SGLT2 inhibitors (‘flozins’) in CKD
SGLT2 inhibitors (SGLT2is) are powerful disease-modifying agents in Chronic Kidney Disease (CKD). They reduce the risk of kidney failure and reduce cardiovascular disease (heart failure and CV death). They are appropriate for many patients with CKD but will require some individualised decision making (see below).
Quick decision tool
| Patient profile | eGFR threshold (ml/min/1.73 m2) | uACR threshold (mg/mmol) | strategy |
|---|---|---|---|
| CKD & T2DM | > 15 | any level | offer SGLT2i (strong recommendation) |
| CKD without diabetes & heavy albuminuria | > 15 | > 25 | offer SGLT2i (strong recommendation) |
| CKD without diabetes & little / no albuminuria | 20 – 45 | < 25 |
offer SGLT2i (slightly weaker recommendation) |
| symptomatic heart failure | > 15 | any level | offer SGLT2i (strong recommendation) |
Notes:
- these thresholds broadly align with NICE / SMC / UKKA guidelines but have been simplified slightly here
- a uPCR of 35 mg/mmol approximates to uACR 25 mg/mmol and can be used for decision-making purposes
- most guidelines recommend using the highest tolerated / licensed dose of an ACE inhibitor or ARB before initiating an SGLT2i but this sequencing has not been tested in large clinical trials
Prescribing principles
Individualise treatment decisions. At the risk of stating the obvious, treatment decisions should be individualised to factor in anticipated benefits (reductions in risk of kidney failure AND cardiovascular disease), potential side-effects, tablet burden, patient preferences etc.
Broadly, SGLT2i can be expected to reduced the relative risk of kidney failure by ~35% and heart failure / cardiovascular death by ~25%. Therefore, absolute benefits are greater in patients at greatest risk of these outcomes: hence their recommended use in patients with heavy proteinuria or lower eGFR. The NNT to prevent kidney failure / progression over 2 years will vary from ~10 – 20 in high-risk groups (e.g. eGFR < 30 ml/min; uACR > 30 mg/mmol) to ~500 – 1000 in low-risk groups (e.g. eGFR 45 – 60 ml/min; uACR < 3 mg/mmol). (And a similar wide spectrum is seen for NNTs to prevent cardiovascular outcomes.)
There is uncertainty over when to use SGLT2i in co-morbid, frail patients but the available evidence suggests both benefits and harms may be amplified in this group and with the caveat that extreme frailty is not represented in clinical trials.
This is a class effect. All SGLT2i have similar benefit and safety profiles. Therefore NICE recommend using the cheapest licensed option (see East Region Formulary for current local practice).
Expect a haemodynamic “eGFR dip”. This can be up to 30% and occurs within weeks of starting. This differs from the similar phenomenon seen with ACEi / ARBs in that there is not a big associated hyperkalaemia risk. Therefore many prescribers choose NOT to repeat U&Es within 1 – 2 weeks of starting because this is unnecessary. It is almost always appropriate to continue the SGLT2i, but if any uncertainty can discuss with renal. It can be important to explain to patients that this is a haemodynamic effect and does NOT mean that SGLT2is are harming the kidneys. There is strong evidence that SGLT2i slow progression of CKD and prevent AKI.
Referral criteria
Specialist input may be helpful in:
- Confirmed or suspected Type 1 Diabetes (T1DM): Do NOT initiate SGLT2is in primary care. At present SGLT2i are generally considered contra-indicated in T1DM due to the risk of diabetic ketoacidosis (DKA). That said, with increasing options for ketone and sugar monitoring, potentially eligible patients could be referred to the diabetes team for consideration of initiation under their specialist supervision. It may also be helpful to liaise with the diabetes team for patients in whom the type of diagnosis is in doubt (i.e. are they definitely T2DM?).
- Diagnostic uncertainty re: kidney disease: refer to renal. Use standard referral criteria to discuss patients with suspected glomerulonephritis, vasculitis, genetic kidney conditions (such as Polycystic Kidney Disease) etc.
- eGFR < 20 ml/min/1.73m2: usually being seen by the renal team anyway.
Safety & patient counselling
SGLT2i are generally very well tolerated, particularly in patients without diabetes.
The main potential side-effects are:
- genital thrush
- volume depletion / orthostatic hypotension
- DKA including euglycaemic DKA (in patients with diabetes)
- increased hypoglycaemia risk if used with insulin / sulphonylureas (in patients with diabetes)
Many guidelines advocate giving patients “sick day rules” (although the evidence supporting these is mixed). It is usually appropriate to advise patients to temporarily suspend SGLT2i during periods of acute illness, particularly when not eating and drinking (risk of ketosis) and for 48 hours prior to major surgery.
Note that there is very strong evidence that risk UTIs (other than genital mycotic infections) is NOT significantly increased by SGLT2i use. Therefore frequent UTIs are not usually a contra-indication to SGLT2i.
Further information:
Clinical Guidelines
- UK Kidney Association (UKKA) SGLT2 Inhibitor Guideline & 2023 Update
- NICE Technology Appraisal (TA1075): Dapagliflozin for treating chronic kidney disease
- BMJ Clinical Practice Guideline (2024)
- KDIGO 2024 Clinical Practice Guideline: Evaluation and Management of CKD
Patient Information Leaflets
- UKKA Patient Resource: SGLT2i patient information leaflets
