Clinical presentation and diagnosis of delayed postoperative spinal implant infection

Implant-associated Infection After Pediatric Spine Deformity Surgery: Is Removal of Hardware Indicated?

BACKGROUND

Postoperative spinal implant infection following pediatric spinal surgery often presents a management dilemma. We aimed to characterize our experience in this cohort of patients, hoping to provide further insight when approaching these challenging cases.

METHODS

A retrospective, observational study was performed of all children who underwent spinal surgery from January 2015 to June 2021 in a tertiary pediatric spinal surgery referral center in Queensland, Australia. Records were reviewed to identify cases of postoperative surgical site infection, with particular focus on children with infection directly involving implants. Preoperative prophylaxis, microbiology, clinical course and outcomes were analyzed.

RESULTS

A total of 580 children underwent 933 procedures during the study period. The most common pathology requiring surgery was idiopathic scoliosis in 257 children (44.3%) followed by neuromuscular scoliosis in 192 children (33.1%). There were 35 cases of surgical site infection (6.03%), and 9 were implant-associated (1.55%). Infection rate among children with neuromuscular spinal deformity was almost 3-fold higher (11.5%) than idiopathic (3.89%). Methicillin-sensitive Staphylococcus aureus was the most commonly isolated organism (n = 15). Four implant-associated infections were successfully managed with retention of implant; all were diagnosed within 32 days (median = 20 days) and cultured either Staphylococcus aureus or Pseudomonas aeruginosa . Conversely, infections that necessitated implant removal had a more indolent onset (median = 175 days), and more often cultured Cutibacterium acnes and coagulase-negative Staphylococci .

CONCLUSIONS

Postoperative spinal implant infection can be treated successfully with hardware retention in select cases. Earlier presentation and recognition appear to be associated with better rates of retention and are linked to certain organisms. Further exploration of specific preventative strategies may be key in preventing devastating late-onset infections.

Changes of the microbiological spectrum and antibiotic resistance pattern in postoperative spinal implant infections with multiple culture-positive revision surgeries

BACKGROUND CONTEXT

In severe cases of postoperative spinal implant infections (PSII) multiple revision surgeries may be needed. Little is known if changes in the microbiological spectrum and antibiotic resistance pattern occur between revision surgeries.

PURPOSE

Analysis of the microbiological spectrum and antibiotic resistance pattern in patients with multiple revision surgeries for the treatment of PSII STUDY DESIGN: Retrospective database analysis.

PATIENT SAMPLE

Between 01/2011 and 12/2018, 103 patients underwent 248 revision surgeries for the treatment of PSII. Twenty patients (19.4%) who underwent multiple revision surgeries for PSII were included in this study.

OUTCOME MEASURES

Microbiological spectrum, antibiotic resistance pattern.

METHODS

A retrospective analysis of a prospectively maintained single center spine infection database was performed with a minimum follow-up of 3 years. Overall, 20 patients (six male/14 female) underwent 82 revisions for PSII (median 3; range 2-12). There were 55 of 82 (67.1%) procedures with a positive microbiological result. Microbiological analysis was performed on tissue and implant sonication fluid. Changes in microbial spectrum and antibiotic resistance pattern between surgeries were evaluated using Chi-Square and Fisher's exact test.

RESULTS

In total, 74 microorganisms (83.3% gram-positive; 10.8% gram-negative) were identified. The most common microorganisms were Staphylococcus epidermidis (18.9%) and Cutibacterium acnes (18.9%). All S. epidermidis identified were methicillin-resistant (MRSE). Overall, there were 15 of 55 (27.3%) polymicrobial infections. The microbiological spectrum changed in 57.1% (20/35) between the revision stages over the entire PSII period. In 42.9% (15/35) the microorganism persisted between the revision surgeries stages. Overall, changes of the antibiotic resistance pattern were seen in 17.4% (8/46) of the detected microorganisms, comparing index revision and all subsequent re-revisions. Moreover, higher resistance rates were found for moxifloxacin and for ciprofloxacin at first re-revision surgery compared with index PSII revision. Resistances against vancomycin increased from 4.5% (1/23) at index PSII revision to 7.7% (2/26) at first re-revision surgery.

CONCLUSIONS

Changes of the microbiological spectrum and the resistance pattern can occur in patients with severe PSII who require multiple revision surgeries. It is important to consider these findings in the antimicrobial treatment of PSII. The microbiological analysis of intraoperative tissue samples should be performed at every revision procedure for PSII.

Clinical differences between delayed and acute onset postoperative spinal infection

Spine surgeons often encounter cases of delayed postoperative spinal infection (PSI). Delayed-onset PSI is a common clinical problem. However, since many studies have investigated acute PSIs, reports of delayed PSI are rare. The purpose of this study was to compare the clinical features, treatment course, and prognosis of delayed PSI with acute PSI.Ninety-six patients diagnosed with postoperative spinal infection were enrolled in this study. Patients were classified into 2 groups: acute onset (AO) within 90 days (n = 73) and delayed onset (DO) after 90 days (n = 23). The baseline data, clinical manifestations, specific treatments, and treatment outcomes were compared between the 2 groups.The history of diabetes mellitus (DM) and metallic instrumentation at index surgery were more DO than the AO group. The causative organisms did not differ between the 2 groups. Redness or heat sensation around the surgical wound was more frequent in the AO group (47.9%) than in the DO group (21.7%) (P = .02). The mean C-reactive protein levels during infection diagnosis was 8.9 mg/dL in the AO and 4.0 mg/dL in the DO group (P = .02). All patients in the DO group had deep-layer infection. In the DO group, revision surgery and additional instrumentation were required, and the duration of parenteral antibiotic use and total antibiotic use was significantly longer than that in the AO group. Screw loosening, disc space collapse, and instability were higher in the DO group (65.2%) than in the AO group (41.1%) (P = .04). However, the length of hospital stay did not differ between the groups.Delayed-onset PSI requires more extensive and longer treatment than acute-onset surgical site infection. Clinicians should try to detect the surgical site infection as early as possible.

Tissue sampling is non-inferior in comparison to sonication in orthopedic revision surgery

BACKGROUND

The aim of this study was to assess the role of sonication fluid cultures in detecting musculoskeletal infections in orthopedic revision surgery in patients suspected of having peri-prosthetic joint infection (PJI), fracture-related infections (FRI), or postoperative spinal implant infections (PSII).

METHODS

Between 2016 and 2019, 149 cases with a data set including sonication fluid cultures and tissue specimen and histological analysis were included. Accuracy of each diagnostic tool as well as the influence of antibiotic therapy was analyzed. Pathogens identified in the sonication cultures and in the associated tissue samples were compared based on the matching of the antibiograms. Therapeutic benefits were then assessed.

RESULTS

Of 149 cases, 43.6% (n = 65) were identified as PJI, 2.7% (n = 4) as FRI, 12.8% (n = 19) as PSII, 6.7% (n = 10) as aseptic non-union, and 34.2% (n = 51) as aseptic implant loosening. The sensitivity and specificity of tissue and synovial specimens showed no significant difference with respect to sonication fluid cultures (sensitivity/specificity: tissue: 68.2%/96.7%; sonication fluid cultures: 60.2%/98.4%). The administration of antibiotics over 14 days prior to microbiological sampling (n = 40) resulted in a lower sensitivity of 42.9% each. Histological analysis showed a sensitivity 86.3% and specificity of 97.4%. In 83.9% (n = 125) of the cases, the results of sonication fluid cultures and tissue specimens were identical. Different microorganisms were found in only four cases. In 17 cases, tissue samples (n = 5) or sonication (n = 12) were false-negatives.

CONCLUSION

Sonication fluid culture showed no additional benefit compared to conventional microbiological diagnostics of tissue and synovial fluid cultures. Preoperative administration of antibiotics had a clearly negative effect on microbiologic test accuracy. In over 83.9% of the cases, sonication fluid and tissue cultures showed identical results. In the other cases, sonication fluid culture did not further contribute to the therapy decision, whereas other factors, such as fistulas, cell counts, or histological analysis, were decisive in determining therapy.