Case of the Month 13 – Answer

Because hyponatremia is a massive topic that could have its own dedicated blog posts, we will focus on determining the cause of hyponatremia. First thing’s first. Determine if this is true hyponatremia. That is, rule out other causes of artificially low sodium levels such as high glucose, proteins, or triglyceride levels.

Next, determine the volume status. Is this patient hypovolemic, euvolemic, or hypervolemic? People tend to get confused after this step, so let us simplify.

Hypervolemic Hyponatremia: The three most common causes of volume overload are renal, cardiac, and hepatic. When the liver or heart fails, the patients gains more water than salt and will become hyponatremic. Naturally, the kidney responds to this increase in circualting volume through the Renin-Angiotensin-Aldosterone system (RAAS) by retaining salt to maintain serum osmolality. Therefore, hypervolemic hyponatremia with concentrated urine (low urine sodium) is from CHF or cirrhosis. If the kidneys cannot retain salt, the urine sodium will be high. As such, hypervolemic hyponatremia with high urine sodium suggests intrinsic renal disease. In theory, CHF and cirrhosis will only have high urine sodium after the kidney function has decreased. I think the issue is that once the kidneys fail in their concentrating ability such as CKD 4 or 5, the urine sodium will likely be high regardless of the underlying etiology. Most people have mixed pictures but strictly speaking, a hypervolemic patient with high sodium is due to inability to concentrate in tubules.

Hypovolemic Hyponatremia: Again, ask yourself:  Are the kidneys intact and functioning normally? Unfortunately, we can’t see the exact number of functioning nephrons or the tubular function in real time but we have some decent estimates. For the purposes of identifying hyponatremia, the Glomerular Filtration Rate (GFR) can be estimated using the creatinine clearance. To do this, we need the Cockcroft-Gault formula or some other variant of your choice. If the GFR is below 60ml/min, you can safely say they have some sort of renal injury. If the kidneys are functioning normally and the urine sodium is high, the cause is likely iatrogenic from diuretics. Other causes include mineralocorticoid deficiency and cerebral salt wasting. However, in cases such as vomiting, diarrhea, and burns, you will find low urine sodium as the body tries to re-regulate the serum osmolality.

Euvolemic Hyponatremia: I view this as the category of exclusion. Don’t go there until all other options are considered. In this case, the kidneys should be functioning somewhat normally, but there is just an increase in water relative to the sodium. The body is able to maintain euvolemia as the kidney, heart, and lungs are able to redistribute the fluid appropriately. Some causes of euvolemic hyponatremia are SIADH, psychogenic polydipsia, hypothyroid, stress, and drugs. Again, urine sodium isn’t as important or diagnostic with this group because the kidneys are functioning somewhat normally.

Now back to our patient. Although I partially agree with Kaufman and Tedward, I think the primary driving force in our patient is SIADH. Good job, Lee. Don’t be fooled by the high BNP and peripheral edema. Patients who are hypervolemic from right-sided heart failure usually have a straightforward reason to the etiology of their failure, worse peripheral edema, and JVD. Additionally, in a hypervolemic state the kidneys see a decreased flow secondary to a low effective circulating volume. While they can compensate with the Renin-Angiotensin-Aldosterone system (RAAS), this backup system has usually failed by the time they get to the point of hyponatremia. These patients’ kidney dysfunction will be evident through a decreased creatinine clearance and GFR. This is not present in our case. On the other hand, patients with cirrhosis also have much more chronic picture with a lower albumin and higher INR. Don’t be fooled by common uses of specific labs. Yes, it is primarily secreted by the ventricles, but BNP is not specific to the heart. In fact, it was originally called brain-natriuretic peptide because it was identified in the brain. BNP will also rise with portal congestion and acute liver injury. The mechanism behind this is unclear and can likely be explained by increased portal pressures and endothelial dysfunction signaling release of BNP through a shared pathway. (Yilmaz, Eken, Avci, Duman 2010). Though typically thought of as a euvolemic state, SIADH can become hypervolemic. Hyponatremia is a pendulum and a balancing act of various organ systems. Normally with SIADH, the kidneys are overwhelmed with water absorption from the inappropriate Antidiuretic Hormone (ADH) secretion, but can redistribute. If the patient has a “second hit” such as a kidney, heart, or liver disease, the pendulum might start to swing from a primary euvolemic process to a hypervolemic one.

The approach to abnormal liver function tests is complex and also well beyond the scope of a single post. Briefly, whenever a patient has abnormal liver function tests, you want to determine two things. Is this acute/chronic and is this cholestatic (obstructive) or hepatocellular (intrahepatic)? In general, cholestatic pictures will have a greater elevation in Alkaline phosphatase level, whereas hepatocellular have predominantly AST/ALT elevations. The main objective of this question was to convey just how high the elevations are. This can answer both of those questions. The clue is the AST of 1793 pointing in the direction of an acute hepatocellular cause. Acute changes in LFT of > 1000 are most commonly caused by a few things: Ischemic, acute viral hepatitis (not HCV) and drug/toxin induced liver injury, and autoimmune hepatitis. Kaufman, Lee, and Tedward, you are correct. The other causes need to be ruled out. But given the recent change in medications, this is likely a case of INH toxicity. Stop the offending agent! The patient will likely improve. Why is the AST is so much higher than the ALT? Typically, we are thought to believe this is more of an alcohol hepatitis picture. Again, don’t lock in, as this is clearly not the case. Remember the hepatic acinus or functional unit of the liver. Zone 3, the area closest to the blood supply, also has the highest concentration of AST. Damage to this zone is usually from toxins and could explain the greater elevation in AST to ALT.

Given the information above, don’t neglect your physical exam. In this patient, the hint of liver pathology came from the physical exam. A painful, tender, enlarged liver is non-specific and can be caused by almost any liver pathology. In our case, this was due to INH but don’t be fooled.  This could easily have been ischemic, viral autoimmune, alcoholic, or obstructive.

I can’t help myself from offering some background information on TB treatment and liver toxicity. For starters, it is not uncommon. It was estimated that 5-28% of patients being treated for TB will have some LFT abnormality (Ramappa, & Aithal 2013). This is because three out of four anti-TB drugs can cause toxicity (isoniazide, rifampin and pyrazinamide with INH being the most common.)  Most of these cases are self-resolving and serial LFT testing is not recommended. However, in practice, they are checked every few weeks at the beginning and with longer intervals as tolerated.

As Lee alluded to, I hate inappropriate use of isolation, and in the emergency department, we are often the ones who initiate this. The decision to place a patient on airborne isolation should not be taken lightly as a patient will undoubtedly have “different” care. Don’t get me wrong, appropriate isolation is great, but take a step back before you hang that isolation sign. Make sure it is the right decision for your patient. For starters, these patients are more likely to board in the emergency department for extended periods waiting for an isolation room to become available. Providers and staff are less likely to enter the rooms due to fear or the simple inconvenience of personal protective equipment. Visitors might be afraid or unable to tolerate the additional face mask they have to wear and may be reluctant to come. This is not to mention the stigma and isolation that patients themselves feel. Many patients who are placed on isolation are those with HIV who already experience stigma. So, does this patient need airborne isolation? The CDC recommends that patients’ infectivity should be considered when deciding on isolation. Infectivity is defined as the number of mycobacterium nuclei that are shed into the air to transmit infection. Of course, we don’t actually know this number, but infectivity can be approximated and clinical judgement should be based on the following (CDC Morbidity and Mortality Weekly Report 2005):

1. Presence of cough
2. Cavity in the lung
3. Acid-fast bacilli on sputum smear
4. TB disease of the lungs, airway, or larynx
5. Patient not covering mouth and nose when coughing
6. Not receiving adequate treatment or having prolonged illness
7. Undergoing cough-inducing procedures
8. Positive sputum cultures

Yes, we know the patient already tested positive and had the above criteria.  But must isolation be continued? The answer to this is partially clinical and partially practice-specific to your institution. Every hospital has different recommendations as to how they specifically are going to conduct their TB control measures. However, the CDC and American Thoracic Society recommend that patients are no longer infectious when all of the following three criteria are met (CDC Morbidity and Mortality Weekly Report 2005, American Thoracic Society 2005):

1. Three consecutive negative AFB sputum smears collected 8-24 hours apart.
2. 2 weeks of treatment compliance
3. Symptoms are improving. It’s ok to still have a cough but they must be coughing less and not have a fever.

To make a long post shorter, contact the hospital the patient was discharged from, confirm the story, time frame, and the three negative AFB. After that, you could probably get away without placing this patient on isolation.

Special shout out to Lee for her comment and concerns about the use of respiratory fluoroquinolone. Be careful! They are second line treatment for resistant TB and should be used with caution. They can create false negative sputum samples, breed antibiotic resistance, and are associated with poorer outcomes. In a patient with recent travel from a TB endemic area,consider whether or not your macrolide resistance to S. Pneumoniae is higher than the risk of your patient having TB. Additionally, be aware that alternative regimens exist for these situations. Beta-lactams such as amoxicillin or ampicillin/sulbactam + azithromycin or beta-lactam + doxycycline might have been a better initial choice for that patient you suspect of high resistance who could potentially have tuberculosis.