It is recommended that the PCR assays be used on at least three samples collected from fluids by needle aspiration or periprosthetic tissue following a surgical biopsy [32, 33].

Ideally, samples should be sent immediately to the laboratory for analyzation. In case direct transport of fresh specimens to laboratory is not possible, PCR should be performed on paraffin-embedded biopsy samples. However, clinicians should be aware that PCR techniques on paraffin-embedded biopsy samples have shown comparatively less specificity and sensitivity [34].

Before the molecular amplification as per protocol, the specimens are left overnight in a lysis buffer and proteinase K [35].

One of the main disadvantages of the PCR technique is the possibility of false-positive results in the setting of contamination of the samples by DNA from dead or contaminating environmental bacteria, or from primer cross-reactivity with human tissue [36]. For this reason, it is essential to follow standard laboratory precautions that avoid any source of contaminations when collecting and transporting microbiology samples.

Superficial wounds and swabs are not ideal for sample collection when considering PCR techniques for diagnosis because of the frequent colonization by the skin flora that can contaminate the samples and lead to false positives as described above.

False-positive results can be reduced by performing PCR on several independent samples from each patient and by using specific genes primers for the specific bacteria alongside the 16S rRNA gene primers.

The interpretation of the PCR-tested samples is another crucial step, and both positive and negative controls have to be correctly established.

An original sequence observed for the first time in a laboratory can usually be considered to be a true positive [35], while the same sequence found in two samples of two patients may suggest a potential contamination.

Care must be given to sequences from microorganisms that are usually present in water or reagents (Pseudomonas spp. and Acinetobacter spp.) and those from skin (coagulase-negative staphylococci and Propionibacterium acnes), which could contribute to contamination [37].

When a result obtained by PCR testing is doubtful, it is possible to target a second gene using the same DNA, and another sample from the patient is needed for the confirmation.

The first studies of using PCR methods for targeting common prosthesis-related pathogens by targeting the 16S rRNA bacterial gene were performed by Mariani et al. [38, 39].

They demonstrated that the positive predictive value in a cohort of 20 revision TKAs was 100% [38], and in a subsequent series of 50 revision TKAs, they found a concordance between all the culture positive specimens and PCR with no false-positive results [39].

A recent meta-analysis [40] has shown that the specificity and the sensitivity of the PCR technique for diagnosis of PJI varied based on the study design, sample type, PCR type and reference standards. They estimated the sensitivity and specificity of PCR technique on the tissue synovial fluid samples were 0.95 and 0.81 and 0.84 and 0.89, respectively.

In a prospective study Gallo et al. [21] compared PCR and culture techniques in the diagnosis of PJI, they included joint fluid samples from 115 patients, and PCR was positive in 71% of PJI cases resulting in an improved sensitivity, specificity and accuracy.

In another study, Portillo et al. [23] demonstrated that multiplex PCR had better sensitivity and specificity compared to culture from periprosthetic tissue or sonication fluid culture, especially in patients who previously received antibiotics.

Similarly, Achermann et al. [24], in a study comparing PCR techniques versus sonication for improving diagnostic yield in PJI, reported that among 19 cases that received antibiotics, multiplex PCR was positive in all 19, while the sonication cultures grew the organisms in only eight cases.

In a recent study, Kawamura et al. [41] validated a new multiplex real-time polymerase chain reaction assay to detect methicillin-resistant Staphylococcus and to distinguish between gram-positive and gram-negative strains. In this series, in 8 out of 12 samples, the PCR was positive in culture-negative cases that were treated with prior antibiotics at the time of diagnosis.

Jacovides et al. [42] showed that the Ibis T500 biosensor system can improve the utility of PCR in detecting pathogen in culture-negative cases. In this technique, the spectral signals from the mass spectrometer are used to determine the mass of each PCR amplicon and therefore the base pair compositions. This can be used to identify the bacterial species and the abundance of PCR amplicons that are present in the analysed sample [42].

The Ibis device, that is a DNA-based amplification and analysis system, revealed that 88% of cases that were initially considered to have aseptic loosening had a subclinical infection [42].

The PCR has shown another promising application in the intraoperative decision-making pathway in revision surgery. Kobayashi et al. [43] demonstrated in a prospective study that the PCR revealed methicillin-resistant Staphylococcus infection in specimens from 6 of the 30 operations analysed, and the 16S rRNA gene universal polymerase chain reaction analysis was positive for specimens from 13 operations with an overall sensitivity of 0.87 and a specificity of 0.8.

In another study, Bergin et al. [44] showed how the rRNA RT-qPCR assay was able to detect as few as 590 colony forming units/mL of Staphylococcus aureus and 2900 colony forming units/mL of Escherichia coli, demonstrating 100% specificity and positive predictive value with a sensitivity equivalent to that of intraoperative culture that was influenced by antibiotic administration.

Because RNA rapidly degrades upon cell death, rRNA RT-qPCR assay system is useful only to detect living bacteria and the viable bacterial load.

Greenwood-Quaintance et al. [45] compared the PCR-electrospray ionization mass spectrometry to culture using sonication fluid from 152 subjects with PJI and found that the sensitivities for detecting PJI were 77.6% for PCR and 69.7% for culture (p = 0.0105). This difference was even more marked among the patients who had received antimicrobials before surgery, the specificities being 93.5 and 99.3%, respectively, (p = 0.0002).

The same group of researchers [46] published another study in which PCR together with electrospray ionization mass spectrometry applied to synovial fluid specimens had 81% sensitivity and 95% specificity for the diagnosis of PJI.

Alraddadi et al. [47] reported 16S rRNA PCR was a valuable tool in the antimicrobial management and on how clinicians made important therapeutic antimicrobial choices based on the PCR results. In this study, 19 patients, previously considered clinically infected by infectious disease consultants, successfully discontinued their antibiotic therapy based on negative PCR assay.

Despite its high sensitivity, the validity of PCR technique is still limited by the number of false-positive results due to both the high magnification power of DNA amplification and the persistence of bacterial DNA following the bacterial death [48] (Table 11.1).

When universal primers are used to amplify the 16S rRNA genes of all bacteria present, the resulting PCR amplicons must be sequenced and compared to known sequences, and this is a lengthy and costly process that requires quality databases.

Other limitations at the moment of a wider application of PCR are its unavailability in many centres and the costs associated with routine use of this technique.

Widespread use of PCR diagnostic tests is also challenging because they need an operating theatre with direct access to a molecular diagnostics laboratory to efficiently process the samples.

Furthermore, as reported by Saeed and Ahmad-Saeed [49] in their review article, no study has looked at the real cost-effectiveness and the overall burden of the routine use of PCR in the diagnosis of PJI.

The main indication for the use of PCR is when there is a lack of microorganism culture or in a case of suspicious multi-bacterial infection, particularly in the presence of open wounds.

A broad-range PCR is able to detect mixed polymicrobial agents allowing the identification of the main organism. Usually this is not possible with routine tissue culture, where overgrowth of different species can mask the result.

Identification of resistant genes in clinical samples using PCR allows modifying the antibiotic regimens for the most effective treatment. Specific primers for PCR detection have been validated for erythromycin resistance-associated methylase genes ermA, ermB and ermC, macrolide transporter protein gene mefA, ATP-dependent macrolide efflux pump gene msrA, aminoglycoside modifying enzyme gene Aac(6′)-aph(2″), oxacillin resistance gene mecA coding a penicillin-binding protein 2a, penicillin resistance gene blaZ coding beta-lactamase, IMP-1 metallo-beta-lactamase gene (bla _IMP ) and vancomycin resistance gene (vanA, vanB) [50].

The other main advantages of the PCR are in cases that have been exposed to a previous antibiotic therapy or prophylaxis and where antibiotic therapy hasn’t been discontinued for at least 2 weeks prior to surgery [20, 22, 23] (Table 11.2).