Aims. Day-case success rates after primary total hip arthroplasty (THA), total knee arthroplasty (TKA), and medial unicompartmental knee arthroplasty (mUKA) may vary, and detailed data are needed on causes of not being discharged. The aim of this study was to analyze the association between surgical procedure type and successful day-case surgery, and to analyze causes of not being discharged on the day of surgery when eligible and scheduled for day-case THA, TKA, and mUKA. Methods. A multicentre, prospective consecutive cohort study was carried out from September 2022 to August 2023. Patients were screened for day-case eligibility using well defined inclusion and exclusion criteria, and discharged when fulfilling predetermined discharge criteria. Day-case eligible patients were scheduled for surgery with intended start of surgery before 1.00 pm. Results. Of 6,142 primary hip and knee arthroplasties, eligibility rates for day-case surgery were 34% for THA (95% CI 32% to 36%), 34% for TKA (95% CI 32% to 36%), and 52% for mUKA (95% CI 49% to 55%). Surgery before 1.00 pm was achieved in 85% of eligible patients. The day-case success rate among patients with surgery before 1.00 pm was 59% (95% CI 55% to 62%) for THA, 61% (95% CI 57% to 65%) for TKA, and 72% (95% CI 68% to 76%) for mUKA. Overall day-case success rates (eligible and non-eligible) were 19% (95% CI 17% to 20%) for THA, 20% (95% CI 18% to 21%) for TKA, and 42% (95% CI 39% to 45%) for mUKA. Adjusted analysis confirmed higher day-case success in eligible mUKA patients (odds ratio 1.9 (1.6 to 2.3)) compared to TKA and THA patients. Primary causes for day-case failure were mobilization issues (9% to 12% between procedures), prolonged spinal anaesthesia (4% to 9%), and postoperative nausea and vomiting (PONV) (4% to 14%). Conclusion. THA and TKA patients showed comparable eligibility (34%) with similar day-case success rates (59 to 61%), whereas mUKA patients demonstrated higher eligibility (52%) and day-case success (72%). Mobilization issues, prolonged spinal anaesthesia, and PONV were the most frequent causes for not being discharged

Aims. This study aims to evaluate the impact of metabolic syndrome in the setting of obesity on in-hospital outcomes and resource use after total joint replacement (TJR). Methods. A retrospective analysis was conducted using the National Inpatient Sample from 2006 to the third quarter of 2015. Discharges representing patients aged 40 years and older with obesity (BMI > 30 kg/m. 2. ) who underwent primary TJR were included. Patients were stratified into two groups with and without metabolic syndrome. The inverse probability of treatment weighting (IPTW) method was used to balance covariates. Results. The obese cohort with metabolic syndrome was significantly older, more likely to be female, had higher rates of Medicare insurance, and more likely to be non-Hispanic Black than the obese cohort without metabolic syndrome. In the unweighted analysis, patients with obesity and metabolic syndrome were more likely to experience cardiac, gastrointestinal, genitourinary, and postoperative anemia complications, had a longer length of stay, and were less likely to be discharged home compared to obese patients without metabolic syndrome. After adjusting for covariates using IPTW, patients with obesity and metabolic syndrome were more likely to experience postoperative anemia complications only and had lower rates of home discharge, but there were no significant differences in any other complication variables or length of stay. Conclusion. Given the variability of metabolic health in obesity, the development of tailored perioperative protocols and recommendations acknowledging this variability in metabolic health in obese patients would ultimately potentially benefit patients and improve outcomes of TJR. Cite this article: Bone Jt Open 2024;5(10):837–843

Aims

The outcomes of patients with unexpected positive cultures (UPCs) during revision total hip arthroplasty (THA) and total knee arthroplasty (TKA) remain unknown. The objectives of this study were to establish the prevalence and infection-free implant survival in UPCs during presumed aseptic single-stage revision THA and TKA at mid-term follow-up.

Aims. Our aim was to estimate the total costs of all hospitalizations for treating periprosthetic joint infection (PJI) by main management strategy within 24 months post-diagnosis using activity-based costing. Additionally, we investigated the influence of individual PJI treatment pathways on hospital costs within the first 24 months. Methods. Using admission and procedure data from a prospective observational cohort in Australia and New Zealand, Australian Refined Diagnosis Related Groups were assigned to each admitted patient episode of care for activity-based costing estimates of 273 hip PJI patients and 377 knee PJI patients. Costs were aggregated at 24 months post-diagnosis, and are presented in Australian dollars. Results. The mean cost per hip and knee PJI patient was $64,585 (SD $53,550). Single-stage revision mean costs were $67,029 (SD $47,116) and $80,063 (SD $42,438) for hip and knee, respectively. Two-stage revision costs were $113,226 (SD $66,724) and $122,425 (SD $60,874) for hip and knee, respectively. Debridement, antibiotics, and implant retention in hips and knees mean costs were $53,537 (SD$ 39,342) and $48,463 (SD $33,179), respectively. Suppressive antibiotic therapy without surgical management mean costs were $20,296 (SD $8,875) for hip patients and $16,481 (SD $6,712) for knee patients. Hip patients had 16 different treatment pathways and knee patients had 18 treatment pathways. Additional treatment, episodes of care, and length of stay contributed to substantially increased costs up to a maximum of $369,948. Conclusion. Treating PJI incurs a substantial cost burden, which is substantially influenced by management strategy. With an annual PJI incidence of 3,900, the cost burden would be in excess of $250 million to the Australian healthcare system. Treatment pathways with additional surgery, more episodes of care, and a longer length of stay substantially increase the associated hospital costs. Prospectively monitoring individual patient treatment pathways beyond initial management is important when quantifying PJI treatment cost. Our study highlights the importance of optimizing initial surgical treatment, and informs treating hospitals of the resources required to provide care for PJI patients. Cite this article: Bone Joint J 2024;106-B(10):1084–1092

Aims. The influence of metabolic syndrome (MetS) on the outcome after hip and knee arthroplasty is debated. We aimed to investigate the change in patient-reported outcome measure (PROM) scores after hip and knee arthroplasty, comparing patients with and without MetS. Methods. From 1 May 2017 to 30 November 2019, a prospective cohort of 2,586 patients undergoing elective unilateral hip and knee arthroplasty was established in Denmark. Data from national registries and a local database were used to determine the presence of MetS. Patients’ scores on Oxford Hip Score (OHS) or Oxford Knee Score (OKS), EuroQol five-dimension five-level questionnaire (EQ-5D-5L), University of California, Los Angeles (UCLA) Activity Scale, and Forgotten Joint Score (FJS) at baseline, three, 12, and 24 months after surgery were collected. Primary outcome was the difference between groups from baseline to 12 months in OHS and OKS. Secondary outcomes were scores of OHS and OKS at three and 24 months and EQ-5D-5L, UCLA Activity Scale, and FJS at three, 12, and 24 months after surgery. Generalized linear mixed model was applied, adjusting for age, sex, Charlson Comorbidity Index, and smoking to present marginal mean and associated 95% CIs. Results. A total of 62.3% (1,611/2,586) of the cohort met the criteria for MetS. Both groups showed similar increase in mean OHS (MetS group 22.5 (95% CI 21.8 to 23.1), non-MetS group 22.1 (21.3 to 22.8); p = 0.477) and mean OKS (MetS group 18.0 (17.4 to 18.6), non-MetS group 17.8 (17.0 to 18.7); p = 0.722) at 12 months' follow-up. Between groups, similar improvements were seen for OHS and OKS at three and 24 months postoperatively and for the mean EQ-5D-5L, EuroQol-visual analogue scale (EQ-VAS), UCLA Activity Scale, and FJS at every timepoint. Conclusion. Patients meeting the criteria for MetS obtain the same improvement in PROM scores as individuals without MetS up to 24 months after hip and knee arthroplasty. This is important for the clinician to take into account when assessing and advising patients with MetS. Cite this article: Bone Joint J 2024;106-B(10):1074–1083

Aims

The aims of this study were to: 1) report on a cohort of skeletally mature patients with native hip and knee septic arthritis over a 14-year period; 2) to determine the rate of joint failure in patients who had experienced an episode of hip or knee septic arthritis; and 3) to assess the outcome following septic arthritis relative to the infecting organism, whether those patients infected by Staphylococcus aureus would be more likely to have adverse outcomes than those infected by other organisms.

Aims. We investigated the efficacy and safety profile of commonly used venous thromboembolism (VTE) prophylaxis agents following hip and knee arthroplasty. Methods. A systematic search of PubMed, Embase, Cochrane Library, Web of Science, and OrthoSearch was performed. Prophylaxis agents investigated were aspirin (< 325 mg and ≥ 325 mg daily), enoxaparin, dalteparin, fondaparinux, unfractionated heparin, warfarin, rivaroxaban, apixaban, and dabigatran. The primary efficacy outcome of interest was the risk of VTE, whereas the primary safety outcomes of interest were the risk of major bleeding events (MBE) and wound complications (WC). VTE was defined as the confirmed diagnosis of any deep vein thrombosis and/or pulmonary embolism. Network meta-analysis combining direct and indirect evidence was performed. Cluster rank analysis using the surface under cumulative ranking (SUCRA) was applied to compare each intervention group, weighing safety and efficacy outcomes. Results. Of 86 studies eligible studies, cluster rank analysis showed that aspirin < 325 mg daily (SUCRA-VTE 89.3%; SUCRA-MBE 75.3%; SUCRA-WC 71.1%), enoxaparin (SUCRA-VTE 55.7%; SUCRA-MBE 49.8%; SUCRA-WC 45.2%), and dabigatran (SUCRA-VTE 44.9%; SUCRA-MBE 52.0%; SUCRA-WC 41.9%) have an overall satisfactory efficacy and safety profile. Conclusion. We recommend the use of either aspirin < 325 mg daily, enoxaparin, or dabigatran for VTE prophylaxis following hip and knee arthroplasty. Cite this article: Bone Joint J 2024;106-B(9):924–934

Aims. The purpose of this study was to determine the association between prior sleeve gastrectomy in patients undergoing primary total hip and knee arthroplasty, and 90-day complications, incidence of revision arthroplasty, and patient-reported outcome scores at final follow-up. Methods. This is a retrospective, single-centre analysis. Patients undergoing primary hip or knee arthroplasty with a prior sleeve gastrectomy were eligible for inclusion (n = 80 patients). A morbidly obese control group was established from the same institutional registry using a 1:2 match, for cases:controls with arthroplasty based on propensity score using age, sex, pre-sleeve gastrectomy BMI, Current Procedural Terminology code to identify anatomical location, and presurgical haemoglobin A1C. Outcomes included 90-day complications, incidence of revision arthroplasty, and patient-reported outcome scores at final follow-up. Multivariable logistic regressions evaluated associations of underlying preoperative demographic and treatment characteristics with outcomes. Results. Complications within 90 days of surgery were increased in the sleeve gastrectomy group relative to the obese control group after controlling for underlying preoperative demographic characteristics (odds ratio (OR) 4.00 (95% CI 1.14 to 13.9); p = 0.030). Postoperative revisions were similar in the sleeve gastrectomy group relative to the obese control group after controlling for underlying preoperative demographic characteristics (OR 17.8 (95% CI 0.64 to 494.3); p = 0.090). Patient-Reported Outcomes Measurement Information System (PROMIS) depression decreased by a greater amount from pre- to postoperative in the obese controls relative to the sleeve gastrectomy group (OR 4.04 (95% CI 0.06 to 8.02); p = 0.047). PROMIS pain interference and physical function change from pre- to postoperative was not associated with sleeve gastrectomy status. Conclusion. We found a higher rate of complications at 90 days in patients who underwent sleeve gastrectomy prior to primary hip or knee arthroplasty relative to a matched, obese control population. Prosthetic revision rates were similar between the two groups, while improvements in PROMIS depression scores were larger in the obese cohort. This study suggests that sleeve gastrectomy to achieve preoperative weight loss prior to arthroplasty surgery may not mitigate early complication risks in obese patient populations. Cite this article: Bone Joint J 2024;106-B(9):935–941

Aims

The COVID-19 pandemic has disrupted the provision of arthroplasty services in England, Wales, and Northern Ireland. This study aimed to quantify the backlog, analyze national trends, and predict time to recovery.

Aims

The aim of this meta-analysis was to determine the pooled incidence of postoperative urinary retention (POUR) following total hip and knee arthroplasty (total joint replacement (TJR)) and to evaluate the risk factors and complications associated with POUR.

Aims. To examine whether natural language processing (NLP) using a clinically based large language model (LLM) could be used to predict patient selection for total hip or total knee arthroplasty (THA/TKA) from routinely available free-text radiology reports. Methods. Data pre-processing and analyses were conducted according to the Artificial intelligence to Revolutionize the patient Care pathway in Hip and knEe aRthroplastY (ARCHERY) project protocol. This included use of de-identified Scottish regional clinical data of patients referred for consideration of THA/TKA, held in a secure data environment designed for artificial intelligence (AI) inference. Only preoperative radiology reports were included. NLP algorithms were based on the freely available GatorTron model, a LLM trained on over 82 billion words of de-identified clinical text. Two inference tasks were performed: assessment after model-fine tuning (50 Epochs and three cycles of k-fold cross validation), and external validation. Results. For THA, there were 5,558 patient radiology reports included, of which 4,137 were used for model training and testing, and 1,421 for external validation. Following training, model performance demonstrated average (mean across three folds) accuracy, F1 score, and area under the receiver operating curve (AUROC) values of 0.850 (95% confidence interval (CI) 0.833 to 0.867), 0.813 (95% CI 0.785 to 0.841), and 0.847 (95% CI 0.822 to 0.872), respectively. For TKA, 7,457 patient radiology reports were included, with 3,478 used for model training and testing, and 3,152 for external validation. Performance metrics included accuracy, F1 score, and AUROC values of 0.757 (95% CI 0.702 to 0.811), 0.543 (95% CI 0.479 to 0.607), and 0.717 (95% CI 0.657 to 0.778) respectively. There was a notable deterioration in performance on external validation in both cohorts. Conclusion. The use of routinely available preoperative radiology reports provides promising potential to help screen suitable candidates for THA, but not for TKA. The external validation results demonstrate the importance of further model testing and training when confronted with new clinical cohorts. Cite this article: Bone Joint J 2024;106-B(7):688–695

Aims. The aim of this study was to evaluate the healthcare costs and benefits of enoxaparin compared to aspirin in the prevention of symptomatic venous thromboembolism (VTE) after total hip arthroplasty (THA) or total knee arthroplasty (TKA) using data from the CRISTAL trial. Methods. This trial-based economic analysis reports value for money as incremental cost per quality-adjusted life-year (QALY) gained in 2022 Australian dollars, compared to a single threshold value of AUD$70,000 per QALY. Event costs were estimated based on occurrence of VTEs and bleeds, and on published guidelines for treatment. Unit costs were taken from Australian sources. QALYs were estimated using CRISTAL six-month follow-up data. Sensitivity analyses are presented that vary the cost of VTE treatment, and extend the analyses to two years. Results. The CRISTAL trial found that enoxaparin was more effective than aspirin in preventing symptomatic VTE within 90 days of THA or TKA (risk difference 1.97% (95% confidence interval (CI) 0.54% to 3.41%; p = 0.007)). The additional cost after a THA or TKA was AUD$83 (95% CI 68 to 97) for enoxaparin, and enoxaparin resulted in an additional 0.002 QALYs (95% CI -0.002 to 0.005). Incremental cost per QALY gained was AUD$50,567 (95% CI 15,513, dominated) for enoxaparin. We can be 60% confident that the incremental cost per QALY does not exceed the willingness-to-pay threshold of AUD$70,000. Increasing the cost of VTE treatment and extension of costs and consequences to two years suggested greater confidence that enoxaparin is good value for money (70% and 63% confidence, respectively). Conclusion. This analysis provides strong evidence that enoxaparin thromboprophylaxis following THA or TKA reduced VTEs, but weak evidence of net economic benefits over aspirin. If the value of avoiding VTEs is high, and there is a strong likelihood of VTE-related health impairments, we can be more confident that enoxaparin is cost-effective compared to aspirin. Cite this article: Bone Joint J 2024;106-B(6):589–595

Aims. The overall aim of this study was to determine the impact of deprivation with regard to quality of life, demographics, joint-specific function, attendances for unscheduled care, opioid and antidepressant use, having surgery elsewhere, and waiting times for surgery on patients awaiting total hip arthroplasty (THA) and total knee arthroplasty (TKA). Methods. Postal surveys were sent to 1,001 patients on the waiting list for THA or TKA in a single Northern Ireland NHS Trust, which consisted of the EuroQol five-dimension five-level questionnaire (EQ-5D-5L), visual analogue scores (EQ-VAS), and Oxford Hip and Knee Scores. Electronic records determined prescriptions since addition to the waiting list and out-of-hour GP and emergency department attendances. Deprivation quintiles were determined by the Northern Ireland Multiple Deprivation Measure 2017 using postcodes of home addresses. Results. Overall, 707 postal surveys were returned, of which 277 (39.2%) reported negative “worse than death” EQ-5D scores and 219 (21.9%) reported the consumption of strong opioids. Those from the least deprived quintile 5 had a significantly better EQ-5D index (median 0.223 (interquartile range (IQR) -0.080 to 0.503) compared to those in the most deprived quintiles 1 (median 0.049 (IQR -0.199 to 0.242), p = 0.004), 2 (median 0.076 (IQR -0.160 to 0.277; p = 0.010), and 3 (median 0.076 (IQR-0.153 to 0.301; p = 0.010). Opioid use was significantly greater in the most deprived quintile 1 compared to all other quintiles (45/146 (30.8%) vs 174/809 (21.5%); odds ratio 1.74 (95% confidence interval 1.18 to 2.57; p = 0.005). Conclusion. More deprived patients have worse health-related quality of life and greater opioid use while waiting for THA and TKA than more affluent patients. For patients awaiting surgery, more information and alternative treatment options should be available. Cite this article: Bone Jt Open 2024;5(5):444–451

Aims. Periprosthetic joint infection (PJI) demonstrates the most feared complication after total joint replacement (TJR). The current work analyzes the demographic, comorbidity, and complication profiles of all patients who had in-hospital treatment due to PJI. Furthermore, it aims to evaluate the in-hospital mortality of patients with PJI and analyze possible risk factors in terms of secondary diagnosis, diagnostic procedures, and complications. Methods. In a retrospective, cross-sectional study design, we gathered all patients with PJI (International Classification of Diseases (ICD)-10 code: T84.5) and resulting in-hospital treatment in Germany between 1 January 2019 and 31 December 2022. Data were provided by the Institute for the Hospital Remuneration System in Germany. Demographic data, in-hospital deaths, need for intensive care therapy, secondary diagnosis, complications, and use of diagnostic instruments were assessed. Odds ratios (ORs) with 95% confidence intervals (CIs) for in-hospital mortality were calculated. Results. A total of 52,286 patients were included, of whom 1,804 (3.5%) died. Hypertension, diabetes mellitus, and obesity, the most frequent comorbidities, were not associated with higher in-hospital mortality. Cardiac diseases as atrial fibrillation, cardiac pacemaker, or three-vessel coronary heart disease showed the highest risk for in-hospital mortality. Postoperative anaemia occurred in two-thirds of patients and showed an increased in-hospital mortality (OR 1.72; p < 0.001). Severe complications, such as organ failure, systemic inflammatory response syndrome (SIRS), or septic shock syndrome showed by far the highest association with in-hospital mortality (OR 39.20; 95% CI 33.07 to 46.46; p < 0.001). Conclusion. These findings highlight the menace coming from PJI. It can culminate in multi-organ failure, SIRS, or septic shock syndrome, along with very high rates of in-hospital mortality, thereby highlighting the vulnerability of these patients. Particular attention should be paid to patients with cardiac comorbidities such as atrial fibrillation or three-vessel coronary heart disease. Risk factors should be optimized preoperatively, anticoagulant therapy stopped and restarted on time, and sufficient patient blood management should be emphasized. Cite this article: Bone Jt Open 2024;5(4):367–373

Aims. Histology is widely used for diagnosis of persistent infection during reimplantation in two-stage revision hip and knee arthroplasty, although data on its utility remain scarce. Therefore, this study aims to assess the predictive value of permanent sections at reimplantation in relation to reinfection risk, and to compare results of permanent and frozen sections. Methods. We retrospectively collected data from 226 patients (90 hips, 136 knees) with periprosthetic joint infection who underwent two-stage revision between August 2011 and September 2021, with a minimum follow-up of one year. Histology was assessed via the SLIM classification. First, we analyzed whether patients with positive permanent sections at reimplantation had higher reinfection rates than patients with negative histology. Further, we compared permanent and frozen section results, and assessed the influence of anatomical regions (knee versus hip), low- versus high-grade infections, as well as first revision versus multiple prior revisions on the histological result at reimplantation. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), chi-squared tests, and Kaplan-Meier estimates were calculated. Results. Overall, the reinfection rate was 18%. A total of 14 out of 82 patients (17%) with positive permanent sections at reimplantation experienced reinfection, compared to 26 of 144 patients (18%) with negative results (p = 0.996). Neither permanent sections nor fresh frozen sections were significantly associated with reinfection, with a sensitivity of 0.35, specificity of 0.63, PPV of 0.17, NPV of 0.81, and accuracy of 58%. Histology was not significantly associated with reinfection or survival time for any of the analyzed sub-groups. Permanent and frozen section results were in agreement for 91% of cases. Conclusion. Permanent and fresh frozen sections at reimplantation in two-stage revision do not serve as a reliable predictor for reinfection. Cite this article: Bone Joint J 2024;106-B(4):372–379

Aims

Breast cancer survivors have known risk factors that might influence the results of total hip arthroplasty (THA) or total knee arthroplasty (TKA). This study evaluated clinical outcomes of patients with breast cancer history after primary THA and TKA.

Aims

While internet search engines have been the primary information source for patients’ questions, artificial intelligence large language models like ChatGPT are trending towards becoming the new primary source. The purpose of this study was to determine if ChatGPT can answer patient questions about total hip (THA) and knee arthroplasty (TKA) with consistent accuracy, comprehensiveness, and easy readability.

Aims

The primary aim was to assess change in health-related quality of life (HRQoL) of patients as they waited from six to 12 months for a total hip (THA) or total or partial knee arthroplasty (KA). Secondary aims were to assess change in joint-specific function, mental health, quality of sleep, number living in a state worse than death (WTD), wellbeing, and patient satisfaction with their healthcare.

Aims. It is unclear whether mortality outcomes differ for patients undergoing total hip arthroplasty (THA) or total knee arthroplasty (TKA) surgery who are readmitted to the index hospital where their surgery was performed, or to another hospital. Methods. We analyzed linked hospital and death records for residents of New South Wales, Australia, aged ≥ 18 years who had an emergency readmission within 90 days following THA or TKA surgery between 2003 and 2022. Multivariable modelling was used to identify factors associated with non-index readmission and to evaluate associations of readmission destination (non-index vs index) with 90-day and one-year mortality. Results. Of 394,248 joint arthroplasty patients (THA = 149,456; TKA = 244,792), 9.5% (n = 37,431) were readmitted within 90 days, and 53.7% of these were admitted to a non-index hospital. Non-index readmission was more prevalent among patients who underwent surgery in private hospitals (60%). Patients who were readmitted for non-orthopaedic conditions (62.8%), were more likely to return to a non-index hospital compared to those readmitted for orthopaedic complications (39.5%). Factors associated with non-index readmission included older age, higher socioeconomic status, private health insurance, and residence in a rural or remote area. Non-index readmission was significantly associated with 90 day (adjusted odds ratio (aOR) 1.69; 95% confidence interval (CI) 1.39 to 2.05) and one-year mortality (aOR 1.31; 95% CI 1.16 to 1.47). Associations between non-index readmission and mortality were similar for patients readmitted with orthopaedic and non-orthopaedic complications (90-day mortality aOR 1.61; 95% CI 0.98 to 2.64, and aOR 1.67; 95% CI 1.35 to 2.06, respectively). Conclusion. Non-index readmission was associated with increased mortality, irrespective of whether the readmission was for orthopaedic complications or other conditions. Cite this article: Bone Jt Open 2024;5(1):60–68

Aims. A higher failure rate has been reported in haematogenous periprosthetic joint infection (PJI) compared to non-haematogenous PJI. The reason for this difference is unknown. We investigated the outcome of haematogenous and non-haematogenous PJI to analyze the risk factors for failure in both groups of patients. Methods. Episodes of knee or hip PJI (defined by the European Bone and Joint Infection Society criteria) treated at our institution between January 2015 and October 2020 were included in a retrospective PJI cohort. Episodes with a follow-up of > one year were stratified by route of infection into haematogenous and non-haematogenous PJI. Probability of failure-free survival was estimated using the Kaplan-Meier method, and compared between groups using log-rank test. Univariate and multivariate analysis was applied to assess risk factors for failure. Results. A total of 305 PJI episodes (174 hips, 131 knees) were allocated to the haematogenous (n = 146) or the non-haematogenous group (n = 159). Among monomicrobial infections, Staphylococcus aureus was the dominant pathogen in haematogenous PJI (76/140, 54%) and coagulase-negative staphylococci in non-haematogenous PJI (57/133, 43%). In both groups, multi-stage exchange (n = 55 (38%) in haematogenous and n = 73 (46%) in non-haematogenous PJI) and prosthesis retention (n = 70 (48%) in haematogenous and n = 48 (30%) in non-haematogenous PJI) were the most common surgical strategies. Median duration of antimicrobial treatment was 13.5 weeks (range, 0.5 to 218 weeks) and similar in both groups. After six years of follow-up, the probability of failure-free survival was significantly lower in haematogenous compared to non-haematogenous PJI (55% vs 74%; p = 0.021). Infection-related mortality was significantly higher in haematogenous than non-haematogenous PJI (7% vs 0% episodes; p = 0.001). Pathogenesis of failure was similar in both groups. Retention of the prosthesis was the only independent risk factor for failure in multivariate analysis in both groups. Conclusion. Treatment failure was significantly higher in haematogenous compared to non-haematogenous PJI. Retention of the prosthesis was the only independent risk factor for failure in both groups. Cite this article: Bone Joint J 2023;105-B(12):1294–1302