The present study demonstrates that the combined method of serum IL-6 and CRP can be used for the diagnosis of PJI, and the AUC of this combined method was higher than that of either IL-6 or CRP alone (0.9628 versus 0.9224 and 0.9133).
Since there is no gold standard for the diagnosis of PJI, the question of how to accurately judge PJI or aseptic loosening has always been concerning to surgeons, microbiologists and infectious disease specialists. Traditionally, the combined or single use of the serum erythrocyte sedimentation rate (ESR) and/or CRP was most commonly performed in the diagnosis of PJI, which is also one of the criteria of the MSIS definition. One meta-analysis of 12 studies of serum ESR and CRP for the diagnosis of periprosthetic hip infection showed that the sensitivity and specificity were 0.860 (95% CI, 0.825 to 0.890) and 0.723 (95% CI, 0.704 to 0.742) as well as 0.869 (95% CI, 0.835 to 0.899) and 0.786 (95% CI, 0.769 to 0.803), respectively . However, the diagnostic value of the serum ESR and CRP was limited; thus, some low virulence cases may be lost . A new inflammatory marker for the diagnosis of PJI is still needed. Serum IL-6 appears to be a superior postoperative inflammatory indicator compared with ESR and CRP. In a patient without complications from arthroplasty surgery, the IL-6 level reached its peak value more rapidly than the CRP or ESR levels, and it also rapidly returned to normal [26, 27]. Serum IL-6 shows great potential for the diagnosis of PJI. Di Cesare and colleagues found that serum IL-6 had the higher diagnostic accuracy compared with CRP and ESR in diagnosing infection following hip and knee replacement (97% and 83%, 69%, respectively) . A prospective study by Abou et al  also found similar results, namely, that serum IL-6 had better diagnostic accuracy than CRP and ESR (92.5% versus 87.5%, 82.5%, respectively). However, some reports found that the utility of IL-6 was not suitable for diagnosing infection after shoulder arthroplasty and even had worse results than serum ESR and CRP. The reason for the lower diagnostic value for shoulder than for hip or knee infection is probably due to the low virulence of the bacterium Propionibacterium acnes (P. acnes), which commonly occurs in periprosthetic shoulder infections [28, 29]. Previous studies have found that either single CRP or IL-6 has limitations in terms of false-negative results during low-grade infection [25, 28–30], whereas Ettinger et al showed that the combined test of IL-6 > 5.12 pg/mL and CRP > 0.3 mg/dL could detect 94% of low-grade infections in periprosthetic hip, knee and shoulder infections; two of these cases were the result of infection by P. acnes . Although there was no further study performed in periprosthetic shoulder infections to verify the detectability of weakly virulent bacteria, nevertheless, studies of combined IL-6 and CRP detection in periprosthetic hip and knee infections were published in these years [4, 7, 9, 20–23], Some reports have supported that the combination method can improve the diagnostic accuracy of PJI, especially that it can increase the sensitivity and/or specificity compared to their individual detection [4, 7, 9, 20–23]. Interestingly, there were 4 studies that excluded patients with inflammatory diseases as well as those who received antibiotic therapy. This is also consistent with our present study’s subgroup in which no antibiotics were used. The results from this subgroup showed higher sensitivity and specificity than the control group (93% and 99% versus 83% and 86%, respectively). Three studies were related to hip and knee infections, and 2 of them showed a high sensitivity and specificity [4, 7]. Yildirim et al  showed that the combined method had a sensitivity of 99% and a specificity of 98% in periprosthetic knee infections. Abou and colleagues found that detection of CRP plus IL-6 had a sensitivity of 100% and specificity of 99% in 11 infection cases; this combined method had the highest sensitivity and specificity among the single tests of serum ESR, CRP, white blood cell count (WBC) and IL-6 . However, Buttaro et al’s research also found 11 PJI cases in which the sensitivity and specificity was 57% and 100%; the authors of this study considered that the reason was probably due to the limited data and concluded that the growth of bacteria in culture conditions or positive histopathological results was the gold standard .
The results of this meta-analysis show that serum IL-6 and CRP are a valuable combined method for the diagnosis of PJI and that the pooled specificity is higher than the quantification of either individual marker alone. Six of our included studies also supported the conclusion that the combined method resulted in higher specificity than either single IL-6 or CRP [4, 7, 8, 20, 22, 23]. In addition, we found that the results of blood collection from patients without prior antibiotic treatment had a higher specificity (0.99 versus 0.86, p = 0.01) than that of blood collected from patients with prior antibiotic treatment or with an unclear situation. The pooled specificity of serum IL-6 and CRP was higher than its sensitivity, and the pooled sensitivity was lower than that of either IL-6 or CRP tested individually. Unfortunately, the potential factors were unclear, and the combined test probably used a diagnostic tool that had been ruled out. However, based on the antibiotic subgroup analysis and current related literature [4, 7], when testing was performed outside of the situation of inflammatory disease or antibiotic therapy, the combined method showed higher sensitivity and specificity than individual testing in periprosthetic hip or knee infections.
The present meta-analysis has some limitations. First, 2 of the included studies only had 11 infection cases [4, 20], and the small sample size potentially influenced the overall results. Second, although the diagnostic standard was always used for identifying infection cases, the use of different diagnostic standards to estimate the value of diagnostic tools resulted in different sensitivity and specificity values . Third, in the antibiotic subgroup, we found that the combination of serum testing without antibiotic therapy further improved diagnostic accuracy. However, most of the results of the antibiotic therapy group were unclear, and all included those without inflammatory disease, so further study is still needed.