Antibiotic prophylaxis in primary and revision shoulder replacement: a systematic review

Given the high success rate, total shoulder arthroplasty has become an effective surgical procedure for glenohumeral osteoarthritis. However, periprosthetic joint infection (PJI) still remains a devastating complication that negatively affects the overall outcome [1‐6]. Furthermore, the PJI is not only a catastrophic clinical failure, but it also represents an important burden for the health care system that requires long therapeutic support and expensive revision interventions [7].

The most common bacteria responsible for most PJIs are Propionibacterium acnes (P. acnes), Staphylococcus aureus, Staphylococcus epidermidis and coagulase-negative Staphylococcus [8, 9]. In particular, the P. acnes is a gram-positive micro-organism present in the dermis and sebaceous glands of most individuals [6, 10]. In the axilla area, there are numerous sebaceous glands which, by secreting the sebum that protects and lubricates the skin, also offer the opportunity for P. acnes to grow and contaminate the shoulder area [11].

PJIs after total shoulder arthroplasty, especially in patients with P. acnes, are more challenging than those occurring in other joints [12]. In particular, P. acnes is characterized by low virulence and, therefore, the laboratory tests normally used for early diagnosis (C reactive protein and red blood cells, white blood cells, erythrocyte sedimentation rate) appear normal even though the infection is in progress. The atypical characteristics of this bacterial strain explain why the diagnosis of infection is often only possible at the time of revision of the prosthesis when intraoperative cultures are performed.

Even though the rate of infections in patients undergoing total shoulder arthroplasty is increasing, effective tests for diagnosis have not yet been discovered [13]. Therefore, in order to reduce contamination of PJI preventive measures are to be improved, this to guarantee patients health, safety and to optimize forecasts prior to surgical intervention. However, the precautions taken at the time of surgery, such as intravenous antibiotics and standard skin preparation solution, are not yet sufficiently adequate to ensure sterilization of the incision area [11, 14]. Therefore, the surgeon should avoid contact with the skin to minimize the risk of infection [15‐17].

This systematic review aims to evaluate the suitability of antimicrobial prophylaxis in prevention of PJI in shoulder replacement and to provide health workers with the best suggestion for the use of antimicrobial agents based on currently available clinical evidence.

This systematic review was performed in accordance with the Preferred Reported Items of Systematic Review and Meta-analysis Statement (PRISMA) [18].

The articles selection process is illustrated in Fig. 1.

P. Acnes are microorganisms that populate the shoulder area where they found their natural site not only in superficial tissues but also in deep ones [15]. In the pre-operative period, 6 culture swabs were collected to be treated with calcium arginate. Some swabs were made on the surface at a maximum of 15 cm above the deltopectoral incision, others performed more deeply in the choracoid area. Three swabs were used to rule out pre-contamination and confirm the sterility of the operating room and analysis laboratories [20]. Moreover, a correlation between the male gender and the presence of P. Acnes has been demonstrated in 30 consecutive series of 30 patients undergoing primary shoulder arthroplasty. In fact, the hair of male gender in the shoulder area are known as important factors in colonization. The result expressed by the analyzed swabs showed that the male gender with the presence of hair is significantly correlated to the positive presence of P. acnes in the superficial area. Contrary in the female gender, there are no superficial or profound positives in P. acnes cultures [20].

The reduction of the surgical site infection rate is possible through the standardized administration of prophylactic antimicrobials, particularly in arthroplasty surgery. Due to its broad-spectrum, the use of cephalosporin is an unscientific approach to avoid shoulder PJI [24]. Furthermore, penicillin, known as very effective against P. Acnes, has shown good results as a preoperative antibiotic, but it is still not completely used [26]. Currently, the combination of an antimicrobial beta-lactam and cefazolin is considered the most appropriate preoperative prophylaxis due to their broad coverage spectrum, especially in shoulder arthroplasty. Cefazolin is active against a large number of infective organisms, skin flora and aerobic gram-positive [11]. Cefazolin is also active against bacteria as it is water-soluble and active bactericide capable of inhibiting cell wall biosynthesis bringing to bacterial lysis [11]. Furthermore, due to its good pharmacokinetics, it is able to quickly reach the site at the time of the incision, demonstrating effective bone, synovial and muscular penetration. Further benefits of Cefazolin are low cost, good safety profile and long half-life after intravenous administration [27]. Instead of Cefazolin, which is still the preferred antimicrobial drug, Clindamycin is recommended by the American Academy of Orthopaedic in patients with beta-lactam allergy and in those with methicillin-resistant Staphylococcus aureus [11]. Clindamycin has a higher minimum of bacterial concentration and a minimum concentration of biofilm eradication. However, it offers an important action against aerobic bacteria but not often against gram-negative bacteria. Clindamycin is a semi-synthetic antibiotic, better absorbed and with a greater antibacterial activity against some pathogens. It is active against numerous gram-positive aerobes and some protozoa and is not always active against aerobic tram-negatives [11, 28].

This systematic review showed that the optimal perioperative antimicrobial regimen is controversial. Although the use of antibiotic prophylaxis with a single drug is recommended by clinical guidelines, the increase in antibiotic-resistant infections may suggest that the choice to increase prophylactic antibiotic regimens may not be correct, improving the risk of expanding bacterial resistance.

The efficacy of the combination of Cefazolin and Clindamycin was studied. These drugs lead to potential damage such as acute postoperative renal damage and Clostridium infection and, therefore, the risk-benefit ratio must be carefully evaluated before performing a large-scale operation [7]. However, the use of combined prophylaxis between Cefazolin and Clindamycin in shoulder arthroplasty is increasing and, therefore, further research is needed to assess long-term effects [7].

The purpose of the preoperative administration of antimicrobials is to allow adequate tissue concentrations capable of eliminating the organisms that could occur before surgery. Therefore, the timing of administration and their dosage must be carefully studied to improve the effectiveness of prophylaxis. Single-shot antibiotic prophylaxis is normally sufficient to prevent the onset of bacteria but is not always capable of controlling P. Acnes [19]. No major reduction of P. Acnes was found on the surface and deep of the surgical wound layer after a careful prophylactic antiseptic protocol [20]. The effective treatment of the infection should include both medical and surgical therapy [21, 22]. Antibiotic treatment may not eliminate P. Acnes as it has proven insufficient in eradicating it [23]. The administration of antibiotic prophylaxis does not exclude contamination from P. Acnes, sometimes not administering antibiotics can guarantee a low rate of onset of P. Acnes [25]. As an increase in PJI has been reported when antibiotics are administered 60 min before the incision, current American Academy of Orthopedic Surgeons, Center for Disease Control and Surgical Care Improvement guidelines recommend the administration of antibiotic prophylaxis within 1 h of the surgical procedure; while the European guidelines report the administration of a single dose within 30 min of the incision [11]. In case of significant blood loss or prolonged surgery, additional doses of antibiotic will be required. In these cases, cefazolin should be repeated for 2–5 h during surgery, while clindamycin should be repeated every 3–6 h during surgery unless the patient shows altered renal parameters. Furthermore, the duration of administration of the antibiotic should not exceed 24 h after surgery since no benefit has been demonstrated in extending antimicrobial administration beyond 1 day after surgery [29].

For the prevention of PJI, the administration of preoperative antibiotic prophylaxis seems to be more effective than normal sterilization to minimize bacterial contamination, in particular from Propionibacterium acnes, before the incision [24]. Although guidelines are available for the administration of antibiotic prophylaxis to prevent PJI in shoulder replacement surgery, they are based on a limited number of randomized controlled trials that identify the efficacy of antimicrobial prophylaxis. The American Society of Health-System Pharmacists has demonstrated the effectiveness of cephalosporin in reducing PJI in surgical procedures [29]. The recommended dose of cefazolin in adults is 2 g, while for clindamycin it is 900 mg. The pharmacological characteristics of this antimicrobial class can be modified by the patient’s body weight, therefore its dosage must be specific for the patient [30].

This systematic review has a limitation represented by the small sample size of the included studies. Therefore, it is essential to define the concept of this contamination for the use of the necessary sterilization techniques to be used in the pre-operative phase [11]. Even though the differences in antibiotic prophylaxis between primary implantation and revision replacement may have considerable clinical relevance, it was not possible to perform an analysis since none of the seven studies included reported these data. Further researches are warranted to define the potential role of different antibiotic prophylaxis between primary and revision shoulder replacement.

Further research is needed to find more effective techniques for the prevention of PJIs, to define the preparation of the surgical site in order to minimize the possibility of contamination and identify adequate antibiotic prophylaxis taking into account the risks and benefits of individual drugs.