From the Journals

Fracture liaison services confer benefit on recurrent fracture risk



Implementation of fracture liaison services (FLS) at two Swedish hospitals was associated with an 18% reduction of recurrent fracture over a median follow-up of 2.2 years, results from an observational cohort study found.

“Patients receiving fracture care within an FLS have higher rates of [bone mineral density] testing, treatment initiation and better adherence,” first author Kristian F. Axelsson, MD, and colleagues wrote in a study published in the Journal of Bone and Mineral Research. “However, the evidence regarding FLS and association to reduced risk of recurrent fracture is insufficient, consisting of smaller studies, studies with short follow-up time, and studies with high risk of various biases.”

Dr. Axelsson, of the department of orthopedic surgery at Skaraborg Hospital, Skövde, Sweden, and colleagues used electronic patient records from four hospitals in Western Sweden to identify all patients aged 50 years or older with a major osteoporotic fracture – defined as a fracture of the wrist, upper arm, hip, vertebra, or pelvis – between 2012 and 2017. The study population consisted of 15,449 patients from two hospitals with FLS and 5,634 patients from two hospitals with no FLS. The researchers used multivariable Cox models to compare all patients with a major osteoporotic fracture during the FLS period with all patients with a major osteoporotic fracture prior to the FLS implementation. The FLS hospitals and non-FLS hospitals were analyzed separately using the same methodology.

The mean age of patients was 74 years, 76% were female, and the most common index fracture site was the wrist (42%). In the hospitals with FLS, the researchers observed 1,247 recurrent fractures during a median follow-up time of 2.2 years. In an unadjusted Cox model, the risk of recurrent fracture was 18% lower in the FLS period, compared with the control period (hazard ratio, 0.82; P = .001). This corresponded to a 3-year number needed to screen of 61, and did not change after adjustment for clinical risk factors. In the non-FLS hospitals, no change in recurrent fracture rate was observed.

Osteoporosis medication treatment rates after fracture did not differ between the FLS and non-FLS hospitals, prior to FLS implementation (14.7% vs. 13.3%, respectively; P = .10). However, following FLS implementation, a larger proportion of fracture patients were treated at the FLS hospitals, compared with those at the non-FLS hospitals (28% vs. 12.9%; P less than .001).

Dr. Mattias Lorentzon, head of geriatric medicine, Sahlgrenska University Hospital, Mölndal, Sweden.

Dr. Mattias Lorentzon

“Our study is the largest yet, including both historic controls and controls at nearby hospitals without implementations of fracture liaison services,” one of the study authors, Mattias Lorentzon, MD, said in an interview. “We were able to rule out temporal trends in refracture risk and show that, [in] patients who had an index fracture at a hospital with an FLS, the refracture rate was lower than for patients who had an index fracture before the FLS was started, indicating that FLS reduce the risk of recurrent fracture. No such trends were observed in hospitals without FLS during the same time period.”

Dr. Lorentzon, head of geriatric medicine at Sahlgrenska University Hospital, Mölndal, Sweden, said that FLS implementation “led to a large increase in the use of osteoporosis medication, which we believe is the reason for the reduction in recurrent fracture risk observed. We believe that our results provide solid evidence that FLS implementation can reduce the rate of recurrent fractures, suggesting that all hospitals treating fracture patients should have fracture liaison services.”

Dr. Stuart L. Silverman, clinical professor of medicine, Cedars-Sinai Medical Center and UCLA School of Medicine

Dr. Stuart L. Silverman

In an interview, Stuart L. Silverman, MD, said that the study adds to compelling data on the efficacy and need for patients with clinical fracture to have case management by a FLS. “We recognize that near term risk is substantial in the year following a fracture,” said Dr. Silverman, who is clinical professor of medicine at Cedars-Sinai Medical Center and the University of California, Los Angeles, and is not affiliated with the Swedish study. “For example, the risk of a subsequent fracture after hip fracture is 8.3%, which is similar to the risk of subsequent acute myocardial infarction after an initial acute MI. However, only 23% of patients receive osteoporosis medication after a hip fracture. Yet a fracture is to osteoporosis what an acute MI is to cardiovascular disease. We recognize that men and women age 65 years and older who have suffered a hip or vertebral fracture should be evaluated for treatment, as this subpopulation is at high risk for a second fracture and evidence supporting treatment efficacy is robust. We need a multidisciplinary clinical system which includes case management such as a fracture liaison service. We know FLS can reduce hip fracture rate in a closed system such as Kaiser by over 40%. This manuscript addresses the utility of a FLS in terms of reducing risk of future fracture.”

The researchers acknowledged certain limitations of the study, including its observational design and the fact that patients prior to the FLS period were fewer and had longer follow-up time, compared with patients during the FLS period.

The study was funded by the Swedish Research Council and by grants from the Sahlgrenska University Hospital. Dr. Axelsson reported that he has received lecture fees from Lilly, Meda/Mylan, and Amgen. Dr. Lorentzon has received lecture fees from Amgen, Lilly, UCB, Radius Health, Meda, GE-Lunar, and Santax Medico/Hologic. The other coauthors reported having no financial disclosures. Dr. Silverman reported that he is a member of the speakers’ bureaus for Amgen and Radius. He is also a consultant for Lilly, Pfizer, and Amgen and has received research grants from Radius and Amgen.

SOURCE: Axelsson K et al. J Bone Min Res. 2020 Feb 25. doi: 10.1002/jbmr.3990.

Next Article: