Accepted Manuscript Risk factors for surgical site infection following spinal surgery: A meta-analysis Qi Fei, Jinjun Li, JiSheng Lin, Dong Li, BingQiang Wang, Hai Meng, Qi Wang, Nan Su, Yong Yang PII:
S1878-8750(15)00656-7
DOI:
10.1016/j.wneu.2015.05.059
Reference:
WNEU 2930
To appear in:
World Neurosurgery
Received Date: 29 April 2015 Revised Date:
29 May 2015
Accepted Date: 30 May 2015
Please cite this article as: Fei Q, Li J, Lin J, Li D, Wang B, Meng H, Wang Q, Su N, Yang Y, Risk factors for surgical site infection following spinal surgery: A meta-analysis, World Neurosurgery (2015), doi: 10.1016/j.wneu.2015.05.059. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Risk factors for surgical site infection following spinal surgery: A meta-analysis
Qi Fei*, Jinjun Li*, JiSheng Lin, Dong Li, BingQiang Wang, Hai Meng, Qi Wang,
RI PT
Nan Su, Yong Yang§,
Department of Orthopaedics, Beijing Friendship Hospital, Capital Medical University,
SC
100050, China. *
§
M AN U
The first two authors contributed equally to the work.
Corresponding author: Yong Yang, Department of Orthopaedics, Beijing Friendship
Hospital, Capital Medical University, No.95, Yong'an Road, Xicheng District， 100050 ， Beijing ， China.
Tel:
86-10-63138353,
86-13701117093;
TE D
86-10-83911029; Email:[email protected]
AC C
EP
：Risk factors for spinal surgery infection: A meta-analysis Running title：
Abbreviations: None
Fax:
ACCEPTED MANUSCRIPT Abstract Background: Surgical site infection (SSI) following spinal surgery is the most common complication, which results in higher morbidity, mortality and healthcare
RI PT
costs. Identifying risk factors of SSI is an important point for preventive strategies to reduce the incidence of SSI. The aim of this meta-analysis is to investigate the most important risk factors for SSI following spinal surgery.
SC
Methods: Pubmed, Embase, Web of Science were systematically searched to identify
M AN U
cohort or case-control studies that investigated the risk factors for SSI following spinal surgery. A fixed-effects or random-effects model was used to pool the estimates, depending on the heterogeneity among the included studies. Heterogeneity between the studies was assessed by I2 and Cochran’s Q test.
TE D
Results: Twelve studies with a total of 13476 patients met the inclusion criteria were included in this meta-analysis. Of them, one was nested case-control studies, seven case-control studies, and four cohort studies. The most important predictors of SSI
EP
were diabetes (RR=2.22, 95%CI: 1.38, 3.60; P=0.001), prolonged operative times (>3
AC C
hours) (RR=2.16, 95%CI: 1.12, 4.19; P=0.009), body mass index (BMI) more than 35(RR=2.36, 95%CI: 1.47, 3.80; P=0.000), and posterior approach (RR=1.22, 95%CI: 1.05, 1.41; P=0.009).
Conclusion: Diabetes, prolonged operative times (>3 hours), BMI more than 35, posterior approach, and number of intervertebral levels(≥7) are associated with an increased risk of SSI after spinal surgery. Almost all these risk factors are in line with the known risk factors for SSI in patients who underwent spinal surgery.
ACCEPTED MANUSCRIPT Keywords: surgical site infection, spinal surgery, risk factors, meta-analysis
Introduction
RI PT
Surgical site infection (SSI) following spinal surgery is the most common complication, which has been reported as a higher infection rate in surgically treated patients, resulting in increased rates of morbidity, mortality and health care costs (1, 9,
SC
13, 14). According to the data from the National Nosocomial Infections Surveillance
M AN U
System, approximately 1.2% of patients who underwent the laminectomy and 2.4% of those who received the spinal fusion surgery developed SSI (8). The rates of SSI reported in the literature have ranged from 0.7 to 12.0% (1, 9, 13, 14), depending on the type of surgery, diagnosis, and he use of instrumentation (23, 29, 31).
TE D
Spinal surgeries are at high risk for infection compared with other orthopaedic procedures. This could be explained by its continuous expanding complexity and increasing number of invasive procedures instead of conservative treatments. Recently,
EP
with the help of improved instrumentation and surgical techniques, spinal procedures
AC C
are available to a wide variety of difficult or complicated diagnoses. However, more complicated procedures lead to higher infection rates since the nature of the procedures are responsible for the variability of the infection risk (6, 17, 31). The simple discectomy or laminectomy has infection rate of 1%, whereas the rates for spinal fusion are 2-5%. The addition of an implant to a spinal fusion increases the infection risk, which range from 2.4 to 8.5%. The combined anterior/posterior procedure is the most frequently infected spinal fusion (6). In addition to the
ACCEPTED MANUSCRIPT surgery-related factors, patients’ characteristics are also associated with increased risk for SSI, including age, smoking, diabetes, obesity, bowel and bladder incontinence, malnutrition (6, 12).
RI PT
Knowing about the risk factors of SSI after spinal surgery can help policy makers to plan preventive strategies to reduce the incidence of SSI. Consequently, we conducted this meta-analysis to identify the most important risk factors for SSI in patients who
SC
underwent spinal surgery. Based upon these identified risk factors, we can develop
M AN U
protocols and apply them in the clinical practice to decrease the risk for SSI, thereby decreasing the morbidity, mortality and healthcare costs.
Materials and methods
TE D
Search strategy and study selection
To identify cohort or case-control studies which evaluated the risk factors for SSI after spinal surgery, we conducted a literature search in Pubmed, Embase, and Web of
EP
Science databases for articles published from dates of inceptions up to March 2, 2015.
AC C
The following search terms were used: ("spinal fusion"[MeSH Terms] OR ("spinal"[All Fields] AND "fusion"[All Fields]) OR "spinal fusion"[All Fields]) AND ("surgical procedures, operative"[MeSH Terms] OR ("surgical"[All Fields] AND "procedures"[All Fields] AND "operative"[All Fields]) OR "operative surgical procedures"[All Fields] OR "surgical"[All Fields]) AND site[All Fields] AND ("infection"[MeSH Terms] OR "infection"[All Fields] OR "infections"[All Fields]) AND ("risk factors"[MeSH Terms] OR ("risk"[All Fields] AND "factors"[All Fields])
ACCEPTED MANUSCRIPT OR "risk factors"[All Fields]). Results were limited to human subjects and no language limitation was imposed. We also searched the reference lists of retrieved studies until no potential articles could be found.
RI PT
We included all the studies when the following inclusion criteria were met: (1) have a cohort or case-control study design and with the risk factors for SSI after spinal surgery as the outcomes; (2) provide odds ratio (OR), risk ratio (RR) or required data
SC
to calculate them with 95% confidence intervals (CIs). Exclusion criteria included
available data of interest.
M AN U
reviews, letters, case reports, systematic reviews and studies that did not report
Data extraction and quality assessment
TE D
The following data were extracted from each of the included studies: lead author’s name, year of publication, origin country, sample size, number of patients in the intervention and control groups, and socio-demographic risk factors of surgical site
EP
infection after spinal fusion. The literature search (Yong Yang and Qi Fei), article
AC C
review (Jinjun Li and Jisheng Lin) and data extraction (Dong Li and Bingqiang Wang) were performed by two independently investigators, and disagreements between them were resolved by discussion and consensus. The methodological quality of the studies included in the meta-analysis was assessed by using the modified Newcastle-Ottawa (NOS) scale (32). The scale consists of three items in the report of a cohort study: patient selection, comparability of the intervention/ control group, and outcome assessment (32). The quality scale ranged
ACCEPTED MANUSCRIPT from 0 to 9 points. Articles are considered as high quality if the NOS scale is more than 5 points.
RI PT
Statistical analysis We explored the risk factors of SSI after spinal surgery based on the data from the twelve studies included in this meta-analysis. The incidences of SSI were treated as
SC
dichotomous variables and were expressed as RR with 95%CI for each study. A
M AN U
fixed-effect model (16) was used to pool the estimates of RR when there was no substantial heterogeneity between the studies, otherwise, a random-effect model (5) was used. Heterogeneity among the studies was tested using I2 statistic (10), in which I2 index >50% or P3 hours) and SSI. The aggregated results using a random-effects model (P=0.000) suggest that patients who prolonged operational time (>3 hours) have two times higher risk for SSI (RR=2.16, 95%CI: 1.12, 4.19; P=0.009) (Figure 3). The Egger’s test (P=0.186) and Begg’s test (P=0.260) revealed no publication bias.
ACCEPTED MANUSCRIPT Six studies (21, 22, 24, 25, 27, 28) report the risk factor of posterior approach for SSI. The aggregated results using a random-effects model (P=0.007) show that patients who underwent surgery using a posterior approach have one time higher risk for SSI
RI PT
(RR=1.22, 95%CI: 1.05, 1.41; P=0.009) (Figure 4). The Egger’s test (P=0.750) and Begg’s test (P=0.707) revealed no publication bias.
Seven studies (6, 18, 20, 24, 25, 27, 28) have evaluated the association between
SC
previous surgery and SSI. The pooled estimates using a random-effects model
M AN U
(P=0.000) demonstrated that previous surgery is not a high risk factor for SSI (RR=1.88, 95%CI: 0.90, 3.91; P=0.092) (Figure 5). The Egger’s test (P=0.646) and Begg’s test (P=0.368) revealed no publication bias.
Six studies (18, 21, 22, 24-26) have assessed the association between obesity and SSI.
TE D
The pooled estimates using a random-effects model (P=0.002) demonstrated that patients whose BMI between 30-35 do not have a high risk for SSI (RR=1.11, 95%CI: 0.63, 1.94; P=0.726), whereas patients with a BMI≥35 have two times higher risk for
EP
SSI (RR=2.36, 95%CI: 1.47, 3.80; P=0.000) (Figure 6). The Egger’s test (P=0.456)
AC C
and Begg’s test (P=0.624) revealed no publication bias. Six studies (6, 20, 21, 25, 27, 30) report the risk factor of smoking for SSI. The aggregated results using a random-effects model (P=0.000) suggest that smoking is not a risk factor for SSI (RR=1.80, 95%CI: 0.70, 4.69; P=0.225) (Figure 7). The Egger’s test (P=0.620) and Begg’s test (P=1.000) revealed no publication bias.
Discussion
ACCEPTED MANUSCRIPT In this meta-analysis, we identified independent risk factors for patients who develop SSI following spinal surgery. Diabetes were the strongest independent risk factor for SSI, and this result is consistent with those other studies that described diabetes as
RI PT
significant risk factor for SSI (20, 25, 28). Prolonged operative time (>3 hours) was associated with two times higher risk of SSI. Furthermore, patients who underwent surgery using a posterior approach have one time higher risk for the development of
SC
SSI. Additionally, patients with a BMI≥35 have two times higher risk for SSI.
M AN U
In our analysis of clinically significant SSI’s, we identified more 20 risk factors that might have an impact on the risk of SSI. With the method of meta-analyses to pool the estimates, we were able to identify five independent factors that increased the risk of developing SSI. Diabetes, surgery with a posterior surgical approach, prolonged
TE D
operative time (>3 hours), morbid obesity (BMI≥35), number of intervertebral level (≥7), increased the risk of SSI while location of surgery (cervical), and BMI (>25, ＜ 30) were associated with a lower risk of SSI.
EP
Among the included studies, several studies have reported that diabetes is a risk factor
AC C
for SSI following spinal surgery (6, 22, 24). Our result is consistent with that of these studies. However, diabetes was not identified as a risk factor for SSI in several other studies (18, 20, 25, 27). There is a possibility that diabetes patients in these studies that found no association between diabetes and increased risk of SSI were well-selected and well-controlled cases. Hikata et al. (11) reported that patients with poorly-controlled diabetes (HbA1c≥7.0%) had a higher risk for the development of SSI after posterior thoracolumbar spinal instrumentation surgeries, as compared with
ACCEPTED MANUSCRIPT those with well-controlled diabetes (6.1≤HbA1c＜7.0%). Moreover, none of these patients who had well-controlled diabetes (6.1≤HbA1c＜7.0%) developed SSI in their operated case series (11).
RI PT
Prolonged operative time (>3 hours) was also found to be significantly associated with an increased risk of SSI in our meta-analysis. However, it was not described as a significant risk factor of SSI in several studies (3, 4, 30). In the study conducted by
SC
Masahiko, et al. (30), 64.3% of patients in the SSI group had an operative time >3
M AN U
hours as compared to 41.6% in the non-SSI group. This result seemed that surgical time was significantly higher in the SSI group. However, when the authors used a multivariate logistic regression analysis to analyze the data, no association was found between the surgical time and increased SSI. Since prolonged operative times increase
TE D
the chance of contamination in the surgical wounds, it is not hard to understand that prolonged surgical time would increase the risk of SSI. Several studies have reported on patients who underwent surgery with a posterior
EP
surgical approach and the incidence of SSI (15, 25, 27, 28, 33). Levi, et al. (15)
AC C
reported that all patients who underwent instrumented posterior spinal fusion had deep wound infections. Similarly, Wimmer, et al. (33) reported that most of the SSIs occurred in patients who received spinal surgery involved posterior fusion and placement of instrumentation. The results of our analysis were in line with these reports. Patients with posterior surgical approach had one time higher risk for the development of SSI. Therefore, reductions in the contamination of the posterior wound with fecal, urinary, or skin flora are very necessary interventions to reduce the
ACCEPTED MANUSCRIPT risk of SSI. Moreover, we should pay much more attention to the antisepsis of posterior wounds, not only in the hospital but also the nursing care at home. Previous surgery was not identified as an independent risk factor for SSI in this
RI PT
meta-analysis. This result was in line with several studies (18, 20, 25, 28). From these results, we hypothesize that it is not the presence of old scar tissue that accounts for the increased risk for SSI. And only those cases with a prior infection in the surgical
SC
field had a higher risk for the development of SSI. In theory, bacteria can be
M AN U
encapsulated in the scar tissue after the occurrence of a SSI. Reoperation on the previously infected wound would release the dormant bacteria into the new wound, thus resulting in a greater higher risk of SSI. Consequently, peri-operative antibiotics should be used to the cases of prior SSI whose treatment beyond the standard 24-h
TE D
prophylaxis to reduce the risk of SSI.
Several studies have reported that obesity is a risk factor for SSI (22, 33). However, obesity was not a significant risk factor for SSI in this meta-analysis. Satoshi, et al.
EP
(20) reported that BMI was not a significant risk factor for SSI according to the
AC C
univariate (OR=0.99, 95%CI: 0.90, 1.11; P=0.88) and multivariate analyses (OR=0.98, 95%CI: 0.88, 1.10; P=0.69). The authors attributed this negative result to the lower prevalence of obesity in Japanese adults, because the proportion of adults with obese (≥30,2007 to 2008) reported by Flegal, et al. (7) was 33.8% in the USA, as compared to 6.1% in this study. Mahta, et al. (18, 19) indicated that the thickness of subcutaneous fat at the surgical site was an important risk for the development of SSI, and they found that it was more significant than BMI in the predicting of SSI.
ACCEPTED MANUSCRIPT Contrary to some reports (6, 27, 30), smoking was not an independent risk factor for SSI in this meta-analysis. Fang, et al. (6) reported that almost 33.3% of patients in the SSI group had a s history of smoking as compared to 16.8% in the non-SSI group
RI PT
(OR=2.47, 95%CI: 1.1, 5.6; P=0.03). Conversely, in another study conducted by Margaret, et al (21), 36.6% of patients developing SSI had a history of smoking as compared to 60.2% in those without SSI (OR=0.4, 95%CI: 0.2, 0.8; P=0.006), which
SC
indicated that smoking decreased risk of SSI. However, this result did not reach a
M AN U
significantly difference after correction for multiple testing.
A weakness of the current meta-analysis is the heterogeneity among the included studies, which might have a potential impact on the results of all meta-analyses. This could be attributed to the characteristics of the patients, study design (cohort,
TE D
case-control study), surgical location, surgical approach, and number of levels. Second, majority of the included studies were conducted based on hospital. This might limit the generalizability of our findings to all the population. However, studies
EP
that identify risk factors for SSI are usually carried out on hospital patients, and this
AC C
limitation is inevitable. Third, majority of these studies did not give a brief descript of the life style risk factors such as smoking (for example, the daily consumption of cigarette) or the alcohol use. Future studies should focus on these aspects of risk factors. Fourth, age is recognized as a most important factor (more than 70) for increase rate of infection following spinal surgery. However, because only one study has assessed the association between age and SSI, we were unable to further investigate the association between age and SSI.
ACCEPTED MANUSCRIPT In conclusion, we identified that, diabetes, prolonged operative time (>3 hours), surgery using a posterior approach, morbid obesity (BMI>35), number of intervertebral level (≥7), were independent risk factors for SSI following spinal
RI PT
surgery in adult patients. Identification of these risk factors could facilitate the design of protocols to decrease the rate of SSI in the future.
SC
Conflict of interest statements
M AN U
All the authors declare that they have no conflict of interest.
Acknowledgement
The study was funded by the Capital Health Development Special Scientific Research
AC C
EP
TE D
Projects (2014-2-2023), Beijing, China.
ACCEPTED MANUSCRIPT References 1.
Abbey, DM, Turner, DM, Warson, JS, Wirt, TC, and Scalley, RD: Treatment of postoperative wound infections following spinal fusion with instrumentation. J
2.
RI PT
Spinal Disord 8: 278-83, 1995. Begg, CB and Mazumdar, M: Operating characteristics of a rank correlation test for publication bias. Biometrics 50: 1088-101, 1994.
Blam, OG, Vaccaro, AR, Vanichkachorn, JS, Albert, TJ, Hilibrand, AS,
SC
3.
M AN U
Minnich, JM, and Murphey, SA: Risk factors for surgical site infection in the patient with spinal injury. Spine (Phila Pa 1976) 28: 1475-80, 2003. 4.
Capen, DA, Calderone, RR, and Green, A: Perioperative risk factors for wound infections after lower back fusions. Orthop Clin North Am 27: 83-6,
5.
TE D
1996.
DerSimonian, R and Laird, N: Meta-analysis in clinical trials. Control Clin Trials 7: 177-88, 1986.
Fang, A, Hu, SS, Endres, N, and Bradford, DS: Risk factors for infection after
EP
6.
7.
AC C
spinal surgery. Spine (Phila Pa 1976) 30: 1460-5, 2005. Flegal, KM, Carroll, MD, Ogden, CL, and Curtin, LR: Prevalence and trends in obesity among US adults, 1999-2008. Jama 303: 235-41, 2010.
8.
Gaynes, RP, Culver, DH, Horan, TC, Edwards, JR, Richards, C, and Tolson, JS: Surgical site infection (SSI) rates in the United States, 1992-1998: the National Nosocomial Infections Surveillance System basic SSI risk index. Clin Infect Dis 33 Suppl 2: S69-77, 2001.
ACCEPTED MANUSCRIPT 9.
Glassman, SD, Dimar, JR, Puno, RM, and Johnson, JR: Salvage of instrumental lumbar fusions complicated by surgical wound infection. Spine (Phila Pa 1976) 21: 2163-9, 1996. Higgins, JP and Thompson, SG: Quantifying heterogeneity in a meta-analysis.
RI PT
10.
Stat Med 21: 1539-58, 2002. 11.
Hikata, T, Iwanami, A, Hosogane, N, Watanabe, K, Ishii, K, Nakamura, M,
SC
Kamata, M, Toyama, Y, and Matsumoto, M: High preoperative hemoglobin
M AN U
A1c is a risk factor for surgical site infection after posterior thoracic and lumbar spinal instrumentation surgery. J Orthop Sci 19: 223-8, 2014. 12.
Ho, C, Sucato, DJ, and Richards, BS: Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in
2007. 13.
TE D
adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976) 32: 2272-7,
Keller, RB and Pappas, AM: Infection after spinal fusion using internal
Kostuik, JP, Israel, J, and Hall, JE: Scoliosis surgery in adults. Clin Orthop
AC C
14.
EP
fixation instrumentation. Orthop Clin North Am 3: 99-111, 1972.
Relat Res: 225-34, 1973.
15.
Levi, AD, Dickman, CA, and Sonntag, VK: Management of postoperative infections after spinal instrumentation. J Neurosurg 86: 975-80, 1997.
16.
Mantel, N and Haenszel, W: Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22: 719-48, 1959.
17.
Massie, JB, Heller, JG, Abitbol, JJ, McPherson, D, and Garfin, SR:
ACCEPTED MANUSCRIPT Postoperative posterior spinal wound infections. Clin Orthop Relat Res: 99-108, 1992. 18.
Mehta, AI, Babu, R, Karikari, IO, Grunch, B, Agarwal, VJ, Owens, TR,
RI PT
Friedman, AH, Bagley, CA, and Gottfried, ON: 2012 Young Investigator Award winner: The distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976) 37:
Mehta, AI, Babu, R, Sharma, R, Karikari, IO, Grunch, BH, Owens, TR,
M AN U
19.
SC
1652-6, 2012.
Agarwal, VJ, Sampson, JH, Lad, SP, Friedman, AH, Kuchibhatla, M, Bagley, CA, and Gottfried, ON: Thickness of subcutaneous fat as a risk factor for infection in cervical spine fusion surgery. J Bone Joint Surg Am 95: 323-8,
20.
TE D
2013.
Ogihara, S, Yamazaki, T, Maruyama, T, Oka, H, Miyoshi, K, Azuma, S, Yamada, T, Murakami, M, Kawamura, N, Hara, N, Terayama, S, Morii, J,
EP
Kato, S, and Tanaka, S: Prospective multicenter surveillance and risk factor
AC C
analysis of deep surgical site infection after posterior thoracic and/or lumbar spinal surgery in adults. J Orthop Sci 20: 71-7, 2015.
21.
Olsen, MA, Mayfield, J, Lauryssen, C, Polish, LB, Jones, M, Vest, J, and Fraser, VJ: Risk factors for surgical site infection in spinal surgery. J
Neurosurg 98: 149-55, 2003. 22.
Olsen, MA, Nepple, JJ, Riew, KD, Lenke, LG, Bridwell, KH, Mayfield, J, and Fraser, VJ: Risk factors for surgical site infection following orthopaedic spinal
ACCEPTED MANUSCRIPT operations. J Bone Joint Surg Am 90: 62-9, 2008. 23.
Picada, R, Winter, RB, Lonstein, JE, Denis, F, Pinto, MR, Smith, MD, and Perra, JH: Postoperative deep wound infection in adults after posterior
Spinal Disord 13: 42-5, 2000. 24.
RI PT
lumbosacral spine fusion with instrumentation: incidence and management. J
Pull ter Gunne, AF and Cohen, DB: Incidence, prevalence, and analysis of risk
SC
factors for surgical site infection following adult spinal surgery. Spine (Phila
25.
M AN U
Pa 1976) 34: 1422-8, 2009.
Pull ter Gunne, AF, van Laarhoven, CJ, and Cohen, DB: Incidence of surgical site infection following adult spinal deformity surgery: an analysis of patient risk. Eur Spine J 19: 982-8, 2010.
Rao, SB, Vasquez, G, Harrop, J, Maltenfort, M, Stein, N, Kaliyadan, G, Klibert,
TE D
26.
F, Epstein, R, Sharan, A, Vaccaro, A, and Flomenberg, P: Risk factors for surgical site infections following spinal fusion procedures: a case-control
Schimmel, JJ, Horsting, PP, de Kleuver, M, Wonders, G, and van Limbeek, J:
AC C
27.
EP
study. Clin Infect Dis 53: 686-92, 2011.
Risk factors for deep surgical site infections after spinal fusion. Eur Spine J 19: 1711-9, 2010.
28.
Sponseller, PD, Shah, SA, Abel, MF, Newton, PO, Letko, L, and Marks, M: Infection rate after spine surgery in cerebral palsy is high and impairs results: multicenter analysis of risk factors and treatment. Clin Orthop Relat Res 468: 711-6, 2010.
ACCEPTED MANUSCRIPT 29.
Tenney, JH, Vlahov, D, Salcman, M, and Ducker, TB: Wide variation in risk of wound infection following clean neurosurgery. Implications for perioperative antibiotic prophylaxis. J Neurosurg 62: 243-7, 1985. Watanabe, M, Sakai, D, Matsuyama, D, Yamamoto, Y, Sato, M, and Mochida,
RI PT
30.
J: Risk factors for surgical site infection following spine surgery: efficacy of intraoperative saline irrigation. J Neurosurg Spine 12: 540-6, 2010.
Weinstein, MA, McCabe, JP, and Cammisa, FP, Jr.: Postoperative spinal
SC
31.
Disord 13: 422-6, 2000. 32.
M AN U
wound infection: a review of 2,391 consecutive index procedures. J Spinal
Wells., G, Shea., B, Connell., DO, Peterson., J, and Welch., V: The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized
TE D
studies in meta-analyses. . 3rd Symposium on Systematic Reviews: Beyond the Basics: 3-5., 2000. 33.
Wimmer, C, Gluch, H, Franzreb, M, and Ogon, M: Predisposing factors for
EP
infection in spine surgery: a survey of 850 spinal procedures. J Spinal Disord
AC C
11: 124-8, 1998.
Figure legends
Figure 1 Eligibility of studies for inclusion in meta-analysis
Figure 2 Meta-analysis of the association between diabetes and SSI
ACCEPTED MANUSCRIPT Figure 3 Meta-analysis of the association between prolonged operation time (>3 hours) and SSI
RI PT
Figure 4 Meta-analysis of the association between posterior approach and SSI
SC
Figure 5 Meta-analysis of the association between previous surgery and SSI
M AN U
Figure 6 Meta-analysis of the association between BMI and SSI
AC C
EP
TE D
Figure 7 Meta-analysis of the association between smoking and SSI
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Characteristics of included studies to assess risk factors of surgical site infection following spinal surgery Author Year Study design Sample Case Control Risk factor size Fang A [9] 2005 Case-control 1629 48 1581 BMI/smoking/previous surgery/diabetes/staged procedure/bone allograft/implanted instrumentation/duration of operation >3hours Ho C [11] 2007 Case-control 126 36 90 Implanted instrumentation Alert FP [17] 2010 Cohort 830 75 755 Obesity/prior surgery/smoking/anterior, posterior approach/duration of operation >3 h Schimmel JJP [18] 2010 Nested 171 36 135 BMI/smoking/previous surgery/diabetes/bone case-control autograft/bone allograft/posterior approach/number of levels Sponseller PD [19] 2010 Case-control 157 16 141 Previous surgery/posterior approach/staged procedure/prophylactic antibiotics Olsen MA [20] 2003 Case-control 229 51 178 Smoking/BMI/ prophylactic antibiotics/steroid therapy/cervical level/anterior/posterior approach/lumbar level Ogihara S [21] 2015 Cohort 2736 24 2712 Diabetes/ smoking/steroid therapy/previous surgery/duration of operation >3 hours of Watanabe M [22] 2010 Cohort 223 14 209 Smoking/ diabetes/BMI/duration operation >3hours/implanted instrumentation Pull ter Gunne AF 2009 Cohort 3174 132 3042 Diabetes/ BMI/previous [23] surgery/anterior/posterior/duration of operation >3hours Mehta AI [24] 2012 Case-control 298 24 274 Diabetes/BMI/previous surgery Olsen MA [25] 2008 Case-control 2316 46 2270 BMI/diabetes/cervical level/prophylactic
NOS scale 8
7 7 8
7 8
8 8 8
8 7
ACCEPTED MANUSCRIPT
Case-control
1587
57
1530
RI PT
2011
SC
Rao SB [26]
antibiotics/steroid therapy/number of levels/duration of operation >3hous BMI/diabetes/prophylactic antibiotics/anterior/posterior approach/bone autograft/bone allograft
AC C
EP
TE D
M AN U
NOS, Newcastle-Ottawa Scale
8
ACCEPTED MANUSCRIPT Table 2 Pooled estimates of RRs obtained from meta-analysis of risk factors of SSI following spinal surgery
EP AC C
P value 0.001 0.092 0.726 0.225 0.009 0.009 0.422 0.007 0.393 0.628 0.645 0.034 0.015 0.535 0.332 0.244 0.182 0.022 0.955 0.123 0.023 0.314 0.425 0.356
SC
RI PT
95%CI 1.38-3.60 0.90-3.91 0.63-1.95 0.70-4.69 1.05-1.41 1.12-4.19 0.14-2.31 0.42-0.88 0.77-1.94 0.38-1.22 0.38-4.87 0.54-0.98 1.18-4.59 0.41-5.61 0.49-1.27 0.51-14.60 0.70-6.61 0.47-0.94 0.72-1.42 0.88-3.04 1.11-4.01 0.19-180.78 0.92-1.23 0.40-12.81
M AN U
RR 2.22 1.88 1.11 1.81 1.22 2.16 0.60 0.61 1.22 0.68 1.35 0.72 2.33 1.51 0.79 2.72 2.15 0.67 1.01 1.63 2.11 5.84 1.06 2.26
TE D
Risk factors Diabetes Previous surgery Obesity Smoking Posterior approach Operation time (>3 hours) Anterior approach Cervical level Thoracolumbar level Lumbar level BMI≤25 BMI>25,35 Lumbosacral level Bone autograft Bone allograft Implanted instrument No. of intervertebral levels(1-2) No. of intervertebral levels(3) No. of intervertebral levels(4-6) No. of intervertebral levels(≥7) Staged procedure Prophylactic antibiotics Steroid therapy
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1878-8750(15)00656-7
DOI:
10.1016/j.wneu.2015.05.059
Reference:
WNEU 2930
To appear in:
World Neurosurgery
Received Date: 29 April 2015 Revised Date:
29 May 2015
Accepted Date: 30 May 2015
Please cite this article as: Fei Q, Li J, Lin J, Li D, Wang B, Meng H, Wang Q, Su N, Yang Y, Risk factors for surgical site infection following spinal surgery: A meta-analysis, World Neurosurgery (2015), doi: 10.1016/j.wneu.2015.05.059. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Risk factors for surgical site infection following spinal surgery: A meta-analysis
Qi Fei*, Jinjun Li*, JiSheng Lin, Dong Li, BingQiang Wang, Hai Meng, Qi Wang,
RI PT
Nan Su, Yong Yang§,
Department of Orthopaedics, Beijing Friendship Hospital, Capital Medical University,
SC
100050, China. *
§
M AN U
The first two authors contributed equally to the work.
Corresponding author: Yong Yang, Department of Orthopaedics, Beijing Friendship
Hospital, Capital Medical University, No.95, Yong'an Road, Xicheng District， 100050 ， Beijing ， China.
Tel:
86-10-63138353,
86-13701117093;
TE D
86-10-83911029; Email:[email protected]
AC C
EP
：Risk factors for spinal surgery infection: A meta-analysis Running title：
Abbreviations: None
Fax:
ACCEPTED MANUSCRIPT Abstract Background: Surgical site infection (SSI) following spinal surgery is the most common complication, which results in higher morbidity, mortality and healthcare
RI PT
costs. Identifying risk factors of SSI is an important point for preventive strategies to reduce the incidence of SSI. The aim of this meta-analysis is to investigate the most important risk factors for SSI following spinal surgery.
SC
Methods: Pubmed, Embase, Web of Science were systematically searched to identify
M AN U
cohort or case-control studies that investigated the risk factors for SSI following spinal surgery. A fixed-effects or random-effects model was used to pool the estimates, depending on the heterogeneity among the included studies. Heterogeneity between the studies was assessed by I2 and Cochran’s Q test.
TE D
Results: Twelve studies with a total of 13476 patients met the inclusion criteria were included in this meta-analysis. Of them, one was nested case-control studies, seven case-control studies, and four cohort studies. The most important predictors of SSI
EP
were diabetes (RR=2.22, 95%CI: 1.38, 3.60; P=0.001), prolonged operative times (>3
AC C
hours) (RR=2.16, 95%CI: 1.12, 4.19; P=0.009), body mass index (BMI) more than 35(RR=2.36, 95%CI: 1.47, 3.80; P=0.000), and posterior approach (RR=1.22, 95%CI: 1.05, 1.41; P=0.009).
Conclusion: Diabetes, prolonged operative times (>3 hours), BMI more than 35, posterior approach, and number of intervertebral levels(≥7) are associated with an increased risk of SSI after spinal surgery. Almost all these risk factors are in line with the known risk factors for SSI in patients who underwent spinal surgery.
ACCEPTED MANUSCRIPT Keywords: surgical site infection, spinal surgery, risk factors, meta-analysis
Introduction
RI PT
Surgical site infection (SSI) following spinal surgery is the most common complication, which has been reported as a higher infection rate in surgically treated patients, resulting in increased rates of morbidity, mortality and health care costs (1, 9,
SC
13, 14). According to the data from the National Nosocomial Infections Surveillance
M AN U
System, approximately 1.2% of patients who underwent the laminectomy and 2.4% of those who received the spinal fusion surgery developed SSI (8). The rates of SSI reported in the literature have ranged from 0.7 to 12.0% (1, 9, 13, 14), depending on the type of surgery, diagnosis, and he use of instrumentation (23, 29, 31).
TE D
Spinal surgeries are at high risk for infection compared with other orthopaedic procedures. This could be explained by its continuous expanding complexity and increasing number of invasive procedures instead of conservative treatments. Recently,
EP
with the help of improved instrumentation and surgical techniques, spinal procedures
AC C
are available to a wide variety of difficult or complicated diagnoses. However, more complicated procedures lead to higher infection rates since the nature of the procedures are responsible for the variability of the infection risk (6, 17, 31). The simple discectomy or laminectomy has infection rate of 1%, whereas the rates for spinal fusion are 2-5%. The addition of an implant to a spinal fusion increases the infection risk, which range from 2.4 to 8.5%. The combined anterior/posterior procedure is the most frequently infected spinal fusion (6). In addition to the
ACCEPTED MANUSCRIPT surgery-related factors, patients’ characteristics are also associated with increased risk for SSI, including age, smoking, diabetes, obesity, bowel and bladder incontinence, malnutrition (6, 12).
RI PT
Knowing about the risk factors of SSI after spinal surgery can help policy makers to plan preventive strategies to reduce the incidence of SSI. Consequently, we conducted this meta-analysis to identify the most important risk factors for SSI in patients who
SC
underwent spinal surgery. Based upon these identified risk factors, we can develop
M AN U
protocols and apply them in the clinical practice to decrease the risk for SSI, thereby decreasing the morbidity, mortality and healthcare costs.
Materials and methods
TE D
Search strategy and study selection
To identify cohort or case-control studies which evaluated the risk factors for SSI after spinal surgery, we conducted a literature search in Pubmed, Embase, and Web of
EP
Science databases for articles published from dates of inceptions up to March 2, 2015.
AC C
The following search terms were used: ("spinal fusion"[MeSH Terms] OR ("spinal"[All Fields] AND "fusion"[All Fields]) OR "spinal fusion"[All Fields]) AND ("surgical procedures, operative"[MeSH Terms] OR ("surgical"[All Fields] AND "procedures"[All Fields] AND "operative"[All Fields]) OR "operative surgical procedures"[All Fields] OR "surgical"[All Fields]) AND site[All Fields] AND ("infection"[MeSH Terms] OR "infection"[All Fields] OR "infections"[All Fields]) AND ("risk factors"[MeSH Terms] OR ("risk"[All Fields] AND "factors"[All Fields])
ACCEPTED MANUSCRIPT OR "risk factors"[All Fields]). Results were limited to human subjects and no language limitation was imposed. We also searched the reference lists of retrieved studies until no potential articles could be found.
RI PT
We included all the studies when the following inclusion criteria were met: (1) have a cohort or case-control study design and with the risk factors for SSI after spinal surgery as the outcomes; (2) provide odds ratio (OR), risk ratio (RR) or required data
SC
to calculate them with 95% confidence intervals (CIs). Exclusion criteria included
available data of interest.
M AN U
reviews, letters, case reports, systematic reviews and studies that did not report
Data extraction and quality assessment
TE D
The following data were extracted from each of the included studies: lead author’s name, year of publication, origin country, sample size, number of patients in the intervention and control groups, and socio-demographic risk factors of surgical site
EP
infection after spinal fusion. The literature search (Yong Yang and Qi Fei), article
AC C
review (Jinjun Li and Jisheng Lin) and data extraction (Dong Li and Bingqiang Wang) were performed by two independently investigators, and disagreements between them were resolved by discussion and consensus. The methodological quality of the studies included in the meta-analysis was assessed by using the modified Newcastle-Ottawa (NOS) scale (32). The scale consists of three items in the report of a cohort study: patient selection, comparability of the intervention/ control group, and outcome assessment (32). The quality scale ranged
ACCEPTED MANUSCRIPT from 0 to 9 points. Articles are considered as high quality if the NOS scale is more than 5 points.
RI PT
Statistical analysis We explored the risk factors of SSI after spinal surgery based on the data from the twelve studies included in this meta-analysis. The incidences of SSI were treated as
SC
dichotomous variables and were expressed as RR with 95%CI for each study. A
M AN U
fixed-effect model (16) was used to pool the estimates of RR when there was no substantial heterogeneity between the studies, otherwise, a random-effect model (5) was used. Heterogeneity among the studies was tested using I2 statistic (10), in which I2 index >50% or P3 hours) and SSI. The aggregated results using a random-effects model (P=0.000) suggest that patients who prolonged operational time (>3 hours) have two times higher risk for SSI (RR=2.16, 95%CI: 1.12, 4.19; P=0.009) (Figure 3). The Egger’s test (P=0.186) and Begg’s test (P=0.260) revealed no publication bias.
ACCEPTED MANUSCRIPT Six studies (21, 22, 24, 25, 27, 28) report the risk factor of posterior approach for SSI. The aggregated results using a random-effects model (P=0.007) show that patients who underwent surgery using a posterior approach have one time higher risk for SSI
RI PT
(RR=1.22, 95%CI: 1.05, 1.41; P=0.009) (Figure 4). The Egger’s test (P=0.750) and Begg’s test (P=0.707) revealed no publication bias.
Seven studies (6, 18, 20, 24, 25, 27, 28) have evaluated the association between
SC
previous surgery and SSI. The pooled estimates using a random-effects model
M AN U
(P=0.000) demonstrated that previous surgery is not a high risk factor for SSI (RR=1.88, 95%CI: 0.90, 3.91; P=0.092) (Figure 5). The Egger’s test (P=0.646) and Begg’s test (P=0.368) revealed no publication bias.
Six studies (18, 21, 22, 24-26) have assessed the association between obesity and SSI.
TE D
The pooled estimates using a random-effects model (P=0.002) demonstrated that patients whose BMI between 30-35 do not have a high risk for SSI (RR=1.11, 95%CI: 0.63, 1.94; P=0.726), whereas patients with a BMI≥35 have two times higher risk for
EP
SSI (RR=2.36, 95%CI: 1.47, 3.80; P=0.000) (Figure 6). The Egger’s test (P=0.456)
AC C
and Begg’s test (P=0.624) revealed no publication bias. Six studies (6, 20, 21, 25, 27, 30) report the risk factor of smoking for SSI. The aggregated results using a random-effects model (P=0.000) suggest that smoking is not a risk factor for SSI (RR=1.80, 95%CI: 0.70, 4.69; P=0.225) (Figure 7). The Egger’s test (P=0.620) and Begg’s test (P=1.000) revealed no publication bias.
Discussion
ACCEPTED MANUSCRIPT In this meta-analysis, we identified independent risk factors for patients who develop SSI following spinal surgery. Diabetes were the strongest independent risk factor for SSI, and this result is consistent with those other studies that described diabetes as
RI PT
significant risk factor for SSI (20, 25, 28). Prolonged operative time (>3 hours) was associated with two times higher risk of SSI. Furthermore, patients who underwent surgery using a posterior approach have one time higher risk for the development of
SC
SSI. Additionally, patients with a BMI≥35 have two times higher risk for SSI.
M AN U
In our analysis of clinically significant SSI’s, we identified more 20 risk factors that might have an impact on the risk of SSI. With the method of meta-analyses to pool the estimates, we were able to identify five independent factors that increased the risk of developing SSI. Diabetes, surgery with a posterior surgical approach, prolonged
TE D
operative time (>3 hours), morbid obesity (BMI≥35), number of intervertebral level (≥7), increased the risk of SSI while location of surgery (cervical), and BMI (>25, ＜ 30) were associated with a lower risk of SSI.
EP
Among the included studies, several studies have reported that diabetes is a risk factor
AC C
for SSI following spinal surgery (6, 22, 24). Our result is consistent with that of these studies. However, diabetes was not identified as a risk factor for SSI in several other studies (18, 20, 25, 27). There is a possibility that diabetes patients in these studies that found no association between diabetes and increased risk of SSI were well-selected and well-controlled cases. Hikata et al. (11) reported that patients with poorly-controlled diabetes (HbA1c≥7.0%) had a higher risk for the development of SSI after posterior thoracolumbar spinal instrumentation surgeries, as compared with
ACCEPTED MANUSCRIPT those with well-controlled diabetes (6.1≤HbA1c＜7.0%). Moreover, none of these patients who had well-controlled diabetes (6.1≤HbA1c＜7.0%) developed SSI in their operated case series (11).
RI PT
Prolonged operative time (>3 hours) was also found to be significantly associated with an increased risk of SSI in our meta-analysis. However, it was not described as a significant risk factor of SSI in several studies (3, 4, 30). In the study conducted by
SC
Masahiko, et al. (30), 64.3% of patients in the SSI group had an operative time >3
M AN U
hours as compared to 41.6% in the non-SSI group. This result seemed that surgical time was significantly higher in the SSI group. However, when the authors used a multivariate logistic regression analysis to analyze the data, no association was found between the surgical time and increased SSI. Since prolonged operative times increase
TE D
the chance of contamination in the surgical wounds, it is not hard to understand that prolonged surgical time would increase the risk of SSI. Several studies have reported on patients who underwent surgery with a posterior
EP
surgical approach and the incidence of SSI (15, 25, 27, 28, 33). Levi, et al. (15)
AC C
reported that all patients who underwent instrumented posterior spinal fusion had deep wound infections. Similarly, Wimmer, et al. (33) reported that most of the SSIs occurred in patients who received spinal surgery involved posterior fusion and placement of instrumentation. The results of our analysis were in line with these reports. Patients with posterior surgical approach had one time higher risk for the development of SSI. Therefore, reductions in the contamination of the posterior wound with fecal, urinary, or skin flora are very necessary interventions to reduce the
ACCEPTED MANUSCRIPT risk of SSI. Moreover, we should pay much more attention to the antisepsis of posterior wounds, not only in the hospital but also the nursing care at home. Previous surgery was not identified as an independent risk factor for SSI in this
RI PT
meta-analysis. This result was in line with several studies (18, 20, 25, 28). From these results, we hypothesize that it is not the presence of old scar tissue that accounts for the increased risk for SSI. And only those cases with a prior infection in the surgical
SC
field had a higher risk for the development of SSI. In theory, bacteria can be
M AN U
encapsulated in the scar tissue after the occurrence of a SSI. Reoperation on the previously infected wound would release the dormant bacteria into the new wound, thus resulting in a greater higher risk of SSI. Consequently, peri-operative antibiotics should be used to the cases of prior SSI whose treatment beyond the standard 24-h
TE D
prophylaxis to reduce the risk of SSI.
Several studies have reported that obesity is a risk factor for SSI (22, 33). However, obesity was not a significant risk factor for SSI in this meta-analysis. Satoshi, et al.
EP
(20) reported that BMI was not a significant risk factor for SSI according to the
AC C
univariate (OR=0.99, 95%CI: 0.90, 1.11; P=0.88) and multivariate analyses (OR=0.98, 95%CI: 0.88, 1.10; P=0.69). The authors attributed this negative result to the lower prevalence of obesity in Japanese adults, because the proportion of adults with obese (≥30,2007 to 2008) reported by Flegal, et al. (7) was 33.8% in the USA, as compared to 6.1% in this study. Mahta, et al. (18, 19) indicated that the thickness of subcutaneous fat at the surgical site was an important risk for the development of SSI, and they found that it was more significant than BMI in the predicting of SSI.
ACCEPTED MANUSCRIPT Contrary to some reports (6, 27, 30), smoking was not an independent risk factor for SSI in this meta-analysis. Fang, et al. (6) reported that almost 33.3% of patients in the SSI group had a s history of smoking as compared to 16.8% in the non-SSI group
RI PT
(OR=2.47, 95%CI: 1.1, 5.6; P=0.03). Conversely, in another study conducted by Margaret, et al (21), 36.6% of patients developing SSI had a history of smoking as compared to 60.2% in those without SSI (OR=0.4, 95%CI: 0.2, 0.8; P=0.006), which
SC
indicated that smoking decreased risk of SSI. However, this result did not reach a
M AN U
significantly difference after correction for multiple testing.
A weakness of the current meta-analysis is the heterogeneity among the included studies, which might have a potential impact on the results of all meta-analyses. This could be attributed to the characteristics of the patients, study design (cohort,
TE D
case-control study), surgical location, surgical approach, and number of levels. Second, majority of the included studies were conducted based on hospital. This might limit the generalizability of our findings to all the population. However, studies
EP
that identify risk factors for SSI are usually carried out on hospital patients, and this
AC C
limitation is inevitable. Third, majority of these studies did not give a brief descript of the life style risk factors such as smoking (for example, the daily consumption of cigarette) or the alcohol use. Future studies should focus on these aspects of risk factors. Fourth, age is recognized as a most important factor (more than 70) for increase rate of infection following spinal surgery. However, because only one study has assessed the association between age and SSI, we were unable to further investigate the association between age and SSI.
ACCEPTED MANUSCRIPT In conclusion, we identified that, diabetes, prolonged operative time (>3 hours), surgery using a posterior approach, morbid obesity (BMI>35), number of intervertebral level (≥7), were independent risk factors for SSI following spinal
RI PT
surgery in adult patients. Identification of these risk factors could facilitate the design of protocols to decrease the rate of SSI in the future.
SC
Conflict of interest statements
M AN U
All the authors declare that they have no conflict of interest.
Acknowledgement
The study was funded by the Capital Health Development Special Scientific Research
AC C
EP
TE D
Projects (2014-2-2023), Beijing, China.
ACCEPTED MANUSCRIPT References 1.
Abbey, DM, Turner, DM, Warson, JS, Wirt, TC, and Scalley, RD: Treatment of postoperative wound infections following spinal fusion with instrumentation. J
2.
RI PT
Spinal Disord 8: 278-83, 1995. Begg, CB and Mazumdar, M: Operating characteristics of a rank correlation test for publication bias. Biometrics 50: 1088-101, 1994.
Blam, OG, Vaccaro, AR, Vanichkachorn, JS, Albert, TJ, Hilibrand, AS,
SC
3.
M AN U
Minnich, JM, and Murphey, SA: Risk factors for surgical site infection in the patient with spinal injury. Spine (Phila Pa 1976) 28: 1475-80, 2003. 4.
Capen, DA, Calderone, RR, and Green, A: Perioperative risk factors for wound infections after lower back fusions. Orthop Clin North Am 27: 83-6,
5.
TE D
1996.
DerSimonian, R and Laird, N: Meta-analysis in clinical trials. Control Clin Trials 7: 177-88, 1986.
Fang, A, Hu, SS, Endres, N, and Bradford, DS: Risk factors for infection after
EP
6.
7.
AC C
spinal surgery. Spine (Phila Pa 1976) 30: 1460-5, 2005. Flegal, KM, Carroll, MD, Ogden, CL, and Curtin, LR: Prevalence and trends in obesity among US adults, 1999-2008. Jama 303: 235-41, 2010.
8.
Gaynes, RP, Culver, DH, Horan, TC, Edwards, JR, Richards, C, and Tolson, JS: Surgical site infection (SSI) rates in the United States, 1992-1998: the National Nosocomial Infections Surveillance System basic SSI risk index. Clin Infect Dis 33 Suppl 2: S69-77, 2001.
ACCEPTED MANUSCRIPT 9.
Glassman, SD, Dimar, JR, Puno, RM, and Johnson, JR: Salvage of instrumental lumbar fusions complicated by surgical wound infection. Spine (Phila Pa 1976) 21: 2163-9, 1996. Higgins, JP and Thompson, SG: Quantifying heterogeneity in a meta-analysis.
RI PT
10.
Stat Med 21: 1539-58, 2002. 11.
Hikata, T, Iwanami, A, Hosogane, N, Watanabe, K, Ishii, K, Nakamura, M,
SC
Kamata, M, Toyama, Y, and Matsumoto, M: High preoperative hemoglobin
M AN U
A1c is a risk factor for surgical site infection after posterior thoracic and lumbar spinal instrumentation surgery. J Orthop Sci 19: 223-8, 2014. 12.
Ho, C, Sucato, DJ, and Richards, BS: Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in
2007. 13.
TE D
adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976) 32: 2272-7,
Keller, RB and Pappas, AM: Infection after spinal fusion using internal
Kostuik, JP, Israel, J, and Hall, JE: Scoliosis surgery in adults. Clin Orthop
AC C
14.
EP
fixation instrumentation. Orthop Clin North Am 3: 99-111, 1972.
Relat Res: 225-34, 1973.
15.
Levi, AD, Dickman, CA, and Sonntag, VK: Management of postoperative infections after spinal instrumentation. J Neurosurg 86: 975-80, 1997.
16.
Mantel, N and Haenszel, W: Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22: 719-48, 1959.
17.
Massie, JB, Heller, JG, Abitbol, JJ, McPherson, D, and Garfin, SR:
ACCEPTED MANUSCRIPT Postoperative posterior spinal wound infections. Clin Orthop Relat Res: 99-108, 1992. 18.
Mehta, AI, Babu, R, Karikari, IO, Grunch, B, Agarwal, VJ, Owens, TR,
RI PT
Friedman, AH, Bagley, CA, and Gottfried, ON: 2012 Young Investigator Award winner: The distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976) 37:
Mehta, AI, Babu, R, Sharma, R, Karikari, IO, Grunch, BH, Owens, TR,
M AN U
19.
SC
1652-6, 2012.
Agarwal, VJ, Sampson, JH, Lad, SP, Friedman, AH, Kuchibhatla, M, Bagley, CA, and Gottfried, ON: Thickness of subcutaneous fat as a risk factor for infection in cervical spine fusion surgery. J Bone Joint Surg Am 95: 323-8,
20.
TE D
2013.
Ogihara, S, Yamazaki, T, Maruyama, T, Oka, H, Miyoshi, K, Azuma, S, Yamada, T, Murakami, M, Kawamura, N, Hara, N, Terayama, S, Morii, J,
EP
Kato, S, and Tanaka, S: Prospective multicenter surveillance and risk factor
AC C
analysis of deep surgical site infection after posterior thoracic and/or lumbar spinal surgery in adults. J Orthop Sci 20: 71-7, 2015.
21.
Olsen, MA, Mayfield, J, Lauryssen, C, Polish, LB, Jones, M, Vest, J, and Fraser, VJ: Risk factors for surgical site infection in spinal surgery. J
Neurosurg 98: 149-55, 2003. 22.
Olsen, MA, Nepple, JJ, Riew, KD, Lenke, LG, Bridwell, KH, Mayfield, J, and Fraser, VJ: Risk factors for surgical site infection following orthopaedic spinal
ACCEPTED MANUSCRIPT operations. J Bone Joint Surg Am 90: 62-9, 2008. 23.
Picada, R, Winter, RB, Lonstein, JE, Denis, F, Pinto, MR, Smith, MD, and Perra, JH: Postoperative deep wound infection in adults after posterior
Spinal Disord 13: 42-5, 2000. 24.
RI PT
lumbosacral spine fusion with instrumentation: incidence and management. J
Pull ter Gunne, AF and Cohen, DB: Incidence, prevalence, and analysis of risk
SC
factors for surgical site infection following adult spinal surgery. Spine (Phila
25.
M AN U
Pa 1976) 34: 1422-8, 2009.
Pull ter Gunne, AF, van Laarhoven, CJ, and Cohen, DB: Incidence of surgical site infection following adult spinal deformity surgery: an analysis of patient risk. Eur Spine J 19: 982-8, 2010.
Rao, SB, Vasquez, G, Harrop, J, Maltenfort, M, Stein, N, Kaliyadan, G, Klibert,
TE D
26.
F, Epstein, R, Sharan, A, Vaccaro, A, and Flomenberg, P: Risk factors for surgical site infections following spinal fusion procedures: a case-control
Schimmel, JJ, Horsting, PP, de Kleuver, M, Wonders, G, and van Limbeek, J:
AC C
27.
EP
study. Clin Infect Dis 53: 686-92, 2011.
Risk factors for deep surgical site infections after spinal fusion. Eur Spine J 19: 1711-9, 2010.
28.
Sponseller, PD, Shah, SA, Abel, MF, Newton, PO, Letko, L, and Marks, M: Infection rate after spine surgery in cerebral palsy is high and impairs results: multicenter analysis of risk factors and treatment. Clin Orthop Relat Res 468: 711-6, 2010.
ACCEPTED MANUSCRIPT 29.
Tenney, JH, Vlahov, D, Salcman, M, and Ducker, TB: Wide variation in risk of wound infection following clean neurosurgery. Implications for perioperative antibiotic prophylaxis. J Neurosurg 62: 243-7, 1985. Watanabe, M, Sakai, D, Matsuyama, D, Yamamoto, Y, Sato, M, and Mochida,
RI PT
30.
J: Risk factors for surgical site infection following spine surgery: efficacy of intraoperative saline irrigation. J Neurosurg Spine 12: 540-6, 2010.
Weinstein, MA, McCabe, JP, and Cammisa, FP, Jr.: Postoperative spinal
SC
31.
Disord 13: 422-6, 2000. 32.
M AN U
wound infection: a review of 2,391 consecutive index procedures. J Spinal
Wells., G, Shea., B, Connell., DO, Peterson., J, and Welch., V: The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized
TE D
studies in meta-analyses. . 3rd Symposium on Systematic Reviews: Beyond the Basics: 3-5., 2000. 33.
Wimmer, C, Gluch, H, Franzreb, M, and Ogon, M: Predisposing factors for
EP
infection in spine surgery: a survey of 850 spinal procedures. J Spinal Disord
AC C
11: 124-8, 1998.
Figure legends
Figure 1 Eligibility of studies for inclusion in meta-analysis
Figure 2 Meta-analysis of the association between diabetes and SSI
ACCEPTED MANUSCRIPT Figure 3 Meta-analysis of the association between prolonged operation time (>3 hours) and SSI
RI PT
Figure 4 Meta-analysis of the association between posterior approach and SSI
SC
Figure 5 Meta-analysis of the association between previous surgery and SSI
M AN U
Figure 6 Meta-analysis of the association between BMI and SSI
AC C
EP
TE D
Figure 7 Meta-analysis of the association between smoking and SSI
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Characteristics of included studies to assess risk factors of surgical site infection following spinal surgery Author Year Study design Sample Case Control Risk factor size Fang A [9] 2005 Case-control 1629 48 1581 BMI/smoking/previous surgery/diabetes/staged procedure/bone allograft/implanted instrumentation/duration of operation >3hours Ho C [11] 2007 Case-control 126 36 90 Implanted instrumentation Alert FP [17] 2010 Cohort 830 75 755 Obesity/prior surgery/smoking/anterior, posterior approach/duration of operation >3 h Schimmel JJP [18] 2010 Nested 171 36 135 BMI/smoking/previous surgery/diabetes/bone case-control autograft/bone allograft/posterior approach/number of levels Sponseller PD [19] 2010 Case-control 157 16 141 Previous surgery/posterior approach/staged procedure/prophylactic antibiotics Olsen MA [20] 2003 Case-control 229 51 178 Smoking/BMI/ prophylactic antibiotics/steroid therapy/cervical level/anterior/posterior approach/lumbar level Ogihara S [21] 2015 Cohort 2736 24 2712 Diabetes/ smoking/steroid therapy/previous surgery/duration of operation >3 hours of Watanabe M [22] 2010 Cohort 223 14 209 Smoking/ diabetes/BMI/duration operation >3hours/implanted instrumentation Pull ter Gunne AF 2009 Cohort 3174 132 3042 Diabetes/ BMI/previous [23] surgery/anterior/posterior/duration of operation >3hours Mehta AI [24] 2012 Case-control 298 24 274 Diabetes/BMI/previous surgery Olsen MA [25] 2008 Case-control 2316 46 2270 BMI/diabetes/cervical level/prophylactic
NOS scale 8
7 7 8
7 8
8 8 8
8 7
ACCEPTED MANUSCRIPT
Case-control
1587
57
1530
RI PT
2011
SC
Rao SB [26]
antibiotics/steroid therapy/number of levels/duration of operation >3hous BMI/diabetes/prophylactic antibiotics/anterior/posterior approach/bone autograft/bone allograft
AC C
EP
TE D
M AN U
NOS, Newcastle-Ottawa Scale
8
ACCEPTED MANUSCRIPT Table 2 Pooled estimates of RRs obtained from meta-analysis of risk factors of SSI following spinal surgery
EP AC C
P value 0.001 0.092 0.726 0.225 0.009 0.009 0.422 0.007 0.393 0.628 0.645 0.034 0.015 0.535 0.332 0.244 0.182 0.022 0.955 0.123 0.023 0.314 0.425 0.356
SC
RI PT
95%CI 1.38-3.60 0.90-3.91 0.63-1.95 0.70-4.69 1.05-1.41 1.12-4.19 0.14-2.31 0.42-0.88 0.77-1.94 0.38-1.22 0.38-4.87 0.54-0.98 1.18-4.59 0.41-5.61 0.49-1.27 0.51-14.60 0.70-6.61 0.47-0.94 0.72-1.42 0.88-3.04 1.11-4.01 0.19-180.78 0.92-1.23 0.40-12.81
M AN U
RR 2.22 1.88 1.11 1.81 1.22 2.16 0.60 0.61 1.22 0.68 1.35 0.72 2.33 1.51 0.79 2.72 2.15 0.67 1.01 1.63 2.11 5.84 1.06 2.26
TE D
Risk factors Diabetes Previous surgery Obesity Smoking Posterior approach Operation time (>3 hours) Anterior approach Cervical level Thoracolumbar level Lumbar level BMI≤25 BMI>25,35 Lumbosacral level Bone autograft Bone allograft Implanted instrument No. of intervertebral levels(1-2) No. of intervertebral levels(3) No. of intervertebral levels(4-6) No. of intervertebral levels(≥7) Staged procedure Prophylactic antibiotics Steroid therapy
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
