This study demonstrated that 25% of PJI cases investigated were culture-negative. There were some differences in clinical characteristics between the culture-positive and culture-negative PJI groups. Levels of systemic inflammatory markers such as serum WBC count, CRP level, and ESR in the culture-negative group were significantly lower than those in the culture-positive group. The treatment success rate of culture-negative PJI was no different from that of culture-positive PJI. A low serum WBC count was identified as a risk factor for culture-negative PJI, but prior antimicrobial therapy was not.
In the current study, the observation that several serum inflammatory markers (all of which are important as first-line tests for diagnosis of PJI) [16] were significantly lower in the culture-negative group than those in the culture-positive group has important clinical implications. In agreement with our results, Choi et al. reported that serum ESR in culture-negative patients was lower than in culture-positive patients [7]. In addition, a previous study by Kheir et al. demonstrated that culture-negative PJI cases had lower CRP values than PJI cases caused by Gram-negative organisms, antibiotic-resistant organisms, Staphylococcus aureus, and Streptococcus species; they also had lower WBC counts than PJI cases caused by Staphylococcus aureus and Streptococcus species [17]. The results of our study suggest that culture-negative PJI is caused by less virulent organisms and that the organism count is lower, resulting in less severe systemic inflammation. In this regard, the ideal cut-off value for these markers may need reconsideration for more accurate screening of PJI [18].
We found that the single risk factor for culture-negative PJI was a low serum WBC count. Although several risk factors for culture-negative PJI have been described [7, 8, 11, 12], serum WBC count was not highlighted. This result suggests that culture-negative PJI should be considered in suspected cases with low levels of inflammatory markers. In such cases, additional diagnostic approaches such as polymerase chain reaction or alpha-defensin tests may be required [19, 20].
Prior use of antibiotics had no effect on culture-positive or culture-negative PJI. In addition, the multifactorial logistic regression model did not identify prior antimicrobial therapy as a risk factor for culture-negative PJI. Several studies demonstrate that perioperative administration of prophylactic antibiotics has no effect on culture yield [21–24]. Tetreault et al. reported that prophylactic antibiotics administered before skin incision had no effect on the results of cultures obtained intraoperatively [21]. A prospective study by Bedencic et al. revealed no differences in diagnostic yield between cultures taken before and after administration of antibiotics to patients with suspected PJI [22]. Thus, prior use of antibiotics may not affect culture results. By contrast, several studies demonstrate that prior use of antimicrobial therapy is a risk factor for culture-negative PJI [7, 11, 12]. Malekzadeh et al. reported that antimicrobial therapy before diagnosis of PJI is associated with increased odds (odds ratio, 4.7) of being culture-negative [11]. In addition, Ibrahim M. S. et al. showed that pre-operative use of antibiotics was a risk factor for culture-negative PJI (odds ratio, 4.1) [12]. A reasonable explanation for a causal association between antibiotics administration and culture-negative results is that antibiotic pressure can induce a viable but non-culturable state in a biofilm [25]. Thus, prior antibiotics use in suspected cases of PJI is a controversial issue; therefore, further studies are required.
In this study, the treatment success rate of culture-negative PJI was similar to that of culture-positive PJI. In agreement with our results, previous studies reported similar outcomes for culture-positive and culture-negative PJI patients [12, 26]. In addition, Choi et al. reported that the success rate of infection control was higher in the culture-negative group [7]. Thus, culture-negative PJI may not necessarily be a negative prognostic factor for PJI. Certainly, culture-negative PJI was associated with less virulent organisms and a lower organism count; therefore, the clinical outcome of culture-negative PJI might not be poor despite the lack of culture results and antibiotic sensitivity. On the contrary, inadequate treatment may lead to unsuccessful results for culture-negative PJI. In fact, the rate of treatment success was greater for patients with 2-stage exchange than for those with irrigation and debridement [27]. In addition, selection of antibiotics is difficult in the absence of information about the causative organism.
This study has several limitations. First, the rate of prior antibiotics use was high: 61% and 56% in the culture-positive and -negative groups, respectively. A previous study reported a rate of prior use of antimicrobial therapy of 64% in the culture-negative group and 24% in the culture-positive group [11]. In our study, many PJI cases were referred from other hospitals, and many were had already been treated with antibiotics. This limitation might have led to selection bias.
Second, our sample sizes were relatively small because PJI is a rare condition in a single center.
Further multicenter studies need to be performed with more cases. Third, while 5 days culture period was applied in this study, an international consensus meeting recommended that routine cultures should be maintained for 5–14 days [
28]. It is, therefore, possible that longer culture duration may have yielded different results. However, Kheir et al reported that most organisms were cultured within 5 days [
29]. The current study was conducted on the assumption that most organisms would be cultured within 5 days, and that these durations would limit the risk of isolating contaminant organisms. Fourth, molecular and modern techniques such as sonication of implants, polymerase chain reaction, alpha-defensin tests, and next-generation sequencing were not used in our study, therefore, the proportion of culture-negative PJI might be otherwise. Finally, there is a significant difference in follow-up period between the culture-positive group and culture-negative group. In addition, treatment methods were not standardized; this may have affected the treatment success rates. In fact, four culture-negative PJI cases were treated with one-stage revision arthroplasty, although it is considered a contraindication for culture-negative PJI [
4]. These cases were diagnosed as aseptic loosening before surgery, but pathological results were positive and they were diagnosed as PJI after surgery. These were treated as PJI after surgery, the infection was eradicated.