The aim of our study is to identify the influence of preoperative pain (NRS pain score and EQ5D pain score) and higher levels of anxiety and depression (EQ5D anxiety/depression score) and their interaction before primary total knee arthroplasty on revision surgery for unexplained symptoms after primary surgery.

Knee arthroplasty can decrease pain and improve function in people with advanced osteoarthritis of the knee. It is a common procedure; over 25,000 primary knee arthroplasties are performed in the Netherlands each year[2]. A primary knee arthroplasty is defined as the first implantation of a prosthesis in the knee. Revision surgery is defined as any exchange (placement, replacement, or removal) or addition of 1 or more components of the prosthesis (e.g. patella resurfacing)[3]. About 12% of knee arthroplasties are revised within 10 years.

It is known that 1 in 5 patients rate themselves categorically dissatisfied after knee arthroplasty (the so-called unhappy knee). It seems sensible that the percentage of patients that has some dissatisfaction with their knee arthroplasty is likely much larger. After knee arthroplasty, some patients receive revision surgery based on clear reasons such as periprosthetic fracture and patellar dislocation, while in some patients that are dissatisfied revision surgery might be considered for less clear reasons such as a small technical issue, for a perceived technical issue, or with the idea that there is a low grade infection. Understanding the preoperative factors of postoperative pain and functional impairment, leading to dissatisfaction, may help to inform patients considering primary knee arthroplasty about the risk of revision surgery. Although pain and function of patients with higher preoperative pain scores before primary knee arthroplasty will improve as much as the scores of patients with lower preoperative pain scores, their final postoperative pain levels and function will remain inferior compared to patients with lower preoperative pain scores. Higher levels of anxiety and symptoms of depression before knee arthroplasty have also been related to worse postoperative patient reported outcome measures after surgery. Furthermore, the experience of pain and symptoms of depression are closely related to each other and they may influence each other in a bidirectional way. Determining the influence of preoperative pain and psychological distress (anxiety/depression) and their interaction on postoperative outcome seems to be crucial because mental and social health opportunities might be underdiagnosed and undertreated and the indication for a revision surgery may be based on a misdiagnosis of perceived or actual pathophysiology or technical deficiency in patients without a clear technical or medical indication for revision.

The aim of this study is to identify the influence of the preoperative pain (NRS pain score and EQ5D pain score) and higher levels of anxiety and depression (EQ5D anxiety/depression score) and their association before primary total knee arthroplasty (TKA) on revision surgery for unexplained symptoms after primary surgery.

Inclusion Criteria:

• All patients undergoing primary knee arthroplasty (e.g., total, unicondylar and patellofemoral) reported in the Dutch Arthroplasty Register.
• Patients who filled out the NRS pain score, EQ-5D 3L or EQ-5D 5L pain score or EQ-5D anxiety/depression score.

Exclusion Criteria:

• Other types of (revision) arthroplasty (hip / shoulder / ankle)
• Patients who did not fill out the NRS pain score, EQ-5D (3L or 5L) pain score or EQ-5D (3L or 5L) anxiety/depression score

Patients undergoing primary knee arthroplasty (e.g., total, unicondylar and patellofemoral) reported in the Dutch Arthroplasty Register.

We will derive data of patients undergoing primary knee arthroplasty from the Dutch Arthroplasty Register, the "Landelijke Registratie Orthopedische Implantaten" (LROI).

We have submitted a research application to use data of the LROI for our study, which was judged by the "Wetenschappelijke Adviesraad" (WAR). After approval af the WAR, further agreements have been made between our research group and the LROI about the way in which the data were securely prepared. The specific dataset for our research question was prepared by the LROI and labeled by a specific research number (LROI2020-057 Poolman - Depressie en pijn bij KA). The LROI does not include names and addresses of patients. The citizen service number is encrypted by the LROI. The researchers will have no access to identifiable patient data.