Is Procalcitonin (PCT) a reliable biomarker for preoperative diagnosing of low grade periprosthetic joint infection? A prospective study

After approval of the institutional review board (18–8042-BO), a prospective study was performed of data gathered from Department of Orthopedics and Trauma Surgery from University of Duisburg-Essen, Germany, in patients with persisting pain [22] after hip, knee and shoulder arthroplasty.

All patients signed informed consent forms prior to being enrolled. The study was conducted in accordance with the declaration of Helsinki.

Medical history, clinical examinations, laboratory values including C-reactive protein (CRP) and joint aspiration fluid were gathered preoperatively as routine diagnostic procedures. Based on the findings of the preoperative diagnostic tests, the patients were considered as aseptic or septic according to the 2018 Definition of periprosthetic hip and knee infection [4].

The study focused on the differentiation between low-grade infects and aseptic cases. According to WAIOT definition patients without two or more signs or symptoms of local inflammation (pain, swelling, redness, warmth, function laesa) were classified as low-grade infects [23]. In order to determine the impact of renal dysfunction on serum and synovial values of PCT, serum creatinine concentrations were gathered at the time of joint puncture.

Inclusion criteria were a sufficient amount of synovial fluid for all determinations, and full clinical and laboratory data to allow for diagnosis of periprosthetic infection (PJI). Patients were further excluded, if they showed signs of early postoperative PJI (8 weeks) due to lack of reliability of synovial and serologic markers shortly after surgery [24, 25]. Metallosis, other inflammatory comorbidities (HIV, rheumatic diseases), and previous or concomitant antibiotic therapy were considered as exclusion criteria.

All patients gave their written informed consent that surplus material of their blood and synovial samples which is not needed for standard diagnostics is used for research studies.

Blood was taken from the cubital vein the day before surgery. Synovial aspiration was executed avoiding an admixture of blood with an 18-gauge needle. Synovial fluid was aseptically aliquoted into sterile tubes and centrifuged for 8 min at 4 °C with 2000 g. The synovial fluid samples were put on ice and transported within 60 min to Laboratory of Institute of Medical Psychology and Behavior Science University of Duisburg-Essen and frozen at − 80 °C.

S-PCT levels were quantified under the use of immunoassay (Centaur, Siemens, Germany) with lower limit of detection of 0.02 ng/mL (normal < 0.5 ng/mL). Serum CRP was analyzed by immune turbidimetry (Centaur, Siemens, Germany) (normal < 0.5 mg/dl). Synovial leukocyte level and percentage of polymorphic neutrophils was measured by flow cytometry with EDTA plasma (normal range, < 3000/μl and < 80%). SF-PCT levels were measured using a standard quantitative PCT enzyme immunoassay kit, according to the manufacturer’s instructions (Anti-Procalcitonin antibody ab166963, ABCAM, Cambridge,UK). Synovial alpha-1-Defensin was analyzed using a standard quantitative enzyme immunoassay kit (Human α-Defensin 1 Antibody, R&D Systems Bio-Techne, Minneapolis, USA)(cut-off level 4800 ng/mL). The results were given as standardized signal relative to a tolerance limit value (interpretation values: < 0.9 aseptic, 0.9–0.99 unspecific, ≥ 1.0 septic). Synovial CRP was analyzed under use of a quantitative enzyme-linked immunoassay (CRP ELISA (EU59131), IBL International GmbH, Hamburg, Germany) (cut-off level (> 6,9 mg / l)).

From July 2018 to June 2019, 78 patients introduced themselves with persisting pain after hip, knee and shoulder arthroplasty in the consultation hour. Seventy patients could be included in the study. Three patients were excluded due to insufficient amount of synovial fluid via preoperative puncture. Two patients were excluded due to inflammatory diseases, another two due to early postoperative PJI. One patient was excluded because antibiotic therapy was already started prior to the puncture. All 78 patients who introduced themselves in consultation hour were operatively treated. In all 78 cases histological specimens were taken according to the current recommendations in the 2018 Definition of periprosthetic hip and knee infection [4].

From the 70 included patients, 47 patients were identified as having an aseptic joint effusion according to the Definition of Parvizi et al. (2018) were included into the study. The group included 27 women and 20 men with a mean age of 66 ± 12.5 (38–88) years. There were 18 knees, 27 hips and 2 shoulders. The group consisted of 45 patients with polyethylene wear debris induced osteolysis and 2 hips with corrosion of modular head-neck junction. The mean BMI (Body Mass Index) was 26.7 ± 3.1 (22–37).

In the same period, 23 patients were classified as having a PJI according to the Definition of Parvizi et al. (2018). No patient was classified as having a high-grade infect according to the WAIOT definition. The group consisted of 15 women and 8 men with a mean age of 72 ± 11.3 (47–89) years. There were 3 knees, 17 hips and 3 shoulders. The mean BMI was 27.1 ± 7.3 (19–45). In 16 aspirations joint fluid was tested positive in microbiological culture. Bacteria were identified in 16 (70%) of 23 patients of the infection group. Staphylococci were found in 11 (69%), Propioni bacteria and and Enterococci in each two (13%) and Serratia marcescens were found in one (6%). In 7 patients (29%) in the infection group with positive histologic specimens for infection, no bacteria could be isolated after 14 days incubation. The patients who were identified as having PJI were operatively revised via two- stage revision with implantation of an intermittent antimicrobial-impregnated spacer.

There were no significant differences in age (p = 0.32) sex (p = 0.53) and age at time of surgery (p = 0.70) between the two groups. The distribution of site of joint arthroplasty was significantly different between the two groups (p = 0.01) with higher rates of hip arthroplasties in the PJI group.

The S-PCT measurement was positive in 1 joint and negative in 69 (see Fig. 1). The mean S-PCT level in the PJI and aseptic groups was 0.05 ng/ml (0.00 to 1.03) and 0.02 ng/ml (0.00 to 0.18), respectively (p <  0.001). With a cut-off value of 0.5 ng/ml, S-PCT showed a specificity of 91% and a sensitivity of 13%. Comparing these data with the diagnosis criteria of PJI according to the Definition of Parvizi et al. (2018), it was found that the PCT-assay was false-positive in 0 and false-negative in 22 cases. The mean SF-PCT in the PJI and aseptic groups was 2.7 ng/ml (0.53 to 9.7) and 8.7 ng/ml (0.25 to 87.9), respectively (p <  0.001) (See Fig. 2). SF-PCT showed with a cut-off level of 5.0 ng/ml a sensitivity of 13% and a specificity of 52%. The mean serum CRP values in the PJI and aseptic groups was 2.3 mg/dl (0.0 to 8.6) and 0.35 mg/dl (0.0–1.9) respectively (p < 0.001) (See Fig. 3). The mean SF-CRP in the PJI and aseptic groups was 19.6 μg/ml (0.6 to 339) and 1.4 μg/ml (0.4 to 5.3), respectively (p < 0.001) (See Fig. 4). The mean synovial fluid alpha-1-defensin levels were significantly higher (p = 0.006) in PJI group with 3.6 μg/ml (0.2–5.7) than in aseptic group with 2.0 μg/ml (0.2–5.7). Synovial alpha-1-defensin showed a sensitivity of 52% and a specifity of 88% with a cut-off of 4,8 μg/ml (Fig. 5). The data of statistical analysis are presented in Table 1.

There were no significant differences (p = 0.98) in the creatinine values between the aseptic and the PJI group. Furthermore, there was no significant correlation between PCT and creatinine values (p = 0.68).

In Table 2, the current data are compared to the literature.