Increased Surgical Site Subcutaneous Fat Thickness Is Associated with Infection after Posterior Cervical Fusion

Abstract

Background:

Previous literature has associated increased body mass index (BMI) with risk of surgical site infection (SSI) after posterior cervical fusion (PCF) surgery. However, few studies have examined the association between local adiposity and risk of SSI, re-admission, and re-operation after PCF. Local adiposity is easily measured on pre-operative magnetic resonance imaging (MRI) and may act as a more accurate predictor compared with BMI.

Patients and Methods:

Subjects undergoing PCF from 2013–2018 at a single institution were identified retrospectively. Posterior cervical subcutaneous fat thickness, paraspinal muscle thickness, and lamina-to-skin distance measurements were obtained from computed tomography (CT) or MRI scans. Subjects with active infection, malignancy, or revision procedures were excluded.

Results:

Two hundred five patients were included with 20 developing SSIs. Subjects with SSIs had a longer fusion construct (4.90 vs. 3.71 levels; p = 0.001), higher Elixhauser comorbidity index (ECI; 2.05 vs. 1.34; p = 0.045), had a history of diabetes mellitus (30% vs. 10.8%; p = 0.026), higher subcutaneous fat thickness (30.5 vs. 23.6 mm; p = 0.013), and higher lamina-to-skin distance (66.4 vs. 57.9 mm; p = 0.027). Subcutaneous fat thickness (odds ratio [OR], 1.05; 95% confidence interval [CI], 1.01–1.10]; p = 0.026) and lamina-to-skin distance (OR, 1.05; 95% CI, 1.01–1.09]; p = 0.014) were associated with SSI in multivariable analysis. A subcutaneous fat thickness cutoff value of 23.2 mm had 90% sensitivity and 54.1% specificity for prediction of SSI. There was no association need for re-admission or re-operation.

Conclusions:

Increased posterior cervical fat may increase the risk of SSI after PCF. Pre-operative advanced imaging may be a valuable tool for assisting with patient counseling, optimization, and risk stratification.

The incidence of surgical site infections (SSI) after any cervical spine surgery has been reported to be between 1% and 7% [1 –3]. The published incidence of SSI after posterior cervical spine surgery has been reported as high as 18% [4 –6] but probably occurs in 0.7% to 4.1% of paients according to larger series [7,8]. Surgical site infection is the most common cause of 30-day re-admission after spine surgery and levies on average an additional $12,619 per posterior cervical spine case with infection [9,10]. The pathogenesis of SSI is attributed most commonly to direct inoculation at the time of surgery, likely in the setting of compromised local or systemic host factors [11 –13]. Therefore, exploring host factor interventions that can help reduce these complications is paramount.

Obesity, which affects more than one-third of the United States population, is a well-known risk factor for SSI after spine surgery [1,2,14,15]. Locally, increased adiposity has been shown to contribute to soft tissue dead space and necrosis after wound closure [16,17]. Seminal studies by Mehta et al. [18,19] demonstrated thickness of subcutaneous fat to be a predictor of SSIs in lumbar fusion and posterior cervical fusion (PCF) patients. Using similar methodology, we expand on this to provide information on re-admissions and re-operation rates. Additionally, obesity has profound implications on systemic immune function and tissue vascularity via its pro-inflammatory effects, as well as its contributions to insulin resistance and diabetes [20 –22]. This current study explored the extent to which local adiposity at the posterior neck influences rates of SSI, 90-day re-admission, and re-operation after posterior cervical spine fusion.

Patients and Methods

An Institutional Review Board-approved retrospective cohort study was conducted with patients who had undergone a primary posterior cervical decompression and fusion procedure between January 2013 and December 2018 at a high-volume academic medical center. Patients were identified by a Standardized Query Language (SQL) search of the following Current Procedural Terminology (CPT) codes for posterior cervical fusion procedures: 22590, 22600, 22614, 22840, 22841, 22842, 22843, and 22844.

Inclusion criteria were: pre-operative computed tomography (CT) or magnetic resonance imaging (MRI) scans of the cervical spine; over 18 years of age; and diagnosis of cervical myelopathy or radiculopathy. Exclusion criteria were: lack of data corresponding to post-operative course; revision posterior cervical decompression and fusion; and surgical intervention due to active infection or malignancy.

Patient demographics and surgical characteristics

Patient demographic and surgical characteristics were obtained from the SQL query and via electronic medical record chart review. These data included age, gender, body mass index (BMI), smoking status, Elixhauser comorbidity index (ECI), and Charlson comorbidity index (CCI). Surgical characteristics collected included: primary pre-operative diagnosis, number of levels fused, if the fusion construct extended distal to C7, and length of stay (LOS).

Radiographic measurements

Radiographic parameters were measured on pre-operative T1-weighted MRI or CT scans using the Sectra Workstation IDS7 21.1 software suite (Sectra AB, Linköping, Sweden). All measurements were obtained horizontal to the level of the C5 vertebrae using sagittal cuts. Three different measurements were calculated for each patient (Fig. 1) [18]:

FIG. 1.

T1-weighted sagittal MRI showing radiographic measurements of (A) lamina-to-skin thickness, (B) paraspinal muscle thickness, and (C) subcutaneous fat thickness

Paraspinal muscle thickness (millimeters): distance from the lamina to the posterior edge of the paraspinal musculature.

Subcutaneous fat thickness (millimeters): distance from the paraspinal musculature to the posterior edge of the subcutaneous fat.

Lamina-to-skin thickness (millimeters): distance from the lamina to the skin surface.

An interclass correlation coefficient (ICC) was calculated to show internal validity of radiographic measurements between the multiple observers.

Outcomes

Patient outcomes were obtained via chart review of patient medical records, telephone encounters, and patient navigator data. Outcomes of interest included: SSIs, 90-day re-admissions, and re-operation. Surgical site infections were classified based on the U.S. Centers for Disease Control and Prevention (CDC) surgical site infection criteria [23]. Surgical site infection was either treated with surgical irrigation and debridement with antibiotic agents or antibiotic agents alone depending on severity.

Statistical analysis

Descriptive statistics including mean, standard deviation, proportions, and confidence intervals were calculated. Normality of plots was assessed for each variable via skewness, kurtosis, and Shapiro-Wilk tests. Non-parametric tests were run for skewed (non-normal) data, whereas parametric tests were run for normally distributed data. Sample means were compared using either a parametric one-way analysis of variance (ANOVA) or non-parametric Kruskall-Wallis test. Categorical data were compared using Pearson χ² test. Receiver operating characteristic (ROC) curve analysis was used to establish thresholds for SSIs for radiographic measurements found to predict SSIs. Youden index was used to indicate the point on the curve that achieves the highest sensitivity and specificity as the optimal cutoff value (threshold). An area under the curve (AUC) between 0.70 and 0.80 was considered a fair model, whereas an area under the curve above 0.80 was considered an excellent model [24,25]. Significance was assumed for p ≤ 0.05.

Logistic regression models were run for dichotomous categorical variables, including: SSI, re-admission, and re-operation. Confounding variables with a univariable analysis p < 0.2 were included in the final logistic regression model. All statistical analyses were performed using SPSS statistics, version 26 (IBM Corp, Armonk, NY).

Results

Lamina to skin, muscle thickness, and subcutaneous fat averages all showed strong inter-agreement between observers with ICC values of odds ratio (OR), 0.950; 95% confidence interval (CI), 0.921–0.967; OR, 0.863; 95% CI, 0.814–0.898, and OR, 0.937; 95% CI, 0.808–0.970, respectively.

Patient demographics and radiographic measurements

After applying inclusion and exclusion criteria, a total of 205 patients were identified for final analysis, with 20 (9.76%) patients having SSIs. Univariable analysis of patient demographic data showed a longer fusion construct in patients with SSIs (4.90 vs. 3.71 levels; p = 0.001; Table 1). Higher ECI was also associated with increased risk of SSI (2.05 vs. 1.34; p = 0.045). A higher patient BMI was not associated with increased risk of SSI (30.9 vs. 28.7; p = 0.153), however, a self-reported history of diabetes mellitus was associated with an increased risk of SSI (30% vs. 10.8%; p = 0.026). There was no significant difference in LOS between the cohorts (4.95 vs. 2.88; p = 0.166). Patients with an SSI had a higher subcutaneous fat thickness (30.5 vs. 23.6; p = 0.013) and lamina-to-skin thickness (66.4 vs. 57.9; p = 0.027). No corresponding significance was observed for paraspinal muscle thickness (p = 0.639).

Table 1.

Posterior Cervical Fusion Patient Demographics Based on Surgical Site Infection

	Without SSI (n = 185)	With SSI (n = 20)	p ^a–c
Age	61.5 (11.7)	63.5 (11.3)	0.439
Gender			0.774
Female	76 (41.1%)	7 (35.0%)
Male	109 (58.9%)	13 (65.0%)
BMI	28.7 (5.88)	30.9 (6.52)	0.153
Smoking status			0.55
Current	43 (23.2%)	5 (25.0%)
Former	32 (17.3%)	5 (25.0%)
Never	110 (59.5%)	10 (50.0%)
ECI	1.34 (1.29)	2.05 (1.43)	0.045
CCI	2.69 (1.87)	2.85 (2.43)	0.781
DM	20 (10.8%)	6 (30%)	0.026
Pre-operative diagnosis			0.536
Stenosis	116 (62.7%)	13 (72.2%)
Spondylolisthesis	61 (33.0%)	4 (22.2%)
Trauma	8 (4.32%)	1 (5.56%)
Number of levels fused	3.71 (1.76)	4.90 (1.37)	0.001
Construct proximal to C7	41 (22.2%)	7 (35.0%)	0.263
Construct distal to C7	147 (79.5%)	17 (85.0%)	0.770
Length of stay	2.88 (1.71)	4.95 (6.42)	0.166
Paraspinal muscle thickness (mm)	33.9 (5.57)	34.7 (7.55)	0.639
Subcutaneous fat thickness (mm)	23.6 (11.2)	30.5 (10.8)	0.013
Lamina-to-skin thickness (mm)	57.9 (12.7)	66.4 (15.3)	0.027
Ninety-day re-admission	10 (5.41%)	10 (50.0%)	< 0.001
Re-operation	12 (6.49%)	4 (20.0%)	0.056

SSI = surgical site infection; BMI = Body mass index; ECI = Elixhauser comorbidity index; CCI = Charlson comorbidity index; DM = diabetes mellitus.

Independent samples t-test or Mann-Whitney U test for continuous variables.

Pearson χ² test for categorical variables.

Significance established at p ≤ 0.05.

A total of 20 (9.76%) patients experienced 90-day re-admission; 17 (8.29%) for posterior wound complications (15 patients for wound infection, 1 for a dural fluid leak, and 1 for wound dehiscence) and three (1.46%) because of intractable pain. Surgical site infection was associated with increased risk of re-admission (p < 0.001; Table 1). A total of 16 (12.8%) patients underwent a repeat posterior cerivcal surgery for any reason. These included seven for wound irrigation and debridement, eight for adjacent segment disease, and one for pain/post-laminectomy syndrome. There was no association between SSI and need for re-operation (p = 0.056).

Regression analysis

Subcutaneous fat thickness and lamina-to-skin thickness were significant predictors of SSI on multivariable analysis (Table 2). When including BMI, ECI, and number of levels fused as confounders, for every 1-mm increase in the thickness of posterior cervical subcutaneous fat, the odds of experiencing an SSI increased by 5% (OR, 1.05; 95% CI 1.01–1.10; p = 0.026). Similarly, the odds of developing an SSI increased by 5% for every 1-mm increase in lamina-to-skin thickness (OR, 1.05; 95% CI, 1.01–1.09; p = 0.014). No difference was noted with respect to subcutaneous fat thickness (OR, 1.03; 95% CI, 0.99–1.07; p = 0.161) or lamina-to-skin thickness (OR, 1.03; 95% CI, 0.99–1.07; p = 0.112) for 90-day re-admission (Table 3).

Table 2.

Logistic Regression Table for Predictors of Surgical Site Infections

Predictor	OR	95% CI	p ^a
Subcutaneous fat thickness	1.05	1.01–1.10	0.026
Lamina-to-skin thickness	1.05	1.01–1.09	0.012

Significance established at p ≤ 0.05.

OR = odds ratio; CI = confidence interval; BMI = body mass index; ECI = Elixhauser comorbidity index; DM = diabetes mellitus.

Logistic regression model included BMI, ECI, and number of levels fused, and DM.

Table 3.

Logistic Regression Table for Predictors of Ninety-Day Re-Admissions

Predictor	OR	95% CI	p ^a
Subcutaneous fat thickness	1.03	0.99–1.07	0.161
Lamina-to-skin thickness	1.03	0.99–1.07	0.112

Significance established at p ≤ 0.05.

OR = odds ratio; CI = confidence interval; BMI = body mass index; ECI = Elixhauser comorbidity index.

Logistic regression model included BMI, ECI, and number of levels fused.

Receiver operator curves (ROC; Fig. 2A and 2B) for subcutaneous fat thickness and lamina-to-skin thickness and risk of SSI showed an AUC of 0.7 (95% CI, 0.615–0.785) and 0.666 (95% CI, 0.563–0.770), respectively (Table 4). A subcutaneous fat thickness cutoff of 23.2 mm was associated with a sensitivity of 90% and a specificity of 54.1% for predicting SSI. A lamina-to-skin thickness cutoff of 57.3 mm was associated with a sensitivity of 85% and a specificity of 55.7% for predicting SSI.

FIG. 2.

(A) Receiver operating curve for subcutaneous fat thickness. (B) Receiver operating curve for lamina-to-skin thickness.

Table 4.

Receiver Operating Curve Analysis for Lamina-to-Skin Thickness and Subcutaneous Fat Thickness

Predictor	Cutoff	AUC	95% CI	Sensitivity	Specificity
Subcutaneous fat thickness	23.20	0.700	0.615–0.785	0.900	0.541
Lamina-to-skin Thickness	57.30	0.666	0.563–0.770	0.850	0.557

AUC = area under the curve; CI = confidence interval.

Discussion

Although there are many known comorbidities that increase rates of SSI in spine surgery, obesity maintains a leading role. Studies often use BMI as a surrogate measure of obesity, but BMI does not take body composition, such as increased muscle mass, into consideration [26,27]. Cervical adipose tissue thickness, which is readily available on pre-operative imaging, may be a more accurate proxy for obesity and subsequent risk of SSI. Cervical adiposity measurements could be then used in pre-operative counseling, risk stratification, and patient optimization prior to PCF. Our study found that both posterior subcutaneous fat thickness as well as lamina-to-skin thickness, but not BMI, were predictors of SSI after PCF.

Our results further corroborate findings from Mehta et al. [18] showing higher rates of SSI in patients with increased posterior subcutaneous fat thickness. Mean subcutaneous fat thickness of 23.2 mm and lamina-to-skin thickness of 57.3 mm compared with 27.0 mm and 62.4 mm, respectively, in the study by Mehta et al. [18] were reliable cutoffs for predicting SSI. Lamina-to-skin thickness did not reach significance in the analysis by Mehta et al. [18] analysis but did reach significance in our series. Furthermore, logistic regression analysis in our study showed subcutaneous fat to be a predictor of SSI, although to a lesser degree than their prior study (OR, 1.05 vs. 3.18, respectively). The overall infection rate of 9.76% in our cohort is comparable to previous series examining SSI after posterior cervical fusion surgery [4 –6,18]. Although average BMI was higher among patients who experienced an SSI versus those who did not (30.9 vs. 28.7; p = 0.153), the thickness of posterior cervical subcutaneous fat had a strong predictive value with regards to incidence of SSI. This observation was also noted by Lee et al. [28] in a lumbar spine cohort. Their study retrospectively reviewed 149 patients status post-lumbar spine procedure through a midline approach, and found that subcutaneous fat thickness, BMI, and obesity were associated with SSI [28]. Moreover, multivariable analysis of their data showed that subcutaneous fat thickness led to a 6% increase in odds of SSI, and that patients with fat thickness >50 mm had a four times higher odds of SSI compared with those <50 mm.

The combined results of these findings can alternatively suggest that local adiposity, not just obesity-related systemic effects, may play a role in risk of SSI. The implications of local adiposity versus whole body obesity are not clear. Voluminus subcutaneous fat makes surgical closure more difficult. Closure method has been found to play a role in SSI in other studies [29,30]. Additionaly, fat necrosis and fluid accumulation in so-called dead space may harbor bacteria that can lead to infection. More studies are needed to investigate the local mileau within the posterior cervical spine after PCF.

Fat specifically, not just soft tissue (such as muscle), appears to be responsible as paraspinal muscle thickness had no effect on SSI. In theory, increased muscle thickness could even reduce infection rates due to improved vascular soft tissue coverage, but this was not the case. The effects of paraspinal muscle thickness were potentially limited by radiographic measurements being made at the midline, the point at which paraspinal muscle is the thinnest. Paraspinal muscle thickness could be better evaluated using saggital cuts lateral to the midline. Further studies looking at muscle thicknes and SSI may be an area of interest in the future.

Posterior cervical subcutaneous fat thickness was found to increase risk of 90-day re-admission rates in univariable but not in multivariable analysis. There was multiple reasons for re-admission in our study, the majority of which were for wound complications. Re-admission rates for posterior cervical spine surgery are higher than for anterior procedures [10], with SSI as one of the major causal factors [31]. This study was likely underpowered to detect the signficace of cervical adiposity and risk of re-admission.

Pain is another common reason for post-operative re-admission after spine surgery, representing 6%–22% of re-admissions in some studies [9,10,32]. The muscular dissection necessary in posterior approaches to the cervical spine has been implicated as a substantial pain generator after surgery [33,34]. In our cohort, paraspinal muscle thickness had no bearing on re-admission rate, however, re-admission due to pain was not specifically examined. This may be a topic of future research.

The literature shows that patients undergoing cervical spine surgery via a posterior approach are at higher risk of revision surgery relative to those undergoing anterior procedures [35]. Our study did not find an association between thickness measurements and revision surgery rates. Of the 20 patients who experienced SSI, seven underwent irrigation and debridment procedures. Although reasons for revision posterior cervical spine surgery are not well described, one series of 768 patients at a single institution noted just three (0.39%) revisions, all for persistent or new neurologic symptoms [36]. Our study also found patients needing re-operation for persistent symptoms. Although increased subcutaneous fat may increase the likelihood of requiring a revision procedure, the relatively few patients requiring re-operation for infection can explain the lack of significance [37].

In assessing patients with abundant posterior cervical subcutaneous fat, several additional factors may be considered. Although there are no clinical studies demonstrating that pre-operative weight loss leads to better surgical outcomes, the added risk of complications and infection that high BMI confers is clear [38 –40]. As such, pre-operative reduction of overall adiposity would likely be of general benefit and should be considered [41,42]. Additionally, the spine oncology and scoliosis literature has demonstrated improved outcomes when a plastic surgery closure is employed [43]. Added expertise in soft tissue management may be a consideration for these patients [44 –46]. Finally, the use of incisional negative pressure wound dressings after primary closure in high risk patients undergoing spine surgery has shown promise in reducing the risk of post-operative infection and wound complications [47,48]. This low risk modality may also be of benefit in minimizing the risk of infection in the context of increased posterior cervical subcutaneous fat.

Our study is not without limitations. First, there are inherent variations to manual measurements on pre-operative CT/MRI scans. We calculated inter-and intra-observer reliability to demonstrate accuracy of these measurements. Second, we found a higher number of fused vertebral levels in the subset of cases complicated by SSI. Although previous literature suggests that number of cervical levels fused does increase the risk for SSI, this is not a universal finding [2,49]. More importantly, this variable was controlled for in our regression modeling. Additionally, unavailability of complete pain score records prevented us from further exploring the relation between post-operative pain, re-admissions, and paraspinal muscle thickness. Finally, the retrospective nature of this study limits available data and, as such, further evaluation of other potential risks such as disease severity and intra-operative factors such as skin preparation. Our study controls for some cofounders in that all patients originate from a single center with designated protocols for wound closure and post-operative management.

Conclusions

This study identified increased posterior cervical adiposity a risk factor for SSI after PCF. Body mass index is often used as a surrogate measure of obesity, but BMI does not take body composition, such as increased muscle mass or adiposity, into consideration. The posterior subcutaneous fat thickness as well as lamina-to-skin thickness, but not BMI, were predictors of SSI after PCF. Posterior cervical fat measurements may be more accurate at predicting SSI than BMI alone. Readily available pre-operative imaging can therefore be used as a valuable tool for assisting with patient counseling, optimization, and risk stratification. Multicenter studies and larger datasets are necessary to detect risk factors for re-operation and re-admission, and to identify other risk factors for SSI in the setting of posterior cervical spine surgery.

Footnotes

Funding Information

There is no financial support to disclose for this work.

Author Disclosure Statement

Dr. Donnally has nothing to disclose. Dr. Henstenburg has nothing to disclose. Joshua Pezzulo has nothing to disclose. Dominic Farronato has nothing to disclose. Dr. Patel has nothing to disclose. Matthew Sherman has nothing to disclose. Dr. Canseco has nothing to disclose.

Dr. Kepler discloses the following relationships unrelated to submitted work. Biomet: Research support; Clinical Spine Surgery: Editorial or governing board; Medtronic: Research support; Pfizer: Research support; Regeneration; Technologies, Inc.: Research support.

Dr. Vaccaro discloses the following relationships unrelated to submitted work. Advanced Spinal Intellectual Properties: Stock or stock options; Aesculap: IP royalties; AO Spine: Board membership; Atlas Spine: IP royalties; Paid consultant; Avaz Surgical: Stock or stock options; Bonovo Orthopaedics: Stock or stock options; Clinical Spine Surgery: Editorial or governing board; Computational Biodynamics: Stock or stock options; Cytonics: Stock or stock options; DePuy, A Johnson & Johnson Company: Paid consultant; Dimension Orthotics LLC: Stock or stock options; Electrocore: Stock or stock options; Elsevier: Publishing royalties, financial or material support; Flagship Surgical: Stock or stock options; FlowPharma: Stock or stock options; Franklin Bioscience: Stock or stock options; Gamma Spine: Stock or stock options; Gerson Lehrman Group: Paid consultant; Globus Medical: IP royalties; Paid consultant; Stock or stock options; Guidepoint; Global: Paid consultant; Innovative Surgical Design: Paid consultant; Stock or stock options; Insight Therapeutics: Stock or stock options; Jaypee: Publishing royalties, financial or material support; Medtronic: IP royalties; Paid consultant; none: other financial or material support; Nuvasive: Paid consultant; Stock or stock options; Orthobullets: Paid consultant; Paradigm Spine: Stock or stock options; Parvizi Surgical Innovations: Stock or stock options; Prime Surgeons: Stock or stock options; Progressive Spinal Technologies: Stock or stock options; Replication Medica: Stock or stock options; Spine Journal: Editorial or governing board; Spine Medica: Stock or stock options; SpineWave: IP royalties; Paid consultant; Spinology: Stock or stock options; Stout Medical: Paid consultant; Stock or stock options; Stryker: IP royalties; Paid consultant; Taylor Franics/Hodder & Stoughton: Publishing royalties, financial or material support; Thieme: Publishing royalties, financial or material support; Vertiflex: Stock or stock options; Sentryx: Board membership.

References

Olsen

, Mayfield

, Lauryssen

, et al. Risk factors for surgical site infection in spinal surgery. J Neurosurg, 2003; 98(Suppl 2):149–155.

Olsen

, Nepple

, Riew

, et al. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Joint Surg Am, 2008; 90:62–69.

Watanabe

, Sakai

, Matsuyama

, et al. Risk factors for surgical site infection following spine surgery: Efficacy of intraoperative saline irrigation. J Neurosurg Spine, 2010; 12:540–546.

Kwon

, Fisher

, Boyd

, et al. A prospective randomized controlled trial of anterior compared with posterior stabilization for unilateral facet injuries of the cervical spine. J Neurosurg Spine, 2007; 7:1–12.

Bydon

, Xu

, Sciubba

, et al. Safety and efficacy of rhBMP2 in posterior cervical spinal fusion for subaxial degenerative spine disease: Analysis of outcomes in 204 patients. Surg Neurol Int, 2011; 2:109.

Gruskay

, Kepler

, Smith

, Radcliff

, Vaccaro

. Is surgical case order associated with increased infection rate after spine surgery?. Spine, 2012; 37:1170–1174.

Edwards

, Peterson

, Mu

, et al. National Healthcare Safety Network (NHSN) report: Data summary for 2006 through 2008, issued December 2009. Am J Infect Control, 2009; 37:783–805.

Sebastian

, Huddleston

, Kakar

, et al. Risk factors for surgical site infection after posterior cervical spine surgery: An analysis of 5,441 patients from the ACS NSQIP 2005–2012. Spine J, 2016; 16:504–509.

Bernatz

, Anderson

. Thirty-day readmission rates in spine surgery: Systematic review and meta-analysis. Neurosurg Focus, 2015; 39:E7.

10.

McCormack

, Hunter

, Ramos

, et al. An analysis of causes of readmission after spine surgery. Spine, 2012; 37:1260–1266.

11.

Chahoud

, Kanafani

, Kanj

. Surgical site infections following spine surgery: Eliminating the controversies in the diagnosis. Front Med, 2014; 1:7.

12.

Wiley

, Ha'eri

. Routes of infection. A study of using “tracer particles” in the orthopedic operating room. Clin Orthop Relat Res, 1979; 139:150–155.

13.

Donnally

, Sheu

, Bondar

, et al. Is there a correlation between preoperative or postoperative vitamin D levels with pseudarthrosis, hardware failure, and revisions after lumbar spine fusion?. World Neurosurg, 2019; 130:e431–e437.

14.

Meng

, Cao

, Meng

. Risk factors for surgical site infections following spinal surgery. J Clin Neurosci, 2015; 22:1862–1866.

15.

Patel

, Bagan

, Vadera

, et al. Obesity and spine surgery: Relation to perioperative complications. J Neurosurg Spine, 2007; 6:291–297.

16.

, Sucato

, Richards

. Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in adolescent idiopathic scoliosis patients. Spine, 2007; 32:2272–2277.

17.

Vaidya

, Carp

, Bartol

, et al. Lumbar spine fusion in obese and morbidly obese patients. Spine, 2009; 34:495–500.

18.

Mehta

, Babu

, Sharma

, et al. Thickness of subcutaneous fat as a risk factor for infection in cervical spine fusion surgery. J Bone Joint Surg Am, 2013; 95:323–328.

19.

Mehta

, Babu

, Karikari

, et al. 2012 young investigator award winner: The distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine, 2012; 37:1652–1656.

20.

Ganz

, Wintfeld

, Li

, et al. The association of body mass index with the risk of type 2 diabetes: A case-control study nested in an electronic health records system in the United States. Diabetol Metab Syndr, 2014; 6:50.

21.

Nieman

, Henson

, Nehlsen-Cannarella

, et al. Influence of obesity on immune function. J Am Diet Assoc, 1999; 99:294–299.

22.

Milner

, Beck

. The impact of obesity on the immune response to infection. Proc Nutr Soc, 2012; 71:298–306.

23.

Borchardt

, Tzizik

. Update on surgical site infections: The new CDC guidelines. JAAPA, 2018; 31:52–54.

24.

Schisterman

, Perkins

. Confidence intervals for the Youden index and corresponding optimal cut-point. Commun Stat Simul Comput, 2007; 36:549–563.

25.

Luo

, Xiong

. Youden index and associated cut-points for three ordinal diagnostic groups. Commun Stat Simul Comput, 2013; 42:1213–1234.

26.

Nuttall

. Body mass index: Obesity, BMI, and health: A critical review. Nutr Today, 2015; 50:117–128.

27.

Romero-Corral

, Somers

, Sierra-Johnson

, et al. Accuracy of body mass index in diagnosing obesity in the adult general population. Int J Obes, 2008; 32:959–966.

28.

Lee

, Odeh

, Holcombe

, et al. Fat thickness as a risk factor for infection in lumbar spine surgery. Orthopedics, 2016; 39:e1124–e1128.

29.

Ando

, Tamaki

, Yoshida

, et al. Surgical site infection in spinal surgery: A comparative study between 2-octyl-cyanoacrylate and staples for wound closure. Eur Spine J, 2014; 23:854–862.

30.

Strom

, Pacione

, Kalhorn

, et al. Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder. Spine, 2013; 38:991–994.

31.

Choy

, Lam

, Smith

, et al. Predictors of 30-day hospital readmission after posterior cervical fusion in 3401 patients. Spine, 2018; 43:356–363.

32.

Pugely

, Martin

, Gao

, et al. Causes and risk factors for 30-day unplanned readmissions after lumbar Spine surgery. Spine, 2014; 39:761–768.

33.

Kato

, Namikawa

, Matsumura

, Konishi

, et al. Effect of cervical sagittal balance on laminoplasty in patients with cervical myelopathy. Global Spine J, 2017; 7:154–161.

34.

Nakama

, Nitanai

, Oohashi

, et al. Cervical muscle strength after laminoplasty. J Orthop Sci, 2003; 8:36–40.

35.

King

, Abbed

, Gould

, et al. Cervical spine reoperation rates and hospital resource utilization after initial surgery for degenerative cervical spine disease in 12 338 patients in Washington state. Neurosurgery, 2009; 65:1011–1022.

36.

Akamnonu

, Cheriyan

, Goldstein

, et al. Ninety-day readmissions after degenerative cervical spine surgery: A single-center administrative database study. Int J Spine Surg, 2015; 9:19.

37.

Jiang

, Teng

, Fan

, et al. Does obesity affect the surgical outcome and complication rates of spinal surgery? A meta-analysis. Clin Orthop Relat Res, 2014; 472:968–975.

38.

Sebastian

, Huddleston

, Kakar

, et al. Risk factors for surgical site infection after posterior cervical spine surgery: an analysis of 5,441 patients from the ACS NSQIP 2005–2012. Spine J, 2016; 16:504–509.

39.

Abdallah

, Jadaan

, McCabe

. Body mass index and risk of surgical site infection following spine surgery: A meta-analysis. Eur Spine J, 2013; 22:2800–2809.

40.

Bono

, Poorman

, Foster

, et al. Body mass index predicts risk of complications in lumbar spine surgery based on surgical invasiveness. Spine J, 2018; 18:1204–1210.

41.

Khoueir

, Black

, Crookes

, et al. Prospective assessment of axial back pain symptoms before and after bariatric weight reduction surgery. Spine J, 2009; 9:454–463.

42.

Hooper

, Stellato

, Hallowell

, et al. Musculoskeletal findings in obese subjects before and after weight loss following bariatric surgery. Int J Obes, 2007; 31:114–120.

43.

Imahiyerobo

, Minkara

, Matsumoto

, et al. Plastic multilayered closure in pediatric nonidiopathic scoliosis is associated with a lower than expected incidence of wound complications and surgical site infections. Spine Deform, 2018; 6:454–459.

44.

Weissler

, Jenkins

, Hecht

, et al. Plastic surgeon closure of index spinal cases: A single-institution review of 928 cases. Clin Spine Surg, 2019; 32:E397–E402.

45.

, Rozanski

, Taub

. The value of plastic surgery spinal closures: A Review of 782 spine cases. Ann Plast Surg, 2019; 83:201–205.

46.

Mesfin

, Sciubba

, Dea

, et al. Changing the adverse event profile in metastatic spine surgery: An evidence-based approach to target wound complications and instrumentation failure. Spine, 2016; 41:S262–S270.

47.

Dyck

, Bailey

, Steyn

, et al. Use of incisional vacuum-assisted closure in the prevention of postoperative infection in high-risk patients who underwent spine surgery: A proof-of-concept study. J Neurosurg Spine, 2019; 31:430–439.

48.

Adogwa

, Fatemi

, Perez

, et al. Negative pressure wound therapy reduces incidence of postoperative wound infection and dehiscence after long-segment thoracolumbar spinal fusion: A single institutional experience. Spine J, 2014; 14:2911–2917.

49.

Cizik

, Lee

, Martin

, et al. Using the spine surgical invasiveness index to identify risk of surgical site infection: A multivariate analysis. J Bone Joint Surg Am, 2012; 94:335–342.