#NephMadness 2018: Hyponatremia Region
Submit your picks! | NephMadness 2018 | #NephMadness | #HyponatremiaRegion
Selection Committee Member for the Hyponatremia Region:
Richard H. Sterns, MD
Dr. Sterns is a Professor Emeritus of Medicine at the University of Rochester School of Medicine and Dentistry. He is the author of several papers and book chapters on disorders of water balance and is the section editor for fluids and electrolytes for UpToDate.
Competitors for the Hyponatremia Region
European Guidelines vs US Guidelines
Cerebral Salt Wasting vs SIADH
European Guidelines vs US Guidelines
If you follow Steph Curry’s advice zealously and drink a lot of water as it “tastes amazing and pours immediately,” you might meet Mr. Hyponatremia. Hyponatremia is the bane of residents and attendings everywhere. Hyponatremia has been known to induce frothing and stupor in both the afflicted and the providers. Faced with a patient with hyponatremia, many residents reach weakly for a version of this useful algorithm, and after wrestling for a while with a calculator and sundry formulae, decide to consult the salt whisperers. Does it really have to be this way?
In 2014, NephMadness tackled hypertonic saline and vaptans (with vaptans falling by the wayside quickly). Since then, here have been 2 sets of guidelines with over a 100 pages of explicit and detailed advice on management of this vexing issue, so it is time for us to have another go at hyponatremia.
Perhaps the most important part is that both guidelines agree on a limited daily increase in serum sodium even in severely symptomatic hyponatremia (unlike in the 1980s, when the target was to bring sodium up to 128 mmol/L in a day)! And hypertonic saline is the fastest way to fix it on both sides of the Atlantic, as it has been since 1938.
However, beyond oedema and edema, there are some notable differences in the recommendations—despite almost the same evidence that was reviewed. And just like before every Olympics when basketball fans have to brush up on international versus NBA rules, the question of the day is what do you prefer.
Do you prefer the Europeans with their streamlined and graded algorithmic advice or the Americans with detailed explanations and ungraded boxes of advice? Do you agree with trials of tolvaptan that results in a recommendation against their use and case series driving a urea recommendation or do you think the funding source and a tolvaptan recommendation are uncomfortably aligned?
Read our scouting reports to help make up your mind.
The European guidelines were written by 17 authors from all over Europe (including Montreal) and were developed and funded by 3 European societies: the European Society of Endocrinology, the European Society of Intensive Care Medicine, and the European Renal Association–European Dialysis and Transplant Association (ERA-EDTA). They were discussed on a NephJC chat when they originally came out in 2014.
The EU guidelines contain 7 figures, 10 tables, and 18 boxes of recommendations. They follow the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) methodology and style (i.e., 1 and 2 on strength and A to D for quality). Their structure is also quite reader-friendly. Following the initial description of methods, each section then has 3 subheadings:
- Why this question?
- What did we find?
- How did we translate this into the diagnostic/therapeutic strategy?
Each section also ends with a helpful ‘Questions for future research.’ There are 2 simple algorithms (Figure 6 for asymptomatic, and Figure 7 for severe symptomatic) which quickly tell you which section to go to in case you are flummoxed with a hyponatremic patient seizing in front of you.
Classification: Symptoms and Acuity Matter
These guidelines are simple to follow. Figure 7 on page G27 starts by classifying the symptoms (severe or moderately severe) and the acuity of the hyponatremia (acute or chronic). The algorithm directs the reader from those categories to the emergent management of the same (see below). Only in the absence of symptoms and acute hyponatremia does the algorithm take you to the diagnostic pathways, which uses urine osmolality and sodium concentration to build a differential diagnosis. For someone looking for quick answers on how to treat a patient, the algorithms and directions to the appropriate section are time-saving and useful.
Defining Acute and Symptomatic Hyponatremia
The Europeans use the simplest definition of “acute.” Look at your watch. If the hyponatremia began less than 48 hours ago, it is acute hyponatremia. If it is longer than 48 hours, or you can’t determine the time of onset, the Europeans consider it is chronic hyponatremia. Table 8 lists clinical settings where the hyponatremia is likely to be acute (eg, colonoscopy prep, exercise, post-operative).
But what exactly is symptomatic hyponatremia? And how does one differentiate moderately severe symptoms from severe symptoms? Check out Table 5 for the list:
- Severe hyponatremia: vomiting, cardiorespiratory distress, abnormal and deep somnolence, seizures, coma with a GCS < 8 (important note—the Americans are wishy-washy on vomiting as a symptom in their guidelines!)
- Moderately severe hyponatremia: nausea (without vomiting), confusion, headache
Treating Symptomatic Hyponatremia
One of the goals of the European guidelines was to simplify the treatment of hyponatremia. Mission accomplished. Treatment of symptomatic hyponatremia in 3 steps:
- Give 150 mL of 3% hypertonic saline over 20 minutes.
- After the bolus, recheck the serum sodium and give another 150 mL of 3% without waiting for the repeat sodium level to result. This provides some data about how much the sodium will move in response to 150 mL of 3% but the patient receives 300 mL of 3% without delay.
- The goal is to raise the sodium 5 mmol/L. Additional 150 mL boluses of 3% should be given until the sodium hits that target.
All of this should be done in a setting where close clinical/biochemical monitoring is possible. No formulae. No calculations. Just repeated boluses of 3% saline and watch the serum sodium. Fast, decisive, and effective therapy. If you are looking for solid prospective, randomized trials to validate this approach, you will be disappointed. The data is primarily based on case series and retrospective cohort studies. There is one prospective uncontrolled trial which used a 100 mL of 3% saline over 4 hours and reported a 2-mmol/L serum sodium increase. Hyponatremia is an evidence desert. Chalk this up as a 1D recommendation. An ongoing trial pits intermittent boluses to slower continuous infusions.
The guideline has some caveats:
- A discussion on how to deal with concomitant hypokalemia
- Use of weight-based 3% NaCl dosing (2 mL/kg) for patients who are very large or very small
- Caution about overcorrection—especially if using the Adrogue-Madias change in sodium formula
The caution with the use of this formula stems from the formula not really accounting for urine output and empirical data showing the sodium often (usually) rises quite a bit higher than one would expect from the calculation.
For patients with less severe symptoms, the guidelines are a bit less aggressive but still lean on 150-mL boluses of 3% to slowly increase the sodium. For mildly symptomatic patients, instead of two 150-mL boluses to kick off the treatment, they recommend a single 150-mL dose. Then with repeat sodium assessment at 1, 6, 12, and 24 hours. One should also spend more time, given that the situation is not life-threatening, to do some diagnostic workup, stop contributory medications, and fix hypokalemia.
The Problem of Overcorrection
Overcorrection and the possibility of causing quadriplegia is what causes night sweats and palpitations every time one orders hypertonic saline. This is especially likely when the kidney’s ability to excrete free water is suddenly restored while the patient is receiving hypertonic saline. A kidney able to excrete excess free water will fix hyponatremia on its own, thank you very much, without any need for hypertonic saline. If that ability is restored in the middle of pharmaceutical correction, the sodium will shoot up, disrupting the best-laid plans. When this happens, doctors have a few options:
- Rapid infusions of electrolyte-free water
- Infusions of electrolyte-free water plus desmopressin (DDAVP)
In section 7.5, the guidelines discuss this issue but do not make any recommendations. They punt by suggesting physicians seek additional expertise when in such a predicament. Somewhat conservative advice, to do no harm with zealous adventures, until better data are available. Given these guidelines are written by nephrology/endocrinology/intensive care experts, one might wonder whose additional expertise we should refer to?
Tolvaptan and Urea
In contrast to the American guidelines, which seem to favor vaptans in SIADH but also provide a lukewarm discussion of urea, the European guidelines explicitly advise against vaptans, as a grade 1C recommendation. They review the literature in detail, reporting that vaptans increase the serum sodium number, but with no improvement in mortality, and no reports of patient-reported outcome measures. This is a bit strange because they had no trouble advising use of 3%, which also has the same deficiencies.
In addition, vaptans improve the numbers too rapidly perhaps, with 2.5 to 3 times greater risk of rapid correction compared with placebo. Though no published reports of the dreaded osmotic demyelination syndrome exist after tolvaptan use, there is an advisory on the higher risk issued by the manufacturer. There actually is one case of CPM with tolvaptan but the case was so mismanaged that it is hard to fault the drug (Pro Tip: when the sodium rises 16 mmol/L in 24 hours and the patient is making 12 liters of urine, don’t double the dose of tolvaptan). Additionally, the concern of liver injury—with the FDA warning after TEMPO—creates an unfavorable risk-benefit balance against vaptans.
In contrast to the multiple prospective studies showing vaptans to be effective treatments for hyponatremia that the workgroup found unconvincing, they found the data on urea to be compelling enough to recommend it, after fluid restriction, in SIADH. As the literature consists only of multiple case reports and case series with the outcome being change in serum sodium (no data on mortality or patient-reported outcomes), it makes it in as a 2D recommendation.
The dose recommended is 0.25 to 0.5 g/kg, on the basis of multiple case series. The bitter taste, you say? Just follow a recipe they provide: “prepare the following as sachets: urea 10 g + NaHCO3 2 g + citric acid 1.5 g + sucrose 200 mg to be dissolved in 50-100 mL water. This will result in a more palatable, slightly sparkling solution.” Since then, a commercial preparation is now available, and you can see a notable nephrologist demonstrate how truly palatable:
The (Mostly) American Guidelines
These are not really ‘American’ but written by a group of preeminent American electrolyte experts (along with an Irishman). The guidelines were conceived at Tufts and published in the American Journal of Medicine, so they are The American Guidelines. These were funded by an unrestricted grant from Otsuka (maker of Tolvaptan). Additionally, 5 of 7 authors have disclosures with Otsuka—something to keep in mind.
The guideline document is truly exhaustive, with 5 tables, 3 figures, and the ‘Expert Panel Recommendations’ highlighted in 18 boxes. No algorithms here—and only one mathematical formula. They start off with a clinical significance and classifications section, before diving into details of therapy.
Cause and mechanism matter just as much as symptoms and acuity
The organization of the management of hyponatremia is based on the etiology of hyponatremia—with detailed discussion for each etiology, ranging from SIADH to thiazide diuretics. Acuity and chronicity is important and is addressed with each etiology individually.
- Thiazide diuretic-induced hyponatremia is always chronic. Stopping the drug and correcting the volume deficit will fix the hyponatremia but may result in rapid uncontrolled correction. Practitioners need to be very cautious in these patients to avoid rapid correction of hyponatremia.
- In heart failure, hyponatremia is always chronic, and in severe, symptomatic cases, hypertonic saline needs to be given with furosemide.
- Exercise-associated hyponatremia is always acute, and treatment starts 100 mL of 3% saline which can repeated twice (up to 300 mL) to arrest symptoms.
This disease and condition-specific discussion of the nuances of management is meticulously presented throughout the guidelines. If you know what you are dealing with, this organization is very useful for the advanced practitioner.
In cases where seizures and/or coma accompany hyponatremia, regardless of chronicity, the recommendation is to go ahead and give 3% hypertonic saline over 10 minutes, and repeat up to 3 times as needed. The same prescription also applies in a few other settings: when the duration of hyponatremia is known to be acute (<48 hours); with associated intracranial pathology or increased intracranial pressure; and with self-induced acute water intoxication (e.g., endurance exercise, ecstasy use, psychiatric conditions).
The goal is to increase the sodium urgently by 4-6 mmol/L to prevent neurological damage. This amount of correction has been shown to reverse clinical signs of herniation and reduce intracranial pressure by 50%. In case of mild to moderate symptoms, a lower rate of 0.5-2 ml/kg/h could be infused. While the Europeans state that vomiting indicates severe symptomatic hyponatremia, the Americans are quiet on that particular symptom. Seems wise as vomiting can indeed be a cause, but also a consequence of hyponatremia.
The Problem of Overcorrection
A lot of attention is paid to explain this issue and discuss various ways of ameliorating rapid correction. There is robust data supporting the association of rapid correction of hyponatremia and the development of myelinolysis in chronic hyponatremia. This is backed up with animal data, and even shown to occur in patients who develop acute hypernatremia. Risk factors for osmotic demyelination syndrome, beyond the speed of correction in chronic hyponatremia, include:
- very low serum sodium to begin with (≤105 mmol/L)
- concomitant hypokalemia
- advanced liver disease
Once the daily target sodium increase (8 mmol/L in high risk and 10-12 mmol/L for others) has been achieved, stop the sodium-raising measures. If the sodium continues to rise start giving free water and DDAVP.
In patients at very high risk of ODS, they discuss using the pre-emptive strategy of combined DDAVP every 6-8 hours along with the 3% hypertonic saline infusion, creating iatrogenic SIADH. This is sometimes called a “DDAVP clamp” and it prevents the body from autocorrecting the sodium and allows for a well-regulated, slow rise in sodium.
What if the sodium correction has already broken the speed limit before you are called or during treatment? In cases when the sodium was below 120 mmol/L, they recommend considering dexamethasone 4 mg every 6 hours (especially effective for iatrogenic hyponatremia in lab rats) and re-lowering sodium, with DDAVP and free water until the goal sodium has been reached.
Tolvaptan and/or Urea?
Unsurprisingly, the guidelines have a couple of dedicated pages (and Table 4) on the vasopressin receptor antagonists, with excellent coverage of the biology and clinical trial data, including SALT-1, SALT-2, SALTWATER, and EVEREST. In causes of hyponatremia that are largely driven by vasopressin, the vaptans make sense and the guideline recommends their use when fluid restriction fails or urine indices suggest that fluid restriction will not work (e.g., urine Na + K > serum Na; urine osmolality > 500 mOsm/kg H2O).
Vaptans make sense in SIADH and hypervolemic hyponatremia from heart failure. When using vaptans, one should stop employing other hyponatremia-correcting measures and lift fluid restriction in order to allow patients to drink water based on thirst. The authors point to a lack of evidence that the drugs are effective at sodium levels below 120 mmol/L.
On the other hand, the bitter taste of urea, the data from case series, and observational data alone didn’t convince the panel to recommend urea outside nephrogenic syndrome of inappropriate antidiuresis (NSIAD). Important to remember: unlike for vaptans, there are no RCTs of urea in this setting, only case series.
Cerebral Salt Wasting vs SIADH
Why does the differential diagnosis of hyponatremia when it comes to these 2 conditions matter so much? In SIADH, the first-line intervention is fluid restriction, which will make matters worse in the volume-contracted patient with salt wasting. Conversely, salt wasting syndromes can be easily fixed with intravenous normal saline, which would make hyponatremia worse if the patient has SIADH instead. Let’s nail down these dueling diagnoses in our scouting report.
Cerebral Salt Wasting
The brain-kidney axis. There is something quite elegant about cerebral injury—usually subarachnoid hemorrhage, but sometimes trauma or surgery—resulting in a neurohormonal cascade, culminating in a specific tubular effect manifesting as a perplexing hyponatremia. Befuddling the simplistic heuristic of SIADH in the neurosurgical ward, this entity was first described in 1950 (7 years before Schwartz and Bartter’s more famous SIADH paper), though it was another 30 years before it began to be begrudgingly accepted.
It also goes against the dogma of hyponatremia being usually a water disorder—in this case it is sodium that is lost—and a lot of it, as unambiguous case reports have reported urinary sodium losses of over 600 mmol/24 h. Lastly, it doesn’t help that most of the cases have been reported in the neurosurgical literature, and are not written by famous renal physiologists.
It is instructive to examine the 12 cases from the 1981 case series that revived cerebral salt wasting (CSW). These were all neurosurgical patients who had had blood volume determinations (red cell mass with chromium-55 and plasma volume with radio-iodinated serum albumin) performed pre-operatively—as well as post-operatively. The hyponatremia was mild (ranging from 122 to 134 mmol/L); however, in 10/12 cases, the total blood volume was decreased—despite the blood loss being far less contributory. The patients responded to blood transfusions and volume replacement—and not to fluid restriction.
Subsequently, a high urinary sodium and clinical assessment of volume was used to report CSW, but it should be accepted that change in volume is not always easy to determine—and even in SIADH, a natriuresis may be seen. However, another study also reports a similar story: 8 of 21 patients developed natriuresis and a negative sodium balance before the development of hyponatremia. All of these patients also had a decline in body weight, and in 6 of them, plasma volume fell by at least 10%.
At its most basic level CSW follows this timeline: a neurological event triggers sodium loss from the tubules, causing hypovolemia—and the homeostatic response, including renin, aldosterone, and appropriate ADH release result in hyponatremia. Volume assessment could potentially be useful since, unlike in SIADH, CSW results in hypovolemia; however, measurement of renin, aldosterone is not always done at the right time and has hence not been shown to be discriminating. A central venous pressure (CVP) reading also may not be helpful—a young person with head injury and normal cardiac function will have a low CVP even normally, and that is not useful as a marker of hypovolemia.
While hypovolemia should increase tubular reabsorption of uric acid and urea, this doesn’t happen in CSW because the same signal that causes sodium loss also impairs urea and uric acid reabsorption. This was elegantly reported in a case series, which demonstrates persistence of hypouricemia and increased fractional excretion of urate (FEurate) even after correction of the hyponatremia. The sodium corrected promptly with saline infusion, subsequent water loading did not cause hyponatremia, but uric acid remained low and FEurate remained high.
The figure below, based on the same case series, demonstrates generation of dilute urines with prompt correction of hyponatremia in response to normal saline infusion over 36 hours to support a true case of salt wasting, which can be contrasted with the lack of response in SIADH to just normal saline:
It is not easy to differentiate CSW from SIADH at the first encounter, but it is important to not dismiss it and keep it in the differential especially if the patient does not respond—or becomes worse with fluid restriction for a presumptive SIADH diagnosis. Careful assessment of change in volume status, reviewing the chart for fluid loss, change in weight, vital signs—along with a careful analysis of the laboratory parameters—is needed. Whether it is truly cerebral—or the term ‘renal salt wasting’ could be used instead—is indeed still arguable, as cases have been reported in non-cerebral pathologies. That’s nitpicking on terminology—but there should be no doubt in your brain about the existence, recognition, and management of this entity.
The most cerebral part of about CSW is the label ‘cerebral’, when in fact it is renal salt wasting. The CSW diagnosis is a zebra of doubtful provenance. It is also over called, over thought, and over worried about. Every time you get a consult for cerebral salt wasting from the neurosurgical unit, just go and write SIADH instead—and you will be correct 98% of the time.
In a case series of hyponatremia in neurosurgical unit, CSW was found in only 7 out of 187 cases. SIADH was found in 123. It is (Neph)madness to do otherwise. SIADH is the most common condition to think about when hyponatremia is encountered in the neurosurgical setting—and elsewhere. SIADH indeed was described 7 years after the first description of CSW, but the criteria developed by Schwartz, Bennett, Curelop, & Bartter in that 1957 report are essentially unchanged even now—because they make physiological sense.
ADH is one of the important protective mechanisms your body has against hypovolemia—in which case the rise in ADH is the antecedent cause of hyponatremia but is not inappropriate at all (e.g., with sensed ineffective arterial volume of heart failure or true volume depletion). In the case of SIADH, the inappropriateness of ADH is the elevation in the presence of normal volume status.
In both the patients in Schwartz et al’s first report, the volume status was specifically looked at and was normal—and the patients did not respond to volume, but to water restriction—and ad libitum salt intake. Though ADH could not be measured easily in 1957 (and even now, vasopressin measurement is sensitive to sample handling, storage, and assays), the presence of normal kidney function, hypertonicity of the urine in the face of hyponatremia, and hypotonicity of the extracellular fluid was shrewdly taken as evidence of the presence of ADH.
These pathophysiologic mechanisms underlie the commonly used criteria for diagnosing SIADH:
- Decreased effective osmolality of the extracellular fluid (Posm < 275 mOsmol/kg H2O)
- Inappropriate urinary concentration (Uosm > 100 mOsmol/kg H2O with normal renal function) at some level of plasma hypo-osmolality
- Clinical euvolemia, as defined by the absence of signs of hypovolemia (orthostasis, tachycardia, decreased skin turgor, dry mucous membranes) or hypervolemia (subcutaneous edema, ascites)
- Elevated urinary sodium excretion (>20-30 mmol/L) while on normal salt and water intake
- Absence of other potential causes of euvolemic hypo-osmolality: severe hypothyroidism, hypocortisolism (glucocorticoid insufficiency)
- Normal renal function and absence of diuretic use, particularly thiazide diuretics.
Supplemental criteria described sometimes include:
- Abnormal water load test (inability to excrete at least 80% of a 20 mL/kg water load in 4 h or failure to dilute Uosm to < 100 mOsm/kg H2O)
- Plasma AVP level inappropriately elevated relative to plasma osmolality
- No significant correction of serum [Na+] with volume expansion but improvement after fluid restriction
- Serum uric acid < 4 mg/dL
- Serum urea < 21.6 mg/dL
- FENa > 0.5%
- FEurate > 12%
Urine osmolality doesn’t have to be higher than plasma; even if it is > 100, it may be inappropriately concentrated enough for the existing hyponatremia. Euvolemia is essential to determine that the ADH secretion is inappropriate in the first place, and any concomitant hypovolemia has to be corrected before the diagnosis of SIADH is made. As above, although vasopressin assays are difficult and plasma copeptin represents a more stable fragment to measure for diagnosis, it still remains a research tool.
SIADH usually arises from inappropriate ADH release from the pituitary gland (as a result of stress, pain, nausea, general anesthesia, etc.), or from ectopic production (e.g., small cell lung carcinoma). Progressive hyponatremia occurs as a result of water reabsorption mediated by vasopressin V2 receptors and aquaporin 2 channels. A steady state occurs due to vasopressin escape (at a certain level of hyponatremia V2 receptor and aquaporin channel expression is downregulated so that additional water intake can be cleared).
How does one treat SIADH? Needless to say, hyponatremia with severe symptoms needs prompt management with hypertonic saline. A simple rule-of-thumb applies here when using 3% saline: using the patient’s body weight in kilograms as the rate of infusion per hour will result in approximately 1 mmol/L increase in serum sodium. Concomitant intravenous furosemide can be used in cases where this results in volume overload.
For treatment of chronic hyponatremia, management consists of choosing between ‘several suboptimal therapies’. Considering the pathophysiology, vaptans certainly make intuitive sense. Vaptans cause a brisk aquaretic effect and increase serum sodium. There is a concern, however, about too rapid a rise in serum sodium, and a lack of data showing improvement in patient-oriented outcomes (not to mention cost). This is the root of the transatlantic disagreement about their use for this indication (see discussion above on European vs American guidelines).
Fluid restriction remains the top recommendation. All fluid intake should be counted, of course, but restriction should be about 500 mL less than the average urine volume per day to see a benefit. Several days may be needed to see a significant increase, and it’s also important to allow unrestricted sodium and protein intake. Some early predictors of failure include:
- High urine osmolality (>500 mOsm/kg H20)
- (UNa + UK) > plasma Na
- 24-hour urine volume < 1500 mL
- Initial rise in serum Na < 2 mmol/L/d after fluid restriction
Urea is another option, again with contrasting opinions across US and Europe given its effect on increasing serum sodium in case series alone, with little risk of rapid increase unlike vaptans but with the added discomfort of bitter taste. Lastly, demeclocycline is an option, which actually causes a form of nephrogenic diabetes insipidus. It is unpredictable and does have a lag effect of 3-4 days. It also has an unfortunate effect of causing acute kidney injury, making it a rather unpopular choice amongst nephrologists.
Regardless of what you think about the reality of cerebral salt wasting, SIADH is a condition that every nephrologist worth their salt needs to know how to manage. It is not an easy condition to deal with. Clinicians need to use all their cerebral capacity to recognize symptoms, correct the sodium level with the appropriate level of aggressiveness, and not spend too much time chasing the salt-wasting hobgoblins.
– Post prepared by Swapnil Hiremath. Follow him @hswapnil.
How to Claim CME
US-based physicians can earn 1.0 CME credit for reading this region. Please register/log in at the NKF PERC portal. Click on “Continue,” click on the “Hyponatremia Region,” then click on “Continue” to access the evaluation. You’ll need to click on “Continue” again to complete the evaluation, after which you can claim 1.0 credit and print your certificate. The CME activity will expire on June 15th, 2018.
Submit your picks! | NephMadness 2018 | #NephMadness | #HyponatremiaRegion
Oral low dose furosemide and table salt (or salt tablets) should be included as an effective treatment of chronic SAIDH (NEJM 304:329,1981).