Frozen Shoulder and Glucose Following Corticosteroid Injection

Frozen Shoulder: A Pragmatic Replicated N-of-1 Case Series Investigating Glucose Levels and Patient Reported Outcomes Following a Corticosteroid Injection.

The goal of this clinical study is to look at what happens to blood sugar levels in people who have a frozen shoulder, who do not have diabetes after a corticosteroid injection in people with frozen shoulder.

• The primary aim of this study will be to evaluate alterations in glucose levels for people without diabetes, following corticosteroid injection for frozen shoulder.
• The secondary aims are to observe a range of clinical, physiological, and psychosocial outcomes following corticosteroid injection for people with frozen shoulder.

Participants will wear a glucose measuring device, attached to the upper arm collecting readings of glucose levels for 14 days.

Participants will undergo a corticosteroid injection as part of usual care.

Frozen shoulder is regarded one of the most painful conditions to affect the shoulder. It is associated with substantial and prolonged pain that affects sleep and permeates into every aspect of daily function. Epidemiological research suggests the average onset is in the 6th decade, affecting up to 8% of men and 10% of women of working age may be affected. In people with diabetes this figure dramatically increases, with the incidence of frozen shoulder ranging from 10 to 59%.

Injection for frozen shoulder is a well-established treatment option, particularly in the early phase of the condition. The aim of this study is to assess glycaemic changes in a non-diabetic population following a corticosteroid injection. The cardiovascular risk factors associated with developing diabetes, which are increasing globally, because of ageing, smoking, poor diet and lack of physical activity, means there is good reason to understand better the impact of injection for frozen shoulder on the normo-glycaemic population.

A continuous glucose monitor worn on the upper arm, will collect glucose data at multiple timepoints for the 14 days it is worn both before and following the injection. The device is minimally invasive with the use of a small electrochemical sensor inserted under the skin and takes readings of interstitial (between cells, just below the skin).

In the N-of-1 design the 'N' represents a single participant. This design bridges the gap between evidence-based research and clinical care in health conditions. N-of-1 studies are patient centred as participants become more involved in their condition and treatment, setting outcomes due to the close monitoring of participants. When the 'N' is repeated, this may make the findings more generalisable. The CONSORT guidelines for RCTs were extended and the CENT 15 guidelines for N-of-1 trials were developed. A strength of N-of-1 studies as they are focussed on the individual and whilst their comorbidities are stable, they can be included in the study.

For this study, an A-B-A phase design will be used. The initial phase (A) is where repeated glucose measures are automatically obtained via the sensor and uploaded to the website and to check participants glucose levels do not go to the hyper/hypoglycaemic ranges on the day of the injection.

The next phase (B) is the intervention phase where participants will undergo a corticosteroid injection. The final phase (A) is where 'washout' will occur and this will be for the remainder of the life of the sensor. The term 'washout period' refers to ''a period in which no intervention is administered. A washout may be administered between different treatment periods or may act as a period in itself, as in a "reversal" design (to "wash out" the effects of a treatment….'' Glucose data is continually provided by the sensor to the website, which is around 14 days. It will be important to investigate glucose levels for the life of the CGM to assess if any individuals developed hyperglycaemia following the injection.

Day 1 is the day the CGM is attached to the participant. Data will be collected from day 1 but will not be used as part of the analysis as there is less available data collected by the device, resulting in less accurate readings.

Day 2 to day 5, is the pre-injection phase. Day 5 is the day of the injection and up until the point of the injection is included as pre-injection phase. Immediately following the injection is the early phase post injection up to day 8. Day 9 to day 14 is the late phase post injection.

Sampling technique

The process of sample selection will be as follows:

• Potential participants referred to the musculoskeletal physiotherapy department with a diagnosis of primary idiopathic frozen shoulder and fitting the inclusion criteria.
• Purposive sampling technique to include potential participants who may be under-represented, in the health-care setting, particularly if south Asian or black due to the increase incidence of frozen shoulder in these cohorts.
• The sample is derived from the electronic patient record system.

Potential participants will be recruited by their referring GP to the musculoskeletal service. The study aims to have a maximum of 10 completed participant data sets. A literature review of sample size for N-of-1 trials found the average sample size was 13 (range 2 - 428; IQR, 6.5 - 38.5). Other studies had sample sizes of 4 for mobility rehabilitation 8 for virtual reality exercises for neck pain and 4 participants with low back pain and exercise. There is no formal sample size calculation as there are unknown features for example size of the effect. This study is novel and to the best of the chief investigators knowledge there is no other similar study for comparison.

The primary outcome measures are glucose levels pre and post injection and the shoulder pain and disability index (SPADI). A Microsoft Teams channel will enable secure digital data collection and this data will be captured on 2 separate occasions: at baseline and at 3 weeks post-injection. The baseline data will be captured once at the start of participant involvement.

Inclusion Criteria:

• Aged between 45 and 65
• Without diabetes or pre-diabetes
• Glenohumeral external rotation range is equally limited actively and passively by at least 50% compared to the contralateral side.

Other glenohumeral movements are equally limited actively and passively compared to the contralateral side in one or more additional planes of movement (such as hand behind the back/shoulder flexion/abduction/scaption).

• The symptoms have been present for at least one month.
• No significant abnormality on plain radiographs.
• Ability to provide informed consent

Exclusion Criteria:

• Aged 44 and under and 66 and over
• People with frozen shoulder secondary to significant shoulder trauma (e.g., dislocation, fracture or full thickness tear requiring surgery) or other causes (e.g., recent breast cancer surgery or radiotherapy)
• People with prediabetes, type 1 and 2 diabetes
• A neurological disease affecting the shoulder
• Evidence of other shoulder disorders (e.g., inflammatory arthritis, rotator cuff disorders, glenohumeral joint instability) or with red flags consistent with the criteria set out by British Elbow and Shoulder Society (Rangan, 2015).
• Primary osteoarthritis of the glenohumeral joint on plain radiograph
• Calcific tendinitis
• Who have received corticosteroid injection for shoulder pain in the last 12 weeks
• On long term steroids or a recent short course of oral steroids
• Currently being treated with coumarin anticoagulants, such as warfarin
• Who have participated in another research study involving an investigational medicinal product in the past 12 weeks
• Significant renal or hepatic impairment
• Any other significant disease which, in the opinion of the Investigator, may either put the participants at risk because of participation in the study, or may influence the result of the study.
• Systemically unwell or with infection
• Pregnant or breastfeeding