The specific outcome is to determine whether the colony count of Propionibacterium acnes, one of the commonest causes of shoulder infection and not eradicated by conventional forms of surgical preparatory solutions and antibiotics, in a shoulder surgical wound will be altered by the use of subdermal cefazolin.

Purpose The purpose of the study is to review if the subdermal injection of antibiotics intraoperatively reduces the bacterial load in the subcutaneous tissue during the course of an open shoulder operation.

Null Hypothesis There will be no change in the colony count of P.acnes following administration of subdermal antibiotics.

Justification. P. Acnes is the most common cause of shoulder infection. The incidence of post-operative shoulder joint infection varies from 4-18%. The P.acnes organism is a Gram positive, largely anaerobic but aerotolerant commensal of the deep skin follicles and standard skin preparation may inadequately penetrate the pilosebaceous follicles. It has also been shown in a study by Levy et al that 41.8% of patients had a positive biopsy or aspirate result for P.acnes prior to implantation of components. This study will determine whether intraoperative subdermal antibiotic administration reduces the bacterial load and in turn reduce the incidence of shoulder joint infection. Cefazolin has been chosen because cephalosporins and penicillins have the lowest minimum inhibitory concentration (MIC) against P.acnes. There is a lower incidence of allergy to cephalosporins than penicillins. This study has the potential to change clinical practice world wide.

Objectives The specific outcome is to compare the bacterial cell count on either side of a surgical wound and determine if subcutaneous antibiotics on a single side reduces subcutaneous cell counts over various time courses throughout the operation.

Research Method. Targeted recruitment is 20 patients in this pilot study. This figure has been arrived at after determining the power of the study, significance levels and the anticipated drop in P.acnes count colonisation after the use of local antibiotics. It is also a number that can be achieved rapidly, enabling us to assess whether there is an observed difference between the groups and thereby gauge the feasibility of a larger trial.

The patients will receive the standard of care pre-operative antibiotics 30 minutes prior to incision and surgical intervention. The skin will also be prepped using a chlorhexidine alcohol based solution as is standard at the investigator's institution. The investigators will divide the normal surgical incision into three even sections on either side of the incision, thus forming six segments. One side of the wound will be injected with three aliquots of 3 mLs using a concentration 100mg/mL of cefazolin. This equates to 900mg of subcutaneous cefazolin antibiotic. There are no specific efficacy criteria for agents in this setting that have been standardised and approved by an official oversight agency. We have selected this quantity based upon safe and wide-spread use of this concentration in a systemic manner as well as being mindful of the readily available concentrations. There is no known study describing any harm on introduction of antibiotics into the soft tissue, e.g. vancomycin used topically in prosthetic joints as a prophylaxis against infection, and the widespread use of intra-muscular (IM) penicillin products. Subcutaneous cefazolin has been demonstrated as safe in a porcine model while delivering higher cutaneous concentrations than with comparative systemic administration. Subcutaneous ceftriaxone is approved for use in France and is of increasing use for geriatric populations as well as by infectious disease practitioners. There is published evidence of safe administration of cefazolin in tumescent anesthesia antibiotic delivery mixed with lidocaine, epinephrine, and sodium bicarbonate.

Standardized samples from all six sections will be sampled using a disposable 6mm Mitex biopsy punch before inoculation and this process will be repeated at the 60 minute mark. The samples will be sent to the microbiology lab where colony count will be obtained and long term cultures i.e. 3 weeks will be performed.

It is routine in surgery to inject subcutaneous substances and not infrequently biopsies are taken. The only potential side effect is allergy to cephalosporins, which would have been determined preoperatively and is part of the exclusion criteria. There will be no expected increased incidence of infection or bleeding.

Collection, Handling and Storage Specimens will be labeled with appropriate patient identifiers and clearly labeled as to the site of collection, and sent to the laboratory immediately for processing as soon as possible. Tissue biopsies will be sent either in anaerobic transport media or in a sterile container with a piece of sterile gauze moistened with non-bacteriostatic saline, or if piece is large enough, just a few drops of non-bacteriostatic in the container to prevent the tissue from desiccation.

Tissue Processing All specimen manipulations will be carried out in the Biological Safety Cabinet and processed as soon as possible. Tissues will be inoculated to media to maximize the recovery of Propionibacterium acnes which will include a 5% sheep's blood agar and chocolate agar plates and thioglycollate (THIO) broth incubated under 5% carbon dioxide (CO2) at 35 Celsius. Brucella and Columbia Naladixic acid (CNA) agars will be incubated in an anaerobic conditions at 35 Celsius.

With sterile forceps, the tissue will be transferred to a sterile Petri plate. Using a disposable sterile scalpel the tissue will be chopped into enough small pieces to inoculate all the media required. Several pieces of tissue will be added to THIO broth. The remaining media will be inoculated if order of least selective to the most selective: the blood agar plate, chocolate agar plate, Brucella and CNA agar.

Lastly "touch preps" for the gram smears will be performed by gently touching a fresh piece of tissue onto the microscope slides. This piece of tissue will be discarded as it now is potentially contaminated.

Tissue Investigation

All media will be incubated for 5 days and read daily. The broth will be held for 21 days and examined at 7, 14 and 21 days after the initial 5 day reads. There is no need to follow the patients up for 2 years as we are only looking to determine change in intra operative colony counts which can be determined within 21 days and may be linked to a reduction in infection.

All bacterial growth will be investigated per usual laboratory procedures for identification and antimicrobial susceptibility testing.

Results will be reported to the study investigators in a timely fashion

Statistical Analysis. Samples will be compared with either side of the wound over the respective time periods. Count averages will be obtained and analysised using a paired t-test.

Inclusion Criteria:

• age greater than or equal to 18
• patients requiring deltopectoral approach and opening of joint capsule for primary shoulder surgery

Exclusion Criteria:

• allergy to cefazolin or cephalosporins
• tattoos over operative site
• previous surgical site infections in either shoulder
• revision surgical procedure
• immunocompromised patients