“Sick day” guidance in patients at risk of acute kidney injury

“Sick day” guidance in patients at risk of acute kidney injury: Position Statement from the UKKA AKI SIG

As health professionals one of our key aims is to reduce the risk of avoidable harm to our patients. Some people are at increased risk of AKI, for example those with chronic kidney disease (CKD), heart failure, or those taking particular medications (2). Many health care professionals provide advice to such patients that certain drugs should be temporarily discontinued during acute intercurrent illnesses, particularly where there is disturbed fluid balance. This advice is commonly described as ‘sick day rules’, similar to those provided to patients with diabetes treated with insulin.

This Position Statement refers to such guidance given to patients when they are well, about how to manage their own medication should they become unwell. It does not relate to clinical management of known or suspected AKI by health care professionals.

The definition of ‘acute illness’ in this context has to be simple enough to be understood by patients at risk. The NHS Highland definition (1) is a good example:

‘when you are unwell with any of the following;

• vomiting or diarrhoea (unless minor)
• fevers, sweats and shaking’

There are three main reasons for providing such advice:

• Non-steroidal anti-inflammatory drugs impair renal autoregulation by inhibiting prostaglandin-mediated vasodilatation of the afferent arteriole and may increase the risk of AKI.
• Drugs that lower blood pressure, or cause volume contraction, might increase the risk of AKI by reducing glomerular perfusion. These drugs include:
  • Diuretics, which can cause a reduction in GFR if their use results in hypovolaemia. This group also includes the mineralocorticoid receptor antagonists spironolactone and eplerenone, used frequently in heart failure.
  • Blood-pressure-lowering medications, which will lower systemic blood pressure.
  • SGLT2i, which can cause a reduction in GFR if their use results in hypovolaemia.
• Drugs might accumulate as a result of reduced kidney function in AKI, increasing the risks of adverse effects. These drugs include:
  • Metformin, accumulation of which may be associated with an increased risk of type A lactic acidosis in high risk patients.
  • Some sulfonylurea drugs, accumulation of which increase the risk of hypoglycaemia.
  • Other renally excreted medications may accumulate — see the table in the UK Renal Pharmacy Group AKI Medicines Optimisation leaflet for more information.

Although the NICE AKI guideline (2) advocates provision of sick day guidance, the evidence that provision of such advice reduces net harm is very weak.

RAAS inhibitors have the highest level of evidence to maintain health and improve prognosis for people living with common and serious long-term health conditions such as heart failure, diabetes mellitus and proteinuric chronic kidney disease (CKD). These same people are at increased risk of developing AKI, especially if they develop acute illness. However, it is pathology associated with these long-term health conditions that most often underlies their high risk of AKI, rather than RAAS inhibitor therapy. Large observational analyses from the UK did not find associations between RAAS inhibitors use and increased AKI risk in these groups at high absolute risk of AKI, when compared to use of other anti-hypertensive medications (5). Although RAAS inhibitor use across the whole study population was associated with a very small increase in relative AKI risk compared to use of other antihypertensive medication, this was observed amongst people with lowest absolute risk of AKI and not in patients with heart failure or advanced CKD.

Whilst there are limitations to interpretation of observational data, these findings do not support the convention that RAAS inhibitors are “nephrotoxic” (i.e. commonly and independently cause AKI). This is in contrast to truly nephrotoxic drugs (e.g. gentamicin, amphotericin B) which commonly cause AKI regardless of clinical context, through direct tubular nephrotoxic effects. The association between RAAS inhibitor use and AKI risk instead usually reflects the common use of these drugs in populations who have the highest absolute risk of AKI. Conventional teaching and practice have incorrectly assumed causality from such association, driving clinical practice for decades despite a lack of rigorous hypothesis testing. Continuing to label RAAS inhibitors as nephrotoxic frequently risks these clinically important drugs being stopped unnecessarily or not restarted within a clinically effective time frame, because of professional or public perception that these drugs have caused AKI. There is no evidence to show routinely stopping these RAAS inhibitors to prevent AKI is of overall patient benefit (3, 4) and delays restarting these drugs likely contributes to very high rates of emergency readmission to hospital post AKI

There are potential harms associated with widespread provision of ‘sick day’ rules or guidance, particularly when the patients have not been clinically assessed and where it is unclear at what level of ill health the medication should be discontinued. These include:

• Decompensated heart failure when drugs blocking the RAAS system and diuretics are
• AKI caused by right-sided congestion complicating withdrawal of diuretics, particularly in right heart failure.
• Development of poorly controlled hypertension with cessation of antihypertensive
• Reduced adherence to drug treatment which may have been incorrectly described as ‘nephrotoxic’. Patients may consider that the potential harm outweighs the potential benefit and decide to stop taking the drug permanently, even in the absence of, or after recovery from, an acute illness.
• Patients may over-interpret the advice and stop their drug treatment during even minor illnesses.
• Patients may not re-start their drug treatment on
• The drugs may not be titrated back to the previous evidence-based levels even when there has been no evidence of AKI.
• People may self-manage inappropriately and not seek professional help at an appropriate stage.
• Making changes to regular medication in patients reliant on medicines compliance aids is complex and can have unintended consequences.
• Diabetes control may be adversely affected by inappropriate cessation of glucose lowering treatment.

It is also a theoretical possibility that ACEI and ARB treatment might reduce the severity or duration of AKI, at least in a subset of patients. These drugs, by causing efferent arteriolar vasodilatation, increase blood flow to the renal tubules: and it is tubular injury that causes persistent AKI and the increased risk of subsequent chronic kidney disease.