| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Special Report |
1 Department of Pathology and Laboratory Medicine, Hartford Hospital, Hartford, CT 06102.
2 Department of Emergency Medicine, Medical Toxicology, Hartford Hospital, Hartford, CT 06102.
3 Department of Clinical Laboratory Sciences, Louisiana State University Health Sciences Center, New Orleans, LA 70112.
4 Department of Emergency Medicine, Bellevue Hospital Center, New York, NY 10016.
5 Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, NY 14642.
6 Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN 55905.
7 Department of Emergency Medicine, University of Cincinnati Hospital, Cincinnati, OH 45267.
8 Department of Pathology, Kaiser Permanente Regional Laboratory, Clackamas, OR 97015.
9 Department of Medical Toxicology Good Samaritan Regional Medical Center, Phoenix, AZ 85006
aCommittee Chairman. Address for correspondence: Department of Pathology and Laboratory Medicine, Hartford Hospital, Hartford, CT 06102.
 |
Abstract |
|---|
 Top Abstract PreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
Methods: Recommendations for the use of clinical laboratory tests were prepared by an expert panel of analytical toxicologists and ED physicians specializing in clinical toxicology. These recommendations were posted on the world wide web and presented in open forum at several clinical chemistry and clinical toxicology meetings.
Results: A menu of important stat serum and urine toxicology tests was prepared for clinical laboratories who provide clinical toxicology services. For drugs-of-abuse intoxication, most ED physicians do not rely on results of urine drug testing for emergent management decisions. This is in part because immunoassays, although rapid, have limitations in sensitivity and specificity and chromatographic assays, which are more definitive, are more labor-intensive. Ethyl alcohol is widely tested in the ED, and breath testing is a convenient procedure. Determinations made within the ED, however, require oversight by the clinical laboratory. Testing for toxic alcohols is needed, but rapid commercial assays are not available. The laboratory must provide stat assays for acetaminophen, salicylates, co-oximetry, cholinesterase, iron, and some therapeutic drugs, such as lithium and digoxin. Exposure to other heavy metals requires laboratory support for specimen collection but not for emergent testing.
Conclusions:Improvements are needed for immunoassays, particularly for amphetamines, benzodiazepines, opioids, and tricyclic antidepressants. Assays for new drugs of abuse must also be developed to meet changing abuse patterns. As no clinical laboratory can provide services to meet all needs, the National Academy of Clinical Biochemistry Committee recommends establishment of regional centers for specialized toxicology testing.
|
Preamble |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
These recommendations were presented in open forum at several meetings during the year 2001: a local clinical chemistry section meeting at the Royal Brisbane Hospital (Brisbane, Australia) in January; the Midwest Association for Toxicology and Therapeutic Drug Monitoring, William Beaumont Hospital (Royal Oak, MI), in May; Edutrak Sessions at the AACC Annual Meeting (Chicago, IL) in August; and in October, The North American Congress of Clinical Toxicology (Montreal, Canada), The Society of Forensic Toxicology (New Orleans, LA), and the Scientific Assembly Toxicology Section meeting of the American College of Emergency Physicians (Chicago, IL). Participants at each meeting discussed the merits of the recommendations. A summary of these discussions are presented herein.
These guidelines cover four major areas. The section on each recommendation contains background information and summarizes the discussions by the Committee and participants of the various sessions on the rationale for that recommendation. We also provide qualitative ratings for the degree of consensus for adoption of the recommendations, based on discussions among the participants at the various presentations and correspondence received: "A" indicates general consensus by most participants, whereas "B" indicates either no consensus or that the recommendation was not applicable to all situations.
|
Introduction and Needs Assessment |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
lists results for 2002. The statistics refer to patients 6–97 years of age whose primary presenting problem was associated with drug use but was not necessarily the sole reason for the ED visit. This database is not a measure of illicit drug or substance abuse prevalence in the general population. Moreover, these statistics are based on self-reporting by the user and were not necessarily confirmed by laboratory testing. Some drugs (e.g., cocaine and heroin) may have a higher association with ED visits than others because they produce greater acute toxicity. Alcohol is not tabulated separately by DAWN; however, many studies have demonstrated a high prevalence of alcohol and substance abuse in ED patients, particularly trauma patients. Prevalence rates of 
25%, along with other data, suggest that nearly 30 million ED visits per year could be associated with some form of drug use (3).
|
There are other substances that can contribute to significant acute clinical problems for which laboratory testing might play an important role. Some of these are tabulated each year by the Toxic Exposure Surveillance System of the American Association of Poison Control Centers (4). In 2000, for example, there were 13 000 reported exposures to organophosphorus compounds, 16 000 to rodenticides (anticoagulants), 12 500 to heavy metals, 17 000 to carbon monoxide, and 1000 to toluene. It should be noted that the majority of these exposures were managed in a non-healthcare facility, usually at the site of exposure.
There will always be pressure between the need to make decisions quickly in the ED setting and the availability of reliable information to guide those decisions. This is true for all aspects of care, from obtaining a history and physical exam to laboratory testing and evaluating response to therapy or likely course of illness and future care. With regard to analyses performed in toxicology laboratories, it is known that a thorough toxicology screening incorporating various methods can identify many more substances than are clinically suspected. At the same time, this information often has no clinical utility because of the time required for specimen delivery, preparation, analysis, and reporting or because the presence of the substance is inconsequential. This has contributed to a range of clinical opinions and practices, from a minimalist approach to a "shotgun" approach of broad-based laboratory testing. These guidelines can be used to discuss the pros and cons of both approaches.
Finally, these recommendations identify laboratory support measures that can improve patient care. They will need to be adapted to specific situations, such as the evaluation of possible child abuse or so-called emergency psychiatric clearance. They do not address every question or identify every substance that might cause an individual to seek emergency care. For example, certain substances are explicitly identified as not requiring stat analytic identification, whereas many are not mentioned at all. The former are often agents that manufacturers have historically included on instrument menus or that certain third-party regulators or agencies have required. When there is no current rationale for these practices, we hope this document can be used cooperatively by laboratory and ED directors and their respective organizations in concert with manufacturers to make changes locally and nationally. On the other hand, these recommendations are designed to be useful to the 30 000 physicians making decisions in EDs across the US. As such, they do not represent the minimum laboratory evaluation that may be used by a specialist in medical toxicology, nor do they reflect all of the current analytic limitations present in various areas of the country. However, they can serve as a forum for discussion of the toxicology laboratory support that can and should be provided within a given patient population, institution, or geographic region of the country.
|
Part I. General Principles for Drug Testing to Support ED Toxicology |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
The basis and inclusive requirements behind the tests listed under the first tier include clinical relevance, available analytical assays for stat testing, and results that may have an immediate impact on subsequent management decisions or patient care. Management decisions need not necessarily relate to acute overdose therapy because results of urine drug testing are also used to determine admission to psychiatric wards. Given the problems with existing immunoassays with regard to sensitivity and specificity, the need for such drug testing is questionable if the patient is not in acute distress. Both falsely positive and negative results can lead to unnecessary investigations. 2
Recommendation.
The clinical laboratory should provide two tiers of drug testing. The first tier includes stat testing of selected target quantitative tests in serum or plasma (Table 2
) and qualitative tests in urine (Table 3
). If the patient is in no acute distress, additional or even most initial toxicology testing may be unnecessary. Degree of consensus: A for the analytes listed in Table 2
, B for the analytes listed in Table 3
.
|
|
Discussion.
The need for stat quantitative serum or plasma assays was not heavily debated during the presentations. Testing for acetaminophen, salicylates, iron, co-oximetry, iron, and toxic alcohols is discussed in Parts III and IV. The toxicities and need for monitoring of the therapeutic drugs listed in Table 2
are not discussed in this document but are listed under the Tier I tests for completeness purposes. The Tier 1 stat quantitative serum/plasma toxicology test menu (Table 2
) is nearly identical to a recent joint guideline published by the United Kingdom Group of the National Poisons Information Service (NPIS) and the Association of Clinical Biochemists (ACB) (5). The differences are the inclusion of the anticonvulsants in the NACB guideline and the omission of paraquat, which was listed in the NPIS/ACB guideline. In the US, paraquat exposure is rare: there were 120 total reported exposures and no deaths in 1998 (4).
In contrast to the situation with serum/plasma testing, the need for stat qualitative urine assays was questioned by many ED physicians. Discussion focused not only on which assays should be included or omitted, but the need for urine drug screening itself, given the inaccurate data that many immunoassays provide and a perceived lack of impact on acute patient care. The finding of a positive result in urine may be completely incidental to the presentation given the lack of an association between clinical impairment and the presence of a drug or its metabolite in urine and considering the length of time that the drug or analyte may be detectable in urine (2–30 days or more depending on the analyte and use pattern). As such, this document’s conclusions are in disagreement with the Laboratory Guideline for the Investigation of the Poisoned Patient prepared by the Alberta Medical Association (Alberta, Canada), which recommended that "A nonspecific toxicology screen is of limited value in the majority of cases and is rarely indicated" (6). The NPIS/ACB have recommended that urine toxicology screens be listed under the category of specialist or infrequent assays (5). The Medicare Carrier Advisory Committee have recently proposed a policy that a qualitative drug screen is not medically reasonable or necessary in known overdose cases when the patient is asymptomatic (i.e., responsive to verbal stimuli and has no seizures, hypoventilation, or cardiac abnormalities other than sinus tachycardia after several hours of observation). Although the NACB Committee agrees in principle with these guidelines and recommendations, the Committee is also of the opinion that few EDs and clinical laboratories will abandon their reliance on urine drug screens on the basis of its recommendations. This may be particularly true in small or rural hospitals that do not have clinical toxicology specialists available who understand these analytical limitations. In addition, many practitioners utilize urine drug screens to increase their confidence in making diagnoses and disposition decisions. Therefore, if urine drug testing is inevitable, ED physicians and clinical laboratories should encourage the in vitro diagnostic industry to take steps to improve analytical methodologies so that they can deliver more accurate drug testing services for the identified drug class. Important issues include assay specificity, cross-reactivity studies, cutoff concentrations (and the definition of false positive and negatives), confirmation testing, and issues regarding chain of custody.
There was considerable debate as to the wisdom of including tricyclic antidepressants (TCAs) to the Tier 1 test list, given the poor specificity of existing immunoassays for TCAs. Despite these problems, the finding of a positive assay may increase the awareness of other possible ingested medications. Therefore, a qualifier has been included in Table 3
to indicate that this is an area where ongoing education of clinical staff is necessary for optimum utilization of this test. In the absence of such education programs, the ED staff might consider this test too inaccurate for naive users and choose to omit it in the Tier 1 panel. Another issue with TCA testing is the appropriate urine cutoff concentration. If the intent is to determine therapeutic and toxic concentrations, a low cutoff might be appropriate, such as 300 µg/L. If the intent is to detect toxic concentrations, which is the most likely the goal in the ED patient, then a cutoff of 1000 µg/L may be more appropriate. This latter cutoff might also reduce the number of false-positive urine drug results attributable to non-TCA drugs. A clinical evaluation of TCA cutoff concentrations in urine to justify these statements has not been published to date (7). It is likely, however, that a single TCA cutoff consistent with toxicity cannot be established.
During the open discussion, a participant questioned the applicability of the Tier 1 test menu for pediatric patients. It is the opinion of the Committee that individual hospitals and medical centers should revise this list as appropriate to handle special populations and needs. Accidental overdoses of prescription or over-the-counter medications such as acetaminophen, salicylates, and iron tablets may be more of a concern in children than street drugs such as marijuana and cocaine, which are abused by older adolescents and adults. Fortunately, the implementation of "tamper-resistant" packaging has greatly reduced the number of accidental poisonings by toddlers (8). Other substances continue to be a problem for children under 6 years, including cosmetics, cleaning substances, foreign bodies, plants, topicals, cough and cold preparations, pesticides, vitamins, gastrointestinal preparations, antimicrobials, arts/crafts/office supplies, antihistamines, hormone and hormone antagonists, and hydrocarbons (4). Results of drug testing can also have a forensic impact on allegations of child abuse or neglect and custody disputes for children and their caregivers.
Drugs-of-abuse patterns among teenagers and young adults may be dramatically different from patterns in older adults, e.g., abuse of lysergic acid diethylamide (LSD), methylenedioxymethamphetamine (Ecstasy), and 
-hydroxybutyrate (GHB). However, the frequency of ED presentation for acute care related to these drugs or the utility of an analytic tool in changing management do not justify their inclusion as a Tier 1 recommendation at this time. Proproxyphene use can contribute to significant toxicities, including cardiac arrhythmias, possibly delayed in appearance. Nevertheless, a laboratory may choose to exclude propoxyphene or any other drug from the Tier 1 list if the prevalence of this drug is low in the geographic region served by that laboratory. It should be noted that drugs that show geographic variability may change over time, requiring reevaluation by the clinical and laboratory staff.
At this time, the Committee does not recommend regular testing of urine for benzodiazepines until the problems with immunoassays, as discussed in Part 2D, are resolved.
It is also important to note that the list of tests shown in Tables 2
and 3
are analytes the laboratory should make available on a stat basis, allowing the causative agent to be identified when a patient presents with signs and symptoms suggesting exposure to one or more of these drugs. Tables 2
and 3
do not imply an ED panel of tests that should be ordered on all patients. The ED physician and toxicologist must decide on the most appropriate plan of action based on the clinical presentation on a case-by-case basis.
b. assay turnaround times for tier i tests
Introduction.
The recommended turnaround time (TAT) for reporting of assay results is a consistent theme among the previous NACB Laboratory Medicine Guidelines because it is readily quantifiable and is an area where most laboratories can find improvement. Among the factors important for achieving acceptable TATs for laboratory tests are the availability of resources, cooperation with the house staff, efficient specimen delivery systems, reductions in the complexity and number of steps in the preanalytical processes, effective laboratory and hospital information systems, and the establishment of a priority list of analytes considered critical to patient care.
Recommendation.
The ideal TAT for Tier 1 toxicology tests is 1 h or less except where noted on Table 3
. Degree of consensus: B
Discussion.
Reporting TATs for laboratory tests continues to be a major issue for all stat testing, not just for drugs of abuse. The overwhelming majority of participants were in favor of the 1-h TAT. The major issue, however, is the appropriate definition of reporting TATs. Some felt that this should be defined from the time of specimen receipt within the laboratory to the availability of results, either by phone or via electronic reporting to the record. Others felt that the TAT should be defined from the time the test is ordered to the reporting time, although laboratory personnel may not have the responsibility of specimen collection and delivery to the laboratory. Nevertheless, proponents of this definition argue that the laboratory should take on this role or at least have some influence on this practice. An important part of meeting this expanded definition is the availability of a rapid specimen delivery system to the laboratory, i.e., a pneumatic tube or use of point-of-care (POC) testing within the ED. Without these conveniences, it is unlikely that clinical laboratories can meet the 1-h TAT as defined from the time of ordering and that the 1-h TAT might be better applied to the time of specimen receipt. Although toxicology data may impact on the efficiency of ED triaging decisions, most acute management decisions are based on the patient’s vital signs and mental status, irrespective of what the laboratory results shows and when they are received.
A continuing issue that was raised in several sessions is the potential legal impact for institutions that are unable to meet the 1-h TAT. These guidelines were developed as goals in hope of providing justification for hospitals and laboratories to improve their testing services. The Committee reemphasizes that these recommendations are not standards of care and may not even be a consensus of the current practice; they, however, reflect ideal goals for both laboratory practice and incentives for the in vitro diagnostics industries.
c. tier ii testing: comprehensive or broad-spectrum testing
Introduction.
Stat testing is adequate to support ED evaluation of acute toxicity for the specific toxins listed in Tables 2
and 3
for which a specific therapy or antidote may be available. A second tier of more complicated and time-consuming tests is recommended for patients with continuing medical problems from toxicologic exposure to drugs and chemicals not identified in Tier I. General screening for a broad spectrum of toxins is generally not necessary for patients who are asymptomatic or clinically improving in the ED. In the event that a patient presents with or develops coma or other clinical signs that cannot be explained by one of the Tier 1 toxins, further evaluation by a trained clinical toxicologist is indicated. These evaluations usually occur outside the ED setting. The availability of tests for these toxins and their recommended TATs are different from those for Tier I tests. The laboratory should be advised if there is a need for a broader panel of tests.
Recommentation.
The second tier of drug tests is for patients admitted to the hospital who remain intoxicated, obtunded, or comatose, where a broad-spectrum ("comprehensive") screening panel is necessary to cover drugs and substances that may have clinical significance and would not be identified based on the findings of the first tier of laboratory tests. Results of these tests might be used for more long-term management and/or counseling of patients. Laboratorians should work closely with intensive care providers to determine the appropriate menu of tests and TATs that are necessary. Degree of consensus: B
Recommendation.
Testing for toxins beyond those outlined in Tables 2
and 3
should be performed only after the patient is stabilized and the attending physician has received toxicology input from a poison control center or, preferably, bedside evaluation by personnel trained in medical toxicology. Degree of consensus: A
Discussion.
A two-tiered mechanism allows a laboratory to concentrate the majority of its resources on providing for the needs of the sick ED patient. The Committee has excluded certain drugs from the first tier of testing because they do not have significant toxicities or they are difficult to measure on a stat basis. These include the phenothiazines, calcium channel blockers, beta blockers, hypnotics and tranquilizers (e.g., chloral hydrate, ethchlorvynol, and glutethimide), some anticholinergic drugs (e.g., atropine), muscle relaxants (e.g., carisoprodol and cyclobenzaprine), some antidepressants (e.g., fluoxetine), behavioral drugs (e.g., clonidine and methylphenidate), date rape drugs (GHB and flunitrazepam), certain anesthetics (e.g., ketamine), and analgesics (fentanyl and analogs). Some of these drugs will be detected on the broad-spectrum tier of testing.
The issue of GHB testing was specifically raised during one of the sessions, i.e., "if a commercial immunoassay were available would there be requests for this test?" The majority of the AACC meeting participants thought that a test for GHB would not be used at their institutions. In general, the clinical presentation and course of this agent are characteristic and shorter than the usual assay TAT.
Methodologies for a comprehensive urine drug screen profile include thin-layer chromatography, liquid chromatography with full-scan spectrophotometric detection, gas chromatography (GC) with mass spectrophotometry (MS), and liquid chromatography–tandem MS. Hospitals that do not have adequate resources to perform a broad-spectrum screening panel can elect to send these specimens to a reference laboratory or regional toxicology laboratory (see Part IV-K). In this situation, laboratory personnel should ascertain the methodologies used by the reference laboratory, including the expected TATs and assay limitations, and communicate this information to the attending physician because this may have an impact on the interpretation of results.
d. selectivity of testing
Introduction.
It is not appropriate for the clinical laboratory to provide test results for all classes of drugs simply because an automated and inexpensive immunoassay is available. Certain tests may not be indicated because the substances they measure do not contribute to significant toxicologic sequelae or their prevalence in that particular geographic location is very low. There is the additional problem of diagnostic inaccuracy if these tests are included in a general drug-screening panel, particularly with tests for phencyclidine (PCP) and TCAs, because existing immunoassays exhibit significant cross-reactivity toward other drugs, e.g., dextromethorphan, diphenhydramine, and sertraline for PCP (9) and phenothiazines, cyclobenzaprine, and diphenhydramine for TCAs (10). Because of the low prevalence of TCA and PCP abuse and the high prevalence of diphenhydramine use in the general population, these tests often have low positive predictive values.
Recommendation.
Stat testing for the following drugs is not recommended for ED patients presenting with acute symptoms: tetrahydrocannabinol (THC; marijuana), LSD, methaqualone, ibuprofen, and cotinine (nicotine metabolite). Testing for some other drugs, such as amphetamines, PCP, and propoxyphene, should be conducted in areas where these drugs exhibit notable prevalence. Degree of consensus: B
Discussion.
The prevalence of methaqualone abuse is very low in the US today (11). The number of ED visits resulting from PCP use exhibit marked geographic variation, but the total number is relatively low (12). Although THC and LSD are more widely abused, many clinical toxicologists do not want or need to know whether a patient is positive for these drugs because they do not contribute to major acute clinical problems (13). THC testing may be useful in drug compliance and rehabilitation programs that are outside the usual objectives of ED testing. Of course, testing for some of these agents (e.g., PCP and THC) may well be indicated in the evaluation of a child with altered mental status, where intoxication may be difficult to discern by history alone.
e. drug panels by "toxidromes"
Introduction.
Patients who are drug intoxicated or overdosed often present to the ED with a collection of physical findings and symptoms that are consistent with a particular drug or class of agents. Recognition of these "toxidromes" can be important in the effective clinical management of ED patients (14). Proper identification of a particular toxidrome could be used to exclude some drug classes as the cause of the symptoms without urine drug testing. Drug testing panels can be established that link specific symptoms to a particular menu of tests: e.g., sympathetics (cocaine and amphetamines), sedatives (benzodiazepines, tranquilizers, and barbiturates), and hallucinogenic agents (THC, LSD, and PCP). Although implementation of such an approach could reduce unnecessary utilization of laboratory tests, the opportunity to identify the causative agent could be missed if the initial clinical impressions were wrong.
Recommendation.
Clinical laboratories should not set up specific drug testing panels based on toxidromes. The failure to recognize a particular toxidrome may lead to the failure to order an important drug test. Degree of consensus: A
Discussion.
Although clinical laboratories are under tremendous pressure to reduce costs and utilization of laboratory services, in cases in which urine has been collected and sent to the laboratory, elimination of a few drug tests that are available (i.e., regularly calibrated and quality controlled) as part of the test menus on automated testing platforms will not greatly impact the cost of delivering laboratory services. On the other hand, a delay in the triage and management of the overdosed patient because of inappropriate laboratory orders can greatly affect the cost for treating that patient and may have an adverse effect on patient outcomes. There has been at least one study that examined the potential success of linking toxidromes to particular ED drug-screening patterns (15). When ED nurses, clinical pharmacists, and medical residents were asked to choose among eight toxidromes, the diagnostic accuracy was 79–88%, with the medical residents scoring the lowest of the group. Although these results indicate a reasonable degree of performance, the critical question is the clinical and fiscal impact for the 12–21% of patients incorrectly diagnosed, if the toxicologic causes of these cases were not identified. Inaccuracies in the assessment of toxidromes may be attributable to the presence of polydrug overdoses, delayed-onset toxicities (16), or clinical inexperience. The importance of clinical experience in the ED is a major factor in the success of toxidrome accuracy and the potential use of specific drug panels. In a study of periodicity of drug overdose presentations, Raymond et al. (17) concluded that EDs are most likely to encounter overdosed patients in the early evening. This is also a time of peak activity in the ED, when problems with resident supervision, thoroughness of evaluation, and delays in patient management may be critical.
f. gastric samples
Introduction.
Gastric contents can be sent to the laboratory for identification of orally ingested drugs. Gastric lavage is used on rare occasions to remove unabsorbed toxicsubstances. Administration of activated charcoal has become the treatment of choice for decontamination of toxins and poisons, with decreased utilization of gastric lavage (18)(19).
Recommendation.
There is no role for the testing of gastric contents in clinical management, although premortem collection and specimen retention may be important for cases with medico-legal considerations. Degree of consensus: A
Discussion.
Although the analysis of gastric contents for drugs provides verification of what was consumed, unabsorbed substances do not contribute to the clinical presentation, and test results would not supercede toxicology data obtained on blood and urine. ED physicians must provide supportive care on the basis of existing signs and symptoms in the event that this information can be useful in analyses for forensic purposes, such as postmortem investigations.
g. "chain-of-custody" for clinical specimens
Introduction.
Chain-of-custody documentation is a basic tenet of forensic and workplace drug testing. When a specimen is handled under chain-of-custody conditions, each individual who handles the specimen must sign a form that indicates when that specimen was in that individual’s possession and when it was transferred to the next individual involved with the processing of that specimen. If the sample is to be stored for any reason, it must be in a secure and locked location, with limited access by only qualified personnel.
Recommendation.
Maintenance of chain-of-custody documentation is unnecessary for samples collected for clinical toxicology purposes, and such practice should be discouraged. As with any laboratory specimen, proper procedures for collection, transport, results reporting, and storage are necessary. Degree of consensus: A
Discussion.
In contrast to workplace drug testing, the principal aim of drug testing for hospitalized patients should be for diagnostic and treatment purposes. It therefore is unnecessary for the ED and the clinical laboratory to maintain a chain of custody for all urine specimens that are tested for drugs of abuse. Although results of laboratory tests may be introduced into court proceedings, this is an insufficient reason to require such documentation. The process is time-consuming, burdensome, and expensive; does not contribute to and may delay patient care; and should be discouraged. If it is known in advance that a specimen will likely be involved in a medico-legal matter, chain-of-custody procedures may be warranted and the ED staff should seek the assistance of qualified members of the laboratory staff.
|
Part II. Recommendations on Analytical and Reporting Issues for Drugs-of-Abuse Testing by Immunoassays |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
Recommendation.
Optimum use of urine drug testing assays for ED patients requires an understanding of the limitations of existing commercial immunoassays for drugs of abuse. A close relationship between the clinical laboratory and ED staffs is necessary. The laboratory should clearly communicate to the ED staff the extent of the toxicology services available to them, such as the menu, target TATs, cross-reactivity data, and contact information for consultations. Degree of consensus: A
Discussion.
The Committee feels that it is the joint responsibility of the ED and clinical laboratories to provide initial and continuing education on the limitations of drug testing to new house officers and to maintain a continuous medical education program for existing practitioners. It is also the responsibility of the in vitro diagnostics industry and, in some cases, the Food and Drug Administration (FDA) to alert the laboratory when changes to assays are made that could affect the performance and interpretation of urine drug testing results. It is clear that this knowledge is not adequately taught in medical schools, and it is inappropriate to expect that senior residents or attending staff will have sufficient understanding of these limitations to effectively educate their junior house staff. The laboratory must inform the ED staff when there are changes concerning the specificity and performance of commercial drugs-of-abuse immunoassays. They should also discuss the availability of new drug assays or assay platforms (e.g., POC testing devices) and the appropriateness of implementing such assays in that institution. On the other hand, ED physicians must inform the laboratory staff of changing drug utilization patterns; the appearance of new drugs or analogs, such as designer drugs (21)(22); or testing and reporting needs that are not currently being met.
b. listing of cross-reacting substances on immunoassays
Introduction.
As described in the previous section, current immunoassays have considerable specificity limitations with respect to other compounds that are not members of the particular drug class being tested. In addition to direct educational activities, the laboratory should also document these specificity limitations when reporting results.
Recommendation.
When immunoassays are used, the laboratory should list the major cross-reacting substances for each drug class when a positive result is reported. It may also be appropriate to indicate in a final report (e.g., in the "notes" section) that a negative urine drug result does not indicate absence of all drugs of abuse. Degree of consensus: A
Discussion.
The issue of cross-reactivity for immunoassays is further complicated by the fact that there is heterogeneity in methodologies and formulations among assay manufacturers. The laboratory is directed to the package insert for specific cross-reactivity data and any changes in antibody cross-reactivity that might have occurred because of differences between reagent lots. The laboratory should compile an abbreviated list of major cross-reacting substances and make them available to the ED staff. Laboratory personnel should also be aware of additional data reported in the literature after the production of the assay and package insert or on new drugs marketed after the initial cross-reactivity testing was conducted. For example, oxaprozin, which was approved by the FDA in 1993, unexpectedly produced interference in most of the commercial immunoassays for benzodiazepines (23). Validation studies are necessary if the laboratory modifies these assays to suit its needs (e.g., diluting the reagent to reduce costs). When in vitro interference studies are conducted, inclusion of drug metabolites is equally important to testing of parent compounds. Because it may be difficult to obtain supplies of the important metabolites, the use of urine from patients given the drugs to be tested is an alternative.
Given the importance of cross-reactivities to clinical practice, it was suggested during the open presentation of these guidelines that data be made available by manufacturers on their internet sites. It may be useful for clinical laboratories to note the manufacturer of the specific drug assay used in the final report, so that users of this information might know where to search for cross-reactivity data.
Another issue that was discussed in open sessions was how cross-reactivity data should be generated and listed in manufacturers’ package inserts and on their web sites. Tests of potential interferents should be conducted at multiple concentrations because the cross-reactivity effects do not produce a parallel response. The concentrations tested should bracket the concentrations seen in routine therapeutic use and under toxic conditions. Studies of potentially interfering compounds should be conducted in both the presence and absence of the target analyte because the interferent can suppress the signal of the analyte, producing false-negative results (24). The combining of different interferents within the same sample, as a means to reduce the amount of work necessary to generate interference data, should not be done because of potential drug–drug interactions. The data should be expressed as a percentage of interference. Methods for calculating cross-reactivities for immunoassays have been published previously (25). After consultation with these investigators, the Committee recommends the use of the 50% displacement method for homogeneous nonisotopic competitive immunoassays.
Given the heterogeneity of approaches toward conducting and presenting cross-reactivity data, the Committee feels that a standardized approach would be in the best interests of the industry. This issue extends beyond the toxicology community and branches into all aspects of immunoassay testing; thus, it is beyond the scope of this work. It is unlikely, however, that the in vitro diagnostics companies will organize a committee with such an agenda. Establishing guidelines for cross-reactivity documentation could be a topic for a future AACC or NACB committee.
c. immunoassay cutoffs
Introduction.
Qualitative assays for drugs-of-abuse testing require cutoff concentrations to distinguish between positive and negative results. Cutoffs are set on the basis of signal-to-noise ratio considerations, which are dependent on the precision, analytical sensitivity, and specificity of the assay. To reduce the number of false-positive results, the cutoff is set at a concentration that is higher than the assay limit of detection. As a consequence, there are urine samples that contain the target drug that are reported as negative because they are below the "administrative" cutoff concentration. The precision of automated immunoassay analyzers enables the use of lower cutoff concentrations without sacrificing specificity.
Recommendation.
Cutoff concentrations optimized for workplace drug testing are not necessarily appropriate for clinical toxicology. Although a true-positive result indicates use, it does not presume impairment or intoxication of the patient at the time of specimen collection. Degree of consensus: A.
Discussion.
The Substance Abuse and Mental Health Services Administration recently raised the workplace drug testing opiate cutoff from 300 to 2000 µg/L to reduce the number of opiate-positive results attributable to poppy seed consumption (26). This reduced the frequency of false-positive indications of drug abuse. Although raising the opiate cutoff concentration may be appropriate for workplace testing, for clinical toxicology, the lower opiate cutoff concentration may be more desirable because the objective is to determine whether any opiates are present that may contribute to the clinical presentation or suggest the need for substance abuse counseling. The cutoff concentrations for other drugs should also be reviewed for appropriateness. For example, the cutoff for cocaine metabolites is set at 300 µg/L for workplace drug testing. Because of the cardiotoxic effects of cocaine, a lower cutoff, e.g., 100 µg/L, may be appropriate for ED patients and will increase the reported prevalence of cocaine use (27). Lowering the cutoff concentration will also increase the number of cases of incidental drug-positive findings, i.e., those that do not contribute to the clinical symptoms of the patient. Manufacturers can assist laboratorians in designing or modifying assays that enable the use of more ideal drug cutoff concentrations.
d. inadequate spectrum of benzodiazepine detection by immunoassays
Introduction.
Traditionally, antibodies used in immunoassays for benzodiazepines were directed against either the parent compound or an unconjugated form of a metabolite (such as oxazepam). For many benzodiazepines, however, this is inappropriate because the parent compound is not excreted into urine in high concentrations and the metabolites are in the conjugated forms. Furthermore, since the development of these assays, many additional benzodiazepines have been approved for use that do not metabolize to oxazepam. If the degree of cross-reactivity is low, these drugs have the potential to produce a false-negative result for the benzodiazepine class, even at higher concentrations than are typically seen in an overdose.
Recommendation.
Some immunoassays for testing benzodiazepines are inadequate. Antibodies in optimum assays should be targeted toward the parent compound and principal conjugated metabolites or should utilize an online hydrolysis procedure to convert the conjugated metabolites to the unconjugated forms. Degree of consensus: A
Recommendation.
Antibodies for benzodiazepines should be updated to identify the newer drugs in this class as they become approved and available for clinical use. Degree of consensus: A
Discussion.
Immunoassays that are not sensitive to conjugated metabolites of all benzodiazepines on the market will produce false-negative results. Many investigators have shown that this problem can be overcome by treating the sample with a ß-glucuronidase before immunoassay screening (28). Although this step improves the usefulness of the assays, it is time-consuming and not practical for emergency (stat) testing. Some manufacturers have reformulated their benzodiazepine assays to incorporate an online hydrolysis step (29)(30). Others have directed their antibodies toward conjugated metabolites (31). The Committee feels that either of these approaches substantially improves the detection of benzodiazepines in urine. The Committee recognizes that updating immunoassays to include new drugs will be costly and time-consuming. One participant felt that lowering the benzodiazepine cutoff concentration to 50 µg/L in urine or adapting urine assays for use in serum or plasma was useful in improving the sensitivities of the assays.
e. opiate vs opioid detection by immunoassay
Introduction.
The immunoassay for opiates is a source of much confusion because there is an expectation by many physicians that this assay will detect any opioid compound. Most commercial immunoassays, however, are directed toward free morphine and have various degrees of cross-reactivity toward codeine, 6-monoacetylmorphine, oxycodone, and hydromorphone and conjugated metabolites of these drugs. Current assays do not detect any of the synthetic opioids.
Recommendation.
Immunoassays should detect most opioids (e.g., oxycodone, hydromorphone, meperidine, tramadol, buprenorphine, propoxyphene, and pentacozine) and not just codeine and morphine. Degree of consensus: B
Discussion.
The opioid class of drugs can contribute to significant toxicities and clinical problems; however, a urine drug screen for "opiates" will usually produce a falsely negative result. Separate immunoassays are available for some of these opioids. The assay for methadone as an independent test is justified because of its specific use for the large number of methadone clinics worldwide. Individual assays for the other opioids may not be financially justified because of the lower prevalence of abuse of these agents. ELISAs for hydromorphone and other semisynthetic opiates are available for veterinary and dog- and horse-racing drug-testing laboratories (32) but are not adaptable to automated chemistry analyzers. Thus, the development of a "cocktailed" assay may be warranted in which a mixture of antibodies is added to detect the presence of these opioids. It is the opinion of the Committee that if such an assay were developed, it would be useful in ED settings. From the acute care viewpoint, however, the clinical response to an appropriate dosage and duration of therapy with naloxone in suspected opioid cases is a sufficient "diagnostic test", obviating the need for a positive urine test. In addition, problems with the specificity of the opiate assay have been identified, with cross-reactivities with such entities as the quinolone antibiotics (33).
f. immunoassays for amphetamines vs sympathomimetic drug class
Introduction.
The term "amphetamines" has been inappropriately applied to a family of amines that have stimulant and sympathomimetic properties. Drugs in the former category include D-isomers of amphetamine, methamphetamine, phentermine, and the designer amines, methylenedioxyamphetamine and methylenedioxymethamphetamine. They are used as appetite suppressants and are abused as recreational drugs (34). The sympathomimetic amines are present in nonprescription cold medications, such as decongestants, and in diet pills. Some of these include ephedrine, pseudoephedrine, phenylpropanolamine (recently removed from the US market), and phenylephrine. These over-the-counter sympathomimetic amines are abused and can produce significant toxicity. For workplace drug testing, highly selective immunoassays have been developed that use monoclonal antibodies targeted toward detection of the illicit amphetamine and methamphetamine (35). Other immunoassays that use polyclonal antibodies are also available that are able to detect both illicit and sympathomimetic amines (36). Tests for all of these drugs may be important in the ED evaluation of a patient with agitation or "sympathetic" toxidrome.
Recommendation.
The optimum immunoassays to test for amphetamines in ED patients are those directed toward a broad spectrum of amines as a class, rather than assays that are directed specifically toward the illicit amines. An assay directed toward phenylethyl amines would largely cover this class. The name of the test should be changed from "amphetamines" to "sympathomimetic amines" or "stimulant amines". Degree of consensus: A
Discussion.
Some manufacturers of immunoassays offer two amphetamine assays. In this case, the laboratory should select the more nonspecific sympathomimetic amine assay for ED practice and reserve the monoclonal amphetamine assay for specific detection of the illicit drugs. For manufacturers who offer only the monoclonal assay for the illicit amphetamines, laboratory personnel should communicate the sensitivity limitations of this assay to the ED staff. Manufacturers are urged to provide a broad-spectrum "amine" assay for testing on automated chemistry analyzers should such an assay not be available.
g. confirmation of positive immunoassays
Introduction.
A basic tenet for forensic drug testing analysis is the use of two analytical methods that differ from one another in the basic chemical principles (37)(38). GC/MS is the definitive and defensible method for analysis for drugs of abuse. Given the difficulties and expense of performing GC/MS in real time, the need for obtaining stat results negates the value of confirmation analysis in ED cases. If the clinician anticipates subsequent involvement with medico-legal or social services or there is a clinical need to identify the specific drug yielding a positive immunoassay result, the staff should notify the laboratory of the need for a confirmative analysis.
Recommendation.
When reporting results of immunoassay screening, there must be proper notation given that the assay used is considered as a "screening test" and that any positive results are to be considered as "presumptive". Degree of consensus: A
Recommendation.
The laboratory should not routinely perform confirmative analyses on positive screening results. Degree of consensus: A
Recommendation.
When confirmation is needed, the laboratory should store these specimens for an indefinite period or until the case is resolved. The laboratory should consult with the hospital’s risk management department for further guidance. Degree of consensus: B
Discussion.
The standards for forensic toxicology are different from those for clinical toxicology, and the results of unconfirmed urine drug testing should be used only as a single data point to assist patient management decisions. The costs of providing GC/MS analysis on a stat basis are prohibitive for all but the most specialized of academic centers. The TAT for typical GC/MS confirmation is long (>4 h) and likely to be unacceptable for real-time use by ED physicians. Nevertheless, laboratory tests can be entered into court proceedings, and there can be inappropriate interpretations made because of the inaccuracies of immunoassays. Confirmation may be necessary if there is anticipated involvement of medico-legal or social services. Because these situations will be unknown to the laboratory, appropriate communication is required from the clinical service. In the absence of confirmative testing, the limitations of immunoassays and their effect on the interpretation of drug testing results will need to be presented and argued in court proceedings. As a compromise, it may be useful for a laboratory to store positive urine drug screen results for a period of time, e.g., 3 months to 1 year, to enable confirmative testing of challenged cases at a later date. The ED staff should be notified as to the policy for toxicology specimen retention. This may not be ideal, however, because some cases do not surface for many years after an ED episode, and adequate storage space is likely to always be an issue.
|
Part III. Recommendations for Specific Analysis of Ethyl Alcohol and Other Toxic Alcohols |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
Although the whole-blood alcohol concentration is important from a law enforcement viewpoint, e.g., the determination of driving under the influence, forensic cutoffs (e.g., 0.08% or 0.1%) have no relevance from a clinical management viewpoint because there is no concentration that consistently defines clinical intoxication. This justifies the use of samples other than whole blood, including serum or plasma, saliva, and breath. Although there are subtle differences in results among these specimens, the magnitude of these differences, i.e., 10–20%, are also without clinical significance. Therefore, a laboratory or ED has choices regarding the optimum specimens and testing methods that best meet its clinical needs. As such, many hospitals have implemented alternative samples for alcohol testing because of their low cost and convenience.
|
a. need for a breath alcohol quality-assurance/quality-control program |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
Recommendation.
Clinical breath alcohol testing is POC testing and must meet the same quality-assurance (QA)/quality-control (QC) requirements as any POC test. As a part of the laboratory’s ongoing QA effort, a program must be in place to monitor and evaluate policy, protocols, and the total testing process so that breath alcohol results are accurate and reliable. The clinical laboratory should be involved in the design, implementation, and monitoring of the quality assurance program. Degree of consensus: A
Discussion.
Elements of an effective QA program include monitoring and evaluating the overall quality of the total testing process (preanalytic, analytic, and postanalytic steps) as well as the evaluation of effectiveness of policies and procedures; identification and correction of problems; assurance of accurate, reliable, and prompt reporting of test results; and affirmation of the competency of operators. It is necessary to have a comprehensive up-to-date accessible Standard Operating Procedure manual, operator training and evaluation of competency, and a QC program. Each device must be checked for accuracy each day by use of a dry gas standard and an air blank. The recovery of alcohol must be within the tolerance established by the manufacturer. The essential operator procedures are listed in Table 4
.
|
|
b. selection and validations of breath alcohol devices |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
Recommendation.
The laboratory should be involved in the selection, validation, and deployment of the breath alcohol devices used. Degree of consensus: A
Discussion.
The selection of breath alcohol devices should be based on performance and features that meet the requirements of clinical services. Only devices listed in the National Highway Traffic Safety Administration Conforming Product List should be used (40). Table 5
lists the Committee’s recommendation for specific device specifications and desirable attributes for breath alcohol measurements in the ED.
|
The clinical laboratory has the responsibility to validate that the device performance meets or exceeds specifications before release of the technology for clinical use at near-patient sites. This includes familiarization with the technology by use of vendor-supplied educational materials or programs and adjustment of manufacturer’s guidance to clinical service requirements. Verification of device performance characteristics requires use of a National Highway Traffic Safety Administrationapproved breath alcohol simulator and certified alcohol solutions or certified dry gas alcohol standards. The analytical performance should be evaluated against experimental protocols established by the NCCLS (41). The accuracy and precision studies should be performed at clinically relevant alcohol concentrations. The specificity should be challenged with aqueous solutions of volatiles (acetone, methanol, isopropanol) at concentrations likely to be encountered in clinical settings. The calibration stability should be verified with suitable QC materials on each day that the device is used.
|
c. reporting units for ethyl alcohol |
|---|
|
TopAbstractPreambleIntroduction and Needs...Part I. General Principles...Part II. Recommendations on...Part III. Recommendations for...a. need for a...b. selection and validations...c. reporting units for...d. assays for methanol...e. osmolality measurements for...f. isopropyl alcohol propylene...Part IV. Recommendations on...References |
|---|
86% (with a normal hematocrit). Therefore, whole-blood alcohol concentrations are lower than serum or plasma values. However, a constant conversion factor cannot be applied because the hematocrit can dramatically change from individual to individual. It should be noted that these legal definitions have little or no clinical meaning in the ED.
Recommendation.
Alcohol concentrations should be reported in units clearly defined by the laboratory, with a notation as to the sample matrix that was tested (serum or plasma, urine, whole blood, breath) and methodology. Degree of consensus: A
Discussion.
A wide variety of technologies are available to quantify alcohol in biological fluids. The laboratory, with advice from the ED, should clearly identify the type of technology used, the specimen of choice, and the reporting units. The vast majority of clinical alcohol assays are based on the alcohol dehydrogenase enzymatic assay of serum or plasma. An absolute conversion of serum or plasma alcohol concentration to whole-blood alcohol concentration should not be made.
There was some discussion on the applicable reporting units for breath alcohol. It is most scientifically correct to express the breath concentration per 210 L of expired air. Use of this unit will also make it more obvious that a breath sample was tested. Most physicians, however, are unfamiliar with the subtle differences between mg/210 L (breath) and mg/dL (blood) and are likely to remember only the actual number. The Committee has decided to keep the mg/dL designation because these are the units used in law enforcement applications where breath alcohol testing is most frequently used. Some countries report in units of mmol/L. These units can be directly used in osmolality calculations.
Other types of samples, such as saliva and sweat, have been used for workplace drug testing applications. For clinical toxicology, the Committee felt that there is insufficient experience or peer-reviewed evidence in a clinical setting to render a recommendation regarding these matrices at this time.
|
d. assays for methanol and ethylene glycol |
|---|
|
TopAbstractPreamble |
|---|
