Diabetes-related screw loosening: the distinction of surgical sites and the relationship among diabetes, implant stabilization and clinical outcomes

doi:10.21203/rs.3.rs-1970866/v1

Introduction.

Diabetes mellitus (DM) is correlated with poor clinical outcomes in spinal surgery. However, the effect of it on screw stabilization has not been investigated. The aim of this study was to evaluate the screw loosening rate and postoperative outcomes in diabetic patients and to identify potential risk factors associated with loosening.

Methods.

Two hundred and forty-three patients who received cervical or lumbar internal fixation between 2015 and 2019 were enrolled in this study. Screw loosening was assessed on radiography, and clinical outcomes were evaluated by the improvement of Visual Analog Scale (VAS), Oswestry Disability Index (ODI) or Japanese Orthopaedic Association (JOA) scores. The relationship of DM, screw loosening and clinical outcomes were analyzed with chi-square tests and regression analyses.

Results.

One hundred and twenty-two patients (50.2%) with diabetes were included in this study. Diabetes led to the increase of the rate of screw loosening in the lumbar spine, while the loosening rate did not vary significantly in the cervical spine. The occurrence of screw loosening in the lumbar spine was more likely to be associated with clinical outcomes for motor performance including walking and sitting. However, no significant effect on JOA and VAS scores in the cervical spine of screw loosening was found. Moreover, the history of DM affected the outcomes of the patients who underwent spinal surgery.

Conclusions.

DM had an adverse effect on screw stabilization. The impaired improvement of clinical outcomes in diabetics after spinal surgery was related to screw loosening. In addition to the direct effects on operative wounds and neural function, the impact on the screws due to DM was also worth noting.

Diabetes mellitus

Screw loosening

Spine surgery

Clinical outcomes

In the past few decades, the rapid development of spinal internal fixation has made it a standard of surgical treatment in orthopedics. Described by King in 1944 and introduced by Roy-Camille et al. for lumbar spinal fusion in 1986 [1], the use of screws has become increasingly widespread in spinal surgery for various vertebral diseases. The accuracy and safety of screws were well established [2–4], while complications were also indicated intraoperatively or postoperatively. The common complications of internal fixation include deep wound infection, screw loosening, breakage, et al.[5, 6]. These complications result in the increasing likelihood of negative outcomes. Screw loosening is one of severe complications which may occur postoperatively in patients, leading to adverse consequences like back pain, deformity and disability [6]. The risk factors of implant failure were considered multifactorial. In a literature review of 43 articles, Galbusera et al. reported that the rate of screw loosening ranged from 1–15%, and even to 60% in osteoporotic patients [7].

The global prevalence of diabetes mellitus (DM) among adults increased from 4.7% in 1980 to 8.5% in 2014, giving rise to heavy social burden[8]. Meanwhile, it also caused a considerable number of complications with harmful influences on patients. A growing number of studies have illustrated that DM posed adverse effect on the quality of bone in normal or even increased bone mineral density (BMD) situations [9]. Many studies confirmed that DM prominently affected dental implant stabilization [10] and elevated overall risk of revision for aseptic loosening in total joint arthroplasty [9]. The failure of serum glucose control in DM patients who took pedicle-based dynamic stabilization system treatment showed a higher visual analogue scale (VAS) score, presumably due to postoperative instability of pedicle screws [11]. Apart from negative impact on bone quality, DM also adversely affected bone regeneration and healing process at bone-implant interface, thus it could lead to higher implant failure risk [12]. Provided that early spinal screw loosening is often asymptomatic, little attention was paid to by patients and relatives.

To the best of authors’ knowledge, there is few systematic follow-up study on screw loosening of DM patients. Therefore, the purpose of this study is to assess the incidence of screw loosening and the impact of loosening on clinical outcomes in patients with DM, and then to further analyze potential risk factors for screw loosening.

Patient enrollment and surgical information collection

This was a retrospective study included 122 patients with DM and 121 normal patients who once accepted spine internal fixation operation from 2015 to 2019. The diagnosis of DM was made according to the guidelines for classification and diagnosis of diabetes formulated by the American Diabetes Association. Patients who had degenerative disc disease (DDD), received spine surgery with internal fixation and agreed to participant in the investigation and had complete imaging data were included in our study. The age of patients was 31 to 77 years old and the follow-up ranged between 12–66 months. Data were obtained from the medical platform of Xijing Hospital. All surgeries were carried out by two same experienced surgeons from the Orthopedics Department of Xijing Hospital. The following procedural codes were included: anterior cervical discectomy and fusion (ACDF, the instrument produced by Medtronic, USA), posterior cervical fusion (PCF, the instrument produced by WEGO, China) and posterior lumbar interbody fusion (PLIF, the instrument produced by WEGO or SANYOU, China). Patients who did not undergo preoperative evaluation or did not cooperate with the follow-up were excluded. Besides, patients who presented with osteoporosis, Parkinson’s disease, traumatic, tumoral and infectious spinal surgery and history of severe fractures were also excluded.

Clinical evaluations

In consideration of the potential effect of DM on the fixation of screws, serum glucose levels were measured preoperatively and postoperatively for all patients. Besides, hemoglobin A1c (HbA1c) was evaluated in DM patients. Meanwhile, prospective collections of scores for Oswestry Disability Index (ODI), Japanese Orthopaedic Association (JOA) and visual analogue scale (VAS) were conducted. Moreover, general conditions of the patients including age, height, weight, operation note, and smoking status, were also investigated. VAS was measured on a scale ranging from 0 (no pain) to 10 (severe pain), and the assessment of VAS scores improvement was set was according to the reduction of scores. Besides, the VAS of low back pain was carried out by sit-to-stand test and passive lumbar extension test, and the average score was calculated. Total points of ODI were set on a 0-to-100- point scale, while subitems of ODI were scored ranging from 0 to 5 points. The improvement of ODI scores was valued as the change between preoperative scores and postoperative scores. Moreover, the change of JOA scores was defined as JOA improvement rate according to the formula: rate = (postoperative JOA-preoperative JOA) / (17-preoperative JOA) *100%. The improvement of JOA scores was defined as whether the recovery was complete (the postoperative JOA scores fully achieved 17).

Radiological evaluations

Screw loosening or dislocation was evaluated on X-rays with posterior-anterior positions, which were received from each patient and saved in the form of photographs. The measurements were conducted postoperatively at the last follow-up. The parameter to determine the screw loosening was ‘double halo’, that is, the breach increments as an inner zone around the implant surrounded by the outer rim of vertebral cancellous bone. The measurements of sagittal alignment during radiographic assessment were lumbar lordosis (LL), pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), sagittal vertical axis (SVA), pelvic incidence minus lumbar lordosis (PI-LL) and Global Alignment and Proportion (GAP) Score. The GAP scores were calculated from obtained data according to the original article[13]. The quantitative measurements were performed using Sante DICOM software (Santesoft ltd., Nicosia, Cyprus), Digimizer (MedCalc Software ltd, Belgium) and ImageJ (National Institutes of Health, Maryland, USA). In this study, we selected a minimum thickness of 1mm as the criterion of screw loosening. The radiographic evaluation was carried out by two independent surgeons and one radiologist.

Statistical analysis

The data were analyzed with SPSS 20.0 Software (SPSS Inc, Chicago, USA) and STATA/SE 12.0 (StataCorp LLC, Texas, USA). Independent t-tests and Chi-square tests were used to examine the relationship between diabetes and screw loosening in different surgical sites (cervical spine and lumbar spine) as well as evaluate several potential risks of screw loosening, and the results were presented as mean values ± standard deviation (SD) or odd ratios (OR). Meanwhile, multivariate linear regression analyses were performed to assess the association between demographic or clinical data and ODI scores. In addition, binary Llogistic regression analyses were used to explore the relationship between the aforementioned data and VAS as well as JOA scores. Fleiss kappa coefficients were calculated for intra-observer and inter-observer reliability. The level of significance was set at p < 0.05.

Characteristics of the patients

A total of 243 patients who received degenerative spinal surgery from Jan 2015 to Dec 2019 were enrolled in the present study. In these patients whose complete data were available, 122 had DM and 121 had not. The sample comprised of 119 men and 124 women (with a mean age of 53.53 ± 10.47 years old). The mean follow-up duration was 36.14 ± 23.22 months. All patients with normal bone mineral density received cervical or lumbar internal fixation (Table 1). The complete demographic and patient characteristics grouped by surgical site and surgical invasiveness were shown in Tables 1 and 2.

Table 1

Demographic and Patient Characteristics (N = 243)
Lumbar spine (N = 170)
	Non-Diabetic Group (N = 84)	Diabetic Group (N = 86)	P Value
Age (yr)	52.17 ± 10.98	55.30 ± 8.79	0.128
Sex			0.265
Male	39(46%)	41(48%)
Female	45(54%)	45(52%)
Body mass index	24.13 ± 3.00	25.26 ± 3.34	0.097
Duration (m)	35.65 ± 19.65	34.33 ± 20.16	0.246
Operative time (h)	3.16 ± 0.94	3.87 ± 1.20	0.002*
Smoker (y/n)	10(12%)	11(13%)	0.724
Segment number	2.32 ± 0.58	3.00 ± 1.68	0.003*
Glucose (mmol/L)	5.10 ± 0.93	8.32 ± 2.63	< 0.001*
Bone Mineral Density	-0.60 ± 0.25	-0.57 ± 0.37	0.351
Estimated blood loss (mL)	325.69 ± 171.98	605.33 ± 846.49	0.008*
Undertake transfusion (y/n)	6(7%)	24(28%)	0.006*
Fusion	82(98%)	82(95%)	0.201
Spondylolisthesis	9(11%)	11(13%)	0.270
Loosening rate	6(7%)	16(19%)	0.021*
Cervical spine (N = 73)
	Non-Diabetic Group (N = 37)	Diabetic Group (N = 36)	P Value
Age (yr)	51.86 ± 9.88	54.45 ± 8.49	0.105
Sex			0.012*
Male	15(41%)	24(67%)
Female	22(59%)	12(33%)
Body mass index	24.08 ± 3.24	25.17 ± 2.58	0.200
Duration (m)	39.19 ± 17.43	37.00 ± 19.60	0.278
Operative time (h)	2.21 ± 0.75	2.38 ± 0.97	0.465
Smoker (y/n)	4(11%)	5(14%)	0.926
Segment number	3.03 ± 1.04	3.60 ± 1.19	0.066
Glucose (mmol/L)	5.01 ± 0.72	7.19 ± 2.24	< 0.001*
Bone Mineral Density	-0.72 ± 0.23	-0.66 ± 0.31	0.266
Estimated blood loss (mL)	164.59 ± 219.87	244.50 ± 361.44	0.303
Undertake transfusion (y/n)	1(3%)	2(6%)	0.660
Fusion	36(97%)	34(94%)	0.355
Loosening rate	2(5%)	10(28%)	0.135
*These values were significant at p < 0.05. y/n indicates yes/no.

Table 2

Patient characteristics according to surgical invasiveness
	Anterior cervical discectomy and fusion (N = 51)	Posterior cervical fusion (N = 22)	Lumbar fusion (N = 170)	P Value
Age (yr)	52.67 ± 9.84	53.73 ± 8.88	52.97 ± 10.72	0.053
Sex				0.040*
Male	24(47%)	15(68%)	80(47%)
Female	27(53%)	7(32%)	90(53%)
Body mass index	24.03 ± 2.93	25.68 ± 3.13	24.46 ± 3.15	0.208
Duration (m)	39.38 ± 20.04	36.27 ± 26.73	34.21 ± 20.10	< 0.001*
Operative time (h)	2.02 ± 0.70	2.97 ± 0.75	3.36 ± 1.20	< 0.001*
Smoker (y/n)	6(12%)	3(14%)	21(12%)	0.900
Segment number	2.71 ± 0.74	4.67 ± 0.62	2.52 ± 1.07	< 0.001*
Glucose (mmol/L)	5.80 ± 1.77	5.69 ± 1.83	6.05 ± 2.18	0.704
Estimated blood loss (mL)	99.52 ± 105.85	453.33 ± 419.38	407.94 ± 492.88	< 0.001*
Undertake transfusion (y/n)	0(0%)	3(14%)	30(18%)	0.051
Fusion	50(98%)	20(91%)	164(96%)	0.152
Loosening rate	7(14%)	5(23%)	22(13%)	0.956
*These values were significant at p < 0.05. y/n indicates yes/no.

Radiographic follow-up on loosening

More screw loosening cases were identified in patients with DM than those in patients without DM. Among the 73 patients, radiographically screw loosening was found in 12 patients in the cervical spine (2 in Non-Diabetic Group vs 10 in Diabetic Group). In patients receiving lumbar surgery, the rate of screw loosening in Diabetic Group was higher than that in Non-Diabetic Group (19% vs 7%, respectively, p = 0.021) (Table 1). Screw loosening was valued as at least one breach increment around the screw was over 1mm shown on the X-ray (Fig. 1). The intra-observer and inter-observer reliability according to the kappa value were 0.91 and 0.85, respectively. We calculated the fusion rate, the results indicated that patients without screw loosening could obtain fusion, while 75 percent of the patients with cervical screw loosening could obtain fusion and 73 percent of those with lumbar screw loosening could obtain fusion. This illustrated that the poor fusion is also an important factor of screw loosening (Tables 1 and 2). According to the subgroup analysis, the risk of loosening in the lumbar spine of DM patients was higher than that of Non-DM patients, with statistical significance (p < 0.001) (Fig. 2 and Table S1). Besides, no significant difference was found between the loosening and non-loosening group in the lumbar spine in the main characteristics. In the cervical spine, no significant difference was found in all evaluated factors (Table S1). Moreover, there was no significant differences (p > 0.05) (Table S2) between posterior and anterior approach group in cervical surgery. In patients with DM, several factors including preoperative HbA1c, postoperative HbA1c, fasting blood glucose and postprandial glucose showed no statistically significant influence on screw loosening (p > 0.05) (Table S3).

Clinical evaluations

Multiple logistic regression analysis was performed to evaluate whether VAS and JOA improved. For lumbar surgery, multiple linear regression was conducted to determine ODI change and the risk factors of ODI improvement. The improvement of ODI score was 37.22 ± 28.29 in Diabetic group and 55.23 ± 24.55 in Non-Diabetic group (Table S4), respectively. Moreover, the changes of leg VAS scores in Diabetic Group were lower than those in Non-Diabetic Group (lower limb pain VAS 1.67 ± 2.25 vs 3.85 ± 2.09, low back pain VAS 1.63 ± 1.85 vs 2.88 ± 1.94, p < 0.01) (Table S4). In terms of subitems of ODI scores, the improvement of sleeping, standing, pain and sitting was significantly associated with DM (β, -0.995, -0.752, -1.021, -0.821; p < 0.05) (Table 3). Moreover, diabetic loosening and high BMI had an adverse effect on walking and sitting improvement (β, -0.806, -0.956; -0.129, -0.102; p < 0.05) (Table 3), and intraoperative blood loss and operative time were also significant predictors of patient-reported outcomes. Besides, patients with diabetes showed higher PI-LLs and less changes of GAP scores (Table S7), and the change of GAP score was related with standing improvement (β, -0.873; p < 0.05) (Table 3) as well as low back pain VAS (β, -0.932; p < 0.05) (Table 4). However, no significant effect on ODI improvement in the history of degenerative spondylolisthesis (grade 1) was found (Table 3).

Table 3

Multivariable Regression Analysis of subitems of ODI scores in patients undergone lumbar spine surgery with or without diabetes
	Sleeping	Walking	Standing	Pain	Sitting	Self-care
History of diabetes	-0.995*	-0.122	-0.752*	-1.021*	-0.821*	0.051
Loosening	-0.065	-0.806*	-0.048	-0.102	-0.956*	-0.006
Estimated blood loss (mL)	-0.001	-0.001*	-0.158	-0.001*	-0.041	-0.165
Segment number	-0.059	0.189	-0.099	0.370	0.065	0.362
BMI	0.010	-0.129*	-0.080	-0.103	-0.102*	-0.075
Operative time (h)	-0.051	-0.126	-0.171	-0.204	-0.403*	-0.569*
Spondylolisthesis	0.130	0.306	0.043	0.588	-0.316	0.763
GAP score	-0.087	-0.108	-0.873*	-0.126	-0.425	-0.062
*These values were significant at p < 0.05. The values are given as the beta.

Table 4

Multivariable Regression Analysis of postoperative outcomes in patients undergone spine surgery with or without diabetes
Cervical spine (N = 73)
	VAS (cervical vertebrae)		VAS (upper limbs)	JOA
History of diabetes	-2.367		-1.761	-3.611*
Loosening	-0.124		0.909	-0.620
Undertake transfusion (y/n)	-17.720		30.173	14.947
Lumbar spine (N = 170)
		VAS (lumbar vertebrae)	VAS (lower limbs)	ODI
History of diabetes		-1.975*	-2.201	-0.995*
Loosening		-0.368	-1.339	-0.065
Spondylolisthesis		0.125	-0.359	5.044
GAP score		-0.932*	-0.391	-0.537
Estimated blood loss (mL)		0.002	-0.003	-0.001*
*These values were significant at p < 0.05. y/n indicates yes/no. The values are given as the beta.

Besides, the comparison of arm pain VAS and ODI scores improvement between Non-Diabetic and Diabetic Group in the cervical spine was found (upper limb pain VAS 3.24 ± 2.78 vs 1.25 ± 3.28, JOA improvement rate (94.74 ± 18.79)% vs (59.00 ± 51.08)%, p < 0.05) (Table S5), which indicated that these outcomes were significantly associated with DM. However, there was no statistically significant difference in neck pain VAS scores between Non-Diabetic and Diabetic Group (p > 0.05) (Table S5).

It was indicated by multivariable regression analysis that diabetes was a significant predictor in both cervical and lumbar clinical outcomes. History of diabetes was found to be highly correlated with JOA scores improvement in patients receiving cervical surgery (β, -3.611; p < 0.05) (Table 4) and with low back pain VAS scores as well as ODI scores in those with lumbar surgery (β, -1.975, -0.995; p < 0.05) (Table 4).

It is noteworthy that both degenerative spinal diseases and diabetes mellitus have become widespread suffering for older people and social burden throughout the world. Furthermore, screw loosening has been suggested as one of the most common complications of spine surgery, and it has been widely reported in previous studies. Armaghani et al found that diabetes might be associated with poor postoperative outcomes in arthroplasty and spine surgery [9, 14], while the relationship of diabetes, screw loosening, and outcomes has not been fully investigated yet. Thus, our study was intended to explore the effect of screw loosening on postoperative recovery of patients with diabetes undergoing cervical or lumbar surgery.

In this study, it was indicated that screw loosening occurred at a higher rate in patients with diabetes. This finding was consistent with the results of several studies conducted by Wu et al and Zou et al to some extent [11, 15]. However, previous studies mainly focused on limited portions, such as the effect of diabetes on Dynamic stabilization system [11], or the screw loosening rate in the lumbar spine without detailed symptom scores [14]. Considering that the distribution of age and bone mineral density are similar in both DM and non-DM group, in this study, potential correlations have been found among DM, screw loosening and clinical outcomes. Although screw loosening was regarded as an inevitable problem, and it was reported asymptomatic by a relatively sparse amount of literature previously [16, 17], the effect of loosening on postoperative outcomes gained very little attention. Moreover, so far, few studies involving screw loosening in both cervical and lumbar spine simultaneously have been reported, while the results in our cohort study suggested that diabetes had a relatively larger impact on the lumbar spine than that on the cervical spine.

The risk factors of screw loosening were identified in cervical and lumbar spine. The interference of BMD was excluded, for the patients with osteoporosis or severe fracture history underwent spine surgery using expandable pedicle screw or bone cement augmentation. Our findings demonstrated that the influence of diabetes could be of importance, while other potential factors (age, sex, BMI, operative time, smoker) had little effect on loosening. Besides, there is evidence that the BMD of several diabetic patients was high while the screw loosening occurred, since the bone quality but not the bone density would be essential for screw stabilization. In particular, diabetes-related indicators, such as glucose, preoperative and postoperative HbA1c were found not to be significantly independent predictors for screw loosening as well. This result indicated that the glucose control may lead to better outcomes in screw stabilization, while diabetes had marked influence and the risk of loosening was dependent on the surgical site.

As for bone quality, an ideal test approach lacks at present. As mentioned above, patients with type 2 diabetes mellitus (T2DM) often perform normal even increased BMD [18, 19]. Several studies have shown that high resolution peripheral quantitative computed tomography (HRpQCT) [20–22] and bone material strength index (BMSi) [23–25] could be employed to evaluate bone quality, and our previous studies and some researches also demonstrated that diabetes had adverse effect on bone quality, while medical examination is lacking in the clinical use. Hence, the assessment of bone quality would be one of the research orientations in our future work.

In diabetes, the structural and functional changes in microvasculature often occur[26]. Meanwhile, diabetic microvascular and macrovascular complications are the commonest cause of mortality and morbidity[27]. A previous study on dental implant indicated that the bleeding on probing under diabetes mellitus condition was worse than the normal one[28]. Besides, our research group previously studied the relationship between diabetes and bone vessel and demonstrated that diabetes led to the holistic abnormality of the vascular system in bone[29]. Hence, in our opinion, the vascular conditions of diabetics were not ideal, which may result in longer surgery duration and more blood loss.

Although Silvestre, Huang and Mac-Thiong et al discovered that patients with screw loosening were asymptomatic [6, 16, 17, 30], we drew a relatively distinct conclusion. According to the logistic regression and the multivariable linear regression in our study, it was revealed that the relevance on postoperative outcome improvement for the occurrence of the screw loosening could exert a special impact on several subitems of symptom scores. Given that loosening cases were more common in patients with diabetes, we evaluated symptom scores including VAS, ODI and JOA in this study. With regard to the lumbar spine, generally, VAS scores and ODI scores were affected by diabetes and screw loosening. It was revealed that the low back VAS scores and ODI scores were related to history of diabetes, with statistical significance. DM especially had an adverse effect on the improvement of outcomes including sleeping, sitting, standing and pain, this may be due to the potential damage to microvasculature or neural changes by diabetic condition. Moreover, we found that loosening in the lumbar spine had more adverse effects on walking and sitting than that on sleeping when compared with DM patients without loosening, and sagittal misalignment could affect the postoperative improvement of standing function. This was because the screw fixation and stabilization were mainly correlated to lower limb function. In particular, the lumbar spine played a pivotal role in bearing the body weight regardless of the condition of rest or motion. A biomechnical study conducted by Zhang et al. clarified the compressive loading of vertebral bodies. In different states of motion including flexion, extension, lateral bend and axial rotation, the maximum loads of cervical vertebra were about 250N, while those of lumbar vertebra were 4400N[31]. Hence, the load bearing might be an influence factor of screw stabilization. Interestingly, an association was found between high BMI and postoperative outcomes, which is likely due to the lack of exercise after spine surgery. By contrast, cervical spine underwent more motion rather than load-bearing, which could reduce the adverse effect of screw loosening on the improvement of function. Hence, in addition to the direct effects of diabetes on wounds and nerves, the impact on the internal fixation due to DM is also a critical factor for poor postoperative recovery and decline of functional indicators such as ODI after spinal surgery.

Notably, the mechanisms of diabetes-induced screw loosening have not been fully disclosed yet. The screw stabilization might be related to bone substance. Many studies have shown that patients with type 2 diabetes represented normal, or even increased bone mineral density (BMD) by DXA[32]. Notwithstanding, the bone of individuals with diabetes shows a number of structural characteristics predisposing it to suffer from destruction, including smaller cortical area, greater cortical porosity and decreased strength[33, 34]. Besides, the bone metabolic disorders in diabetes also suppress bone turnover such as increased advanced glycation end products (AGEs) levels[35]. The accumulation of AGEs disrupts bone cellular biology and microarchitecture[36], and the bone strength is deteriorated by the poor material properties resulting from elevated AGE levels[37]. Therefore, prolonged hyperglycemia has potential detrimental effects on bone material properties[38], so we focused on the screw stabilization under diabetic conditions in this study. Previously, we demonstrated that diabetes could lead to the instability of titanium implant by inducing increased reactive oxygen species level and poor bone regeneration due to decreased osteoblasts function at cellular and molecular levels [12]. Besides, the deterioration of bone quality caused by diabetes might be another important reason for screw loosening. However, previous studies mainly concentrated on molecular mechanism based on animal experiments. The clinical studies on screw stabiliztion under diabetic condition are lacking. We then discovered that porous titanium alloy structure, chitosan/hydroxyapatite composites coating and adiponectin could improve bone-implant osseointegration [39–41]. Meanwhile, our research group explored expandable pedicle screw for use in the treatment of osteoporotic lumbar spine [42, 43]. Nevertheless, whatever treatment strategy has been selected for patients with DM, the entire glucose control is always essential to maintain relatively good bone quality to reduce the possibility of screw loosening. To note, screw augmentation might be a good strategy[44].

Considering the incidence of loosening varies widely in several series [45–50], the standard of screw loosening for radiological diagnose remains to be discussed. B. Sandén et al. found that the radiolucent zone in the radiograph could indicate the loosening of pedicle screw well [51]. Meanwhile, and Dakhil-Jerew et al and Ko et al described this inner zone surrounded by bone tissue as ‘double halo’ [52, 53]. Herein, we determined the criteria of screw loosening as a 1 mm zone.

However, there were several limitations in this study. First, the evaluation methods might be relatively subjective, and a dichotomous outcome improvement standard set in this study might be one-sided. Second, CT was not included in our study because X-ray was the conventional use of postoperative follow-up in China. Further CT evaluation of screw loosening would be conducted in our prospective study in the future [54, 55].

Our study indicated that patients with DM were more likely to experience screw loosening after spinal surgery, while the risk of loosening was related to diabetes but not pre-op or post-op glycemic level. Furthermore, the history of diabetes was closely associated with clinical outcomes after spinal surgery. Specifically, it was highly related to pain perception in the lumbar spine, while it was closely concerned with motor and sensory functions in the cervical spine. More specifically, pedicle screw loosening in the lumbar spine had a major adverse impact on functional prognosis such as standing, walking, and sitting, which suggested the particularity of internal fixation at the load-bearing site. It should be emphasized to avoid poor recovery of motor function, especially in the treatment of diabetes. However, the etiology of screw loosening has not been extensively elucidated yet. The results of our study were in accordance with our previous studies about mechanisms of implant instability under DM. Moreover, previous literature has demonstrated that diabetics could benefit from glycemic control after spinal surgery[14], and this study provided a reference for clinicians’ decision making in treatment of diabetics and patients’ glycemic control after spinal surgery.

DM had an adverse effect on screw stabilization, and the screw loosening may lead to the impaired improvement of clinical outcomes postoperatively in diabetics. In addition to direct effects on operative wounds and neural function, the impact of DM on the screws was also worth noting.

Ethical Approval

The protocol for the study was approved by the Medical Ethics Committee of Xijing Hospital of the Fourth Military Medical University. Each author certifies that all investigations were conducted in conformity with ethical principles. All participants provided informed consent before their participation in the study and written were obtained from all participants. All patients have expressly approved the use of their data for the current study. All clinical investigations had been conducted according to the principles expressed in the Declaration of Helsinki.

Competing interests

The authors declare no conflicts of interest. The authors declare they have no financial interests or non-financial interests.

Authors' contributions

T. Wang, J. Wang and X. Hu contributed equally to this work. Conception and design: T. Wang and Y. Feng. Acquisition of data: T. Li, X. Zhao, Y. Zhang, T. Ma and Y. Chen. Analysis and Interpretation of data: T. Wang, J. Wang and T. Li. Drafting the article: T. Wang. Critically revising the article: W. Lei, Z. Wu and Y. Feng. Statistical analysis: X. Hu and G. Xiang. Reviewed submitted version of manuscript: all authors. The authors read and approved the final manuscript.

Funding

This work is supported by the National Natural Science Foundation of China under Grant Nos. 82172393, 81871799, 81902184 and 82102603, the Key Research and Development Program of Shaanxi Province of China under Grant No. 2020SF-186 and the Natural Science Foundation of Shaanxi Province of China under Grant No. 2019JM-465.

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article. Raw data can be requested from the corresponding author.

Roy-Camille R, Saillant G, Mazel C (1986) Internal fixation of the lumbar spine with pedicle screw plating. Clinical Orthopaedics and Related Research® 203:7–17
Boos N, Webb J (1997) Pedicle screw fixation in spinal disorders: a European view. European Spine Journal 6:2–18
Chiu C, Kwan M, Chan C, Schaefer C, Hansen-Algenstaedt N (2015) The accuracy and safety of fluoroscopically guided percutaneous pedicle screws in the lumbosacral junction and the lumbar spine: a review of 880 screws. The bone & joint journal 97:1111–1117
Glaser J, Stanley M, Sayre H, Woody J, Found E, Spratt K (2003) A 10-year follow-up evaluation of lumbar spine fusion with pedicle screw fixation. Spine 28:1390–1395
Faraj A, Webb J (1997) Early complications of spinal pedicle screw. European spine journal 6:324–326
Di Silvestre M, Parisini P, Lolli F, Bakaloudis G (2007) Complications of thoracic pedicle screws in scoliosis treatment. Spine 32:1655–1661
Galbusera F, Volkheimer D, Reitmaier S, Berger-Roscher N, Kienle A, Wilke H-J (2015) Pedicle screw loosening: a clinically relevant complication? European Spine Journal 24:1005–1016
Collaboration ERF (2010) Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. The Lancet 375:2215–2222
Kremers HM, Schleck CD, Lewallen EA, Larson DR, Van Wijnen AJ, Lewallen DG (2017) Diabetes mellitus and hyperglycemia and the risk of aseptic loosening in total joint arthroplasty. The Journal of arthroplasty 32:S251-S253
Oates TW, Dowell S, Robinson M, McMahan CA (2009) Glycemic control and implant stabilization in type 2 diabetes mellitus. Journal of dental research 88:367–371
Wu J-C, Huang W-C, Tsai H-W, Ko C-C, Wu C-L, Tu T-H, Cheng H (2011) Pedicle screw loosening in dynamic stabilization: incidence, risk, and outcome in 126 patients. Neurosurgical focus 31:E9
Feng Y-F, Wang L, Zhang Y, Li X, Ma Z-S, Zou J-W, Lei W, Zhang Z-Y (2013) Effect of reactive oxygen species overproduction on osteogenesis of porous titanium implant in the present of diabetes mellitus. Biomaterials 34:2234–2243
Yilgor C, Sogunmez N, Boissiere L, Yavuz Y, Obeid I, Kleinstück F, Pérez-Grueso F, Acaroglu E, Haddad S, Mannion A, Pellise F, Alanay A (2017) Global Alignment and Proportion (GAP) Score: Development and Validation of a New Method of Analyzing Spinopelvic Alignment to Predict Mechanical Complications After Adult Spinal Deformity Surgery. The Journal of bone and joint surgery American volume 99:1661–1672. doi: 10.2106/jbjs.16.01594
Armaghani SJ, Archer KR, Rolfe R, Demaio DN, Devin CJ (2016) Diabetes is related to worse patient-reported outcomes at two years following spine surgery. JBJS 98:15–22
Zou D, Muheremu A, Sun Z, Zhong W, Jiang S, Li W (2020) Computed tomography Hounsfield unit–based prediction of pedicle screw loosening after surgery for degenerative lumbar spine disease. Journal of Neurosurgery: Spine 32:716–721
Di Silvestre M, Lolli F, Bakaloudis G, Parisini P (2010) Dynamic stabilization for degenerative lumbar scoliosis in elderly patients. Spine 35:227–234
Huang Wm, Yu Xm, Xu Xd, Song Rx, Yu Ll, Yu Xc (2017) Posterior Lumbar Interbody Fusion with Interspinous Fastener Provides Comparable Clinical Outcome and Fusion Rate to Pedicle Screws. Orthopaedic Surgery 9:198–205
Hamann C, Kirschner S, Günther K, Hofbauer L (2012) Bone, sweet bone–osteoporotic fractures in diabetes mellitus. Nature reviews Endocrinology 8:297–305. doi: 10.1038/nrendo.2011.233
Leslie W, Rubin M, Schwartz A, Kanis J (2012) Type 2 diabetes and bone. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 27:2231–2237. doi: 10.1002/jbmr.1759
Burghardt A, Issever A, Schwartz A, Davis K, Masharani U, Majumdar S, Link T (2010) High-resolution peripheral quantitative computed tomographic imaging of cortical and trabecular bone microarchitecture in patients with type 2 diabetes mellitus. The Journal of clinical endocrinology and metabolism 95:5045–5055. doi: 10.1210/jc.2010-0226
Shu A, Yin M, Stein E, Cremers S, Dworakowski E, Ives R, Rubin M (2012) Bone structure and turnover in type 2 diabetes mellitus. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 23:635–641. doi: 10.1007/s00198-011-1595-0
Patsch J, Burghardt A, Yap S, Baum T, Schwartz A, Joseph G, Link T (2013) Increased cortical porosity in type 2 diabetic postmenopausal women with fragility fractures. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 28:313–324. doi: 10.1002/jbmr.1763
Nilsson A, Sundh D, Johansson L, Nilsson M, Mellström D, Rudäng R, Zoulakis M, Wallander M, Darelid A, Lorentzon M (2017) Type 2 Diabetes Mellitus Is Associated With Better Bone Microarchitecture But Lower Bone Material Strength and Poorer Physical Function in Elderly Women: A Population-Based Study. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 32:1062–1071. doi: 10.1002/jbmr.3057
Malgo F, Hamdy N, Papapoulos S, Appelman-Dijkstra N (2015) Bone material strength as measured by microindentation in vivo is decreased in patients with fragility fractures independently of bone mineral density. The Journal of clinical endocrinology and metabolism 100:2039–2045. doi: 10.1210/jc.2014-4346
Furst J, Bandeira L, Fan W, Agarwal S, Nishiyama K, McMahon D, Dworakowski E, Jiang H, Silverberg S, Rubin M (2016) Advanced Glycation Endproducts and Bone Material Strength in Type 2 Diabetes. The Journal of clinical endocrinology and metabolism 101:2502–2510. doi: 10.1210/jc.2016-1437
Cheung C, Ikram M, Klein R, Wong T (2015) The clinical implications of recent studies on the structure and function of the retinal microvasculature in diabetes. Diabetologia 58:871–885. doi: 10.1007/s00125-015-3511-1
Rahman S, Rahman T, Ismail A, Rashid A (2007) Diabetes-associated macrovasculopathy: pathophysiology and pathogenesis. Diabetes, obesity & metabolism 9:767–780. doi: 10.1111/j.1463-1326.2006.00655.x
Jiang X, Zhu Y, Liu Z, Tian Z, Zhu S (2021) Association between diabetes and dental implant complications: a systematic review and meta-analysis. Acta odontologica Scandinavica 79:9–18. doi: 10.1080/00016357.2020.1761031
Hu X, Xiang G, Wang T, Ma Y, Zhang Y, Yan Y, Zhao X, Wu Z, Feng Y, Lei W (2021) Impairment of type H vessels by NOX2-mediated endothelial oxidative stress: critical mechanisms and therapeutic targets for bone fragility in streptozotocin-induced type 1 diabetic mice. Theranostics 11:3796–3812. doi: 10.7150/thno.50907
Mac-Thiong J-M, Parent S, Poitras B, Joncas J, Hubert L (2013) Neurological outcome and management of pedicle screws misplaced totally within the spinal canal. Spine 38:229–237
Zhang C, Mannen E, Sis H, Cadel E, Wong B, Wang W, Cheng B, Friis E, Anderson D (2020) Moment-rotation behavior of intervertebral joints in flexion-extension, lateral bending, and axial rotation at all levels of the human spine: A structured review and meta-regression analysis. Journal of biomechanics 100:109579. doi: 10.1016/j.jbiomech.2019.109579
Vestergaard P (2007) Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes–a meta-analysis. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 18:427–444. doi: 10.1007/s00198-006-0253-4
Farr J, Drake M, Amin S, Melton L, McCready L, Khosla S (2014) In vivo assessment of bone quality in postmenopausal women with type 2 diabetes. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 29:787–795. doi: 10.1002/jbmr.2106
Lecka-Czernik B, Rosen C (2015) Energy Excess, Glucose Utilization, and Skeletal Remodeling: New Insights. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research 30:1356–1361. doi: 10.1002/jbmr.2574
Yamamoto M, Sugimoto T (2016) Advanced Glycation End Products, Diabetes, and Bone Strength. Current osteoporosis reports 14:320–326. doi: 10.1007/s11914-016-0332-1
Henderson S, Ibe I, Cahill S, Chung Y, Lee F (2019) Bone Quality and Fracture-Healing in Type-1 and Type-2 Diabetes Mellitus. The Journal of bone and joint surgery American volume 101:1399–1410. doi: 10.2106/jbjs.18.01297
Sihota P, Yadav R, Dhaliwal R, Bose J, Dhiman V, Neradi D, Karn S, Sharma S, Aggarwal S, Goni V, Mehandia V, Vashishth D, Bhadada S, Kumar N (2021) Investigation of Mechanical, Material, and Compositional Determinants of Human Trabecular Bone Quality in Type 2 Diabetes. The Journal of clinical endocrinology and metabolism 106:e2271-e2289. doi: 10.1210/clinem/dgab027
Farr J, Khosla S (2016) Determinants of bone strength and quality in diabetes mellitus in humans. Bone 82:28–34. doi: 10.1016/j.bone.2015.07.027
Hu X-F, Wang L, Lu Y-Z, Xiang G, Wu Z-X, Yan Y-B, Zhang Y, Zhao X, Zang Y, Shi L (2017) Adiponectin improves the osteointegration of titanium implant under diabetic conditions by reversing mitochondrial dysfunction via the AMPK pathway in vivo and in vitro. Acta Biomaterialia 61:233–248
Li X, Ma X-Y, Feng Y-F, Ma Z-S, Wang J, Ma T-C, Qi W, Lei W, Wang L (2015) Osseointegration of chitosan coated porous titanium alloy implant by reactive oxygen species-mediated activation of the PI3K/AKT pathway under diabetic conditions. Biomaterials 36:44–54
Ma X-Y, Feng Y-F, Ma Z-S, Li X, Wang J, Wang L, Lei W (2014) The promotion of osteointegration under diabetic conditions using chitosan/hydroxyapatite composite coating on porous titanium surfaces. Biomaterials 35:7259–7270
Wu Z-x, Cui G, Lei W, Fan Y, Wan S-y, Ma Z-s, Sang H-x (2010) Application of an expandable pedicle screw in the severe osteoporotic spine: a preliminary study. Clinical and Investigative Medicine:E368-E374
Wu Z-x, Gong F-t, Liu L, Ma Z-s, Zhang Y, Zhao X, Yang M, Lei W, Sang H-x (2012) A comparative study on screw loosening in osteoporotic lumbar spine fusion between expandable and conventional pedicle screws. Archives of orthopaedic and trauma surgery 132:471–476
Massaad E, Rolle M, Hadzipasic M, Kiapour A, Shankar GM, Shin JH (2021) Safety and efficacy of cement augmentation with fenestrated pedicle screws for tumor-related spinal instability. Neurosurgical Focus FOC 50:E12. doi: 10.3171/2021.2.Focus201121
Röllinghoff M, Schlüter-Brust K, Groos D, Sobottke R, Michael JW-P, Eysel P, Delank KS (2010) Mid-range outcomes in 64 consecutive cases of multilevel fusion for degenerative diseases of the lumbar spine. Orthopedic reviews 2
Yuan HA, Garfin SR, Dickman CA, Mardjetko SM (1994) A historical cohort study of pedicle screw fixation in thoracic, lumbar, and sacral spinal fusion. Spine 19:2279S-2296S
Schizas C, Tzinieris N, Tsiridis E, Kosmopoulos V (2009) Minimally invasive versus open transforaminal lumbar interbody fusion: evaluating initial experience. International orthopaedics 33:1683–1688
Pihlajamäki H, Myllynen P, Böstman O (1997) Complications of transpedicular lumbosacral fixation for non-traumatic disorders. The Journal of bone and joint surgery British volume 79:183–189
Esses SI, Sachs BL, Dreyzin V (1993) Complications associated with the technique of pedicle screw fixation a selected survey of ABS members. Spine 18:2231–2239
Okuyama K, Abe E, Suzuki T, Tamura Y, Chiba M, Sato K (1999) Posterior lumbar interbody fusion: a retrospective study of complications after facet joint excision and pedicle screw fixation in 148 cases. Acta Orthopaedica Scandinavica 70:329–334
Sanden B, Olerud C, Petren-Mallmin M, Johansson C, Larsson S (2004) The significance of radiolucent zones surrounding pedicle screws: definition of screw loosening in spinal instrumentation. The Journal of bone and joint surgery British volume 86:457–461
Dakhil-Jerew F, Jadeja H, Cohen A, Shepperd J (2009) Inter-observer reliability of detecting Dynesys pedicle screw using plain X-rays: a study on 50 post-operative patients. European Spine Journal 18:1486–1493
Ko C-C, Tsai H-W, Huang W-C, Wu J-C, Chen Y-C, Shih Y-H, Chen H-C, Wu C-L, Cheng H (2010) Screw loosening in the Dynesys stabilization system: radiographic evidence and effect on outcomes. Neurosurgical focus 28:E10
Xu F, Zou D, Li W, Sun Z, Jiang S, Zhou S, Li Z (2020) Hounsfield units of the vertebral body and pedicle as predictors of pedicle screw loosening after degenerative lumbar spine surgery. Neurosurgical focus 49:E10
Clerk-Lamalice O, Irani Z, Growney M, Beall DP, Hirsch JA (2018) Parapedicular vertebral augmentation with polymethylmetacrylate for pedicle screw loosening. Journal of Neurointerventional Surgery 11:neurintsurg-2017-013548

No competing interests reported.

SI2022817new.docx

Diabetes-related screw loosening: the distinction of surgical sites and the relationship among diabetes, implant stabilization and clinical outcomes

Status:

Version 1

Abstract

Introduction.

Methods.

Results.

Conclusions.

Figures

Introduction

Material And Methods

Patient enrollment and surgical information collection

Clinical evaluations

Radiological evaluations

Statistical analysis

Results

Characteristics of the patients

Radiographic follow-up on loosening

Clinical evaluations

Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1