COMMENT|ONLINE FIRST
Six pills less: no benefit for bicarbonate supplementation in renal allograft recipients
Klemens Budde, Fabian Halleck
Lancet Published:January 25, 2023
DOI:https://doi.org/10.1016/S0140-6736(22)02630-7
Metabolic acidosis (defined as serum bicarbonate <22 mmol/L) is frequent in patients with chronic kidney disease and is associated with disease progression.1Several prospective trials and meta-analyses2, 3, 4, 5, 6 have suggested potential positive effects (of low certainty) of sodium bicarbonate treatment on chronic kidney disease progression, and bicarbonate supplementation is recommended in patients with chronic kidney disease with serum bicarbonate less than 22 mmol/L.1However, protective mechanisms are not fully understood and it is difficult to determine whether acidosis is a cause or consequence of renal function decline.
In kidney transplant recipients, metabolic acidosis is even more frequent than in patients with chronic kidney disease,7 as other factors, such as calcineurin inhibitors, might aggravate renal tubular acidosis in a transplanted kidney. In The Lancet, Nilufar Mohebbi and colleagues8 present the results of a large, randomised, placebo-controlled, 2-year, phase 3, multicentre trial in Switzerland, providing the first rigorous evidence on the effect of sodium bicarbonate treatment in 240 maintenance renal allograft recipients with metabolic acidosis. The mean participant age was 55·5 years (SD 13·5), 167 (70%) of 240 participants were male and 73 (30%) were female, and 202 (84%) of 240 participants were White. 121 patients were randomly assigned to placebo and 119 patients were randomly assigned to sodium bicarbonate. The primary endpoint was the creatinine-based estimated glomerular filtration rate (GFR) slope. Besides efficacy, the investigators thoroughly evaluated safety, as sodium bicarbonate might have adverse effects—such as hypertension due to sodium load and gastrointestinal discomfort—and causes additional pill burden. The trial was transparently reported and well executed, and treatment adherence was excellent. The data clearly show no positive effects on renal function (mean difference in estimated GFR slope 0·032 mL/min/1·73 m2 per year, 95% CI –1·644 to 1·707) and all other outcome parameters. Thorough sensitivity analyses provided further evidence that there was no detectable effect of sodium bicarbonate on renal function decline in renal allograft recipients. Although sodium bicarbonate had no relevant side-effects, around 3 g sodium bicarbonate treatment (six pills) led to correction of acidosis and serum bicarbonate concentrations, whereas placebo-treated patients had a constant mild acidosis.
What are the lessons learned from such a negative trial? There is no negative trial, as all rigorous, adequately powered, and well executed trials provide an answer to a relevant question, irrespective of whether we like the outcome or not. This trial fills an important knowledge gap in the daily treatment of renal allograft recipients. Although these patients frequently have mild metabolic acidosis, there is now good evidence that correction of acidosis will not have any detectable effect on renal function. Thus, sodium bicarbonate can safely be omitted, reducing the pill burden in patients. As always in medicine, patients should be treated based on the best evidence and not laboratory values or hypotheses. The current study shows that bicarbonate treatment has no measurable benefit for renal allograft recipients, and the abnormal laboratory value can be left uncorrected until evidence evolves.8 Of course, it can be argued that the Swiss population and 2-year intervention was suboptimal, and a longer and more aggressive treatment aimed at complete acidosis correction in a population with more severe acidosis would have shown some potential benefits on renal function or other endpoints, such as nutritional status and bone or muscle health. However, pill burden was already high and with a 2-year study duration and excellent protocol adherence the results of this study have good generalisability. Until new evidence shows a clear benefit of acidosis correction, the practical and evidence-based treatment decision should be to avoid the additional pill burden and the costs for the health system. Another learning point is that evidence from chronic kidney disease in patients who have not received kidney transplants cannot be transferred to kidney transplant recipients with chronic kidney disease, and the additional effects of calcineurin inhibitors might not easily be antagonised with alkali treatment. Another example for a different treatment response in transplanted kidneys is the reduced effect of renin–angiotensin system blockade.9 A timely example is the use of SGLT2 inhibitors, where the beneficial effects in patients with chronic kidney disease should not be extrapolated to renal transplant recipients. Rigorously conducted trials (eg, NCT05374291) are needed to generate the evidence for the benefits and risks of this promising class of drugs in kidney transplant recipients. Finally, we need to learn more about the mechanisms that cause chronic kidney disease progression in patients with metabolic acidosis. The present study protocol describes additional analyses on acid-base handling in renal allograft recipients.4 Such insights, together with the absence of effect of sodium bicarbonate supplementation in renal allograft recipients, might improve understanding of the pathophysiology and consequences of metabolic acidosis in renal allograft recipients and in patients with chronic kidney disease.
Ultimately, patients deserve more trials like this, which answer relevant practical questions in daily clinical practice.
