Different biomarker ratios in peripheral blood have limited value in diagnosing periprosthetic joint infection after total joint arthroplasty: a single-center, retrospective study

Currently, the diagnosis of periprosthetic joint infection (PJI) still relies heavily on the culture of pathogenic microorganisms, which is widely considered the "gold standard" [1]. However, with the emergence of specific pathogen infections such as mycobacteria, Brucella, and fungi and the formation of bacterial biofilms, these factors may diminish the detection rate of pathogens and exacerbate the challenges in PJI diagnosis and treatment [2‐4]. Approximately 5%-42% of PJI patients are estimated to demonstrate negative results for pathogenic microorganisms [5]. To increase the accuracy of diagnosing PJI, clinicians must integrate clinical manifestations, imaging evidence, and a variety of laboratory tests [1‐3, 6, 7]. Classic biomarkers of human inflammation, C-reactive protein (CRP) and the erythrocyte sedimentation rate (ESR), have been extensively employed to assist in PJI diagnosis [8]. Sigmund et al. demonstrated that the serum ESR has a sensitivity range of 33% to 95% and a specificity range of 60% to 100%, while the sensitivity range of CRP is from 62 to 100%, and the specificity range is from 64 to 96% [9]. Although the ESR and CRP can be useful in assisting with the diagnosis of PJI, their diagnostic accuracy may not always meet the needs of clinical physicians. In recent years, several researchers have proposed innovative molecular biological methods, such as polymerase chain reaction (PCR), multiplex PCR, mass spectrometric analysis, and next-generation sequencing (NGS), as new approaches for diagnosing PJI [10, 11]. Although these methods can further improve the accuracy of PJI diagnosis, they present significant challenges for implementation in primary hospitals due to their professional operation and high cost [10‐12].

As a conventional diagnostic biomarker for hospitalized patients, the peripheral blood test is a simple, cost-effective, and widely adaptable test for various infectious diseases. Several biomarkers derived from peripheral blood, such as the ESR, CRP, white blood cell (WBC) count, and leukocyte classification, can assist medical professionals in determining inflammatory status, severity of infection, and type of infection within the body [9, 13]. Recent studies have revealed that changes in the PLR, PVR, NLR and MLR are closely related to inflammatory or infectious status in the body [13, 14]. It is reported that the diagnostic accuracy of PVR in patients with fracture-related infection was no less significant than that of ESR or CRP, suggesting that PVR could function as an auxiliary diagnostic biomarker of fracture-caused infection [15]. The PLR and NLR have been found to have good diagnostic value for predicting surgical site infection, while the MLR and NLR demonstrate significant variation in feverish patients due to bacterial infections [13‐15]. A retrospective study also revealed that the PLR and NLR in the PJI group were significantly greater than those in the aseptic loosening (AL) group [16]. These different biomarker ratios reflect the severity of the infection and make an early diagnosis of the disease. Therefore, further study of the correlation between these biomarkers and PJI is significant.

To explore the diagnostic value of inexpensive peripheral blood biomarkers in periprosthetic infections, we conducted a further analysis of these biomarkers in this single-center retrospective study. We compared these different biomarker ratios with the classical serum biomarkers ESR and CRP to explore the importance of PVR, the PLR, the NLR and the MLR in the diagnosis of PJI after TJA. Furthermore, we combined these biomarkers with the ESR and CRP to evaluate the potential value of each combination in diagnosing PJI.

In this retrospective study, we compared the different biomarker ratios in peripheral blood with traditional inflammatory biomarkers ESR and CRP to explore the potential value of these biomarkers in the diagnosis of PJI. The results confirmed significant differences in WBC, PLT, PLR, PVR, NLR, and MLR between PJI and AL patients. Still, the AUC value of the ROC curve showed that the diagnostic value of these different biomarker ratios in PJI is lower than ESR or CRP. Interestingly, when all ratios were combined with ESR and CRP, the diagnostic capability of this novel combination was significantly enhanced, surpassing the performance of ESR and CRP individually or in combination.

PJI is a severe postoperative syndrome that occurs after TJA, which has been identified as a crucial contributor to knee pain, dysfunction, disability, and even mortality following TKA [6, 19]. Studies have revealed that more than 25% of TKA revisions are attributable to PJI [20], underscoring the importance of timely diagnosis for effective treatment planning and optimal therapeutic outcomes. However, preoperative diagnosis of PJI is still difficult due to the emergence of infections caused by specific pathogens and bacterial biofilm formation [1‐4]. Although diagnostic techniques such as leukocyte esterase (LE), ɑ-defensin, PCR, NGS and others have shown the potential to improve diagnostic accuracy, their high technical complexity and medical costs limit their widespread use [10‐12, 21]. Against this backdrop, our study focused on the clinical utility of readily available and cost-effective blood routine biomarkers, specifically the PLR, PVR, NLR, and MLR, either individually or in combination with the classical serum markers ESR and CRP recommended in the MSIS guidelines, for aiding in the diagnosis of PJI. By comparing the AUC, sensitivity, and specificity of these parameters with those of the ESR and CRP, we aimed to provide initial insights into the potential value of these indicators in assisting PJI diagnosis.

The ESR and CRP are important biomarkers reflecting the presence of inflammation in the body. Due to their high sensitivity and ease of detection, they have become important screening tools for PJI [9]. In a study involving 156 patients, Yang et al. reported that the AUCs of ESR and CRP for assisting in diagnosing PJI were 0.822 and 0.901, with sensitivities of 70.2% and 91.2%, respectively, and specificities of 85.9% and 82.7%, respectively, confirming that both the ESR and CRP have good diagnostic performance for PJI [22]. However, our study results indicate that the clinical value of the ESR and CRP in the diagnosis of PJI is medium, with AUCs of 0.760 (95% CI: 0.688–0.823) and 0.758 (95% CI: 0.687–0.821), respectively. Wu et al. also reported the limited utility of the traditional biomarker CRP (AUC = 0.737) in diagnosing PJI [23]. Although the numerical changes in traditional biomarkers ESR and CRP are critical reference criteria for diagnosing PJI, their diagnostic performance is not entirely satisfactory due to limitations of many factors, such as sample size, disease progression, and different types of pathogens.

The PLR and PVR represent the ratio of platelet count to lymphocyte count and mean platelet volume, respectively, which can be used to reflect the degree of systemic inflammatory response and infection status [24, 25]. It has been reported that elevations in the PLR and PVR are closely associated with bacterial infections, possibly due to platelets being activated and promoting immune cell aggregation and the production of proinflammatory cytokines during the inflammatory response [25, 26]. These biomarkers offer a range of advantages, including simplicity, speed, affordability, and so on. A study by Wang et al. revealed that PVR is a valuable biomarker for the early diagnosis of infectious bone nonunion [27]. Recent investigations have also shown the potential of the PLR and PVR in aiding PJI diagnosis [28, 29]. In a retrospective analysis of 464 patients who underwent revision surgery (191 of whom had PJI), Klemt et al. demonstrated that the PLR and PVR were valuable diagnostic tools, with an AUC of 0.86 for both indicators. A cut-off value of 237.9 for the PLR yielded sensitivity and specificity rates of 75.9% and 82.8%, respectively, while a cut-off of 27.8 for PVR led to sensitivity and specificity rates of 86.4% and 75.5%, respectively [30]. However, Wang H et al. posited that the PLR and PVR are of limited use in diagnosing PJI, with an AUC of 0.700, sensitivity rates of 51.22% and 48.78%, and specificity rates of 80.58% and 86.33% for PLR and PVR, respectively [31]. Our study produced similar results to those of Wang H et al., indicating that PLR and PVR are not superior to the ESR in aiding PJI diagnosis, as their AUC and sensitivity values are lower compared to ESR and CRP. Although the specificity of PVR was greater than that of the ESR in our study (81.8% vs 76.4%), its sensitivity was only 51.7%, which greatly increases the risk of missed diagnosis.

The NLR and MLR offer insight into the absolute value alterations ofin neutrophils, lymphocytes, and monocytes; these changes can reflect respiratory system and urinary system infections and play important roles in the diagnosis of early infectious diseases, evaluation of infection severity and prediction of clinical prognosis [32, 33]. Our research team conducted a retrospective analysis of 168 patients and revealed significantly greater NLR and MLR values among PJI patients than those in the AL patients, with statistical significance (P < 0.001), suggesting that NLR and MLR can be used to assist in the diagnosis of PJI, which was consistent with the findings of Maimaiti et al. [24]. However, Xu et al. contend that the NLR and MLR have limited diagnostic potential for PJI due to their lower sensitivity and specificity compared to CRP, which implies that the NLR and MLR cannot replace the CRP as a diagnostic tool for PJI [34]. Similarly, Jiao et al. demonstrated that the NLR did not demonstrate a significant advantage in terms of sensitivity (73.58%) or specificity (70.97%) compared to the CRP or ESR [35]. Nonetheless, the results of Yu et al. and Zhao et al. indicate that the NLR is more valuable than the CRP and ESR in early PJI detection following TJA, which may be owing to the persistently high levels of CRP and the ESR in the early stages after joint replacement, while the NLR can return to preoperative levels relatively quickly [36, 37]. In our study, the AUC, sensitivity, and specificity of the NLR were 0.723, 62.1%, and 73.6%, respectively, which were all lower than ESR. These findings suggest that the NLR is less effective than the ESR in aiding PJI diagnosis and cannot be a substitute biomarker. It is noteworthy that MLR showed the highest specificity of 83.6% compared to the ESR and CRP in the diagnosis of PJI, but this was achieved by sacrificing the sensitivity of the MLR in diagnosing PJI.

Our study suggests that changes in the level of inexpensive biomarkers such as the WBC, PLT, PLR, PVR, NLR and MLR in blood routine test are significantly different between PJI and AL groups. However, these biomarkers are not found to be superior to the classical markers ESR or CRP in diagnosing PJI and have limited value in this regard. On the other hand, when all these biomarkers are combined together with the ESR and CRP, a new combined model is formed. This combined model can effectively improve the diagnostic capability of PJI.