Improved Diagnostic Accuracy of Periprosthetic Breast Infection: Novel Application of the Alpha Defensin-1 Biomarker.

Prompt, accurate diagnosis of breast implant infection is critical to minimizing patient morbidity. Bacterial culture false negative rate approaches 25%-30%, and better costeffective testing modalities are needed. Alpha defensin-1 (AD-1) is a neutrophil-mediated biomarker for microbial infection. With sensitivity/specificity of 97% and 96%, it has replaced culture as the preferred diagnostic modality for orthopedic periprosthetic infection, but has yet to be investigated in breast reconstruction. This pilot study compares the diagnostic performance of AD-1 to bacterial culture in suspected periprosthetic breast infection.
METHODS: Patients with prosthetic breast reconstruction and suspected periprosthetic infection were prospectively studied. Implant pocket fluid was analyzed with gram stain and culture, AD-1 assay, and adjunctive markers. Demographics, operative history, prosthetic characteristics, and antibiotic exposure were collected, and diagnostic performance of each test was compared.
RESULTS: Fifteen breasts with suspected periprosthetic breast infection were included, 10 (66.7%) of which were acutely infected. Gram stain correctly identified only 1 of 10 infections, whereas culture failed to identify 1 infection and reported equivocal/false-positives in 2 noninfected samples. AD-1, however, correctly classified all 15 samples. AD-1 exhibited 100% sensitivity and specificity, comparing favorably to culture (sensitivity: 90%, specificity: 60%), although this did not reach significance (P=0.22). Infected breasts also demonstrated significantly higher adjunctive marker levels compared to noninfected breasts.
CONCLUSIONS: This study demonstrates the utility of AD-1 in diagnosing periprosthetic breast infection. Combining AD-1 with adjunctive inflammatory markers may allow more accurate, prompt detection of implant infection which may reduce morbidity and reconstructive failures.

INTRODUCTION

Implant-related infections complicate 20% of breast reconstructions and are associated with implant loss, multiple reoperations, and substantially diminished quality of life.[1] As approximately 75% of all breast reconstructions today are implant based, periprosthetic infection is among the most feared complications and prompt, accurate diagnosis of infection is critical to minimizing patient morbidity.[2]

Currently available diagnostic tests for breast implant-related infection are often unreliable and lack sensitivity in the setting of antibiotic exposure.[3] Bacterial culture, the current diagnostic standard, is falsely negative in 25%–30% of untreated patients.[4] Furthermore, microbiological data require several days to result, leading to delays in treatment for subclinical infections as well as unnecessary antibiotic exposure in culture-negative patients. Thus, a critical need exists for more accurate testing modalities that remain cost-effective.

Alpha defensin-1 (AD-1) is an antimicrobial peptide released from neutrophils in response to local pathogen invasion. In addition to inducing bactericidal effects through disruption of bacterial membranes, it serves as an indirect biomarker for infection.[5] Exhibiting several unique biological characteristics, AD-1 is ideally suited for detecting infection and may be a superior indicator of breast periprosthetic infection compared to currently available diagnostic tests.

Unlike bacterial cultures, which may provide misleading results in colonized tissues, AD-1 specifically targets metabolically active microbes in the setting of active infection. It also demonstrates microbicidal activity against the spectrum of bacteria as well as fungi, spirochetes, protozoa, and enveloped viruses. Bacterial cultures, in contrast, are frequently negative in atypical infections. Most importantly, studies have demonstrated that previous antibiotic exposure does not affect AD-1 performance, whereas culture sensitivity drops to 50%–70% in patients receiving antibiotics.[4]

With sensitivity and specificity of 97% and 96%, it has replaced bacterial culture as the preferred diagnostic modality for orthopedic periprosthetic infection.[6,7] However, it has yet to be evaluated in the setting of breast implant-related infection. This pilot study aimed to evaluate the diagnostic sensitivity of AD-1 in suspected breast periprosthetic infection and compare performance to bacterial culture.

METHODS

A prospective, Institutional Review Board-approved study of all adults with prior prosthetic breast reconstruction (expander or implant) and suspected periprosthetic infection requiring operative washout was conducted at a single institution. Periprosthetic infection was defined by cellulitis with abnormal drainage found intraoperatively. All included patients underwent intraoperative fluid sampling; however, the advent of the AlloX2 dual-port tissue expander has facilitated percutaneous sampling of implant pocket fluid in suspected infections and is widely used at our institution now. Implant pocket fluid was sent for gram stain, bacterial culture, AD-1 assay, and surgical pathology if indicated. The AD-1 Synovasure assay test provides a binary result of positive or negative for infection based upon average AD-1 and C-reactive protein biomarker levels in each sample[8] and is adjusted for cell lysis using hemoglobin concentration. Adjunctive infectious markers, including lactate, human neutrophil elastase, cell counts, and differentials, were also analyzed for each sample. The Synovasure assay was sent to an independent outside laboratory for analysis (Citrano Diagnostic Labs, Baltimore, MD). The AD-1 Synovasure assay test kit utilized for each patient in this study was donated by the manufacturer, CD Diagnostics Incorporated (Baltimore, MD). Demographics, operative history, prosthetic characteristics, antibiotic exposure, and postoperative course were collected prospectively. Summary statistics and nonparametric tests of association and Wilcoxon rank-sum and Fisher’s exact tests were performed; P < 0.05 was considered significant.

RESULTS

Thirteen patients and 15 breasts with suspected periprosthetic breast infection met criteria and were included, 10 (66.7%) of which were acutely infected. Women averaged 59.8 years (range: 33–77), and presented with infection an average of 84 days after reconstruction (range:7–255). All 10 infections demonstrated cellulitis, 9 had abnormal drainage indicating infection, and 5 patients additionally presented with fever and/or sepsis. Eleven samples demonstrated positive cultures: 4—coagulase-negative Staphylococcus, 3—methicillin-sensitive Staphylococcus aureus, 2—Staphylococcus lugdunensis, 1—Pseudomonas A., 1—Stenotrophomonas maltophilia, 1—Serratia marcescens, and 2—mixed cutaneous flora. Eleven patients (85%) had received oral or intravenous antibiotics before any surgical intervention.

Gram stain correctly identified 1 of 10 infections, whereas bacterial culture failed to identify 1 infection and reported mixed cutaneous flora in 2 cultures, subject to interpretation (Fig. 1). AD-1, however, correctly identified all 10 infections unambiguously. With combined analysis of sensitivity and specificity through area under the curve (AUC), AD-1 significantly outperformed gram stain (AUC = 1.0 versus AUC = 0.55, P < 0.001) and nearly reached significance compared to culture (AUC = 1.0 versus AUC = 0.75, P = 0.06) (Fig. 2).

Fig. 1.

Comparison of diagnostic tests and final diagnosis. Sensitivity comparison (gram stain vs AD-1: P = 0.01), (culture vs AD-1: P = 0.22). Each number represents the number of samples within each subgroup.

Fig. 2.

Receiver operator curves comparing performance of diagnostic tests for implant infection. P < 0.001 for AD-1 vs gram stain. P = 0.06 for AD-1 vs culture.

Infected breasts demonstrated significantly higher levels of adjunct markers compared to noninfected breasts (C-reactive protein: 22.5 ± 13.6 mg/dL versus 6.3 mg/dL, P = 0.02), (lactate: 121 ± 59 mg/dL versus 70 ± 32 mg/dL, P = 0.03), (polymorphonuclear cell %: 86.5% ± 18.3% versus 56.3% ± 21.1%, P = 0.02) (Fig. 3).

Fig. 3.

Comparison of average biomarker levels among infected and notinfected samples. CRP (mg/dL), lactate (mg/dL), PMN %. P < 0.05 for all comparisons. CRP, C-reactive protein; PMN %, polymorphonuclear cell percentage.

DISCUSSION

This prospective pilot study demonstrates the utility of AD-1 in diagnosing periprosthetic breast infection. As implant infection often mandates reoperation and may lead to reconstructive failure, early diagnosis and appropriate treatment are critical priorities in management.[9] Orthopedic prosthetic joint literature has consistently shown the diagnostic superiority of AD-1 compared to bacterial culture, particularly in complex patients with equivocal clinical presentations. The Musculoskeletal Infection Society Workgroup has now included AD-1 as a major criterion in its latest guideline defining periprosthetic joint infection.[10] The findings from this study show promise that AD-1 may behave similarly in periprosthetic breast infection. Although sample size limits the current study and statistical significance is not reached, an adequately powered prospective case–control study is currently ongoing to more definitively assess diagnostic performance. Similarly, the authors recognize that culture is not truly the gold standard in diagnosing breast implant infection; however, it is currently the best test that is both widely available and cost-effective. Finally, patient management continues to be driven by clinical evaluation due to the preliminary nature of this study. Although results should be interpreted with the above in mind, combining AD-1 with adjunctive inflammatory markers may allow more accurate, prompt detection of breast implant infection. Furthermore, incorporation of AD-1 as a diagnostic criterion for breast implant infection may better inform and standardize evaluation and management of patients with suspected infections.

ACKNOWLEDGENT

This study complies with institutional research policies and procedures and has received and maintained approval from the Institutional Review Board throughout the study period.