• 
• 

Three Part Question

In [patients presenting with a possible drug intoxication] does a [urine drugs-of-abuse screen] change the [clinical management]?

Clinical Scenario

A 30 year old man is admitted to the Emergency Department with a history of an unknown drug ingestion. He is stuporous and unable to give a coherent history. His observations are abnormal and he undergoes emergency tracheal intubation to protect his airway due to a decreasing level of consciousness.

Search Strategy

Pubmed
[urine drug screen OR urine toxicology screen OR drugs of abuse]AND[emergency department] in titles and abstracts. LIMIT to English and Human.

Search Outcome

151 articles identified of which 20 were relevant and citation searched. Three were irrelevant and one was of insufficient quality.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Qirbi 1977 Canada	235 patients with suspected or confirmed drug overdose.	Retrospective cohort study	Evaluation of toxicology service.	54% of patient had positive drug screen (49% for patients who were drowsy and 90% for patients deeply unconscious). Multiple drug overdoses more common in comatose patients. No change in patient management.	No mention of which tests performed on urine. Outcome poorly defined.
Kellerman 1987 USA	196 patients excluding major trauma, alcohol intoxication without suspicion of other drugs and/or non-pharmacological complications of drug abuse. Questionnaire completed by doctors. Tested urine and/or gastric fluid and blood for drugs of abuse: benzodiazepines (EMIT), amphetamines, cocaine, opiates and barbiturates (TLC).	Prospective cohort study	Change in management, presumptive diagnosis, diagnostic certainty and suspected agents based on drug screen result.	Three urine-tested patients had management changed, but the changes are not applicable to today’s standards of care. No change in subsequent hospital management. Qualitative drug screens associated with change in diagnosis or diagnostic certainty in 66.3% of cases.	Trauma patients only. Inadequate description of testing procedures.
Brett 1988 USA	198 patients admitted with diagnosis of intentional drug overdose. Blood, urine and gastric contents tested.	Retrospective cohort study	Change in management of patients based on toxicology results.	Management changed in 3 patients by initiation of forced alkaline diuresis for barbiturate poisoning, but they were recovering regardless and it is doubtful this therapy made any important contribution to the management.	Single drug overdoses with delayed deleterious effects excluded. Inadequate explanation of testing methods.
Sloan 1989 USA	623 trauma patients with altered mental status. Urine and blood tested.	Retrospective cohort study	How drugs influence the need for therapeutic interventions.	Urine tox screens positive in 84% of patients. Ethanol, cannabinoids and cocaine most commonly found. No change in therapeutic interventions.	Trauma patients only studied.
Rygenstad 1990 Norway	145 self-poisoned patients or patients who warranted admission to hospital in 1987 study group. Blood and urine samples taken. Urine tested for amphetamines, benzodiazepines, barbiturates, cannabis, opioids and cocaine (EMIT).	Prospective cohort study	Change in outcome of patients.	Many false positive (opioid) and false negative (benzodiazepine) urine test results. No change in the outcome or management of patients.	Patients inappropriately treated by medical personnel excluded. Two-part study ten years apart with differing tests.
Clark 1991 USA	177 trauma patients with altered mental status. Urine and blood tested. Drugs of abuse tested for: cocaine, marijuana, benzodiazepines, opiates.	Retrospective cohort study	Usefulness of toxicology screens.	72% positive results for one or more substances. Ethyl alcohol, marijuana and cocaine most commonly found. Drugs can alter trauma score results, but does not change clinical management.	Screening at discretion of treating doctor. Testing methodology poorly described.
Sporer 1992 USA	61 patients with a history of overdose and were alert and arousable and remained so for 30 minutes in the Emergency Department. Blood and urine samples tested. Urine drugs of abuse tested for (EMIT): barbiturates, cocaine, opiates, phencyclidine, amphetamines and cannabinoids.	Randomised trial	Influence on clinical management and outcome.	Recommend not doing screening in minimally symptomatic patients. One patient had management changed, but it is doubtful this therapy made any important contribution to the patient’s outcome.	Patients with a depressed or deteriorating (within 30 minutes of arrival) mental state, unstable vital signs or requiring advanced life support were excluded. Randomised by time of presentation (bias). No power analysis.
Olshaker 1997 USA	345 psychiatric patients screened for ‘medical clearance’. Blood and urine tested for drugs of abuse: cocaine, opiates, amphetamines, tetrahydrocannabinol and phencyclidine (EMIT)	Retrospective cohort study	Frequency of medical conditions and yield of toxicology investigations.	History alone had a 94% sensitivity for identifying medical conditions. Routine toxicology screening was unlikely to produce information that was not available on history alone and does not change clinical management.	Psychiatric patients only. Selection bias.
Sugarman 1997 USA	338 Patients aged 0-18, of which 294 had blood and/or urine tested for cocaine, benzodiazepines, cannabinoids and opiods (GC-MS).	Retrospective cohort study	Clinical utility of toxicology screens in a paediatric emergency department.	57% screens positive for at least one drug. The change in management of three patients was based on suspected medication taken in overdose, confirmed by laboratory results. No change in management related to drugs of abuse.	Children only included in study.
Skleton 1998 Australia	200 patients with deliberate self-harm. Urine tested for opiates, cannabinoids, and amphetamines.	Retrospective cohort study	Assessment of current limited toxicology screening and if it needs to change.	Questionnaire given to doctors. Authors still recommend routine testing for alcohol, benzodiazepines, paracetamol, salicylate and tricyclic antidepressants.	Only deliberate self-harm patients reviewed.
Belson, Simon, Sullivan 1999 USA	55 patients <18 years with suspected drug ingestions in two hospitals. Blood and urine tested. Urine drugs of abuse tested for (EMIT): cocaine, barbiturates, benzodiazepines, amphetamines, opiates and marijuana.	Prospective cohort study	How positive as well as negative drug tests affected patient management.	Two of the four were 1-year old twins admitted for social reasons, the third was a 14 year old boy referred for psychiatric evaluation and the fourth was a 15 year old boy who discharged himself against medical advice.	Urine tests differed between two sites. Children only. Inadequate explanation of management changes in the four cases.
Belson, Simon 1999 USA	270 patients <19 years who had comprehensive and limited drug screens included. Urine and blood tested. Urine tested for benzodiazepines, cocaine, amphetamines, opiates, phencyclidine and barbiturates (EMIT).	Retrospective cohort study	Comparison between utility and cost-effectiveness of HPLC vs. limited toxicology screen	The comprehensive drug screen was of no additional value in the clinical management of the patients. Limited screening for drugs of abuse also did not change any management.	Patients <19 years old only tested. Inconclusive cost analysis.
Bast 2000 USA	Drug screening done on 2678 patients of which 414 positive and 401 available charts reviewed. Urine drug screens for opiates, barbiturates, amphetamines, phencyclidine, cocaine, marijuana and benzodiazepines.	Retrospective cohort study	Impact of positive drug screen in early management of trauma patients.	Most commonly encountered substance was marijuana (39%), benzodiazepines (33%) and cocaine (25%). Treatment is based on clinical assessment rather than a positive drug screen result. No change in clinical management.	Trauma patients only. Only positive screen results reviewed. Testing method not described.
Korn 2000 USA	212 patients >16 years who presented with a psychiatric complaint or required psychiatric evaluation before discharge. Used blood and urine for screening.	Retrospective cohort study	Differentiate screening requirements of patients with psychiatric and medical complaints from isolated psychiatric complaints.	No change in the management of psychiatric patients or psychiatric patients with medical problems based on toxicology laboratory results.	No mention of drugs tested for or method of testing. No mention of toxicology results. Outcomes poorly measured.
Schiller 2000 USA	392 psychiatric patients, randomised to mandatory testing and ‘usual care’. Urine drug screens using ‘standardised immunoassays’ for ethanol, amphetamine, metamphetamine, barbiturates, benzodiazepines, cocaine, opiates and methadone.	Randomised controlled trial	Change in disposition of psychiatric patients with manadatory urine drug screen vs. optional urine drug screen (usual-care group).	No significant difference in disposition between mandatory and usual-care group. Clinicians were accurate in their suspicion of substance abuse. Routine urine drug screening not recommended and does not change disposition.	Psychiatric patients only. Type of test used not mentioned.
Montague 2001 Australia	107 patients within 24h of a self-harm drug ingestion. Urine drug screens (EMIT) for opiates, benzodiazepines, cannabis, cocaine, amphetamines and methadone.	Prospective cohort study	Clinical usefulness of urine drug screens in management of overdose patients.	Benzodiazepines most common self-administered drug. Concluded that it is unnecessary to perform urine drug screens on a routine basis on overdose patients.	Management strategies outdated. Inappropriate conclusions extrapolated from study.

Comment(s)

Sixteen publications over the last few decades have addressed the change in clinical management of patients presenting with possible drug ingestion. Eleven of the papers appraised did not find any change in the management of patients based on the results of the urine drug screen. Kellerman et al noted a change in the management of three patients based on the urine drug screen results, but the changes are not applicable to today’s accepted standard of medical practice. Management was changed by alkalinising urine in four patients, but their clinical state did not warrant the change and they were improving regardless of therapy. Sugarman et al noted a change in management of three patients, but it was not related to drugs of abuse. Four children had their disposition changed because of cocaine detected on the urine drug screen, but little detail is provided of these cases. Their conclusion was still that “Qualitative drug screens provide minimal useful information” and continue to suggest urine drug screen findings are unlikely to have any impact on clinical management decisions.

Clinical Bottom Line

In patients presenting to an Emergency Department with a possible drug intoxication, the use of a urine drugs-of-abuse screen does not appear to change the clinical management.

References

1. Qirbi AA, Poznanski WJ Emergency toxicology in a general hospital Can Med Assoc J 1977;116:884-888
2. Kellerman AL, Fihn SD, LoGerfo JP, Copass MK Impact of drug screening in suspected drug overdose. Ann Emerg Med 1987;16:1206-1216
3. Brett AS Implications of discordance between clinical impression between clinical impression and toxicology analysis in drug overdose. Arch Intern Med 1988;148:437-441
4. Sloan EP, Zalenski RJ, Smith RF et al. Toxicology Screening in Urban Trauma Patients: Drug Prevalence and Its Relationship to Trauma Severity and Management. J of Trauma 1989;29(12):1647-1653
5. Rygenstad T, Aarstad K, Gustafsson K, Jenssen U. The clinical value of drug analysis in deliberate self-poisoning. Hum Exp Toxicol 1990;9:221-230
6. Clark RF, Harchelroad F. Toxicology Screening of the Trauma Patient: A Changing Profile. Ann Emerg Med 1991;20:151-153
7. Sporer KA, Ernst AA. The effect of toxicological screening on management minimally symptomatic overdoses Am J Emerg Med 1992;10:173-175
8. Olshaker JS, Browne B, Jerrard DA, et al. Medical Clearance and Screening of Psychiatric Patients in the Emergency Department. Acad Emerg Med 1997;4(2):124-128
9. Sugarman JM, Rodgers GC, Paul RI. Utility of toxicology screening in a paediatric emergency department. Paed Emerg Care 1997;13(3):194-197
10. Skelton H, Dann LM, Ong RTT, et al. Drug Screening of Patients Who Deliberately Harm Themselves Admitted to the Emergency Department. Ther Drug Monit 1998;20(1):98-103
11. Belson MG, Simon HK, Sullivan K, et al. The utility of toxicologic analysis in children with suspected ingestions. Paed Emerg Care 1999;15(6):383-387
12. Belson MG, Simon HK. Utility of Comprehensive Toxicologic Screens in Children. Am J Emerg Med 1999;17:221-224
13. Bast RP, Helmer SD, Henson SR, et al. Limited Utility of Routine Drug Screening in Trauma Patients. South Med J 2000;93(4):397-399
14. Korn CS, Currier GW, Henderson SO. “Medical Clearance” of Psychiatric patients without medical complaints in the Emergency Department. J Emerg Med 2000;18(2):173-176
15. Schiller MJ, Shumway M, Batki SL. Utility of Routine Drug Screening in a Psychiatric Emergency Setting. Psych Serv 2000;51(4):474-478
16. Montague RE, Grace RF, Lewis JH, et al. Urine drug screens in overdose patients do not contribute to immediate clinical management. Ther Drug Monit 2001;23:47-50