This action might not be possible to undo. Are you sure you want to continue?
The effects of local insulin delivery on diabetic fracture healing
Ankur Gandhi, Heather A. Beam, J. Patrick O’Connor, J. Russell Parsons, Sheldon S. Lin*
Department of Orthopaedics, University of Medicine and Dentistry of New Jersey – New Jersey Medical School, 185 South Orange Avenue, MSB-G574, Newark, NJ 07103, USA Received 14 December 2004; revised 16 March 2005; accepted 29 April 2005 Available online 18 July 2005
Abstract Several studies have documented that diabetes impairs bone healing clinically and experimentally. Systemic insulin treatment has been shown to ameliorate impaired diabetic bone healing. However, these studies failed to distinguish between a direct and a systemic effect of insulin upon bone healing. A novel intramedullary insulin delivery system was used in the diabetic BB Wistar femur fracture model to investigate the potential direct effects of insulin on bone healing. Insulin delivery at the fracture site normalized the early (cellular proliferation and chondrogenesis) and late (mineralized tissue, cartilage content and mechanical strength) parameters of diabetic fracture healing without affecting the systemic parameters of blood glucose. These results suggest a critical role for insulin in directly mediating fracture healing and that decreased systemic insulin levels in the diabetic state lead to reduced localized insulin levels at fracture site with concomitant increases in diabetic fracture healing time. D 2005 Elsevier Inc. All rights reserved.
Keywords: Diabetes mellitus; Fracture; BB Wistar rat; Insulin; Drug delivery
Introduction The association between diabetes mellitus and impaired osseous healing has been documented in clinical and experimental settings. Several clinical series have noted that the healing time for diabetic patients is approximately twice as long as for non-diabetic patients [7,19]. Chemically-induced and spontaneous diabetic animal models have demonstrable impairments in fracture healing. In the various fracture healing models, diabetes has led to reduced biomechanical properties of the healing fracture, reduced cellular proliferation in the early callus and reduced collagen synthesis and content compared to non-diabetic control animals [2,11,14,21,31,33]. The mechanism by which diabetes impairs fracture healing is unknown. However, previous studies have demonstrated that high glucose concentrations impair the proliferative response of osteoblast-like cells to insulin like
* Corresponding author. Fax: +1 973 972 5294. E-mail address: firstname.lastname@example.org (S.S. Lin). 8756-3282/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.04.039
growth fractor-I (IGF-I) and delay osteoblast differentiation indicating that defective osteoblast function might contribute to impaired bone formation [1,32]. Sustained hyperglycemia increases formation of advanced glycation endproducts (AGEs) that can signal through a cell-surface receptor, receptor for AGE (RAGE) [6,29]. Previous studies have demonstrated that osteoblasts express RAGE and that RAGE is expressed at significantly higher levels during diabetic vs. non-diabetic bone healing [6,29]. In nondiabetic animals, AGE treatment caused a dose-dependent reduction in bone healing [6,29]. These studies suggest that elevated AGE levels caused by systemic hyperglycemia contribute to diminished bone healing in diabetes, possibly mediated through RAGE. Systemic insulin treatment reverses impaired bone healing in diabetic animals, possibly through enhancement of bone formation and inhibition of bone resorption [2,14,15,21]. However, these studies did not discriminate between an indirect systemic effect on skeletal metabolism or a direct action of insulin on bone healing. It is difficult to ascertain whether the amelioration of impaired healing in the diabetic state through insulin therapy is a result of its
As per the manufacturer. 86% palmitic acid. insulin-releasing) was placed within the intramedullary rod prior to stabilization of the femur fracture allowing immediate release directly to the fracture site. loss of fracture reduction secondary to rod migration (n = 9) and post-operative infection (n = 15). Linshin Canada. Roche Diagnostics. OH). Ontario. histomorphometric analysis (n = 30) and biomechanical testing (n = 34). a bovine insulin-releasing palmitic acid implant (Linplant. 8 T 1 mg. blood obtained from the tail vein was tested for blood glucose levels (ACCUCHEK Advantage. the intramedullary insulin implant has a release rate of 0. The rationale for the later time points was to determine whether a correlation exists between impaired early inflammation/proliferation and the histomorphometric and mechanical properties of the fracture callus at 6 and 8 weeks. The rats were housed under pathogen-free conditions and fed ad libitum. (ii) diabetic rats that received an insulin releasing subcutaneous implant that . Degradation of the intramedullary implant and release of insulin begins immediately upon implantation.12]. with an outer diameter of 1. Canada) as either diabetic resistant (non-diabetic) or diabetic prone at 60 days of age. Blood glucose levels were measured three times a week after initial treatment. Male BB Wistar rats were purchased from Health Canada Animal Research Division (Ottowa. Intramedullary insulin delivery depot Insulin was delivered directly to the fracture site using an insulin-palmitic acid implant placed within a hollow rod that was inserted into the femoral canal in order to stabilize the fracture. Within 4 to 7 days after the onset of glycosuria. IN). All research protocols were approved by the Institutional Animal Care and Use Committee at UMDNJ-New Jersey Medical School. Once glycosuria was detected. The intramedullary insulin-palmitic acid implant consisted of 95% micro-recrystallized palmitic acid and 5% bovine insulin (1 mm diameter. Previous studies have documented that rats will heal by 6 weeks and that diabetic rats will show delayed healing that is not resolved by 8 weeks [2.A. Three experimental groups were used: (i) control. We hypothesize that local insulin delivery to the fracture site directly ameliorates impaired diabetic fracture healing independent of the systemic metabolic state. nondiabetic rats that received a sham subcutaneous implant and a sham intramedullary implant. autoimmune destruction of pancreatic h cells and is associated with genetic and immune factors . Non-diabetic rats received a similar subcutaneous sham implant which does not release insulin. An intramedullary sham implant consisting entirely of palmitic acid was used for the non-diabetic and diabetic without intramedullary insulin treatment groups. The rationale for the early time points (days 2. The amount of subcutaneous insulin implant each diabetic rat received was adjusted accordingly to achieve the appropriate experimental systemic blood glucose level of 300 –400 mg/dl. At sacrifice. The remaining animals were utilized for insulin quantitation (n = 30). weight 26 T 2 mg/implant. Norton. LinShin Canada. The fixation device consisted of a stainless-steel hollow rod. systemic insulin on bone repair. 2 ml of whole blood was collected and glycosylated hemoglobin (HbA1c) levels were determined using the Glyc-AffinGHb kit (Perkin Elmer Life Sciences. 4 and 7 post-fracture) was to look at the inflammatory/early proliferative phase of healing. Canada). Canada) was aseptically placed subcutaneously in the dorsal neck which provides constant insulin release for approximately 30 days (14% bovine insulin. HbA1c is a time-averaged measure of blood glucose control and can be twice as high in patients with poor blood glucose control when compared to normal patients [4. 2 U/ day release rate). The rod was laser machined to produce a regular array of 100 micron diameter holes around the circumference. To isolate the effects of local vs. the beta cells are destroyed completely and if insulin is not administered. a diabetic animal model was used to evaluate the direct effects of insulin on bone repair. Urine from the diabetic prone rats was checked for glycosuria three times a week. / Bone 37 (2005) 482 – 490 483 direct and local effect on the fracture repair process or is a secondary effect due to systemic alterations in the metabolic state.25 mm and an inner diameter of 1. The appropriate implant (sham vs. These time points were chosen based on our previously published data demonstrating reduced platelet-derived growth factor (PDGF) expression in the diabetic fracture callus possibly indicating impaired platelet function/aggregation and a corresponding decrease in cell proliferation . dehydration and ketoacidosis leading to animal death occur. This array was placed to coincide with the mid-diaphyseal region of the femur upon implantation. A local intramedullary insulin delivery depot was evaluated in a rat femur fracture model. cellular proliferation analysis (n = 51). The BB Wistar rat represents an animal model with a close homology to human type I diabetes mellitus. When blood glucose levels became greater than 250 mg/dl. in the absence of systemic metabolic alterations. Indianapolis. The total number of BB Wistar rats utilized for this study (n = 177) included animals sacrificed due to unstable femur fracture pattern (n = 8). muscle wasting.18]. such as by reducing hyperglycemia or formation of advanced glycation end products .05 mm.5 U/day for a minimum of 10 days. 7 mm long. Gandhi et al. Ontario. Materials and methods Animal model The onset of insulin-dependent diabetes mellitus in the BB Wistar rat occurs through the development of insulitis accompanied by a selective.
Ltd. the supernatant was isolated by centrifugation (12. decalcified. confirmed by radiographs. Sections were cut sagitally through the fracture callus using an Isomet diamond saw (Buehler. Mineralized tissue appears red-blue and cartilage appears dark blue.. Histomorphometric analysis was performed to determine areas of cartilage formation and new bone growth as a percentage of total fracture callus area using Scion Image software (Scion Corp. IL). MA). exterior callus. In order to isolate the effect of callus area on cell proliferation.000 rpm for 10 min) and stored at À80-C until testing. NH). The intramedullary rod . were used in this study. Insulin quantitation was performed in non-diabetic rats using a rat specific ELISA assay (ALPCO Diagnostics. 4 and 7 post-fracture. Lake Bluff. 4 and 7 post-fracture in the nondiabetic. Plasma was isolated and stored as above. Areas of proliferation were separated into regions representing either periosteal (intramembranous bone formation) or gap (cartilage formation) callus. Late histomorphometry Fractured femora were resected at 6 and 8 weeks and embedded in polymethylmethacrylate. diabetic and diabetic treated with intramedullary insulin groups. pulverized and total protein extracted using an acid extraction protocol as described . a closed middiaphyseal fracture was created in the right femur using a modification of the method described by Bonnarens and Einhorn . The bone diaphysis or callus was flash frozen in liquid nitrogen. The fracture callus was defined as the region located on either side of the cortices. a thymidine analog) into replicating DNA. Cellular proliferation with 30 mg/kg of BrdU (Sigma Chemical Co. Windham. / Bone 37 (2005) 482 – 490 only poorly controlled systemic blood glucose levels. The fractured femora were decalcified. biebrich scarlet and analine blue. Histomorphometric analysis was performed to determine areas of cartilage formation as a percentage of total fracture callus area using Scion Image software (Scion Corp. since endogenous rat insulin is not detectable in the diabetic rats . the sections were stained with Weigerts iron hematoxylin. Ltd. Only animals that exhibited non-comminuted. St. After fracture. The fractured and contralateral femora were resected and the fracture callus and mid-diaphyseal region corresponding to the fracture callus were isolated. transverse fractures. California) primary antibody. The samples were viewed and digital photomicrographs were taken using an Olympus BH2-RFCA microscope (Olympus Optical Co. Carpinteria.. Waltham. Local insulin levels were normalized to total protein concentration measured using a bicinchoninic acid (BCA) protein assay (Pierce.8 mm Â 4. Japan) and a Polaroid DMC1e digital camera (Polaroid.. embedded and sectioned using standard histological techniques. Cellular proliferation was determined by using monoclonal mouse anti-Brdu (Clone Bu20a. Insulin quantitation Systemic insulin levels (plasma) were measured prior to fracture and on days 2.. To identify cartilage formation. Early histomorphometry The fractured femora were resected at days 2. Tokyo. external to the intramedullary marrow cavity. The assay and analyses were performed according to the instruction of the manufacturer. IL). DAKO Corp. Local insulin levels were measured from the fractured and contralateral femora on days 2. Mineralized tissue appears orangered and cartilage appears dark blue. Prior to fracture. as described . total cell proliferation (periosteal + gap callus proliferation) was normalized to callus area. 4 and 7. mounted on plexiglass slides and polished to a thickness of 100 um. embedded and sectioned using standard histological techniques. blood was collected by tail bleed and the plasma was isolated by centrifugation (1000 rpm for 30 min) and stored at À80-C until testing. After extraction. 4 and 7 post-fracture. Fracture model Fourteen days after the onset of diabetes. The animals were allowed to ambulate freely immediately post-fracture. Rockford. Frederick. Frederick. The fracture callus was defined as the region located on either side of the cortices. Gandhi et al. 4 and 7 post-fracture in the non-diabetic and diabetic treated with intramedullary insulin groups. Windham. A rectangular area of 0. blood was collected by cardiac puncture from rats euthanized at days 2. Louis. MO) 1 h prior to sacrifice at days 2.. All animals were age-matched (106 T 7 days) at the time of fracture.484 A. external to the intramedullary marrow cavity. Mechanical testing Cellular proliferation was measured by counting cells that had incorporated 5-bromo-2-deoxyuridine (BrdU. MD). and an insulin-releasing intramedullary implant. The slides were stained with Stevenel’s blue and Van Gieson picro-fuchsin to identify cartilage formation and new bone growth . Systemic insulin levels were expressed as units per milliliter of plasma. and a sham intramedullary implant and (iii) diabetic rats that received an insulin releasing subcutaneous implant that only poorly controlled systemic blood glucose levels. Shinjuku-ku.0 mm adjacent to the cortex was used to count positively stained cells on the anterior and posterior. Animals were injected Fractured and contralateral femora were resected at 6 and 8 weeks and cleaned of soft tissue. NH) and in diabetic rats using bovine specific ELISA assays (ALPCO Diagnostics. MD).
99). (power = 0. d Represents values statistically higher than diabetic treated with intramedullary insulin values. the number of groups and corresponding sample sizes and an alpha of 0..99).b 9. Ward Hills. P < 0. Cellular proliferation At day 2. Gandhi et al. The blood glucose and HbA1c levels between the diabetic and diabetic treated with intramedullary insulin groups were not different indicating that insulin from the intramedullary depot had no effect on systemic blood glucose (power = 0.05 was considered statistically significant.A. P < 0.1 T 1. P < 0.8a. Systemic rat insulin levels in the non-diabetic group were significantly greater than the systemic bovine insulin levels in the diabetic and diabetic treated with intramedullary insulin groups indicating the presence of a hypoinsulinemic state in both diabetic groups (Table 2).10]. Systemic bovine insulin levels in both diabetic groups were comparable indicating that insulin from the intramedullary insulin delivery system did not have a significant systemic effect (power = 0.01. the difference between the group means of the non-diabetic and diabetic groups.. bovine insulin levels in the fractured femora were approximately fivefold higher than in the contralateral femora demonstrating the efficacy of the intramedullary insulin delivery system (Table 3). the standard deviation of the population.99). b Represents values statistically lower than diabetic treated with intramedullary insulin values. AZ) and tested to failure at a rate of 2. c Represents values statistically higher than diabetic values.0 (SPSS Inc. In the diabetic rats treated with intramedullary insulin. Systemic rat insulin levels in the non-diabetic group and systemic bovine insulin levels in both diabetic groups remained relatively constant through day 7 post-fracture (Table 2). periosteal and gap callus proliferation in the diabetic callus was significantly reduced compared to the Results Animal model—general health The blood glucose and HbA1c levels for the non-diabetic group were significantly lower compared to the diabetic and diabetic treated with intramedullary insulin groups (Table 1).d 10 T 6 9T4 The data represent average value T standard deviation. / Bone 37 (2005) 482 – 490 485 was removed from the fractured femur. In addition. a Represents values statistically lower than diabetic values. Torsional testing was conducted using a servohydraulics machine (MTS Systems Corp. torsional rigidity (TR) and the maximum shear stress (s) were determined through standard equations modeling each femur as a hollow ellipse [9. One-way analyses of variance (ANOVA) were performed followed by Holm – Sidak post-hoc tests to determine specific differences. Statistical analysis All statistical analyses were performed using SigmaStat 3.05. The power for a one-way ANOVA was determined by specifying in this study. Chicago. Insulin quantitation Diabetic BB rats do not produce detectable levels of endogenous rat insulin and therefore the only source of insulin is from a subcutaneous bovine insulin-releasing implant .4 T 1.6 9. Percent weight gain from the time of surgery in both diabetic groups was comparable and significantly lower than in the non-diabetic groups at 6 (power = 0.92) post-fracture (Table 1). Bovine insulin levels in the fractured femora remained relatively constant through day 7 post-fracture indicative of a constant release profile. Illinois).9 T 1. The intramedullary insulin delivery system attained bovine insulin levels in the diabetic fractured femur that were significantly lower (approximately 70%) than rat insulin levels in the non-diabetic fractured femur (power = 0.001. there was no angulation of the fractured femora evident in either the X-rays or the resected bone.001. Scottsdale. A P value <0. Normalized data were obtained by dividing each fractured femur value by its corresponding contralateral femur value. Based on X-rays taken at the time of sacrifice in the AP and lateral planes. the intramedullary rods showed no evidence of bending.0 deg/s. The proximal and distal ends of the fractured and contralateral femora were embedded in Wood’s metal (Alfa Aesar.d 8T6 8T5 8 weeks 25 T 12c. In the non-diabetic group.01. P < 0. . MN) with a 20 Nm reaction torque cell (Interface. The peak torque (T). MA).99) or HbA1c levels (power = 0. Eden Prairie.94) and 8 weeks Table 1 General health Blood Glucose (mg/dl) 83 T 16a.9 Glycosuria Percent weight gain 6 weeks Non-diabetic (n = 44) Diabetic treated with intramedullary insulin (n = 56) Diabetic (n = 45) À + + 20 T 7c.b 367 T 42 380 T 43 HbA1c 4. rat insulin levels in the fractured femur and contralateral femur were comparable and constant through day 7 post-fracture.
the callus area was not significantly different between the three groups but there was more significantly more percent cartilage in the nondiabetic compared to the diabetic callus (Table 5).04 The data represent average value T standard deviation. At week 8.11 T 0.35 0. At week 6. The diabetic group also showed no evidence of remodeling resulting in the continued presence of the fracture callus. 1). all three groups showed reduced periosteal and gap callus proliferation compared to day 2.32 0. the periosteal callus was comparable in all three groups but the gap callus in non-diabetic and diabetic treated with intramedullary insulin animals was more advanced with increased percent cartilage compared to the diabetic group (power = 0.21a 0. hematoma was adjacent to the fracture site with an elevated periosteum above the cortical bone with no significant difference in callus area between the three groups (Table 4).96) compared to the non-diabetic and diabetic treated with intramedullary insulin groups (Table 6).19a 0.78 T 0. Early histomorphometry At day 2. The diabetic group Table 3 Local insulin levels Insulin levels normalized to total protein levels (pg/mg) Non-diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 6) Fractured femur Contralateral femur Fractured femur Contralateral femur Day 2 post-fracture (pg/mg) 1. The cartilage in the nondiabetic and diabetic treated with intramedullary insulin consisted of more proliferating chondrocytes and hypertrophic chondrocytes as compared to the diabetic fracture callus.03 Day 4 post-fracture (pg/mg) 1. the non-diabetic rats and the diabetic rats treated with intramedullary insulin showed almost complete restoration of the cortical bone whereas the diabetic group showed incomplete bridging (Fig. Total cell proliferation normalized to the fracture callus area in the diabetic group was significantly decreased compared to the nondiabetic and diabetic treated with intramedullary insulin at days 2 and 4 post-fracture (Table 4. P < 0. P < 0. / Bone 37 (2005) 482 – 490 Prior to fracture (pg/ml) Non-diabetic (n = 6) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 6) 343 T 12 20 T 7 24 T 8 a Day 2 post-fracture (pg/ml) 398 T 14 24 T 7 27 T 6 a Day 4 post-fracture (pg/ml) 381 T 15 29 T 8 25 T 7 a Day 7 post-fracture (pg/ml) 352 T 11a 22 T 7 25 T 8 The data represent average value T standard deviation. histologic analysis showed bridging of the cortice/fracture callus in the non-diabetic and diabetic treated with intramedullary insulin groups whereas the diabetic group showed incomplete bridging (Fig.23 0. 1). Local insulin levels were measured in the non-diabetic group using a rat insulin ELISA and in the diabetic treated with intramedullary insulin group using a bovine insulin ELISA. cell proliferation for all three groups was not significantly different. non-diabetic and diabetic treated with intramedullary insulin callus (Table 4).26 0. Plasma insulin levels were measured in the non-diabetic group using a rat insulin ELISA and in the diabetic and diabetic treated with intramedullary insulin groups using a bovine insulin ELISA. The diabetic treated with intramedullary insulin also showed increased percent cartilage compared to the diabetic group but the difference was not significant (power = 0. Late histomorphometry At week 6. At day 7.28 0. The diabetic treated with intramedullary insulin showed some remodeling resulting in a reduced fracture callus size.10 T T T T 0. power = 0.76 0.23a 0.486 Table 2 Plasma insulin levels A. The intramedullary insulin depot normalized cell proliferation in the diabetic treated with intramedullary insulin group to levels not statistically different from non-diabetic group (power = 0. Gandhi et al.87). The diabetic fracture callus also contains a greater proportion of fibrous tissue. compared to periosteal and gap callus proliferation in the non-diabetic and diabetic treated with intramedullary insulin group.69 0.85).84 0.97) and significantly increased percent cartilage area (power = 0. At day 7. a Represents values statistically higher than diabetic and diabetic treated with intramedullary insulin values.12 T T T T 0. The intramedullary insulin depot normalized periosteal and gap callus cell proliferation in the diabetic treated with intramedullary insulin group to levels not statistically different from non-diabetic group (power = 0.89).08 0.86). . a Represents values statistically higher than contralateral femur values.02 Day 7 post-fracture (pg/mg) 1.01. No differences between the three groups were demonstrated at day 7 post-fracture.92 T 0.96 0.24 0.14 T 0. At day 4. Periosteal and gap callus proliferation in the diabetic fracture calluses also was significantly reduced at day 4.001. At day 4. the diabetic group exhibited significantly decreased percent mineralized area (power = 0.24 0.92).
b 8.51) were all significantly reduced in the diabetic animals compared to the non-diabetic (Table 7). torsional rigidity (power = 0. Mechanical testing The fractured femora.97) compared to the non-diabetic and diabetic treated with intramedullary insulin groups (Table 6).01 3.97).26a 0. b Represents values statistically lower than diabetic treated with intramedullary insulin values. Gandhi et al.94 T 4. and the contralateral.34 4. P < 0.94 T 1. the torque to failure (power = 0.83 T 0.80) and percent maximum shear stress (power = 0. P < 0.b 328 T 157 1155 T 231 604 T 212a.69).84 T 0.24 0. continued to exhibit delayed remodeling resulting in the presence of the fracture callus. The diabetic femora had significantly reduced mechanical properties at weeks 6 and 8 post-fracture as compared to the non-diabetic and diabetic treated with local intramedul- Table 5 Early histomorphometry Percent cartilage content (cartilage / total callus area) Four days Non-diabetic (n = 6) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 6) 1. a Represents values statistically lower than non-diabetic values.13 3.51 T 3. percent torsional rigidity (power = 0.28 T 2. P < 0.98) and significantly increased percent cartilage area (power = 0.15 388 T 127 204 T 35a. the percent torque to failure (power = 0. a Represents values statistically lower than non-diabetic values. When these values were normalized to the contralateral limb.01.15 12.32 T 3.b 1840 T 624 0.A. Femora that did not heal failed transversely through the non-union site (fracture gap).02 0.05.b 404 T 155 Periosteal callus (positive cells) 1400 T 384 722 T 168a.91). When the data were normalized to the contralateral limb.b 1010 T 287 381 T 125 208 T 91a. The maximum shear stress was significantly reduced in the diabetic animals compared to the non-diabetic animals (power = 0. At week 8 post-fracture. the diabetic group exhibited significantly decreased percent mineralized area (power = 0. P < 0. . / Bone 37 (2005) 482 – 490 Table 4 Cellular proliferation analysis Callus area (mm2) Two days Non-diabetic (n = 5) Diabetic (n = 5) Diabetic treated with intramedullary insulin (n = 6) Four days Non-diabetic (n = 6) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 6) Seven days Non-diabetic (n = 6) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 5) Gap callus (positive cells) 499 T 84 305 T 41a.89 T 1.92 T 0.78 lary insulin groups (Table 7). The diabetic fracture callus also contains fibrous tissue.33 T 0.31 0. Discussion Diabetes impairs the fracture healing process beginning with a reduction in early cellular proliferation. b Represents values statistically lower than diabetic treated with intramedullary insulin values. All three groups had increased percent mineralized area at week 8 compared to week 6 but the difference was not statistically significant.08 4. No significant differences were detected between the femora of the non-diabetic group and the diabetic group treated with local intramedullary insulin.01.57 T 1. The mechanical properties of the intact contralateral femur showed no significant difference between the three groups.98) and torsional rigidity (power = 0. At week 8.97) were significantly reduced in the diabetic group when compared to the nondiabetic and diabetic treated with intramedullary insulin groups. At week 6 post-fracture. intact femora failed spirally in the mid-diaphyseal region following torsional mechanical testing to failure.42 Seven days 9.94) and maximum shear stress (power = 0.81) were significantly reduced in the diabetic animals compared to the non-diabetic and diabetic treated with intramedullary insulin animals. continuing with a delay in endochondral ossification and ending with a decrease in the biomechanical properties of the fracture The data represent average value T standard deviation. torque to failure (power = 0.37 T 2.b 1290 T 346 487 Positive cells per unit area (cells/mm2) 1980 T 601 1130 T 512a.95) were significantly reduced in the diabetic fractured femora compared to the non-diabetic and diabetic treated with intramedullary insulin groups. percent torque to failure (power = 0.63a. exhibiting complete or partial union.98) and percent torsional rigidity (power = 0.b 339 T 134 12.49 T 2.02 11.05 T 1.05.96 T 0.45 216 T 148 141 T 84 189 T 75 599 T 304 445 T 152 505 T 234 65 T 21 46 T 13 57 T 25 The data represent average value T standard deviation.
b Represents values statistically lower than diabetic treated with intramedullary insulin values. Local insulin treatment.b 77. These results suggest that insulin is critical to the diabetic fracture healing process through a . To isolate the effects of insulin on bone healing. the mechanism through which diabetes impairs bone healing is currently unknown. this study does not address the role of insulin on impaired healing associated with hypoinsulinemia and hyperglycemia present in the diabetic condition. these studies demonstrate a failure to distinguish between the direct and indirect effects of insulin upon bone healing.5 T 4. Within the early diabetic fracture callus.6 T 7. 1. This study suggests a direct effect of insulin on bone formation in normoglycemic animals. P < 0.0 The data represent average value T standard deviation.24. hyperglycemia or AGE production) upon the healing process [1.01. However. However. In a diabetic bone explant model. (B) diabetic and (C) diabetic treated with intramedullary insulin groups and 8 weeks in the (D) non-diabetic. However.2 T 8. insulin delivery to calvaria of a normoglycemic animal increased the histomorphometric indices of bone formation . callus. local insulin normalized impairments in cellular proliferation and chondrogenesis.1c. orange-red = mineralized tissue.3 51. Few studies have isolated the individual effects of diabetes (i.d 11. P < 0.9 T 4. P < 0.6 T 7.7 90.e.01. The normalization of diabetic fracture healing through systemic insulin delivery could be attributed to the indirect actions of insulin on blood glucose levels and AGE production. Gandhi et al.29.9a. immediately post-fracture.1 61. Slides were stained with Stevenel’s Blue and Van Gieson’s Picrofuchsin (dark blue = cartilage. chondrogenesis and biomechanical parameters of the diabetic fracture callus . The present study provides new insight into the role of local insulin on fracture healing. However. insulin treatment has been shown to ameliorate Table 6 Late histomorphometry Percent bone content (bone / total callus area) Six weeks Non-diabetic (n = 5) Diabetic (n = 5) Diabetic treated with intramedullary insulin (n = 5) Eight weeks Non-diabetic (n = 5) Diabetic (n = 5) Diabetic treated with intramedullary insulin (n = 5) 84.5 T 2. insulin treatment sufficient to achieve physiologic blood glucose control resulted in normalization of impaired thymidine uptake reflective of increased DNA synthesis and proliferation.488 A. P < 0.9 Percent cartilage content (cartilage / total callus area) 8. / Bone 37 (2005) 482 – 490 Fig. callus bone content and biomechanical properties in the late diabetic callus. c Represents values statistically higher than diabetic values. This model also demonstrated that early insulin treatment restored proteoglycan production and chondrogenesis .01. hypoinsulinemia.6 T 4.7 T 6. 4Â magnification). normalized mineralization. impaired bone healing in vitro and in vivo.2 T 5. a Represents values statistically lower than non-diabetic values.2 4.7 T 7. d Represents values statistically higher than diabetic treated with intramedullary insulin values.6a. In the present study.5 22.0 27.4 T 4. In a diabetic femur fracture model.6c.32].d 6. (A – F) Late fracture callus histology: Histology of gap callus (area of endochondral ossification) at 6 weeks in the (A) non-diabetic.01. tight systemic blood glucose control achieved through increased systemic insulin supplementation normalized cell proliferation. (E) diabetic and (F) diabetic treated with intramedullary insulin groups.b 84.9 T 3. the role of local insulin delivery upon fracture healing was investigated in rats that spontaneously develop insulindependent diabetes mellitus.
b 401 T 80 81 T 21 42 T 15c 65 T 28 26.0 T 21. Local insulin delivery directly to the fracture site could have significant clinical implications. To our knowledge.5 500 T 54 250 T 108a.b 19.8 8.9 T 7.4b 15. Another possible etiology of impaired diabetic fracture healing is the role of insulin signaling on expression of specific factors critical for bone healing.26. diabetes impaired expression of Runx-2. Future work will include examining the relationship between the insulin signaling pathway and the expression of genes critical for the regulation of the fracture healing process.b 80 T 34 Maximum shear stress (Mpa) 7. IRS-1À/À mice exhibited increased incidence of non-union and decreased bone mineral content compared to wild-type . in mediating the fracture healing process. / Bone 37 (2005) 482 – 490 Table 7 Mechanical testing Torque to failure (Nm) Six weeks Non-diabetic (n = 6) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 5) Eight weeks Non-diabetic (n = 5) Diabetic (n = 6) Diabetic treated with intramedullary insulin (n = 6) 414 T 50 178 T 84a. The various stages of fracture healing have been correlated to the temporal expression of three groups of soluble factors (pro-inflammatory cytokines. IGF-I plays an important role in the anabolic regulation of bone and cartilage metabolism.7a 6.23.8 T 7. One potential mechanism involves the role of insulin receptor substrate-1 (IRS-1). Another potential mechanism involves the existence of local insulin gradients within the fracture callus that are sufficient to stimulate IGF-I receptor signaling.1 T 1. The previous study in conjunction with the present study demonstrating impaired cell proliferation and impaired chondrogenesis suggests that diabetes disrupts the regulation and interaction of factors critical for the normal progression of the inflammatory phase and chondrogenesis leading to decreased biomechanical properties of the diabetic femora. Previous studies have theorized that reduced PDGF expression in the diabetic callus led to decreased cellular proliferation and that impaired cellular proliferation in the diabetic fracture callus led to reduced mechanical properties [21.b 54 T 13 Torsional rigidity (Nm/rad) 20.01.27]. Gandhi et al.700 T 4120a. In IRS-1À/À mice.2 T 1. specific regulators of osteoblast differentiation . These results in conjunction with the results obtained in the present study indicate a central role for insulin signaling through the IRS1 pathway in mediating the fracture healing process . Histologically.1 3. P < 0. attempting to explain the impaired fracture healing process associated with the diabetic state.2 16. potentially through the IRS-1 pathway.b 360 T 77 Percent torque to failure 69 T 17 25 T 10a.2 3. b Represents values statistically lower than diabetic treated with intramedullary insulin values.b 23. the TGF-h superfamily and angiogenic factors) that play a critical role in modulating the repair process .2 T 0.5 T 23. direct effect on the fracture callus and indicates that insulin may have a direct effect on fracture healing in normal animals.5c 32. Insulin treatment ameliorated impaired expression of Runx-2 and Dlx5 as well as bone matrix osteocalcin and collagen type I . P < 0. it is difficult to determine the exact mechanism through which insulin signaling plays a role in the normalization of diabetic fracture healing. P < 0.05. In a tibia fracture model. the human homolog of the Drosophila distal-less gene. fibrous tissue was present at 3 and 6 weeks. bFGF stimulates fracture repair through a mitogenic effect on mesenchymal cells and through enhancement of TGF-h expression which is involved in chondrogenesis and bone formation [16. Several mechanisms have been proposed.100 T 8880 89 T 31 36 T 19c 71 T 33 8.2 The data represent average value T standard deviation.34].A. In a model of intramembranous bone formation. The impaired expression of these key regulators was followed by decreased expression of osteocalcin and collagen type I corresponding to decreased bone formation measured through histomorphometric analysis.b 8. . runt domain factor-2.9 40.7 T 4.500 T 4180 10. While the present study along with previous studies suggests insulin and its signaling pathways initiate a complex cascade of factors.1 T 2. This study confirms the importance of local insulin in ameliorating the impaired early and late parameters of diabetic fracture healing. this is the first study looking at the effect of local insulin delivery to the fracture site in diabetic animals.1 T 1.8 489 Percent maximum shear stress 10. an essential protein in the signaling pathway of insulin and IGF-I.900 T 3940 9420 T 3760a. c Represents values statistically lower than non-diabetic values. and Dlx5.5a.1 T 4.6 T 1. A previous fracture healing study demonstrated reductions in bFGF expression in the early diabetic fracture callus which was restored with insulin treatment . wild-type mice progressed through normal endochondral ossification at 3 weeks and bony union at 6 weeks. Detailed histologic analysis revealed that IRS-1À/À fracture site was associated with a decrease in chondrocyte proliferation.01.800 T 6150 Percent torsional rigidity 72 T 26 29 T 18a. a Represents values statistically lower than non-diabetic values.
The effects of blood glucose control upon fracture healing in the BB Wistar rat with diabetes mellitus. Correction of abnormal bone and mineral metabolism in chronic streptozotocin-induced diabetes mellitus in the rat by insulin therapy. Chase HB. Hoshi K. Diabetes 2003. J Orthop Res 2002. Theory and practice of histological techniques. Cerami A.135:774 – 81.59:492 – 5. Stevens A.  Nakhooda AF. Centrella M. Matrix 1992.  Cortizo AM. J Biol Chem 2004. Engesaeter LB. Clin Orthop 1988:210 – 6. Biomechanical evaluation of early fracture healing in normal and diabetic rats.490 A. Borretos J. Stiles AD.279:15314 – 22.35:109 – 21. Parsons JR. Deporter DA. Fracture healing as a post-natal developmental process: molecular. Does diabetes delay fracture healing? Clin Orthop 1972. Ogata E.28:21 – 8. J Cell Biochem 2003. Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation.88:873 – 84. Acta Orthop Scand 1978. Fallon MD.1:31 – 7. Hiyama Y. Chappel CI. Correlation of glucose regulation and hemoglobin AIc in diabetes mellitus. Fibroblast growth factor enhances type beta 1 transforming growth factor gene expression in osteoblast-like cells.  Santana RB. J Orthop Res 2000. Endocrinology 1981. Bolander ME. Szabo P. Tamura M. et al.  Maniatopoulos C.  Bonnarens F. Growth-inhibitory effect of a high glucose concentration on osteoblast-like cells.20:1210 – 6. .  McNair P. Avioli LV.  Cornish J.125:2118 – 26. Rodriguez A. Barnes GL.  Lu H.  Shimoaka T. Murray FT. Underwood LE. Hirschman A. Hurwitz SR.108:2228 – 34. Dan Med Bull 1988. An improved method for preparing histological sections of metallic implants. Beam HA. Jingushi S. Lin SS. References  Balint E. Canalis E. Peterson CM. Hasuma T. Marshall CF. Zarro C. Lettieri MG. Kurokawa T. Reid IR.  Ekeland A. Barrio DA. The influence of diabetes mellitus on the healing of closed fractures.  Noda M. Int J Oral Maxillofac Implants 1986. Etcheverry SB. Effects of fibroblast growth factors on deoxyribonucleic acid and collagen synthesis in rat parietal bone cells. Lin SS.  Page K. and Aircast Foundation Grant (SSL). Hanada K.  Funk JR. Graves DT.  Spanheimer RG. Type X collagen in fracture callus and the effects of experimental diabetes. Endocrinology 2003. Defects of early fracture-healing in experimental diabetes. Bergfeld MA. 144:346 – 52.20:587 – 92. Chikuda H. Ekeland A.  Macey LR. Beam HA. Vogel R. Langeland N. Bolander ME.52:1502 – 10. Parsons JR. J Orthop Res 2002. Endocrinology 1989. 21:199 – 209. Hale JE. Yano Y.  Beam HA. Impairment of bone healing by insulin receptor substrate-1 deficiency.71:722 – 33. Insulin increases histomorphometric indices of bone formation in vivo. Biochem Biophys Res Commun 1975. McCarthy AD. 1996.  Topping RE. Calcif Tissue Int 1996. Sprague SM. Saudek C. Mol Cell Biochem 2003. Morii H. Endocrinology 1989.22:17 – 23. N Engl J Med 1976. Further identification of the nature and linkage of the carbohydrate in hemoglobin A1c. Gerstenfeld LC. Lin SS. Clin Orthop 2003:319 – 30. Graves DT. Marliss EB. Graves DT. Gabbay KH.8:424 – 31.  Kawaguchi H. Advanced glycation end-products (AGEs) induce concerted changes in the osteoblastic expression of their receptor RAGE and in the activation of extracellular signal-regulated kinases (ERK).  Tyndall WA. Inaba M. Shaftan GW. O’Connor JP.  McCarthy TL. Vieira P. Stimulation of fracture repair by recombinant human basic fibroblast growth factor in normal and streptozotocin-diabetic rats. Jingushi S. Bone mineral metabolism in human type 1 (insulin dependent) diabetes mellitus.67:103 – 9.52:605 – 13. Cullinane DM. Acknowledgments Funded by AOFAS Research Grant (SSL). Akune T.81:935 – 9. et al. Glucose-induced inhibition of in vitro bone mineralization. Einhorn TA.250:1 – 10. Bone 1998. Powers JC. Like AA. Gandhi et al. Mechanical properties of fractured and intact rat femora evaluated by bending. Trackman PC. Carmines D. Langeland N. Terek RM. Acta Orthop Scand 1981. Lehrman M.18:126 – 32. Correlation between decreased collagen production in diabetic animals and in cells exposed to diabetic serum: response to insulin. Camilliere JJ.26:100 – 12.  Engesaeter LB. Melcher AH. OREF Research Grant (SSL). Kraut D. Einhorn TA. Decreased platelet derived growth factor expression during fracture healing in diabetic animals. Xu L. 82:134 – 40. Metabolic and morphologic studies. Gooch HL. spatial. Endocrinology 1994. Jones RL. Retardation of fracture healing in experimental diabetes. Production of a standard closed fracture in laboratory animal bone. Clin Orthop 1994: 220 – 8.  Cozen L. A role for advanced glycation end products in diminished bone healing in type 1 diabetes. Proc Natl Acad Sci U S A 1984. Haney DN..124:2991 – 4. and temporal aspects of its regulation. Mercer N. Orthop Clin North Am 1990. Slatopolsky E.  Loder RT.12:101 – 7. Gallop PM. Diabetes 1977. In vivo stimulation of bone formation by transforming growth factor-beta.  Herbsman H.2:97 – 101. New York’ Churchill Livingstone Inc.  Terada M. Callon KE. Nishizawa Y.  Koenig RJ.295:417 – 20. Teitelbaum SL. / Bone 37 (2005) 482 – 490  Joyce ME. Transforming growth factor-beta in the regulation of fracture repair.  D’Ercole AJ.109:2529 – 35. torsional and tensile tests. Low-intensity pulsed ultrasound increases the fracture callus strength in diabetic BB Wistar rats but does not affect cellular proliferation.  Gebauer GP.49:512 – 8. Tissue concentrations of somatomedin C: further evidence for multiple sites of synthesis and paracrine or autocrine mechanisms of action. J Orthop Res 1984. J Surg Res 1968. Balian G. Methods for testing the mechanical properties of the rat femur. Bone 2001.  Noda M.  Bunn HF. J Cell Biol 1989. The spontaneously diabetic Wistar rat. Kana SM. Bancroft JD. Kamekura S. Chung UI. Bolander ME. Amar S.  Gerstenfeld LC. J Bone Joint Surg Am 1989.  Hough S. et al. Lowe J.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.