1174 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 53, NO.

6, JUNE 2006

Microwave Drilling of Bones

Yael Eshet, Ronit Rachel Mann, Abby Anaton, Tomer Yacoby, Amit Gefen, Member, IEEE, and
Eli Jerby*, Member, IEEE

Abstract—This paper presents a feasibility study of drilling in

fresh wet bone tissue in vitro using the microwave drill method
[Jerby et al., 2002], toward testing its applicability in orthopaedic
surgery. The microwave drill uses a near-field focused energy
(typically, power under 200 W at 2.45-GHz frequency) in order
to penetrate bone in a drilling speed of 1 mm/s. The effect
of microwave drilling on mechanical properties of whole ovine
tibial and chicken femoral bones drilled in vitro was studied
using three-point-bending strength and fatigue tests. Properties
were compared to those of geometrically similar bones that were
equivalently drilled using the currently accepted mechanical
rotary drilling method. Strength of mid-shaft, elastic moduli, Fig. 1. Debris around the hole after mechanical drilling in vitro in cortical bone
of chicken femora.
and cycles to failure in fatigue were statistically indistinguishable
between specimen groups assigned for microwave and mechanical
drilling. Carbonized margins around the microwave-drilled hole
were 15% the hole diameter. Optical and scanning electron the microwave energy characteristics for in vitro drilling in
microscopy studies showed that the microwave drill produces sub- fresh (wet) bones as well as the results of mechanical and
stantially smoother holes in cortical bone than those produced by histological tests conducted in the microwave-drilled bones.
a mechanical drill. The hot spot produced by the microwave drill
has the potential for overcoming two major problems presently A. Surgical Drilling in Hard Tissues: Review Up-to-Date
associated with mechanical drilling in cortical and trabecular bone
during orthopaedic surgeries: formation of debris and rupture of Drilling in bone is a common surgical procedure, which may
bone vasculature during drilling. be required during preparation for insertion of a fixative or-
thopaedic implant such as nail, screw or wire [5], or before in-
Index Terms—Carbonization, mechanical properties, orthopaedic
surgery, thermal damage. sertion of a bone graft to enhance bone healing [6]. Presently,
mechanical rotary drillers are the only type used in the clinical
setting. Rotary drilling is performed at a wide range of speeds,
I. INTRODUCTION from low to moderate ( rpm) up to ultra-high (
rpm) [7]. During such rotary drilling, bone debris accumulates
around the drilling site (Fig. 1). Although measures are taken
D RILLING in bones for orthopaedic and dental purposes
has been a common practice for decades, using a variety
of drilling bits [1], [2]. Rotary drills used today are efficient;
to carefully clean bone debris at the preclosure stage of surgery,
remaining particles may induce a foreign-body-reaction around
however, they suffer several drawbacks including debris and the implantation site. Such effects may delay bone healing or
chips spread resulting in foreign-body-reactions, substantial interfere with the process of osseointegration of the implant
hematoma at the drilling site, heat generation, difficulties in [8]. Rotary drilling also ruptures the vasculature at the drilling
attaining geometrical accuracy, and wobbling [1], [2]. An alter- site. Substantial hematomas around the sites of rotary drilling
native, more recent approach employs lasers for bone drilling were demonstrated using histology of rat femoral diaphysis [5].
[3], however, this method may be too costly for large-scale use Hematoxylin and eosin (H&E) histological staining in the same
in the clinical setting. This paper presents a novel approach for study also revealed coagulum material in the blood vessel lu-
drilling of bones, using near-field microwaves [4], and provides mens around the drilling site. The coagulum partially or fully
occluded the vasculature at the site of drilling [5], imposing a
second inhibiting effect on bone healing and osseointegration.
Manuscript received November 25, 2004; revised September 25, 2005. This Worse still, the ruptured vasculature and lymph systems provide
work was supported by the Israel Science Foundation under Grant 1270/04. As-
terisk indicates corresponding author. a portal for infections and for entrance of toxins or wear parti-
Y. Eshet is with the Departments of Biomedical Engineering and Physical cles that may originate from the implant [9]. Since the stability
Electronics and Faculty of Engineering, Tel-Aviv University, Tel Aviv 69978,
Israel (e-mail: jerby@eng.tau.ac.il).
of an orthopaedic fixation or a bone graft highly depend on the
R. R. Mann is with the Departments of Biomedical Engineering and Faculty quality and quantity of the host bone, the above disadvantages of
of Engineering, Tel-Aviv University, Tel Aviv 69978, Israel. mechanical drilling may lead to longer postoperative recovery
A. Anaton and T. Yacoby are with the Departments of Physical Electronics
and Faculty of Engineering, Tel-Aviv University, Tel Aviv 69978, Israel.
periods.
A. Gefen is with the Departments of Biomedical Engineering and Faculty of Excessive heat generation in bone tissue during mechanical
Engineering, Tel-Aviv University, Tel Aviv 69978, Israel. drilling has been reported, and was attributed mainly to friction
*E. Jerby is with the Departments of Physical Electronics and Faculty of during penetration of the rotary drill [7]. The heat generated by
Engineering, Tel-Aviv University, Ramat Aviv, Tel Aviv 69978, Israel (e-mail:
jerby@eng.tau.ac.il). the friction between the drilling bit and bone tissue was shown
Digital Object Identifier 10.1109/TBME.2006.873562 to increase moderately between low and high drilling speeds,
0018-9294/$20.00 © 2006 IEEE

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 ESHET et al.: MICROWAVE DRILLING OF BONES 1175

and excessively when chips clog the flutes of the drill [7]. Tem-
peratures of 89 C up to 185 C were recorded in distances of
0.5 mm from the mechanically drilled holes [10], [11].
Rotary mechanical drills are difficult to be guided accurately.
When touching the smooth bone cortex, rotary drills tend to
slide and may dislocate or misalign [2]. Moreover, the diam-
eter of the hole created is larger than the drill bit diameter, due
to wobbling effects [1]. After prolonged or repeated use, the
dulling of the drill-bit edges increases heat generation and de-
creases geometrical accuracy. Several attempts for improving
the drilling process have been described, all concerned with drill
bit geometry [1], drilling mechanism [12] or optimization of pa-
rameters (applied force, drilling speed, etc.).
Laser ablation of hard tissues including bones has been com-
pared to rotary drills for a variety of wavelengths and pulse
structures [3]. Under proper cooling conditions, the results ob-
tained with laser ablation generally show high accuracy and
clean cuts. The main disadvantage in laser drilling is the absent Fig. 2. Scheme of the microwave drill system. (a) Microwave energy is con-
or delayed healing of the ablation site due to photo-acoustic and centrated in a small hot spot in front of the drilling bit, thus enabling its inser-
tion into the bone. (b) Block diagram of the microwave drill cascade consisting
thermal damage effects [3]. of a 2.45-GHz magnetron tube protected by an isolator, a reflectometer unit,
Taken together, the literature above indicates that a new an impedance-matching tuner (to minimize the reflected power), and the mi-
drilling method is needed in orthopaedic surgery. Specifically, crowave drill device [shown schematically in (a)].
such new method should not produce bone debris, should not
rupture blood vessels in bone during penetration and should
not involve wobbling effects during drilling. Microwave-based mechanism has no rotating or vibrating parts, and it does not
drilling can potentially fill these requirements. produce debris particles. The heat generated under the drilling
tip is anticipated to immediately fuse the bone vasculature
B. Microwave-Tissue Interactions crossing the drilling path, thus eliminating hematomas and
Microwave interactions with biological tissues were studied related coagulum. The microwave drill apparatus (Fig. 2) is
during the last several years [13], [14] but no attempt was made considerably less expensive than any laser-based drill.
to employ microwave energy for the purpose of drilling in The objectives of this study were to 1) characterize the
hard tissues. Reported microwave-tissue interactions include microwave parameters (power, drilling speed) for drilling in
heat generation by resistive losses of moving charged-ions fresh wet bones in vitro; 2) determine if microwave drilling
and oscillations of charged molecules, and heat transfer by has degrading effects on mechanical strength and stiffness of
induced movements of charged ions. At temperatures over 50 fresh whole bones in vitro compared with standard mechanical
C, tissues undergo vaporization and carbonization. Higher drilling; 3) approximate the extent of tissue carbonization at the
temperatures may cause desiccation, protein denaturation, co- margins of holes produced by the microwave drill in vitro.
agulation and finally welding and cavitation [13], [14]. It is not
yet determined whether RF radiation has additional nonthermal II. METHODS
effects on biological tissues, though a recent study indicated the
possibility of its carcinogenicity [13]. Tumor ablation by RF A. Experimental Setup of the Microwave Drill
radiation has drawn a growing interest recently [15]. The aim The experimental setup of the microwave drill is shown in
of this procedure is to cause coagulation necrosis of cancerous Fig. 2(b). The microwave cascade consists a power-controlled
tissue and, thus, tumor lysis and ablation. A percutaneous magnetron (2.45 GHz) protected by an isolator, a reflectometer
electrode is inserted under imaging guidance (CT, MRI or US) for measuring the transmitted and reflected wave power, an E-H
to ablate subdermal lesions. The method was also used on tuner for impedance matching, and a transition from the WR340
bones for ablation of tumors such as osteoid osteoma [16] and rectangular waveguide to the coaxial microwave drill device
metastases [17]. shown in Fig. 2(a). The coaxial structure was cooled by pres-
surized air ( bars). A constant external force of N was
C. The Microwave Drill applied axially on the center electrode by means of weights (2
A new method for general purpose drilling which employs Kg), mounted so that their center of mass was aligned with the
near-field microwaves was recently introduced by Jerby et al. axis of the electrode.
[4]. The microwave drill device is depicted in Fig. 2(a). The For drilling in bovine trabecular bone (from the proximal
coaxial electrode radiates the microwave energy in the near tibial diaphysis), an effective power of 150–200 W was
field, thus producing a confined hot spot under the tip of the required to produce 2.4-mm-diameter holes within 2–5 s. For
electrode. The hot spot increases the local dielectric losses ovine tibias with diaphysial cortical thickness of – mm,
of the material in a thermal runaway process [18], allowing an effective microwave power of W was sufficient for
the electrode to penetrate deeper into the molten region. This drilling 2.4-mm-diameter penetrating holes (using an electrode

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 1176 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 53, NO. 6, JUNE 2006

TABLE I
DIMENSIONS OF OVINE TIBIAS AND CHICKEN FEMORA USED FOR DRILLING
6
(MEANS STANDARD DEVIATIONS)

connective tissues were gently removed from 22 ovine tibia and

18 chicken femur specimens that were obtained fresh from a
local butchery. While sheep are a well-accepted and commonly
used orthopaedic model, chicken bones are not as widely
used, however, in a previous publication [19] we discussed
the similarities between mechanical properties of human and
chicken cortical bone, particularly, under three-point bending.
These similarities in mechanical properties allow to use chicken
Fig. 3. Microwave drilling in cortical bone in vitro. (a) Transverse slice in the
shaft after microwave drilling (ovine tibia). The carbonized margins around the femora as an orthopaedic model in the present study. Sheep
hole can be identified. The scheme on the right frame demonstrates the method were 5 months old at the time of sacrifice, and hens were 10
of calculation of the carbonization effect ratio. (b) Microwave drilling showed months old (i.e., both had completely ossified bones). It has
no visible thermal damage to bone marrow (ovine tibia, with marrow exposed
under the fractured bone). been reported that mechanical properties of bone specimens
that were frozen at C (for less than 100 days) and sub-
sequently thawed to room temperature were indistinguishable
with the same diameter) within less than 5 s. Hence, the ef- from properties of fresh bones (for trabecular tissue, cortical
fective power needed to drill in bovine trabecular and ovine tissue and whole bones) [20], [21]. Based on these reports,
cortical bones generally resembled. For drilling in chicken we stored the fresh cleaned tibial and femoral bones frozen at
femora (with thinner cortex, mm), a lower effective power C until the day of drilling (no longer than 1 month from
of W was required, to produce 1-mm-diameter holes the time of freezing). Specimens were thawed to room temper-
(using a 1-mm-diameter electrode) within less than 1.8 s. Fig. 3 ature (25 C) 2 h before drilling, and were kept moist (using
shows for instance microwave drilling results in cortical ovine a spray of normal saline) until drilling and during mechanical
tibia bone in vitro. testing.
A motion detector installed on the microwave drill bit in- In preparation for fixation of long bones (e.g., using screws
dicated the depth of penetration into the bone. Preliminary and plates) drilling is typically performed through cortical bone
drilling experiments were conducted with fresh wet cortical in the mid-shaft [7]. The experimental design was, therefore,
and trabecular bovine bones, to optimize the following pa- focused on the potential effects of microwave drilling on cortical
rameters for an efficient, low-energy and shortest drilling bone around that site.
process: drill-bit depth, impedance matching (minimizing the Means and standard deviations of the dimensions of ovine
microwave reflected power), hole geometry, and the scorching tibias and chicken femora were obtained using a caliper (res-
radius. olution 0.01 mm) and are provided in Table I. In order to en-
sure uniformity of specimens assigned for testing, we allowed
B. Mechanical Property Studies standard deviations of less than 10% for bone diameter, less
We determined the feasibility of drilling in bones in vitro than 5% for bone length, and less than 30% for bone thickness
using the new microwave system, by drilling in bovine cortical (Table I). Using two-tailed unpaired -tests we verified that spec-
and trabecular bone, ovine cortical bone and chicken cortical imens assigned for mechanical drilling (ovine tibia: ;
bone. In order to test the effect of microwave drilling on bone chicken femora: ) had mid-shaft diameter, cortical thick-
mechanical properties, we drilled holes of the same diameter, in ness, length and cross-sectional area that were statistically indis-
the same location in diaphyses of long bones, using a standard tinguishable from those of specimens assigned for microwave
mechanical (rotary) drill and the new microwave drill described drilling (ovine tibia: ; chicken femora: ).
above, and compared the mechanical performances of the bones, For mechanical drilling, we used a standard rotary drill
in terms of bending strength, elastic modulus of the bone shaft (MultiPro, Dremel Co.) mounted on a drill press device. The
following the drilling, and tolerance to fatigue. drill press device ensured stability and reproducibility of
1) Strength and Elastic Modulus: The success of a fracture drilling sites and orientations among bone specimens assigned
fixation or insertion of bone graft material is determined mainly for mechanical drilling. The rotary drilling rate was 20 000
by the strength and elastic modulus of the repaired bone. In rpm, corresponding to mid-range drilling rates used clinically
preparation for mechanical testing of these properties, the soft [7]. Using either mechanical (rotary) or microwave drills, we

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 ESHET et al.: MICROWAVE DRILLING OF BONES 1177

Fig. 4. Mechanical testing of drilled ovine tibias using three-point bending. (a) Experimental apparatus. (b) Scheme of the experiment showing the lower and
upper supporting jigs (triangles). The upper jig applies a flexural force F which is balanced by reaction forces F=2 at each supporting lower jig. Lower jigs are
L distance apart. The drilled hole is positioned opposed to the point of application of the flexural force. (c) Schematic cross-section through the bone mid-shaft
defining the major and minor diameters of the shaft.

created a single penetrating hole in the center of the shaft, with as function of the flexural displacement caused by the
diameter of mm and mm for the ovine tibias upper support [Fig. 4(b)]. The failure load was used to
and chicken femora, respectively. In mechanical drills, pene- calculate the strength of bone under bending [22]
tration speed was mm/s. In microwave drills, penetration
speed was mm/s for drilling in ovine tibias, and
mm/s for drilling in chicken femora. The center of the shaft (1)
was identified and marked for each specimen as half the bone
length (Table I). where is the minor diameter and is the moment of inertia of
After drilling, each bone was subjected to a three-point the cross-sectional area around the axis of the major diameter
bending test at a deflection rate of 1 mm/min using an Instron (Fig. 4(c), Table I). The slope of the force-displacement curve,
5544 testing machine [Fig. 4(a)]. The span between the lower , was further used to calculate the elastic modulus of
supports [ , Fig. 4(b)] was scaled for bone length, and ranged cortical bone tissue [22]
between 13 and 14 cm for ovine tibias, and was set as 4.6 cm
for chicken femora. The upper support was pressing against
the bone cortex at the side opposed to the location of the hole (2)
[Fig. 4(b)], so that bending-related tensile stresses were applied
around the hole. A load cell with maximum capacity of 2 KN
We compared each mechanical property across groups
was used to measure the applied load , which was recorded
assigned for mechanical and microwave drilling, separately for

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 1178 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 53, NO. 6, JUNE 2006

ovine tibias and chicken femora, using 2-tailed unpaired -tests.

A value less than 0.05 was considered statistically significant
in all statistical tests.
2) Fatigue: Fatigue studies were conducted in chicken
femora to simulate the endurance of drilled bones to a more
physiological loading scenario (i.e., repetitive, and less than
ultimate strength loads). We assigned bones to three experi-
mental groups: 1) drilled by microwave ; 2) drilled
mechanically ; 3) controls, which were not drilled
. One-way ANOVA for each geometrical dimension
(Table I) confirmed that all bone dimensions were insignificant
across groups and hence, geometrical uniformity of specimens
was verified. Specimen preparation and the drilling process
were as described in Section B–1.
Chicken femora from all groups were subjected to fatigue
in three-point bending. The load amplitude was set as 220 N,
which is % of the load to failure in three-point bending
for the bone geometries considered herein (based on prelimi-
nary three-point bending strength studies, ). The loading
speed for fatigue was set as 50 mm/min, and the span between
the lower supports was again set as 4.6 cm. To determine if
microwave drilling had a different effect on cycles to fatigue
failure compared with mechanical drilling and controls, we ran
Fig. 5. The microwave effective power (solid line) and drilling depth (dashed
a one-way analysis of variance (ANOVA). line) versus time during microwave drilling through (a) mid-shaft of ovine tibial
bone (cortical bone) and (b) bovine trabecular bone.
C. Area of Carbonization
Subsequent to mechanical testing we measured the area of
tissue carbonization induced by the hot spot for ovine tibias the bone samples on aluminum discs, using conductive carbon
drilled using the microwave drill. For that purpose, slices of paint. The samples were coated with gold using a sputter coater
bone from both sides of the microwave-generated hole were (SC500, Polaron Co., UK). The acceleration voltage of the
cut transversally [Fig. 3(a)] using an electrically powered di- SEM was set as 15 KV and the filament current was
amond-coated disk saw (MultiPro, Dremel Co.). Measurements Amperes.
of area of carbonization were conducted using image analysis
software (SigmaScan Pro, SPSS Inc.) on digital photos of the III. RESULTS
bone slices (taken at a high-resolution of 2.3 mega pixels). A
graph paper was included in the images for calibration of di- A. Characterization of Microwave Parameters
mensions. We measured the areas of carbonization in the mar- The variation of effective microwave power with depth of
gins of the hole and the projected area of the hole (“effective penetration and time during the drilling process is demonstrated
drilling area”) on each slice as shown in Fig. 3(a), right frame. in Fig. 5 for ovine cortical [Fig. 5(a)] and bovine trabecular
We then divided the effective drilling area plus its carbonized [Fig. 5(b)] bone specimens. The effective microwave power for
margins by the net effective drilling area to obtain a carboniza- drilling in cortical and trabecular bone components in vitro was
tion effect ratio. Ideally, when minimal carbonization occurs, kept under W and holes were produced after no more than
this ratio should approach unity. 5 s. The effective microwave power varied during the drilling
process (Fig. 5) due to the variations in the microwave load-
D. Optical and Scanning Electron Microscopy (SEM) impedance and the consequent reflection from the bone as the
Additional chicken femora were drilled and assigned for drilling process evolved.
optical and SEM (microwave-drilled: ; mechanically
B. Mechanical Property Studies
drilled: ). Segments of 9 mm 9 mm were cut from
these bones using a disk saw, so that they contained the hole in 1) Strength and Elastic Modulus: Unpaired 2-tailed -tests
their center. Samples were first studied under digital optical mi- showed that mechanical properties of ovine tibias subjected
croscopy (Axiolab A, Zeiss Co., reflective, magnification ). to microwave drilling were statistically indistinguishable from
Measurements of spots of carbonization in microwave-drilled those of tibias drilled mechanically [bending strength
specimens were taken using a special ruler for optical mi- MPa, elastic modulus GPa, Fig. 6(a) and (b)]. Simi-
croscopy (resolution 100 m) and a designated micrograph larly, mechanical properties of chicken femora drilled with
image processing software (SigmaScan Pro). microwave or drilled mechanically were indistinguishable
Second, samples were prepared for SEM studies (JSM840A, [bending strength MPa, elastic modulus GPa,
Jeol Co., MA, USA). In preparation for SEM, we mounted Fig. 6(a), (b)].

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 ESHET et al.: MICROWAVE DRILLING OF BONES 1179

2
Fig. 7. Digital optical microscopy images (magnification 30) of hole bound-
aries in chicken femora drilled-mechanically (a) and (b) and with microwave
(c). For microwave-drilled bone, spots of carbonization (size 815–2600 m )
are marked with black arrows around the hole perimeters. Nevertheless, the hole
geometry in microwave-drilled samples (c) is substantially smoother than in me-
chanically drilled samples (a). Specifically, the mechanically drilled holes are
characterized by sharp fragments that were still attached to the hole surface.
Scratches around the mechanically drilled holes were also identified (b).

C. Area of Carbonization
For trabecular bone (bovine), no peripheral surface scorching
was apparent, but crosscuts sometime showed slight scorching
at the margins of the hole and in bone marrow contained in
trabecular spaces. Contrarily, none of the ovine tibias drilled
with microwave showed visual evidence of thermal damage to
bone marrow contained within the mid-shaft cortex [Fig. 3(b)].
Scorching of mm was visible on the cortical bone sur-
face around the drilled hole [Fig. 3(a)]. The carbonization effect
ratio [Fig. 3(a)] for tibias drilled with the microwave drill was
.

D. Optical and Scanning Electron Microscopy

The optical microscopy images of the boundaries of the holes
Fig. 6. Comparison of mechanical properties of long bones after drilling: (a)
(Fig. 7) show remarkable differences between holes drilled
bending strength; (b) elastic modulus; (c) fatigue. Bars indicate means and ver- mechanically and with aid of microwave. Specifically, spots
tical lines indicate standard deviations. of carbonization [size 815–2600 m , measured using image
analysis software, Fig. 7(c)], not seen by visual inspection, can
be identified on the hole perimeter in microwave-drilled bone.
We conclude that if drilling (microwave or mechanical) re- However, the hole geometry is substantially smoother than
duces mechanical properties of cortical bone in vitro, then mi- in mechanically drilled bone. The mechanically drilled holes
crowave drilling does not induce a greater deteriorating effect were characterized by sharp fragments that were still attached
on bone properties than does mechanical drilling. to the hole surface [Fig. 7(a)]. Around the mechanically drilled
2) Fatigue: Using ANOVA, we found that the number hole, there were some scratches which were likely caused by
of cycles to failure in microwave-drilled chicken femora the initial drilling maneuver to place and stabilize the drill
( deviation cycles) was statistically [Fig. 7(b)].
indistinguishable from those drilled mechanically ( The results of the SEM analyses are consistent with those
cycles). Femora which were not drilled (controls) showed a of optical microscopy, but the high-resolution of the SEM de-
slightly higher number of cycles to failure ( cycles) tected additional details not seen in optical microscopy (Fig. 8).
[Fig. 6(c)], but without statistically significant difference SEM images of the entire holes demonstrated that hole geom-
compared to the drilled femora, very likely because the hole etry is substantially smoother in microwave drilling [Fig. 8(a)
diameter was very small ( mm) compared to bone dimen- and (b)]. Specific examples of defects in mechanical drilling
sions (length mm, diameter mm). We conclude that are the large fragment of bone partially detached from the hole
microwave radiation did not influence the resistance of chicken surface [marked by arrow “A” in Fig. 8(a), right frame], sharp
femora to fatigue. bone edges with characteristic size of 100 m that were still

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 1180 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 53, NO. 6, JUNE 2006

Fig. 8. SEM images of chicken femora drilled with microwave (left column) and mechanically (right column): (a) view of the holes from above; (b) and (c) typical
imperfections attached to the hole perimeters. Hole geometry is substantially smoother in microwave drilling (a). Typical defects in mechanical drilling are the
large fragment of bone partially detached from the hole surface (arrow “A”), sharp bone edges still attached to the contour of the hole (arrow “C”), and scratches
[(c), right frame]. Typical imperfections in microwave drilling are strut-like elements around the hole perimeter (arrow “B”) and small lumps (region “D”), which
were apparently formed by melted bone minerals. Magnifications and scales are provided under each SEM frame.

attached to the contour of the hole after drilling [arrow “C,” hole perimeter, with length of less than 1 m and thickness of
Fig. 8(b), right frame], and scratches with width of about 10 m less than 0.2 m [arrow “B,” Fig. 8(b), left frame], which ap-
around the hole, which were probably caused by drill vibrations pear to be hardened fibers of melted bone minerals. A second
during the first contact with bone [Fig. 8(c), right frame]. SEM defect characteristic of microwave drilling was small lumps [di-
images of microwave drilling also showed some typical imper- mensions of 100–200 m, region “D” in Fig. 8(c), left frame],
fections likely to be related with the heat generated during the apparently of melted bone minerals, which were pushed out of
drill. Specifically, we observed strut-like elements around the the hole by the microwave electrode.

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 ESHET et al.: MICROWAVE DRILLING OF BONES 1181

IV. DISCUSSION rotary drilling maneuvers [7]. The authors of this paper believe
that the advantages of site reproducibility and stability provided
The feasibility of drilling in cortical and trabecular bone by the drill press apparatus were required for the present exper-
tissues using microwave radiation was demonstrated in this in imental design, which was aimed at comparing “ideal” rotary
vitro study. The hot spot produced by the microwave drill has and microwave drills.
the potential for overcoming a major problem currently related Safety of the patient undergoing surgery and of the staff who
with mechanical drilling in bone: formation of debris leading to operates the drill is a critical issue to consider at this stage of
foreign-body-reaction (Figs. 1, 6, and 7). When passing through research, as microwave radiation may be hazardous. A tailored
blood vessels in bone, the microwave drill can potentially weld design of the microwave drill and its screening for each specific
the vessels contacting the hot spot (as opposed to rupture surgical application is expected to significantly reduce the radi-
of vessels during mechanical drilling). Hence, the risk for ation exposure of both the patient and staff to a permitted min-
infection may be lower in microwave drilling. The microwave imum. It should be noted that existing RF ablation procedures
drilling process is relatively quick and is geometrically precise [e.g., 16] use RF power of up to 150 W for about 15 min (
(no mechanical wobbling is involved). Drilling penetration s), much longer than needed for microwave drilling in bones.
can be monitored and controlled on a timescale in the order Operated in a similar power level, microwave drilling through
of fractions of a second (Fig. 5). Importantly, the microwave cortical bone only lasts 2–3 s (Fig. 5). Nevertheless, research ef-
drill system is substantially more economical than laser-based forts are required to determine whether microwave drilling in-
drills. The microwave drill technology is versatile and with duces a risk for carcinogenic effects, and how such risk can be
tailored design, can be considered for a wide range of clinical minimized.
applications. These may include insertions of orthopaedic
pins, nails and screws into bones, wire fixation, neurosurgery,
REFERENCES
dentistry, cranial surgeries, open-chest cardiac surgeries, and
any other procedure requiring accurate, carefully controlled [1] S. Saha, S. Pal, and J. A. Albright, “Surgical drilling: Design and
performance of an improved drill,” J. Biomech. Eng., vol. 245, pp.
bone drilling. 252–109, 1982.
In vitro drilling in bone provided several encouraging results: [2] K. L. Wiggins and S. Malkin, “Drilling of bone,” J. Biomech., vol. 553,
1) microwave drilling did not degrade the mechanical proper- pp. 559–9, 1976.
[3] J. T. Payne, G. M. Peavy, L. Reinisch, and D. C. Van Sickle, “Cortical
ties of bone in vitro more than rotary drilling did; 2) carboniza- bone healing following laser osteotomy using 6.1 m wavelength,”
tion at the margins of the hole in cortical bone was relatively Lasers Surg., Med., vol. 38, pp. 43–29, 2001.
small ( % of the effective hole diameter); 3) holes produced [4] E. Jerby, V. Dikhtyar, O. Aktushev, and U. Groslick, “The microwave
drill,” Science, vol. 587, pp. 589–298, 2002.
by microwave drilling were substantially smoother than those [5] S. Chiba, K. Okada, K. Lee, G. V. Segre, and R. M. Neer, “Molecular
produced mechanically, and no partially attached bone frag- analysis of defect healing in rat diaphyseal bone,” J. Veterinary Med.
ments were observed under microscopy (optical or SEM) in mi- Sci., vol. 603, pp. 608–63, 2001.
[6] M. J. Park, M. C. Lee, and S. C. Seong, “A comparative study of the
crowave-drilled holes. healing of tendon autograft and tendon-bone autograft using patellar
We assume that carbonized bone in vitro represents the min- tendon in rabbits,” Int. Orthop., vol. 35, pp. 39–25, 2001.
imal volume of damaged bone tissue around the microwave [7] S. R. Davidson and D. F. James, “Drilling in bone: Modeling heat gen-
eration and temperature distribution,” J. Biomech. Eng., vol. 305, pp.
drilling site. We cannot rule out further tissue damage in non- 314–125, 2003.
carbonized regions in vivo in bones drilled with microwave, e.g., [8] J. F. Kay, L. Gilman, and T. C. May, “The tri-spade drill for endosseous
due to thermal and microwave radiation effects, and this should dental implant installation,” J. Oral Implantol., vol. 424, pp. 8–17,
1991.
be quantified in in vivo (animal) studies. However, it may be [9] S. B. Goodman, M. Lind, Y. Song, and R. L. Smith, “In-vitro, in-vivo,
possible to reduce the carbonization margins further before pro- and tissue retrieval studies on particulate debris,” Clin Orthop., vol. 25,
ceeding to in vivo studies, using adaptive control on the mi- pp. 34–352, 1998.
[10] M. B. Abouzgia and D. F. James, “Measurements of shaft speed
crowave power, by a liquid or air cooling mechanism. Basic while drilling through bone,” J. Oral Maxillofac. Surg., vol. 1308, pp.
studies on thermal properties and thermal tolerance of bone 1315–53, 1995.
tissue under microwave radiation can also be beneficial in order [11] A. R. Eriksson and T. Alberktsson, “Temperature threshold levels
for heat-induced bone tissue injury: A vital-microscopic study in the
to minimize the carbonized margins of the hole. rabbit,” J. Prosthet. Dent., vol. 101, pp. 107–50, 1983.
Some limitations in this study should be recognized. First, in [12] L. S. Nichter, S. Richlin, P. M. Navarrette, and K. Kosari, “The biome-
vitro bone specimens, rather than living bones, were used for the chanical efficacy of an oscillating K-wire driver,” Ann. Plast. Surg., vol.
289, pp. 292–29, 1992.
rotary and microwave drilling tests. Although bone specimens [13] G. S. Gazelle, S. N. Goldberg, L. Solbiati, and T. Livraghi, “Tumor ab-
were obtained fresh and kept moist with a saline spray during lation with radio-frequency energy,” Radiology, vol. 633, pp. 646–217,
testing to closely represent the living bone, the important in vivo 2000.
[14] A. Copty, F. Sakran, M. Golosovsky, M. D. Davidov, and A. Frenkel,
effects such as viability of osteocytes, tissue necrosis, and hy- “Low-power near-field microwave applicator for localized heating of
peremia could not be analyzed at this stage and are awaiting soft matter,” Appl. Phys. Lett., vol. 5109, pp. 5111–84, 2004.
further studies. The interaction of microwave heating with fluid [15] M. Mashevich, D. Folkman, A. Kesar, A. Barbul, R. Korenstein, E.
Jerby, and L. Avivi, “Exposure of human peripheral blood lymphocytes
flow in bone vessels is similarly not considered in vitro. Second, to electromagnetic fields associated with cellular phones leads to chro-
a drill press was used during the rotary drilling experiments mosomal instability,” Bioelectromagnetics, vol. 82, pp. 90–24, 2003.
rather than a hand-held orthopaedic surgical drill. To replicate [16] I. Ghanem, L. M. Collet, K. Kharrat, E. Samaha, H. Deramon, P. Mertl,
and F. Dagher, “Percutaneous radiofrequency coagulation of osteoid
clinical conditions as closely as possible, a drill speed of 20 000 osteoma in children and adolescents,” J. Pediatr. Orthop. B., vol. 244,
rpm, representative of actual orthopaedic drills was used for all pp. 252–12, 2003.

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.
 1182 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 53, NO. 6, JUNE 2006

[17] M. R. Callstrom, “Painful metastases involving bone: Feasibility of per- Abby Anaton was born in Istanbul, Turkey, in 1977.
cutaneous CT- and US- guided radio-frequency ablation,” Radiology, He received the B.Sc. degree in electrical engineering
vol. 87, pp. 97–224, 2002. from Tel Aviv University, Tel-Aviv, Israel, in 2003.
[18] E. Jerby, O. Aktushev, and V. Dikhtyar, “Theoretical analysis of the mi- He is currently working toward the M.Sc. degree.
crowave-drill near-field localized heating effect,” (in 034 909) J. Appl. His thesis concerns microwave heating of concrete.
Phys., vol. 97, 2005. His areas of interest are microwave system engi-
[19] N. Passi and A. Gefen, “Trabecular bone contributes to strength of the neering and computational electromagnetics.
proximal femur under mediolateral impact in the avian,” J. Biomech
Eng., vol. 198, pp. 203–127, 2005.
[20] J. C. Goh, E. J. Ang, and K. Bose, “Effect of preservation medium on
the mechanical properties of cat bones,” Acta. Orthop. Scand., vol. 465,
pp. 467–60, 1989.
[21] F. Linde and H. C. Sorensen, “The effect of different storage methods
on the mechanical properties of trabecular bone,” J. Biomech., vol.
1249, pp. 1252–26, 1993. Tomer Yacoby was born in Israel in 1978. He re-
[22] J. M. Gere and S. P. Timoshenko, Mechanics of materials. Boston, ceived the B.Sc. degree in electrical engineering from
MA: PWS-KENT, 1972. Tel Aviv University, Tel-Aviv, Israel, in 2003. He is
currently working toward the M.Sc. degree, special-
izing in electromagnetics.
His research areas are microwave heating and
processing of materials, and microwave system
engineering.

Amit Gefen (S’96–M’01) was born in Ramat-Gan,

Israel, in 1971. He received the B.Sc. degree in
mechanical engineering and the M.Sc. and Ph.D.
Yael Eshet was born in Tel-Aviv, Israel, in 1971. She degrees in biomedical engineering from Tel Aviv
received the M.D., degree in 1996, the B.Sc. degree University, Tel Aviv, Israel, in 1994, 1997, and 2001,
in electrical engineering in 2005, the B.Sc. degree in respectively.
physics in 2005, and the M.Sc. degree in biomed- During 2002–2003, he was a Postdoctoral Fellow
ical engineering in 2004, all from Tel-Aviv Univer- at the Injury Biomechanics Laboratory of the Bio-
sity, Tel-Aviv, Israel. engineering Department at the University of Pennsyl-
She is currently a second-year Medical Resident vania, Philadelphia. He is currently a Senior Lecturer
with the Department of Imaging, Sheba Medical with the Department of Biomedical Engineering, in
Centre, , Tel-Hashomer, Israel. Her residency com- the Faculty of Engineering of Tel Aviv University. His research interests and ex-
bines active work as a physician and research work. perience are in studying normal and pathological effects of mechanical factors
She is also participating in her free time in a research on the structure and function of tissues, with emphasis on the musculoskeletal
group in the Department of Biomedical Engineering and Physical Electronics, system.
Faculty of Engineering, Tel-Aviv University, Israel.

Eli Jerby (M’04) received the M.Sc. and Ph.D. de-

Ronit Rachel Mann was born in Ramat-Gan, Israel, grees in electrical engineering, from Tel-Aviv Uni-
in 1976. She received the B.Sc. degree in mechan- versity, Tel-Aviv, Israel, in 1980 and 1989, respec-
ical engineering from the Technion, Haifa, Israel, in tively.
2003. She is currently working toward the M.Sc. de- He worked at the Massachusetts Institute of Tech-
gree, specializing in biomedical engineering. Her re- nology with the late Prof. G. Bekefi on free-electron
search is microwave drilling in bone and in particular lasers and cyclotron-resonance masers. Since his re-
and the effects of microwave drilling on bone mor- turn to Tel Aviv University in 1991, he has developed
phology and mechanical properties. and studied several microwave sources and applica-
tions, including the microwave-drill invention and its
experimental, theoretical, and practical aspects.

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY ROORKEE. Downloaded on May 28,2021 at 10:37:38 UTC from IEEE Xplore. Restrictions apply.