| BACKGROUND | ||
| The Centers for Medicare and Medicaid Services (CMS) Appropriate Use Criteria (AUC) program takes effect January 2020, and requires ambulatory and emergency medicine providers to consult AUC using a CMS approved clinical decision support mechanism when ordering advanced imaging (CT, MRI or nuclear medicine) in eight priority clinical areas (PCAs). Physicians from the High Value Practice Academic Alliance evaluated the evidence in the literature pertaining to the use of CT and MRI for patients with radiographically occult hip fracture and compiled evidence to guide decision-making about CT versus MRI. | ||
|
LITERATURE REVIEW Abstract at Association of University Radiologists 2019 Conference. Award: Trainee Prize “The Frequency of Radiographically Occult Operative Hip Fracture in Patients with Acute Hip Pain: A Systematic Review and Meta-Analysis of Individual Patient Data” Arya Haj-Mirzaian | Pamela Johnson, MD | Shadpour Demehri, MD, PhD |
||
| PURPOSE: Owing to the low sensitivity of radiographs for diagnosing hip fractures, a large number of patients require cross-sectional imaging. We performed a systematic review and meta-analysis of studies evaluating the frequency of radiographically occult operative hip fracture to define a diagnostic algorithm based on strong evidence and current-generation high-resolution CT and MRI. We aimed to define a subgroup(s) of patients with the highest risk of occult fracture, and the most appropriate imaging modality to evaluate these patients. | ||
| METHOD AND MATERIALS: A literature search was performed, using PubMed, Medline and Embase, for original articles and abstracts published before Sept. 20, 2018, with no language restrictions. Titles and abstracts from 2,219 publications were screened by two authors in duplicate to determine eligibility. If necessary, the full text was assessed. Inclusion criteria were: 1. patients with suspected hip fracture, 2. no radiographic evidence of operative hip fracture (femur head/neck, intertrochanteric or subtrochanteric), 3. further evaluation with MRI, CT scan or bone scan for the final diagnosis of fracture. Data were extracted in duplicate, and overall strength of evidence was assessed. | ||
| RESULTS: The number of eligible studies identified was 44, including 3,307 patients. The overall frequency of occult hip fracture was 40% (95% Confidence Interval(CI), 35–45). The prevalence of occult fracture was 40% (37–47) and 30% (21–41) based on MRI and CT/bone scan, respectively. Sensitivity analyses showed a significantly higher prevalence of occult fracture in patients with radiographic isolated greater trochanteric fracture (versus no definitive fracture) — 74% (54–87). Patients with no definitive radiographic fracture, age >80 (versus ≤80) or with suspicious radiograph (versus normal) were significantly associated with presence of occult fracture. | ||
| CONCLUSION: Occult hip fracture was diagnosed in five to seven of every 10 patients (number needed to test: 2.5). Using MRI was associated with a 1.3-fold higher probability of occult fracture diagnosis in comparison with CT or bone scan. Patients with suspected hip fracture and normal radiographs, isolated greater trochanter fracture, suspicious radiograph or age >80 should be evaluated by MRI. | ||
EVIDENCE TABLE: Pending publication in PubMed journal.
APPROPRIATE USE CRITERIA
| Title | Clinical scenario 1: Traumatic hip pain and risk factors for occult fracture | Clinical scenario 2: Traumatic hip pain and risk factors for occult fracture |
| Definition | Definition: high suspicion and 1 or more of the following • age > 80 • isolated greater trochanter fracture on radiograph • equivocal radiograph |
Definition: high suspicion and 1 or more of the following • age < 80 • equivocal radiograph |
| AUC rules | ||
| Consistent with AUC | • Hip MRI | • Hip CT |
| Allowable by AUC | • Hip CT | • Hip MRI |
| Does not meet AUC | ||
| Not applicable (No AUC) | • Bone scan | • Bone scan |
| Evidentiary vs Consensus | Evidentiary: Oxford Grade 1 (Meta-analysis) | Evidentiary: Oxford Grade 1 (Meta-analysis) |
MULTIDISCIPLINARY TEAM
HVPAA requires that all practicing physicians participating in the development of AUC disclose any conflicts of interest using the International Community of Medical Journal Editors (ICJME) form. This information is publicly available in a timely fashion upon request, for a period of not less than five years after the most recent published update of the relevant appropriate use criteria. Members of the hip pain AUC development team are:
- Ali Raja, Emergency Medicine, Massachusetts General Hospital
- Mustapha Saheed, Emergency Medicine, The Johns Hopkins Hospital
- Steven Blash, Family Medicine, Johns Hopkins Community Physicians
- Clifton “Bing” Bingham, Internal Medicine and Rheumatology, Johns Hopkins Bayview Medical Center
- Danny Lee, Internal Medicine, Johns Hopkins Community Physicians
- Adam Levin, Orthopaedic Surgery, Johns Hopkins Hospital
- Shadpour Demehri, Radiology, The Johns Hopkins Hospital
- Pamela Johnson, Radiology, The Johns Hopkins Hospital
- Ramin Khorasani, Radiology, Brigham and Women’s Hospital
- Stacy Smith, Radiology, Brigham and Women’s Hospital
Disclosure: AUC developers may receive future royalties from licensure of AUC to CMS-approved clinical decision support mechanisms.
