Loss of joint functioning and severe joint pain are present in a strongly advanced state of osteoarthritis. It is the sole reason for hip and knee joint replacement in around 90% of cases. Although hip and knee replacements are routine surgeries nowadays, serious complications occur in 2% of patients. Up to 20% of patients experience persistent pain and are not satisfied with their implants whereby obese patients are more often affected than normal weight patients.
Weight loss surgery presents an effective way to lose weight for obese patients. It reduces the available space in the stomach or lowers the amount of energy intake of eaten food, depending on the type of procedure. A recent study demonstrated clinical improvement at one year check up in three-fourths of patients who underwent weight loss surgery with prior severe knee or hip pain. To date, no study has addressed the question whether weight loss surgery in obese patients can prevent first time hip or knee replacements. We will therefore investigate the association between weight loss surgery and first-time hip or knee replacements. Beforehand, we will check the records of weight loss surgery entries in a primary care database for its correctness by comparing them with hospital data.
Among a population of patients who are linked to HES data, we will first validate the records of bariatric surgery (BS) in the CPRD through a comparison with HES APC data by estimating sensitivity and specificity and also describe BS patients.
Secondly, we will conduct a 1:4 matched cohort study to assess the association between incident BS and incident total hip or knee replacement (THR/TKR) in obese patients. Each BS patient will be matched to four patients who did not undergo BS on age, sex, body mass index (BMI), presence or absence of obesity-related metabolic disease, calendar time, and years of recorded history on the CPRD. We will follow all patients from the day of BS or the matched date (comparison group) to the first of the following: elapse of 15 years of follow-up, a Read code for THR/TKR (the outcome), end of the patient record, development of an exclusion criterion (see below), or end of the study period. We will perform subgroup analyses by sex, age, BMI groups, arthroplasty location, and duration of follow up (if necessary, see below). An extended Cox model will be applied to calculate hazard ratios (HR) with 95% confidence intervals (CI) of incident THR/TKR in association with incident BS adjusted for socioeconomic status (SES), sleep apnoea (time-varying), cardiovascular disease status (time-varying), Charlson comorbidity score (time-varying), and GP practice (random effect).
Knee arthroplasty
Cataract surgery
Spinal surgery (decompression and fusion)
Hip arthroplasty
Inguinal hernia surgery
Colonoscopy
Cholecystectomy
Susan Jick - Chief Investigator - BCDSP - Boston Collaborative Drug Surveillance Program
Marlene Rauch - Corresponding Applicant - University of Basel
Christoph Meier - Collaborator - University of Basel
Raoul Furlano - Collaborator - University Children's Hospital Basel
Theresa Burkard - Collaborator - University of Basel
Thomas Huegle - Collaborator - University of Basel
Tracy Glass - Collaborator - University of Basel
Christoph Meier - Collaborator - University of Basel
HES Admitted Patient Care;Patient Level Index of Multiple Deprivation