You are on page 1of 4

Toll-Like Receptor 2 in Serum: a Potential Diagnostic Marker of

Prosthetic Joint Infection?
Emanuela Galliera,a,b Lorenzo Drago,b,c Christian Vassena,b Carlo Romanò,b Monica Gioia Marazzi,c Lucia Salcito,c
Massimiliano M. Corsi Romanellic,d
Department of Biomedical, Surgical and Oral Science, Università degli Studi di Milano, Milan, Italya; IRCCS Galeazzi Orthopaedic Institute, Milan, Italyb; Department of
Biomedical Science for Health, Università degli Studi di Milano, Milan, Italyc; IRCCS Policlinico San Donato, San Donato (Milan), Italyd

Prosthetic joint infection (PJI) is a severe complication of arthroplasty and is still lacking diagnostic gold standards. PJI patients
display high Toll-like receptor 2 (TLR2) serum levels, correlating with canonical inflammatory markers (C-reactive protein
[CRP], interleukin 6 [IL-6], tumor necrosis factor alpha [TNF-␣], and IL-1). Therefore, TLR2 serum levels could be considered a
new potential diagnostic tool in the early detection of PJI.

T oll-like receptor (TLR) represents the first line of defense
against invading pathogens (1) by recognizing the invading
bacteria and activating the inflammatory response aimed to elim-
TABLE 1 Patient clinical features

Characteristic Controls PJI patients
inate the pathogen and repair the damaged tissue. Among TLRs,
Amt of TLR (pg/ml) ⫾ SD
TLR4 and TLR2 recognize a broad spectrum of Gram-positive and TLR4 348.81 ⫾ 28.61 384.28 ⫾ 43.84
Gram-negative bacteria, respectively, and induce the main in- TLR2 171.04 ⫾ 34.36 504.19 ⫾ 64.96

flammatory response (2, 3). Periprosthetic joint infection (PJI) Amt of inflammatory marker (pg/ml) ⫾ SD
(4–8) is one of the main adverse events of orthopedic surgical C-reactive protein 0.48 ⫾ 0.27 2.63 ⫾ 1.79
TNF-␣ 1.78 ⫾ 0.51 6.70 ⫾ 1.94
procedures (9–11). Currently, a large number of tests are available IL-1 2.19 ⫾ 0.52 15.79 ⫾ 3.62
for PJI diagnosis, ranging from hematological markers of infec- IL-6 1.822 ⫾ 0.75 11.2 ⫾ 3.86
tion and inflammation to intraoperative culture and histology No. of patients with:
analysis. Nevertheless, there is still a lack of gold standards for the Staphylococcus aureus (Gram positive) 11
Staphylococcus epidermis (Gram positive) 7
diagnosis of PJI (12, 13), because the clinical presentation of PJI is Staphylococcus xylosus (Gram positive) 1
often ambiguous (14), and classical inflammatory markers can be Staphylococcus warneri (Gram positive) 1
Staphylococcus caprae (Gram positive) 1
misleading (15–17). In order to optimize the diagnostic process, Staphylococcus aureus (Gram positive) 1
infection biomarkers with fast response and high sensitivity and Streptococcus anginosus (Gram positive) 1
Streptococcus agalactiae (Gram positive) 1
specificity for infection are needed (7). Enterococcus faecalis (Gram positive) 2
In this context, TLR could be useful for PJI diagnosis. Indeed, Corynebacterium striatum (Gram positive) 4
Klebsiella pneumonia (Gram negative) 1
TLR expression has already been considered of relevance in dif- Pasteurella multocida (Gram negative) 1
ferent inflammatory conditions and infections (18), but so far the Staphylococcus aureus (Gram positive)/Acinetobacter 1
baumannii (Gram negative)
potential diagnostic use of this molecule remains unexplored. For Staphiylococcus aureus (Gram positive)/Pseudomonas 1
this reason, in this study the serum levels of the two main TLRs aeruginosa (Gram negative)
involved in bacterial infection, TLR2 and TLR4, were measured in
No. of females and males 16 females, 12 16 females, 12
PJI patients and in nonseptic patients undergoing implant revi- males males
sion and were compared to classical inflammatory parameters,
Mean age (yrs) ⫾ SD 68 ⫾ 22 63 ⫾ 21
such as C-reactive protein (CRP), and with the main inflamma-
tory cytokines, interleukin 1 (IL-1), IL-6, and tumor necrosis fac-
tor alpha (TNF-␣), in order to explore the potential use of TLR2
and TLR4 serum levels as novel diagnostic tools for PJI identifica- comorbid conditions that could affect the expression of TLR2 and
tion. other markers, no antibiotic therapy in progress, and no diabetes
Patient population was described in Table 1. We selected 32 mellitus type 2 or obesity.
patients undergoing revision of total hip or total knee joint arthro- PJI and control patients were matched for age, sex, and severity
plasty and displaying prosthetic chronic infection for at least 6
months, as demonstrated by clinical and laboratory signs typical
of bone joint infection: swelling, erythema, joint pain, and secre- Received 30 September 2013 Returned for modification 29 October 2013
tion of purulent material. Diagnosis of infection was confirmed Accepted 6 November 2013
according to the criteria set forth by Spangehl et al. (19): at least Published ahead of print 13 November 2013
three positive results for (i) erythrocyte sedimentation rate, (ii) Editor: G. V. Doern
C-reactive protein and aspiration, (iii) frozen section, or (iv) in- Address correspondence to Emanuela Galliera,
traoperative culture. As a control, we selected 28 noninfected pa- Copyright © 2014, American Society for Microbiology. All Rights Reserved.
tients undergoing routine orthopedic surgery without any other doi:10.1128/JCM.02727-13
underlying disease or infection of inflammation and showing no

620 Journal of Clinical Microbiology p. 620 – 623 February 2014 Volume 52 Number 2

05 were considered tein. normality of distribution of sion strictly reflects the progression of the infection in the host. while the present work is the first. Human TLR4 serum Accordingly. (A) Serum levels of TLR2 and TLR4 in prosthetic joint infection patients (gray bars) and noninfected patients (white bars). bound by TLR4. after microbial joint infection (23). USA). Blood was drawn from all patients for serum separation. So the three groups was verified by the Kolmogorov-Smirnov test for far. the alteration of TLR2 and TLR4 has been evaluated only at the normal distribution.005 were considered very signifi- aliquoted. IL-6. USA). 25). and stored at ⫺80°C until further analysis. cant. as a con- to the manufacturer’s protocols (USCN LifeScience Inc. and the 95% confidence interval of the regression Valley. sequence.0 software. Serum TLR2: a Diagnostic Marker of PJI? FIG 1 TLRs and inflammatory mediators. serum using an enzyme-linked immunosorbent assay (ELISA) 21). of illness. gene expression level (24. (R&D Systems. making it suitable for routine clinical diagnosis. while a sandwich duo set assay. a significant increase of TLR2 but not of TLR4 (Fig. according to the manufacturer’s protocols small amount is due to Gram-negative bacteria. catalog number E90753Hu). (C) serum levels of C-reactive protein in prosthetic joint infection patients (gray bars) and noninfected patients (white bars). Human IL-1. PA. which measures the amount of circulating pro- variance (ANOVA) test. in infected patients. Hubei. IL-6. indicating that TLR2 expres- For all the parameters analyzed. CRP was measured using immunoturbidimetry on an auto. con- TLR2 ELISA kit (DY2612.5% of Gram-negative infection 621 . line was calculated by using PRISM 3. TLR2 has been described to be crucial in joint infection (20). and P values of ⬍0. we observed 87.asm.. MN. In order to February 2014 Volume 52 Number 2 jcm. Linear regression analysis was performed between the different mated biochemical analyzer (CRP-Latex assay. (B) serum levels of inflammatory cytokines (IL-1. to Statistical analysis was performed using a one-way analysis of our knowledge. 1). a Gram-positive bacterium recognized by TLR2 (22). Minneapolis. and TNF-␣ and TLR2 were measured in Surgical infection is due mainly to Staphylococcus aureus (20.5% of Gram- concentration was measured by ELISA sandwich assay according positive and only 12. and TNF-␣) in prosthetic joint infection patients (gray bars) and noninfected patients (white bars). R&D Systems) was optimized for the tributing to the degenerative process and destructive arthropathy analysis of TLR2 in cell supernatants and serum. significant and P values of ⬍0. Olympus. Central groups of data. Wuhan. People’s Republic of China.

Koc M. Top left. 11.17150. Crook DW. http://dx. Mattina R.5000054. Ries MD. 2). . Infected patients displayed a significant in. Given the small sample size of patient biol. which is increased in Orthop. Berendt AR. TLR2 and C-reactive protein (PCR). Hanssen AD.1007/s10354-010-0765-6. Nephrol. Micro- detection of PJI (Fig. Leung DT. Gold HS. previously 8. 2011. Odabasi Z. Arikan H. 87:1746 –1751. we mea. diagnosis. 2011.E. J. Frank KL. REFERENCES 12. Atkins BL. J.08. in dence of deep prosthetic infections in a specialist orthopaedic hospital: a 15-year prospective survey.2004. Qian Q. but taken 10. 93:2242–2248. Bone Joint Surg.bbrc. 2009. Parvizi J. 1998. Wochenschr.doi. the 5.doi. E. Akoglu E. Corsi MM. we compared joint infection by Gemella sanguinis and associated “pseudosatelliting” them with canonical markers of infection and plasty on hospital and surgeon resource utilization. ney disease and haemodialysis: relation with inflammation. Krebs V. TLR2 and IL-1. indicating that the serum TLR2 molecule is able to 6.doi. Bone Joint Surg. 4. phenomenon on culture. Walinchus L. Prosthetic joint infections: microbiology. 15. Kumar H. Commun. Parvizi J.doi.01149. Microbiol. Toll-like receptors and innate immu.03. Phillips JE. Patel R. and IL-1␤. http://dx. Am. FIG 2 Correlation analysis. 88:943–948. Prospective evaluation of criteria for microbiological diag- aureus infection (26). TLR2 and IL-6. 26:955–963. Biophys. 88:869 – 882. 2011. Bozic KJ. . Am.Galliera et al. since TLR2 mecha. Dozio E.4846-4849. Simpson H.doi. Artif.doi. Bone Joint Surg. Parvizi J. nism of action leads to an inflammatory response (3).2106/JBJS. The OSIRIS Collaborative Study IL-6 and IL-1␤. which exert a protective role on the tissue in Journal of Clinical Microbiology . In PJI stand? J. Update on the role of Toll-like receptors during bacterial thetic infection.2106/JBJS. the results of this pilot study are preliminary. Am. 2006. J. 1B). 40:614 – 615.1016/j.88B7.1007/s00104-010-2010-7. 2004. http://dx. Noy M. Int. Organs 34:847– 855. Group. http://dx.doi. Br. Clin. thetic joint infection: the utility of a simple yet unappreciated enzyme. in particular IL-6. De Vecchi E. http://dx association with canonical parameter of inflammation. Joint arthroplasty and infection: where do we described to be a significant marker of PJI (16). Jacovides C. 2011.J . Implants 21:225–232. 2011. coefficient of correlation r2 is reported. 49:3395–3397.010. Adeli 13. 388:621– 625. 287–293. Am. /10. http://dx .org for systemic response. 42:4846 – 4849. McLardy-Smith P. Knapp. Vanholder R. /10. Elliott TS. First report of prosthetic 16. Microbiol.1128/JCM. Crane TP. tential diagnostic marker of PJI. 3. 622 jcm. Moreover. Agents 32: Transplant. bottom right.2106/JBJS. S. 14. Biochem. gold standard clinical marker of infection. they indicate that serum TLR2 can be considered. http://dx.2008. Grimer RJ. evaluate TLR2 circulating levels as diagnostic tools. TLR2 displayed a strong positive correlation with both 9. a new po. Clin. Asicioglu marker for prosthetic joint Clin. patients. TLR2 and TNF-␣ (TNF-a). Toprak A. Res. Wien. groups. Kawai T. J. Eksioglu-Demiralp E.5301/ijao. Esposito S. J. Antimicrob. The inci- together.doi. http://dx. Tulunay A. Parvizi J.42. Pathophysiological basis of surgery-linked sepsis. 2005. Med. Toll-like receptor expression in monocytes in patients with chronic kid. Long Term Eff. Drago L. Bauer TW. Linear regression analysis between TLR2 and inflammatory parameters. Athanasou N. Davis EM. 2011. Peto TE. bottom left.01050-11. sured the circulating levels of the three main inflammatory cyto- 7. top right. The impact of infection after total hip arthro- nity. Chirurg 82:199 –207.D. Elbir Y. Med.1093/ndt/gfq500. Diagnosis of peripros- detect an inflammatory Parvizi J. For each analysis. Deeks JJ. Vollmar B.doi .1302/0301-620X.1016/j. http://dx. confirming the importance of TLR2 in the nosis of prosthetic-joint infection at revision arthroplasty. Ghanem E. 2010. infections and sepsis. Int. Leone S. Kobayashi N. http://dx. J. 160:107–111. Vassena C.doi. 2008. Vegari DN. Glorieux G.062. J. Bone Joint Surg. icaA is not a useful diagnostic 2. Antoci V.doi. 36:2932–2939. Diagnosis of peripros- 1. J. 2011.1128/JCM. Molecular diagnostics in peripros- kines: IL-1␤ and TNF-␣ for local inflammatory response and IL-6 thetic joint infection. PJI patients. Akira S. management and prevention. TLR2 showed a strong positive correlation with CRP (Fig.02937.doi.2009. crease of all the cytokines analyzed. New definition for periprosthetic joint infection.

Immunol. 34: 29:247–251. Int. . Microbiol. Lee K. Semin. Am. Improved detection of infection in hip re. Tsezou A. Pulcrano G. Lissauer ME. Johnson SB. Toll-like receptor kinetics in septic shock patients: a vance in patients with Staphylococcus aureus infection of joint prosthesis. able RR. P. Bone Joint Surg. McAvoy E. Muruve DA. preliminary study. J. Francis KP. Masri BA. C-reactive protein. Fournier B. Papathanasiou I. Feild CJ. Bernthal NM. Ho M. Krishna S. 2013. Billi F. 2011. O’Connell JX. Procalcitonin. Gorman SP. J. Gralewski JH. Piazza O. 24. Immunopathol. Br. Iwakura Y. Rossano F.00167.doi.doi. Waterhouse CC. Miller LS. Lonardo M.asm. Malizos KN. Anderson N. Front. Karachalios T. 2011. Kapadia R. sterile inflammation 1 day before sepsis diagnosis. 40:1639 –1650. Cho JS. Hum. 81:672– 683.2012 Hanna D. against post-arthroplasty Staphylococcus aureus infection. 72:47–53. Protective role of IL-1beta 21. 2009. 2011. Tufano R. Bone Joint Surg. 2011. 2:167. 23.1097/SHK. and receptor 2 in response to lipoteichoic acid and Staphylococcus aureus soluble intercellular adhesion molecule-1 as markers of postoperative or. 25:425– 433. Gillrie MR. Res. Pharmacol. Pribaz JR.doi. The function of TLR2 during staphylococcal diseases. Orthop.1002/jor.200939929.1016/j. Patrick S. . Bochicchio GV. Osmon DR. J. Functional assessment of Toll-like receptor 2 and its rele. Oikonomou 80:568 –572. Poultsides L. plasties. Res.doi. Innate and adaptive immune responses mediated by the NF-kappaB pathway in septic arthritis. 2012.doi. Divergent roles of Toll-like 29:1621–1626.001. Hasday 19. 24:433– 440. Shock 31:238 –244. placements. 2012. Zbytnuik L. http://dx. Kubes P. Towns Orthop. Eur.21414. 25. Cross AS. Duncan CP. Miller LS.10 26. Pharmacol. Cell Infect. Davis RI. Ramage G. Stavrakis AI.1002/jor. Fiori PL. http://dx. Davis SP. 261–280. Int. http://dx. analysis of preoperative and intraoperative investigations for the diagnosis Differential expression of Toll-like receptor genes: sepsis compared with of infection at the sites of two hundred and two revision total hip arthro. 623 . J. 22. Whiteford CC. Ramos RI. interleukin-6.1002/eji thopaedic joint prosthesis infections. Berbari EF. Prospective JD. J. Brown RA.1007/s00281-011-0292-6. J. Catania MR. El-Helou O. 1999. Immunol. A currently underestimated problem. Tunney MM. Nussbaumer WA. http://dx. against Staphylococcus aureus skin Spangehl MJ. . Serum TLR2: a Diagnostic Marker of PJI? Romano C. February 2014 Volume 52 Number 2 jcm. 2010.0b013e3181834991. Immunopathol. in vivo. Nixon JR. Scalea Immunopathol.3389/fcimb.doi.21239. 17. The catabolic role of Toll-like receptor 2 (TLR-2) 18. http://dx. J. 20.