Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 May 1.
Published in final edited form as: Spine J. 2019 Nov 20;20(5):800–808. doi: 10.1016/j.spinee.2019.11.011

The effect of insulin dependent diabetes on bone metabolism and growth after spinal fusion

Zachary NaPier a, Linda EA Kanim a,b, Trevor J Nelson a,c, Khosrowdad Salehi a,b,d, Yasaman Arabi a,b,d, Juliane D Glaeser a,b,d, Dmitriy Sheyn a,b,d,e,f, Melodie F Metzger a,c,*
PMCID: PMC7234904  NIHMSID: NIHMS1564710  PMID: 31759133

Abstract

STUDY DESIGN:

Experimental animal model.

OBJECTIVE:

The purpose of this study was to evaluate the hypothesis that insulin dependent diabetes mellitus (IDDM) will inhibit the formation of a solid bony union after spinal fusion surgery via an alteration of the microenvironment at the fusion site in a rat model.

SUMMARY OF BACKGROUND DATA:

Previous studies report diabetes mellitus (DM) and specifically IDDM as a risk factor for complications and poor surgical outcomes following spinal fusion.

METHODS:

Twenty control and 22 diabetic rats were obtained at 5 weeks of age. At 20 weeks of age, all animals underwent posterolateral lumbar fusion surgery using a tailbone autograft with diabetic rats receiving an implantable time release insulin pellet. A subset of rats was sacrificed 1-week postsurgery for growth factor (PDGF, IGF-I, TGF-β, and VEGF) and proinflammatory cytokine ELISA analysis. All other rats were sacrificed 3-months postsurgery for fusion evaluation via manual palpation and micro CT. Glycated hemoglobin (HbA1c) was measured at surgery and sacrifice on all animals.

RESULTS:

Compared with healthy rats undergoing spinal fusion, rats with IDDM demonstrated a significant reduction in manual palpation fusion rates (16.7% vs. 43%, p<.05). Average bone mineral density, bone volume, and bone volume fraction were also significantly reduced and negatively correlated to blood glucose levels. IL-1B, IL-5, IL-10, TNF-α, and KC/GRO were significantly elevated in fusion beds of IDDM rats.

CONCLUSIONS:

This study demonstrates that rats with IDDM demonstrate a reduced rate and quality of spinal fusion with increased local levels of inflammatory cytokines. Targeted modalities are required to improve bone healing in this growing, high-risk population.

CLINICAL SIGNIFICANCE:

This is the first translational animal model of IDDM to evaluate the rate and quality of spinal fusion while controlling for other surgical and patient-related risk factors. Our findings demonstrate the complex nature by which IDDM impairs bone healing and highlight the need for additional basic science research to further elucidate this mechanism in order to develop more effective therapeutic interventions.

Keywords: Diabetes, Growth factors, Insulin dependence, Nonunion, Pseudarthrosis, Spinal fusion

Introduction

Diabetes mellitus (DM), a chronic metabolic disorder related to impaired blood glucose regulation, has become ubiquitous in both United States and global populations. A 2012 National Health and Nutrition Examination Survey found that 49%–52% of the US population was either diabetic or prediabetic [1]. The global prevalence of diabetes in patients aged 20–79 years was estimated to be 415 million in 2015, with this number projected to increase to 642 million by 2040 [2]. Spine surgeons must understand the unique characteristics of DM as they relate to spinal surgery in order to provide optimal care to this rapidly growing patient population.

Type 1 or insulin-dependent diabetes mellitus (IDDM) is a chronic inflammatory disease in which the insulin producing islet cells in the pancreas are destroyed by the body’s own immune system [3]. At the onset of IDDM, there is a dysregulation of systemic growth factors and cytokine levels, including IL-1β and IL-10 [4].

Previous studies report diabetes as a risk factor for increased incidence of complications and poor surgical outcomes following lumbar spine surgery including infection, pseudarthrosis, and lack of symptomatic improvement after discharge [5–13]. Several of these studies have specifically identified IDDM as a risk factor for perioperative complication and nonunion compared with non–insulin-dependent diabetes mellitus (NIDDM) in patients undergoing posterior instrumented fusion [5–8,14]. In 2013, Deyo et al. reported that patients with IDDM had more than twice the rate of major complications and mortality after lumbar spine surgery when compared with nondiabetic controls or patients with NIDDM [14]. Similarly, Golinvaux et al. performed a retrospective multivariate analysis in which they concluded diabetic patients using insulin have a significantly higher surgical risk compared with NIDDM [7].

Research regarding the association of IDDM with complications in lumbar spinal fusion surgery has been limited to retrospective clinical analyses. The purpose of this study was to evaluate the hypothesis that IDDM will inhibit the formation of a solid bony union after spinal fusion surgery via an alteration of the microenvironment at the fusion site in an experimental animal model.

Material and methods

Study design

Approval for this study was obtained from the Institutional Animal Care and Use Committee (IACUC#005364). Twenty control (Biobred Diabetic Resistant, BBDR) and 22 diabetic (Biobred Diabetic Prone, BBDP, Biomere Wor-chester, MA) rats were obtained at 5 weeks of age. Biobred Diabetic Prone rats spontaneously develop diabetes between 50 and 90 days of age [15]. Nonfasting blood glucose (BG) and weight were monitored on BBDP diabetic rats 3 times a week and monthly on BBDR controls. When BBDP rats achieved an overt diabetic status (BG>300 mg/dL) chronic insulin therapy was administered. Rats were anesthetized via isofluorane and a Linplant time-released insulin pellet (LinShin Canada, Inc) was implanted subcutaneously into the nape of the neck. Each provided 2U insulin per day for 45–60 days. Additional pellets were administered if blood glucose levels exceeded 450 mg/dL to prevent ketoacidosis. Glycated hemoglobin (HbA1c), a long-term measure of glucose health, was measured at surgery and sacrifice on all animals. At 20 weeks of age all animals underwent posterolateral and laminar fusion surgery using a tailbone autograft implanted into the L4/L5 transverse processes (TPs), as previously described [16]. Four BBDR controls and seven BBDP diabetic rats were sacrificed 1-week postsurgery for growth factor and proinflammatory cytokine ELISA analysis. Remaining rats were sacrificed 3-months postsurgery (32 weeks of age) for fusion evaluation via manual palpation and μCT, Faxitron. Rats were housed in a vivarium and allowed to drink and feed ad libitum.

Surgical procedure

Posterolateral interlaminar fusion using tail vertebral bone autograft was performed, as previously described [16,17]. Briefly, the tail was amputated and six tail vertebrae were extracted and prepared for use as an autograft. A midline posterior incision was made centered over L4–L5 and a high-speed burr was used to decorticate the TPs, facets, and lamina. The autograft was packed into the posterolateral gutter spanning the TPs and across the posterior laminar space. The fascia, subcutaneous tissues, and skin were closed in a layered fashion. Postoperatively, Bupre-norphine and lactated ringers were administered subcutaneously. Rats were individually caged after surgery until their wounds healed. Oral antibiotics (0.1 mg/mL Enrofloxacin) were administered to both groups in drinking water for 2 weeks postsurgery. Health status was evaluated daily throughout the follow-up period. Rats were allowed to drink and feed ad libitum and sacrificed at 12 weeks postoperatively via CO2. Lumbar spine segments were harvested en bloc for fusion evaluation.

Fusion assessment—manual palpation

Gross explanted motion segments were inspected and tested via manual palpation and confirmed by two researchers blinded to experimental groups. Manual palpation is a binary measure of fusion success that is tested by bending the index level motion segment in both the sagittal and coronal planes, as previously described [18,19]. No motion at the segment was determined as fusion success, whereas any motion was deemed fusion failure.

Fusion assessment–μCT

Ex vivo microcomputed tomography (μCT, vivaCT 40, Scanco USA Inc.) was performed on explanted spines to determine the microarchitectural properties of the fusion mass. The fused segment was contoured and sectioned from soft tissue using a binary thresholding procedure. Direct three-dimensional morphometry was used to determine – volume of mineralized bone tissue (BV, mm3), bone volume fraction (BV/total volume [TV]), bone mineral density (BMD, HA/cm3), structural model index, trabecular thickness, number, and connectivity density. These measurements were calibrated to a commercially available μCT phantom containing hydroxyapatite (HA).

Protein extraction and quantification

Seven days postsurgery spinal fusion sites from the short-term cohort were explanted and surrounding soft tissue was removed. To extract proteins, each sample was lysed using RIPA buffer supplemented with Halt protease inhibitor cocktail (Thermo Scientific, USA) and homogenized using scissors. Samples were incubated on ice for 15 minutes and centrifuged at 13000 g for 15 minutes and the protein containing supernatant was collected. Total protein was determined using BCA protein assay (Thermo Fisher Scientific, USA). Equal concentration of total protein (27 mg/mL) was used for subsequent assay analyses and further diluted to achieve best detection efficiency. Quantification of growth factor levels in the protein extracts was performed using Quantikine ELISA kits for Mouse/Rat PDGF-AB, Mouse/Rat IGF-I, Rat VEGF, and Mouse/Rat/Porcine/Canine TGF-beta 1 (R&D Systems, USA). The following dilutions were used for assay performance: 1:10 for PDGF-AB, 1:500 for IGF-I, 1:100 for VEGF, and 1:2 for TGF-beta 1. For quantitative determination of the preinflammatory proteins IFN-γ, IL-1β, IL-4, IL-5, IL-6, KC/GRO (a CXC chemokine that is known to play a role in inflammation), IL-10, IL-13, and TNF-α, the V-PLEX Proinflammatory Panel 2 Rat Kit (Meso Scale Discovery, USA) was employed. Samples were diluted 1:3 for assay performance.

Statistical analysis

Data were analyzed using SAS software (SAS, 9.3 2002–2010 by SAS Institute Inc., Cary, NC, USA) with significance set at p<.05. Continuous data are presented as mean±standard deviation (SD) across BBDP versus BBDR groups and were evaluated with Student t test applying Satterthwaite corrections for unequal variance with adjustments for unequal sample size when appropriate. A repeated measure ANOVA (GLM, MIXED) was employed to test the relationships among blood measures of diabetes (BG and HbA1c) over time intervals as a function of rat strain. Tukey test was utilized to determine significance for intergroup and pairwise comparisons. Frequency data of manual palpation fusion (fused=1, not fused=0, dependent variable) were tested between groups via Fishers Exact test (Freq procedure, “Exact”) and reported as rates. Inter-relationships among covariates (BG, HbA1c, rat weight) evaluated with Pearson Correlation. ANOVA (GLM) and Regression (REG) were employed to evaluate and compare the slopes of Micro CT variables.

Results

Three of the 42 rats died before sacrifice, resulting in an overall mortality rate of 7.1%. All three deaths occurred within the diabetic BBDP group (BBDP 3/15, 20% vs. BBDR 0/16 p=.10, NS) and were either due to an overt diabetic status (n=1) or postsurgical infection (n=2). All data are presented as mean and standard deviation (SD) for the long-term cohort (BBDP: n=12 and BBDR: n=16, Table 1), except for growth factor and proinflammatory cytokine data which are reported for the short-term cohort of rats (BBDP: n=7 and BBDR: n=3, Table 2) that were sacrificed 1-week postsurgery.

Table 1.

Physical and biomechanical parameters for BBPD biobred diabetic prone rats vs. BBDR biobred diabetic resistance (control) rats − long-term cohort

Measurement BBDP n=12 BBDR n=16 p Value
Characteristics of rats
Intake weight (g) 82.7±9.7 100.2±12.7 p<.001
At surgery weight (g) 343.6±13.3 373.3±21.3 p<.001
Sacrifice weight (g) 378.6±22.3 414.4±23.0 p<.001
Intake BG (mg/dL) 122.6±10.2 131.8±24.7 NS
At surgery BG (mg/dL) 257.9±119.1 146.6±21.2 p<.01
Sacrifice BG (mg/dL) 404.7±113.1 203.8±130.7 p<.01
Sacrifice BG≥250 mg/dL (%) 91.7% 25.0% p<.001
At surgery HbA1c (%) 9.1 ±2.2 5.3±0.2 p<.001
Sacrifice HbA1c (%) 8.2±2.1 4.8±0.2 p<.0001
Sacrifice HbA1c >6.0% (%) 83.3% 0% p<.0001
Outcome variables
Manual palpation Cconfirmed fusion (#/n, %) 2/12, 16.7% 9/16, 56.3% p<.05
MicroCT
Apparent BMD (HA/cm3) 448.2±43.5 519.1±44.4 p<.001
Connectivity density (mm3) 5.0±0.9 3.7±1.6 NS
Bone volume fraction (BV/TV) 0.60±0.05 0.69±0.04 p<.0001
Bone volume (BV, mm3) 498.8±118.4 649.0±109.0 p<.01
Total volume (TV, mm3) 829.7±167.0 943.2± 142.4 NS
Structural model index (SMI) −2.9±0.9 −4.9±2.2 p<.01
Trabecular number (Tb.N) 1.7±0.4 1.6±0.3 NS
Trabecular thickness (Tb.Th) 0.5±0.1 0.6±0.1 p<.0001
Trabecular separation (Tb.Sp) 0.8±0.2 0.9±0.3 NS
Open in a new tab

Mean±SD. or#/n, %

Values given as mean±SD across rat strain groups (BBDP vs. BBDR), NS, Not Significant; SAS-t-test procedure, Satterthwaite utilized for unequal variance.

Table 2.

Growth factors and inflammatory markers for BBPD biobred diabetic prone rats vs. BBDR biobred diabetic resistance (control) rats measured in the short term cohort

Growth factors (pg/mL) BBDP n=7 BBDR n=3 p Value
IGF 3.2±0.3 3.8±0.9 NS
PDGF 20.4±9.5 27.7±5.7 NS
TGF-β 11.8±2.9 11.2±2.9 NS
VEGF 730.9±128.1 743.3±103.5 NS
Inflammatory cytokines (pg/mg)
IFN-γ 4.1±0.9 2.9±0.6 NS
IL-1B 1699.2±601.7 104.8±64.7 p<.001
IL-4 0.61±0.04 0.60±0.01 NS
IL-5 10.8±0.3 10.0±0.3 p<.01
IL-6 2902.5±1533.9 1306.3±477.0 NS
IL-10 12.8±1.5 8.6±0.6 p<.01
IL-13 1.7±0.1 1.6±0.1 NS
KC/GRO 436.6±67.1 123.8±55.6 p<.001
TNF-alpha 41.1±11.5 11.3±3.3 p<.01
Open in a new tab

Mean±SD.

Values given as mean±SD across rat strain groups (BBDP vs. BBDR), NS, Not Significant; SAS-t-test procedure, Satterthwaite utilized for unequal variance.

Intake baseline weights were significantly different between the two groups (BBDP: 82.7±9.7 g vs. BBDR: 100.2±12.7 g) p<.001. BBDR rats also had significantly greater weight at both the surgical and sacrifice timepoints, p<.001 (Table 1).

Diabetic status (BG and HbA1c)

Intake BG levels were not significantly different between the diabetic BBDP (132.5±22.2 mg/dL) and control BBDR (128.6±14.6 mg/dL) groups. BG measured at both the surgical and sacrifice timepoints were significantly higher among diabetic BBDP rats compared with BBDR controls (Fig. 1).

Fig. 1.

Fig. 1.

Open in a new tab

Blood glucose levels measured at surgery (left) and sacrifice (right) was significantly higher in BBDP rats compared with control BBDR rats. The dotted line indicates the cut-off for diabetic status (BG>250 mg/dL). Box and whisker plots with whiskers representing min to max, box represents 25th to 75th percentile, and the line is plotted as the median.

HbA1c levels were also significantly higher among BBDP rats compared with BBDR at the surgery (9.1±2.2% vs. 5.3±0.2%, p<.001) and sacrifice (8.2±2.1% vs. 4.8± 0.2%, p<.0001) timepoints. HbA1c levels at both timepoints were highly correlated (R=0.95, p<.0001). At sacrifice 83.3% of BBDP rats had HbA1c levels greater than 6.0% threshold for diabetes and none of the BBDR controls levels were greater than 6.0%, p<.0001. For the short-term cohort (sacrificed 1-week postsurgery), mean HbA1c (BBDP 8.7±2.16 vs. BBDR 5.3±0.06%, p<.05) and BG (BBDP 268.3±72.5 vs. BBDR 135.0±38.7, p<.05) values measured at surgery were significantly higher in BBDP diabetic compared with BBDR control rats (Fig. 2).

Fig. 2.

Fig. 2.

Open in a new tab

HbA1c levels were significantly higher in diabetic BBDP rats compared to BBDR-controls at both surgery (left) and sacrifice (right), p<.001. The dotted line indicates the cut-off for diabetic status (HbA1c>6%). Box and whisker plots with whiskers representing min to max, box represents 25th to 75th percentile, and the line is plotted as the median.

Manual palpation

The number of fusions as determined by manual palpation were significantly lower in the BBDP group (2/12, 16.7%; confidence interval: 2.1%–48.4%) compared with BBDR rats (9/16, 56.3%: confidence interval: 30.0%–80.25%), p<.05 (Fig. 3). Manual palpation fusion rate was negatively correlated to BG levels measured at sacrifice (r=−0.42, p<.05).

Fig. 3.

Fig. 3.

Open in a new tab

Manual palpation results plotted with 95% confidence intervals. BBDP diabetic rats had a significantly lower rate of fusion comparted to BBDR controls.

Micro CT measures

Average BMD as determined by micro CT was significantly lower in the diabetic BBDP group (448.2±43.5 HA/ cm3) compared with controls (519.1±44.4 HA/cm3) and was significantly negatively correlated to HbA1c when both groups were combined, p<.001 (Fig. 4). There was no significant difference in mean TV (Table 1). Average BV and BV/TV were both significantly lower in BBDP rats compared with BBDR control rats and were significantly correlated to BG and HbA1c measured at sacrifice (Fig. 5). The average structural model index was significantly greater in BBDP rats (−2.9±0.9) than in BBDR rats (−4.9±2.2), p<.01. Mean trabecular thickness was significantly lower in BBDP (0.5±0.1 mm) than in BBDR (0.6± 0.1 mm), p<.0001. Connective density, trabecular number, and trabecular separation were not significantly different between groups (Table 1).

Fig. 4.

Fig. 4.

Open in a new tab

Bone mineral density (BMD) as determined by micro CT was significantly lower in BBDP diabetic rats (red) compared with BBDR controls (red) and was negatively correlated to HbA1c levels measured at sacrifice when the two groups were combined (r=−0.58, R2 =0.30, slope=−11.9), p<.001. Box and whisker plots with whiskers representing min to max, box represents 25th to 75th percentile, and the line is plotted as the median.

Fig. 5.

Fig. 5.

Open in a new tab

Bone Volume (BV) and bone volume density (BV/TV) were both significantly negatively correlated to HbA1c levels measured at sacrifice. (BBDR in blue, BBDP in red).

Growth factor and cytokine analysis

Eleven rats were sacrificed 1-week postsurgery for growth factor and proinflammatory cytokine ELISA analysis. One BBDR specimen was removed from short-term data set due to inflammatory factor levels that were 20 times the standard deviation of the cohort, while the remaining ten (BBDP: n=7, BBDR: n=3) were analyzed. There were no significant differences in any of the growth factors measured (IGF, PDGF, TGF-β, and VEGF) in the fusion bed between the two groups, Table 2. IGF was significantly negatively correlated to BG (r=−0.69, p<.05) (Fig. 6). Proinflammatory factors IL-1β, IL-5, IL-10, KCO/GRO, and TNF-α, were all significantly upregulated in fusion beds of diabetic BBDP compared with BBDR controls (Table 2). There were no significant differences in IFN-γ, IL-4, IL-6, or IL-13 levels between the two groups, although IFN-γ was significantly positively correlated to BG (r=0.76, p<.01) and HbA1c (r=0.67, p<.05) levels (Fig. 7).

Fig. 6.

Fig. 6.

Open in a new tab

Insulin like Growth Factor (IGF) was significantly negatively correlated to blood glucose, (r=−0.69, p<.05). Blue dots represent control (BBDR) and red dots represent diabetic (BBDP) values.

Fig. 7.

Fig. 7.

Open in a new tab

IFN-γ as a function of (A) Blood glucose (mg/dL) and (B) HbA1c (%). Blue dots represent control (BBDR) and red dots represent diabetic (BBDP) values.

Discussion

This study demonstrates that insulin dependent DM reduces the rate of fusion consolidation and quality of the formed fusion mass in an experimental rodent model. Compared to healthy rats undergoing spinal fusion, rats with IDDM demonstrated a threefold reduction in fusion as determined by manual palpation. Furthermore, diabetic rats had significantly lower measures of bone quality as determined by micro CT. Investigation of local cytokine and growth factor regulation in the fusion beds of IDDM rats postsurgery demonstrated a significant elevation of IL-1B, IL-5, IL-10, TNF-α, and KC/GRO, while levels of IGF, PDGF, TGF-β, VEGF IFN-γ, IL-4, IL-13 remained unchanged.

The reduction in fusion success observed in diabetic rats is consistent with existing clinical literature demonstrating reduced fusion rates in diabetic patients undergoing spinal fusion surgery. In a study of 43 insulin-dependent diabetic patients and 43 normal controls, Glassman et al. reported a nonunion rate of 26% in insulin-dependent diabetic patients versus 5% in nondiabetic patients undergoing lumbar spinal fusion [5]. Previous studies report diabetes as a risk factor for increased incidence of complications or poor surgical outcomes following lumbar spine surgery including infection, pseudarthrosis, and poor symptomatic improvement after discharge [5–13]. Likewise, a large nationwide database study of patients undergoing lumbar fusion concluded diabetes was associated with increased risk for postoperative complications, including nonroutine discharge, increased hospital charges, and increased length of stay [20]. However, since ICD-9 codes were used, it was impossible to control for specific variables important to fusion outcomes, including body mass, tobacco use, number of levels fused, and method of arthrodesis. In contrast to these studies, Bendo et al. reported on a series of 32 diabetic patients treated by instrumented spinal fusion and found results comparable to nondiabetic patients [21]. The detected differences in the fusion success in our rat model seem to be IDDM-specific, since no differences in fusion rates were detected in a prior study of our lab that investigated spine fusion of NIDDM rats versus control rats under similar experimental conditions [17].

Clinical studies have specifically identified IDDM as a risk factor for perioperative complication compared with NIDDM in patients undergoing posterior instrumented fusion [5–8,14]. In the present study, oral antibiotics were prophylactically administrated to all animals after surgery, since we experienced a high rate of mortality and postoperative infection in our NIDDM study [17]. This modification in postoperative care likely decreased the mortality rate in the present study. Therefore, we cannot compare infection rates between these two studies. The detected significant reduction in BMD, BV, structural mechanical index, and trabecular thickness is consistent with existing literature that demonstrates decreased BMD and trabecular density in insulin dependent diabetic humans and rats [22–24]. Insulin-dependent diabetic humans have an increased risk for osteoporotic fractures compared with controls (Relative risk=12.25) as do non–insulin-dependent diabetics (Relative risk=1.7). Although IGF-1, an osteogenic growth factor that promotes bone healing, is reduced in IDDM patients with poor glycemic control [22–24], we did not detect a statistically significant difference in the IGF-1 levels of diabetic versus control assayed from tissue explants.

A significant increase in IL-1β, IL-5, IL-10, TNF-α, and KC/GRO (CXCL1) levels were detected locally in the fusion beds of IDDM rats, while IFNγ, IL-4, IL-6, and IL-13 as well as growth factor levels of IGF, PDGF, TGF-β, VEGF remained unchanged between IDDM and control rats at day 7 postsurgery. The observed local increase of cytokines, such as IL-1β, TNF-α, IL-5, and IL-10 as well as lack of increase of IFNγ, IL-4, IL-6, and IL-13 correlates with a systemic increase of these proteins, detected in blood samples of animal models of IDDM as well as humans suffering from IDDM compared with healthy individuals [4,25,26]. Therefore, the elevated cytokine levels in the fusion site of IDDM rats compared with control rats may be rather a result of systemic cytokine dysregulation than a local response only. Although an inflammatory reaction involving immune cells and molecular factors is activated immediately in response to tissue damage and is thought to initiate the repair cascade, a strong increase in proinflammatory cytokines has been shown to delay or inhibit bone healing [27]. For example, IL-1β treatment has been demonstrated to inhibit bone formation both in vivo [28] and in vitro [29] via mechanisms including inhibition of multiple osteoblast functions including protein synthesis, cell replication, and alkaline phosphatase expression. In the present study, we hypothesized that the increased levels of cytokines negatively affect spinal fusion healing success, potentially via mechanisms involving local cells such as osteoblasts.

In line with our study, an upregulation of IL-10 in plasma samples of type 1 diabetic patients has been described [4,30]. Although the role of systemic IL-10 dysregulation is not fully understood, it is assumed that the elevated levels of IL-10 are a result of a compensatory mechanism to the increased proinflammatory cytokines. The effect of the increased IL-10 levels on bone healing could be beneficial or detrimental. Although IL-10 is known to promote osteogenesis via P38/MAPK pathway in its role as an anti-inflammatory cytokine, higher IL-10 doses were shown stimulate inflammatory NF-kB signaling pathways [31]. Future studies are needed to evaluate the role of IL-10 on bone healing in type I diabetic patients in more detail.

Although the association of elevated levels of IL-1β, IL-5 IL-10, and TNF-α levels and type I diabetes has been described in the literature, publications on the role and regulation of KC/GRO (CXCL1), a class of chemotactic factors known to stimulate inflammation and tumor growth, is rather sparse in this context. A study by Takahashi et al. proposed elevated serum CXCL1 is a good marker of type I diabetes after demonstrating elevated serum KC/GRO (CXCL1) concentrations in type 1 diabetic patients [32]. Furthermore, inhibition of chemokine receptor CXCR1/2 has shown to block and reverse type 1 diabetes in mice [33]. Since CXCL1 has also been described to accelerate osteoclast maturation and promote bone osteolysis [34], this protein may be another interesting target for future investigation of bone regeneration in type 1 diabetics.

The lack of growth factor regulation at the fusion site of our present study suggests that IGF, PDGF, TGF-β, and VEGF are either not involved in the mechanisms that resulted in the detected reduced bone quality and fusion rate or may be regulated at a different time point postsurgery that was not tested.

The BBDP rat model used in this study is increasingly preferred compared with traditional streptozotocin-induced models of IDDM since the systemic toxicity of streptozotocin makes long-term evaluation of bone healing difficult. BBDP rats develop the diabetic phenotype during adolescence in a condition that more closely mimics the human condition of IDDM, as the rats spontaneously develop inflammation of pancreatic islet cells followed by selective destruction and overt diabetes [35]. This present study is the first to analyze the effect of IDDM on the unique healing environment that occurs during spinal fusion [36,37] using a well-validated rodent model of spinal fusion [17,38,39]. This study has several limitations, including the generalizability of a rodent model to a human population. Furthermore, literature regarding human IDDM does not typically specify whether patients had juvenile onset and thus do not distinguish between autoimmune DM versus poorly controlled NIDDM that progressed to IDDM.

In summary, this is the first study to evaluate the rate and quality of spinal fusion in a translational animal model of IDDM that allows control for other surgical- and patient-related risk factors. Retrospective clinical data have demonstrated an increased risk of complication in lumbar spinal fusion surgery in patients with IDDM. Similarly, our data show that rats with IDDM demonstrate a lower rate of fusion consolidation and a lower quality fusion mass along with increased levels of inflammatory cytokines. These findings demonstrate the complex nature of IDDM and highlight the need for additional basic science research to further elucidate the mechanism of impaired bone healing to develop more effective therapeutic intervention modalities for this growing, high-risk patient population.

Acknowledgment

This study was funded in part by a 2014 Scoliosis Research Society Grant.

Footnotes

FDA device/drug status: Not applicable

Author disclosures: ZN: Nothing to disclose. LEAK: Nothing to disclose. TJN: Nothing to disclose. KS: Nothing to disclose. YA: Nothing to disclose. JDG: Nothing to disclose. DS: Nothing to disclose. MFM: Nothing to disclose.

References

  • [1].Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA 2015;314(10):1021–9. [DOI] [PubMed] [Google Scholar]
  • [2].Ogurtsova K, da Rocha Fernandes J, Huang Y, Linnenkamp U, Guar-iguata L, Cho N, et al. IDF Diabetes atlas: global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract 2017;128:40–50. [DOI] [PubMed] [Google Scholar]
  • [3].Fatima N, Faisal SM, Zubair S, Amjal M, Siddiqui S, Moin S, et al. Role of pro-inflammatory cytokines and biochemical markers in the pathogenesis of Type 1 diabetes: correlation with age and glycemic condition in diabetic human subjects. PLoS One 2016;11(8): e0161548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Alnek K, Kisand K, Heilman K, Peet A, Varik K, Uibo R. Increased blood levels of growth factors, Proinflammatory Cytokines, and Th17 Cytokines in patients with newly diagnosed Type 1 diabetes. PLoS One 2015;10(12):e0142976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Glassman SD, Alegre G, Carreon L, Dimar JR, Johnson JR. Perioperative complications of lumbar instrumentation and fusion in patients with diabetes mellitus. Spine J 2003;3(6):496–501. [DOI] [PubMed] [Google Scholar]
  • [6].Tang H, Zhu J, Ji F, Wang S, Xie Y, Fei H. Risk factors for postoperative complication after spinal fusion and instrumentation in degenerative lumbar scoliosis patients. J Ortho Surgery Res 2014;9(1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Golinvaux NS, Varthi AG, Bohl DD, Basques BA, Grauer JN. Complication rates following elective lumbar fusion in patients with diabetes: insulin dependence makes the difference. Spine 2014;39 (21):1809–16. [DOI] [PubMed] [Google Scholar]
  • [8].Takahashi S, Suzuki A, Toyoda H, Terai H, Dohzono S, Yamada K, et al. Characteristics of diabetes associated with poor improvements in clinical outcomes after lumbar spine surgery. Spine 2013;38 (6):516–22. [DOI] [PubMed] [Google Scholar]
  • [9].Veeravagu A, Patil CG, Lad SP, Boakye M. Risk factors for postoperative spinal wound infections after spinal decompression and fusion surgeries. Spine 2009;34(17):1869–72. [DOI] [PubMed] [Google Scholar]
  • [10].Browne JA, Cook C, Pietrobon R, Bethel MA, Richardson WJ. Diabetes and early postoperative outcomes following lumbar fusion. Spine 2007;32(20):2214–9. [DOI] [PubMed] [Google Scholar]
  • [11].ter Gunne AFP, Cohen DB. Incidence, prevalence, and analysis of risk factors for surgical site infection following adult spinal surgery. Spine 2009;34(13):1422–8. [DOI] [PubMed] [Google Scholar]
  • [12].Simpson JM, Silveri CP, Balderston RA, Simeone FA, An HS. The results of operations on the lumbar spine in patients who have diabetes mellitus. JBJS 1993;75(12):1823–9. [DOI] [PubMed] [Google Scholar]
  • [13].Freedman MK, Hilibrand AS, Blood EA, Zhao W, Albert T, Vacarro A, et al. The impact of diabetes on the outcomes of surgical and nonsurgi-cal treatment of patients in the spine patient outcomes research trial. Spine 2011;36(4):290–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Deyo RA, Hickam D, Duckart JP, Piedra M. Complications following surgery for lumbar stenosis in a veteran population. Spine 2013;38 (19):1695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Bortell R, Yang C. The BB rat as a model of human type 1 diabetes. Methods Mol Biol 2012;933:31–44. [DOI] [PubMed] [Google Scholar]
  • [16].Metzger MF, Kanim LE, Zhao L, Robinson ST, Delamarter RB. The relationship between serum vitamin D levels and spinal fusion success: a quantitative analysis. Spine (Phila Pa 1976) 2015;40(8): E458–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Bhamb N, Kanim LE, Maldonado RC, Nelson T, Salehi K, Glaeser J, et al. The Impact of Type 2 Diabetes on bone metabolism and growth after spinal fusion. Spine J 2018. [DOI] [PubMed] [Google Scholar]
  • [18].Bae HW, Zhao L, Kanim LE, Wong P, Marshall D, Delamarter RB. Bone marrow enhances the performance of rhBMP-2 in spinal fusion: a rodent model. J Bone Joint Surg Am 2013;95 (4):338–47. [DOI] [PubMed] [Google Scholar]
  • [19].Robinson ST, Svet MT, Kanim LA, Metzger MF. Four-point bending as a method for quantitatively evaluating spinal arthrodesis in a rat model. Comp Med 2015;65(1):46–50. [PMC free article] [PubMed] [Google Scholar]
  • [20].Browne JA, Cook C, Pietrobon R, Bethel MA, Richardson WJ. Diabetes and early postoperative outcomes following lumbar fusion. Spine 2007;32(20):2214–9. [DOI] [PubMed] [Google Scholar]
  • [21].Bendo JA, Spivak J, Moskovich R, Neuwirth M. Instrumented posterior arthrodesis of the lumbar spine in patients with diabetes mellitus. Am J Ortho 2000;29(8):617–20. [PubMed] [Google Scholar]
  • [22].AboElAsrar MA, Elbarbary NS, Elshennawy DE, Omar AM. Insulinlike growth factor-1 cytokines cross-talk in type 1 diabetes mellitus: relationship to microvascular complications and bone mineral density. Cytokine 2012;59(1):86–93. [DOI] [PubMed] [Google Scholar]
  • [23].Moyer-Mileur LJ, Slater H, Jordan KC, Murray MA. IGF-1 and IGF-binding proteins and bone mass, geometry, and strength: relation to metabolic control in adolescent girls with type 1 diabetes. J Bone Miner Res 2008;23(12):1884–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Van Sickle BJ, Simmons J, Hall R, Raines M, Ness K, Spagnoli A. Increased circulating IL-8 is associated with reduced IGF-1 and related to poor metabolic control in adolescents with type 1 diabetes mellitus. Cytokine 2009;48(3):290–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Qiao YC, Chen YL, Pan YH, Tian F, Xu Y, Zhang X, et al. The change of serum tumor necrosis factor alpha in patients with type 1 diabetes mellitus: a systematic review and meta-analysis. PLoS One 2017;12(4):e0176157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Talebi-Garakani E, Safarzade A. Resistance training decreases serum inflammatory markers in diabetic rats. Endocrine 2013;43 (3):564–70. [DOI] [PubMed] [Google Scholar]
  • [27].Claes L, Recknagel S, Ignatius A. Fracture healing under healthy and inflammatory conditions. Nat Rev Rheumatol 2012;8(3):133–43. [DOI] [PubMed] [Google Scholar]
  • [28].Nguyen L, Dewhirst F, Hauschka P, Stashenko P. Interleukin-1 beta stimulates bone resorption and inhibits bone formation in vivo. Lym-phokine Cytokine Res 1991;10(1–2):15–21. [PubMed] [Google Scholar]
  • [29].Stashenko P, Dewhirst FE, Rooney ML, Desjardins LA, Heeley JD. Interleukin-1 β is a potent inhibitor of bone formation in vitro. J Bone Mineral Res 1987;2(6):559–65. [DOI] [PubMed] [Google Scholar]
  • [30].Chatzigeorgiou A, Harokopos V, Mylona-Karagianni C, Tsouvalas E, Aidinis V, Kamper EF. The pattern of inflammatory/anti-inflammatory cytokines and chemokines in type 1 diabetic patients over time. Ann Med 2010;42(6):426–38. [DOI] [PubMed] [Google Scholar]
  • [31].Chen E, Liu G, Zhou X, Zhang W, Wang C, Hu D, et al. Concentration-dependent, dual roles of IL-10 in the osteogenesis of human BMSCs via P38/MAPK and NF-kappaB signaling pathways. FASEB J 2018;32(9):4917–29. [DOI] [PubMed] [Google Scholar]
  • [32].Takahashi K, Ohara M, Sasai T, Homma H, Nagasawa K, Takahashi T, et al. Serum CXCL1 concentrations are elevated in type 1 diabetes mel-litus, possibly reflecting activity of anti-islet autoimmune activity. Diabetes Metab Res Rev 2011;27(8):830–3. [DOI] [PubMed] [Google Scholar]
  • [33].Citro A, Valle A, Cantarelli E, Mercali A, Pellegrini S, Liberati D, et al. CXCR1/2 inhibition blocks and reverses type 1 diabetes in mice. Diabetes 2015;64(4):1329–40. [DOI] [PubMed] [Google Scholar]
  • [34].Hardaway AL, Herroon MK, Rajagurubandara E, Podgorski I. Marrow adipocyte-derived CXCL1 and CXCL2 contribute to osteolysis in metastatic prostate cancer. Clin Exp Metastasis 2015;32(4):353–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].King AJ. The use of animal models in diabetes research. Br J Pharmacol 2012;166(3):877–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Boden SD. Biology of lumbar spine fusion and use of bone graft substitutes: present, future, and next generation. Tissue Eng 2000;6 (4):383–99. [DOI] [PubMed] [Google Scholar]
  • [37].Morone MA, Boden SD, Hair G, Martin G, Racine M, Titus L, et al. The Marshall R. Urist Young Investigator Award. Gene expression during autograft lumbar spine fusion and the effect of bone morphogenetic protein 2. Clin Orthop Relat Res 1998(351):252–65. [PubMed] [Google Scholar]
  • [38].Glaeser J, Behrens PH, Salehi K, Kanim L, Sheyn D, NaPier Z, et al. Wednesday, September 26, 2018 7: 35 AM-9: 00 AM ePosters: P1. NF-κB inhibitor suppresses edema and promotes rhBMP-2-mediated bone formation in spinal fusion. Spine J 2018;18(8):S142. [Google Scholar]
  • [39].Kang JR, Glaeser J, Karamian BA, Kanim L, Koltsov J, Thio T, et al. The effects of varenicline (Chantix) on lumbar spinal fusion in a rat model. Spine J 2018;18(8):S117–S8. [DOI] [PubMed] [Google Scholar]

RESOURCES