Osmotic demyelination syndrome: revisiting the diagnostic criteria through two fatal cases

Abstract

Background

Osmotic Demyelination Syndrome (ODS) encompasses Central Pontine Myelinolysis and Extrapontine Myelinolysis, both of which are serious neurological conditions linked to the overly rapid correction of hyponatremia. Despite growing evidence, the exact etiology of ODS remains incompletely understood. The present paper describes two case studies, aiming to provide a comprehensive overview of the pathological findings and clinical outcomes associated with ODS.

Case presentation

Case #1. A 74-year-old woman was admitted to the emergency department following a head trauma caused by a loss of consciousness. Initial laboratory tests revealed severe hyponatremia (sodium level of 101 mmol/L) and hypokalemia (potassium level of 2.9 mmol/L). The patient underwent corrective therapy with saline and potassium chloride. Despite the correction of electrolyte imbalances, the patient developed a hyperintense lesion in the median portion of the pons on T2-fluid-attenuated inversion recovery (FLAIR) MRI sequence 14 days post-treatment, consistent with ODS. The patient’s condition deteriorated, leading to irreversible coma and status epilepticus, culminating in death 32 days after admission. Case #2. An 81-year-old woman with a medical history of hypothyroidism, hypertension, major depression, and stage 3 chronic kidney disease presented with mild gait disturbances. Subsequent testing revealed severe hyponatremia (sodium level of 100 mmol/L). Following an initial clinical improvement due to sodium correction, the patient’s condition worsened, with symptoms progressing to confusion, lethargy, and eventually, ODS. Dermatological manifestations, including blistering lesions and facial edema, appeared as the condition advanced. The patient succumbed to irreversible coma 47 days after admission.

Conclusion

ODS traditionally carried a poor prognosis, with high mortality rates and diagnoses often made postmortem. However, recent advances in understanding the pathophysiology, along with improvements in diagnostic techniques such as MRI and intensive care treatments, have led to earlier identification, treatment, and recognition of milder forms of the syndrome. Despite these advancements, ODS remains a critical condition with significant risks, particularly following the rapid correction of severe hyponatremia.

Introduction

Central Pontine Myelinolysis (CPM) is an acquired life-threatening neurologic disorder, first described by Adams et al. in 1959 [1], as a focal area of myelinolysis in the pons, differing in its distribution and clinical presentation from any other previously recognized form of demyelination. Initially reported as a disease related to alcohol abuse and malnutrition, it was later identified by Kleinschmidt-DeMasters and Norenberg in 1981 as a consequence of too rapid correction of hyponatremia [2]. The lesion typically involves damage to the myelin sheath of brain cells in the pons, without an inflammatory or vascular basis, while nerve cells and axons are mostly spared. Subsequently, other areas of myelinolysis, more commonly affecting the midbrain, thalamus, cerebellum, lateral geniculate body, external capsule, extreme capsule, hippocampus, and putamen, have also been described and are referred to as Extrapontine Myelinolysis (EPM). Typically, periventricular and subpial areas are not affected [3].

Therefore, CPM and EPM were combined in a clinical syndrome called Osmotic Demyelination Syndrome (ODS) [4].

Case analysis

Case #1

Clinical manifestations

A 74-year-old woman presented to the emergency department (ED) of a tertiary-care hospital due to head trauma following a loss of consciousness. The patient had tested positive for SARS-CoV-2 five days prior, with clinical manifestations including gastrointestinal symptoms (watery diarrhea and vomiting) and generalized asthenia. At home, in the five days before admission, the patient had received fluid therapy with saline and glucose. On admission, she appeared drowsy but was arousable to verbal stimuli, with ideomotor slowing. Earlier that day, she had undergone blood chemistry tests at another laboratory, which showed sodium levels at 101 mmol/L and potassium levels at 2.9 mmol/L; these electrolyte levels were later confirmed in the ED. The electrocardiogram (ECG) showed only an increase in left atrial size, while brain computed tomography (CT) revealed no morphological alterations. Therapy was initiated with 500 mL of 0.9% saline solution and 6 vials of NaCl for 8 h, plus KCl 40 mEq/l diluted in 500 mL saline; antiviral therapy with Remdesivir for SARS-CoV-2 infection was also started.

The following evening, the patient was admitted to the Geriatrics ward, where she appeared alert, oriented, and cooperative with mild slowing. During the hospital stay, close monitoring of sodium and potassium levels in arterial blood gases was carried out (Fig. 1); the infusion rate of sodium chloride was adjusted based on changes in ionic concentrations.

Fig. 1.

Timeline of blood sodium and potassium levels during hospitalization in Case #1

On the subsequent day, given the overall increase in sodium levels of 20 mmol/L over approximately 29 h, the administration of saline was discontinued. Furthermore, the patient experienced an episode of Acute Respiratory Failure (ARF) due to airway obstruction from fluid dysphagia, necessitating the intervention of the resuscitation team.

Over the next few days, sodium levels continued to rise and stabilized returning to baseline levels. Additionally, starting four days after the correction of hyponatremia, the patient, who had tested negative for SARS-CoV-2, exhibited neurological symptoms including slowed and slurred speech, difficulty performing index-nose tests, slowing of finger-tapping, followed by a positive Romberg test and positive left Babinski sign.

A new brain CT scan was performed, which showed no distinguishable changes from the prior scan. A few days later, she experienced an episode of tight bronchospasm with ARF. A chest CT scan revealed pulmonary interstitial edema and bilateral new-onset pleural effusion. Additionally, she presented with dyspnea on mild physical activity/exercise and hypoxemia on arterial blood gases testing. An echocardiogram documented akinesia and ballooning of the ventricular apex, mild heart failure, and pericardial effusion. Considering the symptoms exhibited, the patient received treatment with a Venturi mask for oxygen therapy, bronchodilators, corticosteroids, and diuretics.

Nevertheless, after neurological improvement, the patient was subsequently described as alert, oriented, and cooperative. Additionally, there was progress in apraxia, and the patient subjectively reported a resolution of symptoms consistent with brain fog.

Due to the increase in troponin levels and pathological findings in serial ECGs, the patient underwent percutaneous coronary angiography with percutaneous coronary intervention and placement of three drug-eluting stents. An electroencephalogram (EEG) performed later showed normal brain electrical activity. However, there was a new deterioration in neurological status with worsening vigilance and ideomotor slowing.

On the sixth day of hospitalization, the immunoglobulin levels (IgG, IgA, IgM) in the blood sample were found to be within normal range. Eight days after admission, TSH, FT3, and FT4 levels were within normal ranges. Ten days after admission, the patient was transferred to another facility, where she presented as unresponsive but awake, and disoriented in both space and time. The neurological examination revealed dysarthria, right-sided facial hypomimia with flattening of the nasolabial fold, pronation of the right upper limb, and reduced pain and temperature sensation on the right hemibody. These findings were consistent with involvement of the pons, which affects motor functions related to speech, as well as vigilance and alertness, and the corticobulbar tracts of the brainstem, which can result in facial hypomimia. Additionally, the patient demonstrated generalized fatigability impacting both voice and respiration, leading to a whispery voice and poor coordination between breathing and phonation. The clinical presentation was characterized by fluctuations, including transient improvements in dysarthria and orientation. On the eleventh day, laboratory tests showed a slight increase in FT4 levels (1.69 pg/ml; normal range 0.61–1.50). However, the overall condition progressively deteriorated over the following days. The patient became increasingly unresponsive to verbal stimuli but occasionally followed simple commands. Speech production deteriorated from unintelligible moans to complete absence of speech, despite preserved oral movements. Ideomotor slowing worsened, and motor symptoms developed in all four limbs, with a tendency for pronation and dropping of the right upper limb. Additionally, the left upper limb exhibited clonus, which was initially sporadic but became more pronounced over time. A neurological consultation performed 13 days after admission raised suspicion of herpetic encephalitis, while a subsequent consultation on the fourteenth day suggested autoimmune encephalitis. On the fifteenth day, samples were sent to the laboratory to test for autoimmune encephalitis markers (anti-NMDA-R, anti-CASPR-2, anti-AMPA-1/2, anti-LGI1, anti-DPPX, and anti-GABAB-R antibodies) and paraneoplastic neurological syndromes (anti-amphiphysin, anti-CV2, anti-PNMA2, anti-Yo, anti-Ri, anti-Hu, anti-GAD65, anti-Titin, anti-Recoverin, anti-Sox1, anti-Zic4, and anti-Tr antibodies). Starting from the sixteenth day of hospitalization, neurological consultations converged towards the diagnosis of ODS.

After a few days, the patient exhibited a complete absence of voluntary movements in the upper limbs, accompanied by bilateral clonus, more pronounced on the left side, which showed some regression with benzodiazepines (diazepam) administration. Subsequently, there was a total loss of upper limb movements, with the limbs presenting in spastic rigidity. In a later stage, lower limb movements were also affected, predominantly in the left limb, which exhibited hypotonia. The eyes were alternately open or semi-open, with a gaze preferentially oriented towards the right visual field. Initially, eye movements could follow the examiner, but later there was a restricted exploration of the right visual field, and the patient became unable to follow the examiner’s movements. The pupils were of normal size and shape, and were isotropic and isocyclic, reacting appropriately to both direct and consensual light stimuli.

Ultimately, the terminal clinical presentation involved a complete absence of response to both verbal and later painful stimuli, with closed eyes and a lack of tone in all four limbs, accompanied by an extensor plantar response.

During hospitalization, the patient received antiepileptic therapy and intravenous immunoglobulin therapy for 7 days, but with no apparent benefit. No investigations to assess plasma and urinary osmolality or cortisol levels are documented in the medical records. The patient deceased 32 days after admission to the ED due to irreversible coma and status epilepticus following ODS.

MRI and EEG evidence

During hospitalization, serial brain MRIs were conducted. Initially, an area of subacute ischemia was noted in the left thalamus, along with evidence of chronic ischemia. Starting from 14 days after correction of hyponatremia, a hyperintense lesion was detected in the median portion of the pons on the T2-FLAIR sequence (Figs. 2 and 3). Additionally, there weak hyperintensity was observed in T2-FLAIR images with restricted diffusion signal in the mesial temporal and parahippocampal cortices bilaterally, more pronounced on the right side. Similar changes were observed in both the left and right putamina, involving the outer capsules, deep mesial temporal regions, and juxtacortical white matter, with minimal involvement of the corticospinal tracts. Subsequently, the lesions showed evolution with a reduction in hyperintensity.

Fig. 2.

MR images acquired 14 days after hyponatremia correction in Case #1, demonstrating a hyperintense lesion in the median portion of the pons

Fig. 3.

MR images acquired 14 days after hyponatremia correction in Case #1. Details of the hyperintense lesion in the median portion of the pons in T2-FLAIR sequences (top and bottom left) and DWI sequence (bottom right)

The initial EEG recordings revealed punctate delta and theta waves, occasionally accompanied by right centro-temporo-parietal spikes that tended to spread to other regions. Over time, the EEG tracings deteriorated, showing diffuse slowing and paroxysmal abnormalities, particularly on the right side, which were suggestive of nonconvulsive status epilepticus. This condition responded to the administration of lacosamide. Subsequently, the EEG displayed a burst-suppression pattern.

Case #2

Clinical manifestations

An 81-year-old female patient with a medical history of hypothyroidism, hypertension, major depression, and stage 3 chronic kidney disease presented initially with mild difficulty walking. A few days later, blood tests revealed a serum sodium level of 100 mmol/L. The clinical condition subsequently worsened, manifesting as an inability to walk, confusion, and lethargy. Upon admission to the ED, the patient was found to be amimic and uncooperative. Blood sodium levels were confirmed to be below 100 mmol/L along with concurrent hypokalemia (K⁺ 2.7 mmol/L). A brain CT scan revealed no acute abnormalities but did show evidence of chronic cerebral vasculopathy. It was hypothesized that the decrease in serum sodium levels might have been induced by a thiazide diuretic (hydrochlorothiazide), which was immediately discontinued. The presentation was thus presumed to be due to chronic hyponatremia.

Therapy was initiated with 500 mL of 0.9% saline solution and 2 vials of NaCl, infused at a rate of 21 ml/h, along with 1 vial of KCl diluted in 500 mL of saline. Electrolyte levels were monitored at scheduled intervals (Fig. 4). Prompt clinical improvement was observed, leading to an increase in the infusion rate to 83 ml/h the same day. Approximately 12 h after the initiation of therapy, serum sodium had risen to 109 mmol/L, while potassium was measured at 2.2 mmol/L. The infusion rate of KCl was then adjusted. After a further 12 h, blood tests indicated an increase in sodium and potassium levels. Despite the rapid correction, hypertonic intravenous therapy was continued into the following day. Clinically, the patient exhibited only mild psychomotor slowing. On the second day of hospitalisation, a serum osmolarity of 261.6 mOsm/L was detected (normal range 275–295 mOsm/L).

Fig. 4.

Timeline of blood sodium and potassium levels during hospitalization in Case #2

The patient was then transferred from the ED to the nephrology unit, where, due to a further increase in sodium levels, the infusion regimen was adjusted to include 2 vials of KCl in 3 L of 0.9% saline solution administered at a rate of 60 ml/h. Subsequently, despite an increase in serum sodium levels of 29 mmol/L over 48 h, the scheduled infusion therapy was maintained.

The following day, in addition to an increase in sodium levels, a pH of 7.25 and bicarbonate (HCO₃⁻) levels of 19.3 mEq/L were noted. Intravenous therapy continued with a prescription for a 1.4% NaHCO₃ solution. However, instead of the prescribed solution, an 8.4% NaHCO₃ solution was inadvertently administered.

As a result, from the fourth day of hospitalization, the patient’s clinical status deteriorated significantly, presenting with drowsiness and minimal response to stimuli. Sodium levels continued to rise, reaching hypernatremia (155 mmol/L) accompanied by metabolic alkalosis. In parallel with the clinical neurological deterioration, laboratory tests on day four showed an increase in serum osmolality (322.2 mOsm/L). For approximately one week thereafter, sodium levels remained elevated, and hypotonic solutions were administered. Subsequently, sodium levels stabilized within the normal range. Neurological consultation was requested on the fourth, fifth and ninth days of hospitalisation. The first neurological consultation, on the fourth day of hospitalisation, recommended a brain CT scan for a suspected acute cerebrovascular event. He also added ASA (acetylsalicylic acid) 100 mg to the therapy. As the CT scan was negative for an acute cerebrovascular event, the second neurological follow-up recommended another CT scan and also scheduled a brain MRI. The diagnosis of pontine myelinolysis was confirmed at the neurological examination on the ninth day of hospitalisation. The clinical and instrumental data (brain MRI) were consistent with cerebral edema secondary to overcorrection of hyponatremia.

A clinical examination conducted 18 days after sodium correction revealed that the patient was non-alert, unconscious, and uncooperative. The examination also noted eye opening in response to verbal and painful stimuli, absence of response to verbal commands, lack of pupillary reactivity, and flaccid areflexic tetraparesis.

In the subsequent days, dermatological manifestations (facial erythema with diffuse blistering lesions and edema), along with hypothermia and anemia, emerged. No investigations to assess cortisol levels, TSH, FT3 and FT4 levels or urine osmolality were documented in the medical records.

Ultimately, the patient deceased 47 days after hospital admission due to irreversible coma following ODS, which was caused by the overcorrection of severe hyponatremia.

MRI and EEG evidence

A brain MRI performed 5 days after clinical deterioration revealed T2-FLAIR hyperintensities with diffusion restriction involving the corticospinal junction bilaterally, the thalamic regions, and notably the midbrain and pons. Further, an MRI performed 8 days later revealed an increased extent of signal abnormalities with a bilateral and symmetrical distribution. These abnormalities predominantly affected the pontine and midbrain regions, with partial involvement of the cerebellum, basal ganglia, thalami, and both subcortical and periventricular white matter. The MRI also demonstrated the typical “piglet” sign. (Fig. 5). An EEG performed during hospitalization had revealed slowing of background activity.

Fig. 5.

MR images acquired 13 days after hyponatremia correction in Case #2. Bilateral and symmetrical abnormalities affecting the pontine and midbrain regions (“piglet” sign) in T1 (top left), T2-FLAIR (top right), DWI (bottom left), and T2-FSE (bottom right)

Autopsy and histopathology findings

An autopsy was conducted two days after death. At gross examination, the brain was normal in shape and size, with slightly reduced weight and moderately reduced consistency. Although it appeared morphologically normal and symmetric, it showed signs of congestion and edema. The midbrain, pons, and medulla oblongata showed no macroscopic pathological alterations. Consequently, the entire brain was fixed in formalin for further macroscopic and microscopic examination. Upon dissection of the brainstem, areas of grayish color were noted at the base of the pons, surrounded by brain tissue of normal morphology and color (Fig. 6).

Fig. 6.

Gross examination of the brain fixed in formalin, showing areas of grayish discoloration at the base of the pons

The samples underwent H&E staining. In the brain and brainstem, there were findings of pericellular and perivascular edema, along with areas indicating prior ischemia and reactive gliosis. Deeper within the brainstem, reduced cellularity was noted, accompanied by optically empty vacuoles in the parenchyma, sparse representation of nervous cells, numerous reactive glial cells, and occasional perivascular macrophages and small lymphocytes. Intense blood stasis was also evident in the region (Fig. 7).

Fig. 7.

Histopathological findings in the brainstem showing reduced cellularity, optically empty vacuoles in the parenchyma, and glial reactivity (H&E, 10x)

Subsequent Luxol Fast Blue staining, which targets myelin, revealed areas of reduced parenchymal staining with a symmetrical distribution in the central regions of the brainstem, midbrain, and pons (Fig. 8). The staining also showed diffuse rarefaction of neurofibrils and several optically empty areas disrupting the nerve sheaths. Additionally, an abundance of reactive glial cells surrounded by large optically empty vacuoles was appreciable.

Fig. 8.

Histopathological findings in the brainstem showing diffuse rarefaction of neurofibrils and several optically empty areas disrupting the nerve sheath (Luxol Fast Blue)

Discussion

Traditionally, ODS is described as a rare neurological disorder caused by several conditions, mostly accounting for rapid osmolar shifts, more frequent in males with a mean age of about 50 years old [5].

In the US, it is estimated that ODS accounts for 0.4–0.56% of all neurological admissions to tertiary referral hospitals [6] and 0.06% of all medical hospital admissions [7]. Post-mortem MRI-based studies on consecutive unselected autopsies have described an incidence of 0.3–1.1%, suggesting that it is still an under-diagnosed condition. In contrast, at-risk groups such as patients with chronic liver disease show incidence of 9.5%, with rates of 9.8–29% after liver transplantation [8]. A 2015 clinical study found a 2.5% incidence of ODS among patients in intensive care units [9].

The etiology of CPM and EPM is not fully understood, but evidence indicates that ODS is a serious consequence of overly rapid correction of hyponatremia. Several studies have identified non-modifiable factors that may increase the risk of developing CPM or EPM when treating hyponatremia, including initial sodium blood levels below 120 mmol/l; alcoholism, which is the most commonly associated condition [5]; malnutrition; liver transplantation or cirrhosis; end-stage renal disease; diabetes mellitus; concomitant hypokalemia or hypophosphatemia [10–12]. Furthermore, a significant risk factor is the correction of chronic hyponatremia, defined as circulating sodium deficiency developed in more than 48 h [13]. Yet, among physicians, it is commonly accepted that the most relevant trigger factor is a rapid osmolar shift, especially sodium correction rate. Nevertheless, since the evidence on this topic is limited, international guidelines disagree about the definition of such overcorrection [14]. The US Recommendations state that overcorrection occurs when sodium blood levels increase more than 10–12 mmol/L in 24 h or 18 mmol/L within 48 h (for patients at high risk of developing ODS the threshold is lowered to 8 mmol/L in 24 h) [15]. Differently, according to European Guidelines the correction exceeds therapeutic limits if sodium concentration rises more than 10 mmol/L in the first 24 h or, after that delicate period, more than additional 8 mmol/L in 24 h [13].

It is important to emphasize that these thresholds represent the upper limits rather than the treatment goals. It is recommended to avoid exceeding these limits, particularly in patients with additional risk factors for ODS. Furthermore, due to concerns about the risk of neurological damage, some authors suggest restricting the daily increase in sodium levels to 6–8 mmol/L, regardless of the severity of hyponatremia.

When managing highly symptomatic patients, the “rule of sixes” should be applied, aiming to achieve a daily correction of 6 mmol/L within the first six hours of treatment initiation, with continuation of therapy on subsequent days as needed [16].

As previously stated, the risk of developing ODS is higher if hyponatremia is caused by a reversible condition, due to the onset of hypotonic diuresis. However, this factor may be managed thanks to the administration of desmopressin along with hypertonic saline solution [16].

Strict monitoring of blood sodium levels and urine output is crucial during the treatment of hyponatremia, with checks recommended at least every 6 h. In case of an increase in urine output, it is advised to transition to more frequent monitoring of serum sodium levels, ideally every 2 h [13].

To avoid excessively rapid correction of blood sodium levels, several modern formulas have been developed, including the following equations: (i) the Adrogué–Madias, (ii) the Barsoum–Levine, (iii) the Electrolyte Free Water Clearance, and (iv) the Nguyen–Kurtz (Table 1) [17]. All of these equations are derived from the original equation proposed by Edelman in 1958 [18]. This equation was based on exchangeable sodium (eNa⁺), exchangeable potassium (eK⁺), and total body water (TBW), with α and β being constants extrapolated from linear regression, as follows:

Table 1.

Equations to determine serum sodium correction rate

Adrogué–Madias

Nguyen–Kurtz

Barsoum–Levine

Electrolyte Free Water Clearance

Total body water

Abbreviations: ΔV (volume change); EFWC (Electrolyte Free Water Clearance); [K]i (concentration of potassium in input fluid); [K]o (concentration of potassium in output fluid); [K]urine (potassium in urine); Na₁ (original sodium); Na₂ (predicted sodium); [Na]I (concentration of sodium input fluid); [Na]o (concentration of sodium in output fluid); [Na]urine (urinary sodium concentration); s[Glucose] (serum glucose concentration); TBW (total body water; ) Vi (volume of input fluids); Vo (volume of output fluids); Vurine (volume of urine)

The limitation of the Edelman equation lies in its inability to precisely determine the constants for every patient due to multiple disturbances from acid-base or osmotic disorders [19].

These formulas allow for predicting blood sodium levels after fluid infusion is given initial concentration, type and amount of fluid infusion and total body water.

Nevertheless, a 2017 Consensus Statement released by the Italian Societies of Endocrinology, Nephrology, and Oncology states that, while formulas may be useful for calculating the initial infusion rate at the beginning of treatment, these equations are not fully reliable for mitigating the risk of ODS. The infusion rate should be subsequently adjusted based on the patient’s characteristics and their response to therapy. These adjustments should consider additional factors, such as the reactivation of normal renal response to osmotic stimuli or the concomitant correction of hypokalemia, which might contribute to an increase in sodium levels [20].

When hyponatremia occurs, due to the osmolarity gap between plasma and the brain, water tends to move from the extracellular space to the intracellular space, causing nerve cells to swell. Eventually, over time, brain cells attempt to adapt to these environmental changes [13]. To restore osmotic equilibrium between intracellular and extracellular fluids, brain cells typically shed osmotically active particles—initially electrolytes and later organic osmolytes like myo-inositol. Typically, this equilibrium is restored within 24 to 48 h from the onset of the process [4]. After this period, hyponatremia is considered “chronic”, and brain tissue becomes highly at risk for demyelination. Following an excessively rapid infusion of electrolytes (such as 3% NaCl saline injection), there is an increase in brain electrolytes alongside a decrease in organic osmolytes, resulting in water leakage and cell shrinkage, which triggers apoptosis, particularly in oligodendrocytes with a low apoptotic threshold. Additionally, in response to sudden osmotic stress, neuronal cells tend to release myelotoxic agents. These, along with inflammatory cells and activated glial cells, contribute to damage to the myelin sheath [5].

Another histopathological finding that aids in understanding the pathology of this disease, alongside the loss of myelin and programmed death of oligodendrocytes, is the infiltration of macrophages, which degrade myelin. The most vulnerable structures to damage are those with high concentrations of myelin and oligodendrocytes, such as the pons, which contain a dense network of crossing and descending myelin fibers and a transitional area between white and gray matter. Additionally, other potentially affected cerebral areas, such as the cerebellum, midbrain, thalamus, lateral geniculate body, external and extreme capsules, hippocampus, and basal ganglia, exhibit similar features [7].

The shift of osmotic pressure occurs more rapidly in conditions where the underlying cause of hyponatremia is reversible, allowing the kidneys to resume normal function and produce highly diluted urine. Examples include hypovolemia, cortisol deficiency, or the use of certain drugs (e.g., desmopressin, thiazides). Traditionally, inflammation was not considered a significant factor in the disease’s pathogenesis. However, the detection of macrophages, reactive astrocytes, and myelotoxic substances in demyelinated areas suggests it is now an active area of research.

At necropsy, a red triangular area, with softened consistency in the central part of the pons is reported, surrounded by intact parenchyma [21]. According to the first autopsy description by Adams et al. [1], following fixation of the organs in formalin, a “grayish” color and “unusually soft” texture are reported in the affected pontine areas. These areas exhibit variable shapes, which may resemble the ‘bat-wing shape’ observed in MRI scans.

Histologically, after myelin-specific staining, the area appears pale compared to surrounding tissues, indicating a complete absence of myelin. The borders appear well-defined but irregular, extending into the surrounding parenchyma without any apparent vascular connection. While myelin is absent in the lesion’s core, swollen myelin sheaths and fragments can be detected in the peripheral areas [2].

Nevertheless, apoptosis and loss of oligodendrocytes are considered the primary characteristics of CPM, even though the exact mechanisms remain unclear. These factors are typically regarded as the principal indicators of the disease. Other histological findings include abundant infiltration of large macrophages, particularly in the peripheral areas, containing products from myelin degradation. Occasionally, there is a minimal reaction of T lymphocytes within demyelinating lesions and around blood vessels. These findings can help distinguish acute lesions from remote ones, as the latter typically lack evidence of macrophages.

As soon as activated astrocytes were found in the affected area, different explanations and theories were proposed. Popescu et al. [22] proposed that astrocytes may contribute to the development of CPM. They noted a reduced expression of astrocytic Aquaporins AQ1 and AQ4 in demyelinated areas, specifically in male subjects. It remains unclear whether this reduction may represent a protective response of astrocytes against apoptosis [23]. The appearance and number of axons and nerve cells are generally unaffected, although a few may appear pale, and axonal swelling may be detected.

There is no classical presentation of CPM and EPM, and clinical manifestations can vary depending on which part of the brain is affected. However, the presentation of CPM typically occurs in two stages. In the first stage, there may be encephalopathy, ranging from mild to severe, including seizures and coma in rare cases, due to the electrolyte disequilibrium [5]. These symptoms may initially improve as sodium levels in the blood normalize. In the second stage, various disorders may occur, such as quadriplegia, first with flaccid quadriparesis and then, when the upper motoneuron becomes predominant, spastic paresis, eventually accompanied by clonus and positive Babinski sign.

Other possible clinical manifestations of CPM are pseudobulbar palsy, dysarthria, dysphagia, and ophthalmoplegia. In severe cases, ODS may progress to a locked-in syndrome, a condition in which only blinking and vertical eye movements are possible [7].

If extrapontine structures are involved, ODS might manifest with various movement disorders, such as mutism, tremor, gait disorders, chorea, cogwheel rigidity, dystonia, and involuntary jerking movements of the abdomen known as “Belly dancer’s syndrome”. Additionally, symptoms may include ataxia, catatonia, seizures, altered mental status, and in some cases, patients may be oligosymptomatic with milder symptoms such as lethargy, confusion, emotional instability, altered mental status, decreased level of consciousness, or disorders of conjugate gaze [4, 21, 24] (Table 2).

Table 2.

Clinical symptoms by observational topography and relative classification

CPM	EPM
Symptoms	Movement disorders	General symptoms
Mental status alteration, coma	Dystonia, cogwheel rigidity	Encephalopathy, lethargy, confusion
Paresis (flaccid and spastic)	Tremor, chorea, ataxia	Seizures
Locked-in syndrome	Akinetic-rigid disorders	Mutism
Pupillary disorders	Other gait disorders	Depression, apathy
Eye motility disorders, ophthalmoplegia	Involuntary movements	Emotional dysregulation
Impair of reflexes		Dementia
Dysarthria, dysphagia

The timing of symptom onset is not clearly defined, but typically symptoms begin within a few days after the triggering factor, ranging from 1 day to 2 weeks [7]. The prevalence of asymptomatic cases remains unclear, as they are often diagnosed randomly through autopsy or imaging studies. During autopsies of asymptomatic patients, CPM is estimated to be found in approximately 0.5% of brains [25].

The most sensitive diagnostic modality is brain MRI, which allows to prove the presence of demyelination even in patients with milder presentation of the disease. Demyelination sites are identified as hyperintense areas on T2-weighted and T2-FLAIR sequences and as corresponding hypointense areas in T1-weighted sequences. These findings are more often localized in the central part of the pons in cases of CPM, but also in various extrapontine brain areas, such as the cerebellum, thalamus, and basal ganglia in EPM [6, 7]. The distribution of the lesions in the pons is typically central, roughly symmetrical, and presents with oval, bat-wing, trident, or pig-snout shape (“piglet” or “trident” sign in axial T2 or FLAIR sequences), as corticobulbar and descending corticospinal tracts, corresponding respectively to tegmentum and ventrolateral pons, are partially not involved [26].

Traditionally, the prognosis of ODS showed an extremely poor outcome. Initially, mortality rates were as high as 50%, and most diagnoses were made postmortem [27]. Nevertheless, more recent reports indicate a more favorable prognosis. Improved understanding of the pathophysiology, advancements in intensive care treatments, and the introduction of modern diagnostic techniques, especially MRI, have facilitated the diagnosis of milder forms and early identification and treatment of the syndrome. These achievements have led to clinical improvement and a significant increase in survival rates [28]. According to a retrospective observational study by Fitts et al. [29], after 6 months from diagnosis, 16% of patients had passed away, while approximately 60% experienced mild impairment or no disability at all. Patients with additional risk factors, such as those who have undergone liver transplantation, tend to have a worse outcome.

Some authors suggest that if overly rapid correction of hyponatremia occurs, a possible solution to re-lower sodium blood levels and prevent ODS could be the infusion of 5% glucose solution, which does not contain electrolytes, or the administration of desmopressin to avoid aquaresis [13].

Other possible options could be myo-inositol, urea, and high doses of corticosteroids, but these have only been tested in experimental animal models [30].

About the specific treatment of ODS, there are no standardized guidelines, due to the heterogeneity of the syndrome and the lack of clinical trials; the only data found in the literature consists of some case reports. Treatment with plasmapheresis, started with the aim to reduce the amount of circulating myelotoxic substances in ODS, is reported in some case series, with variable outcomes, from residual motor symptoms to complete neurological recovery [31]. The potential therapeutic effect of immunomodulatory drugs, such as intravenous immunoglobulin and steroids has also been reported [23].

In summary, the proposed case analysis offers critical learning opportunities that enhance understanding of risk factors, diagnostic challenges, imaging techniques, management strategies, and the value of a multidisciplinary approach. Similar insights could contribute to more effective prevention, diagnosis, and treatment of ODS, ultimately benefiting patient care and clinical practice.

One of the crucial learning points is the importance of early recognition of risk factors. Rapid correction of hyponatremia, particularly in patients with chronic low sodium levels, is a well-established risk factor for ODS. Case studies often emphasize the need for careful monitoring of serum sodium levels and adherence to recommended correction rates. Understanding these risk factors can significantly improve clinical management and prevent the onset of ODS.

ODS frequently presents with a range of neurological symptoms that can mimic other conditions, making diagnosis challenging. Learning from case presentations underscores the necessity of considering ODS in the differential diagnosis when patients exhibit unexplained neurological deficits following sodium correction. Detailed analysis of the clinical presentation and progression of symptoms can aid clinicians in distinguishing ODS from other neurological disorders.

Further, traditional imaging techniques, such as MRI, may show subtle changes in the brain’s structure that are characteristic of ODS. Understanding the imaging findings and their evolution over time is essential. Learning points from these cases highlight the importance of using advanced imaging modalities and recognizing patterns associated with ODS, such as CPM, to aid in accurate diagnosis [31].

Moreover, the case presentations provide valuable insights into effective management strategies for patients at risk of ODS. Key takeaways include the importance of gradual correction of sodium levels and the implementation of monitoring protocols to prevent overcorrection. Learning from these cases can guide the development of best practices and protocols to minimize the risk of ODS and improve patient outcomes.

Effective management of ODS often requires a multidisciplinary approach involving neurologists, endocrinologists, and other specialists. Case presentations can demonstrate the benefits of collaboration among healthcare providers in addressing complex cases of ODS. Such an interdisciplinary approach ensures comprehensive care and facilitates better management of both the underlying condition and the resulting neurological symptoms.

Conclusively, the diagnosis of ODS remains a significant challenge due to its nonspecific early clinical signs and the overlap of its imaging features with other neurological conditions. Early recognition is critical for preventing irreversible damage, yet the lack of specific early biomarkers complicates timely diagnosis.

Conclusion

The present paper highlights the importance of a thorough clinical assessment, including careful monitoring of serum sodium levels and considering ODS in the differential diagnosis when patients present with neurological symptoms following rapid correction of hyponatremia. Advanced imaging techniques and awareness of subtle diagnostic clues can aid in distinguishing ODS from other conditions. Future research should focus on developing more specific diagnostic criteria and algorithms to facilitate early detection and improve patient outcomes.

Abbreviations

ARF

Acute Respiratory Failure

CPM

Central Pontine Myelinolysis

CT

Computed tomography

ECG

Electrocardiogram

ED

Emergency Department

EEG

Electroencephalogram

EPM

Extrapontine Myelinolysis

FLAIR

Fluid-Attenuated Inversion Recovery

H&E

Hematoxylin and Eosin

MRI

Magnetic Resonance Imaging

ODS

Osmotic Demyelination Syndrome

TBW

Total Body Water

Author contributions

B.T. and F.C. conteptualized the study. V.F. and A.S. designed the methodology. G.D., F.C., and M.P. collected the data. F.C., A.S., and V.F. found the resources. B.T., G.D., and D.M. wrote the original draft. M.S., A.S., and V.F. reviewed and edited the final version of the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The processing of the data presented in this paper is covered by the general authorization to process personal data for scientific research purposes granted by the Italian Data Protection Authority (1 March 2012 as published in Italy’s Official Journal no. 72 dated 26 March 2012) since the data do not entail any significant personalized impact on the subjects. As the study does not involve the application of experimental protocols, approval from an institutional and/or licensing committee, is not required. The patients are deceased and are legal cases. In case 2 prosecutors opened an investigation; an autopsy was ordered to clarify the exact cause of death. Italian legislation does not require consent for publishing this type of study.

Consent for publication

All co-authors of the present manuscript can certify that it has not been submitted to more than one journal for simultaneous consideration and that the manuscript has not been published previously (partly or in full). The authors also can certify that our main study is not split into several parts to increase the number of submissions, that none of the data presented here have been fabricated or manipulated, and that we present our data/text/theories/ideas. All authors and authorities have explicitly provided their consent to submit the present manuscript; in general, we all agree with the ethical responsibilities of the authors of the journal. Finally, all authors permit publication in BMC Neurology.

Competing interests

The authors declare no competing interests.