Osteoprotegerin gene polymorphism is not associated with prosthetic joint infection after total joint arthroplasty in the Czech population

Background and aims. Osteoprotegerin (OPG; official gene symbol: TNFRSF11B) is considered a negative regulator of bone resorption via inhibition of osteoclast differentiation. Further, OPG expression has been detected in Prosthetic Joint Infection (PJI) a serious complication limiting the overall outcome of total joint arthroplasty (TJA). As OPG may be a candidate molecule for PJI pathogenesis, we investigated whether genetic variation in the OPG promoter, namely the SNP at position -163 was associated with PJI. Methods. OPG -163 T/C SNP (rs3102735) was genotyped by polymerase chain reaction with sequence specific primers (PCR-SSP) in 98 Czech patients with PJI and two Czech control groups: 1) aseptic TJA control [251 patients with TJA who did not develop PJI at least 6 yrs. after the surgery] and 2) population control (185 healthy control subjects without TJA). Results. The distribution of OPG -163 SNP genotypes complied with the Hardy-Weinberg equilibrium in all three groups. The allele frequencies of OPG -163 SNP were similar in patients with PJI (minor allele frequency: 0.14), those with aseptic TJA (0.13) and population controls (0.14, P>0.05). Further, there was no significant difference in genotype or phenotype frequency (carriage rate) between patients with PJI and both control groups (P>0.05). Conclusions. In a Czech population, the OPG -163 T/C SNP has not been found to be associated with PJI.


INTRODUCTION
Prosthetic joint infection (PJI) is a serious complication threatening both the early and late period after total joint arthroplasty (TJA).Current estimates suggest that despite careful management to avoid all sources of infection, PJI can develop in up to 1.7% of primary hip arthroplasties and 2.5% of primary knee arthroplasties 1 .Given that each year at least two millions TJAs are performed worldwide, the number of infected cases is enormous and is predicted to increase 2 .
PJI is caused by bacterial colonisation of the prosthetic joint.Regardless of particular route of infection it is clear that host factors may increase the risk of PJI by compromising the immune response 3 .Bacteria and their products are also potent inducers of bone resorption.Experimentally, significant reduction in bone mineral density shortly after the beginning of Staphylococcus aureus infection of the bone has been revealed 4 .This was effectively prevented by inhibition of RANKL (receptor activator of nuclear factor -kappa B ligand) signalling.This and other experiments support the clinical observation of a tight association between the development of infection and increased resorption of periprosthetic bone 5 .For this reason, cytokine ligands and the corresponding receptors that regulate osteoclast accumulation, maturation and viability, are probably involved in the complex pathogenesis of PJI.The axis RANK-RANKL-OPG is the key regulator of the bone multicellular unit, and genetic factors related to this axis have been shown to affect bone mineral density [6][7][8] , osteoporosis 7 and periprosthetic osteolysis 9 .A recent study demonstrated deregulation of OPG expression in osteoblasts after Staphylococcus aureus infection 10 .Also, expression of OPG was observed in periprosthetic interface membranes retrieved from septic prostheses 11 .
Osteoprotegerin (OPG; official gene symbol: TNFRSF11B) acts as a soluble decoy receptor that blocks binding of RANKL to its receptor (RANK) inhibiting osteoclast differentiation.The human OPG represents a single copy gene located on chromosome 8 (8q24).It spans 29 kb and consists of 5 exons, 4 introns, and is transcribed into four transcripts, 2.4 kb one representing the major transcript 12 .There is a degree of genetic variation in the gene with at least 10 SNP localised mostly within the 5´flanking region and also exons/introns 13 .A plausible contribution of a genetic variation in the OPG gene to PJI has so far been investigated only in a single centre and with inconsistent results 9,14 .In agreement with the current rules for conducting genetic association stud-ies 15 , and especially those relevant to areas of rheumatology and osteoimmunology 16 , we therefore investigated if OPG -163 T/C polymorphism is associated with PJI in a Czech population.

Study population
OPG -163 SNP T/C (rs3102735) was genotyped in 98 patients with PJI and two control groups: i) aseptic TJA control (251 patients with TJA who did not develop PJI at least 6 yrs.after the surgery) and ii) population control (185 healthy control subjects without TJA).Of 98 infected cases, 16 patients were recruited at the Dept. of Orthopaedics Frýdek Místek and Znojmo, Czech Republic.Both patients and controls were unrelated individuals of Czech origin.Clinical and laboratory characteristics of patients with PJI and controls are listed in Table 1.PJI was diagnosed according to the following criteria 17 : i) presence of sinus tract communicating with a joint and/or intra-articular pus; ii) coincidentally positive results of histological examination (five or more neutrophils per high power field) and culture of intraoperative samples; iii) if only intraoperative culture or histological results were positive, then at least two of the following signs had to be present: high clinical suspicion of infection (acute onset, fever, erythema, edema, persistent local pain, early prosthetic failure, wound healing disturbances, etc.), erythrocyte sedimentation rate >30 mm/ hr, C-reactive protein elevated more than 1.5 times above the laboratory reference value, positive 99m Technetium leukocyte scintigraphy (the last criterion included only in patients from the University Hospital Olomouc).Patients without PJI did not fulfil these criteria.
Informed consent for the anonymous use of their DNA for the purposes of this study was obtained from all subjects.The study was performed with the approval of the local Ethics Committee.

Statistics
The significance of differences in allelic, genotype and phenotype frequencies between all groups was determined by means of the χ2 test applying the Woolf-Haldane correction for small numbers.The relative ratio (RR), 95% confidence interval (CI) and P-value were calculated.The statistical power was determined according to the protocol described elsewhere 19 .

RESULTS
The distribution of alleles, genotypes, and phenotypes of OPG -163 SNP in patients with PJI and two control samples is presented in Table 2.The distribution of OPG -163 SNP genotypes complied with the Hardy-Weinberg equilibrium in all three groups.The allele frequencies of OPG -163 SNP were similar in the patients with PJI (minor C allele frequency: 0.14), those with aseptic TJA (0.13) and population controls as well (0.14, P>0.05).Further, the genotype and phenotype frequency (carriage rate) did not differ between the patients with PJI and either of the two control groups (P>0.05).
When the patients with PJI were stratified according to the affected joint, there was no association between OPG -163 SNP and PJI of Total Hip Arthroplasty -C allele frequency was equal in both aseptic THA and in THA/PJI groups: 0.12; P>0.05).Similarly, the distribution of OPG -163 SNP did not differ between the patients with aseptic Total Knee Arthroplasty (TKA, C allele frequency: 0.14) and those with TKA/PJI (0.16; P>0.05).
The statistical power of the present study to replicate observations reported for the British population in the initial report 9 was higher than 99%.The power to detect differences between cases and controls was 94% if aiming at an odds ratio (OR) 2.0, and 82% for OR 1.75.

DISCUSSION
The results show no association between -163 T/C single nucleotide polymorphism in the Osteoprotegerin gene and the risk of prosthetic joint infection.This finding may be interpreted in two ways, either 1) "individually" as an independent study of a candidate SNP in the OPG gene in the Czech population, or 2) in the context of previously reported controversial data from the British population: Here, Malik et al. 9 first reported OPG -163 as a risk factor for PJI.However only after three years Malik et al. 14 published a brief note that this was/might not be true due to "inadvertent data manipulation".
Given the fact, that the latter brief note 14 was not formulated as a typical erratum statement and thus has not been linked to the original report 9 in databases such as PubMed, the association between the investigated polymorphism and PJI can still be considered to be present in the "universe" of genetic association data as it may be exemplified e.g. by the current review 20 .0 ) the replication of these data coming from a single centre is required to prove there is adequate statistical power.If we understood our current work as a  The data are given as frequency of a particular genotype/allele/phenotype with absolute numbers in parentheses.
replication, it was important to note that we investigated a larger number of patients with PJI (n=98) than the original British study (n=63).Thus the statistical power of our study was more than adequate (>99%) to replicate the association of PJI with OPG -163 SNP observed in the original, first British report 9 .
It may be argued that despite the low public visibility of the "corrected" data, a replication study cannot be performed in the present.However, if we consider our study merely as an individual undertaking aimed at the assessment of the plausible association between OPG -163 T/C SNP and PJI, it has very good power (94%) to detect association in the range of OR values around 2.0, and is still adequately powered around OR 1.7 (81%); the acceptable limit being 80% (ref. 19).For the detection of weaker associations (OR <1.7), however, the number of study subjects will have to be increased which we acknowledge is a limitation of the study.
We cannot exclude the possibility that the observed lack of genetic association between OPG -163 SNP and PJI in our Czech cohort was due to inappropriately targeted selection of candidate SNP.Indeed, it has been very recently reported that the OPG -163 T/C may not affect the expression of the OPG gene 6 , however this was an isolated in vitro report.In this context, there are more SNPs located on the OPG gene 13 , However the data on their functionality and mutual relationships are mostly limited to non-Caucasian populations 21,22 .

CONCLUSION
In conclusion, even if the present study had sufficient statistical power both as an individual "exercise" and/or a replication study, it provided no further evidence that OPG -163 T/C SNP is associated with prosthetic joint infection.After more information is obtained on the spectrum of genetic variation in the OPG gene, analyses with further SNPs may allow us to draw definite conclusion about a possible role of Osteoprotegerin genetic variation in PJI.

Table 1 .
Characteristics of Czech patients with PJI and two control groups of the patients with aseptic TJA and Czech population controls.

Table 2 .
Genotype, allele and phenotype (carriage rates) frequencies of the OPG -163 SNP in the patients with PJI and two control groups of the patients with aseptic TJA and Czech population controls.