within 10-15 minutes, are seemingly less susceptible to GAG depletion and probably rather amenable for an accelerated rehabilitation. Other adverse effects of the fluid irrigation, which is used routinely in contemporary arthroscopic surgery, have been described repeatedly in the past. For example, irrigation affects viability of chondrocytes and affects the compressive properties of cartilage(Gulihar et al., 2013; Huang et al., 2015; Jurvelin et al., 1994; Kocaoglu et al., 2011; Reagan et al., 1983). In a best case scenario the cartilage recovers from the iatrogenic damage caused by the irrigation. However, it was demonstrated in different settings that single events of mechanical (and possibly chemical injury) can cause permanent damage and initiate osteoarthritis(Chu et al., 2010; Huser and Davies, 2006). Therefore, it
is possible that prolonged irrigation during arthroscopy could initiate long term changes of the cartilage. While evidence from in vitro or animal models could be questioned for its practical relevance, some clinical studies also suggest potential adverse effects of the procedure itself. Studies investigating meniscal or ACL surgery often find a similar prevalence of osteoarthritic changes on a long term for non-operative as for arthroscopic treatment despite successful reconstruction of deficient ligaments or menisci. Secondary morbidity due to the procedure itself, including the irrigation, might be involved in this phenomenon. Recently it was shown in a human in vivo study, that adverse biochemical changes and a decrease of chondrocyte viability are time-dependent on a practically relevant scale, similarly to our own study(Shen et al., 2016). While those effects were only moderate at shorter intervals, significant changes were demonstrated after more than 45 minutes. This again suggests that arthroscopic procedures should be kept as short as possible and might not be as harmless as commonly thought. An interesting question arising from these insights is, whether it might beneficial to split complex reconstructive procedures, e.g. combined meniscal replacement and cruciate ligament surgery, into two session, in order to minimize iatrogenous injury to the cartilage on a cellular level.
The type of irrigation fluid had also an effect on the CoF in our study. The Purisole solution yielded a significantly smaller CoF than either lactated Ringer’s or normal saline. Thus, it must be assumed that the Purisole solution caused less GAG washout than the other irrigation fluids under investigation. A potential explanation for this phenomenon is that Purisole is an ion-free monosaccharide solution, while Ringer’s and normal saline are salt solutions. It is a well-known phenomenon that salts can increase or decrease the solubility of macromolecules in water depending on the type and concentration of the salt, as initially described by Hofmeister(Hofmeister, 1888; Zhang and Cremer, 2006). Gradinger et al. did also provide evidence that monosaccharide solutions cause less GAG depletion than saline or Ringer’s solution(Gradinger et al., 1995). In terms of the loss of lubricating substances, salt free solutions might thus be preferable. However, this interpretation remains speculative as the detailed biophysical mechanisms behind the different effects of the irrigation solutions were beyond the scope of the study. Further investigations are warranted to prove this hypothesis.
Limitations: In this study we decided to measure the friction of cartilage against glass with a linear sliding motion. This inherits certain limitations. While the cartilage-on-glass approach was used before, the resultant coefficient of friction is certainly much higher than cartilage-on-cartilage friction. Nevertheless, it is reasonable to assume the effect in general would occur also with a cartilage-on-cartilage pairing. The challenge in using the latter approach is to establish a situation in which an even and consistent linear or rotational motion pattern is attained for the friction tests. Natural cartilage surfaces are always
irregularly curved to a degree, which potentially affects friction measurements. Often, small samples are obtained from joints of large animals, to achieve a sufficient conformity of the friction partners. However, taking cartilage samples out of their structural environment might affect the mechanical properties. Freezing the specimens prior to the tests could potentially affect the results. However, Schwarz et. al found no effect of a single freeze-thaw cycle on the results of friction measurements in a porcine model(Schwarz et al., 2012). The results apply to slow motions with a boundary lubrication mode(Hills, 2000). Faster motion with a hydrodynamic mode of friction might yield different results. Also, it is not clear whether human cartilage would behave in a similar way, although Lubricin is a highly conserved proteoglycan across species. Thus it remains unclear whether the effects observed in our study would occur in a similar way in a clinical setting. It is also unclear whether the effect size found would be practically relevant.
Conclusion The results indicate that the increase in cartilage friction, as a result of superficial proteoglycanswashout during arthroscopy, is dependent on the duration of irrigation and the type of irrigation fluid. This underscores that arthroscopic procedures, although commonly considered minimally invasive, should be kept as short as possible. The choice of the irrigation fluid should also take into account the capability of washing out lubricating substances.
Funding Sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
References Chu, C.R., Coyle, C.H., Chu, C.T., Szczodry, M., Seshadri, V., Karpie, J.C., Cieslak, K.M., Pringle, E.K., 2010. In vivo effects of single intra-articular injection of 0.5% bupivacaine on articular cartilage. J. Bone Joint Surg. Am. 92, 599–608. doi:10.2106/JBJS.I.00425 Gleghorn, J.P., Jones, A.R.C., Flannery, C.R., Bonassar, L.J., 2007. Boundary mode frictional properties of engineered cartilaginous tissues. Eur. Cell. Mater. 14, 20-28; discussion 28-29. Gradinger, R., Träger, J., Klauser, R.J., 1995. Influence of various irrigation fluids on articular cartilage. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 11, 263–269.
Gulihar, A., Bryson, D.J., Taylor, G.J.S., 2013. Effect of different irrigation fluids on human articular cartilage: an in vitro study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 29, 251–256. doi:10.1016/j.arthro.2012.07.013 Hills, B.A., 2000. Boundary lubrication in vivo. Proc. Inst. Mech. Eng. [H] 214, 83–94. doi:10.1243/0954411001535264 Hofmeister, F., 1888. Zur Lehre von der Wirkung der Salze. Arch. Für Exp. Pathol. Pharmakol. 25, 1– 30. doi:10.1007/BF01838161 Huang, Y., Zhang, Y., Ding, X., Liu, S., Sun, T., 2015. Osmolarity influences chondrocyte repair after injury in human articular cartilage. J. Orthop. Surg. 10, 19. doi:10.1186/s13018-015-0158-z Huser, C.A.M., Davies, M.E., 2006. Validation of an in vitro single-impact load model of the initiation of osteoarthritis-like changes in articular cartilage. J. Orthop. Res. Off. Publ. Orthop. Res. Soc. 24, 725–732. doi:10.1002/jor.20111 Jahn, S., Seror, J., Klein, J., 2016. Lubrication of Articular Cartilage. Annu. Rev. Biomed. Eng. 18, 235–258. doi:10.1146/annurev-bioeng-081514-123305 Jay, G.D., Waller, K.A., 2014. The biology of lubricin: near frictionless joint motion. Matrix Biol. J. Int. Soc. Matrix Biol. 39, 17–24. doi:10.1016/j.matbio.2014.08.008 Jurvelin, J.S., Jurvelin, J.A., Kiviranta, I., Klauser, R.J., 1994. Effects of different irrigation liquids and times on articular cartilage: an experimental, biomechanical study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 10, 667–672. Kocaoglu, B., Martin, J., Wolf, B., Karahan, M., Amendola, A., 2011. The effect of irrigation solution at different temperatures on articular cartilage metabolism. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 27, 526–531. doi:10.1016/j.arthro.2010.10.019 Lane Smith, R., Trindade, M.C., Ikenoue, T., Mohtai, M., Das, P., Carter, D.R., Goodman, S.B., Schurman, D.J., 2000. Effects of shear stress on articular chondrocyte metabolism. Biorheology 37, 95–107. Peng, G., McNary, S.M., Athanasiou, K.A., Reddi, A.H., 2015. The distribution of superficial zone protein (SZP)/lubricin/PRG4 and boundary mode frictional properties of the bovine diarthrodial joint. J. Biomech. 48, 3406–3412. doi:10.1016/j.jbiomech.2015.05.032 Pooley, C.M., Tabor, D., 1972. Friction and Molecular Structure: The Behaviour of Some Thermoplastics. Proc. R. Soc. Lond. Math. Phys. Eng. Sci. 329, 251–274. doi:10.1098/rspa.1972.0112 Reagan, B.F., McInerny, V.K., Treadwell, B.V., Zarins, B., Mankin, H.J., 1983. Irrigating solutions for arthroscopy. A metabolic study. J. Bone Joint Surg. Am. 65, 629–631. Schwarz, M.L.R., Schneider-Wald, B., Krase, A., Richter, W., Reisig, G., Kreinest, M., Heute, S., Pott, P.P., Brade, J., Schütte, A., 2012. [Tribological assessment of articular cartilage. A system for the analysis of the friction coefficient of cartilage, regenerates and tissue engineering constructs; initial results]. Orthopade 41, 827–836. doi:10.1007/s00132-012-1951-6 Shen, P., Li, X., Xie, G., Huangfu, X., Zhao, J., 2016. Time-Dependent Effects of Arthroscopic Conditions on Human Articular Cartilage: An In Vivo Study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 32, 2582–2591. doi:10.1016/j.arthro.2016.07.021 Smith, R.L., Carter, D.R., Schurman, D.J., 2004. Pressure and shear differentially alter human articular chondrocyte metabolism: a review. Clin. Orthop. S89-95.
Teeple, E., Karamchedu, N.P., Larson, K.M., Zhang, L., Badger, G.J., Fleming, B.C., Jay, G.D., 2016. Arthroscopic irrigation of the bovine stifle joint increases cartilage surface friction and decreases superficial zone lubricin. J. Biomech. 49, 3106–3110. doi:10.1016/j.jbiomech.2016.07.024 Waller, K.A., Zhang, L.X., Elsaid, K.A., Fleming, B.C., Warman, M.L., Jay, G.D., 2013. Role of lubricin and boundary lubrication in the prevention of chondrocyte apoptosis. Proc. Natl. Acad. Sci. U. S. A. 110, 5852–5857. doi:10.1073/pnas.1219289110 Wong, B.L., Bae, W.C., Gratz, K.R., Sah, R.L., 2008. Shear Deformation Kinematics During Cartilage Articulation: Effect of Lubrication, Degeneration, and Stress Relaxation. Mol. Cell. Biomech. MCB 5, 197–206. Zhang, Y., Cremer, P.S., 2006. Interactions between macromolecules and ions: The Hofmeister series. Curr. Opin. Chem. Biol. 10, 658–663. doi:10.1016/j.cbpa.2006.09.020
b
Figure 1: Custom made tribometer: a) load cell to measure friction force, b) fixture to hold the femur, c) linear stage with glass slide as friction partner
Figure 2: Kinetic CoF vs irrigation time: longer irrigation increases friction.
Figure 3: Static CoF vs irrigation time: longer irrigation increases friction.
Figure 4: Type of fluid versus resultant CoF, white columns kinetic CoF, gray columns static CoFg
www.elsevier.com/locate/jmbbm
PII: DOI: Reference:
S1751-6161(17)30391-0 http://dx.doi.org/10.1016/j.jmbbm.2017.09.008 JMBBM2492
To appear in: Journal of the Mechanical Behavior of Biomedical Materials Received date: 5 July 2017 Revised date: 29 August 2017 Accepted date: 4 September 2017 Cite this article as: F. Stärke, F. Awiszus, C.H. Lohmann and C. Stärke, The effect of irrigation time and type of irrigation fluid on cartilage surface friction, Journal of the Mechanical Behavior of Biomedical Materials, http://dx.doi.org/10.1016/j.jmbbm.2017.09.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The effect of irrigation time and type of irrigation fluid on cartilage surface friction F. Stärke, F. Awiszus, C.H. Lohmann, C. Stärke Orthopedics Department, University Hospital Magdeburg, Germany
Corresponding author: Dr. med. Christian Stärke, MD, PhD Orthopedics Department University Hospital Magdeburg, Germany Leipziger Straße 44 39120 Magdeburg E-Mail: [email protected] Tel.: 0049 391 6714022
Abstract: It is known that fluid irrigation used during arthroscopic procedures causes a wash-out of lubricating substances from the articular cartilage surface and leads to increased friction. It was the goal of this study to investigate whether this effect depends on the time of irrigation and type of fluid used. Rabbit hind legs were used for the tests. The knees were dissected and the friction coefficient of the femoral cartilage measured against glass in a boundary lubrication state. To determine the influence of irrigation time and fluid, groups of 12 knees received either no irrigation (control), 15, 60 or 120 min of irrigation with lactated Ringer’s solution or 60 minutes of irrigation with normal saline or a sorbitol/mannitol solution. The time of irrigation had a significant effect on the static and kinetic coefficient of friction (CoF), as had the type of fluid. Longer irrigation time with Ringer’s solution was associated with increased friction coefficients (relative increase of the kinetic CoF compared to the control after 15, 60 and 120 min: 16%, 76% and 88% respectively). The sorbitol/mannitol solution affected the static and kinetic CoF significantly less than either Ringer’s or normal saline. The washout of lubricating glycoproteins from the cartilage surface and the associated increase of friction can be effectively influenced by controlling the time of irrigation and type of fluid used. The time of exposure to the irrigation fluid should be as short as possible and monosaccharide solutions might offer a benefit compared to salt solutions in terms of the resultant friction.
Keywords: knee cartilage friction arthroscopy lubrication
Introduction
Arthroscopic techniques are commonly considered to be less invasive and less harmful than open procedures. During arthroscopy an irrigation fluid, e.g. normal saline, lactated Ringer’s solution, or a carbohydrate solution, is commonly used for the distension of the joint and to allow for a sufficient vision. While certain adverse effects of irrigation fluids have been demonstrated repeatedly in the past(Jurvelin et al., 1994; Reagan et al., 1983), these insights have little impact on the clinical practice.
However, recently it was also demonstrated in a bovine model that irrigation with Ringer’s solution causes a wash-out of Lubricin (aka PRG4 or Superficial Zone Protein, SZP), which is associated with an increased (CoF)(Teeple et al., 2016). Lubricin has been recognized as an important surface proteoglycan, that is mandatory to maintain the ultra-low friction properties of cartilage and joint homeostasis(Jay and Waller, 2014; Peng et al., 2015; Waller et al., 2013). A depletion of surface proteoglycans would increase friction, which again increases shear forces in the cartilage(Wong et al., 2008). Increased shear has been linked to chondrocyte apoptosis(Lane Smith et al., 2000; Smith et al., 2004). Thus, a depletion of surface proteoglycans and altered friction properties could make the joint cartilage more vulnerable to external loads and be an underrecognized adverse effect of standard arthroscopic procedures. While the effects described above could be clearly shown for a bovine knee model using a fixed amount of Ringer’s solution as irrigation fluid(Teeple et al., 2016), it remains unclear whether other types of irrigation fluids would affect friction properties of cartilage in the same way and whether the duration of irrigation is a factor. Both questions could have an impact on clinical practice as they can potentially be addressed during surgery.
The goal of the study was therefore to investigate in a rabbit model whether the CoF of cartilage is affected by the type of the irrigation fluid used and by the time of exposure. It was our hypothesis that irrigation causes an increased friction, which is time dependent and related to the type of fluid.
Methods
Commercially available whole rabbit legs, which had been deep frozen immediately after death of the animals, were used for the study. Prior to the tests, the legs were thawed at room temperature for 6 -8 hours. The legs were then carefully dissected to expose the knee joint. All knee ligaments were sequentially cut until the tibia and femur were separated. Particular care was taken at this point not to touch or wipe the articulating surfaces to avoid mechanical removal of surface proteoglycans. The knees were then temporarily stored at 6° C in boxes with saturated moist air. Generally, friction tests were done within 6 – 8 hours of specimen preparation. Prior to the tests the knees were warmed to ambient temperature, which was 23° C during all experiments.
There were 72 knees used in the study with N=12 per group. Following scenarios were investigated:
-control group, i.e. no irrigation (t0)
-15 min irrigation with lactated Ringer’s solution (t15_ring) -60 min irrigation with lactated Ringer’s solution (t60_ring) -120 min irrigation with lactated Ringer’s solution (t120_ring)
-60 min irrigation with Normal saline (t60_nacl) -60 min irrigation with a sorbitol/mannitol solution (Purisole SM, Fresenius Kabi, Germany; 27g Sorbitol + 5,4g Mannitol /l; t60_puri).
For the process of irrigation, groups of 6 knees were placed in 1000 ml of fresh irrigation fluid of the respective type. The fluids were sterile, standard hospital preparations. An electrical pump kept the fluid moving continuously to simulate intraoperative conditions. The knees were stored in saturated moist air at room temperature until the friction test, which was accomplished generally within one hour after the irrigation.
Friction testing A linear reciprocating tribometer was custom-designed to meet the specific requirements of the experiments (Fig. 1). Interchangeable glass slides were used as the friction partner for the cartilage. The rabbit femora were directly attached to the friction force sensor using custom-made 3D-printed fixtures (Fig. 1).
A gauged weight, born on the friction force sensor, was used to apply a defined normal force to the specimen. As variations of the individual bone anatomy could affect the true normal force, a load cell was placed below the tribometer, allowing to measure the effective normal force accurately for each experiment. The function of this custom-made tribometer was validated with a number of pilot experiments, e.g. testing 5 mm PTFE spheres against glass (12 tests, speed and normal load as in the actual study) yielded highly repeatable measurements with a kinetic CoF of 0.04 (SD 0.004). This value fits published data, gained under comparable conditions, very well(Pooley and Tabor, 1972).
For the main experiments, the rabbit femora were fixed to the friction force sensor at an angle of 45 degrees in its long axis and with an internal or external rotation of approximately 20 degrees. Thereby, only one condyle was in contact with the glass slide, which served as the friction partner (Fig. 1). For each test a fresh glass slide was used. The average (measured) normal force was 0.82 Newton (SD
0,048). Although not necessary for the calculation of the CoF, the average contact area was determined in a pilot study in 12 specimens. For this test the knees were dissected and fixed in the tribometer as described earlier. The condyle in contact with the glass slide was dried with a paper towel and stained with black ink. A sheet of scaled paper was placed between the condyle and the glass slide, thus leaving the footprint of the contact zone on the paper. A photo copy of the paper was imported into ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997-2016 ) and the area of the contact zone exactly determined. An average contact area of 4.22 (SD 0.45) mm^2 was found, giving an average contact pressure of 0.19 MPa. The test protocol was adopted from earlier studies published elsewhere(Gleghorn et al., 2007; Peng et al., 2015): First, the respective femur was fixed to the tribometer. Four drops of phosphate buffered saline (PBS, ~ 200µl) were applied to the glass slide, covering the contact area with the femoral condyle and thus preventing drying out. The condyle was then brought in contact with the glass slide, loaded with the normal force and allowed to reach equilibrium for 2 minutes. The sliding motion was then started with a speed of 0.5 mm/s and a way of 10 mm in alternating directions for 3 minutes. The slow speed was chosen to allow the measurements in a boundary lubrication state and to minimize the effect of hydrodynamic lubrication(Hills, 2000; Jahn et al., 2016). Data was directly recorded to a file using a standard PC and MATLAB/DAQtoolbox software (Mathworks, Natick, MA; USA). The friction force was determined from the recorded files. The values for the kinetic CoF were extracted from the data in the middle of the second plateau. The friction force for the static CoF was determined from the second peak, at which point the direction of motion changed from one direction to the other. The static and kinetic CoF(µ) for each knee was calculated from the normal force and the friction force with µ=Ff/Fn.
Statistics Each series of tests (effect of irrigation time and the type of irrigation fluid) was analyzed using an ANOVA, followed by a post hoc comparison with a Scheffé adjustment for multiple comparisons. The SPSS software suite ( v. 23, IBM Corp., Armonk, NY, USA) was used for the analysis.
Results Both, static and kinetic CoF were highly significantly dependent on the irrigation time (p<0.001) and on the type of irrigation fluid (p<0.001).
Irrigation with lactated Ringer’s solution increased the kinetic friction coefficient by 16%, 76% and 88% after 15, 60 and 120 min minutes of exposure respectively (Fig. 2, 3). Post hoc comparison showed
significant differences between 60 min and 120 min of treatment and the control but not between 15 min and control and 60 min and 120 min of exposure. The same effects were shown for the static coefficient of friction.
Concerning the type of irrigation fluid, a highly significant effect was found for the treatment with Purisole, which increased kinetic and static CoF less than either normal saline or lactated Ringer’s solution. The latter both fluids instead did not differ from each other (Fig. 4).
Discussion Exposing knee joint cartilage to arthroscopic irrigation fluids increased generally the coefficient of friction in our model. We assume that this effect is caused by a depletion of surface-glycoproteins like Lubricin(Teeple et al., 2016). Both, irrigation time and type of the irrigation fluid, were related to the magnitude of the effect. While the difference between the non-irrigated controls and 15 min of irrigation was not statistically significant, there is probably not a clear-cut threshold beyond which cartilage friction would be affected. It is reasonable to assume a continuous relationship between the time of exposure and the depletion of surface glycoproteins and thus the resultant friction, although this relationship is probably not linear. This has potential implications for the clinical practice. As increased cartilage friction is associated with several adverse effects on a cellular level(Lane Smith et al., 2000; Wong et al., 2008), limiting irrigation time to the necessary amount seems advisable. Dry arthroscopy might be considered for steps in the arthroscopic procedure not relying on continuous irrigation. Furthermore, immediate weight bearing and motion exercises, that have been proposed in the context of accelerated rehabilitation after ligament surgery for example, might be re-considered in that light. It seems rather advisable to allow the cartilage for a certain recovery time after the procedure, although it is yet unknown how long it takes to restore normal concentrations of lubricin or other surface proteoglycans after an arthroscopy. Quick procedures like a partial meniscectomy, that can often be accomplished within 10-15 minutes, are seemingly less susceptible to GAG depletion and probably rather amenable for an accelerated rehabilitation. Other adverse effects of the fluid irrigation, which is used routinely in contemporary arthroscopic surgery, have been described repeatedly in the past. For example, irrigation affects viability of chondrocytes and affects the compressive properties of cartilage(Gulihar et al., 2013; Huang et al., 2015; Jurvelin et al., 1994; Kocaoglu et al., 2011; Reagan et al., 1983). In a best case scenario the cartilage recovers from the iatrogenic damage caused by the irrigation. However, it was demonstrated in different settings that single events of mechanical (and possibly chemical injury) can cause permanent damage and initiate osteoarthritis(Chu et al., 2010; Huser and Davies, 2006). Therefore, it
is possible that prolonged irrigation during arthroscopy could initiate long term changes of the cartilage. While evidence from in vitro or animal models could be questioned for its practical relevance, some clinical studies also suggest potential adverse effects of the procedure itself. Studies investigating meniscal or ACL surgery often find a similar prevalence of osteoarthritic changes on a long term for non-operative as for arthroscopic treatment despite successful reconstruction of deficient ligaments or menisci. Secondary morbidity due to the procedure itself, including the irrigation, might be involved in this phenomenon. Recently it was shown in a human in vivo study, that adverse biochemical changes and a decrease of chondrocyte viability are time-dependent on a practically relevant scale, similarly to our own study(Shen et al., 2016). While those effects were only moderate at shorter intervals, significant changes were demonstrated after more than 45 minutes. This again suggests that arthroscopic procedures should be kept as short as possible and might not be as harmless as commonly thought. An interesting question arising from these insights is, whether it might beneficial to split complex reconstructive procedures, e.g. combined meniscal replacement and cruciate ligament surgery, into two session, in order to minimize iatrogenous injury to the cartilage on a cellular level.
The type of irrigation fluid had also an effect on the CoF in our study. The Purisole solution yielded a significantly smaller CoF than either lactated Ringer’s or normal saline. Thus, it must be assumed that the Purisole solution caused less GAG washout than the other irrigation fluids under investigation. A potential explanation for this phenomenon is that Purisole is an ion-free monosaccharide solution, while Ringer’s and normal saline are salt solutions. It is a well-known phenomenon that salts can increase or decrease the solubility of macromolecules in water depending on the type and concentration of the salt, as initially described by Hofmeister(Hofmeister, 1888; Zhang and Cremer, 2006). Gradinger et al. did also provide evidence that monosaccharide solutions cause less GAG depletion than saline or Ringer’s solution(Gradinger et al., 1995). In terms of the loss of lubricating substances, salt free solutions might thus be preferable. However, this interpretation remains speculative as the detailed biophysical mechanisms behind the different effects of the irrigation solutions were beyond the scope of the study. Further investigations are warranted to prove this hypothesis.
Limitations: In this study we decided to measure the friction of cartilage against glass with a linear sliding motion. This inherits certain limitations. While the cartilage-on-glass approach was used before, the resultant coefficient of friction is certainly much higher than cartilage-on-cartilage friction. Nevertheless, it is reasonable to assume the effect in general would occur also with a cartilage-on-cartilage pairing. The challenge in using the latter approach is to establish a situation in which an even and consistent linear or rotational motion pattern is attained for the friction tests. Natural cartilage surfaces are always
irregularly curved to a degree, which potentially affects friction measurements. Often, small samples are obtained from joints of large animals, to achieve a sufficient conformity of the friction partners. However, taking cartilage samples out of their structural environment might affect the mechanical properties. Freezing the specimens prior to the tests could potentially affect the results. However, Schwarz et. al found no effect of a single freeze-thaw cycle on the results of friction measurements in a porcine model(Schwarz et al., 2012). The results apply to slow motions with a boundary lubrication mode(Hills, 2000). Faster motion with a hydrodynamic mode of friction might yield different results. Also, it is not clear whether human cartilage would behave in a similar way, although Lubricin is a highly conserved proteoglycan across species. Thus it remains unclear whether the effects observed in our study would occur in a similar way in a clinical setting. It is also unclear whether the effect size found would be practically relevant.
Conclusion The results indicate that the increase in cartilage friction, as a result of superficial proteoglycanswashout during arthroscopy, is dependent on the duration of irrigation and the type of irrigation fluid. This underscores that arthroscopic procedures, although commonly considered minimally invasive, should be kept as short as possible. The choice of the irrigation fluid should also take into account the capability of washing out lubricating substances.
Funding Sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
References Chu, C.R., Coyle, C.H., Chu, C.T., Szczodry, M., Seshadri, V., Karpie, J.C., Cieslak, K.M., Pringle, E.K., 2010. In vivo effects of single intra-articular injection of 0.5% bupivacaine on articular cartilage. J. Bone Joint Surg. Am. 92, 599–608. doi:10.2106/JBJS.I.00425 Gleghorn, J.P., Jones, A.R.C., Flannery, C.R., Bonassar, L.J., 2007. Boundary mode frictional properties of engineered cartilaginous tissues. Eur. Cell. Mater. 14, 20-28; discussion 28-29. Gradinger, R., Träger, J., Klauser, R.J., 1995. Influence of various irrigation fluids on articular cartilage. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 11, 263–269.
Gulihar, A., Bryson, D.J., Taylor, G.J.S., 2013. Effect of different irrigation fluids on human articular cartilage: an in vitro study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 29, 251–256. doi:10.1016/j.arthro.2012.07.013 Hills, B.A., 2000. Boundary lubrication in vivo. Proc. Inst. Mech. Eng. [H] 214, 83–94. doi:10.1243/0954411001535264 Hofmeister, F., 1888. Zur Lehre von der Wirkung der Salze. Arch. Für Exp. Pathol. Pharmakol. 25, 1– 30. doi:10.1007/BF01838161 Huang, Y., Zhang, Y., Ding, X., Liu, S., Sun, T., 2015. Osmolarity influences chondrocyte repair after injury in human articular cartilage. J. Orthop. Surg. 10, 19. doi:10.1186/s13018-015-0158-z Huser, C.A.M., Davies, M.E., 2006. Validation of an in vitro single-impact load model of the initiation of osteoarthritis-like changes in articular cartilage. J. Orthop. Res. Off. Publ. Orthop. Res. Soc. 24, 725–732. doi:10.1002/jor.20111 Jahn, S., Seror, J., Klein, J., 2016. Lubrication of Articular Cartilage. Annu. Rev. Biomed. Eng. 18, 235–258. doi:10.1146/annurev-bioeng-081514-123305 Jay, G.D., Waller, K.A., 2014. The biology of lubricin: near frictionless joint motion. Matrix Biol. J. Int. Soc. Matrix Biol. 39, 17–24. doi:10.1016/j.matbio.2014.08.008 Jurvelin, J.S., Jurvelin, J.A., Kiviranta, I., Klauser, R.J., 1994. Effects of different irrigation liquids and times on articular cartilage: an experimental, biomechanical study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 10, 667–672. Kocaoglu, B., Martin, J., Wolf, B., Karahan, M., Amendola, A., 2011. The effect of irrigation solution at different temperatures on articular cartilage metabolism. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 27, 526–531. doi:10.1016/j.arthro.2010.10.019 Lane Smith, R., Trindade, M.C., Ikenoue, T., Mohtai, M., Das, P., Carter, D.R., Goodman, S.B., Schurman, D.J., 2000. Effects of shear stress on articular chondrocyte metabolism. Biorheology 37, 95–107. Peng, G., McNary, S.M., Athanasiou, K.A., Reddi, A.H., 2015. The distribution of superficial zone protein (SZP)/lubricin/PRG4 and boundary mode frictional properties of the bovine diarthrodial joint. J. Biomech. 48, 3406–3412. doi:10.1016/j.jbiomech.2015.05.032 Pooley, C.M., Tabor, D., 1972. Friction and Molecular Structure: The Behaviour of Some Thermoplastics. Proc. R. Soc. Lond. Math. Phys. Eng. Sci. 329, 251–274. doi:10.1098/rspa.1972.0112 Reagan, B.F., McInerny, V.K., Treadwell, B.V., Zarins, B., Mankin, H.J., 1983. Irrigating solutions for arthroscopy. A metabolic study. J. Bone Joint Surg. Am. 65, 629–631. Schwarz, M.L.R., Schneider-Wald, B., Krase, A., Richter, W., Reisig, G., Kreinest, M., Heute, S., Pott, P.P., Brade, J., Schütte, A., 2012. [Tribological assessment of articular cartilage. A system for the analysis of the friction coefficient of cartilage, regenerates and tissue engineering constructs; initial results]. Orthopade 41, 827–836. doi:10.1007/s00132-012-1951-6 Shen, P., Li, X., Xie, G., Huangfu, X., Zhao, J., 2016. Time-Dependent Effects of Arthroscopic Conditions on Human Articular Cartilage: An In Vivo Study. Arthrosc. J. Arthrosc. Relat. Surg. Off. Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 32, 2582–2591. doi:10.1016/j.arthro.2016.07.021 Smith, R.L., Carter, D.R., Schurman, D.J., 2004. Pressure and shear differentially alter human articular chondrocyte metabolism: a review. Clin. Orthop. S89-95.
Teeple, E., Karamchedu, N.P., Larson, K.M., Zhang, L., Badger, G.J., Fleming, B.C., Jay, G.D., 2016. Arthroscopic irrigation of the bovine stifle joint increases cartilage surface friction and decreases superficial zone lubricin. J. Biomech. 49, 3106–3110. doi:10.1016/j.jbiomech.2016.07.024 Waller, K.A., Zhang, L.X., Elsaid, K.A., Fleming, B.C., Warman, M.L., Jay, G.D., 2013. Role of lubricin and boundary lubrication in the prevention of chondrocyte apoptosis. Proc. Natl. Acad. Sci. U. S. A. 110, 5852–5857. doi:10.1073/pnas.1219289110 Wong, B.L., Bae, W.C., Gratz, K.R., Sah, R.L., 2008. Shear Deformation Kinematics During Cartilage Articulation: Effect of Lubrication, Degeneration, and Stress Relaxation. Mol. Cell. Biomech. MCB 5, 197–206. Zhang, Y., Cremer, P.S., 2006. Interactions between macromolecules and ions: The Hofmeister series. Curr. Opin. Chem. Biol. 10, 658–663. doi:10.1016/j.cbpa.2006.09.020
b
Figure 1: Custom made tribometer: a) load cell to measure friction force, b) fixture to hold the femur, c) linear stage with glass slide as friction partner
Figure 2: Kinetic CoF vs irrigation time: longer irrigation increases friction.
Figure 3: Static CoF vs irrigation time: longer irrigation increases friction.
Figure 4: Type of fluid versus resultant CoF, white columns kinetic CoF, gray columns static CoFg