Development of Atomic Force Microscope for ...

Japanese Journal of Applied Physics Vol. 45, No. 3B, 2006, pp. 2319–2323 #2006 The Japan Society of Applied Physics

Development of Atomic Force Microscope for Arthroscopic Knee Cartilage Inspection Raphae¨l I MER, Terunobu A KIYAMA1 , Nicolaas F. DE ROOIJ1 , Martin S TOLZ2 , Ueli A EBI2 , Niklaus F. FRIEDERICH3 , Uwe K OENIG3 , Dieter W IRZ4 , A. U. D ANIELS4 and Urs S TAUFER1 Institute of Microtechnology, University of Neuchatel, Jaquet-Droz 1, CH-2002 Neuchatel, Switzerland 2 Maurice E. Mueller Institute, Biozentrum, University of Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland 3 Clinics for Orthopaedic Surgery and Traumatology, Kantonsspital Bruderholz, CH-4101 Bruderholz, Switzerland 4 Laboratory for Orthopaedic Biomechanics, University of Basel, Klingelbergstrasse 50-70, CH-4056 Basel, Switzerland (Received July 4, 2005; accepted November 4, 2005; published online March 27, 2006)

A recent study, based on ex vivo unconfined compression testing of normal, diseased, and enzymatically altered cartilage, revealed that a scanning force microscope (SFM), used as a nano-intender, is sensitive enough to enable measurement of alterations in the biomechanical properties of cartilage. Based on these ex vivo measurements, we have designed a quantitative diagnosis tool, the scanning force arthroscope (SFA), able to perform in vivo measurements during a standard arthroscopic procedure. For stabilizing and positioning the instrument relative to the surface under investigation, a pneumatic system has been developed. A segmented piezoelectric tube was used to perform the indentation displacement, and a pyramidal nanometer-scale silicon tip mounted on a cantilever with an integrated deflection sensor measured the biomechanical properties of cartilage. Mechanical means were designed to protect the fragile cantilever during the insertion of the instrument into the knee joint. The stability of the pneumatic stage was checked with a prototype SFA. In a series of tests, loaddisplacement curves were recorded in a knee phantom and, more recently, in a pig’s leg. [DOI: 10.1143/JJAP.45.2319] KEYWORDS: scanning force microscope, articular cartilage, arthroscopy, biomechanics, degradation, diagnostics, indentation, minimally invasive instrument, stiffness

1.

2.

Introduction

The hyaline cartilage is a thin layer of soft tissue that covers the ends of bones. This tissue distributes load and provides lubrication in the joint. The biomechanical properties of articular cartilage directly depend on its composition and structure.1) Water content is responsible for the deformation of the cartilage surface in response to stress. At up to 80%, wet weight, it is the main constituent of the organic matrix. The tensile strength and the shear properties of cartilage are provided by collagen fibers. These fibers immobilize proteoglycans in the cartilaginous framework. Proteoglycans are complex macromolecules responsible for compressive strength. Chondrocytes maintain articular cartilage; these cells control the production of collagen fibers, proteoglycans, and enzymes, but due to the avascular, aneural, and alymphatic nature of cartilage, the healing system of the human body has a very poor ability to selfrepair lesions. Trauma is one of the main reasons for osteoarthritis of the knee joint. Neyret et al.2) were able to show that the incidence of radiographic osteoarthritis is about 65% at 27 years in patients with a ruptured ligament. The incidence of osteoarthritis actually increases, if the patients have combined lesions of multiple ligaments or other structures such as the meniscus. Depending on the age of the patient and the stage of osteoarthritis, a clinician decides the course of treatment. For some lesions, osteotomy to correct the joint axes is sufficient. For others, osteotomy may also be combined with either cartilage cell transplantation or osteocartilaginous cylinder grafting (Mosaik-Plasty). A minimally invasive device that both facilitates operative decisions by diagnosing cartilage degeneration in an early stage and allows post operative monitoring of the affected areas is highly desirable.

E-mail address: [email protected]

Evaluation of Cartilage

Different methods are available for evaluating the properties of cartilage. Magnetic resonance imaging (MRI), X-ray and computer tomography provide images of the surface only on the macroscopic scale. They can reveal changes in the composition of cartilage, but only at an advanced stage of degradation. An ultrasonic indentation instrument3) uses sound waves to probe the elasticity of the cartilage. This method enables the assessment of the cartilage at the millimeter to upper micrometer scale. Water or air-jet indenter probes4) allow the calculation of the stiffness of cartilage from simultaneously recorded optical images of the induced deformation. This method provides macroscopic measurements. However, classic mechanical indentation is actually the easiest, the most used, and the most reproducible technique available. The mechanical properties of biological tissues measured with this kind of clinical indenter represent an average over several square millimeters of material. This kind of device does not allow the detection of cartilage diseases at an early stage. Measurements performed by Stolz et al.5) on porcine articular cartilage showed a radical improvement in the sensitivity of the measurements when the lateral size of the indenter probes was below 1 mm. However, if the size of the indenter probe becomes smaller than the diameter of a single fiber, variations in elasticity occur due to the three-dimensional organization of the collagen fibers. Therefore, the combination of a micrometersized and a nanometer-sized indenter can be used to resolve local changes in the stiffness of cartilage and to detect cartilage diseases at an early stage. An atomic force microscope (AFM) has already been used by Van Landingham et al.6) to characterize the nanometerscale properties of the interphase regions of fiber-reinforced polymers, and by Rho et al.7) and Turner et al.8) to measure the elastic properties of human cortical and trabecular lamella bones by nano-indentation.

2319

Jpn. J. Appl. Phys., Vol. 45, No. 3B (2006)

3.

R. IMER et al.

Indentation-Type AFM Measurements

Stolz et al.5) developed a method of measuring the dynamic elastic modulus jE j in compression by indentationtype atomic force microscopy (IT AFM). This method involves recording cyclic load–displacement curves at different sites on the sample’s surface at a vertical displacement frequency of 3 Hz. The mechanical properties of cartilage were tested on two different scales using different scanning force microscope (SFM) probes: On the micrometer scale, the dynamic elastic modulus jE j was computed from the average slope of cyclic load–displacement curves according to the modeling method for mechanical indenters developed in 1992 by Pharr et al.9) The unloading data could be fitted to a simple power-law equation: p ¼ Bðh hf Þm ; where p is the indenter load, h is the vertical displacement of the indenter, hf represents the final unloading depth, and B and m are fitting parameters. m ¼ 1:5 for a spherical indenter. The load was obtained simply by multiplying the cantilever’s deflection by its spring constant k. The powerlaw fitted to the averaged load–indentation curve was determined using the Levenberg–Marquardt algorithm. The stiffness S ¼ d p=dh was calculated by linear regression. For finally calculating the dynamic elastic modulus jE j, it was assumed that jE j E, where the indentation modulus E is pffiffiffi S ð1 2 Þ pffiffiffi ; E¼ 2 A where is Poisson’s ratio, S the contact stiffness, and A an area function related to the effective cross section of the indenter. At the nanometer scale, a sharp pyramidal tip was used. Rather than modeling the pyramidal tip data, the slope of the unloading part of the IT AFM load–displacement curve was directly converted into a dynamic modulus by means of a calibration curve recorded on agarose gels. To generate this calibration curve, macrometer scale measurements of agarose gel stiffness were performed in unconfined compression by following a standard protocol for polymer testing using a universal mechanical testing apparatus. 4.

Description of the Instrument

Arthroscopy is a minimally invasive surgery performed in an operating room, in most cases under local anesthesia, during which only small openings are needed for inspecting or treating a joint. To be used during this kind of surgical operation, the scanning force arthroscope (SFA) must have dimensions comparable to those of classical arthroscopic tools such as tweezers and scissors. Therefore, we miniaturized and integrated a complete AFM setup, including stabilization, scanning, and indentation stages, into a standard arthroscopic sheath with a cylindrical outer diameter of 5 mm. Considering the desired resolution of the measurements and the special environment of an operating room, the major challenges for recording IT AFM measurements are the stabilization of the instrument inside the knee joint and the damping of vibrations. The stabilization stage specially designed for that purpose consisted of eight inflatable highpressure angioplasty balloons. These balloons, inflated with physiological liquid to damp outside vibrations and prevent

damage in the case of leakage, wedged the SFA into the joint by filling the gap between the instrument and the body cavity. To increase the stabilization, the balloons were arranged in two sets of four; the first set was rotated against the second by an angle of 45 . This inflatable balloon system was not only responsible for the fixation of the instrument in the body cavity and for the damping of the vibrations, but also was effective for carrying out the coarse approach. Once the SFA was wedged in the knee joint, the distal end of the instrument could be approached from the surface by adjusting the differential pressure between opposite balloons. Each balloon was connected to a syringe pusher controlled by a computer, and a pressure sensor allowed a very well-defined pressure to be applied in the balloons. The scanning stage was essential in carrying out the indentations during the measurements. An encapsulated multi-electrode piezoelectric tube made of lead–zirconium–titanate (PZT) was used to apply a controlled load on the surface of cartilage with the AFM cantilever. Depending on the voltage applied between the inner and outer electrodes, the four-segmented piezoelectric tube contracted laterally or longitudinally, thereby providing, an XYZ motion. With a three-dimensional scanner onboard, the scanning stage is effective not only for recording accurate measurements, but also for creating a two-dimensional map of the stiffness of the cartilage. This XYZ scanning stage operated at a maximum voltage of 200 V and allowed, a volume of 1 6 6 mm3 ðX; Y; ZÞ to be addressed. The indentation stage contained the AFM chip required for probing the biomechanical properties of the cartilage. Due to the very limited space available in the knee joint, a piezoresistive stress sensor was directly integrated in the cantilever, replacing the classic laser and photodiodebased detection. The probes used for the first measurements consisted of a rectangular silicon AFM cantilever on which a 10-mm-high pyramidal tip was integrated. The spring constant of this cantilever was approximately 10 N/m. An important feature of the indentation stage is the ability to quickly exchange the probe in the case of failure. For that purpose, a small connector that could be easily detached from the rest of the instrument was designed. To facilitate the approach, the chip was tilted by an angle of 15 . The outside vibrations introduced by the surgeon’s hand could be reduced by decoupling the tool tip or ‘‘SFA-satellite’’ (Fig. 1) formed by the stabilization stage, scanning stage and indentation stage from the rest of the instrument. During the measurements, electrical and pneumatic conduits were the only connections between the handle and the SFAsatellite. 5.

Experimental Procedures

Due to the very limited dimensions of the knee cavity and the constant flow of fluid (physiological liquid or CO2 ) applied to wash the knee during arthroscopy, we defined a procedure that protected the fragile cantilever during its insertion inside the joint. (1) The surgeon introduced the arthroscope inside the knee and performed a visual inspection. (2) Once an area of interest was located, the surgeon positioned a trocar sleeve to create an entry site for the instrument [Fig. 2(a)]. (3) The trocar was removed from the cannula, and

2320


R. IMER et al.

Fig. 1. The SFA-satellite contains the stabilization stage, the scanning stage and the indentation stage and can be separated from the handling part, connected only by electrical and pneumatic conduits. The piezoelectric tube was protected by a transparent thin polymer tube

Fig. 3. The instrument was inserted inside a knee phantom. Different parts of this knee had been removed, and a PVC tube was used to mimic the joint cavity.

Fig. 2. A safe insertion procedure was defined to protect the fragile cantilever. (a) First, to find the right place, a trocar was insert in the knee under arthroscopic control. (b) The trocar was replaced by our instrument which was protected inside a cannula. (c) The protection was removed and the balloons were inflated.

replaced by the SFA [Fig. 2(b)]. The SFA-satellite was carefully positioned under optical control of the arthroscope. (4) All the balloons were inflated simultaneously to clamp the instrument near the cartilage surface [Fig. 2(c)]. (5) The coarse approach was carried out by changing the pressure in the balloons. (6) Finally, before starting measurements, the SFA-satellite was decoupled from its handle. The reverse procedure was used to safely withdraw the instrument after measurements. During the tests performed to assess the quality of the stabilization, the insertion procedures were simplified because the instrument was

directly inserted inside a knee model or in an open joint without any liquid flow. The first test of stabilization was performed with an early version of the SFA in an anatomical, flexible knee phantom (Fig. 3). In this prototype, the indentation stage connector had not yet been implemented, and the AFM chip, was directly glued at the end of the scanning stage. The knee model included the tibia, the femur, the patella, the meniscus, and all the ligaments, but not the skin, the fat pad, or the muscles present during the stabilization of the instrument. For that reason, a poly(vinyl chloride) (PVC) tube was added between the lateral condyle of the femur and the tibial plateau on the left side of the model. A 1 Hz sine wave with an amplitude of 125 V was applied to the z-axis of the piezoelectric tube during the indentation of the tibial plateau. Figure 4 shows the first load–displacement curve recorded using an arthroscopic AFM. A second experiment, performed at the Institute of Anatomy of the University of Basel, was carried out to determine if the dimensions and the design of the tool allowed a surgeon to reach all the places in the joint where cartilage could be injured. During this surgical simulation, a large opening was made at the place where the arthroscope is usually inserted to allow good optical control of the localization of the instrument inside the cadaver’s knee. During this test, no load–displacement curves were recorded. At that time, the mechanical sleeve for protecting the

2321


R. IMER et al.

Fig. 4. Load–displacement curve recorded inside knee phantom.

Fig. 6. The SFA was inserted inside the pig’s leg, and measurements of the ankle were recorded.

Fig. 5. The instrument inserted inside a knee cadaver. This test was carried out to test the stabilization of the instrument under surgical conditions.

cantilever during the insertion of the instrument inside the joint was not yet implemented. A third set of experiments was performed inside the leg of a pig obtained from a slaughter house. A large incision was cut at the top of the ankle, into which the instrument could be inserted (Fig. 6). The spaces available for stabilization in a pig’s ankle were considerably fewer than in a human knee. For that reason, only the first stack of balloons could be used to wedge the instrument and to carry out the coarse approach. Once again, a 10 Hz sine wave with an amplitude of 125 V was applied to the z-axis of the piezoelectric tube during the indentation. Figure 7 shows the first load–displacement curve recorded with an AFM in a joint. 6.

Results and Discussion

The goal of the experiments performed inside the knee of a cadaver was to prove the ability to fix an arthroscopic instrument inside an articulation using a pneumatic system. During this test, the stabilization stage enabled the instrument to be quickly wedged between the patellar ligament, the tibia, the femur, and the fat pad. The fixation was secure enough to prevent the surgeon from removing or even from shifting the instrument without deflating the balloons. The SFA was inserted through a portal commonly used during

Fig. 7. Load–displacement curve recorded inside pig’s leg.

arthroscopy and was able to reach most of the frequently injured sites in the knee joint. The measurements recorded in the knee phantom (Fig. 4) and in the pig’s leg (Fig. 7) were performed to assess the stabilization of the SFA. Qualitatively, the load–displacement curves showed the expected characteristics. The typical snap out of contact could be observed on the retracting curve in the knee phantom but not in the pig’s leg. The environmental conditions can explain this difference; the first measurement was perform in air, whereas inside the pig’s joint, the cantilever was completely immersed in liquid. As expected, a comparison between the slope of the two indentation curves showed higher values for the plastic surface in the knee phantom than for the cartilage in the pig’s leg. A complete quantitative analysis, using the method presented in §3, would require the piezoelectric tube to be calibrated. 7.

Conclusions

In this paper, we have presented an arthroscopic instrument for measuring in vivo the biomechanical properties of

2322


R. IMER et al.

cartilage on the nanometer scale. The first experimental results of recording load–displacement curves showed that using inflatable balloons is a simple, quick, and efficient method for fixing the instrument inside the body cavity and for suppressing outside vibrations. The instrument will be inserted inside a cadaver knee under arthroscopic control to record the first nano-indentation measurements with an AFM inside a human body in the next steps. Two-dimensional images of load–displacement curves will be recorded using all electrodes of the segmented piezoelectric tube. In the future, the instrument may be inserted inside other joints, such as the hip or the shoulder. We will undertake to record topographic images of collagen fibers as well.

This work was financially supported by NCCR Nanoscale Science of the Swiss National Science Foundation and la Re´publique et Canton de Neuchaˆtel.

Acknowledgements The authors thank Josef Kapfhammer from the Institute of Anatomy of the University of Basel, Switzerland, for giving us the opportunity to conduct experiments on a cadaver. We also thank Bernard Scanio for helpful discussions and acknowledge the technical support of the staff of ComLab, the joint IMT CSEM clean room facility.

2323

1) T. A. Einhorn, S. R. Simon and J. A. Buckwalter: in Orthopaedic Basic Science, ed. American Academy of Orthopaedic Surgeons (2000) 2nd ed., Chap. 17, p. 443. 2) P. Neyret, ST. Donell and H. Dejour: J. Bone Joint Surg. Br. 75 (1993) 36. 3) M. S. Laasanen, S. Saarakkala, J. Toÿra¨s, J. Hirvonen, J. Rieppo, R. K. Korhonen and J. S. Jurvelin: J. Biomech. 36 (2003) 1259. 4) G. N. Duda, R. U. Kleemann, U. Bluecher and A. Weiler: Am. J. Sports Med. 32 (2004) 693. 5) M. Stolz, R. Raiteri, A. U. Daniels, M. R. VanLandingham, W. Baschong and U. Aebi: Biophys. J. 86 (2004) 3269. 6) M. R. VanLandingham, S. H. McKnight, G. R. Palmese, R. F. Eduljee, J. W. Gillespie and JR. R. L. McCulough: J. Mater. Sci. Lett. 16 (1997) 117. 7) J. Y. Rho, T. Y. Tsui and G. M. Pharr: Biomaterials 18 (1997) 1325. 8) C. H. Turner, J. Rho, Y. Takano, T. Y. Tsui and G. M. Pharr: J. Biomech. 32 (1999) 437. 9) G. M. Pharr, W. C. Oliver and F. R. Brotzen: J. Mater. Res. 7 (1992) 613.