Comparing clinical manifestations, treatment strategies and outcomes of Staphylococcus lugdunensis bacteraemia versus Staphylococcus aureus bacteraemia – a retrospective cohort study Authors: Gemma Webb, Jasmine Choi, Angela Younes, Emily Schembri, Roy Chean, Stephen Guy Published: 2026 ABSTRACT AbstractBackground Staphylococcus lugdunensis is a coagulase‐negative Staphylococcus that has been increasingly recognised as a significant pathogen capable of causing invasive infections similar to Staphylococcus aureus. Data comparing the clinical manifestations and outcomes of S. lugdunensis bacteraemia to S. aureus bacteraemia remain limited. This article seeks to compare the two diseases in order to better understand the clinical course of S. lugdunensis bacteraemia.AimsThis study aims to compare the clinical course, management strategies, outcomes and complications of S. lugdunensis and S. aureus bacteraemia in order to better characterise S. lugdunensis bacteraemia.MethodsWe conducted a retrospective cohort study of 57 patients with S. lugdunensis bacteraemia and 66 matched patients with S. aureus bacteraemia across a metropolitan health service in Melbourne, Australia, between 2012 and 2024. Patient demographics, clinical features, biochemical markers, management strategies and outcomes were compared.ResultsBaseline demographics and comorbidities were similar between the groups. S. lugdunensis was more frequently isolated in polymicrobial cultures compared to S. aureus (14/57 vs 2/66) and was more frequently viewed as a contaminant by treating clinicians (12/57 vs 1/66). S. aureus bacteraemia was associated with longer duration of bacteraemia, and more frequent osteoarticular and spinal epidural infections. However, rates of infective endocarditis were similar between the two groups (9/57 vs 15/66), and inpatient and 90‐day mortality outcomes were comparable.ConclusionOur findings support the idea that S. lugdunensis bacteraemia should be approached with similar clinical vigilance as S. aureus, given comparable mortality and potential for serious complications, namely infective endocarditis. Improved clinician awareness is needed to reduce misclassification of S. lugdunensis as contaminants to allow for timely diagnosis and management. Abbreviations antibiotics Acquired Immunodeficiency syndrome alanine transaminase congestive heart failure confidence interval chronic kidney disease clumping factor A central nervous system coagulase negative Staphylococcus chronic obstructive pulmonary disease C‐reactive protein computed tomography cerebrovascular accident diabetes mellitus fibrinogen‐binding protein hospital in the home intensive care unit infectious diseases infective endocarditis interquartile range intravenous myocardial infarction magnetic resonance imaging methicillin sensitive Staphylococcus aureus number odds ratio positron emission tomography peptic ulcer disease peripheral vascular disease quality assurance transient ischaemic attack transoesophageal echocardiogram transthoracic echocardiogram treatment von Willebrand factor white cell count Introduction Staphylococcus lugdunensis is a coagulase‐negative Staphylococcus (CoNS) that forms part of the normal skin flora, most notably in the groin and lower extremities. 1 , 2 Historically, S. lugdunensis was often grouped with other CoNS and often dismissed as a contaminant when identified in blood cultures. However, accumulating evidence over the past two decades has challenged this perception, revealing its potential to cause more severe infections, including bacteraemia, infective endocarditis, abscesses and osteomyelitis. 3 By comparison, Staphylococcus aureus bloodstream infections are well known to place a significant burden on health services globally. 4 Despite its rising recognition as a pathogen, limited data exist regarding the clinical course of S. lugdunensis bacteraemia, particularly in comparison to the more well‐characterised S. aureus bacteraemia; however, several recent studies suggest that it may exhibit similar pathogenic potential. A 2023 retrospective study from Japan found comparable 7‐day mortality rates between S. lugdunensis and S. aureus bacteraemia cohorts (9% vs 7%, P = 0.77). 5 Furthermore, S. lugdunensis bacteraemia has been associated with similar rates of infective endocarditis to S. aureus bacteraemia, especially in the event of multiple positive blood cultures. 6 Other cohort studies have shown that a single positive blood culture for S. lugdunensis can result in a similar clinical course to that of multiple positive S. lugdunensis blood cultures, particularly when supported by clinical context. 7 In this retrospective review conducted across three hospitals within a metropolitan health service in Melbourne, Australia, we compare the clinical and biochemical features, treatment strategies and outcomes of 57 patients with S. lugdunensis bacteraemia to 66 patients with S. aureus bacteraemia, with the aim to improve understanding of S. lugdunensis bacteraemia and support more informed, evidence‐based approaches to its management. Methods This was a retrospective cohort study conducted across a metropolitan health service in Melbourne, Victoria. The study period spanned from January 2012 to December 2024. All S. lugdunensis‐positive blood cultures processed by the microbiology laboratory during this period were screened for inclusion. Cases were excluded if the positive blood culture could not be linked to a known patient based on hospital unique patient identifiers or if case records were not available. For each year between 2012 and 2024, one S. aureus positive blood culture was randomly selected as a comparator for each included S. lugdunensis case, matched by year of collection. To perform this matching randomly, each unique episode of S. aureus bacteraemia and S. lugdunensis bacteraemia per year was assigned a case number. S. aureus cases were selected by randomly generating one number for each S. lugdunensis bacteraemia case number. This one‐to‐one matching by year was undertaken to account for temporal changes in diagnostic approaches and treatment approaches over the 12‐year study period. Bacteraemia was defined as the presence of a bacterial species in at least one blood culture, including a single bottle of a set. Data were extracted from electronic medical records and included patient demographics and comorbidities, biochemical markers, number of positive blood cultures throughout the hospital admission and antimicrobial susceptibility profiles of the organism isolated. Additional data included antibiotic choice upon identification of bacteraemia, treatment duration, investigations undertaken to evaluate the source or complications of bacteraemia, evidence of infective source and seeding as determined by treating clinicians, inpatient length of stay, and inpatient and 90‐day mortality outcomes. All analyses were performed using STATA version 18 (StataCorp, College Station, TX, USA). Patient and disease characteristics were summarised in available data using frequencies and proportions, means and standard deviations, and medians and interquartile ranges dependent on data distribution. Comparisons between the S. lugdunensis and S. aureus groups were conducted using the chi‐squared test, Fisher's exact test, Student's t‐test or Mann–Whitney U test, as appropriate. Associations between pathogen and selected study outcomes were assessed using univariate and multivariate logistic regression, which adjusted for age and sex, white cell count (WCC), C‐reactive protein (CRP) and platelet count. A two‐tailed P‐value of <0.05 was considered statistically significant. Quality assurance (QA) ethics approval was obtained from the health service's Human Research Ethics Committee (Reference Number: QA25‐010‐116 142), and all patient data were de‐identified prior to analysis. Results Eighty‐two unique episodes of S. lugdunensis bacteraemia were identified between January 2012 and December 2024. Twenty‐five episodes were excluded due to the blood culture result not corresponding to a unique patient identifier with a positive blood culture result or no clinical information being available, resulting in 57 S. lugdunensis bacteraemia cases included in the study. A comparator group of 81 S. aureus bacteraemia cases identified at the same health service between January 2012 and December 2024 were randomly selected from the microbiology database, of which 15 were excluded for the same reason, resulting in 66 S. aureus bacteraemia cases being included in the study. Baseline demographic profiles were similar between the two groups. Comorbidities assessed using the Charlson Comorbidity Index (CCI) were comparable between the S. lugdunensis and S. aureus groups with a median CCI of 5 in both groups. No individual comorbidity assessed in the CCI reached differences of statistical significance between the two groups (Table 1). Table 1 S. lugdunensis versus S. aureus bacteraemia, patient demographics and comorbidity dataTotal population S. lugdunensis S. aureus P‐value N 1235766Sex, n (%)Male78 (63.4%)32 (56.1%)46 (69.7%)0.13Female45 (36.6%)25 (43.9%)20 (30.3%)Age, median (IQR)73 (58.5–82)76 (57–85)72 (60–79)0.457ComorbidityMI, n (%)15 (12.2%)4 (7%)11 (16.7%)0.14CHF, n (%)14 (11.4%)7 (7%)6 (9.1%)0.56PVD, n (%)13 (10.6%)8 (14%)5 (7.6%)0.210CVA/TIA, n (%)13 (10.6%)6 (10.5%)7 (10.6%)0.999Dementia, n (%)7 (5.7%)3 (5.3%)4 (6.1%)0.999COPD, n (%)11 (9.0%)7 (12.3%)4 (6.1%)0.343Connective tissue disease, n (%)4 (3.3%)2 (3.5%)2 (3.0%)0.999PUD, n (%)7 (5.8%)4 (7.0%)3 (4.5%)0.700Liver disease, n (%)None116 (94.3%)54 (94.7%)62 (94.0%)0.870Mild3 (2.4%)1 (1.8%)2 (3.0%)Moderate/severe4 (3.3%)2 (3.5%)2 (3.0%)DM, n (%)None/diet controlled91 (74%)45 (79%)46 (69.7%)0.560Uncomplicated21 (17.1%)8 (14%)13 (19.7%)Endorgan11 (8.9%)4 (7%)7 (10.6%)Hemiplegia, n (%)2 (1.7%)0 (0)2 (3.0%)0.204Moderate/Severe CKD, n (%)15 (12.2%)5 (8.8%)10 (15.2%)0.285Solid tumour, n (%)None98 (79.7%)45 (79%)53 (80.3%)0.820Localised11 (8.9%)6 (10.5%)5 (7.6%)Metastatic14 (11.4%)6 (10.5%)8 (12.1%)Leukaemia, n (%)3 (2.4%)1 (1.8%)2 (3.0%)0.999Lymphoma, n (%)4 (3.3%)1 (1.8%)3 (4.6%)0.622AIDS, n (%)0 (0%)0 (0%)0 (0%)‐Charlson comorbidity index, median (IQR)5 (3–7)5 (3–7)5 (3–7)0.972Iatrogenic immunosuppression, n (%)16 (13.0%)7 (12.2%)9 (13.6%)0.999IV drug use last 12 months, n (%)4 (3.3%)1 (1.8%)3 (4.6%)0.623Prosthetic IV/endovascular device, n (%)24 (19.5%)7 (12.2%)17 (25.8%)0.060Previous IE, n (%)2 (1.6%)2 (3.5%)0 (0)0.501Underlying structural cardiac abnormality, n (%)18 (14.6%)8 (14.0%)10 (15.2%)0.817Abbreviations: AIDS, Acquired Immunodeficiency syndrome; CHF, congestive heart failure; CVA, cerebrovascular accident; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; IE, infective endocarditis; IQR, interquartile range; IV, intravenous; MI, myocardial infarction; PUD, peptic ulcer disease; PVD, peripheral vascular disease; TIA, transient ischaemic attack. S. lugdunensis versus S. aureus bacteraemia, patient demographics and comorbidity data Abbreviations: AIDS, Acquired Immunodeficiency syndrome; CHF, congestive heart failure; CVA, cerebrovascular accident; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; IE, infective endocarditis; IQR, interquartile range; IV, intravenous; MI, myocardial infarction; PUD, peptic ulcer disease; PVD, peripheral vascular disease; TIA, transient ischaemic attack. On day 1 of blood culture positivity, median C‐reactive protein (CRP) was more elevated at 176.8 mg/L in S. aureus group compared to 107.4 mg/L in the S. lugdunensis group (p = 0.001, normal range < 5 mg/L). 8 This trend was also observed across the first 14 days since first blood culture positivity. Alanine transaminase (ALT) was also significantly higher in the S. aureus group on day 1 (27 U/L in S. aureus group vs 19 U/L in S. lugdunensis group, P = 0.016, normal range 0–40 U/L in males and 0–35 U/L in females) and across days 1–14. 9 Thrombocytopaenia was observed in the S. aureus group, with the median platelet count reaching a nadir of 136 × 109 /L, compared to 175 × 109/L in the S. lugdunensis group (P = 0.011, normal range 150–400 × 109/L). 10 There were no significant differences in white cell count, creatinine or lactate at any time point between the two groups. The S. aureus group demonstrated a significant persistence of bacteraemia, with a median of 3.5 positive blood cultures during the hospital admission versus 1 in the S. lugdunensis group (P < 0.001), and 7y5fcbvj 89.4% of all S. aureus cases had more than one positive blood culture, compared to 27.5% of S. lugdunensis cases (P < 0.001) (Table 2). Table 2 S. lugdunensis versus S. aureus bacteraemia, microbiology resultsTotal population S. lugdunensis S. aureus P Value N 1235766Total positive cultures, median (IQR)2 (1–4)1 (1–2)3.5 (2–6)<0.001Number of cases with >1 positive culture (%)75 (60.5%)16 (27.5%)59 (89.4%)<0.001Total positive days, median (IQR)1 (1–3)1 (1–1)3 (1–4)<0.001Penicillin susceptibility, n (%)31 (26.3%)26 (46.4%)5 (8.0%)<0.001 Not reported 5 1 4 Oxacillin/methicillin susceptibility, n (%)113 (93.4%)55 (98.2%)58 (89.2%)0.068 Not reported 2 1 1 Vancomycin susceptibility, n (%)12 (100%)5 (100%)7 (100%)— Not reported 111 52 59 Additional organisms identified, n (%)16 (13%)14 (24.6%)2 (3%)0.002Which additional organisms Staphylococcus epidermidis – 4 Other coagulase negative Staphylococcus – 9 Enterococcus faecalis – 1 Staphylococcus aureus – 2 Streptococcus species – 3 Klebsiella pneumoniae – 1 E. faecalis – 1 Streptococcus anginosus – 1 Not reported = sensitivity testing not performed.Abbreviation: IQR, interquartile range. S. lugdunensis versus S. aureus bacteraemia, microbiology results Staphylococcus epidermidis – 4 Other coagulase negative Staphylococcus – 9 Enterococcus faecalis – 1 Staphylococcus aureus – 2 Streptococcus species – 3 Klebsiella pneumoniae – 1 Staphylococcus epidermidis – 4 Staphylococcus epidermidis – 4 Other coagulase negative Staphylococcus – 9 Other coagulase negative Staphylococcus – 9 Enterococcus faecalis – 1 Enterococcus faecalis – 1 Staphylococcus aureus – 2 Staphylococcus aureus – 2 Streptococcus species – 3 Streptococcus species – 3 Klebsiella pneumoniae – 1 Klebsiella pneumoniae – 1 E. faecalis – 1 Streptococcus anginosus – 1 E. faecalis – 1 E. faecalis – 1 Streptococcus anginosus – 1 Streptococcus anginosus – 1 Not reported = sensitivity testing not performed. Abbreviation: IQR, interquartile range. A similar proportion of S. lugdunensis and S. aureus patients were given IV antibiotic therapy (93% vs 98.8%, P = 0.181); however, significantly more S. aureus patients were managed with antibiotics typically used for staphylococcal cover such as cefazolin and flucloxacillin compared to the S. lugdunensis group, in which ceftriaxone was the most used agent. Antibiotic duration also differed between the two groups as outlined in Table 3. Table 3 S. lugdunensis versus S. aureus bacteraemia, treatmentTotal population S. lugdunensis S. aureus P‐valueTreatment given, n (%)Yes118 (95.8%)53 (93.0%)65 (98.5%)0.174Backbone IV therapy, n (%)Ceftriaxone18 (15.3%)16 (30.2%)2 (3.1%)<0.001Ampicillin1 (0.9%)1 (1.9%)0 (0)Benzylpenicillin5 (4.2%)4 (7.6%)1 (1.5%)Cefazolin32 (27.1%)12 (22.6%)20 (30.8%)Daptomycin3 (2.5%)1 (1.9%)2 (3.1%)Flucloxacillin48 (40.7%)15 (28.3%)33 (50.8%)Meropenem4 (3.4%)3 (5.7%)1 (1.5%)Vancomycin7 (5.9%)1 (1.9%)6 (9.2%)IV Abx duration, n (%)1–7 days27 (25.0%)20 (42.6%)4 (6.5%)0.0058–14 days17 (15.7%)8 (17.0%)9 (14.7%)15–21 days13 (12.0%)4 (8.5%)9 (14.7%)22–28 days12 (11.1%)4 (8.5%)9 (14.7%)29–35 days5 (4.6%)2 (4.3%)3 (4.9%)36–42 days17 (15.7%)2 (4.3%)15 (24.6%)>42 days17 (15.7%)7 (14.9%)10 (16.4%) Not reported 12 6 6 Not reported = Patient transferred to alternate healthcare provider prior to antibiotics course completion.Abbreviations: IV, intravenous; Abx, antibiotics. S. lugdunensis versus S. aureus bacteraemia, treatment Not reported = Patient transferred to alternate healthcare provider prior to antibiotics course completion. Abbreviations: IV, intravenous; Abx, antibiotics. Transthoracic echocardiogram (TTE) was performed in 93.9% of S. aureus bacteraemia cases versus 61.4% of S. lugdunensis bacteraemia cases (P < 0.001). Of the 16 cases of S. lugdunensis bacteraemia with ≥2 positive blood cultures, 14 (87.5%) underwent echocardiography. The remaining two were transferred to different hospitals within the first 3 days of admission. Other imaging performed to identify infective sources and/or seeding took place often in the S. aureus bacteraemia group. S. aureus bacteraemia patients were also referred for infectious disease consultation 93.9% of the time compared to 64.9% of S. lugdunensis cases (P < 0.001) (Table 4). Table 4 S. lugdunensis versus S. aureus bacteraemia, investigations and infective fociTotal population S. lugdunensis S. aureus P ValueInvestigationsTTE, n (%)97 (78.9%)35 (61.4%)62 (93.9%)<0.001TOE, n (%)22 (17.9%)5 (8.8%)17 (25.8%)0.014Whole body PET/CT/NM, n (%)27 (22.0%)8 (14.0%)19 (28.8%)0.049MRI, n (%)37 (30.1%)10 (17.5%)27 (40.9%)0.005ID consultation (in person)99 (80.5%)37 (64.9%)62 (93.9%)<0.001Infective fociInfective endocarditis (IE)24 (19.5%)9 (15.8%)15 (22.7%)0.333Site of IEAortic8 (34.8%)3 (33.3%)5 (35.7%)0.816Mitral11 (47.8%)6 (66.7%)5 (35.7%)Pulmonary1 (4.3%)0 (0)1 (7.1%)Tricuspid2 (8.7%)0 (0)2 (14.3%)Right atrium1 (4.3%)0 (0)1 (7.1%)If IE, native or prosthetic valveNative17 (70.8%)6 (66.7%)11 (73.3%)0.999Prosthetic7 (29.2%)3 (33.3%)4 (26.7%)Native osteoarticular infection, n (%)27 (22.0%)5 (8.8%)22 (33.3%)<0.001Pleuropulmonary infection, n (%)19 (15.5%)10 (17.5%)9 (13.6%)0.646Skin/soft tissue infection, n (%)28 (22.8%)9 (15.8%)19 (28.8%)0.066Intra‐abdominal infection, n (%)10 (8.1%)7 (12.3%)3 (4.6%)0.143IV‐line associated infection, n (%)16 (14.0%)5 (8.8%)11 (16.7%)0.157Urinary tract infection, n (%)7 (5.7%)2 (3.5%)5 (7.6%)0.452Spinal epidural abscess, n (%)12 (10.8%)1 (1.8%)11 (16.7%)0.004CNS infection, n (%)8 (6.5%)3 (5.3%)5 (7.6%)0.720Bacteraemia with no definite source, n (%)27 (22.0%)15 (26.3%)12 (18.2%)0.232Abbreviations: CT, computed tomography; CNS, central nervous system; ID, infectious diseases; IE, infective endocarditis; IV, intravenous; NM, Nuclear Medicine Scan; MRI, magnetic resonance imaging; PET, positron emission tomography; TOE, transoesophageal echocardiogram; TTE, transthoracic echocardiogram. S. lugdunensis versus S. aureus bacteraemia, investigations and infective foci Abbreviations: CT, computed tomography; CNS, central nervous system; ID, infectious diseases; IE, infective endocarditis; IV, intravenous; NM, Nuclear Medicine Scan; MRI, magnetic resonance imaging; PET, positron emission tomography; TOE, transoesophageal echocardiogram; TTE, transthoracic echocardiogram. Infective endocarditis was diagnosed in 19.5% of all cases, without a significant difference between groups (15.8% of S. lugdunensis cases vs 22.7% of S. aureus cases, P = 0.333). All cases of infective endocarditis were retroactively assessed for, and met, the 2023 Modified Duke Criteria for Infective Endocarditis. 11 On univariate analysis, S. aureus bacteraemia and S. lugdunensis bacteraemia confer similar odds of developing infective endocarditis (odds ratio (OR) 0.64, 95% confidence interval (CI): 0.26–1.59, P = 0.335). This finding persisted when analysis was adjusted for age, sex, WCC, CRP and platelet count of day of first culture positivity (OR 1.39, 95% CI: 0.47–4.15, P = 0.554). S. aureus bacteraemia was associated with significantly higher prevalence of native osteoarticular infection and spinal epidural abscesses (Table 4). In patients with S. lugdunensis bacteraemia, the presence of a long‐term endovascular device was associated with increased odds of developing infective endocarditis on univariate analysis (OR 12.00, 95% CI: 2.07–69.69, P = 0.006) and adjusted for age and sex (OR 11.72, 95% CI: 1.86–76.58, P = 0.009). Likewise, patients with S. lugdunensis bacteraemia with >1 positive blood culture had increased odds of developing infective endocarditis on univariate analysis (OR 40.00, 95% CI: 4.37–365.76, P = 0.001) and adjusted for age and sex (OR 91.42, 95% CI: 4.87–1716.59, P = 0.003). Variations in WCC, CRP and platelet count on the day of first culture positivity were not associated with increased odds of developing infective endocarditis (WCC OR 1.09, 95% CI: 0.98–1.21, P = 0.125; CRP OR: 1.01, 95% CI: 0.99–1.02, P = 0.084; platelets OR: 0.99, 95% CI: 0.98–1.00, P = 0.330). In‐hospital mortality was similar between groups (S. lugdunensis 10.5% vs S. aureus 9.2%, P = 0.811), as was 90‐day mortality (S. lugdunensis 13.0% vs S. aureus 12.5%, P = 0.940). S. lugdunensis patients had significantly shorter inpatient stays (median 8 days in S. lugdunensis group vs 14 days in S. aureus group, P = 0.004) (Table 5). Odds of inpatient mortality between the two groups were also similar on univariate modelling (OR 1.15, CI: 0.35–3.81, P = 0.811) and multivariate analysis adjusting for age, sex, WCC, CRP and platelet count (OR 1.45, CI: 0.36–5.87, P = 0.605). S. lugdunensis bacteraemia was associated with lower odds of an inpatient stay >17 days compared to S. aureus bacteraemia on univariate analysis (OR 0.26, CI: 0.12–0.55, P < 0.001) and adjusted for age, sex and the previously listed biochemical markers (OR 0.31, CI: 0.13–0.74, P = 0.008). Mortality data and inpatient stay duration were recorded as ‘not reported’ if patients were transferred to another health service provider without any further correspondence from that provider. Table 5 S. lugdunensis vs S. aureus bacteraemia, complications and outcomesTotal population S. lugdunensis S. aureus P‐valuePitt bacteremia score, median (IQR)1 (0–2)1 (0–2.5)1 (0–2)0.830Dialysis, n (%)14 (11.4%)5 (8.8%)9 (13.6%)0.388ICU admission, n (%)42 (34.1%)18 (31.6%)24 (36.4%)0.703Discharge destination, n (%)Deceased10 (8.1%)4 (7.0%)6 (9.1%)0.013Home25 (20.3%)18 (31.6%)7 (10.6%)Home with HITH38 (30.9%)11 (19.3%)27 (40.9%)New nursing home2 (1.6%)2 (3.5%)0 (0)Rehab19 (15.4%)8 (14.0%)11 (16.7%)Transferred to other hospital29 (23.6%)14 (24.6%)15 (22.7%)Days on ward since positive culture, median (IQR)12 (6–20)8 (5–16.5)14 (8–25)0.004In hospital mortality, n (%)12 (9.8%)6 (10.5%)6 (9.2%)0.811 Not reported 1 0 1 90‐day mortality, n (%)15 (12.7%)7 (13.0%)8 (12.5%)0.940 Not reported 5 3 2 Total in patient days in 90 days, median (IQR)17 (8–40)12 (6–28)27 (13–43)0.001 Not reported 110Was the organism, n (%):Contaminant13 (10.6%)12 (21.0%)1 (1.5%)<0.001Indeterminant9 (7.3%)9 (15.8%)0 (0)Pathogen101 (82.1%)36 (63.2%)65 (98.5%)Abbreviations: HITH, hospital in the home; ICU, intensive care unit; IQR, interquartile range. S. lugdunensis vs S. aureus bacteraemia, complications and outcomes Abbreviations: HITH, hospital in the home; ICU, intensive care unit; IQR, interquartile range. Clinician perceptions upon identifying each bacterium in blood cultures also varied significantly; S. lugdunensis was regarded as a contaminant by treating clinicians in 21.0% of cases, and of uncertain significance in 15.8% of cases. S. aureus was considered significant in all but 1.5% of cases (P < 0.001). All diagnoses of infective endocarditis in the S. lugdunensis group occurred in cases where the bacteria was considered a pathogen (9 vs 0 cases, P = 0.019). Cases viewed as ‘true’ S. lugdunensis bacteraemia had a longer median inpatient stay across 90 days compared to those viewed as contaminants or indeterminant (14 vs 7 days, P = 0.031). In hospital, all‐cause mortality was similar between the S. lugdunensis bacteraemia group viewed as pathogens and those viewed as contaminants/indeterminant, at 14.3% in the contaminant/indeterminant group versus 8.3% in the pathogen group (P = 0.659). Among S. lugdunensis bacteraemia episodes, 43 (75.4%) were monomicrobial and 14 (24.6%) were polymicrobial. On day 1, median CRP was significantly higher in polymicrobial cases (159 vs 88 mg/L, P = 0.0267), while other biochemical markers were similar between groups. There were no statistically significant differences in infective foci, with differing rates of infective endocarditis not reaching statistical significance (7.2% in polymicrobial cases vs 18.6% in monomicrobial cases, P = 0.565) (Table 2). Discussion In this 12‐year retrospective study, we compared the clinical characteristics, management and outcomes of S. lugdunensis bacteraemia with those of S. aureus bacteraemia. In our cohort, S. aureus bacteraemia produced a longer duration of bacteraemia and higher prevalence of native osteoarticular infection and spinal epidural abscess compared to S. lugdunensis. The two demonstrated similar prevalences of infective endocarditis despite significantly less S. lugdunensis bacteraemia cases undergoing echocardiography (93.9% S. aureus vs 61.4% S. lugdunensis). Compared with S. aureus bacteraemia, clinicians were less likely to pursue advanced imaging and infectious disease speciality consultation for S. lugdunensis bacteraemia. Previous cohort studies on the role of infectious disease speciality input in the management of S. aureus bacteraemia noted improved mortality outcomes when infectious disease teams were involved in patient care and higher rates of appropriate workup, including echocardiography. 12 , 13 Given that in our cohort S. lugdunensis bacteraemia resulted in similar rates of infective endocarditis and similar mortality compared to S. aureus, it seems advisable that infectious disease teams are involved in the care of all S. lugdunensis bacteraemia patients. We found that patients with S. lugdunensis bacteraemia had increased odds of developing infective endocarditis if they had >1 positive blood culture or if they had a pre‐existing endovascular device in situ; however, given the wide 95% CI obtained during these analyses, the magnitude of the effect of these variables cannot be assessed acutely. Clinicians should request echocardiography for patients presenting with S. lugdunensis in ≥2 separate blood culture sets and/or with endovascular devices in situ or other clinical features suggestive of infective endocarditis. A notable contributor to the relatively high rate of infective endocarditis in S. lugdunensis bacteraemia may be its unique virulence factors that enhance its endothelial and valvular adherence compared to other CoNS species. S. lugdunensis expresses the fibrinogen‐binding protein (Fbl), a surface‐anchored adhesin with approximately 60% sequence homology to the S. aureus clumping factor A (ClfA), capable of binding fibrinogen and promoting colonisation of thrombotic or damaged surfaces. 14 , 15 , 16 Animal models have also demonstrated S. lugdunensis binding to von Willebrand factor (vWF) in flow environments, promoting adhesion to blood vessels and valvular surfaces. 17 The S. lugdunensis microbial surface components recognising adhesive matrix molecule (MSCRAMM) known as vWbl have been considered as a potential ligand for vWF to facilitate this adhesion; however, animal model studies have not identified a statistically significant change in rates of infective endocarditis when the gene encoding this protein is deleted. 17 , 18 S. lugdunensis also produces a carbohydrate‐rich glycocalyx that behaves as a biofilm, though the exact mechanisms of biofilm formation, as well as the roles of other components of the S. lugdunensis biofilm, remain an area of research. 19 , 20 , 21 The high risk of endocarditis as a complication of S. lugdunensis bacteraemia has been increasingly recognised, and S. lugdunensis has been included in the ‘typical organisms’ list in the 2023 update to the Modified Duke Criteria for infective endocarditis. 11 Our findings concur that infective endocarditis is a notable complication of S. lugdunensis and identifies having >1 positive blood culture and/or the presence of long‐term endovascular devices as risk factors for the development of infective endocarditis. The finding that polymicrobial blood cultures were significantly more common in S. lugdunensis bacteraemia cases compared to S. aureus bacteraemia (24.6% vs 3%, P = 0.002) raises important considerations regarding the clinical significance of these isolates. While polymicrobial cultures often suggest contamination, particularly given the skin commensal nature of S. lugdunensis, the possibility of true polymicrobial infection cannot be easily excluded. Careful clinical correlation, including assessment of patient symptoms, source identification and repeat cultures, seems essential to guide management decisions in these cases. Ceftriaxone was the antibiotic most frequently used in the treatment of S. lugdunensis bacteraemia, despite 98.2% of isolates demonstrating oxacillin/methicillin susceptibility and 46.4% of isolates demonstrating penicillin susceptibility. In the S. aureus bacteraemia population, pharmacodynamic studies have found cefazolin to have more bactericidal activity at appropriate dosing compared to ceftriaxone, and standard of care for methicillin sensitive S. aureus (MSSA) bacteraemia is currently established to be the use of cefazolin or penicillin agents with anti‐staphylococcal activity. 22 , 23 Large‐scale studies addressing different antibiotic pharmacodynamics and clinical outcomes of S. lugdunensis bacteraemia with ceftriaxone versus standard of care MSSA therapy have not been performed; however, the use of the narrowest spectrum, targeted therapy for S. lugdunensis bacteraemia, seems prudent. The comparable mortality rates between S. lugdunensis bacteraemia and S. aureus bacteraemia found in our cohort highlight the importance of approaching S. lugdunensis with a similar diagnostic rigour that clinicians would use for S. aureus bacteraemia. Additionally, the similar mortality rates between S. lugdunensis cases classified by clinicians as contaminants or indeterminate and the 36 cases deemed true pathogens raise the possibility that some dismissed or unacknowledged blood cultures may have been clinically significant. However, this interpretation is limited by the retrospective design of the study, especially since most cases labelled as contaminants involved patients that were hospitalised with other serious medical conditions that could also impact mortality. Hence, the attributable mortality of S. lugdunensis bacteraemia remains unknown. This study has several important limitations. The classification of S. lugdunensis isolates as true infection versus contamination relied on the discretion of treating clinicians, and retrospective evaluation of these assessments was constrained by the quality of hospital records. Follow‐up assessments undertaken by treating clinicians, including subsequent blood cultures and further investigations, varied significantly depending on each clinician's interpretation of the S. lugdunensis finding. Several results demonstrated high statistical significance but were accompanied by extremely wide 95% CIs. This instability occurred due to the relatively small size of the S. lugdunensis infective endocarditis cohort, limiting the conclusions that can be drawn from those data points. Additionally, the retrospective design inherently limits control over data collection and introduces potential biases, most notably confounding bias. Furthermore, several patients were transferred to specialised centres for cardiothoracic or neurosurgical care, resulting in loss to follow‐up and incomplete capture of outcomes such as mortality and metastatic complications, which may have influenced the study's findings. Strengths of this study are the relatively large number of S. lugdunensis cases and the breadth of aspects of patient workup evaluated. In future, prospective evaluations of S. lugdunensis bacteraemia would contribute greatly to global understanding of this disease. Conclusion This study reinforces the idea that S. lugdunensis is a clinically significant pathogen capable of causing severe invasive infections that are in some respects comparable to those caused by S. aureus. Although S. aureus bacteraemia showed longer persistence and higher rates of some infective complications, including osteoarticular infections and spinal epidural abscesses, S. lugdunensis bacteraemia exhibited comparable rates of infective endocarditis and similar mortality outcomes. Infectious disease teams should be readily involved in the care of patients presenting with S. lugdunensis bacteraemia to assist in determining the most appropriate workup of potential complications of the bacteraemia and the most appropriate antimicrobial therapy.