Chronic infections of the Spine

Abstract

Chronic infections following posterior fusion are relatively uncommon. They develop in a previous asymptomatic patient at a distant time from the surgery. Chronic infections arise from direct inoculation or hematogenous seeding. To eradicate a chronic infection, the pathogens, biofilm, non-viable tissues, adherence on surfaces, and instrumentation must be removed. The appropriate antibiotherapy is used in a short (4 weeks) or long protocol (9 weeks). Some patients may need repeated surgeries (leaving the instrumentation in situ) to avoid progressive deformity or symptomatic pseudoarthrosis in cases of implant removal.

Introduction

Late infections following posterior fusion are relatively uncommon. The incidence ranges from 1.7 to 6.9 % [4, 8, 14]. They develop in a previous asymptomatic patient and their onset is distant from the time of surgery.

There is no agreement on the definition of the term “chronic infection” of the spine. It is unclear when surgical site infections may be termed chronic or delayed versus acute infection. Chronic infection is suspected when the infection starts at 3 months, 6, or even more than 12 months after the surgery (6 months or >12 months postoperatively) [3, 5, 11, 14]. Chronic infection is considered when the original surgical site has healed and the likelihood that an infected retained hematoma or seroma is not present from the index operation [10].

Etiology

Microorganisms adhere and rapidly bond to non-viable tissue and implant surfaces. Their attachment initiates a physiologic transformation into colony-forming pathogens that continue to multiply within an extracellular, protein–polysaccharide biofilm. In weeks, these colonies mature within an impenetrable glycocalyx, which isolates them from host defenses and systemic antimicrobials [2]. Dubousset et al. [6] considered fretting corrosion, the cause of late infection after CD instrumentation; in many patients, it is caused by micromotion between the components of the instrumentation producing debris and granulomatous reaction.

Pathophysiology of delayed spinal infection

Chronic bacterial infection can arise from direct inoculation or hematogenous seeding [10].

• Direct inoculation

Long-duration scoliosis surgery increases the risk of exposure to saprophytic germs. The identification of low-virulence skin microorganisms as the main cause of delayed infections [3, 14, 16, 17] tends to support this hypothesis.

However, it is questionable whether such mechanisms apply for late infections with Propinobacterium acnes, which is a member of normal human skin flora. A study by Jakab et al. [12] clearly documented the pathogenic potential and role of P. acnes in late postoperative infections. P. acnes has a high affinity for deep skin structures, which makes it difficult to eradicate with commonly used disinfectants.

Low inocula during surgery may enable P. acnes to reside intracellularly and remain in a dormant state. Due to unknown factors, the inoculation changes into a clinical infection after a period of time, even after a number of years. Mechanical irritation and metal fretting might be cofactors in this process.

• Haematogenous seeding of bacteria

Seeding of bacteria from a local process (genitourinary tract, dental caries, or gastrointestinal tract) to spinal implants (the glycocalyx allows bacteria to adhere on the instrumentation) [10].

Clinical signs

The usual presenting symptoms of late implant infections are pain, fever, fluctuating mass, fistula, and sepsis [4].

Patient evaluation and medical history are important in determining the characteristics of the infection as well as the patient’s comorbidities. These comorbidities may affect the final outcome when the demands of treatment overwhelm the patient’s capacity to heal.

Significant comorbidities include local factors such as chronic edema, venous stasis, large vessel disease, arteritis, and extensive scarring, as well as systemic factors such as malnutrition, immune deficiency, hypoxia, malignancy, diabetes, extremes of age, organ failure, substance abuse, and drug inhibition [2].

Imaging tests are used in order to determine bone characteristics, presence of implants, and the extent of the inflammatory zone.

This information lets the surgeon weigh the impact of thorough debridement against the feasibility of successful reconstruction, taking into account the following principles: (1) infected tissue must be resected to live margins; (2) previous fixation failures and structural deficiencies must be addressed; and (3) the patient must fully benefit from treatment.

The patient must understand and accept the risk–benefit ratio. If the risks outweigh the benefits, the patient is offered palliative treatment.

Diagnosis

The microbiologic diagnosis of spinal infection can be challenging. Peri-implant tissue is usually cultured for the diagnosis of implant infection; however, this approach is inaccurate. Specificity may be an issue because the isolated microorganisms may be normal flora isolated as contaminants. Sensitivity is also an issue because the number of microorganisms found in the tissue can be very small. Accordingly, it has been recommended that 5–6 peri-implant tissue specimens be submitted for culture diagnosis of prosthetic hip and knee infections. In case of spinal implants, for an accurate diagnosis of late spinal infection, more than six peri-implant tissue specimens may be required [4, 8, 9].

The major causes of spinal implant infection are [1]:

• Propionibacterium acnes.

• Stahpylococcus spp.

As a general rule, microbiologic specimens, both from preoperative wound and intraoperative swabs, were cultured aerobically and anaerobically for at least 10 days. Media were examined for growth of both high-grade pathogens (Staphylococcus aureus) and low-virulence organisms (Propionibacterium spp., Corynebacterium spp., and Coagulase-negative staphylococcus [8, 9].

Actually, implant sonication is more sensitive that peri-implant tissue cultures for microbiological diagnosis in late implant infections. PCR detection of Staphylococcus spp. is equivalent to culture for detection of these organisms in the sonicated spinal implant. [7].

Risk factors of chronic or late development infection are comorbidities that include: obesity, hypertension, poor nutrition, distal level of fusion, longer surgery, and no use of surgical wound drains. Ho et al. [11] found that the use of drains reduced the risk of delayed infections (from 14 to 4 %).

Antibiotic treatment

Before surgery, empiric preoperative broad-spectrum antibiotics are used. Postoperative antibiotic treatment is tailored to the intraoperative specimens’ microbiologic analysis as soon as antibiograms were available. All patients received a minimum of 48 h to 3 weeks of parenteral drug coverage, followed by a 4- to 20-week period of oral therapy. Those authors that use shorter antibiotic therapy (4 weeks) had the same final satisfactory outcome as those with longer (8–9 weeks) therapy [4].

Surgical treatment—reconstruction

There is no clear protocol for the treatment for patients who present with delayed surgical site infections after spinal deformity.

Eradication of chronic infection is difficult because:

• 1. the colonization of the implants by bacteria

• 2. debridement of the infection with the implants in place is impossible.

There are two treatment options:

• 1. Leave the previous instrumentation in situ

Studies have shown that coronal and sagittal deformities may worsen after implant removal even in spite of apparent solid fusion at exploration. Removal of implants may cause progression of deformity and pain if pseudoarthrosis exists.

The trauma literature supports the concept that stability is necessary in the face of infection to achieve union in infected long fracture bones.

Retention of spinal implants until union is achieved despite the infection and as mentioned above, implant retention in the face of nonunion may be needed to prevent deformity progression.

Attempted implant retention in delayed infection led to increase the number of operations, lengths of hospital stay, number of hospitalizations, and financial cost. The problem is these patients may unfortunately need retention of implants if a nonunion is present and may require rod exchange and perhaps anterior approach [4, 9] (Figs. 1, 2, 3, 4, 5).

Fig. 1

A 14-year-old female with scoliosis showing a prominent right thoracic gibus

Fig. 2

Standing AP (a) and L (b) view and left (c) and right (d) lateral bendings before surgery. Postoperative standing AP (e) and L (f) view 10 months after the surgery

Fig. 3

Intraoperative images showing posterior arthrodesis with titanium instrumentation

Fig. 4

Six months after the initial procedure, a second surgical look with surgical wound cleaning without implant removal was required (a, b). Methicilin-resistant Staphylococcus aureus (MARSA) growth in wound swelling cultures. However, at 1 year after the first procedure, a third surgery was done in order to remove spinal implants, and a 4-week intravenous combined with a 6-month oral therapy was administered until CRP and ESR decreased to normal levels

• 2. Complete removal of the instrumentation

To prevent recurrences, total implant removal followed by intravenous antibiotic therapy is the treatment of choice.

Deformity surgeons are less likely to remove implants in chronic infection in order to avoid the possibility of pain, pseudoarthrosis, and curve progression. There are isolated reports of successful reinstrumentation of the spine after eradication of the infection [13].

Some authors [4, 8, 9, 15] have not noted a progression of deformity or significant complaints requiring revision. However, some patients had a significantly destabilized spine and in others, the fusion mass was still immature and they presented with a delayed infection. These patients may unfortunately need retention of implants if a nonunion is present [3, 8, 9] (Figs. 6, 7, 8, 9).

Fig. 5

Two years after spinal instrumentation removal, the patient complained about pain and a granuloma in surgical scar. We performed an MRI study showing a fistulous tract from T3 to T4 to the epidural space. We performed a fourth surgery in order to clean and evacuate the spinal abscess. A month after this surgical procedure, CRP levels decreased from 200 to 0 mg/L

Fig. 6

A 66-year-old woman with degenerative scoliosis treated with anterior and posterior approach

Fig. 7

The patient had an E. coli infection that did not resolve completely with antibiotics. When she was 68, a total removal of instrumentation was performed, with recurrence of the deformity

Fig. 8

The patient being 70 years old, a double anterior and posterior approach was performed in order to correct spinal deformity. However, 1 year after the intervention, she suffered from pain with swelling. She had another positive culture for E. coli. She underwent two surgical cleanings with intravenous antibiotics for 2 months. Unable to control the ESR or CRP, 2 years after, we proceeded to the removal of the instrumentation

Fig. 9

Today, the patient is 72 years old, the spinal deformity is still present but she does not wish to undergo further surgery

Conclusion

• Chronic infections are refractory to nonsurgical management

• To prevent recurrence, the biofilm burden must be excised and the wound revitalized (instrumentation removal).

• The antibiotic therapy in a short or long protocol is necessary

• The factors affecting treatment outcomes include the health of the host, the fusion mass, and the disability.