Additional imaging is of limited value in traumatic hip fractures with a history of distant malignancy and no suspicious lesion on plain radiographs

Abstract

Objective:
To determine the value of obtaining additional pre-operative imaging in patients with a traumatic hip fracture and a history of malignancy in whom plain radiographs show no lesion suspicious for metastases.

Design:
Retrospective review.

Setting:
Teaching NHS Trust in the UK, over an 8-year period treating 4,421 hip fractures.

Patients/Participants: 367 patients with hip fracture and a history of malignancy at a site distant to the hip. 330 had a history of trauma and no lesion on the plain radiograph suspicious for metastases.

Main Outcomes Measurements:
Whether obtaining additional imaging pre-operatively (MRI, CT, bone scan) identified metastases or affected management.

Results:
32/330 patients had further pre-operative imaging, none of which demonstrated a pathological fracture secondary to malignancy. On follow up, 3/330 (0.9%) cases were found to have occult metastasis at the hip fracture site. All 3 had only plain radiographs prior to surgery. In 2 this was identified on histological examination of intra-operative samples, and in 1 radiologically as a metastatic metaphyseal lesion 18 months following a hemiarthroplasty. Only in the latter case, pre-operative identification of hip metastasis could have altered surgical management. Patients undergoing further pre-operative imaging waited significantly longer for surgery (35±26 versus 51±26 hours, P=0.0011).

Conclusions:
In the absence of a suspicious metastatic lesion on initial plain radiographs, further pre-operative imaging is unlikely to identify a lesion that will affect management and confers significant delays to surgery. Sending intra-operative histological samples may help guide post-operative oncological management but further work is needed to prove its utility.

Level of Evidence:
Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.