Treatment of bone infection is a difficult procedure. At present, the overall treatment strategy includes conservative treatment, palliative debridement, en bloc resection, and amputation, but there is a lack of uniform treatment standards. Surgery is the usual method chosen by clinicians, but not all cases require surgical debridement , and many infections can be effectively treated with oral therapy [8-10]. Hamed  reported that intravenous antibiotics (six weeks) and oral suppression therapy were administered for cases of PJI infection, with reliable clinical efficacy. Research has shown that drug therapy alone has an average success rate of 71.1%, while surgical treatment alone has an average success rate of 88.9% . Systemic intravenous antibiotics after debridement can achieve a rapid reduction in the bacterial load at the site of infection . Therefore, clinicians are more inclined to use surgical treatment combined with antibiotic treatment to stop the infection.
According to previous studies, the optimal treatment of acute haematogenous osteomyelitis involves intravenous antibiotics for several weeks and then oral antibiotics until the symptoms and signs are alleviated [14, 15]. However, for chronic bone infection, debridement is often the first choice. Although research has shown that oral and parenteral therapies achieve similar success rates , clinicians are accustomed to considering the use of intravenous antibiotics [16, 17], followed by oral antibiotics for several weeks, during hospitalizations. It has been reported that the use of intravenous antibiotics for less than one week has no significant effect on prognosis [18-20]. However, only limited data comparing intravenous antibiotics and oral antibiotics or concerning proper treatment durations are available. We applied debridement plus systemic antibiotic treatment. Most patients (80.6%) were treated with intravenous antibiotics for only two weeks, although long-term antibiotic use has achieved good results in previous reports. Our results showed that the recurrence rate of the IV group was not significantly different from that of the other two groups, and the overall infection control rate was 83.6%. The infection control rate is comparable to that of previous reports, providing support for the treatment of these patients with short-term systemic antibiotic regimens following debridement.
Multiple and refractory bacterial infections are common in bone infections, especially chronic infections, so combined antibiotic treatments are commonly applied in the clinic. The most common oral combination drug is rifampicin, which has a bactericidal effect on multiple gram-positive and gram-negative bacteria . Rifampicin can achieve high intracellular levels and is capable of penetrating bacterial biofilms [21, 22]. However, it is not advisable to only apply rifampicin for anti-infection treatment because resistance can quickly develop . Thus, rifampicin should be started after the bacterial load is reduced by surgery and when the wound is dry [23, 24]. In this study, 307 (48.0%) patients were infected with Staphylococcus aureus and 56 (8.8%) with Pseudomonas aeruginosa, accounting for more than 50% of the isolated bacteria, so levofloxacin was applied more often in our clinic. Levofloxacin alone was unable to eradicate methicillin-susceptible S. aureus [25, 26]. The classic antibiotic combination for bone infections caused by Staphylococcus aureus and P. aeruginosa is levofloxacin plus rifampicin.
It is difficult to assess how long it will take for an infection to clear following the treatment of bone infection. Urania Rappo  evaluated the efficacy of infection after six weeks of treatment. Daver NG  indicated that the infection is apparently cured if it does not recur within six months. We evaluated infection control at six months after the operation. Although this timing may miss some cases of recurrence, our statistical analysis shows that recurrence occurs within two months in most cases.
Although rifampicin is rarely used alone in the treatment of bone infections, when this drug used in combination with other antibiotics, rifampicin can increase the treatment outcomes of other antibiotics. Rifampicin diffuses very well within bone tissue and bacterial biofilms  and is a valuable treatment option. However, rifampicin often elicits a hepatotoxic response and leads to gastroenteropathy. Our results indicated that the rate of abnormal ALT levels in the rifampicin group was significantly higher than that in the other groups, suggesting that postoperative combinations including rifampicin may lead to liver damage. In addition, Sahoko  reported that rifampicin can cause acute renal damage. Our results show that the overall complication rate was acceptable in the three groups. The recurrence rate of the IV group was not significantly different from that of the other two groups. Moreover, the oral group and rifampicin group had higher ALT levels and proteinuria rates, and the abnormal levels were probably caused by the additional oral therapies. Although the proteinuria rate in the rifampicin group was the highest, the rate of abnormal creatinine levels was not significantly different, suggesting that the additional rifampicin therapies may cause early or mild renal damage.
There are many strengths in this article, including the large sample size and its use of standard techniques. However, there are several drawbacks. First, although our results showed that there was no significant difference in the general data, including the type of bacterial infection and fixation method, the heterogeneity within the cohort may have affected the results. Second, the abnormal ALT and proteinuria levels suggested possible liver and renal damage, but the extent of the damage was not determined. Third, recurrence is common in bone infection, and the long-term recurrence rate for patients (after the six-month follow-up) is unknown.