Total joint replacement is a common clinical practice for patients suffering from disabling arthritis, since it provides significant pain relief and functional recovering. Nevertheless, its outcome is compromised by complications such as periprosthetic joint infection (PJI), which is reported to occur in 1 to 4% of primary total knee arthroplasties (TKA), and approximately 1% of primary total hip replacements (THR). Despite all efforts to restrain PJI, its prevalence may reach even higher proportions if patients undergo a resection arthroplasty or irrigation and débridement for infected prosthesis. That said, timely diagnosis and early isolation of the infected microorganism is utterly important, if proper care is to be delivered.

The gold standard for the diagnosis of PJI is the isolation of a microorganism from the intraoperative cultures, combined with the sonication from retrieved joint implants1. This technique applies sound energy to agitate and disrupt biofilm, dislodging adherent bacterias to the bone cement, which has been proved to be a more sensitive method than conventional intraoperative cultures. False-negative percentages were reported to be 15% in patients who did not receive extended antibiotic prophylaxis and 60% if extended antibiotic therapy was administered.

Regardless of an adequate clinical, radiographic and surgical suspicion confirming PJI, an organism is not always successfully isolated from the intraoperative cultures, which increases false negatives results. This fact has been trying to be explained by several authors, some of which postulate that antibiotic prophylaxis could interfere with the isolation of the microorganism from the intraoperative cultures. As a result, and acting accordingly to this hypothesis, preoperative antibiotics are often withheld until intraoperative cultures are obtained, hoping that tissues are not loaded with antibiotics. Nevertheless, one should be aware of the adverse consequences of this practice that may result in systemic dissemination of infection.

Moreover, Ghanem and Stephen recently concluded that antibiotic prophylaxis does not interfere with the isolation of the microorganism from intraoperative cultures, despite being studies that lack statistical power.

Therefore, it is clear that reported studies in this field support both preoperative antibiotic prophylaxis administration, as well its withdrawal, until intraoperative cultures are obtained.

This decision in the department study depends exclusively on the treating surgeon judgment. In fact, 48% of all patients admitted at the study hospital with PJI receive preoperative antibiotic prophylaxis, which could be related to higher false-negative intraoperative culture and sonication results. Thus, the investigators add substances with chelation properties to hemoculture containers and then inoculate sonication samples. This practice offsets antibiotic interference with intraoperative cultures and has proved to enhance microorganism detection rates.

That said, and given the lack of scientific evidence about this clinical practice the investigators are willing to engage a prospective randomized double-blind clinical trial, that will allow us to determine whether intraoperative cultures and sonication samples are affected by antibiotic prophylaxis.

AIM AND RATIONALE:

Hypothesis:

H0: Preoperative antibiotic prophylaxis in patients with PJI interferes with the isolation of the microorganism from the intraoperative cultures and sonication samples, resulting in higher false-negative results.

H1: Preoperative antibiotic prophylaxis in patients with PJI does not interfere with the isolation of the microorganism from the intraoperative cultures and sonication samples.

Goals:

Main Goal:

- Determine preoperative antibiotic prophylaxis outcome on cultures obtained intraoperatively.

Secondary Goals:

• Ascertain if organisms isolated from preoperative joint aspirate correspond to those isolated from intraoperative cultures;
• Verify if the addition of substances with chelation properties to hemoculture containers filled with sonication samples affects antibiotic prophylaxis outcome;
• Make a descriptive analysis and revision of microorganisms implicated in PJIs.

PATIENTS AND METHODS:

Prospective, Randomized, Double-blind Clinical trial (Level I of Evidence):

Group I (study group): antibiotic prophylaxis + empiric antibiotic treatment once obtained intraoperative culture results.

Group II (control group): Empiric antibiotic treatment once obtained intraoperative culture results.

At the present institution, the investigators routinely administer intravenous prophylactic antibiotic, such as first-generation cephalosporins (2g cefazolin), or Glycopeptide derivatives (1g Vancomycin) if penicillin allergies are reported. This antibiotic administration schedule is initiated 30-60 minutes, or 60 minutes before surgery, respectively.

Empiric antibiotic treatment (vancomycin 1g/12h + ceftazidim 2g/8h) will be administered if aspiration cultures results are negative, or the infecting microorganism is unknown. Contrariwise, if the infecting microorganism is isolated and its antibiogram known, targeted antibiotic therapy will be used accordingly.

Patients were assigned to each group using a computer generated random selection scheme from Statistic Assessment Department. Both doctors and patients have been blinded for this information to ensure there are no differences in the way information is assessed or managed, in order to minimize bias. The study and placebo solutions will be prepared by the nurse on duty according to the anesthesiologist recommendations.

The surgical techniques performed vary according to the type of infection. In chronic infections the investigators will proceed to remove the infected prosthesis and implant the cement spacer, whereas in acute infections the adopted technique will be irrigation and débridment with liner exchange and retention of components. Once obtained intraoperative cultures and sonication samples, antibiotic prophylaxis regimen will be initiated with vancomycin (1g/12h) and ceftazidim (2g/8h), both for acute and chronic infections. In case there is a positive synovial fluid aspiration result, the investigators will perform targeted antibiotic treatment according to antibiogram. Hence, patients assigned to the control group will be able to receive an adequate antibiotic coverage. Additionally, there are no guidelines or clinical recommendations about the use of antibiotic prophylaxis in first-stage exchange, or irrigation and débridment procedures.

The protocol and surgical techniques performed, as well as the prosthetic models and cement spacers are the same the investigators use in the Department of Orthopaedic surgery and Traumatology of the study center, which have already been approved by a proper board review. That said, his clinical trial will not suppose any modification to the current clinical practice, neither to the microbiological techniques and methods applied.

Data collection:

• Preoperative data: gender, age, weight, Body Mass Index, comorbidities, type of prosthesis, mean time to diagnosis, preoperative joint aspirate microbiologic analysis and blood cell count.
• Intraoperative data: Whether or not antibiotic has been used and type of antibiotic used.
• Postoperative data: Intraoperative culture results and microbiological technique employed.

The investigators will consider false-negatives those patients who do not meet the infection criteria referred above (see inclusion criteria) and those who present negative culture results.

The routine surgical follow-up is scheduled 4, 6, 12 and 24 weeks after surgical intervention, as recommended by the study hospital protocol.

STATISTICS:

Statistical analysis:

Firstly, the investigators will perform a descriptive analysis of the study population. Whereas categorical variables will be expressed as numbers and frequencies, parameters measured on a continuous scale will be represented by their mean and standard deviation, in case they follow a normal distribution. Otherwise, the investigators will employ quartiles and median values to describe them.

Secondly, a bivariate analysis comparing each parameter with the group patients were assigned to will be executed, using the Chi Square or Fisher test to describe categorical variables. As to qualitative parameters, a variance analysis will be done if the variables follow a normal distribution. Otherwise, Kruskal-Wallis and Man-Whitney test will be used.

Finally, the investigators will employ a relative risk method to evaluate negativization risk.

The investigators will consider statistically significant p values under 0.05 and use SPSS 18.0 (SPSS inc, Chicago- Illinois) software and Prism (6.01 version, Graphpad, La Jolla, CA) for graphical analyses.

Sample Size:

The investigators undertook a Chi Square difference test to determine the sample size, assuming a 15% false negative rate in the control group, a 60% false negative rate in the study group and a 10% rate of lost to follow-up. The test revealed that 14 patients were needed in each group to detect statistically significant differences, establishing an α error of 0.05 and a statistical power of 80%.

Inclusion Criteria: as defined previously

• All the patients diagnosed with PJI who meet Zimmerli criteria and undergo a first-stage exchange, or débridment and retention procedure, in case there is an acute infection. These criteria are: visible sinus tract or purulence surrounding the prosthesis; detection of an infecting microorganism in the synovial fluid aspiration samples, intraoperative tissue and/or fluid cultures; synovial white blood cell count and histopathology parameters. The investigators will also include an additional criterion, which is prosthetic loosening 2 years after primary arthroplasty implantation following recent Portillo findings.
• The investigators will include hip, knee and shoulder replacements.

Exclusion Criteria:

• Patients receiving any kind of antibiotic regimen during the previous week of surgery;
• Hemodynamically unstable patients in need of antibiotic therapy, previously to the surgery.