Skip to main content
JAMA Network logoLink to JAMA Network
. 2023 Oct 31;6(10):e2339793. doi: 10.1001/jamanetworkopen.2023.39793

Postoperative Staphylococcus aureus Infections in Patients With and Without Preoperative Colonization

Darren P R Troeman 1,✉, Derek Hazard 2, Leen Timbermont 3, Surbhi Malhotra-Kumar 3, Cornelis H van Werkhoven 1, Martin Wolkewitz 2, Alexey Ruzin 4, Herman Goossens 3, Marc J M Bonten 1, Stephan Harbarth 5, Frangiscos Sifakis 6,7, Jan A J W Kluytmans 1,8; and the ASPIRE-SSI Study Team, Jelle Vlaeminck 3, Tuba Vilken 3, Basil Britto Xavier 3, Christine Lammens 3, Marjolein van Esschoten 1, Fleur P Paling 1, Claudia Recanatini 1, Frank Coenjaerts 9, Bret Sellman 7, Christine Tkaczyk 7, Susanne Weber 2, Miquel Bart Ekkelenkamp 8, Lijckle van der Laan 10, Bastiaan P Vierhout 11, Elodie Couvé-Deacon 12, Miruna David 13, David Chadwick 14, Martin J Llewelyn 15, Andrew Ustianowski 16, Antony Bateman 17, Damian Mawer 18, Biljana Carevic 19, Sonja Konstantinovic 20, Zorana Djordjevic 21, María Dolores del Toro-López 22, Juan Pablo Horcajada Gallego 23,24, Dolores Escudero 25, Miquel Pujol Rojo 26, Julian Torre-Cisneros 24,27, Francesco Castelli 28, Giuseppe Nardi 29, Pamela Barbadoro 30, Mait Altmets 31, Piret Mitt 32, Adrian Todor 33, Serban-Ion Bubenek-Turconi 34, Dan Corneci 35, Dorel Săndesc 36, Valeriu Gheorghita 37, Radim Brat 38, Ivo Hanke 39, Jan Neumann 40, Tomáš Tomáš 41, Wim Laffut 42, Anne-Marie Van den Abeele 43
PMCID: PMC10618839  PMID: 37906196

Key Points

Question

What is the cumulative incidence of Staphylococcus aureus surgical site infections (SSIs) and bloodstream infections (BSIs) in Europe, and what factors are associated with an increased risk of SSIs and BSIs?

Findings

In a cohort study of 5004 surgical patients, the weighted cumulative incidence of S aureus SSIs and BSIs was 1.23%. Preoperative S aureus carriage, mastectomy or neurosurgery, higher body mass index, and having nonremovable implants in the body were independently associated with S aureus SSIs and BSIs.

Meaning

Staphylococcus aureus SSIs and BSIs are important postoperative complications, and future interventions aimed at prevention of these infections should focus on at-risk surgical patient groups to achieve a higher efficacy.


This cohort study assesses the incidence of Staphylococcus aureus surgical site and bloodstream infections and examines factors associated with these infections among surgical patients with and without preoperative S aureus colonization.

Abstract

Importance

Staphylococcus aureus surgical site infections (SSIs) and bloodstream infections (BSIs) are important complications of surgical procedures for which prevention remains suboptimal. Contemporary data on the incidence of and etiologic factors for these infections are needed to support the development of improved preventive strategies.

Objectives

To assess the occurrence of postoperative S aureus SSIs and BSIs and quantify its association with patient-related and contextual factors.

Design, Setting, and Participants

This multicenter cohort study assessed surgical patients at 33 hospitals in 10 European countries who were recruited between December 16, 2016, and September 30, 2019 (follow-up through December 30, 2019). Enrolled patients were actively followed up for up to 90 days after surgery to assess the occurrence of S aureus SSIs and BSIs. Data analysis was performed between November 20, 2020, and April 21, 2022. All patients were 18 years or older and had undergone 11 different types of surgical procedures. They were screened for S aureus colonization in the nose, throat, and perineum within 30 days before surgery (source population). Both S aureus carriers and noncarriers were subsequently enrolled in a 2:1 ratio.

Exposure

Preoperative S aureus colonization.

Main Outcomes and Measures

The main outcome was cumulative incidence of S aureus SSIs and BSIs estimated for the source population, using weighted incidence calculation. The independent association of candidate variables was estimated using multivariable Cox proportional hazards regression models.

Results

In total, 5004 patients (median [IQR] age, 66 [56-72] years; 2510 [50.2%] female) were enrolled in the study cohort; 3369 (67.3%) were S aureus carriers. One hundred patients developed S aureus SSIs or BSIs within 90 days after surgery. The weighted cumulative incidence of S aureus SSIs or BSIs was 2.55% (95% CI, 2.05%-3.12%) for carriers and 0.52% (95% CI, 0.22%-0.91%) for noncarriers. Preoperative S aureus colonization (adjusted hazard ratio [AHR], 4.38; 95% CI, 2.19-8.76), having nonremovable implants (AHR, 2.00; 95% CI, 1.15-3.49), undergoing mastectomy (AHR, 5.13; 95% CI, 1.87-14.08) or neurosurgery (AHR, 2.47; 95% CI, 1.09-5.61) (compared with orthopedic surgery), and body mass index (AHR, 1.05; 95% CI, 1.01-1.08 per unit increase) were independently associated with S aureus SSIs and BSIs.

Conclusions and Relevance

In this cohort study of surgical patients, S aureus carriage was associated with an increased risk of developing S aureus SSIs and BSIs. Both modifiable and nonmodifiable etiologic factors were associated with this risk and should be addressed in those at increased S aureus SSI and BSI risk.

Introduction

Surgical site infections (SSIs) and postoperative bloodstream infections (BSIs) are important and common complications of surgical procedures,1,2 causing significant morbidity, extended hospital stays, and increased mortality and health care costs.3,4 A considerable proportion of these postoperative infections are caused by Staphylococcus aureus.5,6,7 In recent years, the management of postoperative S aureus infections has been complicated by the increase and spread of antibiotic-resistant S aureus.8 Because postoperative S aureus infections can have severe consequences for patients, there is an increasing need for effective interventions aimed at prevention and management of these infections. To support the development of these interventions, contemporary information about the incidence of and etiologic factors associated with S aureus postoperative infections is needed. The epidemiology of S aureus SSIs in Europe has not been fully described in part because of variations in case definitions and surveillance systems and a lack of comprehensive preoperative S aureus screening programs. Therefore, we attempted to assess the incidence and quantify the association of patient-related and contextual factors with S aureus SSIs and postoperative BSIs in Europe.

Methods

Study Design and Setting

ASPIRE-SSI (Advanced Understanding of Staphylococcus aureus Infections in Europe–SSI)9 was a prospective multicenter cohort study that recruited surgical patients at 33 hospitals in 10 European countries between December 16, 2016, and September 30, 2019. The last study patient completed follow-up on December 30, 2019. In each European subregion, as described by the United Nations,10 at least 2 countries were included. The study design and rationale have been described in detail elsewhere.11 Additional methods are reported in the eMethods in Supplement 1. Institutional or ethics review boards at each participating hospital or country approved the study protocol. All participants provided written informed consent. This study was conducted in accordance with the principles of the Declaration of Helsinki,12 the Medical Research Involving Human Subjects Act,13 and local guidelines in the participating countries. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.14

Study Participants

Patients 18 years or older undergoing 11 different types of surgical procedures were screened for S aureus colonization in the nose, throat, and perineum within 30 days before surgery and provided a preoperative serum sample. These patients constituted the source population. Subsequent enrollment in the study cohort (main study population) was determined by the preoperative S aureus colonization status of the patients, completeness of preoperative sample collection, and whether the surgery took place. Within each surgical procedure, all preoperative S aureus carriers were eligible for enrollment in the study cohort, together with the first eligible noncarrier after 2 enrolled S aureus carriers. The aim was to enroll 5000 study cohort participants. Exclusion criteria were simultaneous participation in any antistaphylococcal intervention study, an SSI as the reason for undergoing surgery, and expected inability to comply with study procedures or follow-up. The rationale for the sample size is described elsewhere.11 Postoperative follow-up data were collected for all patients at days 7, 14, 21, 28, 60, and 90 after surgery (±3 days each) by medical record review and by contacting the participants or next of kin. If a postoperative infection was suspected, the participant was encouraged to seek medical attention for clinical assessment and collection of microbiological cultures. The list of participating countries and the number of enrolled participants per country are given in eTable 1 in Supplement 1.

Data Collection and Outcomes

Clinical reasoning and a literature review guided the selection of variables that were collected in this study. A description of the collected variables is given in the eMethods in Supplement 1. All data were recorded in web-based case report forms. For this study, S aureus SSIs and postoperative BSIs were combined into 1 composite outcome. Staphylococcus aureus SSIs and postoperative BSIs were defined as the isolation of S aureus from a wound-related (eg, culture from surgical site or purulent drainage) or blood culture, respectively, and fulfilling the criteria for an SSI15 or BSI (criterion laboratory-confirmed bloodstream infection 1)16 according to the Centers for Disease Control and Prevention guidelines.

Laboratory Methods

Staphylococcus aureus screening samples were analyzed locally on chromogenic culture media (Colorex staph aureus; BioTrading Benelux B.V.) using standardized methods. The presence of S aureus was based on phenotypic criteria (pink- or mauve-colored colonies). This culture medium has a reported sensitivity of 95.5% and a specificity of 99.4% for the detection of S aureus.17 All collected S aureus strains, including from clinical samples in the case of SSIs or BSIs, were frozen at −80 °C and shipped to the central laboratory for further analysis. All received isolates (n = 4738; 3345 participants) were aerobically cultured on blood agar (defibrinated horse blood; E & O Laboratories Ltd) at 37 °C for 18 to 24 hours and identified using matrix-assisted laser desorption/ionization–time-of-flight mass spectrometry (Bruker Corp). Methicillin susceptibility of all confirmed S aureus strains (n = 4427; 3148 participants) was determined using high-throughput screening for cefoxitin resistance. Briefly, bacterial suspensions of 0.5 McFarland were spotted on methicillin-resistant S aureus (MRSA) chromogenic media (CHROMagar; BioTrading) using a multiblot replicator (VP 407-96 Pin Multi-Blot Replicator [9-mm centers, 2.36-mm pin diameter, 22-mm long, blunt tip]; V&P Scientific Inc), followed by aerobic incubation at 37 °C for 18 to 24 hours. Suspected MRSA was confirmed by a cefoxitin E-test (bioMérieux). Cefoxitin minimum inhibitory concentration greater than 4 μg/mL was indicative of MRSA according to the European Committee on Antimicrobial Susceptibility Testing breakthrough guidelines, version 10.0. Additionally, all available S aureus isolates from the patients who developed S aureus SSIs and BSIs (screening isolates [n = 84; 54 patients] and infecting isolates [n = 139; 60 patients]) and from a random sample of noninfected patients (n = 221; 162 patients) were characterized by multilocus sequence typing using whole-genome sequencing data (eFigure 1 in Supplement 1). Genomic DNA was extracted from S aureus isolates using a high-molecular-weight DNA extractor (MagAttract HMW DNA Kit; Qiagen), per the manufacturer’s instructions, and quantified with a double-stranded DNA high-sensitivity assay (Qubit dsDNA HS assay kit; Thermo Fisher Scientific). Libraries were then generated (NexteraXT DNA sample preparation kit; Illumina Inc) and sequenced (2 × 250 bp) with a sequencer (MiSeq; Illumina Inc). FastQC, version 0.11.7 (Babraham Bioinformatics) was used to assess the quality of the raw sequence data. Trimming of reads, building a draft genome, and multilocus sequence typing were performed using BacPipe, version 1.2.6.18

Statistical Analysis

Baseline characteristics are presented in aggregate and stratified by preoperative S aureus colonization status. Study estimates were determined in the study cohort and estimated for the source population using weighting methods.19 These weighting methods considered the likelihood of patients to be enrolled in the study cohort to obtain study estimates for the underlying source population. A detailed description of these methods is given in the eMethods in Supplement 1.

Incidence Calculation

The 90-day cumulative incidence of S aureus SSIs and BSIs (only the first episode per patient) was calculated for the entire weighted study population and for S aureus carriers and noncarriers separately. The 95% CIs were determined by bootstrapping and applying the weights.

Risk Factor Analysis

Univariable cause-specific Cox proportional hazards regression models were used to assess associations between each explanatory variable and (1) the outcome S aureus SSIs and BSIs (with censoring for loss to follow-up and death) and (2) the competing event death without S aureus SSIs and BSIs (with censoring for loss to follow-up and S aureus SSIs and BSIs). The latter was done to assess whether any variable could also be associated with informative censoring by being associated with the competing events and thus affect the hazard ratio (HR) of explanatory variables for S aureus SSIs and BSIs. This process yielded univariable cause-specific HRs. Because of anticipated differences among countries, we defined country as a stratum variable. Variables with a P ≤ .157 (corresponding to selection via Akaike information criterion optimization) in either analysis were included in a multivariable model to quantify the relative rates of developing S aureus SSIs and BSIs. A 2-sided P < .05 was considered significant. All analyses were conducted on weighted and unweighted data.

Missing Data

We assumed data were missing at random and used default multivariate imputation by chained equation procedures to acquire 5 suitable imputed data sets for the main analyses. Model estimates were pooled according to Rubin’s rules20 to provide a single mean estimate and adjusted SEs for each variable. Using these results, we derived pooled (adjusted) HRs and robust 95% CIs for each variable.

Sensitivity Analysis

We conducted several sensitivity analyses to assess the robustness of the study results. First, we repeated the multivariable analysis while keeping preoperative S aureus decolonization in the model because it is an important confounder of the occurrence association between preoperative S aureus colonization and postoperative S aureus infection. Second, we repeated the univariable and multivariable analyses using the subdistribution hazards (Fine and Gray)21 approach to assess the independent associations between the explanatory variables and the risk of developing S aureus SSIs and BSIs. Third, we conducted a complete case analysis to assess whether the imputation procedure had any effect on the study results. All statistical analyses were performed between November 20, 2020, and April 21, 2022, using R statistical software, version 4.0.2 (R Foundation for Statistical Computing).22

Results

In total, 5004 patients (median [IQR] age, 66 [56-72] years; 2510 [50.2%] female and 2494 [49.8%] male) were included in the study cohort; 3369 patients (67.3%) were S aureus carriers. Table 1 gives the characteristics of the S aureus carriers and noncarriers of the study cohort. Compared with the noncarriers, carriers were more often male, had a prior history of S aureus colonization or infection, and received preoperative topical decolonization treatment. Other characteristics were similar. The characteristics of the weighted and original source population are presented in eTable 2 in Supplement 1.

Table 1. Baseline Characteristics of Study Participantsa.

Characteristic Unweighted population
Staphylococcus aureus carriers (n = 3369) Noncarriers (n = 1635)
Sex
Female 1642 (48.7) 868 (53.1)
Male 1727 (51.3) 767 (46.9)
Implant before surgery
No 2544 (75.5) 1246 (76.2)
Yes 820 (24.3) 387 (23.7)
Missing 5 (0.1) 2 (0.1)
Type of surgery
Cardiovascular surgeryb 1002 (29.7) 488 (29.8)
Orthopedic surgeryc 1231 (36.5) 609 (37.2)
Neurosurgeryd 612 (18.2) 287 (17.6)
Emergency surgery 207 (6.1) 99 (6.1)
Mastectomy 317 (9.4) 152 (9.3)
ASA score
1 326 (9.7) 136 (8.3)
2 1445 (42.9) 707 (43.2)
3 1269 (37.7) 640 (39.1)
≥4 187 (5.6) 93 (5.7)
Missing 142 (4.2) 59 (3.6)
Receipt of immunosuppressive medication within 2 weeks of surgery
No 3194 (94.8) 1559 (95.4)
Yes 171 (5.1) 75 (4.6)
Missing 4 (0.1) 1 (0.1)
History of S aureus colonization or infection
No 3033 (90) 1589 (97.2)
Yes 329 (9.8) 43 (2.6)
Missing 7 (0.2) 3 (0.2)
Receipt of preoperative decolonization treatment
No 2538 (75.3) 1353 (82.8)
Yes 831 (24.7) 282 (17.2)
Age, median (IQR), y 65 (55;72) 67 (58;73)
Missing 0 0
BMI, median (IQR) 27.9 (24.9-31.6) 27.7 (24.8-31.1)
Missing 44 (1.3) 23 (1.4)
CCI, median (IQR) 1 (0-2) 1 (0-2)
Missing 1 (0.0) 2 (0.1)

Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CCI, Charlson Comorbidity Index.

a

Data are presented as number (percentage) unless otherwise indicated.

b

Including open cardiac surgery, implantable cardioverter defibrillator implantation, peripheral artery bypass surgery, and central artery reconstructive surgery.

c

Including hip prosthesis and knee prosthesis surgery.

d

Including craniotomy, laminectomy, and spinal fusion surgery.

Patient Recruitment

In all, 10 691 patients were included in the source population; 121 patients were not screened correctly, and 1063 patients did not undergo surgery or missed serum sample collection. Of the remaining 9507 patients, 3501 (36.8%) were S aureus carriers and 6006 (63.2%) noncarriers. For the current analysis, 3369 (67.3%) S aureus carriers and 1635 (32.7%) noncarriers were included in the study cohort. Figure 1 shows the flow of patient recruitment and reasons for nonparticipation at each stage.

Figure 1. Flow of Participants in the Advanced Understanding of Staphylococcus aureus Infections in Europe–Surgical Site Infections (ASPIRE-SSI) Study.

Figure 1.

Incidence of S aureus SSIs and BSIs

A total of 100 study cohort participants developed a S aureus SSI or BSI during follow-up (91 SSIs, 4 BSIs, and 5 SSIs with secondary BSIs). Eighty-six (86.0%) of these patients were preoperative S aureus carriers. Of the 96 S aureus SSIs, 47 (49.0%) were superficial, 34 (35.4%) were deep, and 15 (15.6%) were organ-space SSIs.

Within the weighted study cohort (n = 9657), 119 participants developed an S aureus SSI or BSI. Eighty-six participants were S aureus carriers and 33 were noncarriers. The weighted cumulative incidence of S aureus SSIs and BSIs was 1.23% (95% CI, 0.97%-1.58%) for the entire weighted study population, 2.55% (95% CI, 2.05%-3.12%) for S aureus carriers, and 0.52% (95% CI, 0.22%-0.91%) for noncarriers (Table 2 and Figure 2). There were also differences in the incidence of S aureus SSIs and BSIs among different surgical procedures (eTable 3 in Supplement 1). Unweighted incidence calculations are provided in eTable 4 and eFigure 2 in Supplement 1.

Table 2. Weighted Cumulative Incidences by Preoperative Staphylococcus aureus Colonization Statusa.

S aureus colonization status No. of patients No. of S aureus SSI and BSI events Cumulative incidence per 100 patients (95% CI) Time to event, median (IQR), d
Carrier 3369 86 2.55 (2.05-3.12) 19 (13-33)
Noncarrier 6288 33 0.52 (0.22-0.91) 22 (11-28)
Overall 9657 119 1.23 (0.97-1.58) 20 (13-33)

Abbreviations: BSI, bloodstream infection; SSI, surgical site infection.

a

The numbers of patients and events given in the table are the weighted totals. These numbers were derived by bootstrapping. A total of 10 000 bootstrap samples of the study cohort were made, after which the bootstrap samples were inflated using the weights (creating 10 000 weighted bootstrap samples). In each weighted bootstrap sample, the cumulative incidence of S aureus SSIs and BSIs was calculated for S aureus carriers and noncarriers. The sequence of 10 000 cumulative incidences for S aureus carriers and noncarriers was used to derive the median cumulative incidence with 95% CIs.

Figure 2. Cumulative Incidence Function for Staphylococcus aureus Surgical Site Infections (SSIs) and Bloodstream Infections (BSIs) in S aureus Colonized vs Noncolonized Patients.

Figure 2.

Prevalence of MRSA Carriage

Of the 4230 screening S aureus isolates analyzed (3138 participants), 162 (127 participants) were MRSA. Thus, MRSA carriage was observed in 127 of 3138 S aureus carriers (4.0%); the remaining isolates were methicillin-susceptible S aureus (eTable 5 in Supplement 1).

High Concordance in Sequence Type Between Colonizing and Infecting S aureus

Among patients who developed an S aureus SSI or BSI, the colonizing and infecting S aureus isolates were distributed into 23 and 25 sequence types (STs), respectively. The colonizing S aureus isolated from the randomly selected cohort of noninfected patients was distributed into 37 STs. The most prevalent STs in patients who developed an S aureus SSI or BSI and in the randomly selected cohort participants were STs 30, 45, 5, 8, and 398, with no apparent preference of certain STs to cause infection (eTable 6 in Supplement 1). For 53 patients who developed an S aureus infection, both colonizing and infecting S aureus strains were available. For 44 of these 53 patients (83.0%), the ST of the colonizing and infecting strains was identical.

Risk Factor Analysis

A total of 364 records (7.3%) included missing values. Primary reasons for missing data were that the data could not be retrieved from the medical records or that the perineum screening sample was not collected. Results from the weighted univariable and multivariable analyses are presented in Table 3.

Table 3. Weighted Risk Factor Analysis for Staphylococcus aureus SSIs and BSIs.

Risk factor Univariable analysis for the association with S aureus SSIs and BSIs Univariable analysis for the association with death without S aureus SSIs and BSIs Multivariable analysis for the association with S aureus SSIs and BSIsa
Crude HR (95% CI) P value Crude HR (95% CI) P value Adjusted HR (95% CI) P value
Preoperative S aureus colonization statusb
Colonized 4.56 (2.26-9.21) <.001 1.17 (0.70-1.97) .55 4.38 (2.19-8.76) <.001
Noncolonized 1 [Reference] 1 [Reference] 1 [Reference]
Sex
Male 1.35 (0.79-2.30) .27 1.41 (0.83-2.39) .20 Not included Not applicable
Female 1 [Reference] 1 [Reference]
Nonremovable implant before surgery 1.56 (0.91-2.67) .10 0.48 (0.24-0.99) .05 2.00 (1.15-3.49) .01
Type of surgery
Cardiovascular surgery 2.13 (1.10-4.12) .03 5.55 (2.34-13.21) <.001 1.91 (0.86-4.22) .11
Mastectomy 4.12 (1.52-11.20) 006 0.44 (0.06-3.44) .43 5.13 (1.87-14.08) .002
Neurosurgery 2.37 (1.01-5.55) .05 12.80 (5.23-31.33) <.001 2.47 (1.09-5.61) .03
Emergency surgery 1.78 (0.54-5.84) .35 7.40 (2.37-23.06) .001 2.42 (0.67-8.84) .18
Orthopedic surgery 1 [Reference] Not applicable 1 [Reference] Not applicable 1 [Reference] Not applicable
Having no postoperative drains 0.82 (0.47-1.41) .47 1.52 (0.80-2.88) .21 Not included Not applicable
ASA score
1 0.73 (0.32-1.71) .47 0.63 (0.13-3.06) .57 0.59 (0.26-1.34) .21
3 1.46 (0.80-2.69) .22 3.30 (1.48-7.39) .004 1.49 (0.63-3.55) .37
≥4 1.66 (0.67-4.12) .28 7.91 (3.17-19.76) <.001 1.70 (0.59-4.87) .33
2 1 [Reference] Not applicable 1 [Reference] Not applicable Not included Not applicable
Immunosuppressive medication within 2 weeks before surgery 1.77 (0.72-4.36) .21 3.88 (2.29- 6.55) <.001 1.74 (0.62-4.91) .30
Prior history of S aureus colonization or infection 1.11 (0.46-2.67) .82 2.79 (0.47-16.50) .26 Not included Not applicable
Preoperative decolonization 1.35 (0.75-2.44) .32 1.09 (0.49- 2.47) .83 Not included Not applicable
Agec 0.99 (0.97-1.00) .03 1.03 (0.99- 1.06) .12 0.98 (0.97-1.00) .08
BMId 1.04 (1.01-1.08) .009 0.97 (0.90-1.05) .48 1.05 (1.01-1.08) .008
CCIe 1.16 (1.05-1.28) .003 1.49 (1.36- 1.64) <.001 1.09 (0.97-1.22) .17

Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); BSI, bloodstream infection; CCI, Charlson Comorbidity Index; HR, hazard ratio; SSI, surgical site infection.

a

Risk factors with a P ≤ .157 in either univariable analysis were selected for the multivariable analysis.

b

Colonization status before surgery based on S aureus screening of the nose, throat, and perineum.

c

Per 1-year increase in age (range, 18-99 years).

d

Per 1-point increase in BMI (range, 13.5-65.8).

e

Per 1-point increase in the CCI (range, 0-12).

The adjusted HR (AHR) for S aureus carriers compared with noncarriers for developing S aureus SSIs and BSIs was 4.38 (95% CI, 2.19-8.76). Having any noninfected and nonremovable body implant preoperatively (AHR, 2.00; 95% CI, 1.15-3.49), undergoing mastectomy (AHR, 5.13; 95% CI, 1.87-14.08) or neurosurgery (AHR, 2.47; 95% CI, 1.09-5.61) (compared with orthopedic surgery), and having an elevated body mass index (BMI) (AHR, 1.05; 95% CI, 1.01-1.08 per unit increase) also increased the daily likelihood of S aureus SSIs and BSIs. The unweighted AHRs are presented in eTable 7 in Supplement 1.

Sensitivity Analyses

The association of each explanatory variable with the outcome S aureus SSIs and BSIs were reassessed in three sensitivity analyses. Repeating the main analyses while keeping preoperative S aureus decolonization in the multivariable analysis (eTable 8 in Supplement 1), using the subdistribution hazards approach (eTable 9 in Supplement 1), or using complete data only (eTable 10 in Supplement 1) did not yield any relevant changes in the reported estimates or conclusions.

Discussion

In this large, prospective cohort study, we found that preoperative S aureus carriage is still a major etiologic factor for S aureus SSIs and BSIs. The finding that in 83% of colonized patients who developed an S aureus SSI or BSI the colonizing and infecting S aureus strain had identical STs further supports this association. Additionally, we found that certain types of surgery (mastectomy and neurosurgery), an increasing BMI, and having nonremovable implants also independently increased the risk of S aureus SSIs and BSIs. Interestingly, preoperative colonization and postoperative infection with MRSA remained uncommon.

Our study confirms findings from previous studies23,24,25 that endogenous S aureus carriage is an important etiologic factor for postoperative S aureus infections. However, external factors probably also play a significant role in the etiology of postoperative S aureus infections. Furthermore, we observed a low level of preoperative MRSA carriage, whereas a higher MRSA prevalence has been reported in several countries.26,27 This finding could be attributable to a decreasing trend in MRSA prevalence8 or to the finding that the previously reported high prevalence of MRSA carriage was primarily health care associated and not community based.

An interesting finding was that S aureus ST398 was found in isolates from study patients (mostly from Northern and Southern Europe) and that it was identified in approximately 15% of S aureus SSI and BSI events. This ST was originally discovered in livestock animals in the early 2000s.28 Since then, there have been multiple reports of this ST causing infections in humans, especially among those working near livestock or other farm animals, although ST398 S aureus strains (predominantly methicillin susceptible) also cause infections in the absence of livestock exposure.29 Our data suggest that this S aureus ST has become a more common colonizing and infecting strain in certain parts of Europe, although it did not seem to cause more SSIs and BSIs than other STs in our study population.

A high BMI is a well-established risk factor for noncommunicable diseases, such as cardiovascular disease or type 2 diabetes.30 However, its role in postoperative outcomes remains a topic of debate, with some studies suggesting that patients with a higher BMI are not at increased risk of detrimental postoperative outcomes, such as mortality31,32,33 or infection,34,35 while others suggest the opposite.36,37,38,39 In our study, an increase in BMI was associated with an increased risk of S aureus SSIs and BSIs. This finding is in accordance with the hypothesis that suboptimal tissue oxygenation and wound hypoxia, in combination with altered antimicrobial pharmacokinetics and increased hepatic clearance of antimicrobials, which are found in obesity, increase the susceptibility of patients with higher BMI to develop postoperative infections.37

Another interesting finding was that mastectomy and neurosurgery, compared with orthopedic procedures, were associated with an increased risk of S aureus SSIs and BSIs. Although existing data show that certain surgical procedures have a higher SSI rate than other procedures,40 this is the first study, to our knowledge, that has compared head-to-head the risk of S aureus SSIs and BSIs of different surgical procedures from different surgical subspecialties in such a large cohort. It is important that health care interventions aimed at S aureus SSI and BSI prevention target the patient populations who have an increased risk of this outcome first. These data could help in the prioritization of these efforts.

Unexpectedly, preoperative topical S aureus decolonization was not associated with S aureus SSIs and BSIs in any of our analyses, although there is convincing evidence supporting a protective effect of preoperative S aureus decolonization on the occurrence of S aureus SSI.41 However, the participating sites used different preoperative decolonization strategies for different surgical procedures (universal decolonization for certain procedures regardless of colonization status, targeted decolonization in case of S aureus colonization or MRSA carriage only, or no decolonization), using different types of decolonization agents. Therefore, the use of topical S aureus decolonization treatment was probably too diverse; thus, we cannot draw any conclusions regarding the true effectiveness of preoperative S aureus colonization against S aureus SSIs and BSIs based on this finding.

Last, we also found that having nonremovable artificial body implants before surgery increased the risk of developing S aureus SSIs and BSIs. This finding was surprising because there is currently no published evidence to support a biological mechanism or epidemiologic link. It could be that the foreign material predisposes patients to postoperative infections by facilitating hematogenous seeding of the pathogen,42,43 although we do not have information to support this claim.

Limitations

This study had some limitations that merit consideration. First, for the sample size that we included, we expected a higher incidence of S aureus SSIs and BSIs than we observed. Because of this, we were restricted in the number of potential etiologic factors that we could analyze. Despite this, we were still able to address important etiologic factors in our analyses, although we cannot exclude that there still might be some residual confounding present. Second, the evaluation of preoperative S aureus carriage was based only on colonial morphologic findings, and no additional microbiological confirmation was requested. For that reason, we cannot exclude misclassification of S aureus carrier status for some patients. Third, patient recruitment in certain countries and for certain surgical procedures progressed slower than we expected, which resulted in overrepresentation of patients from certain countries and surgical procedures in the study (eTables 1 and 2 in Supplement 1). This overrepresentation may have affected the generalizability of our results.

Conclusions

The findings of this cohort study indicate that preoperative S aureus carriage is still a major predisposing factor for postoperative S aureus infection, increasing the daily risk more than 4-fold. Additionally, surgery type, an increasing BMI, and having any artificial body implants before surgery may also have an impact on the risk of developing these infections. We provide updated data that can be used to design future clinical trials of strategies aimed at S aureus SSI and BSI prevention or that can provide guidance on which surgical patient groups to prioritize when implementing a strategy aimed at S aureus SSI and BSI prevention.

Supplement 1.

eMethods. Supplemental methods

eTable 1. Number of study participants per country and European sub-region

eTable 2. Baseline characteristics of the weighted study population and original source population

eTable 3. Weighted cumulative incidence of SA SSI/BSI within 90 days post-surgery by surgery type

eTable 4. Unweighted cumulative incidence of SA SSI/BSI by preoperative SA colonization status

eTable 5. Methicillin-susceptibility of colonizing strains

eTable 6. ST types of the isolates from the sub-cohort (N = 346)

eTable 7. Unweighted risk factor analysis for SA SSI/BSI

eTable 8. Weighted risk factor analysis for SA SSI/BSI (keeping preoperative SA decolonization in multivariable model)

eTable 9. Weighted Fine and Gray model for SA SSI/BSI

eTable 10. Weighted risk factor analysis for SA SSI/BSI (complete case analysis)

eFigure 1. Subject selection and number of isolates for the MLST analysis

eFigure 2. Cumulative incidence function for SA SSI/BSI (unweighted data)

eReferences

Supplement 2.

Data Sharing Statement

References

  • 1.O’Brien WJ, Gupta K, Itani KMF. Association of postoperative infection with risk of long-term infection and mortality. JAMA Surg. 2020;155(1):61-68. doi: 10.1001/jamasurg.2019.4539 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gottlieb GS, Fowler VG Jr, Kong LK, et al. Staphylococcus aureus bacteremia in the surgical patient: a prospective analysis of 73 postoperative patients who developed Staphylococcus aureus bacteremia at a tertiary care facility. J Am Coll Surg. 2000;190(1):50-57. doi: 10.1016/S1072-7515(99)00211-2 [DOI] [PubMed] [Google Scholar]
  • 3.Badia JM, Casey AL, Petrosillo N, Hudson PM, Mitchell SA, Crosby C. Impact of surgical site infection on healthcare costs and patient outcomes: a systematic review in six European countries. J Hosp Infect. 2017;96(1):1-15. doi: 10.1016/j.jhin.2017.03.004 [DOI] [PubMed] [Google Scholar]
  • 4.Klevens RM, Edwards JR, Richards CL Jr, et al. Estimating health care-associated infections and deaths in U.S. hospitals, 2002. Public Health Rep. 2007;122(2):160-166. doi: 10.1177/003335490712200205 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Marimuthu K, Eisenring MC, Harbarth S, Troillet N. Epidemiology of Staphylococcus aureus surgical site infections. Surg Infect (Larchmt). 2016;17(2):229-235. doi: 10.1089/sur.2015.055 [DOI] [PubMed] [Google Scholar]
  • 6.Haque M, Sartelli M, McKimm J, Abu Bakar M. Health care-associated infections: an overview. Infect Drug Resist. 2018;11:2321-2333. doi: 10.2147/IDR.S177247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.European Centre for Disease Prevention and Control . Healthcare-associated infections: surgical site infections. In: Annual Epidemiological Report for 2017. ECDC; 2019. [Google Scholar]
  • 8.Lee AS, de Lencastre H, Garau J, et al. Methicillin-resistant Staphylococcus aureus. Nat Rev Dis Primers. 2018;4:18033. doi: 10.1038/nrdp.2018.33 [DOI] [PubMed] [Google Scholar]
  • 9.Advanced understanding of staphylococcus aureus infections in Europe—surgical site infections (ASPIRE-SSI). ClinicalTrials.gov identifier: NCT02935244. Updated May 3, 2021. Accessed September 20, 2022. https://classic.clinicaltrials.gov/ct2/show/NCT02935244
  • 10.Statistics Division United Nations . Methodology: standard country or area codes for statistical use (M49). Accessed May 2, 2023. https://unstats.un.org/unsd/methodology/m49/#fn1
  • 11.Troeman DPR, Weber S, Hazard D, et al. Designing the ASPIRE-SSI study: a multicenter, observational, prospective cohort study to assess the incidence and risk factors of surgical site and bloodstream infections caused by Staphylococcus aureus in Europe. medRxiv. Preprint posted online July 10, 2020. doi: 10.1101/2020.07.08.20148791 [DOI]
  • 12.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
  • 13.Wet medisch-wetenschappelijk onderzoek met mensen. Published February 26, 1998. Accessed May 2, 2023. https://wetten.overheid.nl/BWBR0009408/2022-07-01
  • 14.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. doi: 10.1136/bmj.39335.541782.AD [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.National Healthcare Safety Network, Centers for Disease Prevention and Control . Surgical site infection (SSI) event. Published January 2023. Accessed May 2, 2023. https://www.cdc.gov/nhsn/pdfs/pscmanual/9pscssicurrent.pdf
  • 16.National Healthcare Safety Network, Centers of Disease Control and Prevention . Bloodstream infection event (central line-associated bloodstream infection and non-central line associated bloodstream infection). Published January 2023. Accessed May 2, 2023. https://www.cdc.gov/nhsn/pdfs/pscmanual/4psc_clabscurrent.pdf
  • 17.Gaillot O, Wetsch M, Fortineau N, Berche P. Evaluation of CHROMagar Staph. aureus, a new chromogenic medium, for isolation and presumptive identification of Staphylococcus aureus from human clinical specimens. J Clin Microbiol. 2000;38(4):1587-1591. doi: 10.1128/JCM.38.4.1587-1591.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Xavier BB, Mysara M, Bolzan M, et al. BacPipe: a rapid, user-friendly whole-genome sequencing pipeline for clinical diagnostic bacteriology. iScience. 2020;23(1):100769. doi: 10.1016/j.isci.2019.100769 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lee ES, Forthofer RR. Analyzing Complex Survey Data. Vol 71. 2nd ed. SAGE Publications; 2005. [Google Scholar]
  • 20.Rubin DB. Multiple Imputation for Nonresponse in Surveys. John Wiley & Sons Inc; 2004. [Google Scholar]
  • 21.Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94(446):496-509. doi: 10.1080/01621459.1999.10474144 [DOI] [Google Scholar]
  • 22.R Core Team . R: a language and environment for statistical computing [computer program]. R Foundation for Statistical Computing; 2022.
  • 23.Paling FP, Hazard D, Bonten MJM, et al. ; ASPIRE-ICU Study Team . Association of Staphylococcus aureus colonization and pneumonia in the intensive care unit. JAMA Netw Open. 2020;3(9):e2012741. doi: 10.1001/jamanetworkopen.2020.12741 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Albrecht VS, Limbago BM, Moran GJ, et al. ; EMERGEncy ID NET Study Group . Staphylococcus aureus colonization and strain type at various body sites among patients with a closed abscess and uninfected controls at U.S. emergency departments. J Clin Microbiol. 2015;53(11):3478-3484. doi: 10.1128/JCM.01371-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.von Eiff C, Becker K, Machka K, Stammer H, Peters G; Study Group . Nasal carriage as a source of Staphylococcus aureus bacteremia. N Engl J Med. 2001;344(1):11-16. doi: 10.1056/NEJM200101043440102 [DOI] [PubMed] [Google Scholar]
  • 26.Lee AS, Cooper BS, Malhotra-Kumar S, et al. ; MOSAR WP4 Study Group . Comparison of strategies to reduce methicillin-resistant Staphylococcus aureus rates in surgical patients: a controlled multicentre intervention trial. BMJ Open. 2013;3(9):e003126. doi: 10.1136/bmjopen-2013-003126 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.European Centre for Disease Prevention and Control . Antimicrobial Resistance Surveillance in Europe 2014: Annual Report of the European Antimicrobial Resistance Surveillance Network (EARS-net). ECDC; 2015. [Google Scholar]
  • 28.van Cleef BA, Monnet DL, Voss A, et al. Livestock-associated methicillin-resistant Staphylococcus aureus in humans, Europe. Emerg Infect Dis. 2011;17(3):502-505. doi: 10.3201/eid1703.101036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lu H, Zhao L, Si Y, et al. The surge of hypervirulent ST398 MRSA lineage with higher biofilm-forming ability is a critical threat to clinics. Front Microbiol. 2021;12:636788. doi: 10.3389/fmicb.2021.636788 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med. 1999;341(15):1097-1105. doi:MJBA-411501 doi: 10.1056/NEJM199910073411501 [DOI] [PubMed] [Google Scholar]
  • 31.Dotan I, Shochat T, Shimon I, Akirov A. The association between BMI and mortality in surgical patients. World J Surg. 2021;45(5):1390-1399. doi: 10.1007/s00268-021-05961-4 [DOI] [PubMed] [Google Scholar]
  • 32.Mariscalco G, Wozniak MJ, Dawson AG, et al. Body mass index and mortality among adults undergoing cardiac surgery: a nationwide study with a systematic review and meta-analysis. Circulation. 2017;135(9):850-863. doi: 10.1161/CIRCULATIONAHA.116.022840 [DOI] [PubMed] [Google Scholar]
  • 33.Nafiu OO, Kheterpal S, Moulding R, et al. The association of body mass index to postoperative outcomes in elderly vascular surgery patients: a reverse J-curve phenomenon. Anesth Analg. 2011;112(1):23-29. doi: 10.1213/ANE.0b013e3181fcc51a [DOI] [PubMed] [Google Scholar]
  • 34.Thomas EJ, Goldman L, Mangione CM, et al. Body mass index as a correlate of postoperative complications and resource utilization. Am J Med. 1997;102(3):277-283. doi: 10.1016/S0002-9343(96)00451-2 [DOI] [PubMed] [Google Scholar]
  • 35.Waisbren E, Rosen H, Bader AM, Lipsitz SR, Rogers SO Jr, Eriksson E. Percent body fat and prediction of surgical site infection. J Am Coll Surg. 2010;210(4):381-389. doi: 10.1016/j.jamcollsurg.2010.01.004 [DOI] [PubMed] [Google Scholar]
  • 36.Fu CY, Bajani F, Bokhari M, et al. Obesity is associated with worse outcomes among abdominal trauma patients undergoing laparotomy: a propensity-matched nationwide cohort study. World J Surg. 2020;44(3):755-763. doi: 10.1007/s00268-019-05268-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Thelwall S, Harrington P, Sheridan E, Lamagni T. Impact of obesity on the risk of wound infection following surgery: results from a nationwide prospective multicentre cohort study in England. Clin Microbiol Infect. 2015;21(11):1008.e1-1008.e8. doi: 10.1016/j.cmi.2015.07.003 [DOI] [PubMed] [Google Scholar]
  • 38.Galyfos G, Geropapas GI, Kerasidis S, Sianou A, Sigala F, Filis K. The effect of body mass index on major outcomes after vascular surgery. J Vasc Surg. 2017;65(4):1193-1207. doi: 10.1016/j.jvs.2016.09.032 [DOI] [PubMed] [Google Scholar]
  • 39.Sommerstein R, Atkinson A, Kuster SP, et al. ; Swissnoso Network . Association between antimicrobial prophylaxis with double-dose cefuroxime and surgical site infections in patients weighing 80 kg or more. JAMA Netw Open. 2021;4(12):e2138926. doi: 10.1001/jamanetworkopen.2021.38926 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Thorn L, Lamagni T, Harrington P, Wloch C, Hope R, Brown C. Surveillance of surgical site infections in NHS hospitals in England April 2020 to March 2021. Accessed September 20, 2023. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1087917/SSI-annual-report-2020-to-2021.pdf
  • 41.Bode LG, Kluytmans JA, Wertheim HF, et al. Preventing surgical-site infections in nasal carriers of Staphylococcus aureus. N Engl J Med. 2010;362(1):9-17. doi: 10.1056/NEJMoa0808939 [DOI] [PubMed] [Google Scholar]
  • 42.Tande AJ, Patel R. Prosthetic joint infection. Clin Microbiol Rev. 2014;27(2):302-345. doi: 10.1128/CMR.00111-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Rakow A, Perka C, Trampuz A, Renz N. Origin and characteristics of haematogenous periprosthetic joint infection. Clin Microbiol Infect. 2019;25(7):845-850. doi: 10.1016/j.cmi.2018.10.010 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eMethods. Supplemental methods

eTable 1. Number of study participants per country and European sub-region

eTable 2. Baseline characteristics of the weighted study population and original source population

eTable 3. Weighted cumulative incidence of SA SSI/BSI within 90 days post-surgery by surgery type

eTable 4. Unweighted cumulative incidence of SA SSI/BSI by preoperative SA colonization status

eTable 5. Methicillin-susceptibility of colonizing strains

eTable 6. ST types of the isolates from the sub-cohort (N = 346)

eTable 7. Unweighted risk factor analysis for SA SSI/BSI

eTable 8. Weighted risk factor analysis for SA SSI/BSI (keeping preoperative SA decolonization in multivariable model)

eTable 9. Weighted Fine and Gray model for SA SSI/BSI

eTable 10. Weighted risk factor analysis for SA SSI/BSI (complete case analysis)

eFigure 1. Subject selection and number of isolates for the MLST analysis

eFigure 2. Cumulative incidence function for SA SSI/BSI (unweighted data)

eReferences

Supplement 2.

Data Sharing Statement


Articles from JAMA Network Open are provided here courtesy of American Medical Association

RESOURCES