IM BOARD REVIEW A SELF-TEST CME EDUCATIONAL OBJECTIVE: Readers will apply a systematic approach to diagnosing acid-base disturbances ON A CREDIT CLINICAL SHYLAJA MANI, MD GREGORY W. RUTECKI, MD Department of Internal Medicine, Department of Internal Medicine, CASE Cleveland Clinic Cleveland Clinic

A patient with altered mental status and an acid-base disturbance

78-year-old black woman with a his- TABLE 1 A tory of osteoarthrosis and chronic diffuse joint pain presents with altered mental status The patient’s laboratory values and tachypnea, which began 3 hours earlier. She lives alone, and her family suspects she Substance Value abuses both alcohol and her pain medica- Blood chemistry tions. She has not been eating well and has lost approximately 10 pounds over the past 3 Sodium 132 mmol/L 136–144 months. Her analgesic regimen includes acet- Potassium 4.8 mmol/L 3.7–5.1 aminophen and acetaminophen-oxycodone. In the emergency department her tem- 16.0 mmol/L 22–30a perature is 98.6°F (37.0°C), pulse 100 beats Chloride 95 mmol/L 97–105 per minute and regular, respiratory rate 22 per minute, and blood pressure 136/98 mm Hg. 23 mg/dL 7–21 She is obtunded but has no focal neurologic 1.3 mg/dL 0.58–0.96 defects or meningismus. She has no signs of heart failure (jugular venous distention, car- 97 mg/dL 74–99 diomegaly, or gallops), and examination of the Lactate 1.1 mmol/L 0.5–2.2 lungs and abdomen is unremarkable. Suspecting that the patient may have tak- Albumin 4.5 g/dL 3.9–4.9 en too much oxycodone, the physician gives osmolality 318 mOsm/kg 275–295 her naloxone, but her mental status does not improve. Results of chest radiography and cra- Arterial blood gases nial computed tomography are unremarkable. pH 7.25 7.35–7.45 The physician’s initial impression is that the b patient has “metabolic encephalopathy of un- Pco2 28 mm Hg 34–46 known etiology.” Bicarbonate 16 mmol/L 22–26 The patient’s laboratory values are shown in Table 1. aFor acid-base problem-solving, a value of 25 mmol/L is considered normal. b Healthier patients with better pulmonary function may attain lower Pco2 values; ■■ WHICH ACID-BASE DISORDER for clinical problem-solving, a value of 40 mm Hg is considered normal. Pco = partial pressure of carbon dioxide DOES SHE HAVE? 2

Which acid-base disorder does this patient □□ with an elevated 1have? □□ Metabolic acidosis and respiratory alkalosis □□ A triple disturbance: metabolic acidosis, □□ Metabolic acidosis and respiratory acidosis respiratory acidosis, and metabolic alkalosis doi:10.3949/ccjm.84a.16042

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 27 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. ACID-BASE DISTURBANCE

TABLE 2 A 5-step approach Acid-base disorders can be diagnosed and ‘Rules of 5’ for acid-base problem-solving characterized using a systematic approach known as the “Rules of 5” (Table 2)1: 1 Determine the arterial pH status 1. Determine the arterial pH status. pH < 7.40 is acidemic, pH > 7.44 is alkalemic 2. Determine whether the primary process is But a normal pH does not rule out an acid-base disorder respiratory, metabolic, or both. 3. Calculate the anion gap. 2 If the arterial pH is abnormal, determine whether 4. Check the degree of compensation the primary process is respiratory, metabolic, or both (respiratory or metabolic).

pH Pco2 Bicarbonate 5. If the patient has metabolic acidosis with Respiratory acidosis Low High — an elevated anion gap, check whether the bicarbonate level has decreased as much Metabolic acidosis Low — Low as the anion gap has increased (ie, whether Mixed respiratory and Low High Low there is a delta gap). metabolic acidosis Let us apply this approach to the patient Respiratory alkalosis High Low — described above. Metabolic alkalosis High — High 1. What is her pH status? Mixed respiratory and High Low High An arterial pH less than 7.40 is acidemic, metabolic alkalosis whereas a pH higher than 7.44 is alkalemic. (Acidemia and alkalemia refer to the abnormal 3 Calculate the anion gap laboratory value, while acidosis and alkalosis Anion gap = sodium – (chloride + bicarbonate) refer to the process causing the abnormal val- ue—a subtle distinction, but worth keeping in If serum albumin is low, add 2.5 mmol/L to the anion gap mind.) for every 1 g the serum albumin is below normal Caveat. A patient may have a significant An anion gap > 10 mmol/L is elevated acid-base disorder even if the pH is normal. 4 Check the degree of compensation (respiratory or metabolic) Therefore, even if the pH is normal, one should verify that the partial pressure of car- Pco2 and bicarbonate should move in the same direction bon dioxide (Pco2), bicarbonate level, and an- Nominal normal levels: bicarbonate 25 mmol/L and Pco2 40 mm Hg ion gap are normal. If they are not, the patient may have a mixed acid-base disorder such as In respiratory acidosis, for every 10-mm Hg increase in Pco2, bicarbonate should increase by 1 mmol/L in the first 48 hours respiratory acidosis superimposed on metabol- and 4 mmol/L afterward ic alkalosis. In metabolic acidosis, for every 1-mmol/L decrease in bicarbonate, Our patient’s pH is 7.25, which is in the acidemic range. Pco2 should decrease by 1.3 mm Hg

In respiratory alkalosis, for every 10-mm Hg decrease in Pco2, 2. Is her acidosis respiratory, metabolic, bicarbonate should decrease by 2 mmol/L in the first 48 hours or both? and 5 mmol/L afterward Respiratory acidosis and alkalosis affect the

In metabolic alkalosis, for every 1-mmol/L increase in bicarbonate, Pco2. The Pco2 is high in respiratory acido- Pco2 may increase by 0.6 mm Hg sis (due to failure to get rid of excess carbon dioxide), whereas it is low in respiratory alka- 5 If the patient has metabolic acidosis with an elevated an- losis (due to loss of too much carbon dioxide ion gap, check whether the bicarbonate level has decreased through hyperventilation). as much as the anion gap has increased Metabolic acidosis and alkalosis, on the In metabolic acidosis, the anion gap should increase by the same other hand, affect the serum bicarbonate lev- amount that bicarbonate decreases; a difference in these two changes is el. In metabolic acidosis the bicarbonate level called a delta gap is low, whereas in metabolic alkalosis the bi- carbonate level is high. Pco = partial pressure of carbon dioxide 2 Moreover, in mixed respiratory and meta- Based on information in reference 1 bolic acidosis, the bicarbonate level can be

28 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. MANI AND RUTECKI

low and the Pco2 can be high. In mixed meta- Our patient’s bicarbonate level is 16 bolic and respiratory alkalosis, the bicarbon- mmol/L, which is 9 mmol/L lower than nor- ate level can be high and the Pco2 can be low mal (for acid-base calculations, we use 25 (Table 2). mmol/L as the nominal normal level). If she is

Our patient’s serum bicarbonate level is compensating appropriately, her Pco2 should low at 16.0 mmol/L, indicating that the pro- decline from 40 mm Hg (the nominal normal cess is metabolic. Her Pco2 is also low (28 mm level) by about 11.7 mm Hg (9 × 1.3), to ap- Hg), which reflects an appropriate response to proximately 28.3 mm Hg. Her Pco2 is, indeed, compensate for the acidosis. 28 mm Hg, indicating that she is compensat- 3. What is her anion gap? ing adequately for her metabolic acidosis. Always calculate the anion gap, ie, the serum If we use Winter’s formula instead (Pco2 = 3 sodium concentration minus the serum chlo- [1.5 × the bicarbonate level] + 8 ± 2), the ride and serum bicarbonate concentrations. If lowest calculated Pco2 would be 30 mm Hg, the patient’s serum albumin level is low, for which is within 2 mm Hg of the Rules of 5 every 1 gram it is below normal, an additional calculation. Other formulas for calculating 2.5 mmol/L should be added to the calculated compensation are available.3 anion gap. We consider an anion gap of 10 This information rules out the first two an- mmol/L or less as normal. swers to question 1, ie, metabolic acidosis with Caveats. The blood sample used to calcu- respiratory alkalosis or acidosis. late the anion gap should be drawn close in time to the arterial blood gas sample. 5. Is there a delta gap? Although the anion gap is an effective Although we know the patient has metabolic tool in assessing acid-base disorders, further acidosis with an elevated anion gap, we have investigation is warranted if clinical judg- not ruled out the possibility that she may have ment suggests that an anion gap calculation a triple disturbance. For this reason we need to is inconsistent with the patient’s circum- check her delta gap. 2 In metabolic acidosis with an elevated an- stances. The patient Our patient’s anion gap is elevated (21 ion gap, as the bicarbonate level decreases, mmol/L). Her serum albumin level is in the the anion gap should increase by the same lives alone, normal range, so her anion gap does not need amount. If the bicarbonate level decreases and her family to be adjusted. more than the anion gap increases, the ad- ditional decline is the result of a second pro- suspects she 4. Is the degree of compensation appropriate cess—an additional normal-anion-gap acido- abuses alcohol for the primary acid-base disturbance? sis. If the bicarbonate level does not decrease The kidneys compensate for the lungs, and and pain as much as the anion gap increases, there is an vice versa. That is, in respiratory acidosis or additional metabolic alkalosis. medications alkalosis, the kidneys adjust the bicarbonate levels, and in metabolic acidosis, the lungs ad- Our patient’s bicarbonate level decreased 9 mmol/L (from the nominal normal level of just the Pco2 (although in metabolic alkalosis, it is hard for patients to breathe less, especially 25 to 16), and therefore her anion gap should if they are already hypoxic). have increased approximately the same In metabolic acidosis, people compensate amount—and it did. (A normal anion gap for by breathing harder to get rid of more carbon problem-solving is 10, and this patient’s anion dioxide. For every 1-mmol/L decrease in the gap has increased to 21. A difference of ± 2 is insignificant.) This conclusion verifies that a bicarbonate level, the Pco2 should decrease by 1.3 mm Hg. triple acid-base disturbance is not present, so Compensation does not return pH to nor- the last answer is incorrect. mal; rather, it mitigates the impact of an acid So, the correct answer to the question or alkali excess or deficit. If the pH is normal- posed above is metabolic acidosis with an ized with an underlying acid-base disturbance, elevated anion gap (that is, metabolic aci- there may be mixed acid-base processes rather dosis with appropriate respiratory compen- than compensation. sation).

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 29 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. ACID-BASE DISTURBANCE

TABLE 3 , and their many metabolic products. In our patient, ingestion of both metha- ‘MUD PILES’ then and now: nol and should be considered, Metabolic acidosis with elevated anion gap since she lives alone and has been suspected of Then Now alcohol and opioid abuse. Her calculated os- mol gap is 278 mOsm/kg. Her measured osmo- Methanol Methanol lality is 318 mOsm/kg (Table 1). The osmol Uremia Uremia gap is 40 mOsm/kg (normal is ≤ 10).4,5 There- Diabetic ketoacidosisa Diabetic ketoacidosis fore, her is elevated. c Identifying the specific substance the pa- Paraldehyde Pyroglutamate and tient ingested that caused metabolic acidosis Isoniazid Isoniazid and ingestions with anion gap may be difficult. Poisonings Lactic acidemia (l and d)b Lactic acidemiad with these agents do not always increase the osmol gap.6 A high index of suspicion is essen- Ethylene glycol (di)Ethylene glycol tial. It is helpful to have the family search for Salicylates Salicylates any sources of ethylene glycol and methanol aKetoacidosis can be a complication of alcoholism (betahydroxybutyric acid). at home and initiate treatment early if an in- bd-Lactate is a consequence of short-bowel syndrome. If more common causes of gestion is suspected, using fomepizole (an al- metabolic acidosis are not present and the patient has a history suggesting short-bowel cohol dehydrogenase inhibitor) or parenteral syndrome, d-lactate can be considered. 7 cIf the patient is older, female, and has a history of acetaminophen ingestion—and if and hemodialysis. Liquid chromatog- more common causes of metabolic acidosis are not present—consider pyroglutamic raphy identifies these two toxins, but results acid metabolic acidosis as an etiology. are not available emergently. dd-Lactate can also be a consequence of propylene glycol metabolic acidosis. Diethylene glycol ingestion should also be considered.8 Since it is diagnosed and treated ■■ ‘MUD PILES’: FINDING THE CAUSE like ethylene glycol intoxication, it can be OF ANION GAP METABOLIC ACIDOSIS placed with the “E” of (di)ethylene glycol in the mnemonic. The possible causes of metabolic acidosis with A patient an elevated anion gap (as in our patient) can Uremia may have be summarized in the mnemonic MUD PILES Renal failure can lead to metabolic acidosis.9 a significant (methanol, uremia, diabetes, paraldehyde, Our patient has no history of kidney disease, isoniazid, lactate, ethylene glycol, and salicy- but her blood urea nitrogen and creatinine acid-base lates), which has been used for many years. concentrations are above normal, and her es- disorder Parts of it are no longer useful, but rather than timated glomerular rate by the Mod- discard it, we propose to update it (Table 3). ification of Diet in Renal Disease formula is 48 even if the pH mL/min/1.73 m2—low, but not uremic. is normal Methanol and ethylene glycol Rhabdomyolysis (suspected by elevated We will address toxic ingestion of methanol values) should be considered and ethylene glycol (the “M” and “E” of MUD in any patient with mental status changes, sus- PILES) at the same time. pected toxic ingestion, and metabolic acidosis In cases of suspected ingestion of toxic sub- (see the “I” in MUD PILES below). Compart- stances such as these, it is useful to examine ment syndromes with muscle necrosis may the osmol gap, ie, the difference between the present in a subtle fashion. Therefore, renal calculated and the measured serum osmolality. failure from rhabdomyolysis may complicate Serum osmolality (in mOsm/kg) is calculated this patient’s course later, and should be kept as the sodium concentration in mmol/L times in mind. 2, plus the glucose concentration in mg/dL di- vided by 18, plus the blood urea nitrogen con- Diabetes centration in mg/dL divided by 2.8 (Table 4). The patient has no history of diabetes and has If the measured osmolality is higher than this a normal blood glucose level. Blood testing did calculated value, the difference may be due to not reveal ketones. She is not taking metfor- solutes in the blood that should not be there min (alleged to cause lactic acidosis) or a so- such as ethylene glycol, diethylene glycol, dium-glucose cotransporter 2 inhibitor (which

30 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. MANI AND RUTECKI

10 have been associated with ketoacidosis). TABLE 4 There is another, less common cause of ketoacidosis: alcohol.11 Although alcoholism Increasing diagnostic utility of the osmol gap is common, alcoholic ketoacidosis is uncom- mon, even in heavy drinkers. Ethyl alcohol The osmol gap is the difference between the calculated serum osmolality and measured osmolality causing metabolic acidosis is similar to meta- bolic acidosis with (di)ethylene glycol and Calculated serum osmolality = methanol, and if suspected it should be treated (Sodium x 2) + (Glucose/18) + (Blood urea nitrogen/2.8) empirically (first with thiamine, then dextrose An elevated osmol gap indicates osmotically active solutes in plasma and saline, and correcting other that are not typically present under normal conditions, such as disturbances such as hypokalemia and hypo- ethylene glycol, diethylene glycol, methanol, and their many metabolic magnesemia) before specific identification is products made. Ketones (predominantly beta-hydroxy- butyrate) may persist up to 2 weeks after al- ■■ UPDATING THE ‘P’ IN MUD PILES cohol ingestion has stopped.11 Ketosis in the setting of alcoholic ketoacidosis is frequently Paraldehyde is rarely prescribed anymore. A accompanied by other markers of alcohol tar- PubMed search on December 21, 2015 ap- get organ injury: elevated , aspartate plying the terms paraldehyde and metabolic aminotransferase, alanine aminotransferase, acidosis yielded 17 results. Those specific to and gamma-glutamyl transferase levels. The anion gap metabolic acidosis were from 1957 12–20 term “ketohepatitis” has been suggested as an to 1986 (n = 9). alternative to alcoholic ketosis.11 Therefore, we can eliminate paraldehyde This patient did not have an elevated from the MUD PILES mnemonic and replace blood ethanol level, and her liver markers it with pyroglutamic acid and propylene glycol. were otherwise normal. 5-Oxoproline or pyroglutamic acid, a metabolite of acetaminophen ■■ THE NEW MUD PILES Acetaminophen depletes glutathione stores Anion gap: in acute overdoses, in patients with inborn Which of the following is (are) true? Re- sodium minus 2garding the remaining letters of the MUD errors of , and after chronic inges- PILES mnemonic: tion of excessive, frequent doses. Depletion chloride minus of glutathione increases metabolic products, bicarbonate □□ The “P” (paraldehyde) has been replaced including pyroglutamic acid, which dissoci- by pyroglutamic acid (5-oxoproline) and ates into hydrogen ions (leading to metabolic (normal ≤ 10) propylene glycol. acidosis and an anion gap), and 5-oxoproline, □□ There are two isomers of lactate (dextro (which can be detected in the ).21,22 and levo), and consequently two clinical Risk factors for metabolic acidosis with ac- varieties of lactic acidosis. etaminophen ingestion include malnutrition, □□ Isoniazid is no longer associated with chronic alcoholism, liver disease, and female metabolic acidosis with elevated anion gap. sex. In fact, most cases have been reported in □□ Salicylates can paradoxically be associated females, and altered mental status has been both with elevated and low anion gaps. common. Isoniazid is still associated with metabolic aci- Metabolic acidosis with pyroglutamic acid dosis with elevated anion gap, and so the third can occur without elevated acetaminophen answer choice is false; the rest are true. levels. Serum and urine levels of pyroglutamic acid may assist with diagnosis. Since identifi- Paraldehyde, isoniazid, lactate cation of urine pyroglutamic acid usually re- The “P,” “I,” and “L” (d-lactate) of the re- quires outside laboratory assistance, a clinical vamped MUD PILES acronym are less com- diagnosis is often made initially and corrob- mon than the others. They should be consid- orated later by laboratory results. When the ered when the more typical causes of metabolic anion gap metabolic acidosis is multifactorial, acidosis are not present, as in this patient. as it was suspected to be in a case reported by

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 31 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. ACID-BASE DISTURBANCE

Tan et al,23 the osmol gap may be elevated as a Lactic acidosis consequence of additional toxic ingestions, as Lactic acidosis and metabolic acidosis can it was in the reported patient. result from hypoperfusion (type A) or other No controlled studies of treatment have causes (type B). Not all lactic acidosis is been done. n-Acetylcysteine may be of ben- contingent on l-lactate, which humans can efit. Occasional patients have been dialyzed metabolize. Metabolic acidosis may be a con- for removal of excess pyroglutamic acid. sequence of d-lactate (mammals have no d- ). d-Lactic acidosis as Propylene glycol, a result of short bowel syndrome has been a component of parenteral lorazepam 29 Lorazepam is a hydrophobic drug, so when it is known for more than a generation. How- given parenterally, it must be mixed with a suit- ever, d-lactic acidosis occurs in another new able solvent. A typical formulation adds propyl- setting. The new “P” in MUD PILES, propyl- ene glycol, can generate substantial amounts ene glycol. In patients receiving high doses of 29 lorazepam as relaxation therapy for acute respi- of d-lactate. ratory distress syndrome in the intensive care d-lactic metabolic acidosis is always accom- unit, or as treatment of alcohol withdrawal, the panied by neurologic manifestations (slurred propylene glycol component can precipitate an- speech, confusion, somnolence, ataxia, abu- 30 ion gap metabolic acidosis.24,25 sive behavior, and others). With short bowel Although nearly one-half of the adminis- syndrome, the neurologic manifestations occur 30 tered propylene glycol is excreted by the kid- after eating and clear later. neys, the remaining substrate is metabolized by Although our patient’s anion gap is more alcohol dehydrogenase into d,l-lactaldehyde, than 20 mmol/L, her blood level of lactate is then converted into d- or l-lactate. l-Lactate not elevated, and she had no history to suggest can be metabolized, but d-lactate cannot and short-bowel syndrome. leads to anion gap metabolic acidosis. This is Salicylates another toxic metabolic acidosis associated Salicylate overdose can cause a mixed acid- with an elevated osmol gap. An increasing os- An elevated base disorder: metabolic acidosis with elevat- mol gap in the intensive care unit can serve ed anion gap and respiratory alkalosis. osmol gap as a surrogate marker of excessive propylene Although our patient does not have respi- glycol administration.23 may be due ratory alkalosis, an aspirin overdose must be to solutes Isoniazid considered. A salicylate level was ordered; it Although it is uncommon, there are reports was negative. in the blood of isoniazid-induced anion gap metabolic aci- Despite the typical association of salicy- that should not dosis,26 either due to overdoses, or less com- lates with an elevated anion gap, they may be there monly, with normal dosing. Isoniazid should also cause a negative anion gap.31 Chloride- therefore remain in the mnemonic MUD sensing ion-specific electrodes contain a PILES and may be suspected when metabolic membrane permeable to chloride. Salicylates acidosis is accompanied by seizures unrespon- can increase the chloride permeability of sive to usual therapy. The seizures respond to these membranes, generating pseudohyper- pyridoxine. chloremia, and consequently, a negative an- The “I” should also be augmented by ion gap. newer causes of metabolic acidosis associated with “ingestions.” Ecstasy, or 3,4-methylene- ■■ WHAT ELSE MUST BE CONSIDERED? dioxymethamphetamine, can cause metabolic acidosis and seizures. Ecstasy has been asso- In view of her anion gap metabolic acido- ciated with rhabdomyolysis and uremia, also 3sis, elevated osmol gap, and absence of dia- leading to anion gap metabolic acidosis.27 A betes, renal failure, or lactate excess, what newer class of abused substances, synthetic are the remaining diagnoses to consider in cathinones (“bath salts”), are associated with this patient? (Choose all that are potential metabolic acidosis, compartment syndrome, sources of metabolic acidosis and an in- and renal failure.28 creased anion gap.)

32 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. MANI AND RUTECKI

□□ Methanol, ethylene, or diethylene glycol ■■ ACID-BASE DISORDERS □□ Excessive, chronic acetaminophen IN DIFFERENT DISEASES ingestion Diverse diseases cause distinctive acid-base □□ Salicylate toxicity abnormalities. Matching the appropriate acid- □□ Alcoholic ketoacidosis base abnormality with its associated disease All of the above can potentially contribute to metabolic acidosis. may lead to more timely diagnosis and treat- A search of the patient’s home did not re- ment: veal a source of methanol or either ethylene Type 2 diabetes mellitus, for example, can or diethylene glycol. Similarly, no aspirin was lead to lactic acidosis, ketoacidosis, or type 4 33 found, and the patient’s salicylate levels were renal tubular acidosis. not elevated. The patient’s laboratory work Heart failure, although not typically did not reveal increased ketones. framed in the context of acid-base physiology, Since none of the common causes of meta- can lead to elevated lactate, which is associ- bolic acidosis were discovered, and since the ated with a worse prognosis.34 patient had been taking acetaminophen, the Acquired immunodeficiency syndrome. diagnosis of excessive chronic acetaminophen Abacavir can cause normal anion gap meta- ingestion was suspected pending laboratory bolic acidosis.35,36 verification. Identification of 5-oxoproline in Cancer37,38 can be associated with proxi- the urine may take a week or more since the mal tubular renal tubular acidosis and lactic sample is usually sent to special laboratories. acidosis. Acetaminophen levels in this patient were significantly elevated, as were urinary oxypro- An expanding array line levels, which returned later. of toxic ingestions The patient was diagnosed with pyroglu- Metabolic acidosis may be the most promi- tamic acid metabolic acidosis. She was treated nent and potentially lethal clinical acid-base supportively and with n-acetylcysteine intrave- disturbance. When metabolic acidosis occurs nously, although there have been no controlled in certain disease states—lactic acidosis with studies of the efficacy of this drug. Seventy-two hypoperfusion or methanol ingestion with hours after admission, she had improved. Her metabolic acidosis, for example—there is in- acid-base status returned to normal. creased morbidity and mortality. As reflected in the revisions to MUD ■■ GOLD MARK: PILES and in the newer GOLD MARK acro- ANOTHER WAY TO REMEMBER nym, the osmol gap has become more valuable Another mnemonic device for remembering in differential diagnosis of metabolic acidosis the causes of metabolic acidosis with elevated with an elevated anion gap consequent to an anion gap is “GOLD MARK”: glycols (eth- expanding array of toxic ingestions (metha- ylene and propylene), oxoproline (instead of nol, propylene glycol, pyroglutamic acid- pyroglutamic acid from acetaminophen), l- oxoproline, ethylene glycol, and diethylene lactate, d-lactate, methanol, aspirin, renal glycol), which may accompany pyroglutamic failure, and ketoacidosis).32 acid-oxoproline. ■ ■■ REFERENCES centre experience. Clin Kidney J 2012; 5:120–123. 1. Whittier WL, Rutecki GW. Primer on clinical acid-base problem solv- 6. Kraut JA. Diagnosis of toxic alcohols: limitations of present meth- ing. Dis Mon 2004; 50:122–162. ods. Clin Toxicol (Phila) 2015; 53:589–595. 2. Kraut JA, Madias NE. Serum anion gap: its uses and limitations in 7. Ghannoum M, Hoffman RS, Mowry JB, Lavergne V. Trends in toxic clinical medicine. Clin J Am Soc Nephrol 2007; 2:162–174. alcohol exposures in the United States from 2000 to 2013: a focus 3. Adrogué HJ, Madias NE. Secondary responses to altered acid-base on the use of antidotes and extracorporeal treatments. Semin Dial status: the rules of engagement. J Am Soc Nephrol 2010; 21:920– 2014; 27:395–401. 923. 8. Schep LJ, Slaughter RJ, Temple WA, Beasley DM. Diethylene glycol 4. Krasowski MD, Wilcoxon RM, Miron J. A retrospective analysis of poisoning. Clin Toxicol (Phila) 2009; 47:525–535. glycol and toxic alcohol ingestion: utility of anion and osmolal gaps. 9. Kraut JA, Madias NE. Metabolic acidosis of CKD: an update. Am J BMC Clin Pathol 2012;12:1. Kidney Dis 2016; 67:307–317. 5. Latus J, Kimmel M, Alscher MD, Braun N. Ethylene glycol poisoning: 10. Taylor SI, Blau JE, Rother KI. SGLT2 inhibitors may predispose to a rare but life-threatening cause of metabolic acidosis—a single- ketoacidosis. J Clin Endocrinol Metab 2015; 100:2849–2852.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 33 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. ACID-BASE DISTURBANCE

11. Yokoyama A, Yokoyama T, Mizukami T, et al. Alcoholic ketosis: soc Physicians India 2009; 57:70–71. prevalence, determinants, and ketohepatitis in Japanese alcoholic 27. Ben-Abraham R, Szold O, Rudick V, Weinbroum AA. ‘Ecstasy’ intoxi- men. Alcohol Alcohol 2014; 49:618–625. cation: life-threatening manifestations and resuscitative measures in 12. Hayward JN, Boshell BR. Paraldehyde intoxication with metabolic the intensive care setting. Eur J Emerg Med 2003; 10:309–313. acidosis; report of two cases, experimental data and a critical review 28. German CL, Fleckenstein AE, Hanson GR. Bath salts and synthetic of the literature. Am J Med 1957; 23:965–976. cathinones: an emerging designer drug phenomenon. Life Sci 2014; 13. Elkinton JR, Huth EJ, Clark JK, Barker ES, Seligson D. Renal tubular 97:2–8. acidosis with organic aciduria during paraldehyde ingestion; six year 29. Jorens PG, Demey HE, Schepens PJ, et al. Unusual d-lactic acidosis study of an unusual case. Am J Med 1957; 23:977–986. from propylene glycol metabolism in overdose. J Toxicol Clin Toxicol 14. Waterhouse C, Stern EA. Metabolic acidosis occurring during admin- 2004; 42:163–169. istration of paraldehyde. Am J Med 1957; 23:987–989. 30. Kang KP, Le S, Kang SK. d-Lactic acidosis in humans: review and 15. Beier LS, Pitts WH, Gonick HC. Metabolic acidosis occurring during update. Electrolyte Blood Press 2006; 4:53–56. paraldehyde intoxication. Ann Intern Med 1963; 58:155–158. 31. Emmett M. Approach to the patient with a negative anion gap. Am 16. Hiemcke T. Metabolic acidosis due to paraldehyde. Ned Tijdschr J Kidney Dis 2016; 67:143–150. Geneeskd 1964; 108:2165–2167. Dutch. 32. Mehta AN, Emmett JB, Emmett M. GOLD MARK: an anion gap mne- 17. Gailitis RJ. Paraldehyde acidosis syndrome. IMJ III Med J 1966; monic for the 21st Century. Lancet 2008; 372:892. 129:258–262. 33. Palmer BF, Clegg DJ. Electrolyte and acid-base disturbances in pa- 18. Gutman RA, Burnell JM. Paraldehyde acidosis. Am J Med 1967; tients with diabetes mellitus. N Engl J Med 2015; 373:548–559. 42:435–440. 19. Hadden JW, Metzner RJ. Pseudoketosis and hyperacetaldehydemia 34. Park JJ, Choi DJ, Yoon CH, et al; KorHF Registry. The prognostic in paraldehyde acidosis. Am J Med 1969; 47:642–647. value of arterial blood gas analysis in high-risk acute heart failure 20. Linter CM, Linter SP. Severe lactic acidosis following paraldehyde patients: an analysis of the Korean Heart Failure (KorHF) registry. administration. Br J Psychiatry 1986; 149:650–651. Eur J Heart Fail 2015; 17:601–611. 21. Zand L, Muriithi A, Nelsen E, et al. Severe anion gap metabolic aci- 35. Musso CG, Belloso WH, Glassock RJ. Water, , and acid- dosis from acetaminophen use secondary to 5-oxoproline (pyroglu- base alterations in human immunodeficiency virus infected patients. tamic acid) accumulation. Am J Med Sci 2012; 344:501–504. World J Nephrol 2016; 5:33–42. 22. Abkur TM, Mohammed W, Ali M, Casserly L. Acetaminophen- 36. Camara-Lemarroy CR, Flores-Cantu H, Calderon-Hernandez HJ, Diaz- induced anion gap metabolic acidosis secondary to 5-oxoproline: a Torres MA, Villareal-Velazquez HJ. Drug-induced haemolysis, renal case report. J Med Case Rep 2014; 8:409. failure, thrombocytopenia and lactic acidosis in patients with HIV 23. Tan EM, Kalimullah E, Sohail MR, Ramar K. Diagnostic challenge in a and cryptococcal meningitis: a diagnostic challenge. Int J STD AIDS patient with severe anion gap metabolic acidosis. Case Rep Crit Care 2015; 26:1052–1054. 2015; 2015:272914. 37. Miltiadous G, Christidis D, Kalogirou M, Elisaf M. Causes and mecha- 24. Jorens PG, Demey HE, Schepens PJ, et al. Unusual d- aci- nisms of acid-base and electrolyte abnormalities in cancer. Eur J dosis from propylene glycol metabolism in overdose. J Toxicol Clin Intern Med 2008; 19:1–7. Toxicol 2004; 42:163–169. 38. Vlachostergios PJ, Oikonomou KG, Gibilaro E, Apergis G. Elevated 25. Barnes BJ, Gerst C, Smith JR, Terrell AR, Mullins ME. Osmol gap as lactic acid is a negative prognostic factor in metastatic lung cancer. a surrogate marker for serum propylene glycol concentrations in Cancer Biomark 2015; 15:725–734. patients receiving lorazepam for sedation. Pharmacotherapy 2006; 26:23–33. ADDRESS: Gregory W. Rutecki, MD, Department of Internal Medi- 26. Gokhale YA, Vaidya MS, Mehta AD, Rathod NN. Isoniazid toxicity cine, G10, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195; presenting as status epilepticus and severe metabolic acidosis. J As- [email protected]

34 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 1 JANUARY 2017 Downloaded from www.ccjm.org on October 2, 2021. For personal use only. All other uses require permission. CORRECTION

Altered mental status and an acid-base disturbance JANUARY 2017

TABLE 2 In the article “A patient with altered mental status and an acid-base disturbance” (Mani S, ‘Rules of 5’ for acid-base problem-solving Rutecki GW, Cleve Clin J Med 2017; 84:27– 1 Determine the arterial pH status 34), 2 errors occurred in Table 2. The correct- pH < 7.40 is acidemic, pH > 7.44 is alkalemic ed table appears at left, with corrections shown But a normal pH does not rule out an acid-base disorder in red: In addition, two sentences in the text re- 2 If the arterial pH is abnormal, determine whether garding the osmol gap should be revised as fol- the primary process is respiratory, metabolic, or both lows: On page 31, the last 3 lines should read as pH PCO Bicarbonate 2 follows: “When the anion gap metabolic aci- Respiratory acidosis Low High — dosis is multifactorial, as it was suspected to Metabolic acidosis Low — Low be in a case reported by Tan et al,23 the osmol Mixed respiratory and Low High Low gap may be elevated as a consequence of ad- metabolic acidosis ditional toxic ingestions, as it was in the re- ported patient.” Respiratory alkalosis High Low — And on page 33, the last sentence should Metabolic alkalosis High — High read as follows: “As refl ected in the revisions Mixed respiratory and High Low High to MUD PILES and in the newer GOLD metabolic alkalosis MARK acronym, the osmol gap has become more valuable in differential diagnosis of 3 Calculate the anion gap metabolic acidosis with an elevated anion gap Anion gap = sodium – (chloride + bicarbonate) consequent to an expanding array of toxic in- If serum albumin is low, add 2.5 mmol/L to the anion gap gestions (methanol, propylene glycol, ethyl- for every 1 g the serum albumin is below normal ene glycol, and diethylene glycol), which may An anion gap > 10 mmol/L is elevated accompany pyroglutamic acid-oxoproline.”

4 Check the degree of compensation (respiratory or metabolic) Cardiopulmonary exercise testing PCO2 and bicarbonate should move in the same direction FEBRUARY 2017 Nominal normal levels: bicarbonate 25 mmol/L and PCO 40 mm Hg 2 In the article, “Cardiopulmonary exercise In respiratory acidosis, for every 10-mm Hg increase in PCO2,, bicarbonate should increase by 1 mmol/L in the fi rst 48 hours testing: A contemporary and versatile clini- and 4 mmol/L afterward cal tool” (Leclerc K, Cleve Clin J Med 2017; In metabolic acidosis, for every 1-mmol/L decrease in bicarbonate, 84:161–168), an error occurred in Table 1. Heart rate reserve was defi ned as maximum PCO2 should decrease by 1.3 mm Hg heart rate minus resting heart rate. It should In respiratory alkalosis, for every 10-mm Hg decrease in PCO2, bicarbonate should decrease by 2 mmol/L in the fi rst 48 hours be defi ned as (maximum heart rate minus rest- and by 5 mmol/L afterward ing heart rate) divided by (predicted maxi- In metabolic alkalosis, for every 1-mmol/L increase in bicarbonate, mum heart rate minus resting heart rate). PCO2 may increase by 0.6 mm Hg

5 If the patient has metabolic acidosis with an elevated anion gap, check whether the bicarbonate level has decreased as much as the anion gap has increased In metabolic acidosis, the anion gap should increase by the same amount that bicarbonate decreases; a difference in these two changes is called a delta gap

PCO2 = partial pressure of carbon dioxide Based on information in reference 1

214 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 84 • NUMBER 3 MARCH 2017