Influence of Disparities and Risk Factors on Periprosthetic Joint Injection Rates Following Joint Arthroplasties

Hi, my name is Caitlin Donnelly. I'm a rising 4th year medical student at LSU New Orleans, and this project is on the influence of disparities and Risk factors on Periprosthetic joint infection rates following joint arthroplasties. I'd like to start by thanking my co-authors for their contributions.

PGIs occur in approximately 0.5 to 2% of cases, and the mortality rate is approximately 4% 1 year after PGI and increases yearly thereafter. Some of the well-documented Risk factors for PGIs include age, gender, obesity, rheumatoid Arthritis, and diabetes. However, there are inconsistencies in literature for risk associated with alcohol use and liver disease.

Furthermore, patient demographics such as race andur status are frequently unaccounted for in Research. Therefore, our aim was to analyze Risk factors for PJIs following THAs and TKAs and to highlight possible disparities.

To do so, we collected data from a large regional database in Louisiana and Texas called ReachNet, and we extracted data on Primary THAs and TKAs from the years 2014 to 2019. Our exclusions included patients whose first surgical code was a revision THA or TKA. They had a prior PJI before their date of Surgery, and if they had a Fracture code within 30 days of their Surgery to exclude for non-elective cases. We also excluded patients with missing data or no follow-up within one year.

This resulted in a total of 15,979 patients in our cohort.

Data was collected using diagnostic codes for PJIs, and we also collected data on race, ethnicity, alcohol use, smoking, insurance, BMI, age, gender, length of stay, procedure year, liver disease, avascular necrosis, and CCI.

This is the multivariable logistics regression for PJIs within one year of TJA in our cohort.

The significant findings were higher rate of PJIs and THAs versus TKAs, a lower rate of PGIs in black versus white patients, a higher risk of PGIs with alcohol use, a higher rate in male gender, a lower rate of PGIs and private insurance versus public insurance.

A lower risk of PGIs as age increased, but a higher rate of PJIs as CCI increased, and a higher rate of PJIs with BMI over 40.

The male gender increased DCI and BMI over 40 had a significantly increased rate of PGIs consistent with the literature that exists today. THAs having an increased risk compared to TKAs is opposite of what the literature says, But of note in our cohort, 12.5% of avascular necrosis was in our THA patients compared to 0.6% in our TKA patients.

Alcohol use was found to have the greatest influence compared to Other Risk factors analyzed in our cohort. It had an adjusted odds ratio of 2.4. There's some inconsistencies in the literature of alcohol use is an independent risk factor for PJIs. However, there is Research showing that alcohol impairs the phagocytic function and leads to an increased risk of overall infections.

Variables that did not independently affect the rate of PGIs were smoking, year of TJA, length of hospital stay, ethnicity, and liver disease. Interestingly, liver disease and smoking were significant before adjustment. Smoking is something that the literature has shown to be significant before, but we did not find in our cohort, and liver disease is something that has been inconsistent in the literature.

And of note in our cohort, 29.4% of patients with alcohol use had comorbid liver disease, and we feel further studies are needed on liver disease and alcohol use to determine their risk for PJIs and the amount of alcohol they drink, how long they've been drinking alcohol, and the degree of liver damage, and what's the underlying cause of liver damage, to decide if they are independently Risk factors for PJIs or not.

There was decreased risk of PGIs in our cohort with private insurance, older age, and black race.

Finding a decreased risk of PGIs with private insurance is consistent with what the literature assessed today. However, older age and black grace is inconsistent.

One of the possible reasonings is that in our cohort, we had 80.4% white patients and 17.5% black patients compared to 79% and 63% white, and 13% and 33% black in Texas and Louisiana. So not only did we have a lower proportion of black patients compared to white patients, but this is also something that happens very rarely, so there was a decreased power in the black population.

Another theory is the selection bias could have had an effect on the decreased risk of PJIs with older age and race.

So in conclusion, our project further supports literature for increased risk of PGIs in patients with BMI over 40, male gender, public insurance, and increased CCI. Our most noteworthy finding is a substantial increase associated with alcohol use.

And we suggest further studies to clarify and expand on the effects of alcohol use and liver disease. Thank you for listening.