Well, it’s been a while since I did a case study because I never think to remember things like these, but let’s do another. 

A 36 year old male with a history of alcohol abuse is found lying in a pool of vomit at home by his mother. He was apparently drinking heavily for the last 24 hours and was now unresponsive to painful stimuli. He had a grand mal seizure (45s long) followed by two shorter ones. His pupils were fixed and unreactive, so he was given IV Narcan (opiate antidote, did nothing), dextrose (volume expander), vitamin B12. He also showed signs of brain swelling so he was given IV mannitol.

If you were the physician, what would you order STAT for this patient? Remember that the only tests that should ever be requested stat and not routine are ones that will affect the course of treatment for the patient right then and there, so it is important to be judicious. That means no “drug screen” for you. 

Well, in this region, we have a progressive testing system in place for suspected bad alcohol ingestion in addition to the usual stat test battery. Here’s the results (reference ranges in brackets):

Hematology (only abnormal results listed):

RBC: 3.87x10^12/L (4.3-6.0x10^12/L)
Hemoglobin: 117g/L (135-175g/L)
Hematocrit: 0.35 (0.41-0.52)
White Cells: 19.2x10^9/L (4.0-11.0x10^9/L)

This doesn’t really tell us much other than that he is anemic, which is not rare at all in alcoholics (especially megaloblastic anemia), and he probably has an infection happening.

Creatinine: 172umol/L (45-125umol/L)

Creatinine isn’t the best marker for renal function, but it’s the best one in clinical use. It’s a little bit elevated so his kidney isn’t working fabulously, but it’s not high enough to have much relation to the case at the moment.

Glucose: 11.6mmol/L (3.3-11.0mmol/L)

Random glucose levels are usually pretty all over the place, and we actually need it to calculate osmolal gap for the alcohol progressive testing. But now we also know that he is not suffering from diabetic ketoacidosis because DKA glucose levels are way, way higher. I’ve seen them in the order of 50mmol/L.

Acetaminophen was also ordered because the symptoms of overdose are usually not detectable until it is too late, but came back negative.

Electrolytes:
Sodium: 137mmol/L (133-146mmol/L)
Potassium: 6.4mmol/L (3.5-5.0mmol/L)
Chloride: 110mmol/L (96-107mmol/L)
Total CO2: <5.0mmol/L (21-31mmol/L)

The critical potassium aside, the one to look at is the total CO2 here. It is practically undetectable by the instrument. Let’s look at the blood gases:

pCO2: 16mmHg (38-42mmHg)
paO2: 80mmHg (80-100mmHg)
pH: 6.9 (7.25-7.35)

Our patient here is severely, severely acidotic, more specifically, so metabolically acidotic that his body can’t even begin to compensate for it. Which is unsurprising because the alcohols all metabolize into acids.

Urea: 6.5mmol/L (2.5-8.0mmol/L)
Ethanol: <3mmol/L
Osmolality: 430mmol/kg (280-300mmol/kg)

Alright! The first step of progressive alcohol testing can now be done, and that is the calculated osmolal gap. Using the equation for calculated osmolality (2Na+Glucose+urea) we get a value of 292mmol/Kg. The gap between the measured and calculated osmolality is 138mmol/Kg of something in the blood that probably should not be there (an alcohol or some other low molecular weight compound). Taking ethanol into the equation (which is practically zero since it is also undetectable) doesn’t change much. An unaccounted osmolal gap >2mmol/Kg triggers a methanol level, and an unaccounted gap >5mmol/Kg triggers and ethylene glycol level. So let’s do that:

Ethylene Glycol: undetectable
Methanol: 55mmol/L.
And while we’re at it, since they are part of the same analytical run on the gas chromatograph, isopropanol and acetone (its major metabolite) levels were also undetectable.

Well, then. 138-55= 83mmol/kg worth of gap still unaccounted for. But remember how his brain was swelling? And how they gave him mannitol? Mannitol is extremely osmotically active so chances are the bulk of the 83mmol/kg are mannitol. It is always very important to be aware of what the physician has done to your sample before you get it on the instrument. Regardless, the patient has enough methanol in his system to hook him up to hemodialysis. We should follow him through that too.

Estimating the time it takes to bring methanol levels down to the 6mmol/L target is done from a prediction table. The table is actually very good and is based on a lot of elimination kinetic studies. In theory, it should have taken him 7.1hours to be done, but when he was levelled again at 7 hours he was nowhere near 6mmol/L. In the end it actually took about 13 hours. So what happened there and how did the table end up being so far off the mark? We should look at how the dialysis went:

His chart mentioned severe hypotension with a blood pressure of 40/20. He even suffered from cardiopulmonary arrest 15 minutes before dialysis began. So they loaded him way up with epinephrine, norepinephrine bitatrate (Levophed), calcium chloride, and heparin. Nor/epinephrine are really good vasoconstrictors. The infusion rate of levophed was 5mg/min, where normal dosing is 0.5-1.0mg/min. And that’s the problem. Alcohols distribute with total body water. They go where water goes (ie, everywhere). The catheter in hemodialysis just goes into the femoral vein. Sure, there’s a pump that pulls blood into the dialyzer, but the patient had 1)virtually no blood pressure and circulation and 2) very constricted blood vessels so the methanol was still sitting in the interstitial fluid. The double whammy really hurt dialysis progress.

So what happened to the patient?

Nuclear medicine brain scan showed no evidence of arterial inflow, and brain death was ascertained. With methanol ingestion, the presence of an osmolal gap and acidosis is dependent on the time between ingestion and presentation which has a huge impact on prognosis. A person with an osmolal gap and no acidosis probably hasn’t metabolized his methanol yet and they are generally saved. Someone with no osmolal gap/acidosis is a late presenter and is very rarely saved. Our patient didn’t really fall into these categories as he is somewhere in between the two, but with a blood pH of 6.9 like our gentleman here, the chances of saving him were pretty small despite all heroic efforts.