- Email: [email protected]
Miniaturised ultrasonic aspiration handpiece for increased applicability
European Journal of Ultrasound 11 (2000) 41 – 46 www.elsevier.com/locate/ejultrasou
Basic Original
Miniaturised ultrasonic aspiration handpiece for increased applicability Hans Wiksell a,b,*, Harvey Martin c, Hugh Coakham d, Anders Berggren b, Sara Westermark a,b,e a
Department of Urology, Di6ision of Surgical Science, Karolinska Institute and Karolinska Hospital, Stockholm, Sweden b Comair AB, Box 39011, S-100 54 Stockholm, Sweden c Elekta Surgical Instruments Ltd., Ando6er, UK d Department of Neurosurgery, Frenchay Hospital and Uni6ersity of Bristol, Bristol, UK e Department of Laboratory Science and Technology, Di6ision of Medical Engineering, NOVUM and Karolinska Institute, Huddinge Hospital, Huddinge, Sweden Received 17 January 1999; received in revised form 5 July 1999; accepted 23 July 1999
Abstract Objecti6e: At present, ultrasonic aspiration is routinely used in several fields of surgery, especially in brain and spinal micro-surgery for tumour removal. In order to broaden the access to difficult surgical sites, it is important to design highly miniaturised but still efficient handpieces. The internal resonant system, always made of high-grade materials, must be optimally dimensioned. Normally this is done semi-empirically, by successively improving the design during many iterative test steps. This method however involves several additional difficulties when the degree of miniaturisation increases. For example, small transducer weights exacerbate heat-dissipation problems and make design optimisation important. Methods: To resolve these problems we have produced modelling software that makes it possible to simulate and automatically tune each individual interacting section of the design before it is actually manufactured, thereby assuring optimal efficiency. Results: Using a new mini-handpiece, designed via the software, two cases of dissection of acoustic neurinomas were successfully performed. Conclusion: Using conventional physical steps for improving ultrasonic aspiration handpieces, several problems arise when the grade of miniaturisation increases, due to increasing demands. We have designed computer software for handpiece simulation. Using this model it has been possible to manufacture a highly efficient miniaturised handpiece. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ultrasonic surgical aspiration; Micro-surgery; Tissue selective dissection; Brain tumour surgery
* Corresponding author. Tel.: + 46-8-664-4030; fax: +46-8-664-1820. E-mail address: [email protected] (H. Wiksell) 0929-8266/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 9 - 8 2 6 6 ( 9 9 ) 0 0 0 6 9 - 5
42
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
1. Introduction Ultrasonic aspiration methods were initially developed because of their potential for producing a tissue-selective dissection, for enhanced surgical tactility, for the removal of tissues with minimal damage to surrounding areas and for achieving a significant decrease in total blood loss (Wiksell and Granholm, 1986; Kourtopoulos et al., 1989; Thomasson et al., 1990).
1.1. Principles of ultrasonic aspiration handpieces An ultrasonic aspiration handpiece generally consists of an ultrasonic transducer attached to a resonant horn-shaped wave-guide amplitude magnifier that ends in a surgical tip. The transducer and magnifier are housed in an electrically and acoustically isolated hand-held precision casing. The surgical tip is often housed in a high-grade plastic coaxial tip-cover arrangement, in order to protect nearby tissues from accidentally being touched. This is important due to the frequent use of stereo-microscopy during the surgical procedure, which limits the field of view. The gradually tapered cross-section of the wave-guide creates the acoustic gain, due to the law of energy conservation versus the losses. The nearly exponential overall shape of the tapered sections further eliminates unacceptable standing-wave generating reflections. The horn is equipped with a central suction channel, typically of 1.5 mm i.d., for transporting tissue fragments to a container. The suction pressure can be adjusted down to approx. − 0.8 bar, which means an equivalent flat-area force on the target tissues of up to approx. 140 mN. This suction channel could be continuous through the handpiece to facilitate transducer cooling or it could be joined to an external tube, at a velocity node to reduce sealing losses. The tip is coaxially irrigated with sterile isotonic NaCl-solution which serves 1. to increase the acoustic coupling to the target; 2. cool the tip-material and the exposed tissues from internal frictional heating; 3. produce a liquid bulk to soak and enhance vacuum-transport of the tissue fragments
through the suction system without obstructions. The irrigation flow can typically be adjusted up to some ml/s. The mechanical resonance Q-value of the aspirator system must be high, in order to achieve the intended tip-amplitude. This calls for continuous frequency adjustment since the temperature varies slightly during handpiece usage. Changes in the temperature alter the resonant lengths, which therefore must be continuously monitored and adjusted. This is done by detecting the resonance deviations via a feed-back phase-locked loop (PLL), located in the driver console. From this it follows that the resonance bandwidth of each individual section must be carefully designed. The tip amplitude can also automatically compensate for different tissue loads using a quadrature directional-coupler as a power controlling feed-back. The energy stored in the elastic oscillation will be greater than the work done. The stored energy will therefore act as a flywheel and stabilise against fast variations in the acoustic load. An important feature of the ultrasonic handpiece is that it is designed to tolerate repeated sterilisation procedures in an ordinary autoclave at a nominal 135°C. Compared to cold sterilisation this means a significantly higher grade of safety against crosscontamination, since hidden spaces inside the handpiece will also be heated.
1.2. Magnetostricti6e and electrostricti6e transducers Two different types of transducers are used in surgical ultrasonic aspirators: magnetostricti6e and electrostricti6e. Magnetostrictive transducers were historically the first to be used. They consist for instance of a half-wavelength (approx. 10 cm at 24 kHz) resonant nickel plate package placed in a magnetic-field produced by a coil. The package is composed of many thin laminated and end welded nickel alloy plates to reduce eddy currents. In the magnetic field, the package is forced to mechanically oscillate due to magnetostrictive forces. The amplitude of such a transducer can be quite high, but the conversion efficiency is poor and the transducer thus normally needs to be
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
water cooled by a separate high-flow water loop. This loop provides a potentially serious sterility problem due to the risk of non-sterile leakage. The electrostrictive transducer mainly consists of a driving piezoelectric section. In practice, it is necessary to compose a sandwich Langevin-type transducer, which has resonator-bars in both ends. The piezoelectric material can only take a limited negative stress, which is why most high-power transducers use a positive mechanical pre-stress in order to improve the material’s fatigue strength. An electrostrictive transducer generally provides high stability and a high conversion efficiency. It generates a low amplitude but with great strain. The two resonator bars might have different lengths (so-called non-symmetric transducers) in order to further increase the amplitude. Still the output amplitude will be low, approx. 10 mmpeak-topeak, which is why the electrostrictive transducer horn needs a much higher amplitude-gain (typically a factor of 30) than the magnetostrictive. Designing high-gain horns must be done properly to avoid non-linear tendencies, which can act to create dangerous transversal spurious-oscillations by parametric pumping.
1.3. Tip amplitude The surgical tip will oscillate purely longitudinally with an amplitude of up to approx. 300 mmpp. The surgical selective dissection effect starts at an amplitude of approx. 50 mmpp and tissue fragmentation, as well as other remote tissue effects, will increase with increasing amplitude. However, an amplitude of more than 300 mmpp should not be used. Even this amplitude means an extreme strain load on the horn, which explains why it must be manufactured from special titanium alloys emanating from the aircraft turbine-motor industry. Most surgical aspirators are designed to oscillate at approx. 24 kHz. Higher frequencies, for instance 35 kHz, are sometimes used for extremely precise work and for softer tissues. The peak-to-peak equivalent velocity of the tip will reach levels as high as approx. 24 m/s (80 km/h) at the full 300 mm setting. At this velocity the tip acceleration will be in the order of 3.8× 105 ×g where g is 9.81 m/s2.
43
The tip oscillation amplitude is generally measured using a measuring microscope. However, a simpler method using only a dial-gauge also gives satisfactory measurements. The dial-gauge should be mounted on a laboratory clamp to be in straight axial contact with the handpiece tip. In order to reduce tip wear caused by contact with the hardsteel sphere of the dial-gauge, short ultrasound test-pulses (less than 1 s) must be used. The dial-gauge will of course not follow the oscillation but instead statically deflect from the position of rest to the extended peak value. The indicated deflection (peak value) should therefore be multiplied by a factor of two to become the peak-topeak value.
1.4. Radiation force During handpiece usage a steady-state radiation force will be applied to the target tissue in the same direction as the sound path, which partly balances the opposite suction related force. The force induced by reflection is twice the force from absorption. The radiation force can also be used to estimate the acoustic total output power obtained from the handpiece. If a precision balance arrangement is used, P= Fk, where P is the acoustic power in watts (W), F is the absorbed induced force in Newtons (N) and the constant k: 1480. For instance 15 mN means a transmitted acoustic output power of approx. 22 W.
1.5. Fragmentation mechanisms With no load, the handpiece is in steady resonant vibration. This means that there is no strain at the tip end. When the tip approaches the target tissue, the tissue acts as an elongation of the resonance length and an acoustical mismatch arises. This causes the tissue to fail and, close to the tip, cavitation will be induced. Cavitation means the generation of gas-filled micro bubbles that affect the acoustical behaviour. Rectified diffusion (Crum and Hansen, 1982; Crum, 1984) pumps the micro cavities with dissolved gas into a gas phase, which tends to decrease the density and fragments the tissue into small aggregates. This
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
44
phenomenon explains the tissue-selective effect because the fragmentation will occur more strongly in tissues with a high water content. Parenchymal cells will thus be fragmented, while, for instance, blood vessels and stroma structures will be spared to a greater extent. By adjusting the tip amplitude, the surgeon can choose a sufficient level of watercontent related fragmentation. Delicate dissection, for instance of tumour capsule de-bulking procedures, can further be partly explained by the high acoustic acceleration which by its inertia prevents all but small fragments from following the rapid oscillation.
2. Materials Our design is based on an electrostrictive transducer. The parameters necessary for handpiece construction are basically calculated from the following differential equation valid for longitudinally resonant horn shapes (Eisner and Seager, 1965)
d 2p (ln A) + dx l
u¦ + u%
2
u=0
in which u is the particle displacement, A the cut-section area, l the wavelength and x the length co-ordinate. The prime symbol stands for the derivative with respect to time.
2.1. Dedicated computerised design simulation model Initially the simulation was made by independent software modules, for the transducer and for the horn respectively. The transducer is simulated using material data and dimension matrices according to the differential equation. The final code combines resonant solutions involving both the transducer and the horn section. Due to interactions between the sections, this might mean a significant feature to design short and light total systems, in which the horn can even be a part of the transducer. All significant resonator constants can be set into reasonable ranges. Many of these are auto-tuneable, which means that the designer can quickly sweep over a single pre-set range and collect only real solutions. Taken together this
turned out to be valuable since many earlier hidden solutions were uncovered. The numerical technique is based on an improved Runge-Kutta algorithm (Bjo¨rck and Dahlquist, 1969) and the adjustment method (Hazewinkel, 1995) that gives high stability even to intermediate sections with structures orthogonal to the x co-ordinate. The simulation model also indicates important structural planes as velocity nodes, anti nodes, etc. This is important due to the fact that the final assembly will be in an optimally slimmed housing with matched damping, etc. The dynamic simulation model has been progressively developed and refined during feedback from many physical experiments. Initially a DOS-based high-level language was used as an interactive menu-system. Later the program was re-coded in a Win95 visual type language (Visual Basic). The software displays a dedicated solution plot-box which is upgraded each time any parameter is changed.
2.2. Hazard and appro6als An aspirator handpiece as well as its driver console must not be hazardous to the patient or personnel. This of course also sets requirements for periodic maintenance as well as for correct handling. The most important issue is the electric patient leakage current (PLC). For a system in direct contact with, for instance, the central nervous system (CNS) the PLC should be limited to a frequency-weighted current of 10 mA or less (i.e. the Cardiac Floating level, CF), as specified by the EN 60 601-1 standard. It is also necessary to bring this type of surgical device in line with the Medical Devices Directive 93/42 EEC (CE standard), which calls for an appropriate design and intense tests of function, clinical efficiency and electromagnetic compatibility, EMC. 3. Results
3.1. De6elopment of a new highly miniaturised handpiece During 1996–1997, numerous experiments and
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
tests, guided by the dedicated simulation model, were performed and resulted finally in a short light-weight resonator system with acoustic performances almost as a standard large-size handpiece. The resonator system has in turn been adapted to a specially developed high-grade plastic housing as a complete micro-surgical handpiece (Fig. 1). It has been possible to design the transducer-system so that the piezoelectric sections have been fully isolated in the rear so-called dry compartment. This enhances the electrical safety and simultaneously improves the handpiece mechanical balance. The micro-surgical handpiece has been designed so that it is fully compatible with the standard driving console. With this new design the thermal losses from the transducer section have been limited since the dissipation heat decreased significantly on any amplitude level. The handpiece has so far been produced in an initial series that has been approved by the ITS Notified Body in Cranleigh, Surrey, and clinically trialled at two European hospitals.
45
3.2. Clinical results The new highly miniaturised ultrasonic aspiration handpiece was used in two cases of acoustic neurinoma removal.
3.2.1. Case 1 Left acoustic neuroma (2.3 cm diameter) in a 63-year-old male with no useful left sided hearing. The surgery was performed in the semi-prone, park bench position, with electro-myographics monitoring of facial nerve function. The retrosigmoid trans-meatal approach was used and a 4-cm craniotomy was made with an air-driven craniotome. The posterior portion of the mastoid was removed with a high speed burr so that a margin of the sigmoid sinus could be visualised. The dura was then opened close to this margin and sutured back to give a good exposure of the cerebellum and subsequently the cerebello-pontine angle. Intra-operative lumbar drainage was used to facilitate cerebellar retraction and to create good operative conditions. The cerebellum was gently retracted to an extent of 2 cm. This was sufficient
Fig. 1. The new micro-surgical handpiece version, the curved length-extended type. As seen, the surgical approx. 1.9 mm diameter tip is protected by a coaxial transparent tip-cover. The complete instrument including its integrated soft cable can be autoclaved at 135°C. Overall length is 250 mm; weight, 130 g.
46
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
to visualise the tumour surface and the adjacent petrous bone. The internal auditory canal was drilled away and tumour microsurgically loosened from the 7th and 8th nerves. The miniaturised ultrasonic aspirator was then used to carry out a subcapsular removal of the tumour. This was easily and efficiently accomplished. This phase of the surgery lasted for only 10 min, which included occasional homeostasis with bipolar diathermy. Only a thin rim of capsule remained. The capsule was then microsurgically dissected from the brainstem, and the cochlear, the vestibular and the facial nerves. A total tumour removal was achieved in a total procedure time of 4 h. The patient made a good recovery with normal facial nerve function (House grade 1).
3.2.2. Case 2 Retro-sigmoid trans-meatal removal of a 2.5-cm acoustic neurinoma in a 48-year-old female with no useful hearing on the right side. The same operative technique was used. This time the retro-mastoid free flap craniotomy was smaller measuring 3 cm in diameter. Once again, the cerebellum was gently retracted 2 cm, with conditions improved by lumbar spinal drainage. Miniaturised ultrasonic aspirator was used to debulk the tumour. The light weight and handling properties of the instrument were good which allowed a rapid and delicate subcapsular removal of the tumour. This stage took 8 min with some time for homeostasis. A radical central tumour debulking was achieved. The capsule of the tumour could then be easily manipulated and microsurgically dissected from the cranial nerves and brainstem. Procedure time: 5 h.
4. Discussion Neurosurgical techniques are becoming minimally invasive and the introduction of a new miniaturised ultrasonic aspirator is an important advance which permits removal of deeply placed tumours such as acoustic neurinomas using smaller approach corridors. This reduces procedure time and may also reduce patient morbidity and postoperative pain. .
Ultrasonic surgical aspiration has been used for many years to improve surgical efficiency and to improve feed-back tactility. In order to improve the accessibility, it is important to be able to design small, light weight, well-balanced handpieces without losing too much surgical efficiency at the working tip. A high grade of miniaturisation puts special demands on the design when for instance development of heat from losses in the internal parts of the handpiece becomes a more and more pronounced and difficult problem. A micro-surgical handpiece is a delicate resonant system with many interacting sections which must be carefully optimised as one single unit. The shorter handle on the new instrument makes it more comfortable to hold and to control since the centre of gravity is within the surgeon’s hand. Also, there is no problem using this instrument with the operating microscope. The surgical author has found that certain other ultrasonic aspirators have long handles which give a high centre of gravity (above the surgeon’s hand). This makes the instrument more clumsy and less precise. Also, these instruments are sometimes difficult to use in the span between the operating microscope objective and the craniotomy site.
References Bjo¨rck A, , Dahlquist G. Numerical Methods. Lund: CWK Gleerup, 1969:248 – 9. Crum LA. Ultrasonics 1984;22:215. Crum LA, Hansen GM. Growth of air bubbles in tissue by rectified diffusion. Phys Med Biol 1982;27:413 – 7. Eisner E, Seager JS. A longitudinally resonant stub for vibrations of large amplitude, Ultrasonics 1965; April –June: 88 – 98. Hazewinkel M, editor. Encyclopaedia of Mathematics: Nonlinear Equation, Numerical Methods, vol. 4. London: Kluwer, 1995:139 – 42. Kourtopoulos H, Vavruch L, Brunk U. The use of ultrasonic aspirator, Sonocut, for removal of brain tumours. Analysis of 26 consecutive cases. Neurochirurgia 1989;32:154 –6. Thomasson B, Hedenborg L, Wiksell H. Liver resection with the Sonocut ultrasonic knife. Prog Pediatr Surg 1990;25:48 – 57. Wiksell H, Granholm L. Development of a new system for surgical ultrasonic aspiration used in cancer surgery. Lakartidningen 1986;47:3990 – 3.
Basic Original
Miniaturised ultrasonic aspiration handpiece for increased applicability Hans Wiksell a,b,*, Harvey Martin c, Hugh Coakham d, Anders Berggren b, Sara Westermark a,b,e a
Department of Urology, Di6ision of Surgical Science, Karolinska Institute and Karolinska Hospital, Stockholm, Sweden b Comair AB, Box 39011, S-100 54 Stockholm, Sweden c Elekta Surgical Instruments Ltd., Ando6er, UK d Department of Neurosurgery, Frenchay Hospital and Uni6ersity of Bristol, Bristol, UK e Department of Laboratory Science and Technology, Di6ision of Medical Engineering, NOVUM and Karolinska Institute, Huddinge Hospital, Huddinge, Sweden Received 17 January 1999; received in revised form 5 July 1999; accepted 23 July 1999
Abstract Objecti6e: At present, ultrasonic aspiration is routinely used in several fields of surgery, especially in brain and spinal micro-surgery for tumour removal. In order to broaden the access to difficult surgical sites, it is important to design highly miniaturised but still efficient handpieces. The internal resonant system, always made of high-grade materials, must be optimally dimensioned. Normally this is done semi-empirically, by successively improving the design during many iterative test steps. This method however involves several additional difficulties when the degree of miniaturisation increases. For example, small transducer weights exacerbate heat-dissipation problems and make design optimisation important. Methods: To resolve these problems we have produced modelling software that makes it possible to simulate and automatically tune each individual interacting section of the design before it is actually manufactured, thereby assuring optimal efficiency. Results: Using a new mini-handpiece, designed via the software, two cases of dissection of acoustic neurinomas were successfully performed. Conclusion: Using conventional physical steps for improving ultrasonic aspiration handpieces, several problems arise when the grade of miniaturisation increases, due to increasing demands. We have designed computer software for handpiece simulation. Using this model it has been possible to manufacture a highly efficient miniaturised handpiece. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ultrasonic surgical aspiration; Micro-surgery; Tissue selective dissection; Brain tumour surgery
* Corresponding author. Tel.: + 46-8-664-4030; fax: +46-8-664-1820. E-mail address: [email protected] (H. Wiksell) 0929-8266/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 9 - 8 2 6 6 ( 9 9 ) 0 0 0 6 9 - 5
42
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
1. Introduction Ultrasonic aspiration methods were initially developed because of their potential for producing a tissue-selective dissection, for enhanced surgical tactility, for the removal of tissues with minimal damage to surrounding areas and for achieving a significant decrease in total blood loss (Wiksell and Granholm, 1986; Kourtopoulos et al., 1989; Thomasson et al., 1990).
1.1. Principles of ultrasonic aspiration handpieces An ultrasonic aspiration handpiece generally consists of an ultrasonic transducer attached to a resonant horn-shaped wave-guide amplitude magnifier that ends in a surgical tip. The transducer and magnifier are housed in an electrically and acoustically isolated hand-held precision casing. The surgical tip is often housed in a high-grade plastic coaxial tip-cover arrangement, in order to protect nearby tissues from accidentally being touched. This is important due to the frequent use of stereo-microscopy during the surgical procedure, which limits the field of view. The gradually tapered cross-section of the wave-guide creates the acoustic gain, due to the law of energy conservation versus the losses. The nearly exponential overall shape of the tapered sections further eliminates unacceptable standing-wave generating reflections. The horn is equipped with a central suction channel, typically of 1.5 mm i.d., for transporting tissue fragments to a container. The suction pressure can be adjusted down to approx. − 0.8 bar, which means an equivalent flat-area force on the target tissues of up to approx. 140 mN. This suction channel could be continuous through the handpiece to facilitate transducer cooling or it could be joined to an external tube, at a velocity node to reduce sealing losses. The tip is coaxially irrigated with sterile isotonic NaCl-solution which serves 1. to increase the acoustic coupling to the target; 2. cool the tip-material and the exposed tissues from internal frictional heating; 3. produce a liquid bulk to soak and enhance vacuum-transport of the tissue fragments
through the suction system without obstructions. The irrigation flow can typically be adjusted up to some ml/s. The mechanical resonance Q-value of the aspirator system must be high, in order to achieve the intended tip-amplitude. This calls for continuous frequency adjustment since the temperature varies slightly during handpiece usage. Changes in the temperature alter the resonant lengths, which therefore must be continuously monitored and adjusted. This is done by detecting the resonance deviations via a feed-back phase-locked loop (PLL), located in the driver console. From this it follows that the resonance bandwidth of each individual section must be carefully designed. The tip amplitude can also automatically compensate for different tissue loads using a quadrature directional-coupler as a power controlling feed-back. The energy stored in the elastic oscillation will be greater than the work done. The stored energy will therefore act as a flywheel and stabilise against fast variations in the acoustic load. An important feature of the ultrasonic handpiece is that it is designed to tolerate repeated sterilisation procedures in an ordinary autoclave at a nominal 135°C. Compared to cold sterilisation this means a significantly higher grade of safety against crosscontamination, since hidden spaces inside the handpiece will also be heated.
1.2. Magnetostricti6e and electrostricti6e transducers Two different types of transducers are used in surgical ultrasonic aspirators: magnetostricti6e and electrostricti6e. Magnetostrictive transducers were historically the first to be used. They consist for instance of a half-wavelength (approx. 10 cm at 24 kHz) resonant nickel plate package placed in a magnetic-field produced by a coil. The package is composed of many thin laminated and end welded nickel alloy plates to reduce eddy currents. In the magnetic field, the package is forced to mechanically oscillate due to magnetostrictive forces. The amplitude of such a transducer can be quite high, but the conversion efficiency is poor and the transducer thus normally needs to be
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
water cooled by a separate high-flow water loop. This loop provides a potentially serious sterility problem due to the risk of non-sterile leakage. The electrostrictive transducer mainly consists of a driving piezoelectric section. In practice, it is necessary to compose a sandwich Langevin-type transducer, which has resonator-bars in both ends. The piezoelectric material can only take a limited negative stress, which is why most high-power transducers use a positive mechanical pre-stress in order to improve the material’s fatigue strength. An electrostrictive transducer generally provides high stability and a high conversion efficiency. It generates a low amplitude but with great strain. The two resonator bars might have different lengths (so-called non-symmetric transducers) in order to further increase the amplitude. Still the output amplitude will be low, approx. 10 mmpeak-topeak, which is why the electrostrictive transducer horn needs a much higher amplitude-gain (typically a factor of 30) than the magnetostrictive. Designing high-gain horns must be done properly to avoid non-linear tendencies, which can act to create dangerous transversal spurious-oscillations by parametric pumping.
1.3. Tip amplitude The surgical tip will oscillate purely longitudinally with an amplitude of up to approx. 300 mmpp. The surgical selective dissection effect starts at an amplitude of approx. 50 mmpp and tissue fragmentation, as well as other remote tissue effects, will increase with increasing amplitude. However, an amplitude of more than 300 mmpp should not be used. Even this amplitude means an extreme strain load on the horn, which explains why it must be manufactured from special titanium alloys emanating from the aircraft turbine-motor industry. Most surgical aspirators are designed to oscillate at approx. 24 kHz. Higher frequencies, for instance 35 kHz, are sometimes used for extremely precise work and for softer tissues. The peak-to-peak equivalent velocity of the tip will reach levels as high as approx. 24 m/s (80 km/h) at the full 300 mm setting. At this velocity the tip acceleration will be in the order of 3.8× 105 ×g where g is 9.81 m/s2.
43
The tip oscillation amplitude is generally measured using a measuring microscope. However, a simpler method using only a dial-gauge also gives satisfactory measurements. The dial-gauge should be mounted on a laboratory clamp to be in straight axial contact with the handpiece tip. In order to reduce tip wear caused by contact with the hardsteel sphere of the dial-gauge, short ultrasound test-pulses (less than 1 s) must be used. The dial-gauge will of course not follow the oscillation but instead statically deflect from the position of rest to the extended peak value. The indicated deflection (peak value) should therefore be multiplied by a factor of two to become the peak-topeak value.
1.4. Radiation force During handpiece usage a steady-state radiation force will be applied to the target tissue in the same direction as the sound path, which partly balances the opposite suction related force. The force induced by reflection is twice the force from absorption. The radiation force can also be used to estimate the acoustic total output power obtained from the handpiece. If a precision balance arrangement is used, P= Fk, where P is the acoustic power in watts (W), F is the absorbed induced force in Newtons (N) and the constant k: 1480. For instance 15 mN means a transmitted acoustic output power of approx. 22 W.
1.5. Fragmentation mechanisms With no load, the handpiece is in steady resonant vibration. This means that there is no strain at the tip end. When the tip approaches the target tissue, the tissue acts as an elongation of the resonance length and an acoustical mismatch arises. This causes the tissue to fail and, close to the tip, cavitation will be induced. Cavitation means the generation of gas-filled micro bubbles that affect the acoustical behaviour. Rectified diffusion (Crum and Hansen, 1982; Crum, 1984) pumps the micro cavities with dissolved gas into a gas phase, which tends to decrease the density and fragments the tissue into small aggregates. This
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
44
phenomenon explains the tissue-selective effect because the fragmentation will occur more strongly in tissues with a high water content. Parenchymal cells will thus be fragmented, while, for instance, blood vessels and stroma structures will be spared to a greater extent. By adjusting the tip amplitude, the surgeon can choose a sufficient level of watercontent related fragmentation. Delicate dissection, for instance of tumour capsule de-bulking procedures, can further be partly explained by the high acoustic acceleration which by its inertia prevents all but small fragments from following the rapid oscillation.
2. Materials Our design is based on an electrostrictive transducer. The parameters necessary for handpiece construction are basically calculated from the following differential equation valid for longitudinally resonant horn shapes (Eisner and Seager, 1965)
d 2p (ln A) + dx l
u¦ + u%
2
u=0
in which u is the particle displacement, A the cut-section area, l the wavelength and x the length co-ordinate. The prime symbol stands for the derivative with respect to time.
2.1. Dedicated computerised design simulation model Initially the simulation was made by independent software modules, for the transducer and for the horn respectively. The transducer is simulated using material data and dimension matrices according to the differential equation. The final code combines resonant solutions involving both the transducer and the horn section. Due to interactions between the sections, this might mean a significant feature to design short and light total systems, in which the horn can even be a part of the transducer. All significant resonator constants can be set into reasonable ranges. Many of these are auto-tuneable, which means that the designer can quickly sweep over a single pre-set range and collect only real solutions. Taken together this
turned out to be valuable since many earlier hidden solutions were uncovered. The numerical technique is based on an improved Runge-Kutta algorithm (Bjo¨rck and Dahlquist, 1969) and the adjustment method (Hazewinkel, 1995) that gives high stability even to intermediate sections with structures orthogonal to the x co-ordinate. The simulation model also indicates important structural planes as velocity nodes, anti nodes, etc. This is important due to the fact that the final assembly will be in an optimally slimmed housing with matched damping, etc. The dynamic simulation model has been progressively developed and refined during feedback from many physical experiments. Initially a DOS-based high-level language was used as an interactive menu-system. Later the program was re-coded in a Win95 visual type language (Visual Basic). The software displays a dedicated solution plot-box which is upgraded each time any parameter is changed.
2.2. Hazard and appro6als An aspirator handpiece as well as its driver console must not be hazardous to the patient or personnel. This of course also sets requirements for periodic maintenance as well as for correct handling. The most important issue is the electric patient leakage current (PLC). For a system in direct contact with, for instance, the central nervous system (CNS) the PLC should be limited to a frequency-weighted current of 10 mA or less (i.e. the Cardiac Floating level, CF), as specified by the EN 60 601-1 standard. It is also necessary to bring this type of surgical device in line with the Medical Devices Directive 93/42 EEC (CE standard), which calls for an appropriate design and intense tests of function, clinical efficiency and electromagnetic compatibility, EMC. 3. Results
3.1. De6elopment of a new highly miniaturised handpiece During 1996–1997, numerous experiments and
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
tests, guided by the dedicated simulation model, were performed and resulted finally in a short light-weight resonator system with acoustic performances almost as a standard large-size handpiece. The resonator system has in turn been adapted to a specially developed high-grade plastic housing as a complete micro-surgical handpiece (Fig. 1). It has been possible to design the transducer-system so that the piezoelectric sections have been fully isolated in the rear so-called dry compartment. This enhances the electrical safety and simultaneously improves the handpiece mechanical balance. The micro-surgical handpiece has been designed so that it is fully compatible with the standard driving console. With this new design the thermal losses from the transducer section have been limited since the dissipation heat decreased significantly on any amplitude level. The handpiece has so far been produced in an initial series that has been approved by the ITS Notified Body in Cranleigh, Surrey, and clinically trialled at two European hospitals.
45
3.2. Clinical results The new highly miniaturised ultrasonic aspiration handpiece was used in two cases of acoustic neurinoma removal.
3.2.1. Case 1 Left acoustic neuroma (2.3 cm diameter) in a 63-year-old male with no useful left sided hearing. The surgery was performed in the semi-prone, park bench position, with electro-myographics monitoring of facial nerve function. The retrosigmoid trans-meatal approach was used and a 4-cm craniotomy was made with an air-driven craniotome. The posterior portion of the mastoid was removed with a high speed burr so that a margin of the sigmoid sinus could be visualised. The dura was then opened close to this margin and sutured back to give a good exposure of the cerebellum and subsequently the cerebello-pontine angle. Intra-operative lumbar drainage was used to facilitate cerebellar retraction and to create good operative conditions. The cerebellum was gently retracted to an extent of 2 cm. This was sufficient
Fig. 1. The new micro-surgical handpiece version, the curved length-extended type. As seen, the surgical approx. 1.9 mm diameter tip is protected by a coaxial transparent tip-cover. The complete instrument including its integrated soft cable can be autoclaved at 135°C. Overall length is 250 mm; weight, 130 g.
46
H. Wiksell et al. / European Journal of Ultrasound 11 (2000) 41–46
to visualise the tumour surface and the adjacent petrous bone. The internal auditory canal was drilled away and tumour microsurgically loosened from the 7th and 8th nerves. The miniaturised ultrasonic aspirator was then used to carry out a subcapsular removal of the tumour. This was easily and efficiently accomplished. This phase of the surgery lasted for only 10 min, which included occasional homeostasis with bipolar diathermy. Only a thin rim of capsule remained. The capsule was then microsurgically dissected from the brainstem, and the cochlear, the vestibular and the facial nerves. A total tumour removal was achieved in a total procedure time of 4 h. The patient made a good recovery with normal facial nerve function (House grade 1).
3.2.2. Case 2 Retro-sigmoid trans-meatal removal of a 2.5-cm acoustic neurinoma in a 48-year-old female with no useful hearing on the right side. The same operative technique was used. This time the retro-mastoid free flap craniotomy was smaller measuring 3 cm in diameter. Once again, the cerebellum was gently retracted 2 cm, with conditions improved by lumbar spinal drainage. Miniaturised ultrasonic aspirator was used to debulk the tumour. The light weight and handling properties of the instrument were good which allowed a rapid and delicate subcapsular removal of the tumour. This stage took 8 min with some time for homeostasis. A radical central tumour debulking was achieved. The capsule of the tumour could then be easily manipulated and microsurgically dissected from the cranial nerves and brainstem. Procedure time: 5 h.
4. Discussion Neurosurgical techniques are becoming minimally invasive and the introduction of a new miniaturised ultrasonic aspirator is an important advance which permits removal of deeply placed tumours such as acoustic neurinomas using smaller approach corridors. This reduces procedure time and may also reduce patient morbidity and postoperative pain. .
Ultrasonic surgical aspiration has been used for many years to improve surgical efficiency and to improve feed-back tactility. In order to improve the accessibility, it is important to be able to design small, light weight, well-balanced handpieces without losing too much surgical efficiency at the working tip. A high grade of miniaturisation puts special demands on the design when for instance development of heat from losses in the internal parts of the handpiece becomes a more and more pronounced and difficult problem. A micro-surgical handpiece is a delicate resonant system with many interacting sections which must be carefully optimised as one single unit. The shorter handle on the new instrument makes it more comfortable to hold and to control since the centre of gravity is within the surgeon’s hand. Also, there is no problem using this instrument with the operating microscope. The surgical author has found that certain other ultrasonic aspirators have long handles which give a high centre of gravity (above the surgeon’s hand). This makes the instrument more clumsy and less precise. Also, these instruments are sometimes difficult to use in the span between the operating microscope objective and the craniotomy site.
References Bjo¨rck A, , Dahlquist G. Numerical Methods. Lund: CWK Gleerup, 1969:248 – 9. Crum LA. Ultrasonics 1984;22:215. Crum LA, Hansen GM. Growth of air bubbles in tissue by rectified diffusion. Phys Med Biol 1982;27:413 – 7. Eisner E, Seager JS. A longitudinally resonant stub for vibrations of large amplitude, Ultrasonics 1965; April –June: 88 – 98. Hazewinkel M, editor. Encyclopaedia of Mathematics: Nonlinear Equation, Numerical Methods, vol. 4. London: Kluwer, 1995:139 – 42. Kourtopoulos H, Vavruch L, Brunk U. The use of ultrasonic aspirator, Sonocut, for removal of brain tumours. Analysis of 26 consecutive cases. Neurochirurgia 1989;32:154 –6. Thomasson B, Hedenborg L, Wiksell H. Liver resection with the Sonocut ultrasonic knife. Prog Pediatr Surg 1990;25:48 – 57. Wiksell H, Granholm L. Development of a new system for surgical ultrasonic aspiration used in cancer surgery. Lakartidningen 1986;47:3990 – 3.