5f.1 Background and Evidence Review
Multimorbidity and frailty are interrelated but distinct conditions that are important to consider when individualising treatment decisions for SGLT 2 inhibitor prescription. Multimorbidity, (defined as the presence of two or more long-term health conditions) is common in the UK. In a cross-sectional study of 1.75 million people registered with a general practitioner in Scotland, 40% of people had single health conditions and 23% were multimorbid (61). In UK Biobank, which recruited half a million people between the ages of 40 and 69 years living in the UK, 19% of people were multimorbid (62).
The exclusion criteria of trials often result in underrepresentation of people with certain types of comorbidity from large placebo-controlled trials. This is arguably the case for many of the SGLT 2 inhibitor trials. The CANVAS, CREDENCE, EMPA-REG OUTCOME, DAPA-CKD, DAPA-HF, DECLARE-TIMI 58 and VERTIS CV trials all excluded participation where there was evidence of certain conditions (e.g. liver disease, cancer, and haematologic conditions). Such approaches may be justified by the need to ensure participants will survive long enough to be at risk of the studied outcomes. For example, progression of CKD can take years, and so exclusion of people with active cancer is necessary to ensure the trial can address its primary question. However, these exclusion criteria pose a challenge when aiming to generalise the evidence from large placebo-controlled trials to people with these conditions/multimorbidity. Of note, individual trials had additional exclusions which also limit generalisability. For example, people with prior or current immunosuppressive therapy, or people affected by endocrine diseases other than DM, and potentially at risk of certain safety outcomes (e.g. at risk of DKA or amputation) were excluded from many of the SGLT 2 inhibitor trials.
Two of the large SGLT 2 inhibitor trials in heart failure, DAPA-HF and DELIVER, investigated the role of frailty on treatment effects of dapagliflozin in heart failure with reduced, and mildly reduced or preserved, ejection fraction, respectively (63, 64). Both of these trials, in which ̴50% and ̴60% of individuals were categorised as frail respectively, demonstrated consistent results; specifically, that the risk of key clinical endpoints (including cardiovascular death or worsening heart failure) was higher among people with a higher frailty score, while the relative benefits of dapagliflozin were unmodified by frailty. Consequently, the estimated absolute benefits of dapagliflozin would be expected to be greater among more frail individuals. Adverse events were similar between dapagliflozin and placebo arms across the spectrum of frailty. However, such analyses do not mitigate the limitations of stringent exclusion criteria in randomized controlled trials, with the most frail individuals still likely to be excluded from such trials. Evidence among highly frail or multimorbid individuals therefore remains limited.
People with multimorbidity and frailty may suffer from a particularly high burden of treatment. They are at high risk for adverse drug reactions and, in the case of multimorbid people, may be considered for treatment with multiple different drugs for their different health conditions. Conversely, they may also be at high absolute risk of a trial’s key efficacy outcomes and therefore particularly benefit from the effects of SGLT 2 inhibition on risk of cardiovascular death, heart failure complications, AKI and the reduced risk of hospitalisation observed in some of the SGLT 2 inhibitor trials.
For some indications for SGLT 2 inhibition, such as for the treatment of CKD in people without albuminuria, the benefits of SGLT 2 inhibition on CKD progression may be expected to accrue over a few years. Benefits on cardiac disease, however, emerge early. Therefore among frail individuals, the ratio of benefit to risk may need to be considered, with particular emphasis on the absolute benefits and the duration of treatment anticipated for meaningful benefit to accrue.
Our recommendations for this group are in line with the UK guidelines for multimorbidity and frailty (Multimorbidity: clinical assessment and management, NICE Guideline NG56, Published September 2016) which places emphasis on individual preference, awareness of the potential burden of polypharmacy and consideration of life expectancy in balancing risks and benefits of SGLT 2 inhibitor treatment.
5f.2 Recommendations for Implementation
- We suggest an approach to care that takes account of frailty and multimorbidity where these apply. This can include:
- Establishing the person’s goals, values and priorities
- Consideration of the balance of disease and treatment burden (for example, prognostic benefits in people with limited life expectancy or frailty)
- Agreeing an individualised management plan (Grade 2D).
Rationale: When making decisions on which individuals would benefit from SGLT 2 inhibition one has to consider the participants included in the relevant trials that provided the evidence for their use. These trials generally excluded people with greater degrees of frailty and certain comorbidities. Therefore, caution must be exercised when extending evidence of safety (and perhaps also benefit) of SGLT 2 inhibitors to such individuals, although one needs to also consider at the same time that many of these individuals, and particularly those with heart failure, are likely to achieve significant benefit from the use of SGLT 2 inhibitors.
5f.3 Clinical Research Recommendation
- Future trials of SGLT 2 inhibitor use in people with CKD that seek to extend inclusivity to those of advanced age and multimorbid status.