23.3.1 Early versus Late Surgical Site Infection

The CDC does not clearly define early versus late SSI. In clinical practice, an early surgical site infection is defined as a wound/infection occurring within the first 90 days after the index operation. Late surgical site wounds/infections typically present well beyond 1 year following the index procedure, but by definition, these are any infections that occur after the first 90 days postoperatively.

23.3.2 Deep versus Superficial Surgical Site Infection

The criteria defined by the CDC clarify the delineation for a superficial versus a deep SSI. ¹⁰ The criteria for a superficial incisional SSI are the following:

• 
  

  Infection occurs within 30 days after any spine fusion and involves only the skin and subcutaneous tissue of the incision, and the patient has at least one of the following:

  
  
  
  
  • 
    

    Purulent drainage from the superficial incision.

    
    
  • 
    

    Organisms isolated from an aseptically obtained culture of fluid or tissue from the superficial incision.

    
    
  • 
    

    Superficial incision that is deliberately opened by a surgeon and is culture-positive or not cultured, and the patient has at least one of the following signs or symptoms: pain or tenderness, localized swelling, redness, or heat. A culture-negative finding does not meet this criterion.

    
    
  • 
    

    Diagnosis of a superficial incisional SSI by the surgeon or attending.

The criteria for a deep surgical site infection are the following:

• 
  

  Infection occurs within 90 days after any spine fusion and involves deep soft tissues of the incision (e.g., fascial and muscle layers), and the patient has at least one of the following:

  
  
  
  
  • 
    

    Purulent drainage from the deep incision.

    
    
  • 
    

    A deep incision that spontaneously dehisces or is deliberately opened by a surgeon and is culture-positive or not cultured, and patient has at least one of the following signs or symptoms: fever (>38°C), localized pain, or tenderness. A culture-negative finding does not meet this criterion.

    
    
  • 
    

    An abscess or other evidence of infection involving the deep incision that is found on direct examination, during invasive procedure, or by histopathologic examination or imaging test.

    
    
  • 
    

    Diagnosis of a deep incisional SSI by a surgeon or attending physician.

23.4.1 Background and Incidence

Reported rates of acute infection after surgery for AIS generally have been reported to be less than 2%. ^{5 , 11 , 12 , 13 , 14 , 15} This is in contrast to patients with neuromuscular disease where the reported SSI rate after spinal fusion is between 8 and 24%. ^{12 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25} The highest reported rates are in patients with myelodysplasia. Some of these data are derived from older studies and thus the surgical technique and implant type may be outdated. Although it may be difficult to compare these rates to current-day techniques, a recent multicenter study of 946 patients demonstrated the SSI rate to be 9.2 and 2.6% in neuromuscular and AIS populations, respectively. ²⁶ Although anterior surgery has become less common recently, infection rates following anterior spinal surgery are quite low, approaching 0% in some reports. ²⁷ Janik et al ²⁸ reported no deep wound infections in 20 years of experience. Grossfeld et al ²⁹ reported a low deep-wound infection rate of 1.2% in this population. The SSI rate in patients with AIS postoperatively based on a multicenter database is 1.6%. ⁶

There are significant costs related to a postoperative SSI; in a 2009 study by Hedequist et al, direct costs to treat a postoperative SSI were estimated to range between $26,977 and $961,722. ³⁰ There are also considerable indirect costs to the family that are difficult to quantify, such as missed work and school, not to mention the psychological stress associated with this complication. Therefore, there is significant interest on the hospital and national level to understand risk mitigation and infection prevention strategies.

23.4.2 Risk Factors and Prevention

Understanding risk factors may be the most important component of an effective infection prevention strategy. In general, risk factors can be categorized as preoperative or intraoperative. Prevention strategies can target both preoperative optimization and intraoperative techniques in an effort to reduce infections.

In considering the microbiology associated with acute SSIs, the most common infecting organisms are Staphylococcus aureus and S. epidermidis,with S. aureus accounting for about 60% of infections. The microbiology of SSIs in patients with neuromuscular scoliosis is different, with about half having an infection caused by polymicrobial and/or gram-negative bacteria. ^{26 , 31 , 32 , 33} Literature has shown that contamination rates of spinal implants and the surgical wound before closure are 9.5 and 23%, respectively. ^{34 , 35} However, we do not have a firm understanding about what those data mean with regard to possible infection risk and/or how to influence the risk of a postoperative SSI based on this information.

Postulated preoperative risk factors for SSI have included nonidiopathic diagnosis/American Society of Anesthesiologists (ASA) physical status, the presence of a gastrostomy (G)/gastrojejunostomy (GJ) tube, obesity, and malnutrion. ^{11 , 24 , 26 , 33 , 36 , 37 , 38 , 39 , 40} A recent study evaluated patient factors associated with the incidence of infection and found that obesity was the only significant risk factor (odds ratio [OR], 7.6; p ≤ 0.001), with the resultant model demonstrating good discrimination and calibration. ⁴¹ These findings were corroborated by another study with a series of 207 AIS patients in which the infection rate was four times greater in overweight patients. ⁴²

However, when looking at the available evidence, there are little good data to support definitive risk factors for SSI. In a recent systematic review, ⁴³ there was no grade A evidence available to support any preoperative risk factors for SSI. There were only low grades of evidence that incontinence (B), neuromuscular disease (B), malnutrition (C), obesity (C), or a positive urine culture (C) increased SSI risk. In a more recent systematic review of 167 studies, ⁴⁴ there was grade C evidence that obesity increased risk of infection; there were insufficient data to make conclusions regarding any other preoperative factors.

There are some perioperative factors that have been suggested to increase SSI risk. Risk factors that have been identified in the literature include inappropriate antibiotic dosing, ^{33 , 45 , 46 , 47} hypothermia, ³⁹ operative time, ^{34 , 38} implant prominence (unit rods), ^{11 , 16} and instrumentation to the pelvis. ^{19 , 26} A few studies have demonstrated possible benefit in the use of postoperative drains; however, other studies have shown limited benefit. ^{33 , 48 , 49} Topical vancomycin has shown possible SSI risk reduction in adults, but unfortunately there are only safety data for the use of topical vancomycin in children without any clinical efficacy data. ^{50 , 51 , 52 , 53} Other possible topical treatments that have shown benefit in non-AIS populations include the use of topical gentamycin in patients with cerebral palsy and the use of dilute povidone iodine in adult spinal patients. ^{21 , 54 , 55 , 56} In a recent study, a plastic surgery/multilayer closure reduced the risk of a postoperative SSI in nonidiopathic patients from 19 to 0%. ⁵⁷

In recent systematic reviews of intraoperative risk factors, ^{43 , 44} there is overall poor evidence to help guide infection prevention strategies. There is grade B evidence that inappropriate perioperative antibiotic prophylaxis and increased implant prominence increase SSI risk, whereas surgery duration, closure method, and the use of drains have no effect on SSI rate. In the literature, there is only grade C evidence that blood loss, blood transfusions, number of levels fused, extension to the sacrum/pelvis, use of drains, type of allograft, or prolonged operative time increase risk. Unfortunately, there is insufficient evidence to support many other measures discussed previously, which are part of many centers’ perioperative pathways.

There is significant variability with infection prevention strategies between centers, likely a direct result of a lack of evidence to effectively guide processes. ^{58 , 59} Given the lack of available evidence to guide risk reduction in the AIS, we often must translate knowledge gained from populations at higher risk for infection. This can be challenging, as it is not entirely clear how effective these strategies are in a low-risk population.

In other fields, patient outcomes can be improved by reducing variablity. ⁶⁰ Further, when data are limited, ^{43 , 44} consensus-building approaches such as the Delphi method can be used to derive “best practices.” Best practice guidelines for infection prevention have been published and serve as a starting point in reducing variability in high-risk patients (Fig. 23‑1). ⁶¹ Using multidisciplinary pathways to reduce complications and improve outcomes after spinal surgery has been demonstrated in adults. ^{62 , 63} Team-based approaches have recently been shown to improve spine care and reduce SSIs. ^{64 , 65} Whereas it may be intuitive that similar processes would be beneficial after surgery in AIS, it is not clear to what extent these guidelines/adopted measures would be beneficial, and proving any positive impact will be challenging given the low baseline infection rate in this population.

23.4.3 Clinical Presentation and Evaluation

The clinical presentation of an acute SSI after spinal fusion is often obvious. The patient often will have wound drainage, but they may also have incisional erythema, fluctuance, pain, and tenderness. They may also have wound dehiscence. Occasionally, an acute SSI can make the patient systemically ill with fever, chills, sepsis, malaise, and/or anorexia. ⁶⁶ Most present fairly early, with 67% of all SSIs presenting by 1 month post index procedure. ²⁶

In any patient who presents with a suspected infection, standard laboratory evaluation should include a complete blood count (CBC) with differential, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and blood cultures. Blood cultures may help identify the causative organism and allow for more directed antibiotic therapy. Following uncomplicated spine surgery, the ESR peaks at day 5 and may remain elevated for 3 to 6 weeks. CRP peaks earlier, on day 3, and typically normalizes in 1 to 2 weeks. Therefore, these labs must be interpreted with these timetables in mind. ⁶⁷ Despite evidence that other inflammatory markers such as procalcitonin may demonstrate improved sensitivity and specificity compared to CRP in identifying bacterial infections, these markers have not been widely adopted in the workup of postoperative SSIs. ⁶⁸

Evaluation of acute SSI with imaging is challenging and often not useful. Plain radiographs are often normal in the early stages (4–6 weeks) but may demonstrate implant loosening or failure (broken implants) after this. Computed tomography (CT) scan can better characterize bony destruction and pseudarthrosis. ⁶⁹ Bone scan is often not useful in the acute phase as the area may appear reactive from the recent fusion. Three-phase technetium-99m bone scans and gallium-67 citrate scans are sensitive in detecting de novo infection of the spine; however, their role in diagnosing SSI after spinal fusion is unclear. ⁷⁰ They are often not useful in the acute phase as the area may appear reactive from the recent fusion. Magnetic resonance imaging (MRI) with gadolinium can demonstrate marrow changes, rim enhancing and/or epidural collections, as well as bone destruction. However, the metallic implants similarly impede the ability to interpret images due to interference. ⁷⁰ Fluorine-18 fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET) scans may be useful in detecting early SSIs, as they are not affected by implants. The study has relatively low radiation and has a high sensitivity/specificity. Although this may be increasingly used in the future, this diagnostic test has a high cost and limited availability at this point. ⁷¹ CT- and/or ultrasound-guided aspiration can be used to identify an organism prior to antibiotic initiation; however, the yield is low (<40%). ⁷¹ Often when the diagnosis of an acute SSI is made, operative irrigation and debridement is done and cultures are obtained from that operation.

Although the discussed tools to work up an SSI are important, clinicians should always have a high index of suspicion when evaluating a possible infection. A proposed pathway for SSI workup is presented in Fig. 23‑2.

23.4.4 Treatment

Although treatment of chronic SSIs is often unsuccessful if implants are retained, ^{12 , 30 , 32} the outcomes of treatment of acute SSIs with implant retention are not quite as discouraging. Acute infections should be treated with irrigation and debridement (I and D) followed by a course of intravenous (IV) and oral antibiotics. There is no literature to help guide the appropriate number of debridements required, whether topical vancomycin or povidone- iodine irrigation should be used, or whether bone graft should be retained or removed. Most patients are treated with 4 to 8 weeks of IV antibiotics, followed by 6 to 12 months of suppressive therapy.

In a recent multicenter study of 84 patients treated with I and D for an acute infection (<3 months from index procedure), the chance of being successfully treated for that SSI with implant retention was 75% at mean follow-up of 34 months. ⁷² These results are similar to those of Cahill et al where 24/32 patients with acute SSI were successfully treated with implant retention. ¹² One finding of importance in the multicenter study was that the chance of successful treatment of the acute SSI with implant retention was much lower if stainless steel implants were in place compared to other metal types, and patients with stainless steel implants represented 83% of recurrent infections. ⁷²

Therefore, a reasonable algorithm for treatment of an acute infection involves early I and D of the surgical wound. Cultures of the spinal fluid should be sent. Implants can be retained and often loose bone graft is removed. A second washout may be required if significant necrotic tissue is identified at the first I and D. Implant exchange should be considered, especially if stainless steel implants are in place. Closure over drains with organism-specific IV antibiotic treatment should be initiated and continued for 4 to 6 weeks followed by oral antibiotic suppressive therapy. If the patient presents with a recurrent infection, it should be treated as a chronic infection with implant removal. A proposed pathway for the treatment of SSIs is presented in Fig. 23‑3.

23.5.1 Clinical Features

Patients present with late, deep SSIs most often over a year from the time of surgery. Cases have been reported to present as many as 10 years after operation (Fig. 23‑4). Often there is an insidious backache, which is initially unexplained until the infection becomes clinically evident. Fever is very rare with this condition.

On physical examination, the findings are centered around the incision, usually in its lower region. The incision is typically well-healed with no erythema. In some patients, a subcutaneous bulge may develop. Usually, there is little tenderness to palpation and little or no erythema. Eventually, the infection begins to drain through the incision itself and forms a persistent sinus. In most patients, drainage is expressed daily; it slows only when suppressed by antibiotics.

23.5.2 Evaluation and Initial Management

If the patient has a fluctuant subcutaneous mass, a CBC and sedimentation rate are appropriate, though likely to be normal. Plain radiographs should be obtained in order to look for any lucency and sclerosis around the anchors to indicate a chronic inflammatory process. This finding may be better seen on CT imaging. However, it is more specific than sensitive. Loss of fixation, loss of correction, or pseudarthrosis is sometimes seen as well, which can help in the surgical planning process.

Aspiration of the mass is useful to confirm infection and identify an organism. If there is an active sinus, it is always in the line of the incision itself (Fig. 23‑5). It can be probed with a small cotton applicator, usually painlessly. This confirms a surprisingly deep extent of the sinus tract. It is often possible to appreciate the implant or the bone at the base of the sinus tract. Deep material should be sent for culture. If the extent of the sinus is not clear, it is possible to instill some radiographic contrast material with the patient in the prone position and observe that it tracks to the implant. This may be confirmatory, especially to those who are not accustomed to this typical clinical scenario.

The results of the culture may require 1 week or more to be finalized, and low-virulence organisms should not be discounted as contaminants. Among the most common identified are Propionibacterium acnes, Corynebacterium, S. epidermidis, Peptostreptococcus, and many others. Up to one-quarter or one-half of cases may not grow an organism. This paradox may be explained by prior suppression or extreme fastidiousness of the organism. However, it has also raised the question of whether metal sensitivity may play a role as well. Skin testing has typically been inconclusive.

Conservative care with antibiotics alone has not been shown to stop these infections. However, they may suppress the drainage and even cause the sinus to close while the antibiotics are being taken. This may allow patients to function well while they weigh their options or complete pending commitments. There is no evidence that a period of judicious suppression will increase the risk of further osteolysis or of osteomyelitis. The suggested workup for infection is summarized in Fig. 23‑2.

23.5.3 Surgical Management

Surgery is the definitive treatment of late infections just as much as it is for early ones (Fig. 23‑3). This should be done when the patient is optimized for surgery. The implants can be reached by following the sinus tract. Typically, a significant amount of granulation tissue tracks up and down the length of both implants and around the individual anchors (Fig. 23‑6). The standard management at present is to remove the implants. Sometimes the anchors become loose and can be removed with fingers. Because of the chronic inflammation, the field cannot be adequately debrided without removing the implants completely. Antifibrinolytics are recommended as this procedure is often more bloody than index surgery. Care should be exercised in removing loose screws, as they may have expanded their track and eroded into the canal. Transient neurologic injury has sometimes occurred from removing such screws.

After the implants are removed, the granulation tissue is debrided and the fusion mass is explored. This should be correlated with the preoperative radiographs or CT. A pseudarthrosis is often noted by the increased adherence of membrane to the fusion mass, in contrast to the smooth nature of a typical mature fusion mass. Motion, if present, is typically subtle. Levels of pseudarthrosis should be recorded for future monitoring. The wound is virtually always able to be debrided to clean tissues and closed over a drain. Postoperatively, antibiotics are typically continued for 2 to 3 weeks, in contrast to the longer period often used for acute deep SSIs. A brace is not needed if the fusion mass is solid but is an appropriate option in the presence of a pseudarthrosis. If a brace is utilized, it should be worn for 3 to 6 months.

23.5.4 Aftercare

After implant removal, the patient should be counseled of risk of loss of correction. In part, this depends on the nature of the original deformity and the status of the anterior column. If kyphosis was significant in the original deformity, there is a high chance of recurrence unless an anterior procedure was performed. Deformities of all types, which were initially larger, are more likely to recur. In all types of deformities, it is common to lose 10 to 30 degrees of correction, sometimes much more. Patients tolerate this better if they know of the possibility. Interestingly, even with a loss of correction, most patients have outcome scores similar to patients without complications. ⁷⁶

It is because of the risk of loss of correction that some surgeons have adopted the option of a one-stage exchange. The original implants are removed, the tissues thoroughly debrided, and new implants inserted to span the original curves. Titanium seems to be a favored metal, which may allow a high salvage rate with a single-stage exchange. The recurrence risk, while not zero, is low. Further study of this option may provide a valuable alternative to patients who are at risk of progression.

A final option is a planned two-stage exchange. The implants may be removed and the tissues debrided; planned reinsertion may be done after 4 to 6 weeks of antibiotic treatment but before the anchor tracts are overgrown or the deformity is significantly worsened.

Because of the serious and increasing burden of late infection, prevention becomes all the more compelling. New biomaterials may lessen the risk of biofilm development. Until then, current “best practices” (Fig. 23‑1) for infection prevention may help lower the rate of late deep SSI.