Metabolic acidosis

Metabolic Acidosis with an Elevated Anion Gap Determining the cause of metabolic acidosis with a high anion gap may present a diagnostic challenge. Possible causes include ketoacidosis, certain toxic ingestions, renal failure and lactic acidosis. Many of these entities present with nausea, vomiting and changes in mental status; however, there are specific hallmarks in the signs, symptoms and laboratory findings that help to differentiate among them. Patients presenting to the emergency room with metabolic acidosis with a high anion gap may create a diagnostic problem for the examining physician. At times, the etiology is apparent from the history and physical examination. At other times, the diagnosis may be elusive. The history and physical examination may not suggest a cause of the acidosis, or important information may be unavailable, as in the case of an unconscious patient or a nonverbal child. Since many of the causes of metabolic acidosis with a high anion gap are lethal if not properly treated, an awareness of the differential diagnosis and an organized approach to rapid diagnosis are essential.

Illustrative Case

A 44-year-old man with a history of paranoid schizophrenia and substance abuse was brought to the emergency department in an unresponsive condition. The patient was intubated, given naloxone (Narcan), 50 percent dextrose and thiamine. Oxygen was administered by nasal cannula at a rate of 10 L per minute.

Temperature was 36.6|C (97.8|F), pulse 76, respirations 20 and blood pressure 132/88. On physical examination, the patient had pinpoint pupils that were minimally reactive to light. He was unresponsive to painful stimuli. The physical examination was otherwise unremarkable. Laboratory data are summarized in Table 1.

Shortly after the initial presentation, further history was obtained from a friend of the patient. It was revealed that 12 hours before admission, the patient had threatened to drink antifreeze. The friend was uncertain of the amount of antifreeze the patient had actually ingested. The initial blood ethylene glycol level was 1,200 mg per dL (193 mmol per L). The patient was treated with ethanol and hemodialysis. The initial loading dose was 120 mL of 50 percent ethanol, delivered through a nasogastric tube; this was followed by 25 mL hourly. While the patient was on hemodialysis, the dose of ethanol was doubled to achieve a therapeutic blood level of approximately 100 mg per dL.

On the second hospital day, the patient was alert and extubated. His hospital course was complicated by renal failure and sepsis. One month after admission, the patient was discharged. Psychiatric follow-up was arranged at the local community mental health center.

Anion Gap and Osmolal Gap

Anion gap is defined as the difference between measured serum cations and anions. An elevated anion gap reflects an increase in "unmeasured" anions--usually organic acids--that are not normally included in standard determinations of other anions, such as chloride and bicarbonate. The anion gap is calculated by subtracting the sum of serum bicarbonate and chloride from the serum sodium (Table 2). Some formulas include serum potassium; however, the amount is negligible and is deleted from most calculations. A normal anion gap is 8 to 12 mEq per L (8 to 12 mmol per L).

An increased anion gap may be caused by an accumulation of organic acids, such as that seen in lactic acidosis, ketoacidosis, toxic ingestions and acute renal failure. The anion gap may also be increased because of reduced inorganic acid excretion, such as that seen in chronic renal failure.(1) Table 3 lists the causes of metabolic acidosis with a high anion gap.(2,3) Most of the underlying disorders produce similar symptoms, especially nausea, vomiting and a change in mental status, ranging from mild confusion to lethargy. However, a few characteristics signs, symptoms and laboratory findings help to differentiate among the causes (Table 4).

Osmolal gap is defined as the difference between calculated and measured osmolality (Table 5). The osmolal gap is often elevated in diabetic ketoacidosis and in methanol or ethylene glycol poisoning.(4)

Differential Diagnosis

KETOACIDOSIS

Ketoacidosis is caused by either an increase in the free fatty acid load to the liver or an increased conversion of free fatty acids to keto acids. The increased free fatty acid load may result from a high-fat diet. Increased conversion of free fatty acids to keto acids may occur in diabetic ketoacidosis, alcoholism or, to a lesser degree, prolonged starvation or a high-fat diet.(2,5) The most common keto acid formed is Beta-hydroxybutyrate, followed by acetoacetate and hydroxybutyric acid. A nitro-prusside reaction test (Acetest Reagent, Chemstrip K, Ketostix Strips) is commonly used to document ketones in serum and urine. This test reacts with acetoacetate and acetone, but not with Beta-hydroxybutyric acid. Therefore, the test may not reveal the full extent of ketoacidosis.(2)

Diabetic Ketoacidosis. The patient may have a history of diabetes mellitus (usually insulin-dependent), polydipsia, polyuria and polyphagia. On physical examination, the patient may be hyperventilating, hypovolemic and hypotensive, with acetone breath and changes in mental status.(2)

A laboratory definition of diabetic ketoacidosis is a serum glucose level greater than 300 mg per dL (16.7 mmol per L), ketones in the serum and a pH less than 7.3(6) Other laboratory findings may include hypoxia, leukocytosis, increased serum osmolality(7) and increased osmolal gap.(4) Serum sodium is often low secondary to hyperglycemia. Serum potassium is often elevated secondary to the acidosis, in spite of total body potassium deficits of up to 10 mEq per kg (10 mmol per kg) of body weight.(7) Diabetic ketoacidosis is treated with insulin, fluids and correction of the electrolyte disturbances.

Alcoholic Ketoacidosis. The patient, who usually has a history of alcohol abuse, presents with nausea, vomiting and diffuse abdominal pain. Generally, the patient has fasted for 24 to 36 hours, except for ingesting a significant amount of alcohol.(2) On physical examination, the patient may be dehydrated and malnourished and may have epigastric tenderness, an ethanol odor and variable changes in mental status. Laboratory studies usually reveal ketones, a high, normal or low glucose level, an elevated amylase level and hyperuricemia. A variable ethanol level may be present. Treatment of alcoholic ketoacidosis includes administration of glucose and saline-containing fluids, correction of metabolic derangements and nutritional support (including thiamine).

Starvation. This state results in accelerated gluconeogenesis with depletion of liver glycogen stores, hypoinsulinemia and lipolysis. In prolonged starvation (four to six weeks), ketogenesis serves to supply ketones for utilization by the brain.(8) The patient has a history of starvation and is usually cachectic, hypoglycemic and ketotic. Treatment involves nutritional support.

High-Fat Diets. A diet with a high fat content may cause a mildly elevated anion gap due to ketosis from lipolysis and increased Beta-oxidation of free fatty acids in the liver.(8) Treatment is proper dietary modification.

POISONING

Methanol, ethylene glycol, salicylate and paraldehyde poisoning can cause metabolic acidosis with a high anion gap. As mentioned earlier, ethylene glycol and methanol can also cause an increased osmolal gap.(4)

According to the 1987 Annual Report of the Poison Control Center National Data Collecting System,(9) of the more than 1 million reported poisoning cases, over 39,000 were due to alcohols and/or glycols, and over 20,000 were due to salicylates (both alone and in combination preparations). Paraldehyde poisoning is relatively rare.(2)

Methanol. Commonly known as wood alcohol, methanol is a clear liquid found in solvents, shellacs and varnishes. It is sometimes ingested by alcoholics as a substitute for ethanol. The usual lethal dose is 30 mL of absolute methanol, but deaths have been reported after ingestion of as little as 6 mL, and survival has been reported after ingestion of as much as 600 mL.(3) Peak methanol levels develop 30 to 60 minutes after ingestion, but there is usually a 12- to 24-hour latent period before symptom onset.

Classically, the patient describes cloudy, blurred or misty vision similar to "stepping out into a snowfield." The person may see yellow spots, develop a central scotoma or develop blindness, which may or may not be reversible.(3) These symptoms are caused by formaldehyde, a metabolite of methanol.(2) Other symptoms include nausea, vomiting, weakness, epigastric pain, headache, dizziness and central nervous system depression.(3) Examination of the eyes may disclose optic disc hyperemia and pericapillary edema, optic disc pallor and decreased pupillary reflex to light.(3)

Laboratory studies reveal metabolic acidosis with a high anion gap and an elevated osmolal gap. The high anion gap is caused mainly by formic acid, a metabolite of methanol. Lactate, acetate and butyrate contribute a small amount to the high anion gap.(2)

The treatment of methanol poisoning is administration of ethanol, because alcohol dehydrogenase (the enzyme that metabolizes ethanol, ethylene glycol and methanol) has a significantly greater affinity for ethanol than for the other alcohols.(2) Ethanol levels should be maintained at approximately 100 mg per dL (22 mmol per L). Ethanol can be administered either parenterally or enterally.(3) It is interesting that methanol intoxication with a normal anion gap and no visual impairment has been reported when there has been concurrent ethanol ingestion.(10)

An experimental drug, 4-methylpyrazole, has been shown in vitro and in human and animal studies to inhibit alcohol dehydrogenase. In animal studies, this drug lessens or eliminates the toxic effects of methanol poisoning.(11) Often, hemodialysis is required if the methanol blood concentration is greater than 50 mg per dL (16 mmol per L).(3)

Ethylene Glycol. This odorless substance is present in antifreeze, hydraulic brake fluid, cellophane softeners and solvents for paints and plastics.(2) Without treatment, ingestion of over 100 mL is usually lethal in adults.(12) The minimal lethal dose is 1 to 1.5 mL per kg.

In the illustrative case, the patient's ethylene glycol level of 1,200 mg per dL (193 mmol per L) was almost twice the highest level previously reported in a survivor (650 mg per dL [105 mmol per L])(13) and 12 times the minimum lethal dose (98 mg per dL [16 mmol per L]).

Peak ethylene glycol levels are reached after one to four hours, but toxic manifestations are delayed four to 12 hours.(3) Toxicity is classically divided into three stages: central nervous system injury (during the first 12 hours); respiratory depression and cardiopulmonary failure (at 12 to 24 hours), and renal failure (at 24 to 72 hours).(3,13) The toxic effects of ethylene glycol poisoning are produced by metabolites of ethylene glycol, including glycoaldehydes, glycolic acid, glyoxylic acid and oxalate.(13) Glycolic acid and some lactic acids are responsible for the high anion gap. Oxalate is the primary factor in renal toxicity.(3,13)

The diagnosis of ethylene glycol poisoning is supported by a urine sediment with needle-shaped (monohydrate) or octahedral-shaped (dihydrate) calcium oxalate crystals.(2,13) The maximum production of oxalate occurs eight hours after ingestion, so there may be no crystalluria if the patient presents earlier.(12)

Other laboratory findings include hypocalcemia, leukocytosis, and an elevated osmolal gap that is caused by the serum concentration of ethylene glycol. The osmolal gap may be normal if it is measured many hours after ingestion when ethylene glycol is no longer present in the serum. However, the serum level of glycolic acid, which causes the associated toxic effects, remains high.

Ethylene glycol poisoning is treated with supportive measures (e.g., respiratory support) and administration of ethanol.(3) If ethylene glycol levels exceed 50 mg per dL (8 mmol per L), or if renal failure is present, hemodialysis is indicated.(3,13) Some authors recommend thiamine, 100 mg intramuscularly, and pyridoxine, 100 mg intravenously or intramuscularly, although these have not yet been proved to be effective.(3) Both thiamine and pyridoxine are cofactors in the metabolism of ethylene glycol to nontoxic metabolites.(13) As in methanol poisoning, 4-methylpyrazole, administered either orally or intravenously, is being studied as an antidote for ethylene glycol poisoning.(11)

Salicylates. Found in numerous prescriptions and over-the-counter preparations, salicylates constitute a significant source of poisoning. The usual toxic dose is 200 to 300 mg per kg, with blood levels of 500 mg per dL (36.2 mmol per L) reported as potentially lethal. Peak levels occur two to four hours after ingestion of most preparations or six to nine hours after ingestion of enteric-coated tablets.(14)

The first manifestations of salicylate poisoning include tinnitus and hearing impairment. These symptoms occur at an average adult dose of 4.5 g per day. In mild toxicity, there might also be vomiting, which is usually seen three to eight hours after ingestion. Some hyperpnea and central nervous system depression may also occur. In moderate toxicity, symptoms include severe hyperpnea and marked lethargy or excitability. Severe toxicity is manifested by coma and, frequently, seizures.(14) Acute nonoliguric renal failure and hemostatic defects have also been reported.(14,15)

Salicylate directly stimulates the respiratory center, causing respiratory alkalosis. Occasionally, in adults or older children, this is the only acid-base abnormality seen. After the development of respiratory alkalosis, an increase in the metabolic rate with the production of more carbon dioxide may result in respiratory acidosis. At a later stage of poisoning, a direct toxic effect on carbohydrate metabolism produces the classic high anion gap metabolic acidosis.(14)

The Done nomogram, which is based on peak serum salicylate levels, is used to estimate the severity of a single overdose.(16) The severity of the poisoning is determined not by the absolute serum salicylate level but by the brain salicylate level, the peak serum concentration and the rapidity of decline in serum concentration.(14)

Mixing a few drops of the patient's urine with 10 percent ferric chloride produces a purple color if salicylate has been ingested, although only one 325-mg tablet containing salicylate can cause this response. Urinalysis may show proteinuria, tubular cells and granular casts.(12)

Therapy for salicylate poisoning consists of emesis with syrup of ipecac or, in a comatose patient, lavage coupled with activated charcoal. Salicylate excretion may be enhanced through correction of the fluid and metabolic disturbances and alkalinization of the urine with or without concomitant diuresis.(12) For severe poisoning (salicylate level greater than 100 mg per dL [145 mmol per L]), hemodialysis is indicated.(14)

Paraldehyde. This agent is used as a sedative and an antiseizure medication. The average minimal lethal blood level is approximately 500 micro g per mL.(17) Manifestations of toxicity include gastritis, renal failure, fatty changes in the liver, pulmonary hemorrhages, edema and congestive heart failure.(2)

On physical examination, patients with paraldehyde poisoning have a characteristic offensive odor, mild to moderate dehydration, hypotension and Kussmaul respirations. Other manifestations are mental status changes, guaiac-positive gastrointestinal contents and pulmonary edema.(12,17)

In paraldehyde poisoning, the elevated anion gap is caused by acetic acid and chloracetic acid. Diagnosis is made by detection of paraldehyde in the serum. Acetaldehyde can be found in the urine and blood, but many laboratories do not automatically test for this.(2) When a nitroprusside reaction test is used, paraldehyde may cause a false-positive reaction for ketones, called "pseudoketosis."(12)

Treatment of paraldehyde poisoning includes lavage, activated charcoal and supportive measures. Emesis should not be promoted, since paraldehyde is locally corrosive to the gastrointestinal tract and is rapidly absorbed.(17)

RENAL FAILURE

In acute renal failure, the glomerular filtration rate is decreased. Organic acids (phosphates and sulfates) are retained from endogenous metabolic sources, producing metabolic acidosis with a high anion gap. In chronic renal failure, ammonia excretion is diminished, causing the increased anion gap.(2)

Presentations vary with the degree of renal function. Treatment involves (1) correcting the underlying cause of renal failure, if possible, (2) supplying a low-protein, fluid-restricted diet, (3) administering bicarbonate or citrate to correct the acidosis, if necessary, and (4) correcting fluid and electrolyte abnormalities, which may require dialysis.

LACTIC ACIDOSIS

Cohen and Woods have developed a classification of lactic acidosis (Table 6).(12) In Type A lactic acidosis, there is poor tissue oxygenation and perfusion. In Type B, however, no evidence of decreased tissue perfusion is found, and the mechanism of the acidosis is unknown.

Patients with lactic acidosis may present with nausea, vomiting, restlessness, Kussmaul respirations, stupor or coma. The serum lactic acid level (as lactate) is elevated. Other laboratory abnormalities include hyperuricemia, hyperphosphatemia and leukocytosis.(12) Treatment of lactic acidosis is directed toward correction of the underlying cause. If the pH is less than 7.20, bicarbonate should be administered to maintain the pH between 7.20 and 7.25.(12)

Rapid Laboratory Evaluation

Since metabolic acidosis with a high anion gap has many different causes, an organized approach to diagnosis is needed. Rapid, accurate identification of the cause is necessary so that specific treatment measures can be started.

A urine dipstick test that is positive for both ketones and glucose rapidly confirms diabetic ketoacidosis. If the test is negative for glucose, another cause of ketoacidosis should be considered.

If there is a history of paraldehyde ingestion, and/or the urine contains acetaldehyde, paraldehyde poisoning is possible, and the finding of ketonuria may represent a false-positive reaction. A history of starvation and alcoholism suggests alcoholic ketoacidosis.

If the urine dipstick test is negative for ketones, the serum osmolality should be tested to determine whether there is an elevated osmolal gap. The causes of a high osmolal gap include ethylene glycol and methanol poisoning and diabetic ketoacidosis. If calcium oxylate crystals are found in the urine, ethylene glycol poisoning should be suspected. If there is a history of visual impairment or an abnormal funduscopic examination, methanol poisoning is the most likely cause.

When the osmolal gap is normal, the urine should be tested with ferric chloride. If a purple color develops, salicylate poisoning should be considered, although it is important to remember that the test is very sensitive. Renal failure is diagnosed when blood urea nitrogen and creatinine levels are elevated. Lactic acidosis is confirmed by an elevated lactic acid level.

Final Comment

Diagnosing the cause of metabolic acidosis with an elevated anion gap may be a dilemma if there is no helpful history or if the patient is an unconscious adult or a small child with an unknown ingestion. Although metabolic acidosis with an elevated anion gap has a wide range of causes (e.g., ketoacidosis, drug ingestion, renal failure, lactic acidosis), most cases present with similar symptoms (nausea, vomiting and change of mental status). However, certain salient features of the physical examination and laboratory investigation are characteristic of each disorder. An understanding of the causes of metabolic acidosis with a high anion gap, as well as an organized approach to laboratory evaluation and diagnosis, will enable the physician to start treatment expediently and thereby improve survival.

TABLE 1

Laboratory Data in Illustrative Case Arterial blood gases

pH--7.09

Pco2--14

Po2--230 Blood urea nitrogen--6 mg per dL (2 mmol per L) Creatinine--1 mg per dL (88 micro mol per L) Sodium--149 mEq per L (149 mmol per L) Potassium--5.3 mEq per L (5.3 mmol per L) Chloride--113 mEq per L (113 mmol per L) CO2--10 mEq per dL (10 mmol per L) Anion gap--26 mEq per L (26 mmol per L) Acetone--trace Hemoglobin--15 g per dL (150 g per L) Hematocrit--55% (0.55) White blood cell count--12,800 per mm raised to 3

(12.8 X 10 raised to 9 per L) Glucose--127 mg per dL (7.1 mmol per dL) Serum osmolality--290 mOsm per kg serum H2O

(290 mmol per kg serum H2O) Urine sediment--Hippuric acid and calcium oxalate

crystals

TABLE 2

Calculation of the Anion Gap Serum Na - (HCO3 + Cl) = anion gap (Normal anion gap is 8 to 12 mEq per L

[8 to 12 mmol per L].)

TABLE 3

Causes of Metabolic Acidosis with a High Anion Gap Ketoacidosis Diabetes Alcoholism Prolonged starvation (mild acidosis) High-fat diet (mild acidosis) Ingestions Elevated osmolal gap

Methanol

Ethylene glycol Normal osmolal gap

Salicylate

Paraldehyde Renal failure Acute Chronic Lactic acidosis Type A--decrease in tissue oxygenation Type B--no decrease in tissue oxygenation

TABLE 4

Diagnosis and Treatment of Conditions That Produce Metabolic Acidosis with a High Anion Gap

High osmolality High osmolal gap Ketones

Ketones

Ketones Nutritional support

Hypoglycemia Low albumin

High ethylene glycol level

Hyperkalemia Hyperuricemia Hyperphosphatemia Hypermagnesemia Hypocalcemia Normochromic,

normocytic anemia

Hyperphosphatemia Leukocytosis Normokalemia

TABLE 5

Calculation of the Osmolal Gap Measured osmolality - calculated osmolality =

osmolal gap Calculated osmolality =

2 Na + (Blood urea nitrogen/2.8) + (glucose/18) (Normal osmolality is 286 + 4 mOsm per kg serum

H2O [286 plus or minus 4 mmol per kg serum H2O]). Normal

osmolal gap is <10 mOsm per kg serum H2O

[10 mmol per kg serum H2O].)

TABLE 6

Cohen and Woods' Classification of Lactic Acidosis Type A (tissue hypoxia)

Shock states

Profound anemia

Massive catecholamine excess Type B (tissue oxygenation appears normal)

Diabetes mellitus

Liver failure

Renal failure

Carcinoma

Seizures

Alkaloses

Drugs/toxins

Inborn errors of metabolism

Hypoglycemia REFERENCES (1)Krupp MA. Fluid and electrolyte disorders. In: Schroeder SA, ed. Current medical diagnosis and treatment 1988. Norwalk, Conn.: Appleton & Lange, 1988:34-5. (2)Shaffer MA. Acid-base homeostasis. In: Rosen P, ed. Emergency medicine: concepts and clinical practice. 2d ed. St. Louis: Mosby, 1988:1956-62. (3)Litovitz T. The alcohols: ethanol, methanol, isopropanol, ethylene glycol. Pediatr Clin North Am 1986;33:311-23. (4)Jacobsen D, Bredesen JE, Eide I, Ostborg J. Anion and osmolal gaps in the diagnosis of methanol and ethylene glycol poisoning. Acta Med Scand 1982;212:17-20. (5)Andreoli TE. Disorders of fluid volume, electrolyte, and acid-base balance. In: Wyngaarden JB, Smith LH Jr, eds. Textbook of medicine. 18th ed. Philadelphia: Saunders, 1988:551-5. (6)Zack BG. Diabetic ketoacidosis in childhood. Am Fam Physician 1982;26(1):107-13. (7)Foster DW, McGarry JD. The metabolic derangements and treatment of diabetic ketoacidosis. N Engl J Med 1983;309:159-69. (8)Felig P, Havel RJ, Smith LH. Metabolism. In: Smith LH Jr, Thier SO, eds. Pathophysiology: the biological principles of disease. 2d ed. Philadelphia: Saunders, 1985:364-70. (9)Litovitz TL, Schmitz BF, Matyunas N, Martin TG. 1987 annual report of the American Association of Poison Control Centers National Data Collection System. Am J Emerg Med 1988;6:479-515. (10)Palmisano J, Gruver C, Adams ND. Absence of anion gap metabolic acidosis in severe methanol poisoning: a case report and review of the literature. Am J Kidney Dis 1987;9:441-4. (11)Baud FJ, Galliot M, Astier A, et al. Treatment of ethylene glycol poisoning with intravenous 4-methylpyrazole. N Engl J Med 1988;319:97-100. (12)Narins RG, Krishna GG, Bressler L, et al. The metabolic acidoses. In: Maxwell MH, Kleeman CR, Narins RG, eds. Clinical disorders of fluid and electrolyte metabolism. 4th ed. New York: McGraw-Hill, 1987:607-28. (13)Ellenhorn MJ, Barceloux DG. Alcohols and glycols. In: Medical toxicology: diagnosis and treatment of human poisoning. New York: Elsevier, 1988:782-812. (14)Seger DL. Aspirin and acetaminophen. In: Rosen P, ed. Emergency medicine: concepts and clinical practice. 2d ed. St. Louis: Mosby, 1988:2141-9. (15)Ellenhorn MJ, Barceloux DG. Salicylates. In: Medical toxicology: diagnosiand treatment of human poisoning. New York: Elsevier, 1988:562-72. (16)Sallis RE. Management of salicylate toxicity. Am Fam Physician 1989;39(3):265-70. (17)Ellenhorn MJ, Barceloux DG. Anticonvulsants. In: Medical toxicology: diagnosis and treatment of human poisoning. New York: Elsevier, 1988:241-5.

COPYRIGHT 1989 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group