Loading ...

New Orleans — Should serum sodium in chronic hyponatremia be corrected slowly or rapidly? Two expert nephrologists faced off in an hour-long debate on the topic here at the National Kidney Foundation (NKF) 2026 Spring Clinical Meetings. 

Current guidelines advise slow correction, to a maximum of 8 mmol/L/d, to avoid osmotic demyelination syndrome (ODS) for those with risk factors — including baseline sodium of 105 mEq/L or below — along with alcohol use disorder, malnutrition, and advanced liver disease, hypokalemia, and hypophosphatemia. Higher rate limits (10-12 mmol/L/d) are recommended for those at average risk. 

However, recent studies suggest that ODS is extremely rare and that rapid correction may be associated with lower mortality. 

“This is a very clinically relevant topic that has become significantly controversial because of emerging new studies that have challenged what we thought we understood of the risks associated with rapid correction,” Juan Carlos Q. Velez, MD, professor of medicine and interim chair of nephrology at Ochsner Health, New Orleans, told Medscape Medical News in an interview prior to the debate. 

“Many elements of this controversy are worth discussion,” said Velez, who argued in favor of rapid correction. 

Arguing for the more cautious side, Helbert Rondon Berrios, MD, professor of medicine, renal-electrolyte division, University of Pittsburgh School of Medicine, told Medscape, “the controversy, which dates back to the early 1980s, has been revived by these recent studies. But most of the studies have severe methodological flaws…Our position is that there are certain people who we have to be careful with in terms of correcting.” 

PRO: Rapid Correction in Patients at Low Risk of ODS 

Velez went first, arguing in favor of rapid correction for those in whom the risk of ODS is low. “The rate of correction of chronic hyponatremia creates anxiety for practitioners…I would submit that the morbidity of symptomatic hyponatremia is often underestimated,” he said. 

Most nephrologists have never seen ODS in the absence of risk factors such as alcohol abuse or malnourishment. Moreover, there is no evidence from randomized controlled trials showing that slow correction prevented ODS and/or improved survival, he argued. 

In 1986, a paper published in the New England Journal of Medicine reported that 12% of 60 patients with hyponatremia who were corrected at a rate greater than 12 mmol/L/d developed ODS. However, radiological demonstration was only conducted in half of them, and only CT was available at the time, not MRI. “So, we can’t judge this,” said Velez. 

Perhaps the strongest evidence against rapid correction in the literature is a 2019 Swedish national healthcare database study in which a total of 83 cases of ODS were identified over 14 years. Of those, 86.7% were hyponatremic and 69.9% were alcoholic. Of the hyponatremic patients, 92% were corrected at a greater rate than 8 mmol/L/d. All had MRI demonstration of ODS. 

But, of the 7.2% who died within 3 months, it wasn’t clear how many died from hyponatremia. Moreover, there was no control group of hyponatremia patients who did not develop ODS to determine how many of them had been corrected rapidly. “Because we don’t have a control group, we don’t have an ability to extract an odds ratio,” Velez pointed out. 

Low Rates of ODS 

Meanwhile, there have been at least four recent publications essentially finding that ODS is extremely rare and perhaps not even associated with rapid correction. In addition, these studies do not show that slower correction of sodium is associated with improved outcomes and in fact, some showed the opposite, he said. 

The largest of these, published in 2023, included 22,858 hospitalizations with hyponatremia at five Canadian hospitals between 2010 and 2020. Rapid sodium correction was common, in 17.7%, but ODS was rare, occurring in just 0.05%. 

And in the most recent, an observational study of 13,988 patients who were hospitalized with severe hyponatremia (≤ 120mEq/L) between 2008 and 2023, faster sodium correction was associated with lower risk for 90-day death or delayed neurologic events. 

And, Velez pointed out, it’s not always easy to follow the practice guidelines for slow correction. One paper from 2015 reported that rapid correction occurred 28% of the time, despite efforts at slow correction. And in the 2023 Canadian study, 69% of patients with sodium < 110 mmol/l/d ended up with fast correction, despite best efforts. 

“This gives me a bit of anxiety, as we are now vulnerable for lawsuits…Is that really fair when we don't really have the ability to control these events?” asked Velez.

ODS Associated With Alcohol, Nutrition, and Liver Disease 

Velez “dissected” four recent studies suggesting that ODS is a concern with rapid correction. One was a retrospective study of 852 patients admitted to the ED of a tertiary hospital from January 2013 to December 2018 with plasma sodium ≤ 125 mmol/L, of whom 40% had corrections faster than 8 mmol/L/d. A total of four patients (0.5%) with cerebral edema and 11 (1.3%) with ODS were detected.

However, imaging was not obtained systematically. No cases of ODS led to permanent neurological sequelae. Moreover, alcohol, malnutrition, and liver disease were all associated with ODS risk, whereas the rate of correction was not. 

“So, it's not really a study that would clarify and make strong case for ODS being a concern,” Velez said. 

Other studies purporting to show increased risk for ODS with rapid correction, such as a Swedish retrospective cohort study of 7623 patients, relied on chart coding and lacked imaging confirmation, he noted. 

In addition, several of the studies failed to separate clearly between acute hyponatremia, which must be corrected fast to avoid brain edema from the hyponatremia itself and chronic hyponatremia. 

Other studies showing increased ODS from fast correction have mostly included people at high risk, including those with very low sodium levels — ie, between 104-112 mEq/L — often with evidence of alcohol abuse. 

Two meta-analyses addressed this issue, one showing rapid correction associated with ODS, the other suggesting the opposite, said Velez. 

But meta-analyses aren’t informative in this case, because “what we have is a bunch of observational studies, heterogeneous, with different study designs, drastic sample size differences, and not prospective…So I don’t think we should put a lot of weight in these,” he said. 

Velez concluded by listing four main conditions that make people vulnerable to ODS due to impairment of the brain’s ability to adapt to changes in serum sodium: excess alcohol use, malnutrition, hypokalemia, and hepatic disease. 

A “cautious rate of correction should perhaps be reserved to these individuals with extremely high risk,” he said. 

“The current guidance, in my view, leads to ineffective management of symptomatic hyponatremia, and disproportional fear of using the most effective mechanism-of-action-driven medication ever developed for the treatment of hyponatremia — yes, I’m talking about vaptans. And for that reason, I submit that practice guidelines should be revised,” Velez proposed.

CON: Potential for Overshooting the Mark in Patients at High Risk of ODS 

Berrios said that the recommended upper correction limit of 8 mmol/L/d was set “because of the potential for overshooting the mark.” He went on to explain that the authors of the guidelines wanted to set a limit below the traditional 10-12 mmol/L/d, “so the rate of correction is not too close to rates that can potentially result in patient harm, allowing for some room for error.” 

However, he noted that the lower limit “is intended for patients with chronic hyponatremia at higher risk for ODS. It was not intended for patients with acute hyponatremia, or for most patients with sodium of 120 or greater.”

In an editorial accompanying the large 2023 Canadian study mentioned by Velez, that concluded that ODS was rare and not linked to rapid correction, two long-time advocates of rapid correction stated that the current guidelines are too strict and recommended liberalizing the correction limits to 15-20 mmol/L in 48 hours. 

However, Berrios noted that the same editorialists published this view in 1985, “so their position has been very consistent for almost 40 years.” 

Their main arguments, along with those of other rapid correction proponents, are that ODS is rare and that it is not caused by rapid correction of hyponatremia, but rather, that alcohol use disorder and other factors inherent to the patient are responsible for this complication, he said.

They also argue that slow sodium correction in patients with severe hyponatremia is damaging, and that rapid hyponatremia correction actually saves lives. 

But, said Berrios, “the devil is in the details.” 

In the 2023 Canadian study, roughly 90% of the patients included had a sodium of more than 120 mmol/L, so the recommended correction limit of 8 mmol/L/d does not apply to those patients, because their risk is very low, he pointed out. However, in the subgroup of patients with sodium levels of 110 mmol/L or lower, “we notice that the incidence goes up to 2.6%, over 50 times the incidence in the entire cohort. So the effect is diluted by including patients at low risk.” 

Moreover, he argued, none of the studies excluded patients with acute hyponatremia, and ODS was detected based on ICD codes and radiology reports. “We know that ICD codes are not reliable to detect ODS. For instance, in one study, where they detected eight cases of ODS by chart review, none of them had an ICD code for ODS,” he reported. 

Missing ODS Cases? 

Another problem is that not all patients were imaged, so it’s possible that many ODS cases were missed. One example of this is the Canadian study which identified seven patients with ODS, while another report by a group of neurologists from the same institution — and within the same time period — found 45 cases of ODS, said Berrios.

“Now, if you look at older studies with populations at high risk, meaning those with sodium 115 or less, using an adequate method to identify cases of ODS, such as manual chart reviews, we observe that many more cases of ODS start to appear,” he noted. 

In one study, the incidence of ODS went up to 21% among those with sodium levels of 105 or less, and in another, 8% of those with baseline sodium 110 mmol/L or less had ODS, many of whom had experienced rapid correction. 

The emergency department study of 852 patients, cited by Velez, actually showed that over-correction was a risk factor for ODS, with an odds ratio of 4.2, Berrios pointed out. The researchers found other risk factors as well, including alcohol use and liver disease, “but over-correction of hyponatremia was the most important,” he said. 

The ongoing PRONATREOUS study, for which Berrios is an investigator, is using manual chart reviews to examine neurological outcomes of patients with severe hyponatremia with sodium 110 mmol/L or less.

A preliminary analysis of 157 patients who had sodium levels of 105 mmol/L or less showed that 1% developed cerebral edema, 4.5% had confirmed ODS by imaging, and 14.7% had suspected ODS without imaging confirmation. 

All of the confirmed and suspected ODS cases were corrected faster than 8 mmol/L/d. Of all those who were corrected at more than 8 mmol/L/d, 22% developed ODS, compared with none who were corrected at less than 8 mmol/L/d, Berrios reported. 

‘Slow’ Patients Inherently Different From ‘Fast’ Patients 

The problem with recent studies suggesting lower mortality rates with more rapid correction is that they’re all observational and therefore subject to confounding, he argued. 

Indeed, in three of those studies, individuals with observed slower rates of correction had more comorbidities — including cirrhosis, heart failure, and cancer — than those who were corrected more rapidly; and these comorbidities themselves can affect the correction rate. 

“Sodium correction rates in these studies [may be] less influenced by active therapeutic interventions and more by the passive intrinsic characteristics of the patient themselves,” he said. “In other words, patients who correct sodium slowly are inherently different than patients who correct the sodium fast.” 

Berrios concluded, “Recent studies questioning the relationship between rapid correction of chronic hyponatremia and ODS or suggesting potential mortality benefits have limitations that should be considered carefully. Until more conclusive evidence, the existing guidelines for treating severe hyponatremia continue to be a reasonable approach.”

Velez is a speaker for Mallinckrodt Pharmaceuticals and Vertex Pharmaceuticals, and on advisory boards for Travere Therapeutics and Vera Therapeutics. Berrios has been remunerated as an expert witness in cases related to the treatment of hyponatremia by attorneys for both the plaintiffs and the defense. 

Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape, with other work appearing in The Washington Post, NPR’s Shots blog, and Diatribe. She is on X (formerly Twitter) @MiriamETucker and BlueSky @miriametucker.bsky.social