LabMed

Overdose of Propylene Glycol

At a Glance

Propylene glycol (specifically 1,2-propanediol; PG) is widely used as a solvent and antimicrobial preservative, especially in intravenous (IV) pharmaceuticals. Poisoning is, therefore, usually iatrogenic, under-appreciated, and under-diagnosed. Since it is considered less toxic than ethylene glycol, it has been substituted for it in some antifreeze, and the incidence of toxic ingestion is expected to rise. However, rapid recognition and treatment are crucial to prevent serious sequelae. The lethal dose is greater than 10 g/kg.

PG poisoning should be suspected in any patient having received intravenous medications containing PG (>25 mg/kg/day) who develops an elevated osmolality and osmolar gap, anion-gap metabolic acidosis, central nervous system (CNS) depression, cardiovascular system (CVS) depression, seizures, and hypoglycemia. Likewise, in a patient with history of antifreeze ingestion with negative serum ethylene glycol or symptoms not typical of ethylene glycol intoxication, PG intoxication should be considered. Unlike ethylene glycol and other toxic alcohols, the majority of toxicity results from the PG itself, rather than its metabolites, so there is little delay in presentation and little benefit from ethanol administration.

What Tests Should I Request to Confirm My Clinical Dx? In addition, what follow-up tests might be useful?

There are no widely available or easily performed tests for PG. Ideally, the diagnosis should be made by determination of PG in the serum or urine by gas chromatography, but these methods are not standardized. If the PG determination is negative and the history positive for exposure, ethylene glycol or one of the longer glycols (e.g., 1,4-butanediol) should be considered. The limited availability and delay in reporting these tests means that the diagnosis will invariably be based on clinical history and other supporting tests.

It is important to determine whether ethanol is present by a qualitative test, and, if positive, a quantitative serum or breath ethanol should be obtained as its presence may confound the diagnosis. The presence of PG in the serum elevates the osmolal gap, but this declines as the PG is metabolized and is replaced by an anion gap metabolic acidosis (routine chemistries required) as lactate accumulated. Osmolality should be determined by freezing point depression, and the osmolal gap should be calculated. The patient's progress should be monitored regularly by his or her acid-base status and serum osmolality.

These readily available tests are rapid and inexpensive to perform. They are not, however, sufficiently sensitive to detect smaller exposures, since the population variability means that the gap can still lie within the normal range (up to about 75 mg/dL PG). Furthermore, the elevation in lactate is often much greater than the measured value.

Calcium oxalate crystals are not seen in the urine after PG exposure, nor is there hypocalcemia as is seen with ethylene glycol poisoning. Although there is absence of tubular damage from crystal deposits with PG, there can be renal impairment from the osmotic effects of PG. Hypoglycemia is also supportive. Few conditions besides ethylene glycol or methanol intoxication produce such a profound metabolic acidosis (bicarbonate <8 mmol/L), and most of these present in a characteristic fashion with a high measured serum lactate (status epilepticus, shock, ischemic bowel, or diabetic ketoacidosis). If the source of the PG was IV medications, it is likely that the acidosis is already partially corrected by direct intervention before the diagnosis of PG toxicity is considered. (Table 1)

Table 1

Test Results Indicative of the Disorder
Propylene Glycol Electrolytes and Blood Gas Osmolality Acids Routine Chemistry
Serum or urine > 10 mg/dL (Metabolic acidosis, compensating respiratory alkalosis) pH < 7.35 Bicarbonate <8 mmol/L pCO2 decreased Elevated anion gap > 20 mmol/L, especially later Elevated > 300 mOsm/kg Osmolal Gap > 20 mOsm/kg, especially early Lactate variable depending on assay but often < 3 mmol/L Creatinine rising, BUN less so Normo-calcemia Hypoglycemia

Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications - OTC drugs or Herbals - that might affect the lab results?

PG is widely available and has many legitimate uses, such as antifreeze and windshield washers, and there can be confusion regarding whether methanol, ethylene glycol, or PG is present in the product.

Osmolality should be measured by freezing point depression (vapor dew point pressure methods are not reliable with volatile osmolytes). Any small osmolyte, such as methanol, ethanol, isopropanol, acetone, or ethylene glycol, will contribute to an elevated osmolal gap. Of these, however, only PG, ethylene glycol, and methanol produce an elevated anion gap. For calculation of the osmolal gap, a simultaneous measurement of serum sodium, glucose, and blood urea nitrogen (BUN) are required.

Knowledge of ethanol co-ingestion is important only for determining its effects on the patient's mental status; it is not important for the evolution or treatment of PG intoxication.

Because of the nature of the toxicity, serum PG does not relate to the severity of the condition; the osmolality and pH are more informative.

PG does not appreciably affect rapid point of care or enzymatic assays for ethanol or breathe ethanol tests.

Interpret the lactate with caution; D-lactate metabolite of PG is not included in most assays.

What Lab Results Are Absolutely Confirmatory?

PG analysis by gas chromatography is the gold standard for establishing the diagnosis. In addition, this test usually detects ethylene glycol and possibly some longer chain glycols. A variety of home-brew solvent screens may be offered, and it is important to ascertain exactly what compounds are included, otherwise a diagnosis may be missed.

PG is stable in blood, but fluoride oxalate preservative should be used if ethanol determination is also required.

The serum PG concentration does not correlate with the severity of the intoxication, even if the time of ingestion or dose is known, since the degree of acidosis and the osmolar effects are the major determinants of outcome. Once the diagnosis is established and treatment instigated, repeat serum ethylene glycol determinations are not helpful, unless the intoxication does not resolve in the expected time. If the patient presents late, it may not be possible to demonstrate the presence of PG.

In the absence of PG assays, diagnosis and treatment is based predominantly on history (especially administered IV medications), acid-base status, and exclusion of other metabolic causes.

What Tests Should I Request to Confirm My Clinical Dx? In addition, what follow-up tests might be useful?

Additional useful tests

Urine pregnancy test should be performed in age-appropriate females.

A urine drug screen should be requested if there is a suspicion that other agents are involved. This can be a set of simple immunoassays for recreational drugs, which can be performed as a Point of Care or a laboratory-based test. Laboratory tests frequently also include acetaminophen and salicylate. This determines whether the clinical picture is due to PG alone or is confounded by co-ingestion of sedative hypnotic agents, such as benzodiazepines, barbiturates, or opioids, and can provide an indication of other toxins that might contribute to the cardiovascular instability or the metabolic acidosis.

Although the absence of birefringent calcium oxalate crystals in urinalysis directs the diagnosis away from ethylene glycol intoxication and toward PG, it should not be used as confirmatory. Sometimes, a fluorescent marker is added to the product by the manufacturer, and this may be detected in the urine with the use of a UV lamp. This test is subject to false positives, and the absence of fluorescence does not exclude a significant ingestion.

For calculation of the osmolal gap, a simultaneous measurement of serum sodium, glucose, and BUN are required. Once the presence of ethanol has been excluded, this information can be used to cautiously calculate the approximate amount of ethylene glycol present by the observed gap as follows:

Calculated osmolality mOsm/kg = 2[Na mmol/L] = [glucose mg/dL/18] + [BUN mg/dL/2.8]

Osmolal gap = Measured osmolality – Calculated osmolality which is normally <10 mosm/kg

Serum propylene glycol (mg/dL) = (10 – osmolal gap ) x 7.6

Remember to measure and factor in for any ethanol administered. Once the ethanol concentration is known, the portion of the osmolal gap attributed to the ethylene glycol can be used to calculate the methanol concentration:

Serum propylene glycol mg/dL = 7.6 [(10 – Osmolal gap mOsm/kg) – (Serum ethanol mg/dL/4.6)]

Specimens for solvent screens are often requested in specific containers with special handling conditions but should always be drawn with minimal airspace to avoid loss during transport.

Metabolism

The majority of the toxicity observed with PG is attributed to the parent compound. Approximately 50% is excreted unchanged, whereas the remainder undergoes hepatic metabolism via alcohol dehydrogenase (ADH) to lactic acid. It is important to realize that, although the body makes almost exclusively L-lactate, PG will make racemic (both D- and L-) lactate. The body has poor metabolic capacity for D-lactate, and this accumulates. However, because many lactate tests are specific for L-lactate, they grossly under-estimate the severity of the accumulated D-lactate (possibly by a factor of 10) and, therefore, the anion gap. Ethanol administration has little effect on PG elimination, and, since PG is relatively toxic, there is no rationale for delaying its metabolism.

Expected kinetics

Peak PG concentrations in the serum occur 1-2 hours after oral ingestion. The distribution volume is total body water (0.6 L/kg). Elimination half-life is about 3 hours, but is saturated at 50 mg/dL, and then proceeds at zero-order at 10-20 mg/dL/hr (similar to ethanol).

Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications - OTC drugs or Herbals - that might affect the lab results?

If the diagnosis is based on the osmolal gap, the presence of other osmolytes may confound the diagnosis. Critically ill patients may have elevated osmolal gaps. Categorical ruling out of PG poisoning on the basis of an osmolal gap less than 10 mOsm'kg is unjustified, as is assuming that a small elevation in the osmolal gap in a patient with a low pretest probability is due to PG. Osmolal gaps greater than 25 mOsm/kg are caused by few other substances, and a large unexplained gap is presumptive evidence of recent toxic alcohol ingestion in the appropriate clinical setting. Subsequent tests are needed to make the specific identification or narrow down the options.

A number of IV medicines are solubilized in propylene glycol, which allows them to be administered at physiological pH and affords antimicrobial protection. If the administered dose exceeds the patient's ability to clear it (renal elimination accounts for 50% of the dose), it will accumulate in the serum over time. This is relatively easy to achieve, and doses in the order of grams per day can be given without adequate appreciation given the content of some mediations commonly administered in the intensive care situation (i.e., lorazepam 80%, phenobarbital 70%, diazepam, pentobarbital and phenytoin 40%, etomidate 35%). Once the daily dose exceeds 25 mg/kg, accumulation of PG might be expected. Initial serum PG concentration can be calculated from the dose administered:

PG mg/dL = 100 x (mL infused) x (% composition x 1.08 g/mL) / (0.6 L/kg x weight in kg)

The contribution to the osmolal gap is predicted mOsm/kg = Propylene glycol mg/dL/7.6.

Lactate results should be interpreted with caution. Approximately 50% of the dose undergoes hepatic metabolism via ADH to lactic acid. It is important to realize that, although the body makes almost exclusively L-lactate, PG will make racemic (both D- and L-) lactate. The body has poor metabolic capacity for D-lactate, and this accumulates. However, because many lactate tests are specific for L-lactate, they grossly under-estimate the severity of the accumulated D-lactate (possibly by a factor of 10) and, therefore, the anion gap.

PG analysis by gas chromatography is the gold standard for establishing the diagnosis. However, these methods are not standardized, and a number of different internal standards and derivatives are used. Some longer chain glycols are currently abused (e.g., 1,4-butanediol), and it is possible that, in some methods, these or other ingested or endogenous small molecules could coelute; mass spectrometry is not usually used.

You must be a registered member of Renal and Urology News to post a comment.

Sign Up for Free e-newsletters