Prediction of Early Periprosthetic Joint Infection After Total Hip Arthroplasty

Prediction of Early Adverse Events After THA: A Comparison of Different Machine-Learning Strategies Based on 262,356 Observations From the Nordic Arthroplasty Register Association (NARA) Dataset

OBJECTIVE

Preoperative risk prediction models can support shared decision-making before total hip arthroplasties (THAs). Here, we compare different machine-learning (ML) approaches to predict the six-month risk of adverse events following primary THA to obtain accurate yet simple-to-use risk prediction models.

METHODS

We extracted data on primary THAs (N = 262,356) between 2010 and 2018 from the Nordic Arthroplasty Register Association dataset. We benchmarked a variety of ML algorithms in terms of the area under the receiver operating characteristic curve (AUROC) for predicting the risk of revision caused by periprosthetic joint infection (PJI), dislocation or periprosthetic fracture (PPF), and death. All models were internally validated against a randomly selected test cohort (one-third of the data) that was not used for training the models.

RESULTS

The incidences of revisions because of PJI, dislocation, and PPF were 0.8%, 0.4%, and 0.3%, respectively, and the incidence of death was 1.2%. Overall, Lasso regression with stable iterative variable selection (SIVS) produced models using only four to five input variables but with AUROC comparable to more complex models using all 32 variables available. The SIVS-based Lasso models based on age, sex, preoperative diagnosis, bearing couple, fixation, and surgical approach predicted the risk of revisions caused by PJI, dislocations, and PPF, as well as death, with AUROCs of 0.61, 0.67, 0.76, and 0.86, respectively.

CONCLUSION

Our study demonstrates that satisfactory predictive potential for adverse events following THA can be reached with parsimonious modeling strategies. The SIVS-based Lasso models may serve as simple-to-use tools for clinical risk assessment in the future.

Risk equations for prosthetic joint infections (PJIs) in UK: a retrospective study using the Clinical Practice Research Datalink (CPRD) AURUM and GOLD databases

BACKGROUND

Prosthetic joint infections (PJIs) are a serious negative outcome of arthroplasty with incidence of about 1%. Risk of PJI could depend on local treatment policies and guidelines; no UK-specific risk scoring is currently available.

OBJECTIVE

To determine a risk quantification model for the development of PJI using electronic health records.

DESIGN

Records in Clinical Practice Research Datalink (CPRD) GOLD and AURUM of patients undergoing hip or knee arthroplasty between January 2007 and December 2014, with linkage to Hospital Episode Statistics and Office of National Statistics, were obtained. Cohorts' characteristics and risk equations through parametric models were developed and compared between the two databases. Pooled cohort risk equations were determined for the UK population and simplified through stepwise selection.

RESULTS

After applying the inclusion/exclusion criteria, 174 905 joints (1021 developed PJI) were identified in CPRD AURUM and 48 419 joints (228 developed PJI) in CPRD GOLD. Patients undergoing hip or knee arthroplasty in both databases exhibited different sociodemographic characteristics and medical/drug history. However, the quantification of the impact of such covariates (coefficients of parametric models fitted to the survival curves) on the risk of PJI between the two cohorts was not statistically significant. The log-normal model fitted to the pooled cohorts after stepwise selection had a C-statistic >0.7.

CONCLUSIONS

The risk prediction tool developed here could help prevent PJI through identifying modifiable risk factors pre-surgery and identifying the patients most likely to benefit from close monitoring/preventive actions. As derived from the UK population, such tool will help the National Health Service reduce the impact of PJI on its resources and patient lives.

Preoperative dental screening can reduce periprosthetic infections of hip and knee endoprostheses in the first month after surgery: results of a cohort study

PURPOSE

The oral cavity and, in particular, potential oral foci might pose a risk of periprosthetic joint infection (PJI). The aim of this cohort study was to determine whether practical preoperative dental screening would reduce the prevalence of early PJI in the first month after surgery.

METHODS

Patients attending a specialized endoprosthesis implantation clinic between 2018 and 2022 were recruited. Two groups were examined. The test group consisted of patients attending the clinic between 2020 and 2022 and who were referred to their family dentist using a standardized form. The comparison group consisted of patients who were treated in the clinic between 2018 and 2020. They were not referred to their family dentist. The two groups were compared for the prevalence of PJI. Univariate analysis followed by multiple logistic regression was performed to confirm risk factors for PJI in this cohort.

RESULTS

2560 individuals (test group: 1227, comparison group: 1333) were included. The prevalence of PJI was significantly lower in the test group (0.8% vs. 1.8%, p = 0.04). Multiple logistic regression with PJI as the dependent variable showed that a dental referral was a strong predictor of a lower prevalence of PJI (OR: 0.43, CI95 0.205-0.917, p = 0.03). Male gender was also strongly associated with a higher frequency of PJI (OR: 2.68, CI95 1.32-5.42, p = 0.01). Age (OR: 1.06, CI95 1.01-1.10, p = 0.01) and BMI (OR: 1.11, CI95 1.05-1.17, p < 0.01) had little effect on the risk of PJI.

CONCLUSION

Dental referral using a standardized form can reduce the prevalence of early PJI. Accordingly, orthopedists and dentists should collaborate in this practical way.

Incidence of periprosthetic joint infection after primary total hip arthroplasty is underestimated: a synthesis of meta-analysis and bibliometric analysis

The American Musculoskeletal Society updated the diagnostic criteria for periprosthetic joint infection (PJI) in 2011 and 2018. However, the overall incidence of PJI since the introduction of these new standards has not been assessed. In order to fill this knowledge gap, a single-group meta-analysis was conducted using articles obtained from several databases, focusing on the incidence of PJI after primary total hip arthroplasty (THA). Our study revealed a significant difference in the incidence of PJI reported by different national or regional databases. Moreover, most cases of PJI were found to be underestimated. This highlights the crucial need for standardized diagnostic criteria and monitoring methods to accurately identify and track cases of PJI. Furthermore, a bibliometric analysis was conducted to provide a comprehensive overview of the current state of research on PJI after THA. This analysis explored the most productive countries, organizations, journals, and individuals in this research area. Additionally, it identified the research trends and hotspots of the last decade, highlighting the advancements and areas of focus in this field. By conducting these analyses, the study aims to contribute to the understanding of PJI after THA and provide valuable insights for clinicians, researchers, and policymakers involved in the management of this condition.

Preoperative Risk Factor Analysis and Dynamic Online Nomogram Development for Early Infections Following Primary Hip Arthroplasty in Geriatric Patients with Hip Fracture

Background

Hip arthroplasty is in increasing demand with the aging of the world population, and early infections, such as pneumonia, surgical site infection (SSI), and urinary tract infection (UTI), are uncommon but fatal complications following hip arthroplasty. This study aimed to identify preoperative risk factors independently associated with early infections following primary arthroplasty in geriatric hip fracture patients, and to develop a prediction nomogram.

Methods

Univariate and multivariate logistical analyses were performed to identify the independent risk factors for early infections, which were combined and transformed into a nomogram model. The prediction model was evaluated by using the area under the receiver operating characteristic curve (AUC), Hosmer-Lemeshow test, concordance index (C-index), 1000 bootstrap replications, decision curve analysis (DCA), and calibration curve.

Results

One thousand eighty-four eligible patients got included and 7 preoperative variables were identified to be independently associated with early infections, including heart disease (odds ratio (OR): 2.17; P: 0.026), cerebrovascular disease (OR: 2.25; P: 0.019), liver disease (OR: 8.99; P: <0.001), time to surgery (OR: 1.10; P: 0.012), hematocrit (<lower limit; OR: 3.72; P: 0.015), the platelet-to-mean platelet volume ratio (PMR; >44.52; OR: 2.73; P: 0.047), and high-sensitivity C-reactive protein (HCRP; >78.64mg/L; OR: 3.71; P: <0.001). For the nomogram model, AUC was 0.807 (95% confidence interval (CI): 0.742-0.873), the Hosmer-Lemeshow test demonstrated no overfitting (P = 0.522), and C-index was 0.807 (95% CI: 0.742-0.872) with corrected value of 0.784 after 1000 bootstrapping validations. Moreover, the calibration curve and DCA exhibited the tools' good prediction consistency and clinical practicability.

Conclusion

Heart disease, cerebrovascular disease, liver disease, time to surgery, hematocrit, PMR, and HCRP were significant preoperative predictors for early infections following primary arthroplasty in elderly hip fracture patients, and the converted nomogram model had strong discriminatory ability and translatability to clinical application.