The KDIGO AKI guidelines were recently published as a supplement to Kidney International. I have already discussed the guidelines briefly, but wanted to now focus on the specific guideline recommendation.
KDIGO’s recommended position on the use diuretics in AKI caught my eye. Chapter 3.4, pages 47-49, reviews the data for the use of diuretics. The recommendations are as follows:
3.4.1: We recommend not using diuretics to prevent AKI. (1B)
3.4.2: We suggest not using diuretics to treat AKI, except in the management of volume overload. (2C)
The KDIGO panel cite some of the potential benefits of loop diuretic therapy, including reducing oxygen consumption in the loop of Henle by inhibiting active sodium transport - therefore potentially reducing ischemic injury; inducing the loss of the high medullary osmolality by inhibiting the Na-K-2Cl cotransporter – decreasing the ability to reabsorb water; increase intratubular pressure and thereby push out necrotic debris blocking tubules; inhibiting prostaglandin dehydrogenase which reduces renovascular resistance and increases renal blood flow.
The panel then goes on to say: “There is no evidence that the use of diuretics reduces the incidence or severity of AKI. Ho et al conducted two comprehensive systematic reviews on the use of the loop diuretic furosemide (furosemide) to prevent or treat AKI. Furosemide had no significant effect on in-hospital mortality, risk for requiring RRT, number of dialysis sessions, or even the proportion of patients with persistent oliguria."
The question: why is there continued use of loop diuretics in the management of AKI given the lack of evidence and in light of KDIGO guidelines? In an informal survey of my colleagues at the Brigham there was a uniform opinion that a trial of loop diuretics is useful in managing patients with AKI. So what gives?
The quick answer is that there have been very few studies in this area and in each of the studies the sample size has been quite limited – in the 50 to 100 range, and the trials are flawed - some severely, so there is really no basis for making a recommendation either way. The role of diuretics to treat AKI has really not been adequately tested.
To give you a flavor of the type of evidence there is, KDIGO cite a study by Félix Cantarovich and co-workers on behalf of the High-Dose Furosemide in Acute Renal Failure Study Group, a French consortium as evidence to support the recommendation of no benefit of loop diuretics. The study published in AJKD is unfortunately unavailable open access (Vol 44, No 3 (September), 2004: pp 402-409).
Cantarovich et al enrolled 338 consecutive subjects with ARF [AKI] requiring RRT in a prospective, randomized, double-blind, placebo-controlled, multicenter trial. The design of the study was complicated – they had a screening phase where they treated patients with very high dose of furosemide 15 mg/Kg IV over 4 hours (for a 60 Kg patient that is 900 mg of furosemide over 4 hours!) and if they had a further increase in serum creatinine they were randomized to either furosemide 25 mg/kg/d IV (maximum, 2 g/d,), or matched placebo. They also stratified patient’s severity using a Simplified Acute Physiology Scores (SAPS) score. After randomization, patients could enter an optional predialysis period (phase I) for a maximum of 48 hours (days 1 and 2) or a dialysis period (phase II), which began either immediately after the screening period or after phase I, according to the urgency of dialysis requirements. Intermittent or continuous RRT could be performed. RRT modalities differed between intensive care units and nephrology wards and included continuous hemofiltration or hemodiafiltration, when possible, as well as hemodialysis daily or on alternate days.
The trial had many limitations. The most important were:
1. The sample comprised of quite severe AKI. This is because they only enrolled the patient’s if they needed RRT. RRT was defined according to French practice guidelines at the time: a progressive increase in plasma urea levels greater than 180 mg/dL (>30 mmol/L), oligoanuria for 48 hours, or uremic syndrome.
2. The complicated design – with a screening period, stratification, 2 different phases, different dialysis modalities by center.
3. They did not use the “intention to treat” principle because of 338 subjects enrolled and randomized, they excluded 8 subjects because “they did not meet inclusion criteria”. But in actuality these subjects 2 patients in the furosemide group and 6 patients in the placebo group, showed a spontaneous recovery in renal function during predialysis phase I. Notably, though, these excluded patients had all been treated with furosemide at a high dose already. The recovery was not really spontaneous!
4. The randomization was not balanced. Patients randomized to the furosemide versus the placebo arm had worse kidney function at baseline (4.87 +/- 2.61 versus 4.31 +/- 2.78, P=0.013, respectively). Likewise, there was imbalance for BUN at baseline between the 2 arms. Staggeringly, 33 of 166 versus 13 of 164 subjects had diabetes in the furosemide versus placebo arms. One imbalance went in the other direction with more subjects with sepsis and sepsis/shock in the placebo versus furosemide arm, respectively.
5. The surprising use of “death” and “death at the end of the study”. They assumed a 45% baseline value and they observed a 33% rate. They also assumed that they would need to see a 15% effect size for 80% power. Pretty ambitious! Clearly, the study was underpowered.
Despite all of these limitations, the authors report that “More patients in the furosemide group achieved a diuresis of 2 L/d and in a shorter period compared with the placebo group.”
The bottom-line is that KDIGO went on a limb to make the assertion recommending against diuretics in AKI. The evidence is simply not there to make such a statement. The largest randomized trial is deeply flawed but even this showed minimal evidence for harm and a potential benefit. So, I will continue to use diuretics in the treatment of AKI until there is additional data to suggest otherwise.
KDIGO’s recommended position on the use diuretics in AKI caught my eye. Chapter 3.4, pages 47-49, reviews the data for the use of diuretics. The recommendations are as follows:
3.4.1: We recommend not using diuretics to prevent AKI. (1B)
3.4.2: We suggest not using diuretics to treat AKI, except in the management of volume overload. (2C)
The KDIGO panel cite some of the potential benefits of loop diuretic therapy, including reducing oxygen consumption in the loop of Henle by inhibiting active sodium transport - therefore potentially reducing ischemic injury; inducing the loss of the high medullary osmolality by inhibiting the Na-K-2Cl cotransporter – decreasing the ability to reabsorb water; increase intratubular pressure and thereby push out necrotic debris blocking tubules; inhibiting prostaglandin dehydrogenase which reduces renovascular resistance and increases renal blood flow.
The panel then goes on to say: “There is no evidence that the use of diuretics reduces the incidence or severity of AKI. Ho et al conducted two comprehensive systematic reviews on the use of the loop diuretic furosemide (furosemide) to prevent or treat AKI. Furosemide had no significant effect on in-hospital mortality, risk for requiring RRT, number of dialysis sessions, or even the proportion of patients with persistent oliguria."
The question: why is there continued use of loop diuretics in the management of AKI given the lack of evidence and in light of KDIGO guidelines? In an informal survey of my colleagues at the Brigham there was a uniform opinion that a trial of loop diuretics is useful in managing patients with AKI. So what gives?
The quick answer is that there have been very few studies in this area and in each of the studies the sample size has been quite limited – in the 50 to 100 range, and the trials are flawed - some severely, so there is really no basis for making a recommendation either way. The role of diuretics to treat AKI has really not been adequately tested.
To give you a flavor of the type of evidence there is, KDIGO cite a study by Félix Cantarovich and co-workers on behalf of the High-Dose Furosemide in Acute Renal Failure Study Group, a French consortium as evidence to support the recommendation of no benefit of loop diuretics. The study published in AJKD is unfortunately unavailable open access (Vol 44, No 3 (September), 2004: pp 402-409).
Cantarovich et al enrolled 338 consecutive subjects with ARF [AKI] requiring RRT in a prospective, randomized, double-blind, placebo-controlled, multicenter trial. The design of the study was complicated – they had a screening phase where they treated patients with very high dose of furosemide 15 mg/Kg IV over 4 hours (for a 60 Kg patient that is 900 mg of furosemide over 4 hours!) and if they had a further increase in serum creatinine they were randomized to either furosemide 25 mg/kg/d IV (maximum, 2 g/d,), or matched placebo. They also stratified patient’s severity using a Simplified Acute Physiology Scores (SAPS) score. After randomization, patients could enter an optional predialysis period (phase I) for a maximum of 48 hours (days 1 and 2) or a dialysis period (phase II), which began either immediately after the screening period or after phase I, according to the urgency of dialysis requirements. Intermittent or continuous RRT could be performed. RRT modalities differed between intensive care units and nephrology wards and included continuous hemofiltration or hemodiafiltration, when possible, as well as hemodialysis daily or on alternate days.
The trial had many limitations. The most important were:
1. The sample comprised of quite severe AKI. This is because they only enrolled the patient’s if they needed RRT. RRT was defined according to French practice guidelines at the time: a progressive increase in plasma urea levels greater than 180 mg/dL (>30 mmol/L), oligoanuria for 48 hours, or uremic syndrome.
2. The complicated design – with a screening period, stratification, 2 different phases, different dialysis modalities by center.
3. They did not use the “intention to treat” principle because of 338 subjects enrolled and randomized, they excluded 8 subjects because “they did not meet inclusion criteria”. But in actuality these subjects 2 patients in the furosemide group and 6 patients in the placebo group, showed a spontaneous recovery in renal function during predialysis phase I. Notably, though, these excluded patients had all been treated with furosemide at a high dose already. The recovery was not really spontaneous!
4. The randomization was not balanced. Patients randomized to the furosemide versus the placebo arm had worse kidney function at baseline (4.87 +/- 2.61 versus 4.31 +/- 2.78, P=0.013, respectively). Likewise, there was imbalance for BUN at baseline between the 2 arms. Staggeringly, 33 of 166 versus 13 of 164 subjects had diabetes in the furosemide versus placebo arms. One imbalance went in the other direction with more subjects with sepsis and sepsis/shock in the placebo versus furosemide arm, respectively.
5. The surprising use of “death” and “death at the end of the study”. They assumed a 45% baseline value and they observed a 33% rate. They also assumed that they would need to see a 15% effect size for 80% power. Pretty ambitious! Clearly, the study was underpowered.
Despite all of these limitations, the authors report that “More patients in the furosemide group achieved a diuresis of 2 L/d and in a shorter period compared with the placebo group.”
The bottom-line is that KDIGO went on a limb to make the assertion recommending against diuretics in AKI. The evidence is simply not there to make such a statement. The largest randomized trial is deeply flawed but even this showed minimal evidence for harm and a potential benefit. So, I will continue to use diuretics in the treatment of AKI until there is additional data to suggest otherwise.
So there is no evidence to show that it should not be use , And there is no evidence to say it should be used, yet we use it. This is called the art of medicine, this is where we exercise our clinical judgement as physicians, for a majority of clinical situations we do this everyday
ReplyDeleteDr. Singh, can you please educate me on the dosage of loop diuretcs you would try in AKI?
ReplyDeleteIf blood pressure and/or hypertension occur after you purchase affordable cialis and taking it while having sex, never forget diuretics as it is used to treat hypertension and kidney diseases that usually starts from high blood pressure.
ReplyDelete