Periprosthetic Joint Infection in Shoulder Arthroplasty

Periprosthetic Joint Infection in Shoulder Arthroplasty

Peter Lapner MD FRCSC1, Jay Keener MD2 and Thomas Duquin MD3

1Division of Orthopaedic Surgery, Department of Surgery, University of Ottawa, Ottawa, ON, Canada

2Department of Orthopaedic Surgery, Washington University, St. Louis, MO, USA

3Department of Orthopaedic Surgery, State University of New York at Buffalo, Buffalo, NY, USA

Clinical scenario

  • A 55‐year‐old man with advanced glenohumeral arthrosis has failed nonoperative treatment. He is seeking pain relief and improved function and is considering shoulder arthroplasty.
  • Five years following the index anatomic total shoulder arthroplasty, he presents with three months of increasingly severe shoulder pain.
  • The pain occurs deep in the joint, is worse with movement, and is not associated with systemic symptoms including fevers, chills, or sweats. Preoperative serum indices reveal an erythrocyte sedimentation rate (ESR) of 6 and a C‐reactive protein (CRP) of 4. Plain films and computed tomography (CT) scan reveal significant lucencies around the glenoid component suggestive of loosening. The patient is seeking definitive management and pain relief.

Top three questions

  1. Are infection prevention strategies, including modifiable patient factors and perioperative interventions, effective in reducing periprosthetic joint infection (PJI) in patients who undergo shoulder arthroplasty procedures?
  2. In patients with possible PJI, do preoperative serum indices, aspiration, or imaging aid in establishing the diagnosis of infection compared with preoperative tissue culture?
  3. In patients with shoulder PJI, does a two‐stage revision result in lower reinfection rates compared with one‐stage revision?

Question 1: Are infection prevention strategies, including modifiable patient factors and perioperative interventions, effective in reducing periprosthetic joint infection (PJI) in patients who undergo shoulder arthroplasty procedures?


Infection continues to be a devastating complication following shoulder arthroplasty procedures. Preventative strategies, including optimization of patient modifiable risk factors and perioperative interventions, have varying degrees of success. Optimization of these preventative strategies is essential for reducing infection following shoulder arthroplasty procedures.

Clinical comment

Shoulder arthroplasty continues to show significant growth with the number of procedures performed annually projected to increase by 150% by 2020.1 The incidence of PJI following shoulder arthroplasty is between 0.4 and 2.9% in the literature.2,3 Although rare, infection following shoulder arthroplasty results in significant morbidity to the patient and cost to the health care system.4,5 Risk factors for PJI after shoulder arthroplasty include age, sex, medical co‐morbidities, inflammatory arthritis, corticosteroid use, duration of procedure, and blood transfusion. There has been significant attention placed on the development of preventative strategies to reduce the rates of PJI. Identifying effective methods for reduction of infection will result in significant benefit to patients and society. The optimization of patient modifiable risk factors, and the development of effective perioperative interventions that reduce contamination are both important aspects of infection prevention. Developing effective prevention strategies in shoulder arthroplasty is challenging, due to the low incidence of PJI and the prevalence of infection with low virulent organisms.

Available literature and quality of the evidence

There are many studies evaluating different infection prevention strategies in shoulder surgery. Several modifiable patient factors including blood glucose control, obesity, smoking, and substance abuse are commonly thought to influence the risk of PJI in shoulder arthroplasty. Several cohort studies have evaluated the influence of these factors on patient outcomes and complications. There are no level I randomized control studies regarding modifiable risk factors. There is limited evidence that optimization of patient modifiable risk factors influences the risk of PJI in shoulder arthroplasty.


Modifiable risk factors

Obesity The best evidence in this area is level II. Richards et al. did not observe any difference in deep infection rate based on body mass index.6

Blood glucose control The evidence for glucose control is level III. The influence of diabetes and glycemic control has had mixed results in the literature. In a large retrospective database study of patients who underwent joint replacement surgery, Marchant et al. demonstrated increased risk of wound infection in patients with uncontrolled diabetes (odds ratio = 2.28; 95% confidence interval [CI]: 1.36–3.81; p = 0.002).7 In a large retrospective study of lower extremity joint arthroplasty at a single institution, Iorio et al. confirmed the increased rate of infection in diabetic patients compared to nondiabetics (3.43% vs 0.87%) but found no association with HbA1c levels.8 In contrast, a large database study of total knee arthroplasty failed to identify an increased rate of infection in diabetic patients, or in poorly controlled diabetic patients, when compared to patients without diabetes.9

Smoking Smoking has been identified as a risk factor for multiple complications following open and arthroscopic shoulder procedures (level III).10,11 In a retrospective review of 1834 shoulder arthroplasty procedures, Hatta et al. found there was a deep infection rate of 4.7% at 10 years’ follow‐up for smokers versus 0.6% in nonsmokers(11).11 In this study smoking cessation was found to decrease the risk of infection but not to the level of a nonsmoker. The hazard ratio for deep infection decreased from 7.27 in smokers to 4.26 in patients who were classified as former smokers.

Substance abuse Alcohol abuse is a risk factor for complications following total joint arthroplasty.12 Evidence is level IV. There are no studies specifically regarding alcohol consumption in patients undergoing shoulder arthroplasty. The timing or efficacy of alcohol cessation prior to joint replacement surgery has not been reported in the literature. Despite limited evidence, surgeons should counsel patients regarding alcohol cessation prior to shoulder arthroplasty.

Intravenous drug abuse has been associated with unacceptably high rates of infection following joint replacement surgery. Two small retrospective series have demonstrated infection rates greater than 25%.13,14

Perioperative prevention strategies

Skin preparation Evidence for skin preparation is level I based on skin cultures. Saltzman et al. reported decreased rates of superficial skin culture after use of ChloraPrep compared with DuraPrep or povidone‐iodine solution.15 However, all three agents demonstrated limited efficacy against formerly Propionibacterium acnes (now known as Cutibacterium acnes) with positive cultures in 7–22% of patients after skin preparation.

Preoperative skin preparation In a level I study, the use of chlorhexidine wipes showed significant decrease in the culture rate of coagulase negative staphylococcus prior to skin preparation compared to standard soap and water washing. However, there was no significant decrease in the rate of P. acnes cultures.16 Dizay et al. evaluated the use of benzoyl peroxide with clindamycin gel application in a level II study and reported reduction of deep cultures at the time of arthroscopic surgery to less than 4%.17

Hair removal There are no studies specific to the orthopedic literature regarding clipping versus shaving, but several studies from the general surgical literature support this practice (level I).18,19 The removal of axillary hair does not have any influence on the bacterial burden of P. acnes following skin preparation with chlorhexidine.20

Antibiotics The efficacy of antibiotic prophylaxis in joint replacement surgery has been demonstrated with level III evidence.21 The exact timing of antibiotics has been debated. The current guidelines indicate that administration should be performed with in one hour prior to surgical incision. The available level III evidence indicates increased infection risk if antibiotics are administered >1 hour before, or after skin incision. Conflicting evidence exists regarding the administration between 0–30 minutes and 31–60 minutes prior to incision.4,22,23

Local antibiotic administration The only prospective randomized trial in the spine literature (level I) with this technique failed to demonstrate a significant decrease in infection rate compared to standard intravenous antibiotic prophylaxis.24

Intraoperative irrigation A prospective randomized control trial of hemiarthroplasty for hip fractures demonstrated a decreased incidence of PJI with the use of pulsatile lavage compared to bulb syringe irrigation.25

Betadine lavage A retrospective study (level III) of total knee and hip arthroplasty demonstrated a decrease in the 90‐day infection rate with the addition of a dilute betadine lavage at the time of closure (0.97% vs 0.15%).26

Resolution of clinical scenario

Antibiotic prophylaxis with a first‐generation cephalosporin administered within one hour of the surgical incision has demonstrated efficacy. Topical skin treatments prior to admission and the use of a chlorhexidine‐based solution at the time of surgery reduces positive culture rates, but the role of either intervention in reducing PJI is unproven. Local use of antibiotics at the time of routine primary shoulder arthroplasty for the reduction of PJI is not supported. Intraoperative irrigation and betadine lavage use are both supported.

Question 2: In patients with possible PJI, do preoperative serum indices, aspiration, or imaging aid in establishing the diagnosis of infection compared with preoperative tissue culture?


P. acnes, a gram‐positive anaerobic bacteria, is the most common infecting organism in periprosthetic shoulder infection (PPSI).2732 It is very indolent and slow‐growing, and does not elicit an inflammatory reaction or clinical features of typical of infection. Establishing the diagnosis of infection with a painful shoulder after arthroplasty is extremely difficult.

Clinical comment

Establishing the diagnosis of infection with a painful shoulder after arthroplasty is extremely difficult and other than arthroscopic tissue biopsy, few reliable preoperative tests exists. This may lead to highly invasive operations in order to treat shoulder infections that may or may not be present. The current lack of a reliable diagnostic tool for infection may lead to delays in diagnosis, additional operations that could have been avoided, or inappropriate surgery in the presence of an undiagnosed infection.

Available literature and quality of the evidence

Nov 28, 2021 | Posted by in ORTHOPEDIC | Comments Off on Periprosthetic Joint Infection in Shoulder Arthroplasty
Premium Wordpress Themes by UFO Themes