J Neurosurg Spine 21:648–652, 2014 ©AANS, 2014
Risk factors for delayed infections after spinal fusion and instrumentation in patients with scoliosis Clinical article Jianxiong Shen, M.D.,1 Jinqian Liang, M.D.,1 Haiquan Yu, M.D., 2 Guixing Qiu, M.D.,1 Xuhong Xue, M.D.,1 and Zheng Li, M.D.1 Department of Orthopaedic Surgery, Peking Union Medical College Hospital, Beijing; and 2Department of Orthopaedic Surgery, Shi Jiazhuang Center Hospital, Shi Jiazhuang, Hebei, China
1
Object. There are limited published data about the risk factors for the development of delayed infections after spinal fusion and instrumentation in the population with scoliosis. The objective of this study was to evaluate the predictive factors of development of delayed infections in patients with scoliosis who underwent surgical treatment. Methods. A total of 17 patients with scoliosis and delayed infections were identified from 3463 patients with scoliosis who received surgical treatment. The control group was composed of 85 patients with scoliosis without infections, matched for sex, age, approximate date of surgery, and diagnosis. These 2 groups were compared for demographic distribution and clinical data to investigate the predictive factors of delayed infections. Results. The overall incidence rate of delayed infections was 0.49%. The variables of age, body mass index, and number of levels fused were similar between the 2 groups. The average primary curve magnitude for the delayed infection and control (uninfected) groups was 80.4° ± 27.0° (range 47°–135°) and 66.3° ± 11.6° (range 42°–95°), respectively (p = 0.001). Operation time in the group with delayed infections was 384.7 ± 115.9 minutes versus 254.4 ± 79.2 minutes in the control group (p = 0.000), and estimated blood loss was 1342.2 ± 707.2 ml versus 833.9 ± 235.6 ml (p = 0.000) in these 2 groups, respectively. The perioperative mean red blood cell transfusion requirement in the delayed infection group was significantly higher than that found in patients without infections (2.8 ± 2.3 units/patient versus 1.1 ± 1.6 units/patient, respectively; p = 0.000). Logistic regression analysis showed that operation time and allogenic blood transfusion were the 2 independent predictors of delayed infections (odds ratio [OR] 1.021, 95% confidence interval [CI] 1.010–1.033, and OR 1.546, 95% CI 1.048–2.278, respectively). Conclusions. The occurrence of a delayed infection in patients with scoliosis who undergo surgical treatment is most likely multifactorial and is related to surgical time and the use of allogenic blood transfusion. (http://thejns.org/doi/abs/10.3171/2014.6.SPINE13636)
T
Key Words • delayed infection • spinal fusion • scoliosis • risk factor
identification and quantification of risk factors for delayed infections after spine surgery are of paramount importance to the patient and the clinician. In addition to its obvious importance for patient safety, risk factor information becomes critical as health care policy makers implement and enforce “quality” metrics. Numerous authors have reported potential risk factors for postoperative infections after pediatric spinal deformity surgery. Certain patient-related risk factors, such as underlying medical conditions and previous surgeries, are not modifiable. Recent literature has focused on identifying modifiable risk factors and prevention strategies that may decrease the rates of wound infections after spinal fusion. Perioperative factors such as cognitive impairment, use of an allograft in patients with neuromuscular scoliosis, use of preoperative skeletal traction, age greater he
Abbreviations used in this paper: BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; OR = odds ratio.
648
than 20 years, and higher intraoperative blood loss have been reported as risk factors for infection.6,16,25 However, there are limited published data about the risk factors for the development of delayed infections after spinal fusion and instrumentation in the population with scoliosis. This retrospective cohort study was undertaken to investigate 1) the incidence of delayed infections in a larger population with scoliosis treated using spinal fusion and instrumentation, and 2) the predictive factors of the development of delayed infections after spine surgery in unselected patients with scoliosis who underwent spinal deformity surgery.
Methods Study Population
We examined data from a consecutive series of 3463 patients who had undergone scoliosis surgery between January 1, 1997, and January 1, 2011, at 1 academic hospiJ Neurosurg: Spine / Volume 21 / October 2014
Risk factors for delayed infections in scoliosis patients tal (Peking Union Medical College Hospital), a universitybased medical center. Those patients who had a delayed infection, defined as an infection of the incision occurring more than 6 months from the primary surgery, were identified. Only those patients who underwent a primary spinal deformity surgery and had a minimum of 2 years of follow-up were reviewed. All patients with delayed infections were identified, and these patients with scoliosis without infection were chosen and best matched for sex, age, approximate date of surgery, and diagnosis with the patients with infections, and then matched in a 5:1 ratio. Operative Procedures
In all patients, a Cell Saver system (Haemonetics Cell Saver 5, Haemonetics Corp.) was used intraoperatively to re-infuse shed blood. The volume of blood collected by the Cell Saver system was recorded, and estimated blood loss (EBL) was calculated by measuring loss through Cell Saver suction and by weighing the surgical sponges. The surgery was performed on the patient under hypotensive anesthesia in which systolic blood pressure was maintained at less than 90 mm Hg. The same blood transfusion and prophylactic antibiotic guidelines were used for all patients. Allogenic blood transfusion was performed if hemoglobin decreased to less than 7.0 mg/dl or if anemic symptoms developed, such as a decrease in blood pressure to less than 100 mm Hg systolic, tachycardia greater than 100 beats/min, or a low urine output of less than 30 ml/hr, even after initial fluid challenge with 500 ml normal saline in patients with a hemoglobin level between 7.0 and 8.0 mg/dl.18,20 All patients were given a dose of 1 g of first-generation cephalosporin 30 minutes before skin incision and were covered for 24 hours with 3 additional doses.
Predictive Variables
The treatment of all infections was complete by the time of the study. Variables studied for their significance as preoperative predictors of delayed infections included age at surgery, sex, height, weight, body mass index (BMI), previous medical history, and curve magnitude. Intraoperative factors studied in relation to infection included surgical time, use of allograft or autograft, amount and type of blood transfusion, intraoperative blood loss, number of levels fused, use of drains, and postoperative hemoglobin/hematocrit levels.
Statistical Analysis
Factors associated with the occurrence of delayed infections were identified using univariate analysis. The data analysis was performed using SPSS version 19.0 (IBM Corp.). Continuous data were compared between the 2 groups using the Student t-test, whereas discontinuous data were analyzed using the chi-square test. The Fisher’s exact test was used for small data subsets (n < 5). All significance tests were 2-tailed, with p < 0.05 representing statistical significance. In addition, a multivariate logistic regression analysis was performed to identify which factors helped predict the probability of a delayed infection.
J Neurosurg: Spine / Volume 21 / October 2014
Results Demographic Data
A summary of the clinical and radiographic data before spinal fusion for the delayed infection and control (uninfected) groups is presented in Table 1. Of the 17 patients identified as having wound infections, 3 had congenital scoliosis, 4 had idiopathic scoliosis, 7 had adult scoliosis, and 3 had neuromuscular scoliosis. Five of the 17 patients used stainless-steel instrumentation and the other 12 used titanium instrumentation. An additional 85 patients without infections matched for age, diagnosis, and year of surgery were also evaluated, resulting in 17 patients with congenital scoliosis, 21 with idiopathic scoliosis, 34 with adult scoliosis, and 13 with neuromuscular scoliosis. The mean age at surgery was 23.0 years (range 6–57 years) for the patients in whom infection developed and 18.6 years (range 8–55 years) for the matched cohort without infections. The average primary curve magnitude for the infection and control groups was 80.4° ± 27.0° (range 47°–135°) and 66.3° ± 11.6° (range 42°–95°), respectively (p = 0.001). Curve patterns were classified as follows: in the delayed infection group, there were 5 single thoracic, 9 double major, and 3 triple curves; in the control group, there were 21 single thoracic, 38 double major, and 26 triple curves. No significant differences were observed between the 2 groups in terms of age, sex ratio, BMI, or distribution of curve patterns. A strong risk factor for the development of a delayed infection was comorbid medical conditions (p = 0.006), which involved a myriad of different diagnoses. Of the 17 TABLE 1: Clinical and radiological patient data Delayed Infection Characteristics
Yes
No
no. of patients mean age ± SD (yrs) sex males females BMI ± SD (kg/m2) comorbid medical conditions yes no scoliosis diagnosis (%) congenital idiopathic adult neuromuscular mean preop curve magnitude ± SD (°) curve pattern single thoracic double major triple curves
17 23.0 ± 13.8
85 18.6 ± 11.3
5 12 19.5 ± 3.3
30 55 18.7 ± 2.4
5 12
4 81
p Value 0.165 0.641
0.213 0.006
0.991 3 (17.6) 4 (23.5) 7 (41.2) 3 (17.6) 80.4 ± 27.0
17 (20.0) 21 (24.7) 34 (40.0) 13 (15.3) 66.3 ± 11.6
5 9 3
21 38 26
0.001 0.558
649
J. Shen et al. patients with infections, 5 (29.4%) had comorbid medical conditions, including diastematomyelia in 1 case, leukocytopenia in 1 case, severe acne in 1 case, dysembryoma in 1 case, and obsolete tuberculosis of the vertebral column in 1 case. Among the 85 patients without infections, 4 (4.70%) had comorbid medical conditions, including tethered cord in 1 case, syringomyelia in 1 case, hyperthyreosis in 1 case, and mitral valve regurgitation in 1 case. No other comorbidities, such as the presence of any skin lesions, were observed in any of the patients preoperatively. Cultures of specimens obtained from deep within the wound were performed in all patients. Staphylococcus aureus was identified in 3 patients and Staphylococcus coagulase-negative in 2 patients. Twelve patients had negative cultures despite prolonged incubation. All of these 17 patients were taken to the operating room for removal of their instrumentation. The patients were treated with parenteral antibiotics for 1–2 weeks after instrumentation removal, and they were subsequently treated with oral antibiotics for 1–3 weeks. Operative Data
The average operation time was longer in patients with delayed infections (384.7 ± 115.9 minutes, range 180–540 minutes) than in those without infections (254.4 ± 79.2 minutes, range 185–460 minutes; p = 0.000; Table 2). Average EBL was 1342.2 ± 707.2 ml (range 300–2800 ml) and 833.9 ± 235.6 ml (range 400–1800 ml) in the delayed infection and control groups, respectively (p = 0.000). Of the 17 patients with infections, 12 (70.6%) received allogenic blood perioperatively, either as red blood cells or whole blood. On average, 2.8 ± 2.3 units/patient were given perioperatively. However, in the control group
without infections, 27 patients (31.8%) received an allogenic blood transfusion perioperatively, either as red blood cells or whole blood. On average, 1.1 ± 1.6 units/ patient were given perioperatively. The use of an allograft was not statistically associated with an increased risk of infection; it was used in 10 of 17 patients in the group with infection, as compared with 59 of 85 patients in the uninfected matched cohort (p = 0.394, chi-square test). Postoperative factors such as drainage method (subcutaneous or conventional drainage), mean hemoglobin level, and mean hematocrit level were not significantly different between the 2 groups (Table 2). Predictive Factors of Postoperative Complications
In the delayed infection group, according to multivariate logistic regression analysis, operative time and allogenic blood transfusion were identified as 2 factors that remained statistically significant (p < 0.05) in predicting the probability of a delayed infection (R2 = 0.558; Table 3).
Discussion
Delayed infection after spine surgery is believed to be associated with a positive medical history, more extensive implants, and allogenic blood transfusion.10,23 Clinical experiences suggest that surgical time, longer fusion segments of the spine, not using postoperative drains, and increased drainage when drains are used may also correlate with delayed infection. In this retrospective, nonrandomized study we analyzed the predictors of delayed infections after spinal fusion and instrumentation in patients with scoliosis. The overall incidence rate of delayed infections was 0.49% in this study, suggesting that de-
TABLE 2: Predictive factors for postoperative complications in the 2 study groups Delayed Infection
650
Possible Predictive Factor
Yes
No
p Value
mean forced vital capacity ± SD (%) spinal fusion procedure (%) anterior posterior 1-stage anterior & posterior 2-stage anterior & posterior mean no. of levels fused ± SD mean op time ± SD (mins) mean EBL ± SD (ml) mean transfusion ± SD (units/patient) allograft (%) yes no postop drainage (%) subcutaneous conventional mean hemoglobin at discharge ± SD (g/dl) mean hematocrit at discharge ± SD (%)
80.5 ± 12.6
76.5 ± 14.6
0.289 0.280
0 (0) 12 (70.6) 5 (29.4) 0 (0) 11.6 ± 3.2 384.7 ± 115.9 1342.2 ± 707.2 2.8 ± 2.3
3 (3.5) 64 (75.3) 12 (14.1) 6 (7.1) 10.7 ± 1.6 254.4 ± 79.2 833.9 ± 235.6 1.1 ± 1.6
10 (58.8) 7 (41.2)
59 (69.4) 26 (30.6)
0.109 0.000 0.000 0.000 0.394
0.412 5 (29.4) 12 (70.6) 104.9 ± 11.9 30.6 ± 4.7
34 (40.0) 51 (60.0) 102.4 ± 8.9 29.9 ± 3.3
0.334 0.433
J Neurosurg: Spine / Volume 21 / October 2014
Risk factors for delayed infections in scoliosis patients TABLE 3: Multivariate regression model of predicting delayed infection* Predictors
OR (95% CI)
p Value
op time transfusion (units/patient)
1.021 (1.010–1.033) 1.546 (1.048–2.278)
0.000 0.028
* Nagelkerke R 2 = 0.558.
layed infection is a rare problem in clinical practice. Our results also indicate that in contrast to age, sex ratio, BMI, forced vital capacity, and other clinical characteristics, surgical time and allogenic blood transfusion are the 2 independent predictors of delayed infections. The definition of a delayed infection after spinal fusion is controversial and has been described as greater than 1 month,13 2 months,17,26 3 months,1,7 6 months,9 and 1 year3 after the initial procedure. Some published studies on delayed infections in spinal instrumentation have described “delayed infections” as occurring as early as 10 days after the index procedure, although most authors usually describe delayed infections after 9 months from posterior spinal fusion and instrumentation.3,8,24,28 For the purposes of this study, we defined late infections as those that occurred more than 6 months after the index procedure, after an asymptomatic period. It is well established in spine surgery that prolonged operative time predisposes patients to postoperative spinal wound infections.21,29 The results of the present study demonstrate that patients with longer surgical time were more likely to suffer from delayed infections in the postoperative period (OR = 1.021, 95% CI 1.010–1.033). A possible explanation may be that contamination of the sterile field occurs during the extended preoperative setup time. Possible contamination sources include direct contact with the sterile field, airborne contamination from traffic,19 or loss of sterile technique. Dalstrom et al. conducted a study in a simulated operating room environment and found a significant correlation between the length of time that the sterile tray is exposed to the air and the contamination rate of the instruments. Covering the tray with a sterile towel after it is opened reduces this contamination risk.4 The other strength of the present study is that we included a large cohort of patients with scoliosis surgeries. Opinions concerning the correlation between allogenic blood transfusion and infections are not unanimous. Ho et al. reviewed 36 patients with delayed infections in adolescent idiopathic scoliosis who underwent posterior spinal fusion and found that the incidence of delayed infections was directly related to the blood transfusion.10 Christodoulou et al. found that high allogenic blood transfusion rates were significantly associated with increased acute infection in adult spinal surgery.2 However, according to the findings of Master et al. and Linam et al., transfusion requirement17 and volume of blood products transfused15 were not found to be significant risk factors. Consistent with the studies of Ho et al.10 and Christodoulou et al.,2 our study demonstrated that allogenic blood transfusion acted as an independent predictor J Neurosurg: Spine / Volume 21 / October 2014
of delayed infections. We hypothesize that immune status can be disturbed or even weakened by allogenic transfusion. Both the number and activity of polymorphonuclear leukocytes decrease, and furthermore, cellular immunity is suppressed after administration of allogenic blood.5 Conventional closed-suction drainage is widely used in spine surgery. In theory, it prevents the formation of hematomas in the operative field, decreases the tension of closed incisions, contributes to avoiding delayed wound healing, and reduces the risk of infection.11,12 Ho et al. demonstrated that failure to use a drain after surgery was a significant independent risk factor for delayed infections. Patients who did not receive a drain were 3 times more likely to develop an infection.10 However, this modality can decrease wound tamponade, leading to an increase in blood loss and thus potentially increasing the need for transfusion.22,27 Some surgeons used subcutaneous closed-suction drainage as an alternative to avoid increased blood loss, although doing so diminishes the possibility of persistent hematomas or drainage through the incision, both of which create favorable bacterial culture media theoretically. Liang et al. reported that subcutaneous closed-suction drainage offers a reasonable alternative to closed-wound suction drainage in adolescent patients with idiopathic scoliosis undergoing posterior instrumented spinal fusion, without increasing the postoperative wound infection rate.14 Consistent with the results in the study of Liang et al., our study demonstrated no statistically significant differences for delayed infections between the subcutaneous drainage group and conventional drainage group, suggesting subcutaneous drainage could be applied as a safe and effective method in scoliosis surgery to decrease the perioperative blood loss. There are some important limitations to this study. These data represent the experience at a single institution that is an academic tertiary care center with trainees in the anesthesia, orthopedic, neurosurgical, and nursing departments. Another limitation of the study was that our data collection system only captured wound infection information on patients who returned to our hospital for treatment. Patients with localized superficial infections who were managed successfully as outpatients would not be included in this study. However, we consider these infections to be clinically significant because these patients required further hospitalization and/or additional surgeries to treat their infections. Moreover, the results are limited by a relatively small sample size and the broad time period covered. In addition, changes in anesthetic and surgical practices over time may have affected patient outcomes. Finally, we also acknowledge the limitations introduced by our patients’ clinical heterogeneity.
Conclusions
The risks for delayed infections after spinal fusion and instrumentation in patients with scoliosis are multifactorial. We believe that seeding occurs at the time of surgery and is related to the surgeon as well as the surgical technique. We advocate the use of a cell salvage system coupled with other blood-saving techniques, such as antifibrinolytic drugs, fibrin sealant, controlled hypo651
J. Shen et al. tension, autologous platelets, and others, to reduce the chance of perioperative allogenic blood transfusion. This information should be taken into account when surgeons are discussing the risks of postoperative wound complications for scoliosis surgery with the patient and the family. Disclosure This study was supported by grants from the National Natural Science Foundation of China (No. 81301596). Author contributions to the study and manuscript preparation include the following. Conception and design: Shen. Acquisition of data: Liang, Yu, Qiu. Analysis and interpretation of data: Liang, Yu. Drafting the article: Liang. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Shen. Statistical analysis: Xue, Li. Administrative/technical/material support: Xue. Study supervision: Shen. References 1. Cahill PJ, Warnick DE, Lee MJ, Gaughan J, Vogel LE, Hammerberg KW, et al: Infection after spinal fusion for pediatric spinal deformity: thirty years of experience at a single institution. Spine (Phila Pa 1976) 35:1211–1217, 2010 2. Christodoulou AG, Givissis P, Symeonidis PD, Karataglis D, Pournaras J: Reduction of postoperative spinal infections based on an etiologic protocol. Clin Orthop Relat Res 444:107–113, 2006 3. Clark CE, Shufflebarger HL: Late-developing infection in instrumented idiopathic scoliosis. Spine (Phila Pa 1976) 24: 1909–1912, 1999 4. Dalstrom DJ, Venkatarayappa I, Manternach AL, Palcic MS, Heyse BA, Prayson MJ: Time-dependent contamination of opened sterile operating-room trays. J Bone Joint Surg Am 90:1022–1025, 2008 5. Fischer E, Lenhard V, Seifert P, Kluge A, Johannsen R: Blood transfusion-induced suppression of cellular immunity in man. Hum Immunol 1:187–194, 1980 6. Gersoff WK, Renshaw TS: The treatment of scoliosis in cerebral palsy by posterior spinal fusion with Luque-rod segmental instrumentation. J Bone Joint Surg Am 70:41–44, 1988 7. Hedequist D, Haugen A, Hresko T, Emans J: Failure of attempted implant retention in spinal deformity delayed surgical site infections. Spine (Phila Pa 1976) 34:60–64, 2009 8. Heggeness MH, Esses SI, Errico T, Yuan HA: Late infection of spinal instrumentation by hematogenous seeding. Spine (Phila Pa 1976) 18:492–496, 1993 9. Ho C, Skaggs DL, Weiss JM, Tolo VT: Management of infection after instrumented posterior spine fusion in pediatric scoliosis. Spine (Phila Pa 1976) 32:2739–2744, 2007 10. Ho C, Sucato DJ, Richards BS: Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976) 32:2272–2277, 2007 11. Holt BT, Parks NL, Engh GA, Lawrence JM: Comparison of closed-suction drainage and no drainage after primary total knee arthroplasty. Orthopedics 20:1121–1125, 1997 12. Kim YH, Cho SH, Kim RS: Drainage versus nondrainage in simultaneous bilateral total knee arthroplasties. Clin Orthop Relat Res (347):188–193, 1998 13. Labbé AC, Demers AM, Rodrigues R, Arlet V, Tanguay K, Moore DL: Surgical-site infection following spinal fusion: a case-control study in a children’s hospital. Infect Control Hosp Epidemiol 24:591–595, 2003 14. Liang J, Qiu G, Chua S, Shen J: Comparison between sub-
652
cutaneous closed-suction drainage and conventional closedsuction drainage in adolescent idiopathic scoliosis patients undergoing posterior instrumented spinal fusion: a randomized control trial. J Spinal Disord Tech 26:256–259, 2013 15. Linam WM, Margolis PA, Staat MA, Britto MT, Hornung R, Cassedy A, et al: Risk factors associated with surgical site infection after pediatric posterior spinal fusion procedure. Infect Control Hosp Epidemiol 30:109–116, 2009 16. Lonstein JE, Akbarnia A: Operative treatment of spinal deformities in patients with cerebral palsy or mental retardation. An analysis of one hundred and seven cases. J Bone Joint Surg Am 65:43–55, 1983 17. Master DL, Poe-Kochert C, Son-Hing J, Armstrong DG, Thompson GH: Wound infections after surgery for neuromuscular scoliosis: risk factors and treatment outcomes. Spine (Phila Pa 1976) 36:E179–E185, 2011 18. Ovadia D, Luger E, Bickels J, Menachem A, Dekel S: Efficacy of closed wound drainage after total joint arthroplasty. A prospective randomized study. J Arthroplasty 12:317–321, 1997 19. Panahi P, Stroh M, Casper DS, Parvizi J, Austin MS: Operating room traffic is a major concern during total joint arthroplasty. Clin Orthop Relat Res 470:2690–2694, 2012 20. Parker MJ, Roberts CP, Hay D: Closed suction drainage for hip and knee arthroplasty. A meta-analysis. J Bone Joint Surg Am 86-A:1146–1152, 2004 21. Pull ter Gunne AF, Cohen DB: Incidence, prevalence, and analysis of risk factors for surgical site infection following adult spinal surgery. Spine (Phila Pa 1976) 34:1422–1428, 2009 22. Raleigh E, Hing CB, Hanusiewicz AS, Fletcher SA, Price R: Drain clamping in knee arthroplasty, a randomized controlled trial. ANZ J Surg 77:333–335, 2007 23. Richards BR, Emara KM: Delayed infections after posterior TSRH spinal instrumentation for idiopathic scoliosis: revisited. Spine (Phila Pa 1976) 26:1990–1996, 2001 24. Richards BS: Delayed infections following posterior spinal instrumentation for the treatment of idiopathic scoliosis. J Bone Joint Surg Am 77:524–529, 1995 25. Sponseller PD, LaPorte DM, Hungerford MW, Eck K, Bridwell KH, Lenke LG: Deep wound infections after neuromuscular scoliosis surgery: a multicenter study of risk factors and treatment outcomes. Spine (Phila Pa 1976) 25:2461–2466, 2000 26. Szöke G, Lipton G, Miller F, Dabney K: Wound infection after spinal fusion in children with cerebral palsy. J Pediatr Orthop 18:727–733, 1998 27. Tjeenk RM, Peeters MP, van den Ende E, Kastelein GW, Bres lau PJ: Wound drainage versus non-drainage for proximal femoral fractures. A prospective randomised study. Injury 36:100–104, 2005 28. Viola RW, King HA, Adler SM, Wilson CB: Delayed infection after elective spinal instrumentation and fusion. A retrospective analysis of eight cases. Spine 22:2444–2451, 1997 29. Wimmer C, Gluch H, Franzreb M, Ogon M: Predisposing factors for infection in spine surgery: a survey of 850 spinal procedures. J Spinal Disord 11:124–128, 1998
Manuscript submitted July 6, 2013. Accepted June 2, 2014. Please include this information when citing this paper: published online July 4, 2014; DOI: 10.3171/2014.6.SPINE13636. Address correspondence to: Jianxiong Shen, M.D., Department of Orthopaedic Surgery, Peking Union Medical College Hospital, No. 1 Shuai Fu Yuan, Wang Fu Jing St., Beijing 100730, China. email: [email protected].
J Neurosurg: Spine / Volume 21 / October 2014
Risk factors for delayed infections after spinal fusion and instrumentation in patients with scoliosis Clinical article Jianxiong Shen, M.D.,1 Jinqian Liang, M.D.,1 Haiquan Yu, M.D., 2 Guixing Qiu, M.D.,1 Xuhong Xue, M.D.,1 and Zheng Li, M.D.1 Department of Orthopaedic Surgery, Peking Union Medical College Hospital, Beijing; and 2Department of Orthopaedic Surgery, Shi Jiazhuang Center Hospital, Shi Jiazhuang, Hebei, China
1
Object. There are limited published data about the risk factors for the development of delayed infections after spinal fusion and instrumentation in the population with scoliosis. The objective of this study was to evaluate the predictive factors of development of delayed infections in patients with scoliosis who underwent surgical treatment. Methods. A total of 17 patients with scoliosis and delayed infections were identified from 3463 patients with scoliosis who received surgical treatment. The control group was composed of 85 patients with scoliosis without infections, matched for sex, age, approximate date of surgery, and diagnosis. These 2 groups were compared for demographic distribution and clinical data to investigate the predictive factors of delayed infections. Results. The overall incidence rate of delayed infections was 0.49%. The variables of age, body mass index, and number of levels fused were similar between the 2 groups. The average primary curve magnitude for the delayed infection and control (uninfected) groups was 80.4° ± 27.0° (range 47°–135°) and 66.3° ± 11.6° (range 42°–95°), respectively (p = 0.001). Operation time in the group with delayed infections was 384.7 ± 115.9 minutes versus 254.4 ± 79.2 minutes in the control group (p = 0.000), and estimated blood loss was 1342.2 ± 707.2 ml versus 833.9 ± 235.6 ml (p = 0.000) in these 2 groups, respectively. The perioperative mean red blood cell transfusion requirement in the delayed infection group was significantly higher than that found in patients without infections (2.8 ± 2.3 units/patient versus 1.1 ± 1.6 units/patient, respectively; p = 0.000). Logistic regression analysis showed that operation time and allogenic blood transfusion were the 2 independent predictors of delayed infections (odds ratio [OR] 1.021, 95% confidence interval [CI] 1.010–1.033, and OR 1.546, 95% CI 1.048–2.278, respectively). Conclusions. The occurrence of a delayed infection in patients with scoliosis who undergo surgical treatment is most likely multifactorial and is related to surgical time and the use of allogenic blood transfusion. (http://thejns.org/doi/abs/10.3171/2014.6.SPINE13636)
T
Key Words • delayed infection • spinal fusion • scoliosis • risk factor
identification and quantification of risk factors for delayed infections after spine surgery are of paramount importance to the patient and the clinician. In addition to its obvious importance for patient safety, risk factor information becomes critical as health care policy makers implement and enforce “quality” metrics. Numerous authors have reported potential risk factors for postoperative infections after pediatric spinal deformity surgery. Certain patient-related risk factors, such as underlying medical conditions and previous surgeries, are not modifiable. Recent literature has focused on identifying modifiable risk factors and prevention strategies that may decrease the rates of wound infections after spinal fusion. Perioperative factors such as cognitive impairment, use of an allograft in patients with neuromuscular scoliosis, use of preoperative skeletal traction, age greater he
Abbreviations used in this paper: BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; OR = odds ratio.
648
than 20 years, and higher intraoperative blood loss have been reported as risk factors for infection.6,16,25 However, there are limited published data about the risk factors for the development of delayed infections after spinal fusion and instrumentation in the population with scoliosis. This retrospective cohort study was undertaken to investigate 1) the incidence of delayed infections in a larger population with scoliosis treated using spinal fusion and instrumentation, and 2) the predictive factors of the development of delayed infections after spine surgery in unselected patients with scoliosis who underwent spinal deformity surgery.
Methods Study Population
We examined data from a consecutive series of 3463 patients who had undergone scoliosis surgery between January 1, 1997, and January 1, 2011, at 1 academic hospiJ Neurosurg: Spine / Volume 21 / October 2014
Risk factors for delayed infections in scoliosis patients tal (Peking Union Medical College Hospital), a universitybased medical center. Those patients who had a delayed infection, defined as an infection of the incision occurring more than 6 months from the primary surgery, were identified. Only those patients who underwent a primary spinal deformity surgery and had a minimum of 2 years of follow-up were reviewed. All patients with delayed infections were identified, and these patients with scoliosis without infection were chosen and best matched for sex, age, approximate date of surgery, and diagnosis with the patients with infections, and then matched in a 5:1 ratio. Operative Procedures
In all patients, a Cell Saver system (Haemonetics Cell Saver 5, Haemonetics Corp.) was used intraoperatively to re-infuse shed blood. The volume of blood collected by the Cell Saver system was recorded, and estimated blood loss (EBL) was calculated by measuring loss through Cell Saver suction and by weighing the surgical sponges. The surgery was performed on the patient under hypotensive anesthesia in which systolic blood pressure was maintained at less than 90 mm Hg. The same blood transfusion and prophylactic antibiotic guidelines were used for all patients. Allogenic blood transfusion was performed if hemoglobin decreased to less than 7.0 mg/dl or if anemic symptoms developed, such as a decrease in blood pressure to less than 100 mm Hg systolic, tachycardia greater than 100 beats/min, or a low urine output of less than 30 ml/hr, even after initial fluid challenge with 500 ml normal saline in patients with a hemoglobin level between 7.0 and 8.0 mg/dl.18,20 All patients were given a dose of 1 g of first-generation cephalosporin 30 minutes before skin incision and were covered for 24 hours with 3 additional doses.
Predictive Variables
The treatment of all infections was complete by the time of the study. Variables studied for their significance as preoperative predictors of delayed infections included age at surgery, sex, height, weight, body mass index (BMI), previous medical history, and curve magnitude. Intraoperative factors studied in relation to infection included surgical time, use of allograft or autograft, amount and type of blood transfusion, intraoperative blood loss, number of levels fused, use of drains, and postoperative hemoglobin/hematocrit levels.
Statistical Analysis
Factors associated with the occurrence of delayed infections were identified using univariate analysis. The data analysis was performed using SPSS version 19.0 (IBM Corp.). Continuous data were compared between the 2 groups using the Student t-test, whereas discontinuous data were analyzed using the chi-square test. The Fisher’s exact test was used for small data subsets (n < 5). All significance tests were 2-tailed, with p < 0.05 representing statistical significance. In addition, a multivariate logistic regression analysis was performed to identify which factors helped predict the probability of a delayed infection.
J Neurosurg: Spine / Volume 21 / October 2014
Results Demographic Data
A summary of the clinical and radiographic data before spinal fusion for the delayed infection and control (uninfected) groups is presented in Table 1. Of the 17 patients identified as having wound infections, 3 had congenital scoliosis, 4 had idiopathic scoliosis, 7 had adult scoliosis, and 3 had neuromuscular scoliosis. Five of the 17 patients used stainless-steel instrumentation and the other 12 used titanium instrumentation. An additional 85 patients without infections matched for age, diagnosis, and year of surgery were also evaluated, resulting in 17 patients with congenital scoliosis, 21 with idiopathic scoliosis, 34 with adult scoliosis, and 13 with neuromuscular scoliosis. The mean age at surgery was 23.0 years (range 6–57 years) for the patients in whom infection developed and 18.6 years (range 8–55 years) for the matched cohort without infections. The average primary curve magnitude for the infection and control groups was 80.4° ± 27.0° (range 47°–135°) and 66.3° ± 11.6° (range 42°–95°), respectively (p = 0.001). Curve patterns were classified as follows: in the delayed infection group, there were 5 single thoracic, 9 double major, and 3 triple curves; in the control group, there were 21 single thoracic, 38 double major, and 26 triple curves. No significant differences were observed between the 2 groups in terms of age, sex ratio, BMI, or distribution of curve patterns. A strong risk factor for the development of a delayed infection was comorbid medical conditions (p = 0.006), which involved a myriad of different diagnoses. Of the 17 TABLE 1: Clinical and radiological patient data Delayed Infection Characteristics
Yes
No
no. of patients mean age ± SD (yrs) sex males females BMI ± SD (kg/m2) comorbid medical conditions yes no scoliosis diagnosis (%) congenital idiopathic adult neuromuscular mean preop curve magnitude ± SD (°) curve pattern single thoracic double major triple curves
17 23.0 ± 13.8
85 18.6 ± 11.3
5 12 19.5 ± 3.3
30 55 18.7 ± 2.4
5 12
4 81
p Value 0.165 0.641
0.213 0.006
0.991 3 (17.6) 4 (23.5) 7 (41.2) 3 (17.6) 80.4 ± 27.0
17 (20.0) 21 (24.7) 34 (40.0) 13 (15.3) 66.3 ± 11.6
5 9 3
21 38 26
0.001 0.558
649
J. Shen et al. patients with infections, 5 (29.4%) had comorbid medical conditions, including diastematomyelia in 1 case, leukocytopenia in 1 case, severe acne in 1 case, dysembryoma in 1 case, and obsolete tuberculosis of the vertebral column in 1 case. Among the 85 patients without infections, 4 (4.70%) had comorbid medical conditions, including tethered cord in 1 case, syringomyelia in 1 case, hyperthyreosis in 1 case, and mitral valve regurgitation in 1 case. No other comorbidities, such as the presence of any skin lesions, were observed in any of the patients preoperatively. Cultures of specimens obtained from deep within the wound were performed in all patients. Staphylococcus aureus was identified in 3 patients and Staphylococcus coagulase-negative in 2 patients. Twelve patients had negative cultures despite prolonged incubation. All of these 17 patients were taken to the operating room for removal of their instrumentation. The patients were treated with parenteral antibiotics for 1–2 weeks after instrumentation removal, and they were subsequently treated with oral antibiotics for 1–3 weeks. Operative Data
The average operation time was longer in patients with delayed infections (384.7 ± 115.9 minutes, range 180–540 minutes) than in those without infections (254.4 ± 79.2 minutes, range 185–460 minutes; p = 0.000; Table 2). Average EBL was 1342.2 ± 707.2 ml (range 300–2800 ml) and 833.9 ± 235.6 ml (range 400–1800 ml) in the delayed infection and control groups, respectively (p = 0.000). Of the 17 patients with infections, 12 (70.6%) received allogenic blood perioperatively, either as red blood cells or whole blood. On average, 2.8 ± 2.3 units/patient were given perioperatively. However, in the control group
without infections, 27 patients (31.8%) received an allogenic blood transfusion perioperatively, either as red blood cells or whole blood. On average, 1.1 ± 1.6 units/ patient were given perioperatively. The use of an allograft was not statistically associated with an increased risk of infection; it was used in 10 of 17 patients in the group with infection, as compared with 59 of 85 patients in the uninfected matched cohort (p = 0.394, chi-square test). Postoperative factors such as drainage method (subcutaneous or conventional drainage), mean hemoglobin level, and mean hematocrit level were not significantly different between the 2 groups (Table 2). Predictive Factors of Postoperative Complications
In the delayed infection group, according to multivariate logistic regression analysis, operative time and allogenic blood transfusion were identified as 2 factors that remained statistically significant (p < 0.05) in predicting the probability of a delayed infection (R2 = 0.558; Table 3).
Discussion
Delayed infection after spine surgery is believed to be associated with a positive medical history, more extensive implants, and allogenic blood transfusion.10,23 Clinical experiences suggest that surgical time, longer fusion segments of the spine, not using postoperative drains, and increased drainage when drains are used may also correlate with delayed infection. In this retrospective, nonrandomized study we analyzed the predictors of delayed infections after spinal fusion and instrumentation in patients with scoliosis. The overall incidence rate of delayed infections was 0.49% in this study, suggesting that de-
TABLE 2: Predictive factors for postoperative complications in the 2 study groups Delayed Infection
650
Possible Predictive Factor
Yes
No
p Value
mean forced vital capacity ± SD (%) spinal fusion procedure (%) anterior posterior 1-stage anterior & posterior 2-stage anterior & posterior mean no. of levels fused ± SD mean op time ± SD (mins) mean EBL ± SD (ml) mean transfusion ± SD (units/patient) allograft (%) yes no postop drainage (%) subcutaneous conventional mean hemoglobin at discharge ± SD (g/dl) mean hematocrit at discharge ± SD (%)
80.5 ± 12.6
76.5 ± 14.6
0.289 0.280
0 (0) 12 (70.6) 5 (29.4) 0 (0) 11.6 ± 3.2 384.7 ± 115.9 1342.2 ± 707.2 2.8 ± 2.3
3 (3.5) 64 (75.3) 12 (14.1) 6 (7.1) 10.7 ± 1.6 254.4 ± 79.2 833.9 ± 235.6 1.1 ± 1.6
10 (58.8) 7 (41.2)
59 (69.4) 26 (30.6)
0.109 0.000 0.000 0.000 0.394
0.412 5 (29.4) 12 (70.6) 104.9 ± 11.9 30.6 ± 4.7
34 (40.0) 51 (60.0) 102.4 ± 8.9 29.9 ± 3.3
0.334 0.433
J Neurosurg: Spine / Volume 21 / October 2014
Risk factors for delayed infections in scoliosis patients TABLE 3: Multivariate regression model of predicting delayed infection* Predictors
OR (95% CI)
p Value
op time transfusion (units/patient)
1.021 (1.010–1.033) 1.546 (1.048–2.278)
0.000 0.028
* Nagelkerke R 2 = 0.558.
layed infection is a rare problem in clinical practice. Our results also indicate that in contrast to age, sex ratio, BMI, forced vital capacity, and other clinical characteristics, surgical time and allogenic blood transfusion are the 2 independent predictors of delayed infections. The definition of a delayed infection after spinal fusion is controversial and has been described as greater than 1 month,13 2 months,17,26 3 months,1,7 6 months,9 and 1 year3 after the initial procedure. Some published studies on delayed infections in spinal instrumentation have described “delayed infections” as occurring as early as 10 days after the index procedure, although most authors usually describe delayed infections after 9 months from posterior spinal fusion and instrumentation.3,8,24,28 For the purposes of this study, we defined late infections as those that occurred more than 6 months after the index procedure, after an asymptomatic period. It is well established in spine surgery that prolonged operative time predisposes patients to postoperative spinal wound infections.21,29 The results of the present study demonstrate that patients with longer surgical time were more likely to suffer from delayed infections in the postoperative period (OR = 1.021, 95% CI 1.010–1.033). A possible explanation may be that contamination of the sterile field occurs during the extended preoperative setup time. Possible contamination sources include direct contact with the sterile field, airborne contamination from traffic,19 or loss of sterile technique. Dalstrom et al. conducted a study in a simulated operating room environment and found a significant correlation between the length of time that the sterile tray is exposed to the air and the contamination rate of the instruments. Covering the tray with a sterile towel after it is opened reduces this contamination risk.4 The other strength of the present study is that we included a large cohort of patients with scoliosis surgeries. Opinions concerning the correlation between allogenic blood transfusion and infections are not unanimous. Ho et al. reviewed 36 patients with delayed infections in adolescent idiopathic scoliosis who underwent posterior spinal fusion and found that the incidence of delayed infections was directly related to the blood transfusion.10 Christodoulou et al. found that high allogenic blood transfusion rates were significantly associated with increased acute infection in adult spinal surgery.2 However, according to the findings of Master et al. and Linam et al., transfusion requirement17 and volume of blood products transfused15 were not found to be significant risk factors. Consistent with the studies of Ho et al.10 and Christodoulou et al.,2 our study demonstrated that allogenic blood transfusion acted as an independent predictor J Neurosurg: Spine / Volume 21 / October 2014
of delayed infections. We hypothesize that immune status can be disturbed or even weakened by allogenic transfusion. Both the number and activity of polymorphonuclear leukocytes decrease, and furthermore, cellular immunity is suppressed after administration of allogenic blood.5 Conventional closed-suction drainage is widely used in spine surgery. In theory, it prevents the formation of hematomas in the operative field, decreases the tension of closed incisions, contributes to avoiding delayed wound healing, and reduces the risk of infection.11,12 Ho et al. demonstrated that failure to use a drain after surgery was a significant independent risk factor for delayed infections. Patients who did not receive a drain were 3 times more likely to develop an infection.10 However, this modality can decrease wound tamponade, leading to an increase in blood loss and thus potentially increasing the need for transfusion.22,27 Some surgeons used subcutaneous closed-suction drainage as an alternative to avoid increased blood loss, although doing so diminishes the possibility of persistent hematomas or drainage through the incision, both of which create favorable bacterial culture media theoretically. Liang et al. reported that subcutaneous closed-suction drainage offers a reasonable alternative to closed-wound suction drainage in adolescent patients with idiopathic scoliosis undergoing posterior instrumented spinal fusion, without increasing the postoperative wound infection rate.14 Consistent with the results in the study of Liang et al., our study demonstrated no statistically significant differences for delayed infections between the subcutaneous drainage group and conventional drainage group, suggesting subcutaneous drainage could be applied as a safe and effective method in scoliosis surgery to decrease the perioperative blood loss. There are some important limitations to this study. These data represent the experience at a single institution that is an academic tertiary care center with trainees in the anesthesia, orthopedic, neurosurgical, and nursing departments. Another limitation of the study was that our data collection system only captured wound infection information on patients who returned to our hospital for treatment. Patients with localized superficial infections who were managed successfully as outpatients would not be included in this study. However, we consider these infections to be clinically significant because these patients required further hospitalization and/or additional surgeries to treat their infections. Moreover, the results are limited by a relatively small sample size and the broad time period covered. In addition, changes in anesthetic and surgical practices over time may have affected patient outcomes. Finally, we also acknowledge the limitations introduced by our patients’ clinical heterogeneity.
Conclusions
The risks for delayed infections after spinal fusion and instrumentation in patients with scoliosis are multifactorial. We believe that seeding occurs at the time of surgery and is related to the surgeon as well as the surgical technique. We advocate the use of a cell salvage system coupled with other blood-saving techniques, such as antifibrinolytic drugs, fibrin sealant, controlled hypo651
J. Shen et al. tension, autologous platelets, and others, to reduce the chance of perioperative allogenic blood transfusion. This information should be taken into account when surgeons are discussing the risks of postoperative wound complications for scoliosis surgery with the patient and the family. Disclosure This study was supported by grants from the National Natural Science Foundation of China (No. 81301596). Author contributions to the study and manuscript preparation include the following. Conception and design: Shen. Acquisition of data: Liang, Yu, Qiu. Analysis and interpretation of data: Liang, Yu. Drafting the article: Liang. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Shen. Statistical analysis: Xue, Li. Administrative/technical/material support: Xue. Study supervision: Shen. References 1. Cahill PJ, Warnick DE, Lee MJ, Gaughan J, Vogel LE, Hammerberg KW, et al: Infection after spinal fusion for pediatric spinal deformity: thirty years of experience at a single institution. Spine (Phila Pa 1976) 35:1211–1217, 2010 2. Christodoulou AG, Givissis P, Symeonidis PD, Karataglis D, Pournaras J: Reduction of postoperative spinal infections based on an etiologic protocol. Clin Orthop Relat Res 444:107–113, 2006 3. Clark CE, Shufflebarger HL: Late-developing infection in instrumented idiopathic scoliosis. Spine (Phila Pa 1976) 24: 1909–1912, 1999 4. Dalstrom DJ, Venkatarayappa I, Manternach AL, Palcic MS, Heyse BA, Prayson MJ: Time-dependent contamination of opened sterile operating-room trays. J Bone Joint Surg Am 90:1022–1025, 2008 5. Fischer E, Lenhard V, Seifert P, Kluge A, Johannsen R: Blood transfusion-induced suppression of cellular immunity in man. Hum Immunol 1:187–194, 1980 6. Gersoff WK, Renshaw TS: The treatment of scoliosis in cerebral palsy by posterior spinal fusion with Luque-rod segmental instrumentation. J Bone Joint Surg Am 70:41–44, 1988 7. Hedequist D, Haugen A, Hresko T, Emans J: Failure of attempted implant retention in spinal deformity delayed surgical site infections. Spine (Phila Pa 1976) 34:60–64, 2009 8. Heggeness MH, Esses SI, Errico T, Yuan HA: Late infection of spinal instrumentation by hematogenous seeding. Spine (Phila Pa 1976) 18:492–496, 1993 9. Ho C, Skaggs DL, Weiss JM, Tolo VT: Management of infection after instrumented posterior spine fusion in pediatric scoliosis. Spine (Phila Pa 1976) 32:2739–2744, 2007 10. Ho C, Sucato DJ, Richards BS: Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976) 32:2272–2277, 2007 11. Holt BT, Parks NL, Engh GA, Lawrence JM: Comparison of closed-suction drainage and no drainage after primary total knee arthroplasty. Orthopedics 20:1121–1125, 1997 12. Kim YH, Cho SH, Kim RS: Drainage versus nondrainage in simultaneous bilateral total knee arthroplasties. Clin Orthop Relat Res (347):188–193, 1998 13. Labbé AC, Demers AM, Rodrigues R, Arlet V, Tanguay K, Moore DL: Surgical-site infection following spinal fusion: a case-control study in a children’s hospital. Infect Control Hosp Epidemiol 24:591–595, 2003 14. Liang J, Qiu G, Chua S, Shen J: Comparison between sub-
652
cutaneous closed-suction drainage and conventional closedsuction drainage in adolescent idiopathic scoliosis patients undergoing posterior instrumented spinal fusion: a randomized control trial. J Spinal Disord Tech 26:256–259, 2013 15. Linam WM, Margolis PA, Staat MA, Britto MT, Hornung R, Cassedy A, et al: Risk factors associated with surgical site infection after pediatric posterior spinal fusion procedure. Infect Control Hosp Epidemiol 30:109–116, 2009 16. Lonstein JE, Akbarnia A: Operative treatment of spinal deformities in patients with cerebral palsy or mental retardation. An analysis of one hundred and seven cases. J Bone Joint Surg Am 65:43–55, 1983 17. Master DL, Poe-Kochert C, Son-Hing J, Armstrong DG, Thompson GH: Wound infections after surgery for neuromuscular scoliosis: risk factors and treatment outcomes. Spine (Phila Pa 1976) 36:E179–E185, 2011 18. Ovadia D, Luger E, Bickels J, Menachem A, Dekel S: Efficacy of closed wound drainage after total joint arthroplasty. A prospective randomized study. J Arthroplasty 12:317–321, 1997 19. Panahi P, Stroh M, Casper DS, Parvizi J, Austin MS: Operating room traffic is a major concern during total joint arthroplasty. Clin Orthop Relat Res 470:2690–2694, 2012 20. Parker MJ, Roberts CP, Hay D: Closed suction drainage for hip and knee arthroplasty. A meta-analysis. J Bone Joint Surg Am 86-A:1146–1152, 2004 21. Pull ter Gunne AF, Cohen DB: Incidence, prevalence, and analysis of risk factors for surgical site infection following adult spinal surgery. Spine (Phila Pa 1976) 34:1422–1428, 2009 22. Raleigh E, Hing CB, Hanusiewicz AS, Fletcher SA, Price R: Drain clamping in knee arthroplasty, a randomized controlled trial. ANZ J Surg 77:333–335, 2007 23. Richards BR, Emara KM: Delayed infections after posterior TSRH spinal instrumentation for idiopathic scoliosis: revisited. Spine (Phila Pa 1976) 26:1990–1996, 2001 24. Richards BS: Delayed infections following posterior spinal instrumentation for the treatment of idiopathic scoliosis. J Bone Joint Surg Am 77:524–529, 1995 25. Sponseller PD, LaPorte DM, Hungerford MW, Eck K, Bridwell KH, Lenke LG: Deep wound infections after neuromuscular scoliosis surgery: a multicenter study of risk factors and treatment outcomes. Spine (Phila Pa 1976) 25:2461–2466, 2000 26. Szöke G, Lipton G, Miller F, Dabney K: Wound infection after spinal fusion in children with cerebral palsy. J Pediatr Orthop 18:727–733, 1998 27. Tjeenk RM, Peeters MP, van den Ende E, Kastelein GW, Bres lau PJ: Wound drainage versus non-drainage for proximal femoral fractures. A prospective randomised study. Injury 36:100–104, 2005 28. Viola RW, King HA, Adler SM, Wilson CB: Delayed infection after elective spinal instrumentation and fusion. A retrospective analysis of eight cases. Spine 22:2444–2451, 1997 29. Wimmer C, Gluch H, Franzreb M, Ogon M: Predisposing factors for infection in spine surgery: a survey of 850 spinal procedures. J Spinal Disord 11:124–128, 1998
Manuscript submitted July 6, 2013. Accepted June 2, 2014. Please include this information when citing this paper: published online July 4, 2014; DOI: 10.3171/2014.6.SPINE13636. Address correspondence to: Jianxiong Shen, M.D., Department of Orthopaedic Surgery, Peking Union Medical College Hospital, No. 1 Shuai Fu Yuan, Wang Fu Jing St., Beijing 100730, China. email: [email protected].
J Neurosurg: Spine / Volume 21 / October 2014
