Theophylline toxicity: clinical features of 116 consecutive cases

To examine the predisposing factors, clinical and laboratory characteristics, management, course, and outcome of consecutive cases of theophylline toxicity in an outpatient setting.

Toxicology records and hospital charts of consecutive patients with a serum theophylline concentration (STC) greater than 30 mg/L (167 mumol/L) identified in the emergency departments (EDS) of a University Medical Center and a Veterans Administration Medical Center were reviewed.

Ten percent and 2.8% of 5,557 consecutive STCs measured in the EDs over 2 years were greater than 20 mg/L (111 mumol/L) and greater than 30 mg/L (167 mumol/L), respectively. One hundred sixteen cases with STC greater than 30 mg/L were identified. Fourteen (12%) and 102 (88%) were due to acute overdose and chronic overmedication, respectively. Principal predisposing factors included patient and/or physician dosing errors and conditions or medications that reduce theophylline clearance. One or more toxic manifestations were present in 109 (94%) cases. Fifty percent of patients had mild toxicity, 38% had moderate toxicity, and 7% had severe or life-threatening toxicity. Seven (6%) patients died when STC was still in the toxic range and/or as a result of toxicity. Acute overdose was associated with higher peak STC (p less than 0.001), younger age (p less than 0.01), and greater mortality (p less than 0.05) than chronic overmedication. Peak STC correlated significantly with the severity of toxicity for patients with acute overdose (p less than 0.01) but not for patients with chronic overmedication. All three patients with acute overdose and fatal toxicity had peak STCs greater than 100 mg/L (555 mumol/L) and fulminant toxicity, whereas the four patients with chronic overmedication who died during toxicity had peak STCs in the 40 to 60 mg/L (222 to 333 mumol/L) range and most died of respiratory failure rather than directly from toxicity. Patients with acute overdose who had the delayed onset of severe or life-threatening toxicity and/or died from toxicity were accurately identified using previously published criteria for prophylactic charcoal hemoperfusion. In contrast, the predictive value of the criteria applied to patients with chronic overmedication was poor. Two patients with acute overdose underwent charcoal hemoperfusion, but died. No patient with chronic overmedication received charcoal hemoperfusion.

Toxic-range STCs are relatively common in the ED population, occur primarily as a result of patient and physician dosing errors, and cause a broad range of toxic manifestations of varying severity. Peak STC correlates with the severity of toxicity and outcome for acute overdose but not chronic overmedication intoxication. Previously published criteria for prophylactic charcoal hemoperfusion accurately identify patients with acute overdose but not patients with chronic overmedication at risk for serious complications and death.