The Case




45yo M with obesity, HTN, DM, GERD, EtOH abuse, presents feeling unwell and fatigued. Has mild HA, lightheadedness and for the past few days has been having polyuria with home FS in 300s. Was seen in ED yesterday with elevated FS, but given fluids with improvement in symptoms. Afterward, he saw his endocrinologist who adjusted his medications. Denies fever, abd pain, n/v/d, dysuria.

Physical Exam

BP 156/90, HR 101, RR 17, O2 99%, T 97.5F, FS 121
Gen: NAD, AAOx3
CV: RRR, 2+ peripheral pulses 
Pulmonary/Chest: CTAB, no resp distress
Abdominal: Soft, NT/ND
Musculoskeletal: FROM, no edema
Neurological: No deficits
Skin: Warm and dry


CBC: 10 > 15 / 43 < 297, 65% Neut BMP: 138 / 4.8 / 105 / 12 / 15 / 0.6 < 135 VBG: 7.265 / 30 / 32 / 13.7, lactate 3.7 EtOH: <10


  1. What is your differential for this patient?

    Note the lab abnormalities - Anion gap of 21 with a bicarbonate of 12 and a pH of 7.265. This patient has an anion gap metabolic acidosis. MUDPILES is the classic mnemonic used for the differential of an AG metabolic acidosis. CAT MUDPILES and GOLDMARK are other mnemonics you can use. Regardless of your preference, the differential is broad and you should begin narrowing it down based on history, physical, and lab values.

  2. What other lab test can confirm your diagnosis?

    KULT (Ketones, Uremia, Lactate, Toxins). Yes, another mnemonic. KULT is a good way to begin your workup. Whereas MUDPILES gives you the diagnosis, KULT gives you the cause of the elevated anion gap. Every substance or condition in MUDPILES will lead to an elevation
    is one component of KULT. Once you know which component, you can go back to MUDPILES and narrow your diagnosis.
    From the provided lab values we know that the patient is not uremic. There is an elevated lactate, but not high enough to cause an AG of 21. We do not know about ketones or toxins.
    urinalysis of this patient would reveal elevated ketones. This is the cause of the patient’s AG metabolic acidosis. If you were worried about other causes such as salicylate poisoning you could get a salicylate level (which was negative on this patient). If you were worried about a toxic alcohol ingestion you could get a serum osmolarity and check for an elevated osmolar gap (which was normal in this patient). The patient also denied ingestion of other substances that might cause an AG. It is important to remember that history and physical are still the most important aspects of the workup and lab values in isolation will not give you the diagnosis.

  3. How would you manage this patient?

    Now that we know this patient has a ketoacidosis, the next question is, “What type of ketoacidosis?” Diabetic ketoacidosis, alcoholic ketoacidosis, starvation ketoacidosis. The patient denies fasting and has been tolerating PO - unlikely starvation ketosis. He has a history of alcohol abuse but denies drinking in the past 2 years. He also does not have the classic presentation of abdominal pain, nausea, vomiting - unlikely alcoholic ketosis. That leaves DKA. But the patient’s FS is normal, how could it be DKA?.

    Once again, history is key. The patient was started on a new medication, canagliflozin, an SGLT2 inhibitor. This medication causes increased urinary excretion of glucose. One side effect, however, is euglycemic DKA. In these cases, patients will have an anion gap metabolic acidosis, elevated ketones, and a FS less than 200. Other medications in this class include dapagliflozin and empagliflozin.

    Treatment is the same for euglycemic DKA as it is for hyperglycemic DKA - fluids to correct volume losses, insulin to correct anion gap, and electrolyte supplementation. A typical DKA protocol will switch to dextrose containing fluids when the FS drops below 200. This patient is starting at a FS below 200 so the only difference is that fluid resuscitation will begin with D5 ½ NS.