Exploring thermal anisotropy of cortical bone using temperature measurements in drilling

Abstract

Background:

Bone drilling is widely used in orthopaedics for fracture treatment, reconstructive surgery and bone biopsy. Heat generation in bone drilling can cause rise in bone temperature resulting in prolonged healing time or loosening of fixation.

Objective:

The purpose of this study was to investigate thermal anisotropy of bone by measuring the level of temperature in bone drilling with and without cooling conditions in two anatomical directions.

Methods:

Drilling tests were performed on bovine cortical bone. A total of fifteen specimens were used to obtain data for statistical analysis. Temperature near the cutting zone was measured in two anatomical directions. i.e. along the longitudinal and circumferential direction. Temperature distribution was also found in the two prescribed directions. Analysis of variance (ANOVA) was used to identify significant drilling parameter affecting bone temperature.

Results:

Drilling speed, feed rate and drill size were found influential parameters affecting bone temperature. Higher drilling speed, feed rate, and large drill size were found to cause elevated temperature in bone. Much lower temperature was measured in bone when cooling fluid was supplied to the drilling region. Experimental results revealed lower temperatures in the circumferential direction compared to the longitudinal direction.

Conclusions:

Thermal anisotropy for heat transport was found in the bone. This study recommends lower drilling speed and feed rate and cooling for controlling rise in bone temperature.

Keywords

Orthopaedic bone drilling necrosis thermocouple experimental analysis

1. Introduction

Bone drilling is one of the most common surgical procedures widely performed in orthopaedics, dental and neural surgeries. Osteotomies, performed by rotational instruments such as a drill, can cause temperature rise on the bone causing death of the cells near the cut surface. Inactive or dead cells prolong the healing process and weaken the strength of the bond between fixative devices and the bone postoperatively. The heat generation during bone drilling is caused largely by the shear deformations and the friction between the drill and the bone [1]. The thermal threshold level for necrosis of the bone tissue is 47°C for a minute which can be easily achieved by rotational cutters such as drill or burs [2]. As a consequence of elevated temperature, bone is resorbed and replaced with fat cells [2]. Bone temperatures exceeding a critical threshold level causes a denaturation process which seriously affects the mechanical properties of the bone [3]. Higher temperatures in drilling the bone may also cause weakening of the bond between the bone and implant [4,5]. Such condition may lead to repeat of the surgical procedures and put more financial burden on healthcare units. Most orthopaedic and dental surgeon used sterile saline with lower temperature to protect the bone from the thermal injuries in surgical procedures.

Measurement and control of temperature in bone drilling are critical to the success of orthopaedic and dental operations. Several factors such as drilling force, drill speed, feed rate, drill geometry, drill size, drill condition, drilling depth, cooling condition and type of drilling influence heat generation in bone [6–12]. Several studies have been performed which used either internal or external cooling to keep bone temperature well below the thermal threshold level [6,8,13–17]. In those studies, the use of coolant in drilling was found the most important factor to control bone temperature. Some recent studies predicted temperature in bone drilling and plane cutting in the presence of cooling using finite element (FE) analysis [12,18]. Despite mechanical and thermal anisotropy of the bone [19,20] and different thermal conductivity, specific heat of fresh and dry bone [21], previous studies either in dry drilling or using coolant, did not measure the level of temperature and its distribution along major anatomical directions. i.e. along the longitudinal axis of the bone and transverse (circumferential) direction.

The present paper studied the measurement of temperature in two different directions and evaluated the difference between them using either dry drilling or with coolant at reduced temperature using thermocouples. The influence of drill speed, feed rate and drill size on bone temperature near the cutting edge of the drill was found and comparison between dry and wet drilling is performed. Temperature distribution in bone tissue was also found in the longitudinal and circumferential direction for dry and wet drilling using two sizes of drills. Critical drilling parameters responsible for both types of drilling conditions were identified and discussed.

2. Materials and methods

2.1. Specimen preparation

Fresh bovine femoral bone was selected as a candidate material as it replicate the properties of human bone [22] and is widely been used in research studies related to bone drilling [23–25]. The middle section was excised from the femoral bone with a hand saw. A total of fifteen specimens were obtained from femoral shafts each could accommodate around twenty holes. The final shape of the specimen was approximately cylindrical of 50 mm length. The top surface of the bone was cleaned for drilling small holes for thermocouple insertion. The first thermocouple in both anatomical directions was placed at a distance of 2 mm from to the cutting edge of the drill. Five calibrated thermocouples were placed in each direction to obtain a meaningful distribution of temperature. Each thermocouple was placed 2 mm apart from each other. Bone has very low thermal conductivity compare to other engineering materials. A distance of 2 mm was found sufficient for obtaining smooth profile of bone temperature. In addition, crushing of thermocouple bead adjacent to the drill (placed at a distance of 2 mm) was not seen during the drill penetration into the bone. The holes drilled for thermocouples were filled with thermal paste to eliminate the air gap between the bead of the thermocouple and the bone. Each thermocouple bead was kept at a depth approximately half of the wall thickness of the specimen (4 to 5 mm from the top surface). The depth was sufficient enough to record the temperature data since at that depth; the cutting edges of the drill were fully engaged in cutting. The cavities around the thermocouples were covered with water resistant glue to avoid direct contact of thermocouple beads with normal saline. The specimens were kept at room temperature for ten minutes so that the glue becomes hard and dry. Specimen with thermocouple in both directions is shown in Fig. 1.

Fig. 1.

(a) Bone specimen for drilling, (b) schematic of thermocouple locations. (X-longitudinal direction, Y-circumferential direction.)

2.2. Equipment and procedure for drilling

A vertical drilling machine was used in the experiments. The maximum spindle speed and feed rate of the machine was 10,000 rpm and 10 m/min respectively. Experimental set up for drilling is shown in Fig. 2. Thermocouples having 1 mm wire diameter were used for temperature measurement. The estimated uncertainty in the temperature measurement system was 0.5°C. The speed and feed rate, with which a drill is penetrated into the bone tissue, varies from surgeon to surgeon. Selection of drilling parameters was based on those reported in previous research related to bone drilling [11,12,24,25]. Drilling was performed perpendicular to the longitudinal axis of the bone. Two drill sizes, 5 mm and 3 mm were used due to their frequent use in drilling procedures in orthopaedics. Normal saline with a temperature of 10°C was used as a coolant medium. The saline was supplied to the drilling site using a plastic hose of 3 mm diameter with volume flow rate of 0.4 liter per minute. Analysis of variance (ANOVA) was used to identify significant drilling parameter affecting bone temperature.

Fig. 2.

Experimental set up for temperature measurement in bone drilling.

3. Results

Temperature reading from each thermocouple was observed to rise quickly as soon as the cutting edge of the drill approached the location where the thermocouple bead was placed. The temperature value then stayed at a maximum constant value till the drill exit the cortex of the cortical bone. All the data points in the subsequent plots were recorded at the peak values. Each experiment was repeated three times for a particular set of drilling parameters. Error bars in the subsequent plots graphically represent variability or uncertainty in the temperature data. The bars provided a general idea of how the temperature data is far from the mean value in a particular set of experiment. The terminologies “dry” and “wet” are also used in the text representing cooling and no cooling condition respectively. The effect of drilling speed, feed rate, drill size on bone temperature and its distribution in two anatomical directions is discussed below.

3.1. Effect of drilling speed on bone temperature

The rise in bone temperature was measured by varying the drill speed from 500 rpm to 3000 rpm. The feed rate was kept constant at 50 mm/min to see the effect of drill speed on bone temperature alone. The rise in bone temperature with and without cooling is shown in Fig. 3. A linear relationship between the drill speed and bone temperature was found in both anatomical directions. Using a drill size of 5 mm, the bone temperature was observed to rise from a mean value of 66°C to 93°C when the drilling speed was changed from 500 rpm to 3000 rpm in longitudinal direction without cooling. Similarly, an increase of 30°C was noted for the same drilling parameters when cooling was used. Surprisingly, the bone temperature was dropped from 93°C to 37°C in the longitudinal direction when the drilling condition was changed from dry to wet using 5 mm drill and the highest drilling speed used in this study.

Fig. 3.

Variation of bone temperature with drill speed (drill size – 5 mm, feed rate – 50 mm/min); (a) longitudinal direction; (b) circumferential direction.

Fig. 4.

Variation of bone temperature with drill speed (drill size – 3 mm, feed rate – 50 mm/min); (a) longitudinal direction; (b) circumferential direction.

Using a drill size of 3 mm, the temperature was observed to increase from a mean value of 56°C to 83°C when the drilling speed was changed from 500 rpm to 3000 rpm in the longitudinal direction without cooling. Similarly, with highest drilling speed, temperature was dropped to 20°C when cooling was used. The relationship between the temperatures and drilling speed in both types of drilling in directions using a 3 mm drill is shown in Fig. 4. The rise in bone temperature with increase in drilling speed was due to the increase in rate of material deformation, which was proportional to the heat generation, and increased friction between the drill and the bone.

3.2. Effect of feed rate on bone temperature

The effect of feed rate on bone temperature with and without cooling conditions in both directions and two drill sizes was also studied. The drilling speed was kept constant at 2500 rpm to find the effect of feed rate alone. The trend with which the temperature rose with increase in feed rate in the longitudinal direction was similar to that noted in the circumferential direction, however, lower temperature were measured for similar drilling parameters in the later. The variation of bone temperature with feed rate in the longitudinal direction and circumferential direction is shown in Figs 5 and 6, respectively. It was found that increasing feed rate, there was an upward trend of bone temperature. Using the maximum value of the feed rate and 2500 rpm drill speed, the bone temperature was to drop by 70% in the both directions when cooling was used.

Fig. 5.

Variation of bone temperature with feed rate (drill size – 5 mm, drill speed – 2500 rpm); (a) longitudinal direction; (b) circumferential direction.

Fig. 6.

Variation of bone temperature with feed rate (drill size – 3 mm, drill speed – 2500 rpm); (a) longitudinal direction; (b) circumferential direction.

3.3. Temperature distribution in bone

In this section, the distribution of temperature in bone material has been studied in both anatomical directions. A 5 mm drill size with the highest values of drilling speed and feed rates (3000 rpm and 70 mm/min) was used to see the temperature distribution in bone. Temperature distribution in both anatomical directions is shown in Fig. 7. Due to the low heat transport capability of the bone tissue, the temperature was seen to drop to a value which was almost the temperature of saline when cooling was used.

Fig. 7.

Temperature distribution in bone around the drill with and without cooling condition (drill size – 5 mm, drill speed – 3000 rpm, feed rate – 70 mm/min); (a) longitudinal direction; (b) circumferential direction.

3.4. Statistical analysis

Analysis of variance (ANOVA) and regression results for bone temperature distribution against drill speed, drill direction (circumferential or longitudinal), and feed rate are presented below for both 5 mm and 3 mm drill diameters. ANOVA results for 3 mm drill are not shown due to space limitations. Both F-values and P-values (Table 1) indicate that there is significant influence of drill speed, drill direction, and feed rate on variation of bone temperature. It is also clear that effect of drill speed and feed rate on bone temperature is more significant than that of drill direction. Very high goodness of fit ( $R^{2}$ values of more than 99% in both tables) supports earlier conclusions from experimental results, that variation of bone temperature against drill speed and feed rate is almost linear. As expected, bone temperature increases with increasing drill or feed speeds (Fig. 7). For drill speed, rate of increase of bone temperature remains almost constant throughout, while it drops down a little for higher feed rates (though still showing an increasing trend). Bone temperature is lower in the circumferential direction than in the longitudinal one, for both 5 mm and 3 mm drills. Variation of bone temperature against drill speed, drilling direction, and feed rate for 5 mm (top) and 3 mm (bottom) drill diameters is shown in Fig. 8.

Table 1
Analysis of variance (ANOVA) for bone temperature against drill speed and drill direction (a); and against feed rate and drill direction (b); 5 mm drill diameter

(a)

Source DF Seq SS Adj SS Adj MS F P

Speed 5 858.67 858.67 171.73 99.08 0.000

Direction 1 120.33 120.33 120.33 69.42 0.000

Error 5 8.67 8.67 1.73

Total 11 987.67

S = 1.31656; R-Sq = 99.12%; R-Sq(adj) = 98.07%

(b)

Source DF Seq SS Adj SS Adj MS F P

Feed 4 900.00 900.00 225.00 140.62 0.000

Direction 1 211.60 211.60 211.60 132.25 0.000

Error 4 6.40 6.40 1.60

Total 9 1118.00

S = 1.26491; R-Sq = 99.43%; R-Sq(adj) = 98.71%

Fig. 8.

Main effects plot of bone temperature for 5 mm drill (top) and 3 mm drill (bottom).

4. Discussions

The amount of heat generation in a cutting process is largely dependent on the shearing and frictional forces. The amount of heat generated in bone significantly increased with the number of revolutions and the speed of penetration (feed rate) of the drill. The rate of deformation was increased with drilling speed which caused heat flux generated at higher rate. Cooling in drilling region, directly lowered bone temperature by convection and lubrication, which altered frictional condition between the drill and the bone. A possible factor in rise in bone temperature with increase in drill speed was the applied force [26]. The increase in bone temperature with drilling speed as observed in this study was consistent with previous research [8,27]. A similar behaviour of temperature rise was reported in drilling human and bovine cortical bone [25] with drilling speeds exceeding 1200 rpm.

The results of temperature measurements in our experiments contradicted those reported in [28]. The tests were conducted at variable speeds (20000–100000 rpm) which were much higher than those used in this study. The drill was also subjected to a vertical force (1.5–9.0 N) in the drilling direction. The temperature rise and the duration of temperature elevation were found to decrease with drill speed. The results obtained from our experiments also contradicted those measured in conventional drilling of bone [3], where decrease in temperature with an increase in feed rate was found. In that study, increase of feed rate from 1 to 3 mm/s, significantly decreased bone temperature. A drill size of 6.2 mm was used to drill equine cortical bone with significantly lower drilling speed (317 to 1242 rpm) compared to those used in this study. In addition, the thermocouples were placed at a distance of 1 mm, 1.5 mm and 2 mm from the drilling track. Obvious reason discrepancies the current study and previous research was the use of different experimental arrangements and bone types for acquiring thermal data in bone drilling.

The results obtained in this study were similar to those reported before which simulated bone drilling process using numerical simulations [12]. In that study, drilling speed, feed rate and cooling environment were found influential factors affecting bone temperature. Bone temperature was found to increase with increase in drill speed and feed rate with and without cooling. It was concluded that higher drilling speeds and feed rates may be used without inducing thermal necrosis in bone when saline is used as a coolant medium. Similar results were also reported in two-dimensional FE model of bone cutting [18]. The bone temperature, measured with infrared thermography in ultrasonically assisted drilling, was also found to increase with drill speed, feed rate and frequency above 20 kHz imposed on the drill [11]. Interestingly, lower temperatures were measured in bone in vibrational drilling, with frequency below 20 kHz imposed on the drill. In another study [8], external irrigation (water at 26°C) was attributed as the single most important factor in decreasing the increase in bone temperature during drilling.

The result of this study confirmed that thermal heat transport is more in the longitudinal direction than transverse direction since thermal conductivity was found more in that direction [20]. At microstructural level, bone is similar to a fiber reinforced composite. The fibers known as osteons run predominantly along the longitudinal direction of femur in inhomogeneous interstitial bone (matrix). The osteons are separated from the matrix by a thin interface called cement line, which carries different physical properties than osteons and interstitial bone matrix [29,30]. The thermal conductivity of bone constituents is unknown and this study assume heat transport more efficient along the osteons direction than in the direction transverse to them (circumferential direction). Lower heat transport capability in the circumferential direction may be attributed to the barrier offer by the layers of several cement lines accumulated in 2 mm length. In the current study, cooling was identified as an influential factor for reducing bone temperature well below the thermal threshold level causing necrosis.

The temperature was measured using thermocouple placed in bone having lower thermal conductivity. More experiments are required to measure the temperature in bone either by attaching the bead of thermocouple to the tip of the drill or using a non contact method such as infrared thermography. Further study is required to explore the effect of drilling depth on bone temperature in the presence of irrigation. Such measurements can be obtained by inserting thermocouples along the drilling track. In this study, drilling tests are performed using hard cortical bone as a test specimen. Further research is suggested to investigate the thermal anisotropy of bone by conducting similar drilling experiments in spongy (trabecular) bone.

5. Conclusions

Drilling experiments were conducted to investigate thermal response of bone in two anatomical directions. Lower drilling speed, feed rate and small drill size were identified parameters for inducing lower temperature in bone tissue. The study concluded that with all combinations of parameters, the use of cooling produced bone temperatures well below the threshold for thermal osteonecrosis in longitudinal and circumferential directions. Lower temperature in the circumferential direction was expected due to lower thermal conductivity in that direction. This study provided new information on the measurement and controlling of bone temperature in two different directions which may help orthopedic surgeon and technicians understanding thermal anisotropy of bone.

Footnotes

Acknowledgements

The author wishes to thank Engr. Abdullah Al Omarani and Mr. Halwalage Induka Nilupul for providing support in specimen preparation and drilling experiments.

Conflict of interest

The author has no conflict of interest to report.

References

Lee

Rabin

and Ozdoganlar

O.B.

, A new thermal model for bone drilling with applications to orthopaedic surgery, Med. Eng. Phys. 33 (2011), 1234–1244. doi:10.1016/j.medengphy.2011.05.014.

Eriksson

A.R.

and Albrektsson

, Temperature threshold levels for heat induced bone tissue injury: A vital-microscopic study in the rabbit, J. Prosthet. Dent. 50(1) (1983), 101–107. doi:10.1016/0022-3913(83)90174-9.

Toews

A.R.

Bailey

J.V.

Townsend

H.G.G.

and Barber

S.M.

, Effect of feed rate and drill speed on temperatures in equine cortical bone, Am. J. Vet. Res. 60 (1999), 942–944.

Pallan

F.G.

, Histological changes in bone after insertion of skeletal fixation pins, J. Oral. Surg. Anest. Hosp. Dent. Serv. 18 (1960), 400–408.

Christie

, Surgical heat injury of bone, Injury 13 (1981), 188–190. doi:10.1016/0020-1383(81)90236-9.

Augustin

Zigman

Davila

Udiljak

Staroveski

Brezak

et al., Cortical bone drilling and thermal osteonecrosis, Clin. Biomech. 27 (2012), 313–325. doi:10.1016/j.clinbiomech.2011.10.010.

Karmani

, The thermal properties of bone and the effects of surgical intervention, Curr. Orthop. 20 (2006), 52–58. doi:10.1016/j.cuor.2005.09.011.

Augustin

Davila

Mihoci

Udiljak

Vedrina

D.S.

and Antabak

, Thermal osteonecrosis and bone drilling parameters revisited, Arch. Orthop. Trauma Surg. 128 (2008), 71–77. doi:10.1007/s00402-007-0427-3.

Allan

Williams

E.D.

Kerawala

C.J.

and López

L.N.

, Effects of repeated drill use on temperature of bone during preparation for osteosynthesis self-tapping screws, Br. J. Oral. Maxillofac. Surg. 43 (2005), 314–319. doi:10.1016/j.bjoms.2004.11.007.

10.

Sezek

Aksakal

and Karaca

, Influence of drill parameters on bone temperature and necrosis: A FEM modelling and in vitro experiments, Comput. Mater. Sci. 60 (2012), 13–18. doi:10.1016/j.commatsci.2012.03.012.

11.

Alam

and Silberschmidt

V.V.

, Analysis of temperature in conventional and ultrasonically-assisted drilling of cortical bone with infrared thermography, Technol. Health Care 22 (2014), 243–252. doi:10.3233/THC-140813.

12.

Alam

Khan

and Silberschmidt

V.V.

, 3D finite element modeling of drilling cortical bone: Temperature analysis, J. Med. Biol. Eng. 34(6) (2014), 618–623. doi:10.5405/jmbe.1585.

13.

Matthews

L.S.

and Hirsch

, Temperatures measured in human cortical bone when drilling, J. Bone Joint Surg. Am. 54 (1972), 297–308.

14.

Lavelle

and Wedgwood

, Effect of internal irrigation on frictional heat generated from bone drilling, J. Oral. Surg. 38 (1980), 499–503.

15.

Haider

Watzek

and Plenk

, Effect of drill cooling and bone structure on IMZ implant fixation, Int. J. Oral. Maxillofac. Implants 8 (1993), 83–91.

16.

Benington

I.C.

Biagioni

P.A.

Briggs

Shearidan

and Lamey

P.J.

, Thermal changes observed at implant sites during internal and external irrigation, Clin. Oral. Implants Res. 13(3) (2002) 293–297. doi:10.1034/j.1600-0501.2002.130309.

17.

Cardioli

and Majzoub

, Heat generation during implant site preparation: An in vitro study, J. Oral. Maxillofac. Implants 12 (1997) 186–193.

18.

Alam

Mitrofanov

A.V.

and Silberschmidt

V.V.

, Thermal analysis of orthogonal cutting of cortical bone using finite element simulations, Int. J. Exp. Comput. Biomech. 1(3) (2010), 236–251. doi:10.1504/IJECB.2010.035259.

19.

Demirci

and Silberschmidt

V.V.

, Variability and anisotropy of mechanical behavior of cortical bone in tension and compression, J. Mech. Behav. Biomed. Mater. 21 (2013), 109–120. doi:10.1016/j.jmbbm.2013.02.021.

20.

Davidson

S.R.

and James

D.F.

, Measurement of thermal conductivity of bovine cortical bone, Med. Eng. Phys. 22(10) (2000), 741–747. doi:10.1016/S1350-4533(01)00003-0.

21.

Biyikli

Modest

M.F.

and Tarr

, Measurements of thermal properties for human femora, J. Biomed. Mat. Res. 20(9) (1986), 1335–1345. doi:10.1002/jbm.820200908.

22.

Vashishth

, Rising crack-growth-resistance behavior in cortical bone: Implications for toughness measurements, J. Biomech. 37(6) (2004), 943–946. doi:10.1016/j.jbiomech.2003.11.003.

23.

Wang

Cao

Zhao

Zhu

McClean

Zhao

and Fan

, Experimental investigations and finite element simulation of cutting heat in vibrational and conventional drilling of cortical bone, Med. Eng. Phys. 36(11) (2014), 1408–1415. doi:10.1016/j.medengphy.2014.04.007.

24.

Alam

Mitrofanov

A.V.

and Silberschmidt

V.V.

, Experimental investigations of forces and torque in conventional and ultrasonically-assisted drilling of cortical bone, Med. Eng. Phys. 33 (2011), 234–239. doi:10.1016/j.medengphy.2010.10.003.

25.

Hillery

M.T.

and Shuaib

, Temperature effects in the drilling of human and bovine bone, J. Mater. Process Tech. 92–93 (1999), 302–308. doi:10.1016/S0924-0136(99)00155-7.

26.

Abouzgia

M.B.

and James

D.F.

, Measurement of shaft speed while drilling through bone, Int. J. Oral. Maxillofac. Implants 53 (1995), 1308–1315. doi:10.1016/0278-2391(95)90590-1.

27.

Matthews

L.S.

Green

C.A.

and Goldstein

S.A.

, The thermal effects of skeletal fixation pin insertion in bone, J. Bone Joint. Surg. 66 (1984), 1077–1083.

28.

Abouzgia

M.B.

and Symington

J.M.

, Effect of drill speed on bone temperature, Int. J. Oral. Maxillofac. Surg. 25 (1996), 394–399. doi:10.1016/S0901-5027(06)80040-8.

29.

Weiner

Traub

and Wagner

H.D.

, Lamellar bone: Structure-function relations, J. Struc. Biol. 126(3) (1999), 241–255. doi:10.1006/jsbi.1999.4107.

30.

Pidaparti

R.M.V.

Chandran

Takano

and Turner

C.H.

, Bone mineral lies mainly outside collagen fibrils: Predictions of a composite model for osteonal bone, J. Biomech. 29(7) (1996), 909–916. doi:10.1016/0021-9290(95)00147-6.

(a)
Source	DF	Seq SS	Adj SS	Adj MS	F	P
Speed	5	858.67	858.67	171.73	99.08	0.000
Direction	1	120.33	120.33	120.33	69.42	0.000
Error	5	8.67	8.67	1.73
Total	11	987.67
S = 1.31656; R-Sq = 99.12%; R-Sq(adj) = 98.07%
(b)
Source	DF	Seq SS	Adj SS	Adj MS	F	P
Feed	4	900.00	900.00	225.00	140.62	0.000
Direction	1	211.60	211.60	211.60	132.25	0.000
Error	4	6.40	6.40	1.60
Total	9	1118.00
S = 1.26491; R-Sq = 99.43%; R-Sq(adj) = 98.71%