Cocaine Toxicity Workup

Updated: Mar 11, 2023
  • Author: Lynn Barkley Burnett, MD, EdD, JD; Chief Editor: Sage W Wiener, MD  more...
  • Print

Laboratory Studies

No laboratory studies are indicated if the patient has a clear history and mild symptoms.

If history is absent or if the patient has moderate-to-severe toxicity, appropriate laboratory tests may include the following:

  • Complete blood cell (CBC) count
  • Electrolytes, BUN, creatinine, glucose (basic metabolic panel)
  • Glucose
  • Pregnancy test
  • Calcium
  • Creatine kinase (CK) 
  • Troponin
  • Urinalysis
  • Toxicology screens

A normal CK concentration may be used to help rule out rhabdomyolysis. An elevated concentration is nonspecific, with possible causes ranging from local trauma due to intramuscular injection, to myocardial infarction.

Urinalysis should include inspection to detect myoglobinuria. In cocaine-induced rhabdomyolysis, a dipstick urinalysis reveals an orthotoluidine reaction positive for heme in 75% of patients, findings positive for protein in 67%, and microscopic hematuria in some.

On a urine drug screen, drain cleaner or bleach added to urine can mask cocaine; alkaline urine may raise this suspicion. Desipramine and amantadine, prescribed to reduce cravings for cocaine, may cause false-positive results on urine tests for amphetamines. No substances result in a false-positive urine drug screen for cocaine, though false positives may still occur from switched or mislabeled specimens.

Urine, blood, gastric contents, and unknown substances found on patients, such as on a mustache, may be sent for toxicologic evaluation. Include the patient's clinical history and differential diagnosis of the toxins in question to guide the laboratory evaluations. High plasma cocaine concentrations are rarely observed because cocaine has a short half-life of 30-45 minutes. Furthermore, numerous studies have demonstrated that toxicology screening rarely changes the clinical treatment of patients. Although concentrations higher than 1 mg/L are generally associated with toxicity, deaths have been reported with blood levels of 0.1-20.9 mg/L. Because of this wide range of toxicity, quantitative blood levels of cocaine or metabolites are generally not clinically useful.

Cocaine exhibits first-order kinetics over a wide dose range; therefore, after 5 half-lives (approximately 4 h), virtually all of the cocaine taken should have been converted to its metabolites. Hollander et al concur, indicating that urinary cocaine may be detected for 4-8 hours after a single intranasal dose. However, Lewin, Goldfrank, and Weisman maintain that most cocaine is excreted in the urine within 24 hours of ingestion. [40]

Benzoylecgonine, which may induce neurotoxicity, ischemia, and arrhythmias, [41] may be present in urine for as long as 60 hours after single use and for as long as 22 days after the cessation of heavy cocaine use. If the ratio of benzoylecgonine to cocaine found in the urine is less than 100:1, either the cocaine was ingested less than 10 hours before collection of the sample or ongoing liberation of cocaine is occurring from a body package. The “cocaine” assay on most clinical urine drug screens are for benzoylecgonine and not cocaine itself.

Also consider tests of the following:

  • Cardiac markers in patients with chest pain; cocaine use does not affect the specificity of troponin assays [32]

  • Lactate dehydrogenase (LDH) level, aspartate aminotransferase (AST), prothrombin time (PT), activated partial thromboplastin time (aPTT), and cultures of blood and urine in patients with elevated temperatures

  • Evaluation of cerebrospinal fluid (CSF) for patients with altered sensorium and fever in whom meningitis is being considered

  • Serum osmolality and ketones in patients with altered mental status


Imaging Studies

Obtain a chest radiograph in patients with chest pain, hypoxia, or moderate-to-severe toxicity. The chest radiograph may reveal diffuse granulomatous changes resulting from chronic parenteral use due to the injection of inert insoluble ingredients of oral preparations or insolubles used to cut cocaine (eg, talc). Septic pulmonary emboli appear round or wedge shaped; they may clear rapidly or cavitate. Aspiration pneumonitis and noncardiogenic pulmonary edema may also be demonstrated. Pulmonary abscess may become evident after aspiration pneumonitis or after an intravenous injection of bacteria or toxic organic or inorganic materials.

The chest radiograph may reveal a needle that was lost during drug injection. An aneurysm or pseudoaneurysm may be noted with mainlining (directly injecting into a central artery or vein); this finding is an indication for further imaging studies.

Radiographs may be useful to evaluate cellulitis, abscess, or nonhealing wound in an intravenous drug user, and may reveal a foreign body or subcutaneous emphysema produced by gas-forming organisms in an anaerobic infection. Ultrasonography may identify foreign body or abscess.

Skeletal images can reveal osteomyelitis or fractures. However, osteomyelitis may not be demonstrable on plain images for 1-2 weeks, and other imaging studies should be performed if such a diagnosis is considered.

Radiography to investigate body packing

Swallowed packets of cocaine may rupture, resulting in acute cocaine poisoning. They may also critically obstruct the esophagus or small bowel, typically at the ileocecal valve, or may cause bowel ischemia. Drug packets typically weigh 1-12 g each.

Begin with plain images of the abdomen to search for packets. However, the rate of false-negative results is 1.2-33%. Radiographs depicting packets are shown below.

Patient transporting cocaine packets seen on KUB a Patient transporting cocaine packets seen on KUB and lateral radiographs (mostly on left side). The patient was admitted, and a large number of packets was later obtained without procedural intervention or complication.
Patient transporting cocaine packets seen on KUB a Patient transporting cocaine packets seen on KUB and lateral radiographs (mostly on left side). The patient was admitted, and a large number of packets was later obtained without procedural intervention or complication.

Results from ultrasonography are occasionally useful for identifying packets, but not sensitive enough to exclude their presence. Contrast-enhanced study of the bowel or abdominal CT may be the only way of identifying the packages. This is shown in the image below.

CT scan of patient transporting cocaine packets. CT scan of patient transporting cocaine packets.

Regular radiologic examination is imperative to confirm successful transit of packages through the GI tract.

McCarron and Wood reported a series of 75 patients with suspected cocaine body packing evaluated with kidneys, ureters, and bladder (KUB) radiography. Cocaine packages (15-175 per individual) were retrieved from 48 patients. In 73%, the KUB images showed foreign bodies. KUB findings were negative in 3% of patients with cocaine packages in the rectum and in another 16% who subsequently passed 15-135 packages. [42]

McCarron and Wood identified 3 types of packages, with the following physical and radiographic characteristics and risk for rupture: [42]

  • Type 1: Condoms, toy balloons, or fingers of latex gloves contained cocaine in loose white powder form. Typically, the package material was stuffed with cocaine, tied, folded back on itself, and tied again at the opposite end. A variation involved wrapping a package with masking tape to make a small bundle, then covering it with 2 more condoms tied with fishing line. Type 1 packages radiographically appeared as well-defined circular or cigar-shaped white opacities. Ties, if radiographically apparent, had a rosette appearance. If gas halos were observed, they were irregular. This type of cocaine package posed the highest risk for breakage or leaching.

  • Type 2: About 5-7 layers of tubular latex, with a smooth tie on each end, covered white or light yellow matted cocaine powder. On direct inspection of the bundles, they appeared light yellow and were relatively large and uniform in size. These radiopaque bundles were oblong. When viewed radiographically, gas halos were present and regular, but no ties were apparent. Type 2 packages were less susceptible to breaking than type 1 packages.

  • Type 3: Yellow, hardened cocaine paste wrapped in aluminum foil, then wrapped again with 3-5 layers of tubular latex and securely tied at both ends. These packages were hard, smaller than type 1 and type 2 packages, and irregularly sized. They did not appear as foreign bodies on abdominal images. No breakage or leaching of cocaine was reported with this type of package.

The risk of bag rupture or leaching increases with dwell time. In one series, 2 patients had sloughed pieces of wrapping, and 1 had evidence of leaching. After the container believed to be the last has passed, Perrone and Hoffman recommend imaging (eg, Gastrografin upper-GI series with small-bowel follow-through or CT of the abdomen and pelvis) to ensure that the GI tract has been fully purged of all packets. [43]

American College of Emergency Physicians guidelines recommend brain CT for a patient presenting with first-time seizure. Perrone and Hoffman recommend CT scan of the head in all patients with cocaine-associated seizures, as intracranial pathology is often identified. [43]


Other Tests

Obtain a 12-lead electrocardiogram (ECG) in patients with any of the following:

  • Chest pain
  • Hypoxia
  • Dyspnea
  • An irregular, rapid, or slow pulse
  • Altered mental status
  • Moderate-to-severe toxicity

Of 48 patients admitted to an intensive care unit with cocaine-induced chest pain, 86% had abnormal ECG findings, but only 6% were found to have sustained a nyocardial infarction.

The Brugada sign has been noted in cocaine users. This finding should prompt consideration of its implications for sudden cardiac death. Note the images below.

Schematics show the 3 types of action potentials i Schematics show the 3 types of action potentials in the right ventricle: endocardial (End), mid myocardial (M), and epicardial (Epi). A, Normal situation on V2 ECG generated by transmural voltage gradients during the depolarization and repolarization phases of the action potentials. B-E, Different alterations of the epicardial action potential that produce the ECGs changes observed in patients with Brugada syndrome. Adapted from Antzelevitch, 2005.
Three types of ST-segment elevation in Brugada syn Three types of ST-segment elevation in Brugada syndrome, as shown in the precordial leads on ECG in the same patient at different times. Left panel shows a type 1 ECG pattern with pronounced elevation of the J point (arrow), a coved-type ST segment, and an inverted T wave in V1 and V2. The middle panel illustrates a type 2 pattern with a saddleback ST-segment elevated by >1 mm. The right panel shows a type 3 pattern in which the ST segment is elevated < 1 mm. According to a consensus report (Antzelevitch, 2005), the type 1 ECG pattern is diagnostic of Brugada syndrome. Modified from Wilde, 2002.

Coma has several possible etiologies in the cocaine-toxic patient, including the second (nonconvulsive) stage of status epilepticus. Therefore, immediate electroencephalography (EEG) is indicated in patients presenting with unexplained coma in whom this is thought to be a possibility.



Renzi recommends lumbar puncture (LP) to rule out intracranial hemorrhage in patients with persistent headache after the patient's BP is normalized and contraindications are ruled out on head CT. [44] When LP is considered for this possible indication, remember that headaches are common in cocaine users secondary to decreased uptake of serotonin. Consider LP in all patients with hyperthermia or altered mental status.