analysis of bone healing with a novel bone wax substitute … · 4 67 since bone wax is comprised...
TRANSCRIPT
1
Analysis of Bone Healing with a Novel Bone Wax Substitute Compared to Bone Wax in a 1
Porcine Bone Defect Model 2
3
Tristan Tham, MD1; Keith Roberts
2; John Shanahan, PhD
2; John Burban, PhD
2; Peter 4
Costantino, MD FACS1
5 6
1 - New York Head & Neck Institute, Lenox Hill Hospital, New York, USA 7
2 – Hemostasis LLC 8
9 10
Lenox Hill Hospital 11 (Black Hall Building) 12
130 East 77th Street 13 10th Floor 14
New York, NY 10075 15
16 Corresponding author: 17 Tristan Tham, MD – [email protected] 18
19
Senior author: 20 Peter Costantino, MD – [email protected] 21
22 Co-authors: 23 Keith Roberts – [email protected] 24
John Shanahan – [email protected] 25 John Burban – [email protected] 26
27
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
2
Abstract: 250 words 28
Background: Bone wax is used in surgery as a hemostatic device for bone. Despite its good 29
functional capacity as a bone hemostat, Bone wax materials often have very poor long-term 30
interactions with bone. This study describes a novel composite of hydroxyapatite (HA) and 31
biodegradable poly-lactic acid (PLA) with wax-like handling properties (OsteoStat). The goal 32
was to compare qualitative and quantitative measures between OsteoStat versus Bone wax. 33
Methods: The porcine critical size defect model was chosen in this study. OsteoStat and Bone 34
wax were introduced into separate critical size defects located in the femur and humerus of a 35
single porcine specimen. After a duration of 6 weeks, the defect sites were harvested for clinical, 36
histological, and histomorphometric analysis. 37
Results: Both groups had effective hemostatic action when introduced into the defects. Analysis 38
of the histomorphometric data revealed that the amount of new bone was significantly greater at 39
6 weeks in the OsteoStat group (38.05%) versus the Bone wax group (11.88%), p=0.028. 40
OsteoStat also demonstrated less soft tissue and less test material remaining in the defect sites; 41
however, this was not statistically significant. 42
Conclusions: We speculate that the incomplete biodegradation of Bone wax as well as its 43
intrinsic inflammatory properties may have retarded osseous regeneration and promoted fibrosis. 44
In contrast, well known biodegradation pathways for PLA combined with the HA component of 45
OsteoStat may have accounted for the positive results of OsteoStat compared to Bone wax. It is 46
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
3
important that bone hemostat substances have biocompatible, osteoconductive, hemostatic, as 47
well as good handling properties. 48
49
Keywords 50
Biomaterials, hydroxyapatite, biocompatibility, osteoconduction, bone wax, porcine model, 51
hemostasis 52
53
Background 54
Bone wax, first described by Sir Victor Horsley as an antiseptic in 1892, is now 55
commonly used in surgery as a hemostatic device.1,2
In cardiothoracic surgery, the main 56
indication for Bone wax is extensive bleeding from the sternal bone marrow. During sternotomy 57
procedures, intraoperative exposure of the cut surfaces of the sternum, together with concurrent 58
anticoagulant therapies increases the patients’ risk of bleeding into the sternal wound. Surgeons 59
try to minimize such bleeding as the eventual intrasternal hematoma collection could serve as a 60
nidus for bacterial infection, an uncommon but serious sequalae of cardiothoracic surgery.3 In 61
order to prevent extensive hematoma formation, surgeons apply topical hemostatic agents, such 62
as Bone wax, to the cut sternal marrow.4 Surgical Bone wax consists of a naturally occurring 63
substance, sterilized Cera Alba (honeybees wax), and is usually mixed with paraffin wax which 64
acts as a softening agent to enhance handling characteristics. 65
66
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
4
Since Bone wax is comprised of paraffin wax and esterified fatty acids, it is highly 67
hydrophobic. This enables Bone wax to serve as a physical barrier against blood, occluding the 68
bleeding channels and achieving hemostasis by a tamponade and blood stasis effect. However, 69
this hydrophobic property in conjunction with limited enzymatic degradation of waxes in the 70
human body prevents appreciable rates of absorption and/or excretion post-surgical application.. 71
It is well documented that Bone wax impairs optimal bone formation and healing of 72
sternotomies, 3,5
which could be due to, in part, physical inhibition of osteoblast and osteocyte 73
migration to the site of bony injury. Furthermore, the nature of Bone wax which makes it highly 74
resistant to degradation has also been linked with infection,6,7
although large randomized studies 75
have found the infection link to be inconclusive.8 Since intraoperative bone bleeding can be 76
heavy, physicians must weigh the benefits of bone hemostasis using Bone wax versus the risk of 77
decreased bone healing and other complications such as infection. 78
79
It is important that Bone wax like substances have biocompatible, osteoconductive, 80
hemostatic, as well as good handling properties. Several alternative materials have been reported 81
in the literature such as PEG/collagen,9 polyorthoester,
10 fibrin-collagen,
11 chitin-based 82
material,12
, and gelfoam.13
However, none of these alternative materials have yet seen 83
widespread adoption, suggesting that a material which meets the effective hemostatic qualities of 84
Bone wax together with good osseous integration and affordability has not been met. 85
86
Herein we describe OsteoStat (Hemostasis LLC, MN, USA), a novel composite of 87
hydroxyapatite (HA) and biodegradable poly-lactic acid (PLA) with wax-like handling 88
properties. HA (Ca10(PO4)6(OH)2) is a biomaterial similar to the mineral component of natural 89
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
5
bone, and exhibits good osteoconductivity.14
Similar to Bone wax, the OsteoStat creates a 90
physical barrier which is its primary mechanism of hemostasis. Unlike Bone wax, we 91
hypothesized that the OsteoStat would, due to its HA content, have an enhanced bone healing 92
profile in addition to hemostatic qualities. The goal of this experiment was to compare the widely 93
used cardiothoracic surgery hemostatic agent, Bone wax, versus OsteoStat. We investigated 94
qualitative and quantitative measures of bone healing between the two materials. 95
Materials & Methods 96
Details of animal husbandry, diet, care, monitoring, health, and well-being as well as 97
measures to alleviate suffering were all performed in accordance with International Organization 98
for Standardization document ISO 10993-11:2006: Biological evaluation of medical devices. 99
This experiment was conducted in a Sichuan Province Experimental Animal Management 100
Committee Accredited Facility in China. Ethics approval for animal experimentation was 101
approved by the animal ethics board of the Sichuan Province Experimental Animal Management 102
Committee of China. 103
A porcine bone defect model was used for this study. One female young adult Landrace pig 104
weighing 52.8 kg was used for this study, with an acclimation period of 7 days. Two groups of 105
holes were drilled into the femur and humerus and filled with either OsteoStat or Bone wax. 106
After a period of 6 weeks, qualitative (clinical and histology) and quantitative 107
(histomorphometry) analyses were performed on the drilled sites. 108
Surgical Procedure 109
The animal was prepared for operation under general anesthesia. Intraperitoneal 110
pentobarbital sodium was administered and the field of operation was then sterilized and selected 111
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
6
at the right humerus and contralateral left femur. Tissue dissection was performed, exposing the 112
underlying periosteum at these respective sites. Four test sites were chosen for the bone defect, 2 113
holes located in the right diaphyseal humerus and 2 holes located in the left diaphyseal femur. 114
All test sites were drilled to have standardized intraosseous defects with a circumference of 3cm 115
and equal depth. Care was taken to drill the cortical bone down to a similar depth in all the drill 116
sites. One defect on each bone (humerus or femur) was filled with OsteoStat (Hemostasis LLC, 117
MN, USA) and the other defect was filled with Bone wax W31G (Ethicon, NJ, USA). The 118
surgical sites were closed in multiple layers. 119
The animal was given a healing period of 6 weeks before it was sacrificed. During the 120
healing period, the animal displayed no signs of local infection and surgical site incisions were 121
well healed. The animal was fully ambulant during the entire period and was effectively load 122
bearing on all four limbs. After the healing period, the animal was sacrificed with an overdose of 123
pentobarbital sodium. The right humerus and left femur diaphysis were cut to harvest the test 124
sites as discrete blocks. On harvesting of the test sites, no signs of gross inflammation or necrosis 125
were observed in any of the sites. 126
Histology Preparation 127
The harvested blocks of tissue containing the test sites were fixed with a 10% neutral 128
buffered solution of formalin for a period of 7 days. The blocks were then decalcified by a 129
solution of mixed acid decalcification agent for 6 weeks. Next, the sections were dehydrated with 130
ethanol gradient and prepared with paraffin. Hematoxylin and Eosin (H&E) stain was applied to 131
the sections for histological observation at x40, x100, and x 400 magnification. Two sections 132
were prepared from each of the 4 drilled sites, for a total of 8 sections for 133
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
7
histological/histomorphometric analysis (4 sections for the OsteoStat defect and 4 sections for 134
the Bone wax defect). 135
Histomorphometric Evaluation 136
Two sections per drilled site were used for histomorphometric analysis, for a total of 4 137
sections per test material. Analysis was done at the x40 power magnification level, as this would 138
give a balance between tissue resolution and reduction in section variability, which would be 139
greatly biased at higher power magnification. Histomorphometric data was obtained by 140
analyzing the magnified cross sections of the defect sites using National Institute of Health 141
program software, ImageJ 1.48v. Areas of interest were measured in terms of pixel counts. 142
Parameters delineated in the histomorphometric evaluation were % bone area, % soft tissue area, 143
% test material, all of which combined would add up to approximately 100% of the defect size. 144
Statistical Analysis 145
Histomorphometric data was analyzed by comparing the mean between the two test materials 146
using Student’s t-test. Results were considered significant with a p value of less than 0.05. 147
Results 148
Clinical Evaluation 149
Upon drilling defects into the femoral and humeral diaphysis, bone bleeds were observed in 150
all the bone defects. There was effective hemostasis in both materials. The animal remained 151
healthy for the duration of the study with no post-operative or surgical site complications. The 152
rest of the 6 week healing period was uneventful. 153
Histology 154
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
8
A selection of images which are representative of the qualitative findings are included below. 155
X40 power magnification: In the OsteoStat sites, new bone formation was observed, in 156
addition to the soft tissue stroma. Furthermore, new vessel formation was observed in some of 157
the sections (Fig. 1). In the Bone wax sites, fibrous soft tissue stroma was the predominating 158
component. Some new bone can also be seen, and there is also the presence of undegraded test 159
material. The amount of new bone appears visually less than in the OsteoStat sites (Fig. 2). 160
161
Fig. 1 H&E stained view of the bony defect containing OsteoStat at x40 power 162 magnification. 163 Formation of new bone (B) can be seen on the lateral and inferior borders of the image. 164 Haversian/volkman-like vessels are present in the central portion of the image. Fibrous stroma 165 (F) is also present in the middle of the section at the interface of the newly formed bone. 166
167
Fig. 2. H&E section containing Bone wax at x40 power magnification. 168 This section is dominated by fibrous stroma (F) in the middle of the image, and a small amount 169
of new bone (B) formation at the peripheries. 170
171
X100 power magnification: In the OsteoStat sites, new bone formation can be seen which 172
make up a large portions of the images (Fig. 3). Bone wax sites show a large amount of fibrous 173
tissue and some new bone is also present, but as immature bony trabeculae (Fig. 4). 174
175
Fig. 3 Section of OsteoStat site at x100 power magnification. 176 New bone (B) can be seen on the right margins, with the left margin composed of fibrous stroma 177
(F). 178 179
Fig. 4 Section of the Bone wax site at x100 power magnification. 180
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
9
Immature bony trabeculae (B) are seen at the inferior margins. The rest of the image is mainly 181
composed of fibrous stroma (F). 182 183
X400 power magnification: OsteoStat sites shows active osteoblast activity lining the bony 184
trabeculae. Osteocytes are also present in the bony matrix. No osteoclasts are seen (Fig. 5). 185
Similarly, the Bone wax site (Fig. 6) shows osteoblast activity lining the bony trabeculae, with 186
no osteoclasts. The variation in sections at the x400 magnification is inherently greater. 187
188
Fig. 5 OsteoStat site at x400 power magnification. 189 Osteoblasts (OB) are seen lining the newly formed bony trabeculae (B), with osteocytes (OCY) 190 also present in the bony matrix. 191
192
Fig. 6 Bone wax site at x400 power magnification. 193 Similar to figure 5, osteoblasts (OB) are seen lining the newly formed bony trabeculae (B), with 194 osteocytes (OCY) also present in the bony matrix. The bottom portion of the image consists of 195
soft tissue 196 197
Histomorphometry 198
Histomorphometric analysis of the sections at x40 power magnification was performed. 199
Parameters measured were mean volume fraction of test material remaining, soft tissue area and 200
bone area. The mean area fractions of test material, soft tissue and bone in the OsteoStat 201
composite were 12.31%, 49.64%, and 38.05% respectively. In the Bone wax group, the mean 202
area fractions of test material, soft tissue and bone were 16.08%, 72.04%, 11.88% and 203
respectively (Table 1). The difference in bone area fraction between the OsteoStat group 204
(38.05%) and the Bone wax group (11.88%) was significant (p = 0.028) (Table 2). The soft 205
tissue area fraction in the OsteoStat group (49.64%) was also less than the Bone wax group 206
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
10
(72.04%); however, this result was not statistically significant (p = 0.089). Similarly, the amount 207
of test material remaining in the OsteoStat group (12.31%) was less than the Bone wax group 208
(16.08%) and was also not statistically significant (p = 0.421). 209
210
211
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
11
Table 1. Histomorphometric Analysis of OsteoStat versus Bone wax 212
213
Histomorphometric analysis of the sections at x40 power magnification. Parameters measured 214 were mean volume fraction of test material remaining, soft tissue and bone. All area fractions 215
add up to make 100% of the total volume. The difference in areas of bone (p=0.028) between the 216 two groups are statistically significant. The difference in areas of soft tissue (p=0.089) and test 217 material remaining (p=0.421) were not statistically significant. 218
219
220
38.05
11.88
49.64
72.04
12.31 16.08
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
OsteoStat Bone wax
% M
ean
Are
a
Histomorphometric Analysis
Test materialremaining
Soft tissue
Bone area
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
12
Table 2. Bone area seen in histomorphometric analysis in OsteoStat versus Bone wax 221
OsteoStat bone area (%) Bone wax bone area (%)
Sample 1 52.49 Sample 1 9.65
Sample 2 12.42 Sample 2 15.96
Sample 3 44.31 Sample 3 6.45
Sample 4 42.98 Sample 4 15.44
Mean *38.05 Mean *11.88
Standard deviation 17.6 Standard deviation 4.61
Histomorphometric analysis of 4 samples of bone defects for OsteoStat and Bone wax. The mean 222 amount of bone observed in histomorphometric analysis was significantly greater in OsteoStat 223 compared to bone wax. * p<0.05 224
Discussion 225
Histomorphometric data revealed that the amount of new bone was significantly greater at 6 226
weeks in the OsteoStat group (38.05%) versus the Bone wax group (11.88%), p=0.028. This data 227
is consistent with the published literature in that Bone wax is a potent inhibitor of bone healing 228
4,15. Histological examinations in animal models as well as human autopsies demonstrate that the 229
application of Bone wax not only prevents bone healing, but also promotes granuloma formation, 230
chronic inflammation, and fibrotic scar tissue.3,5,16
Furthermore, Bone wax is resistant to 231
degradation and remains in the implanted sites indefinitely.5,16
In a case series of 18 post-mortem 232
examinations, histologically verified Bone wax granulomas were found, in one case as long as 10 233
years after implantation.5 Because of these potential complications, good surgical practices 234
minimize the amount of Bone wax used, whenever needed. Its use is avoided altogether when 235
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
13
fusion of bone is critical for post-operative function, for example, in most orthopedic surgery 236
procedures. 237
Bone wax hemostatic activity is purely mechanical. It physically occludes the bleeding haversian 238
canals in cortical and medullary bone and activates coagulation via the stasis component of 239
Virchow’s Triad. The OsteoStat composite used in this experiment has similar handling 240
characteristics to Bone wax, and was designed to promote hemostasis via a similar tamponade-241
like mechanism. Despite that no objective measures of hemostasis were designed as part of this 242
study, both materials are subjectively reported to have equal hemostatic efficacy. 243
244
Our results showed that the amount of fibrous tissue remaining in the OsteoStat group (49.64%) 245
was less than in the Bone wax group (72.04%), though this was not statistically significant 246
(p=0.089). Furthermore, amount of test material remaining for the OsteoStat group (12.31%), 247
though less than Bone wax (16.08%), was also not significant (p=0.421). One possible 248
explanation for these findings could be that the timeframe of 6 weeks was too short for all the 249
HA particles to take part in resorption and osteointegration. As described earlier, it would be 250
reasonable to assume that the Bone wax particles would be resistant to degradation and 251
resorption. However, HA particles have been shown to have complete osseous integration over a 252
period of time, especially when blended with PLA oligomer.17
253
254
Subjectively, both OsteoStat and Bone wax had effective hemostatic action on the bleeding bone. 255
However, in this experiment, we noted that OsteoStat has the added advantage of having a higher 256
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
14
bone healing capacity compared to Bone wax. One other hemostatic Bone wax substitute that 257
does not inhibit healing is water soluble Bone wax (WSW). WSW was first reported as 258
‘pluronic-based’ wax by Wang et al in 2001, and similar to this study, demonstrated superior 259
bone healing to Bone wax.18
A subsequent study by Vestergaard et al of a randomized trial in 260
humans comparing WSW versus Bone wax found no differences in infection rates, but radiologic 261
bone measurements indicated lower levels of bone healing in Bone wax. Additionally, the study 262
surgeons commented the WSW had some drawbacks, specifically the need to reapply due to 263
dissolution of the WSW and the need to heat the WSW product before application to make it 264
more pliable to smear on trabecular bone surfaces.19
265
Whether or not the lower levels of radiologic bone healing in Bone wax compared to WSW 266
translate into real clinical effects such as sternal bone strength is a matter for debate, as 267
demonstrated in long-term animal trials. A study comparing WSW and Bone wax in porcine 268
sternotomies after a period of 6 months showed that although Bone wax had poorer histological 269
and radiological outcomes, bone mechanical properties were similar. Sternal wounds closed with 270
Bone wax were found to be weaker compared to a negative control, but no difference in sternal 271
strength was observed between Bone wax and WSW.20 272
Similar to WSW, OsteoStat has several advantages over Bone wax and other reported Bone wax 273
substitutes. OsteoStat used in this study has demonstrated a higher amount of bone growth, is 274
easily sterilizable, is biocompatible and would theoretically fully integrate in bony architecture 275
over a longer period of time.21
Furthermore, since it has no biological components, the material 276
would be immunogenic and allergen free. Physical handling characteristics which are similar to 277
Bone wax would make OsteoStat easy to use for surgeons familiar with working with traditional 278
Bone wax. 279
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
15
Further studies are needed to ascertain the efficacy of OsteoStat in the human sternotomy and/or 280
cranial/spine surgery models, particularly comparing it against Bone wax and other common 281
hemostatic agents. Other useful avenues for investigation would be longer-term studies to 282
ascertain the rate of HA integration in bone after application as a hemostat. 283
Conclusion 284
This experiment has demonstrated that at 6 weeks, porcine bone defects will have higher 285
amounts of new bone if filled with OsteoStat than with Bone wax. OsteoStat test sites also 286
demonstrated less soft tissue and test material remaining than the Bone Wax, though the results 287
for these parameters did not meet the threshold for statistical significance. This might be because 288
the time frame for eventual HA osseous integration lasts many months to years. Subjectively, 289
both OsteoStat and Bone wax have effective hemostatic properties. It is important that bone 290
hemostat substances or its substitutes have biocompatible, osteoconductive, as well as hemostatic 291
properties. 292
Bibliography 293
294
1. Horsley V. Antiseptic wax. Br Med J. 1892;1:1165 (letter). 295
2. Light R. Hemostasis in neurosurgery. Journal of Neurosurgery. 1945;2(5):414-34. 296
3. Wellisz T, Armstrong J, Cambridge J, et al. The effects of a soluble polymer and bone 297
wax on sternal healing in an animal model. Ann Thorac Surg. 2008;85:1776–80. 298
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
16
4. Achneck H, Sileshi B, Jamiolkowski R, Albala D, Shapiro M, Lawson J. A 299
comprehensive review of topical hemostatic agents: efficacy and recommendations for use. Ann 300
Surg. 2010;251:217–228. 301
5. Sudmann B, Bang G, Sudmann E. Histologically verified bone wax(beeswax) granuloma 302
after median sternotomy in 17 of 18 autopsy cases. Pathology. 2006;38:138–41. 303
6. Nelson DR, Buxton TB, Luu QN, Rissing JP. The promotional effect of bone wax on 304
experimental Staphylococcus aureus osteomyelitis. The Journal of thoracic and cardiovascular 305
surgery. 1990 Jun;99(6):977-80. 306
7. Johnson P, Fromm D. Effects of bone wax on bacterial clearance. Surgery, 89. 307
1981;89:206–209. 308
8. Prziborowski J, Hartrumpf M, Stock U. Is Bonewax Safe and Does It Help? Ann Thorac 309
Surg. 2008;85:1002–6. 310
9. Orgill D, Ehret F, Regan J, Glowacki J, Mulliken J. Polyethylene glycol/ microfibrillar 311
collagen composite as a new resorbable hemostatic bone wax. J Biomed Matter Res. 312
1998;39:358-363. 313
10. Solheim E, Pinholt E, Andersen R, Bang G, Sudmann E. The effect of a composite of 314
polyorthoester and demineralized boneon the healing of large segmental defects of the radius in 315
rats. J Bone Joint Surg Am. 1992;74A:1456–1463. 316
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
17
11. Solheim E, Pinholt E, Bang G, Sudmann E. Effect of local hemostatics on bone induction 317
in rats: A comparative study of bone wax, fibrin-collagen paste, and bioerodible polyorthoester 318
with and without gentamycin. J Biomed Mater Res. 1992;26:791–800. 319
12. Sugamori T, Iwase H, Maeda M, Inoue Y, Kurosawa H. Local hemostatic effects of 320
microcrystalline partially deacetylated chitin hydrochloride. J Biomed Mater Res. 2000;49:225–321
232. 322
13. Wilkinson HA, Baker S, Rosenfeld S. Gelfoam paste in experimental laminectomy and 323
cranial trephination: hemostasis and bone healing. Journal of neurosurgery. 1981 May;54(5):664-324
7. 325
14. Legeros R, Lin S, Rohanizadeh R, et al. Biphasic calcium phosphate bioceramics: 326
preparation, properties and applications. J Mater Sci Mater Med. 2003;14:201–209. 327
15. Finn MD, Schow SR, Schneiderman ED. Osseous regeneration in the presence of four 328
common hemostatic agents. Journal of oral and maxillofacial surgery. 1992 Jun 30;50(6):608-12. 329
16. Sorrenti S, Cumming W, Miller D. Reaction of the human tibia to bone wax. Clin Orthop 330
Relat Res. 1984:293–6. 331
17. Higashi S, Yamamuro T, Nakamura T, Ikada Y, Hyon S, Jamshidi K. Polymer-332
hydroxyapatite composites for biodegradable bone fillers. Biomaterials. 1986;7(3):183-7. 333
18. Wang M, Armstrong J, Fisher T, Meiselman H, McComb G, Levy M. A New, Pluronic-334
based, Bone Hemostatic Agent That Does Not Impair Osteogenesis. Neurosurgery. 335
2001;49(4):962-967. 336
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
18
19. Vestergaard R, Nielson P, Terp K, Soballe K, Andersen G, Hasenkam J. Effect of 337
Hemostatic Material on Sternal Healing After Cardiac Surgery. Ann Thorac Surg. 2014;97:153-338
160. 339
20. Andelius T, Vestergaard R, Hauge E, et al. The effect of haemostatic devices on bone 340
healing 6 months postoperatively in sternotomized pigs. Eur J Cardiothorac Surg. 341
2015;48(6):850-4. 342
21. Overgaard S, Søballe K, Lind M, Bünger C. Resorption of hydroxyapatite and 343
fluorapatite coatings in man. An experimental study in trabecular bone. J Bone Joint Surg Br. 344
1997;79(4):654-9. 345
346
347
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
19
Figure 1. 348
349
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
20
Figure 2. 350
351
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
21
Figure 3.352
353
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
22
Figure 4.354
355
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
23
Figure 5.356
357
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
24
Figure 6.358
359
360
List of abbreviations 361
PEG – polyethylene glycol, HA –hydroxyapatite, PLA- polylactic acid, mm – millimeter, 362
H&E – hematoxylin & eosin, 363
Acknowledgements 364
365
We would like to acknowledge the assistance and expertise of Liao WenJun, Wang Wei, Bao 366
WenTao and Cheng JingWen for the animal surgery and histology preparation. We would also 367
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
25
like to acknowledge the expertise of Molly Speltz, DVM, in the histomorphometric analysis. 368
Funding Statement 369
The funder provided support in the form of salaries for authors JS JB KR, but did not have 370
any additional role in the study design, data collection and analysis, decision to publish, or 371
preparation of the manuscript. The specific roles of these authors are articulated in the ‘author 372
contributions’ section. 373
Author Contributions 374
Ideas; formulation or evolution of overarching research goals and aims – TT; Development 375
or design of methodology; creation of models – TT; Verification, whether as a part of the activity 376
or separate, of the overall replication/reproducibility of results/experiments and other research 377
outputs - TT PC JS JB KR; Application of statistical, mathematical, computational, or other 378
formal techniques to analyse or synthesize study data – TT; Conducting a research and 379
investigation process, specifically performing the experiments, or data/evidence collection – TT; 380
Provision of study materials, reagents, materials, patients, laboratory samples, animals, 381
instrumentation, computing resources, or other analysis tools - TT PC JS JB KR; Management 382
activities to annotate (produce metadata), scrub data and maintain research data (including 383
software code, where it is necessary for interpreting the data itself) for initial use and later re-use 384
– TT; Preparation, creation and/or presentation of the published work, specifically writing the 385
initial draft (including substantive translation) – TT; Preparation, creation and/or presentation of 386
the published work by those from the original research group, specifically critical review, 387
commentary or revision – including pre- or post-publication stages – TT; Preparation, creation 388
and/or presentation of the published work, specifically visualization/data presentation - TT PC JS 389
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint
26
JB KR; Acquisition of the financial support for the project leading to this publication - TT PC JS 390
JB KR 391
Conflict of Interest 392
The lead author Tristan Tham has no competing interests. Authors Keith Roberts, John 393
Shanahan, and John Burban are employees of Hemostasis LLC. The author Peter Costantino is 394
the co-developer of OsteoStat and has royalty position with Hemostasis LLC. This does not alter 395
our adherence to PLOS ONE policies on sharing data and materials. 396
.CC-BY-NC-ND 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available
The copyright holder for this preprint (which wasthis version posted December 20, 2017. ; https://doi.org/10.1101/236430doi: bioRxiv preprint