Routine x-rays for suspected Frozen Shoulder: a retrospective review


Frozen shoulder is a common condition and current guidelines state that it is a diagnosis of exclusion. Along with a history and clinical examination, routine x-ray is mandated to rule out any masquerading pathology such as fracture, dislocation, metastatic lesions or severe OA. Despite the certainty of the guidelines there is a lack of evidence to support the use of routine x-rays in this situation.

342 of 350 shoulder x-rays
did not have any concerning features.


A retrospective review was performed of all x-rays obtained in the diagnosis of frozen shoulder between February 2014 and August 2017 in an integrated musculoskeletal interface service. Results were screened and the prevalence of masquerading pathology was determined.


350 shoulder x-rays that were performed with a differential diagnosis of frozen shoulder were reviewed. 213 were from female patients (60.9%), mean age was 57.7 years (SD 10.4). 342 (97.7%) did not have any concerning features. Six (1.7%) had severe OA, one (0.3%) had a fracture and one (0.3%) had a lucency. All 8 patients with 'masquerading' pathology had findings from the history and clinical examination that would have warranted an x-ray regardless of differential diagnosis of frozen shoulder.

Cost and savings

Routine x-rays for suspected frozen shoulder offer little over diagnosis based on history and examination alone.

The full cost savings of not doing routine £25/radiographs and the impact on productivity/efficiency the change to our Frozen shoulder clinical pathway has made will be established from October 2018-October 19 data.


Our results show that 342 shoulder x-rays did not have any concerning features and thus conceivably were performed unnecessarily. Considering how common frozen shoulder is, this knowledge could be a stimulus to significantly reduce the number of unnecessary x-rays undertaken per year and also reduce patient exposure to radiation The IPOPS  Frozen shoulder clinical pathway has been amended and x-rays will be requested based on clinical reasoning rather than mandated.

Top three learning points

  1. The importance of collecting, reviewing and acting on the findings of local service data to improve patient care and ensuring resources are used appropriately.  
  2. Not to underestimate the time required for such a project especially in no additional financial support.
  3. Don’t be afraid to reach out to researchers for support.
  4. The peer review process can be challenging but our final manuscript was enhanced by the process.

Funding acknowledgements

This work was unfunded and developed in-house incurring no costs.

This project was undertaken during CPD time allocated to clinicians during their working week.


Additional notes

Presented at Physiotherapy UK 2018.