ABSTRACT
Introduction:
Prosthetic joint infection is a serious complication after primary shoulder arthroplasty, yet no standardized infection prevention protocol exists. This study evaluated current trends in infection prevention among members of the American Shoulder and Elbow Surgeons.
Methods:
A 25-item cross-sectional survey was developed to capture commonly used preoperative, intraoperative, and postoperative infection prevention strategies in primary shoulder arthroplasty. The finalized instrument was administered electronically through REDCap to all active American Shoulder and Elbow Surgeons members in November 2024, with 1 reminder sent 2 weeks later. Anonymized responses were categorized by years in practice (<5, 5–10, 10–20, and >20 years). Statistical analyses assessed variability across experience groups.
Results:
Of 229 responses, 224 were complete and included for analysis. Most surgeons reported using hemoglobin A1c cutoffs (77%), avoiding corticosteroid injections within 3 months of surgery (86%), performing preoperative skin cleansing (78%), using chlorhexidine for preparation (94%), switching scalpel blades (80%) or using electrocautery (79%) after skin incision, and administering perioperative IV antibiotics (76%). Surgeons with <5 years of experience were more likely to use hydrogen peroxide or alcohol-based skin preparation, vancomycin powder, topical skin glue, silver-impregnated dressings, and apply stricter hemoglobin A1c cutoffs, reaching statistical significance in all cases (P < .05). A body mass index cutoff was used by 25% of respondents, with a mean threshold of 42 kg/m2.
Conclusion:
Substantial variation exists in infection prevention strategies for primary shoulder arthroplasty (SA) across experience levels, highlighting the need for standardized protocols.
Level of Evidence:
IV, epidemiological study.
Keywords: prosthetic joint infection, shoulder arthroplasty, ASES, infection prevention, survey, current practices
Introduction
Prosthetic joint infection (PJI) is a devastating complication after shoulder arthroplasty with significant implications for patient outcomes and healthcare system costs.1 In contrast to hip and knee arthroplasty, the more indolent presentation of PJI of the shoulder poses a unique diagnostic challenge because of the predominance of low-virulence organisms such as Cutibacterium acnes (C. acnes).2,3 The annual incidence of shoulder PJI continues to rise, with a 2023 study projecting a 176% increase from 2023 to 2030, highlighting the need to identify effective infection prevention strategies.4
Various measures, including patient optimization, preoperative skin decolonization, antibiotic prophylaxis, antimicrobial surgical materials, and postoperative wound management, have been explored in hopes of reducing the risk of PJI.5–8 Despite extensive efforts, no established standard protocol for infection prevention in primary shoulder arthroplasty (PSA) currently exists, leading to variability in utilization among practicing surgeons.9–11 As such, identifying practice patterns may provide critical insight into prevailing trends, areas of consensus, and opportunities to establish standardized, evidence-based guidelines.
The purpose of this study was to assess current trends in infection prevention strategies for PSA among American Shoulder Elbow Surgeon (ASES) members. Specifically, we aimed to assess utilization patterns of several preoperative, intraoperative, and postoperative protocols and to determine whether these practices differ based on surgeon experience. We hypothesized that prevention methods would vary among surgeons and that variation would correlate with surgeon experience.
Methods
A survey was developed through an iterative, consensus-driven process among the authors. Commonly used and previously reported infection prevention strategies for PSA were identified using an extensive literature review. The survey content was refined through 3 rounds of author review, during which all authors participated in proposing, modifying, and selecting questions. Items were voted on collectively in a nonblinded fashion.
The final survey consisted of 25 items and was administered using REDCap (Vanderbilt University, Nashville, TN). Each emailed invitation contained a unique survey link tied to a single-response ID, preventing multiple submissions from the same recipient. The survey was distributed electronically to all members of the ASES in November 2024. A follow-up email was sent 2 weeks after the initial invitation. The survey link was distributed twice, 2 weeks apart, and remained open for approximately 1 month.
The survey included questions regarding respondents' years of experience and their practices related to infection prevention. Preoperative strategies included the use and type of skin preparation, nasal methicillin-resistant Staphylococcus aureus screening and decolonization, axillary hair removal, criteria for surgical eligibility such as hemoglobin A1c (HbA1c) and body mass index (BMI) thresholds, and corticosteroid injection timing before surgery. Intraoperative questions assessed skin antisepsis agents, surgical helmet use, knife blade changes, electrocautery use, drape application, irrigation techniques, use of antibiotic-loaded cement, and vancomycin powder application. Postoperative questions covered antibiotic type and duration, use of topical surgical glue, and silver-impregnated dressings.
Responses were tallied for the overall cohort and classified into 4 groups based on years of experience: <5 years, 5–10 years, 10–20 years, and >20 years. For questions with an “other” option, free-text responses mentioned by more than 1 participant were grouped into a new answer category for analysis. Continuous variables, specifically HbA1c and BMI, were sorted into categorical bins for statistical analysis.
Statistical analysis
Descriptive statistics were conducted for the overall sample. Counts and proportions were reported for categorical variables, while means and standard deviations (or medians and interquartile ranges, depending on normality) were reported for continuous variables. Bivariate analyses were conducted to examine the association between years in practice and surgical practices. A chi-square test of independence was used for categorical comparisons, while an independent samples t test or Mann-Whitney U test (based on normality) was applied for continuous variables.
For statistically significant variables from the bivariate analyses, binary logistic and linear regression analyses were performed to further assess the relationship between years in practice and surgical practices. All regressions used <5 years of practice experience as a reference. For binary logistic regression models, odds ratios and 95% confidence intervals were reported. For linear regression models, regression coefficients and P-values were presented.
All statistical tests were 2-tailed, with a significance level set at P < .05. Analyses were performed using RStudio (version 4.4.2) for Windows.
Results
Participant demographics
Eighteen percent (224/1412) of ASES members completed the survey. The cohort consisted of 39 (17.4%) surgeons with <5 years of experience, 32 (14.3%) with 5–10 years, 74 (33.0%) with 10–20 years, and 79 (35.3%) with >20 years of experience (Figure 1).
Figure 1.
Distribution of respondents by years of practice. The largest group of respondents reported more than 20 years of practice experience (35.3%), followed by 10–20 years (33.0%), <5 years (17.4%), and 5–10 years (14.3%).
Preoperative strategies
Patient-specific risk factors and preoperative optimization
Seventy-seven percent of respondents reported using a HbA1c cutoff when offering surgery to diabetic patients (Table 1). Most of the respondents reported using a HbA1c cutoff value between 8.0 and 8.4 (54%), and the mean reported value was 7.9 ± 0.6 (Figure 2). A significantly greater percentage of surgeons in practice for less than 5 years reported using a HbA1c cutoff value compared with all other experience groups (5–10 years, 10–20 years, and >20 years; P = .001). Surgeons in practice for greater than 20 years used the lowest mean cutoff value of 7.7 (P = .04).
Table 1.
Patient-specific risk factors
| Questionnaire item | Overall responses (n = 224) | Years in practice (n = 224) | P-value | |||
| <5 years (n = 39) | 5–10 years (n = 32) | 10–20 years (n = 74) | >20 years (n = 79) | |||
| HbA1c cutoff used (n, %) | 173 (77) | 38 (97) | 26 (81) | 58 (78) | 51 (65) | .001 |
| HbA1c cutoff value (mean ± SD)a | 7.9 ± 0.6 | 7.9 ± 0.4 | 8.1 ± 0.6 | 7.8 ± 0.5 | 7.7 ± 0.7 | .037 |
| HbA1c cutoff range used (n, %) | .027 | |||||
| <7.0 | 3 (2) | 0 (0) | 0 (0) | 1 (2) | 2 (4) | |
| 7.0–7.4 | 19 (11) | 2 (5) | 1 (4) | 4 (7) | 12 (25) | |
| 7.5–7.9 | 40 (24) | 8 (21) | 4 (15) | 18 (31) | 10 (21) | |
| 8.0–8.4 | 91 (54) | 25 (66) | 18 (69) | 29 (50) | 19 (40) | |
| 8.5–8.9 | 7 (4) | 1 (3) | 1 (4) | 4 (7) | 1 (2) | |
| 9.0–9.4 | 7 (4) | 2 (5) | 0 (0) | 2 (3) | 3 (6) | |
| ≥9.5 | 3 (2) | 0 (0) | 2 (8) | 0 (0) | 1 (2) | |
| BMI cutoff used (n, %) | 55 (25) | 13 (33) | 7 (21.9) | 18 (24) | 17 (22) | .544 |
| BMI cutoff value (mean ± SD)a | 41.8 ± 4.9 | 41.0 ± 3.8 | 42.5 ± 4.3 | 43.2 ± 5.9 | 40.9 ± 4.9 | .494 |
| BMI cutoff range used (n, %) | .76 | |||||
| <35 | 2 (4) | 1 (8) | 0 (0) | 0 (0) | 1 (6) | |
| 35–39 | 5 (9) | 1 (8) | 1 (14) | 1 (6) | 2 (12) | |
| 40–44 | 33 (61) | 8 (62) | 3 (43) | 12 (71) | 10 (59) | |
| 45–49 | 6 (11) | 2 (15) | 2 (29) | 0 (0) | 2 (12) | |
| ≥50 | 8 (15) | 1 (8) | 1 (14) | 4 (24) | 2 (12) | |
| Perform surgery within 3 months of CSI (n, %) | 31 (14) | 2 (5) | 5 (16) | 11 (15) | 13 (17) | .379 |
BMI, body mass index; CSI, corticosteroid injection; HbA1c, hemoglobin A1c; SD, standard deviation.
t test analysis performed.
Reported HbA1c and BMI cutoff values, stratified by years in practice. The most common HbA1c cutoff range was 8.0–8.4%, and the most common BMI cutoff range was 40–44.
Bolded values are statistically significant at P < 0.05.
Figure 2.
Distribution of HbA1c cutoff ranges reported by surgeons. The most common cutoff range was 8.0–8.4% (n = 91). HbA1c, hemoglobin A1c.
Twenty-five percent of respondents reported using a BMI cutoff (Figure 3). Among those who use a threshold, the mean BMI cutoff was 41.8 ± 4.9. The most common cutoff across all respondents was 40–44, selected by 61% of surgeons. No difference in reported BMI cutoff values was noted among surgeons of varying experience. Finally, 86% of surgeons reported that they would not perform surgery within 3 months of corticosteroid injection (Table 1).
Figure 3.
Distribution of BMI cutoff ranges reported by surgeons. The most common cutoff range was 40–44 (n = 33). BMI, body mass index.
Preoperative skin decolonization
A large majority of surgeons reported use of an at-home skin cleanse before surgery (Table 2), with the most commonly reported agent being chlorhexidine-based (50%). Fifty percent of respondents reported obtaining preoperative nasal swabs to identify methicillin-resistant S. aureus colonization and 53% proceed with a topical nasal solution to address a positive finding. No difference in preoperative skin cleanse, nasal swab use, or axillary hair shaving was noted among surgeons of varying experience.
Table 2.
Preoperative skin cleansing and decolonization
| Questionnaire item | Overall responses (n = 224) | Years in practice (n = 224) | ||||
| <5 years (n = 39) | 5–10 years (n = 32) | 10–20 years (n = 74) | >20 years (n = 79) | P-value | ||
| Use of home skin cleanser (n, %) | 175 (78) | 26 (67) | 24 (75) | 64 (87) | 61 (77) | .1 |
| Home skin cleanser type (n, %) | ||||||
| Benzoyl peroxide | 69 (31) | 12 (31) | 7 (22) | 27 (37) | 23 (29) | .489 |
| CHG | 111 (50) | 18 (46) | 17 (53) | 38 (51) | 38 (48) | .918 |
| Soap and water shower | 29 (13) | 7 (18) | 5 (16) | 5 (7) | 12 (15) | .268 |
| Alcohol | 1 (0) | 0 (0) | 0 (0) | 0 (0) | 1 (1) | .605 |
| Shave axillary hair (n, %) | 58 (26) | 11 (28) | 10 (31) | 14 (19) | 23 (29) | .409 |
| Preoperative MRSA nasal swab (n, %) | 111 (50) | 18 (46) | 15 (47) | 38 (51) | 40 (51) | .939 |
| Topical solution in nares if MRSA swab is positive (n, %) | 118 (53) | 18 (46) | 15 (47) | 44 (60) | 41 (53) | .531 |
CHG, chlorhexidine gluconate; MRSA, methicillin-resistant Staphylococcus aureus.
Reported use of home skin cleansers, shaving practices, and preoperative MRSA decolonization strategies stratified by years in practice. Most respondents reported use of a home skin cleanser (78%), with chlorhexidine-based products being the most common (50%). Half of respondents obtained a preoperative MRSA nasal swab, and more than half prescribed a topical nasal solution if the swab was positive.
Perioperative and intraoperative strategies
Antibiotic prophylaxis
In patients without a history of penicillin allergy, 89% of surgeons reported use of cefazolin for perioperative prophylaxis before PSA (Table 3). For patients with a history of anaphylactic reaction to penicillin, 57% of surgeons reported use of vancomycin and 19% administer clindamycin. In addition, 19% of surgeons use cephazolin despite a history of anaphylaxis. In patients reporting a history of nonanaphylactic allergy to cefazolin, 81% of surgeons administer cefazolin before incision while 9% use vancomycin and 5% clindamycin. No difference in antibiotic prophylaxis utilization was noted among surgeons of varying years of work experience, and all surveyed surgeons reported use of some form of perioperative antibiotic prophylaxis.
Table 3.
Preincision antibiotic prophylaxis stratified by history of reported allergy to penicillin.
| Reaction to penicillin | Overall response rate (n = 224) | Years in practice (n = 224) | P-value | |||
| <5 years (n = 39) | 5–10 years (n = 32) | 10–20 years (n = 74) | >20 years (n = 79) | |||
| No allergy (n, %) | .274 | |||||
| Cefazolin | 199 (89) | 37 (95) | 28 (88) | 67 (91) | 67 (85) | |
| Vancomycin | 9 (4) | 0 (0) | 0 (0) | 5 (7) | 4 (5) | |
| Clindamycin | 2 (1) | 0 (0) | 0 (0) | 0 (0) | 2 (3) | |
| Ceftriaxone/vancomycin | 6 (3) | 1 (3) | 1 (3) | 1 (1) | 3 (4) | |
| Cefazolin/clindamycin | 2 (1) | 0 (0) | 1 (3) | 0 (0) | 1 (1) | |
| Cefazolin/vancomycin | 4 (2) | 1 (3) | 2 (6) | 1 (1) | 0 (0) | |
| Other | 2 (1) | 0 (0) | 0 (0) | 0 (0) | 2 (3) | |
| Anaphylactic allergy (n, %) | .464 | |||||
| Cefazolin | 42 (19) | 10 (26) | 4 (13) | 18 (24) | 10 (13) | |
| Vancomycin | 127 (57) | 19 (49) | 19 (59) | 40 (54) | 49 (62) | |
| Clindamycin | 43 (19) | 9 (23) | 6 (19) | 13 (18) | 15 (19) | |
| Ceftriaxone/vancomycin | 1 (0) | 0 (0) | 1 (3) | 0 (0) | 0 (0) | |
| Clindamycin/vancomycin | 5 (2) | 0 (0) | 1 (3) | 2 (3) | 2 (3) | |
| Aztreonam/vancomycin | 2 (1) | 1 (3) | 1 (3) | 0 (0) | 0 (0) | |
| Other | 3 (1) | 0 (0) | 0 (0) | 1 (1) | 2 (3) | |
| None | 1 (0) | 0 (0) | 0 (0) | 0 (0) | 1 (1) | |
| Nonanaphylactic allergy (n, %) | .260 | |||||
| Cefazolin | 181 (81) | 34 (87) | 27 (84) | 63 (85) | 57 (72) | |
| Vancomycin | 20 (9) | 1 (3) | 1 (3) | 7 (10) | 11 (14) | |
| Clindamycin | 10 (5) | 2 (5) | 1 (3) | 2 (3) | 5 (6) | |
| Cefazolin/vancomycin | 3 (1) | 1 (3) | 2 (6) | 0 (0) | 0 (0) | |
| Ceftriaxone/vancomycin | 6 (3) | 1 (3) | 1 (3) | 1 (1) | 3 (4) | |
| Other | 2 (1) | 0 (0) | 0 (0) | 0 (0) | 2 (3) | |
| None | 2 (1) | 0 (0) | 0 (0) | 1 (1) | 1 (1) | |
Most respondents used cefazolin for patients without an allergy. Among those reporting an anaphylactic allergy, vancomycin was most frequently used.
Surgical preparation and additional practices
Chlorhexidine-based skin preparation was reportedly used by 94% of surgeons before surgery (Table 4). An additional 43% of surgeons use a peroxide-based agent and 34% an alcohol-based solution. Peroxide-based (P = .002) and alcohol-based (P = .01) solutions were used more commonly among surgeons in practice for less than 5 years than in surgeons with greater work experience.
Table 4.
Perioperative and intraoperative strategies
| Questionnaire item | Overall responses (n = 224) | Years in practice (n = 224) | P-value | |||
| <5 years (n = 39) | 5–10 years (n = 32) | 10–20 years (n = 74) | >20 years (n = 79) | |||
| Preincision skin prep type (n, %) | ||||||
| CHG-based | 210 (94) | 38 (97) | 31 (97) | 70 (95) | 71 (90) | .314 |
| Peroxide-based | 96 (43) | 25 (64) | 12 (38) | 36 (49) | 23 (29) | .002 |
| Iodine-based | 26 (1) | 4 (10) | 4 (13) | 5 (7) | 13 (17) | .308 |
| Alcohol-based | 76 (34) | 21 (54) | 8 (25) | 27 (37) | 20 (25) | .012 |
| Other | 1 (0) | 1 (3) | 0 (0) | 0 (0) | 0 (0) | .190 |
| Positive-pressure surgical helmet (n, %) | 104 (46) | 24 (62) | 14 (44) | 29 (39) | 37 (47) | .155 |
| Iodine-impregnated adhesive drape (n, %) | 188 (84) | 35 (90) | 27 (84) | 65 (88) | 61 (77) | .216 |
| Change knife blade after incision (n, %) | 180 (80) | 28 (72) | 27 (84) | 60 (81) | 65 (82) | .503 |
| Switch to electrocautery after incision (n, %) | 177 (79) | 38 (97) | 27 (84) | 62 (84) | 50 (64) | <.001 |
| Antibiotic-loaded cement (n, %) | 103 (46) | 12 (31) | 13 (41) | 39 (53) | 39 (49) | .12 |
| Intraoperative irrigation type (n, %) | ||||||
| Sterile saline | 154 (69) | 30 (77) | 24 (75) | 48 (65) | 52 (66) | .452 |
| Antibiotic-infused sterile saline | 33 (15) | 3 (8) | 5 (16) | 11 (15) | 14 (18) | .547 |
| Sterile dilute povidone iodine | 61 (27) | 11 (28) | 7 (22) | 22 (30) | 21 (27) | .866 |
| Hydrogen peroxide | 25 (11) | 8 (21) | 5 (16) | 7 (10) | 5 (6) | .104 |
| CHG-based solution | 45 (20) | 3 (8) | 6 (19) | 20 (27) | 16 (20) | .112 |
| Citric acid–based solution | 2 (1) | 1 (3) | 0 (0) | 1 (1) | 0 (0) | .492 |
| Other | 2 (1) | 1 (3) | 0 (0) | 1 (1) | 0 (0) | .492 |
| Topical vancomycin powder (n, %) | 137 (61) | 35 (90) | 26 (81) | 36 (49) | 40 (51) | <.001 |
| Topical surgical glue (n, %) | 103 (46) | 23 (59) | 22 (69) | 31 (42) | 27 (34) | .003 |
| Silver-impregnated bandage (n, %) | 81 (36) | 23 (59) | 15 (47) | 22 (30) | 21 (27) | .002 |
Bolded values are statistically significant at P < 0.05.
CHG, chlorhexidine gluconate.
Reported practices regarding skin preparation, intraoperative irrigation, use of antibiotic-loaded cement, surgical adjuncts, and topical antimicrobials. Significant differences by years in practice were noted for iodine-based preparation, switch to electrocautery, topical vancomycin powder, and silver-impregnated bandage use.
Intraoperative wound irrigation consisting of sterile saline was used among 69% of respondents and sterile dilute povidone iodine among 27% (Table 4). The remainder of used irrigation solutions can be found in Table 4. No difference in the rate of utilization of any of the listed irrigants was noted among surgeons by years of work experience.
Surgeons in practice for less than 5 years more commonly reported use of topical vancomycin powder (90%, P < .001), silver-impregnated bandages (59%, P = .002), and switching to electrocautery after making skin incision with a knife (97%, P < .001; Table 4). Surgical glue was most commonly used over the skin incision by surgeons in practice for 5–10 years (69%, P = .003). There was no difference in the use of antibiotic-loaded cement, iodine-impregnated adhesive drapes, positive-pressure helmet systems, or changing of the knife blade after incision between cohorts (Table 4).
Postoperative strategies
Postoperative antibiotic administration
After PSA, 57% of surgeons reported administering antibiotics for less than 24 hours and 24% did not routinely prescribe postoperative antibiotics (Table 5). Of those who reported use of postoperative antibiotics (n = 167), 80% used intravenous cefazolin. Fourteen percent of surgeons reported prescribing oral antibiotics, with doxycycline being the most commonly reported medication (68%). No difference in the duration or type of postoperative antibiotic used after PSA was noted among surgeons of varying work experience.
Table 5.
Postoperative antibiotic protocol
| Questionnaire item | Overall responses (n = 224) | Years in practice (n = 224) | P-value | |||
| <5 years (n = 39) | 5–10 years (n = 32) | 10–20 years (n = 74) | >20 years (n = 79) | |||
| Antibiotic duration (n, %) | .073 | |||||
| <24 hours | 127 (57) | 24 (62) | 22 (69) | 44 (60) | 37 (47) | |
| >24 hours | 43 (19) | 6 (15) | 2 (6) | 11 (14) | 24 (30) | |
| None | 54 (24) | 9 (23) | 8 (25) | 19 (26) | 18 (23) | |
| Antibiotic choice (n, %) | .458 | |||||
| IV cefazolin | 133 (59) | 29 (74) | 20 (63) | 43 (58) | 41 (52) | |
| IV vancomycin | 6 (3) | 0 (0) | 0 (0) | 3 (4) | 3 (4) | |
| IV clindamycin | 2 (1) | 0 (0) | 0 (0) | 0 (0) | 2 (3) | |
| IV ceftriaxone | 2 (1) | 0 (0) | 0 (0) | 0 (0) | 2 (3) | |
| PO antibiotics | 24 (11) | 2 (5) | 3 (9) | 8 (11) | 11 (14) | |
| None | 57 (25) | 8 (21) | 9 (28) | 20 (27) | 20 (25) | |
| Oral antibiotic if given (n, %) | .305 | |||||
| Cefadroxil | 2 (11) | 0 (0) | 0 (0) | 0 (0) | 2 (20) | |
| Doxycycline | 13 (68) | 1 (50) | 0 (0) | 5 (83) | 7 (70) | |
| Cephalexin | 4 (21) | 1 (50) | 1 (100) | 1 (17) | 1 (10) | |
IV, intravenous; PO, oral.
Duration of antibiotic administration, intravenous antibiotic choices, and use of oral antibiotics stratified by years in practice. Most respondents administered <24 hours of antibiotics, with cefazolin being the most common intravenous agent.
Statistical analyses of association
Binary logistic and linear regression analyses of variables found to be statistically significant through bivariate analyses can be found in Table 6. Logistic regression revealed that compared with surgeons in practice for less than 5 years, surgeons in practice for greater than 20 years were significantly less likely to use any of the analyzed variables including hydrogen peroxide or alcohol skin cleansing, electrocautery after skin incision, topical vancomycin powder, surgical glue over the incision, silver-impregnated bandages, and the use of a HbA1c cutoff value to indicate surgery. Linear regression analysis demonstrated that when using a HbA1c cutoff, surgeons in practice for greater than 20 years used a cutoff value that was 0.25 units lower than those in practice less than 5 years. Additional significant comparisons were noted for several variables when comparing surgeons in practice for 5–10 years and 10–20 years to those in practice for less than 5 years (Table 6).
Table 6.
Binary logistic regression assessing association between years in practice and infection prevention strategies
| Questionnaire item | Odds ratioa | 95% CIa | P-value |
| Hydrogen peroxide cleansing solution | |||
| 5–10 years | 0.34 | 0.12–0.87 | .027 |
| 10–20 years | 0.53 | 0.24–1.17 | .119 |
| >20 years | 0.23 | 0.10–0.51 | <.001 |
| Alcohol cleansing solution | |||
| 5–10 years | 0.27 | 0.10–0.77 | .016 |
| 10–20 years | 0.49 | 0.22–1.08 | .078 |
| >20 years | 0.29 | 0.13–0.65 | .003 |
| Electrocautery use | |||
| 5–10 years | 0.14 | 0.01–0.95 | .083 |
| 10–20 years | 0.14 | 0.01–0.73 | .060 |
| >20 years | 0.05 | 0.003–0.24 | .003 |
| Topical vancomycin powder use | |||
| 5–10 years | 0.50 | 0.12–1.91 | .312 |
| 10–20 years | 0.11 | 0.03–0.30 | <.001 |
| >20 years | 0.12 | 0.003–0.33 | <.001 |
| Topical surgical glue use | |||
| 5–10 years | 1.53 | 0.58–4.18 | .396 |
| 10–20 years | 0.50 | 0.23–1.10 | .086 |
| >20 years | 0.36 | 0.16–0.79 | .011 |
| Silver-impregnated bandage use | |||
| 5–10 years | 0.61 | 0.24–1.57 | .310 |
| 10–20 years | 0.29 | 0.13–0.65 | .003 |
| >20 years | 0.25 | 0.11–0.56 | .001 |
| Has HbA1c cutoff | |||
| 5–10 years | 0.11 | 0.01–0.72 | .050 |
| 10–20 years | 0.10 | 0.01–0.50 | .025 |
| >20 years | 0.05 | 0.003–0.24 | .003 |
| HbA1c cutoff of 8.0 | |||
| 5–10 years | 1.50 | 0.46–5.42 | .51 |
| 10–20 years | 0.54 | 0.22–1.31 | .18 |
| >20 years | 0.36 | 0.14–0.88 | .03 |
| HbA1c cutoffb | |||
| 5–10 years | 0.12 (0.14) | .400 | |
| 10–20 years | −0.11 (0.12) | .346 | |
| >20 years | −0.25 (0.12) | .041 |
Bolded values are statistically significant at P < 0.05.
CI, confidence interval; HbA1c, hemoglobin A1c.
<5 years of experience set as reference group.
Linear regression was performed for continuous variable as presented as estimate (standard error) format.
Surgeons in practice for >20 years were significantly less likely to use certain infection prevention strategies including hydrogen peroxide or alcohol skin cleansing, electrocautery after skin incision, topical vancomycin powder, surgical glue over the incision, silver-impregnated bandages, and the use of a hemoglobin A1c (HbA1c) cutoff value to indicate surgery.
Discussion
In this survey study, a strong consensus in several core infection prevention practices were identified. Strategies such as optimization of glycemic control, avoidance of recent steroid injections, use of chlorhexidine-based skin preparation, and use of cefazolin in nonallergic patients were used by most of the respondents, and are practices broadly consistent with American Academy of Orthopaedic Surgeons (AAOS) clinical practice guidelines.12 In addition, ASES surgeons similarly demonstrated high agreement about blade change or electrocautery after incision, use of iodine-impregnated drapes, and application of silver-impregnated dressings. By contrast, use of adjuvants such as topical antibiotics and postoperative antibiotic duration varied substantially, reflecting the absence of corresponding AAOS or ASES-endorsed guidance for PSA.
Experience-dependent differences were also observed, particularly regarding the use of preincision cleansing solutions, electrocautery, topical antibiotics, HbA1c cutoff utilization, and postoperative wound management. Taken together, these findings highlight areas of alignment with available national guidelines and practice heterogeneity, underscoring the need for standardized, shoulder-specific recommendations.
A key finding of this study is the variation in preincision skin cleansing solutions based on the number of years in practice. Surgeons with fewer than 5 years of experience were significantly more likely to use hydrogen peroxide–based and alcohol-based solutions than their counterparts who have been in practice longer. This trend may reflect the increasing adoption of novel antiseptic agents by younger surgeons, potentially influenced by emerging evidence or evolving guidelines emphasizing broader-spectrum antisepsis.10 Additional potential drivers include recent fellowship training environments where sequential skin preparation protocols may be more frequently used and greater exposure with industry-supported products preferentially marketed to early-career surgeons.
Recent data support the adjunctive use of hydrogen peroxide–based protocols to enhance antisepsis. Chalmers et al.5 demonstrated that adding 3% hydrogen peroxide to standard skin preparation significantly reduced positive bacterial cultures from the skin, dermis, and glenohumeral joint, suggesting improved efficacy in both superficial and deep tissue antisepsis. In addition, Mizels et al.13 reported fewer revision surgeries and C. acnes infections in patients receiving hydrogen peroxide as part of their preoperative skin preparation, although the results did not reach statistical significance. These findings align with the broader concept of colonization reduction and its association with lower rates of surgical site infections and PJI. Furthermore, they indicate that hydrogen peroxide shows promise as an adjunct to traditional skin antisepsis and warrants further investigation.
These results demonstrate a similar dichotomy in the use of certain intraoperative infection prevention measures. Surgeons in practice longer were less likely to switch to electrocautery after initial skin incision, apply topical vancomycin powder before wound closure, or use silver-impregnated dressings. The increased use of topical antibiotics and barrier-enhanced dressings by younger surgeons may be a response to recent literature suggesting potential benefits in reducing PJI rates in reverse shoulder arthroplasty.11 For example, silver-impregnated occlusive dressings have been shown to be a cost-effective strategy for minimizing infection risk, with economic analyses demonstrating their viability even if they prevent just 1 infection in 1,471 procedures.7 Recent studies have also demonstrated the potential efficacy of vancomycin powder in infection prevention. Miquel et al.14 found that a 10 mg dose of vancomycin powder prevented C. acnes colony growth within 48 hours in a mimetic shoulder joint model. Garofalo et al.15 reported no periprosthetic infections in a retrospective cohort of 422 total shoulder arthroplasties treated with 1 g of vancomycin powder embedded in a collagen sponge, compared with a 3.2% infection rate in the control group. Finally, electrocautery use after the initial skin incision has been associated with a significant reduction in C. acnes contamination at the incised dermis. A randomized controlled trial by Kim et al.6 found that although 31% of patients in the scalpel group had positive cultures from the incised dermis—most of which were C. acnes—none in the electrocautery group had positive cultures present (P < .001). Despite this initial antibacterial effect, C. acnes was still detected on surgical gloves and instruments in both groups, particularly in male patients, suggesting that bacterial inoculation may occur through alternative routes of contamination. These findings highlight the potential role of the aforementioned strategies in reducing dermal C. acnes contamination, although further research is needed to address ongoing intraoperative transmission risks.
An additional point of variance among respondents was the use of glycemic control thresholds when evaluating diabetic patients for PSA. Surgeons in practice for more than 10 years were significantly less likely to use a specific HbA1c cutoff for surgical eligibility. Among those who did, experienced surgeons tended to use lower cutoff values than their younger counterparts. This discrepancy may reflect differing levels of confidence in managing perioperative hyperglycemia based on clinical experience or variations in institutional protocols.
The role of diabetes mellitus (DM) as a risk factor for postoperative infection in PSA remains controversial. Although large cohort analyses have identified DM as a risk factor for both superficial and deep infections, other studies have found no significant association between glycemic control and postoperative complications including infection risk and revision rates.16–19 Cancienne et al.16 observed a significant increase in wound complications and deep infections with rising perioperative HbA1c levels, identifying an inflection point at 8.0 mg/dL above which patients faced an increased risk of both short-term and long-term infectious complications. Conversely, McElvany et al.18 and Statz et al.19 found no correlation between HbA1c levels and postoperative infection or revision risk, even at lower cutoff values. Findings from the hip and knee arthroplasty literature mirror these inconsistencies. Shohat et al.20 reported higher surgical site infection (SSI) rates with inadequate glycemic control but found no support for an A1c threshold of 7% as a predictor of infection. Maradit Kremers et al.21 similarly observed increased PJI risk in diabetic and hyperglycemic patients, but HbA1c itself showed no independent association with infection after adjustment. Given the inconsistent evidence regarding DM's impact on infection risk in shoulder arthroplasty, further consensus is needed to establish optimal perioperative glycemic control strategies, particularly regarding the role of HbA1c thresholds in surgical eligibility.
Despite these differences, several areas of agreement were observed across experience levels, suggesting that certain infection prevention measures have been widely accepted as best practices. In this study, most of the surgeons (>75%) reported routine use of chlorhexidine-based preincision skin preparation. Chlorhexidine gluconate (CHG) has long been favored as a preoperative skin preparation agent because of its broad-spectrum antibacterial activity and efficacy in reducing overall bacterial colonization.8 However, its effectiveness against C. acnes remains limited because studies have demonstrated persistent C. acnes colonization despite CHG application.8,22 Matsen et al.22 further reinforced this limitation, showing that even with preoperative CHG showers, the skin burden of C. acnes was not significantly reduced. Emerging data suggest that the composition of the shoulder skin microbiome, including the relative abundance, load and genotype of C. acnes, and other resident organisms may better predict infection risk than surface decontamination alone.23–25 These findings highlight the importance of incorporating CHG solutions as part of a multimodal approach rather than relying on them in isolation for infection prevention.
The widespread use of cefazolin for prophylactic antibiotic administration reported in this study aligns with robust clinical evidence supporting its efficacy in reducing prosthetic joint infection (PJI) rates. A study of 7,713 primary shoulder arthroplasties found that cefazolin, administered in 89.2% of cases, was associated with a 69% reduction in overall PJI risk and a 78% reduction in C. acnes infections compared with noncefazolin alternatives such as vancomycin and clindamycin.26 However, although concerns regarding vancomycin's efficac have been raised, recent evidence suggests that improper administration rather than the antibiotic itself may be the underlying issue. In a study of 461 PSA patients, incomplete vancomycin administration was found to be an independent risk factor for PJI, with a fourfold increased risk (hazard ratio 4.22) even after adjusting for patient-related factors.27 Patients who did not receive a full vancomycin dose experienced significantly higher rates of any infectious complications (8% vs. 2.3%), PJIs (5.5% vs. 1%), and infection-related reoperations (4.9% vs. 1%). Survivorship free of PJI at 5 years was markedly lower in the incomplete administration group (95.1% vs 99.0%), underscoring the importance of ensuring complete vancomycin infusion before incision to maximize its protective effect.27 Timing of antibiotic administration is often dependent on anesthesia readiness, nursing workflows, and preoperative coordination among team members, with delays or miscommunication resulting in subtherapeutic dosing preincision.
This study has several limitations. First, the response rate of 18% may introduce selection and nonresponder bias because surgeons with a particular interest in infection prevention may have been more likely to participate. Recent ASES-wide electronic surveys (http://links.lww.com/BH9/A1) have reported response rates ranging from the high teens to approximately 30%, with several publications in the 18–20% range. Although on the lower end of that spectrum, an 18% rate remains within the reported range for society-level electronic surveys of ASES members.28–33 In addition, the respondent pool was weighted toward more experienced surgeons, with nearly 70% reporting 10 or more years in practice, which could skew the results. Furthermore, information was not collected on the geographic distribution, practice model and setting (eg, academic vs private practice), fellowship training of respondents, or whether reported strategies reflected surgeon preference vs institutionally mandated protocols, all of which may influence infection prevention practices. The study is further limited by the design of the survey, which may have influenced surgeons to report practices perceived as evidence-based or guideline-concordant rather than reflecting actual clinical behavior. Similarly, the use of self-reported data introduces the possibility of recall and response biases, potentially affecting the accuracy of the reported strategies. Furthermore, the cross-sectional design of the survey limits the ability to assess causal relationships between specific prevention strategies and rates of PJI.
Conclusion
Among ASES shoulder surgeons, there is considerable variability in preoperative, intraoperative, and postoperative infection prevention strategies among shoulder surgeons especially when stratified by years in practice. Surgeons with <5 years of experience are more likely to use many of the surveyed strategies than surgeons with greater experience. There is a lack of high quality evidence supporting infection prevention strategies, and future research should be directed toward gathering high-quality data to formulate evidence-based infection prevention guidelines.
Funding
No outside funding or grants were received in support of the completion of this study.
Conflict of Interest
Dr. Virk is a paid consultant for Advita, Inc. Dr. Zuckerman is a design surgeon and receives royalties for Advita, Inc. All other authors of this paper certify that they have no affiliations with or involvement in any organization or entity with any financial or nonfinancial interests pertinent to the subject matter discussed in this manuscript.
Institutional Review Board
This study contains only deidentified data, and it does not contain any protected health information. Therefore, this study was exempt from IRB approval.
Acknowledgments
Sophia M. Jacobi, BA; Young Kwon, MD, PhD.
Contributor Information
Jacquelyn J. Xu, Email: Jacquelyn.xu@downstate.edu.
Brian O. Molokwu, Email: brianmolo040@gmail.com.
Joseph D. Zuckerman, Email: joseph.zuckerman@nyulangone.org.
Mandeep S. Virk, Email: mandeep.virk@nyulangone.org.
Dennis A. DeBernardis, Email: dennisdebernardis@gmail.com.
References
- 1.Padegimas EM, Maltenfort M, Ramsey ML, Williams GR, Parvizi J, Namdari S. Periprosthetic shoulder infection in the United States: incidence and economic burden. J Shoulder Elbow Surg 2015;24:741–6. [DOI] [PubMed] [Google Scholar]
- 2.Jauregui JJ Tran A Kaveeshwar S, et al. Diagnosing a prosthetic shoulder infection: a systematic review. J Orthop. 2021;26:58-66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nazzal EM Herman ZJ Como M, et al. Shoulder prosthetic joint infection: principles of prevention, diagnosis, and treatment. J Bone Joint Surg Am. 2024;106:2265-75. [DOI] [PubMed] [Google Scholar]
- 4.Schick S, Elphingstone J, Murali S, et al. The incidence of shoulder arthroplasty infection presents a substantial economic burden in the United States: a predictive model. JSES Int 2023;7:636–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chalmers PN, Beck L, Stertz I, Tashjian RZ. Hydrogen peroxide skin preparation reduces Cutibacterium acnes in shoulder arthroplasty: a prospective, blinded, controlled trial. J Shoulder Elbow Surg 2019;28:1554–61. [DOI] [PubMed] [Google Scholar]
- 6.Kim HM, Huff H, Smith MJ, Nguyen M, Smith C. Effect of making skin incision with electrocautery on positive Cutibacterium acnes culture rates in shoulder arthroplasty: a prospective randomized clinical trial. J Shoulder Elbow Surg 2024;33:6–13. [DOI] [PubMed] [Google Scholar]
- 7.Puzzitiello RN, Menendez ME, Moverman MA, Pagani NR, Salzler MJ, Garrigues GE. Silver-impregnated occlusive dressings are a cost-effective strategy for preventing infection after total shoulder arthroplasty. Semin Arthroplasty: JSES 2021;31:552–6. [Google Scholar]
- 8.Saltzman MD, Nuber GW, Gryzlo SM, Marecek GS, Koh JL. Efficacy of surgical preparation solutions in shoulder surgery. J Bone Joint Surg Am 2009;91:1949–53. [DOI] [PubMed] [Google Scholar]
- 9.Clark JJC, Abildgaard JT, Backes J, Hawkins RJ. Preventing infection in shoulder surgery. J Shoulder Elbow Surg 2018;27:1333–41. [DOI] [PubMed] [Google Scholar]
- 10.Garrigues GE, Zmistowski B, Cooper AM, Green A; ICM Shoulder Group. Proceedings from the 2018 International Consensus Meeting on Orthopedic Infections: the definition of periprosthetic shoulder infection. J Shoulder Elbow Surg 2019;28:S8–12. [DOI] [PubMed] [Google Scholar]
- 11.Saad MA, Moverman MA, Da Silva AZ, Chalmers PN. Preventing infections in reverse shoulder arthroplasty. Curr Rev Musculoskelet Med 2024;17:456–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Diagnosis and Prevention of Periprosthetic Joint Infections Evidence-Based Clinical Practice Guideline. American Academy of Orthopaedic Surgeons. Available at: https://www.aaos.org/pjicpg. Accessed December 12, 2025. [Google Scholar]
- 13.Mizels J, Lewis DC, Tashjian RZ, Chalmers PN. Hydrogen peroxide may reduce the risk for revision surgery and infection in primary shoulder arthroplasty: two-year follow-up from a prospective, blinded, controlled trial. J Am Acad Orthop Surg 2024;32:e769–76. [DOI] [PubMed] [Google Scholar]
- 14.Miquel J, Huang TB, Athwal GS, Faber KJ, O'Gorman DB. Vancomycin is effective in preventing Cutibacterium acnes growth in a mimetic shoulder arthroplasty. J Shoulder Elbow Surg 2022;31:159–64. [DOI] [PubMed] [Google Scholar]
- 15.Garofalo R, Fontanarosa A, De Giorgi S, Lassandro N, De Crescenzo A. Vancomycin powder embedded in collagen sponge decreases the rate of prosthetic shoulder infection. J Shoulder Elbow Surg 2023;32:1638–44. [DOI] [PubMed] [Google Scholar]
- 16.Cancienne JM, Brockmeier SF, Werner BC. Association of perioperative glycemic control with deep postoperative infection after shoulder arthroplasty in patients with diabetes. J Am Acad Orthop Surg 2018;26:e238–45. [DOI] [PubMed] [Google Scholar]
- 17.Rao AJ, Yeatts NC, Reid RT, et al. Is postoperative glucose variability associated with adverse outcomes following shoulder arthroplasty?. J Shoulder Elbow Surg 2021;30:616–24. [DOI] [PubMed] [Google Scholar]
- 18.McElvany MD, Chan PH, Prentice HA, Paxton EW, Dillon MT, Navarro RA. Diabetes disease severity was not associated with risk of deep infection or revision after shoulder arthroplasty. Clin Orthop Relat Res 2019;477:1358–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Statz JM, Wagner ER, Sperling JW, Cofield RH. Outcomes of shoulder arthroplasty in diabetic patients as assessed by peri-operative A1C. Int Orthop 2018;42:1923–34. [DOI] [PubMed] [Google Scholar]
- 20.Shohat N, Muhsen K, Gilat R, Rondon AJ, Chen AF, Parvizi J. Inadequate glycemic control is associated with increased surgical site infection in total joint arthroplasty: a systematic review and meta-analysis. J Arthroplasty 2018;33:2312–21.e3. [DOI] [PubMed] [Google Scholar]
- 21.Maradit Kremers H, Lewallen LW, Mabry TM, Berry DJ, Berbari EF, Osmon DR. Diabetes mellitus, hyperglycemia, hemoglobin A1C and the risk of prosthetic joint infections in total hip and knee arthroplasty. J Arthroplasty 2015;30:439–43. [DOI] [PubMed] [Google Scholar]
- 22.Matsen FA, Whitson AJ, Hsu JE. While home chlorhexidine washes prior to shoulder surgery lower skin loads of most bacteria, they are not effective against Cutibacterium (Propionibacterium). Int Orthop 2020;44:531–4. [DOI] [PubMed] [Google Scholar]
- 23.Moore NF, Batten TJ, Hutton CE, White WJ, Smith CD. The management of the shoulder skin microbiome (Cutibacterium acnes) in the context of shoulder surgery: a review of the current literature. Shoulder Elbow 2021;13:592–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Burden E, Seyoum Y, Evans JP, et al. Mapping the microbial landscape and variations based on biological sex, age, and biopsy location in the shoulder skin microbiome. J Shoulder Elbow Surg 2025;34:e839–47. [DOI] [PubMed] [Google Scholar]
- 25.Kurihara MNL, Santos INM, Eisen AKA, et al. Phenotypic and genotypic characterization of Cutibacterium acnes isolated from shoulder surgery reveals insights into genetic diversity. Microorganisms 2023;11:2594. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Marigi EM, Bartels DW, Yoon JH, Sperling JW, Sanchez-Sotelo J. Antibiotic prophylaxis with cefazolin is associated with lower shoulder prosthetic joint infection rates than non-cefazolin alternatives. J Bone Joint Surg Am 2022;104:872–80. [DOI] [PubMed] [Google Scholar]
- 27.Marigi IM, Yu K, Nieboer MJ, et al. After primary shoulder arthroplasty appropriate vancomycin antibiotic prophylaxis does not lead to increased infectious complications when compared to cefazolin. J Shoulder Elbow Surg 2024;33:2612–8. [DOI] [PubMed] [Google Scholar]
- 28.Smith KM, Chalmers PN, Garrigues GE, Levine WN, Steinmann SP, Klifto CS. Job satisfaction and practice characteristics of shoulder and elbow surgeons: a survey of American Shoulder and Elbow Surgeons Society members. J Shoulder Elbow Surg 2024;33:e356–63. [DOI] [PubMed] [Google Scholar]
- 29.Chen AZ, Knudsen ML, Jobin CM, Levine WN. Early career opportunities and practice characteristics of recent graduates of shoulder and elbow fellowship programs. J Shoulder Elbow Surg 2022;31:e436–43. [DOI] [PubMed] [Google Scholar]
- 30.Garcia GH, Taylor SA, Fabricant PD, Dines JS. Shoulder instability management: a survey of the American Shoulder and Elbow Surgeons. Am J Orthop (Belle Mead NJ) 2016;45:E91–7. [PubMed] [Google Scholar]
- 31.Paxton ES, Matzon JL, Narzikul AC, Beredjiklian PK, Abboud JA. Agreement among ASES members on the AAOS Clinical Practice Guidelines. Orthopedics 2015;38:e169–77. [DOI] [PubMed] [Google Scholar]
- 32.Collins AP, McCall M, Cassinat J, et al. Performance of orthopaedic shoulder and elbow surgeons on a biostatistical knowledge examination. Adv Orthop 2023;2023:8840263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Saunders P, Mathur A, Frausto J, et al. Regional trends in perceptions of American Shoulder and Elbow Surgeons towards barriers to access for patients with Medicaid: limited perioperative service access, low patient engagement, and decreased reimbursement are regionally consistent obstacles. JSES Rev Rep Tech 2025;5:423–9. [DOI] [PMC free article] [PubMed] [Google Scholar]