Arthroplasty and Modifiable Risk Factors

Hip and knee total joint arthroplasty allow effective treatment of disability and pain due to joint destruction, and are one of the most frequently performed surgery worldwide. Surgical-site infections remain a serious complication of total joint arthroplasty, occurring in 1-7% of cases. The most severe form of surgical site infection, prosthetic joint infection is a devastating complication associated with severe morbidity, increased mortality and huge costs. There are several underlying medical conditions that have been identified as independent risk factors for development of prosthetic joint infection that might be improved before operation, called modifiable risk factors, namely pre-operative anemia, obesity, diabetes, poor nutritional status, smoking and physical activity.

The hypothesis of the study is that by utilizing the waiting time for operation (6 -12 months) to improve the state of known modifiable risk factors the frequency of prosthetic joint infection may be decreased. This optimization was established with co-operation between hospital and primary care.

Therefore a prospective non-randomized quality control study before and after an intervention regarding preoperative preparation for total joint arthroplasty of either hip or knee, comparing patient characteristics and outcome was planned.

Control Arm: Inclusion of 'usual care' patients started 20/8 2018 one week prior to surgery at the anesthesia preoperative assessment outpatient clinic. At that point in time there were 561 patients already on a waiting list for either hip or knee arthroplastic surgery. The next four months to 31/12 2018 66 more patients were added to the study. After 1/1 2019 (when collection to the interventional part had started) further 111 patients were added (15% of the 'usual care' group). These were patients where it had been missed to offer or had declined participation in the interventional arm of the study at the orthopedic outpatient clinic and when discovered at the preoperative clinic one week prior to surgery were offered participation in the 'usual care' arm as they had received traditional preoperative preparation. Operation of these patients started at 27/8 2018 and lasted to 7/9 2020 and consists of total of 738 individuals.

Interventional arm: Inclusion into the interventional arm of the study occurred between 2/1 2019 to 30/1 2021, 6 -12 months prior to surgery at the orthopedic outpatient clinic, ensuring an appointment with the patient general physician within 3 weeks afterwards to further analyze results of blood test and assess if further tests, treatments or consultation of other specialists would be preferred to optimize their condition. Patients at extremes of BMI also had a nutritional advice at the hospital. Enlisted have been 1010 patients, operation of them started 25/3 2019 and to date 710 patients have been operated on, but due to delays caused by COVID-19 (SARS-CoV-2) surgery of them is still ongoing.

Outcome variables: Information was gathered regarding age, gender, smoking, comorbidities, type of surgery, type of anesthesia, ASA classification, weight, physical activity, laboratory values (Hb, Glu, HbA1c, Albumin. Lymphocytes, D-vitamin), blood transfusions and length of stay. Postoperative follow up happened first 3 weeks postoperatively at primary care for suture removal and then 6 weeks at the outpatient orthopedic clinic. Notes from these visits were studied to identify surgical site complications, including surgical site infection, prosthetic joint infection, surgical site drainage, bleeding, dehiscence, and hematoma and evaluated according to CDC's National Healthcare Safety Network and PJI diagnosis. Also, a two year follow will occur by analysis of patient journal in regards of complications of the arthroplasty.

Statistical methods. Patient demographics will be described as median (interquartile range) for continuous variables, and number (percentage) for categorical variables. Mean and standard deviation (SD) will be reported for normally distributed data. A univariate association between the presence of modifiable risk factors and the presence of postoperative surgical site infection will be assessed using chi-squared test for categorical variables and t-test for continuous variables. As this is a descriptive study and incidence of modifiable risk factors not known in the cohort no assessment of power was performed prior to analysis. To study if there is a difference between the groups a t-test or non-parametric Wilcoxon rank sum test or Hotelling's t will be used for continuous parameters, and chi-squared test or Fisher's exact test to compare categorical parameters. Poisson regression or negative binominal regression of zero inflated regression will be used to study if a difference in the frequency of complications has occurred after the intervention. These methods might be subjected to change if unexpected factors arise during analysis of data.