Use of Venous Excess UltraSound (VExUS) score in hyponatraemia management in critically ill patient

Background

Hyponatraemia is the most frequent electrolyte disturbance identified in the intensive care unit (ICU),1 representing a decreased sodium concentration below 135 mmol/L.2 Moreover, hyponatraemia may negatively impacts patients’ outcomes in the ICU setting. This is why it should not be dismissed in the neurocritical care patient because even small decreases in serum sodium level can lead to severe prognosis limitation.1 Finally, hyponatraemia management also represents a real challenge for physicians in general since severe complications can occur with either under or overcorrection of sodium levels.2

Excluding pseudohyponatraemia and hypertonic hyponatraemia is one of the primordial stages which aid identifying the cause of hyponatraemia.2

Hypotonic hyponatraemia can be triggered either by an extreme loss of sodium and other electrolytes or by an increasement in free water resulting in fewer total osmoles relative to blood volume.2 In these cases, plasma and urine osmolality comparisons are paramount,1 the same as the urine and serum sodium and the uric acid levels. All of them can demonstrate signs concerning diuretic use and antidiuretic hormone status.2 Altogether with these laboratory analysis, the essential diagnostic step consists on evaluating the patient’s volume status to categorise hyponatraemia as hypovolaemic, euvolaemic or hypervolaemic, to treat the disorder appropriately.2–4

In critical patients, precise assessment of volume status is complex, and analytical or clinical parameters alone often do not suffice.2 3 Central venous pressure (CVP) measurements in the ICU setting remain invasive, and previous studies showed a poor correlation between the real volume status and the CVP.5 6 Variables such as cumulative fluid balance or the development of peripheral oedema have significant limitations, and their relationship with systemic venous pressure is not usually proportional.2 6

In the last years, ultrasound has been used to improve the evaluation of volume status, particularly with point-of-care ultrasound (POCUS) by acute care physicians as an complement to physical examination and decision making at the bedside.2 6 7 In cases of systemic congestion, the increased volume lead to a rise in venous pressures. POCUS enables the clinicians to estimate systemic venous pressure using Doppler evaluation.6 7 Several possible indicators of volume overload and congestive process has been proposed in the literature,8 among them, the Venous Excess UltraSound score (VExUS), which was developed to detect venous congestion in the presence of a plethoric inferior vena cava, using Doppler evaluation of the hepatic vein, portal vein and intrarenal vein.7

The present case highlights the use of POCUS—and specifically VExUS—as a powerful adjunct to physical examination and laboratory evaluation to improve volume status assessment and as a useful tool in the differential diagnosis of hyponatraemia as in clinical decision making on a neurocritical patient.

Case presentation

A female in her late teens presented to the emergency department after being run over, resulting in a closed-skull traumatic brain injury. At the arrival of the prehospital emergency team, she had an initial Glasgow Coma Scale (GCS) of 4, and endotracheal intubation was performed. In the emergency department, the head CT scan showed skull base and bilateral orbital floor fractures, an acute subdural haematoma in the right frontotemporoparietal region, a traumatic subarachnoid haemorrhage, multiple cerebral contusions, and diffuse cerebral oedema. In addition, CT studies showed a sternal fracture and right lung contusion.

An intraparenchymal fibreoptic catheter was placed to measure the intracranial pressure (ICP), and she was transferred to the ICU. An arterial line and central venous catheters were placed for haemodynamic monitoring, fluid and medication management. Later, an intravenous isotonic solution at 80 mL/hour was started. Norepinephrine was begun to maintain cerebral perfusion pressure (CPP) in 70–75 mm Hg. She also received deep sedation and analgesia, same as prophylactic anticonvulsant therapy.

On hospital day 3, the patient’s ICP reached 24 mm Hg and was temporarily controlled with increased sedation and hypertonic sodium solution. Afterwards, a repeated CT head scan revealed the worsening of the right subdural haemorrhage and diffuse cerebral oedema. Subsequently, the patient was taken to the operating room, and a right-sided decompressive craniectomy was performed. Of note, intravenous mannitol administration for ICP control was needed. Postoperatively, the patient did not need hyperosmolar therapy again, with the ICP remaining within the normal range. The head CT scan showed no complications, and weaning from intravenous sedative medications was started. Due to an enteral feeding intolerance, the flow rate of continuous enteral nutrition was reduced, and she remained with 1000 mL/day of intravenous isotonic solution.

On day 6, blood analysis was notable for acute hyponatraemia, with serum sodium of 128 mmol/L, and no other electrolyte disorders (serum potassium of 3.8 mmol/L).

Investigations

Her urine sodium was 175 mmol/L (normal value: 50–133), urine potassium 10.1 mmol/L (normal value: 20–67) and urine osmolality 445 mOsm/kg (normal value: 300–900) (figure 1). Her fluid balance on the day of onset hyponatraemia was neutral, with spontaneous diuresis, and she had preserved renal function (serum creatinine of 0.4 mg/dL and serum urea of 12 mg/dL) and no peripheral oedema. Her pulmonary auscultation was symmetrical without adventitious sounds, and she remained on mechanical ventilation with a FiO₂ requirement of 25%. The patient remained sedated, requiring low-dose norepinephrine (0.1 mcg/kg/min) to maintain adequate CPP. A cerebral salt wasting syndrome (CSW) was presumed as a possible cause of hypotonic hyponatraemia. She was treated with fludrocortisone 0.2 mg/day and isotonic solution at 40 mL/hour. However, there was a worsening of the hyponatraemia, with serum sodium decreasing to 118 mg/dL. Blood analysis also revealed albumin of 2.9 g/dL, haematocrit of 27%, and urea of 15 mg/dL. Her serum osmolality was 260 mOsm/kg (normal value: 275–295), urine sodium was 130 mmol/L, urine potassium 29 mmol/L and urine osmolality 583 mOsm/kg. Although the cumulative fluid balance was positive by more than 7 L, her daily fluid balance was now slightly negative (−267 mL/24 hours) with spontaneous diuresis, and she had no peripheral oedema. The isotonic solution was replaced by 3% saline solution perfusion.

Differential diagnosis

CSW and syndrome of inappropriate antidiuretic hormone secretion (SIADH) have been defined as the most common causes of hyponatraemia in the neurocritical patient. The differential diagnosis between CSW and SIADH in critical patients is often challenging since the physical examination is limited, there are multiple confounding factors (fluid administration, medications such as mannitol), and laboratory analysis information is commonly confusing. Classically, hypovolaemia has been regarded as the factor that enables differentiating CSW, although this is difficult to diagnose in the neurocritical setting.9 In our case, acute hypotonic hyponatraemia with increased natriuresis and physical examination with no peripheral oedema or other signs suggestive of hypervolaemia was assumed to be in a hypovolaemic context, suggesting CSW. Despite treatment with fludrocortisone and 0.9% saline solution, serum sodium values continued to decrease, raising suspicions about hyponatraemia aetiology and, at this point, suggesting SIADH.

Dysfunction of the hypothalamic–hypophyseal–adrenal axis is also common in patients with severe traumatic brain injury, and hypopituitarism is frequent in young patients, as well as in those to whom etomidate, propofol, or phenobarbital have been administered. Although the patient was young and under sedation with propofol and cortisol level measurement was not required, there was no hypoglycaemic, and she remained under low-dose vasopressor to maintain adequate CPP. Although pain can be another potential stimulus for ADH release, adequate analgesia was achieved using a multimodal approach.

Hyponatraemia can also be secondary to the use of certain drugs, such as mannitol. In this case, the development of hyponatraemia occurred several days after mannitol administration for ICP control.

Treatment

In view of this situation, considering the presumed mismanagement of hyponatraemia and the presence of multiple confounding factors, POCUS was performed by an intensive care physician to adequately assess the patient’s volume status. A cardiac evaluation revealed preserved biventricular systolic function (cardiac output of 3 L/min, left ventricular outflow tract velocity time integral of 21 cm, stroke volume of 48 mL, E/A 1.7, E/E’ 9.6, TAPSE 19 mm, tricuspid S’ 13 cm/s, RV-RA gradient of 21 mm Hg, no interventricular septal flattening). Lung ultrasound showed bilateral normal A pattern and no pleural effusion. The evaluation of systemic venous congestion was performed using the VExUS score. Her inferior vena cava appeared with a maximum diameter of 21 mm and a distensibility index of 33% (figure 2A). Hepatic vein Doppler showed systolic (S) and diastolic (D) forward flow waves but with S component lower in magnitude than the D component (figure 2B). Portal vein showed 53% pulsatility (figure 2C), and renal Doppler showed a biphasic interrupted intra-renal venous flow (figure 2D). The overall VExUS pattern was grade 2, denoting moderate venous congestion (figure 3). The ultrasound findings supported hypervolaemia conditioning a change in hyponatraemia management and therapy. As a result, intravenous furosemide was started, and her diuretic response and serum sodium were closely monitored. Intermittent 3% hypertonic saline solution was administered to increase natraemia in the first hours. The ICU team closely monitored blood and urine sodium levels and diuresis to avoid overcorrection, aiming to obtain a sodium improvement of less than 10 mmol in 24 hours and less than 18 mmol over 48 hours.1 After 24 hours of strategy change and negative fluid balance, sodium had increased 10 mmol (118–128 mmol/L). Consequently, a continuous 3% hypertonic saline solution was started, and occasional furosemide boluses were performed to maintain negative fluid balance. Sodium was corrected at an appropriate rate, with a total increase of 13 mmol (119–132 mmol/L) in 48 hours. Over the same period, portal vein pulsatility decreased until it was approximately 30% (figure 4A), and the hepatic vein Doppler got to a normal pattern with S>D forward flow (figure 4B). Given the sodium decreasing whenever hypertonic saline was stopped, continuous infusion of 3% saline solution was maintained and serum sodium continued to improve in the following days.

Outcome and follow-up

The patient’s GCS improved during her ICU stay, giving leading to mechanical ventilation weaning. However, after two failed extubation attempts due to postextubation stridor, a percutaneous tracheostomy was performed. There was a respiratory deterioration due to ventilator-associated pneumonia with acute respiratory distress syndrome during the ICU stay. After the bacterial infection was resolved, GCS improved, and she was successfully weaned from the ventilator and maintained with oxygen supply by tracheostomy. After 53 days in the ICU, she was discharged to the department of neurosurgery with GCS 10 (E4, V1, M5), where she remained after that.