Hypertonic Saline Regimen Linked to Lower Mortality in Severe Brain Injury

A two-part saline treatment commonly used to control dangerous pressure inside the skull was associated with substantially lower mortality among patients with severe brain injuries, according to a retrospective study conducted at a major academic hospital in South Korea. The findings suggest that maintaining a relatively steady osmotic environment with a continuous infusion of 3% saline, while using concentrated 7.5% saline boluses to respond to acute pressure crises, may be more effective than relying on boluses alone. In the most seriously injured patients, those who received the combined treatment had an intensive-care mortality rate of 16.2%, compared with 41.1% among patients treated only with intermittent boluses. Their 28-day mortality was 18.9%, versus 45.7% in the bolus-only group. The results, published in Neurocritical Care, are striking because the combined-treatment patients received a greater total sodium load but nevertheless reached lower peak blood sodium concentrations. That pattern challenges the assumption that the highest sodium concentration necessarily provides the best protection against swelling in the injured brain.

The study addresses one of the most urgent problems in neurocritical care: elevated intracranial pressure, or ICP. The skull is a rigid compartment containing the brain, blood and cerebrospinal fluid. When trauma, bleeding, stroke or another disease causes brain tissue to swell, pressure can rise within this confined space. As ICP increases, it can reduce cerebral perfusion—the pressure gradient that drives blood through brain tissue—and compromise the delivery of oxygen and glucose. In severe cases, high pressure can trigger ischemic injury, distort brain structures and cause herniation, in which brain tissue is forced through openings within the skull. Osmotic therapy is intended to interrupt this process. By increasing the concentration of osmotically active particles in the bloodstream, hypertonic saline draws water out of swollen brain cells and into the circulation, where it can eventually be removed by the kidneys. The treatment can also expand intravascular volume, potentially supporting blood pressure and cerebral perfusion in patients who are simultaneously suffering from blood loss or shock.

The investigation was led by Hye Won Choi and Jeong-Am Ryu at Samsung Medical Center in Seoul. The researchers reviewed the records of 993 adults who had received 7.5% hypertonic saline in a neurosurgical intensive-care unit between January 2015 and December 2024. All participants had received the concentrated saline bolus, allowing the central comparison to focus on whether adding continuous saline infusion was associated with better outcomes. Of the total cohort, 851 patients received bolus therapy alone, while 142 received the boluses along with a continuous intravenous infusion of 3% saline. The primary analysis concentrated on 212 patients classified as having severe brain injury, defined by a Glasgow Coma Scale score from 3 to 9 on admission. The Glasgow Coma Scale measures eye opening, verbal responses and motor responses, producing a score from 3 to 15. Lower scores indicate deeper impairment of consciousness. When speech could not be assessed because a patient was intubated or had a tracheostomy, the investigators estimated the verbal component using a previously described regression method.

The two saline strategies have different pharmacological roles. At the Seoul hospital, a standard 7.5% saline bolus consisted of 100 milliliters infused over 20 to 30 minutes. Clinicians used it when patients experienced sudden neurological deterioration, when monitored ICP remained above 20 millimeters of mercury or when brain scans showed signs of impending herniation. The bolus produces a rapid increase in plasma osmolality, creating an immediate gradient that can shift water away from swollen tissue. Its effect, however, may fade over several hours, and repeated doses can cause abrupt changes in serum sodium. The continuous treatment was delivered as 3% saline at 30 milliliters per hour through a central venous catheter, with clinicians targeting a serum sodium concentration of 145 to 155 milliequivalents per liter. Rather than waiting for pressure to surge and then reacting with another large dose, the infusion was intended to sustain the osmotic gradient between acute interventions. The concept resembles maintaining a stable pressure differential instead of repeatedly applying a sequence of short, powerful pushes.

Brain swelling itself is not a single process. In the early phase of traumatic or other severe brain injury, cytotoxic edema can develop when depleted cellular energy stores disrupt ion pumps. Sodium and other ions accumulate inside cells, and water follows, causing neurons and glial cells to swell. Later, damage to the blood–brain barrier can produce vasogenic edema. This allows plasma proteins and fluid to move into the extracellular space, increasing tissue volume. Hypertonic saline may influence both forms of edema, although its ability to draw water depends on the integrity and permeability of the barrier between blood and brain. The study’s results support the possibility that the duration and steadiness of osmotic exposure may matter as much as the highest sodium value achieved. In the severe-injury subgroup, the combined-therapy group had better ICP control despite lower peak serum sodium. The observation is important because clinical strategies have often emphasized reaching a predetermined sodium target, potentially overlooking how quickly sodium rises, how long the osmotic effect persists and how sharply concentrations fall afterward.

After statistical adjustment for differences between patients, the association with survival remained large. The investigators used inverse probability of treatment weighting, a method that attempts to balance measured characteristics between treatment groups by giving different statistical weights to individual patients. In the severe brain injury analysis, combined therapy was associated with an adjusted odds ratio for ICU death of 0.23, with a 95% confidence interval of 0.08 to 0.65 and a p value of 0.006. For 28-day mortality, the adjusted odds ratio was 0.20, with a 95% confidence interval of 0.07 to 0.55 and a p value of 0.002. The researchers also performed five sensitivity analyses, including analyses using multiple imputation for missing information and an E-value assessment of the potential influence of unmeasured confounding. Each analysis reportedly supported the direction of the main findings. In the overall cohort, however, the mortality differences did not reach conventional statistical significance: the odds ratio was 0.58 for ICU mortality and 0.59 for 28-day mortality. A time-to-event analysis across the full cohort did show a significant survival difference, measured by a Kaplan–Meier log-rank p value of 0.006.

The apparent benefit does not prove that continuous saline caused patients to survive. This was a single-center retrospective cohort study, meaning the researchers examined treatment decisions and outcomes that had already occurred rather than assigning patients to treatment randomly. Physicians chose to add continuous 3% saline when they believed sustained osmotic support was needed. That clinical judgment may have been based on factors not fully captured in the records, such as the trajectory of ICP, subtle neurological changes, the severity of swelling on imaging or the anticipated need for surgery. It is also possible that patients selected for combined therapy received other differences in care, including more intensive monitoring or earlier neurosurgical intervention. Statistical weighting can adjust for recorded variables, but it cannot eliminate confounding by indication when important differences remain unmeasured. The authors note that even the study’s robustness analyses and high E-values cannot fully exclude this problem. The lower mortality could therefore reflect the combined regimen, associated aspects of care, patient selection or some mixture of all three.

The findings also highlight why hypertonic saline has attracted particular interest in critically ill patients. Mannitol, another osmotic agent, can lower ICP but promotes diuresis, increasing urine output and potentially reducing circulating blood volume. That effect may be disadvantageous in patients with traumatic injuries, bleeding or unstable blood pressure, where maintaining cerebral perfusion is already difficult. Hypertonic saline can increase intravascular volume while creating the osmotic gradient needed to move water from brain tissue. Its concentrated formulations also deliver a large osmotic load in a relatively small volume, a useful property when clinicians need a rapid response without administering substantial fluid. In the hospital involved in the study, 7.5% saline was prepared by diluting 60 milliliters of 11.7% sodium chloride with sterile water. Local regulatory guidance discourages direct use of undiluted 11.7% saline, so the compounded 7.5% formulation served as the institution’s standard bolus treatment. These details matter because saline concentrations, dosing practices, monitoring protocols and thresholds for treatment may differ substantially among hospitals.

The work arrives amid uncertainty about whether continuous hypertonic saline improves long-term outcomes. The large COBI randomized trial examined continuous infusion of 20% hypertonic saline in patients with moderate-to-severe traumatic brain injury and did not find a significant improvement in neurological outcomes six months later, although a post hoc analysis suggested a possible mortality benefit. That study and earlier research have helped establish that hypertonic saline can reduce ICP, but lowering pressure is not automatically equivalent to improving recovery. A therapy could normalize a physiological measurement while failing to prevent neuronal death, disability or complications elsewhere in the body. The new study focuses on a different regimen—7.5% boluses combined with 3% continuous saline—and on short-term outcomes in a broader neurocritical population treated at one center. Its most provocative result is not simply the survival association, but the relationship between sodium kinetics and pressure control. If sustained gradients are more important than sodium peaks, future protocols may be designed around the shape and stability of the concentration curve rather than a single target value. That hypothesis will require carefully controlled testing.

Prospective, multicenter trials are now needed to determine whether the strategy can safely improve outcomes and which patients are most likely to benefit. Such studies would need to define when the continuous infusion begins, how long it continues, how sodium and osmolality are monitored, and how clinicians respond to kidney dysfunction, fluid overload or electrolyte disturbances. They would also need to measure neurological outcomes beyond survival, because avoiding death does not necessarily mean preserving independence or cognitive function. Until those data are available, the findings should be viewed as a compelling signal rather than a new standard of care. Still, the study offers a clinically intuitive possibility: in a brain under extreme pressure, repeatedly delivering a rescue dose may not be as effective as combining rapid responses with a carefully controlled background treatment. The result has already sharpened a central question for neurocritical care—whether the brain benefits more from the highest osmotic force clinicians can safely produce, or from a steadier gradient that prevents pressure from rebounding in the first place.