Urine Specimen Dilution: Assessment and Policy Recommendations Dr. Leo Kadehjian uly 21, 2003 The following comments and recommendations are for informational use only. Consultation with program counsel is highly recommended before implementing any of these policy suggestions.
Urine dilution by excess water consumption One of drug users most effective resources in their attempts to thwart detection through drug testing is dilution of their urine specimens through excess fluid consumption. By consumption of excess water (or other fluids) over a short period of time the urine quickly becomes highly dilute, thereby significantly lowering the concentration of any detectable drugs and/or drug metabolites eliminated in the urine. For example, after rapid consumption of 1 liter of fluids shortly before a drug test, urine drug/metabolite concentrations can be reduced several-fold and remain diluted for a few hours. This may allow a drug user to escape detection by reducing the concentration of any drugs/metabolites present in the specimen to levels below the screening assay cut-offs. Although the consumption of excess fluids to avoid detection is commonly called flushing, this term is misleading. The excess fluid consumption and increased urine production rates are not causing a more rapid elimination of drugs/metabolites from the body. Rather, the same amount of drug and metabolites are being filtered through the kidneys, but they now are being sent to the bladder with a much greater amount of water, and so diluting the measured concentration. Dilution should not be confused with adulteration, where chemical adulterants are directly added to a urine specimen. Many products intended for oral consumption and claiming to help rid the body of toxins are sold over the Internet. Although these body cleansing products may claim to rid the body of toxins (i.e. help beat the drug tests), they appear to be effective only because of the large amounts of water the user is instructed to consume along with the teas or powders. Consumption of excess fluids is the most effective way to dramatically increase urine production rates and produce dilute urine specimens. Diuretics, prescribed for those with high blood pressure, increase urine production rates to only a few ml/min compared to a typical 1 ml/min urine production rate. The same is true for alcohol, with only a few ml/min increase in urine production rate. However excess water consumption of 1 2 liters over a short period of time can lead to dramatic increases in urine production rates to 10 or even 20 ml/min as demonstrated in the charts below.
Urine Production Rate After Water Loading Urine production rate (ml/min) 10 9 8 7 6 5 4 3 2 1 Drink 1 L. water 0-60 -30 0 30 60 90 120 150 180 210 Time (min) Baldes and Smirk, 1934 Urine Production Rate After Water Loading Urine production rate (ml/min) 18 16 14 12 10 8 6 4 2 Drink 2 L. water 0 0 50 100 150 200 250 300 Time (min) Macallum and Benson, 1909 Although a drug user may be able to easily dilute their urine by a factor of 5 or more, for many drugs the urine concentrations may be so high that even a 5 10 fold dilution will not allow a drug user to reduce their urine levels below the established cut-offs. For cocaine users, typical urine levels of cocaine metabolite are on the order of many tens of thousands of ng/ml
and even higher, so even a 10-fold dilution will still leave urine levels well above the screening cut-off values (300 ng/ml of cocaine metabolite). cf. the chart below. 20000 Dilution: Cocaine 15000 ng/ml 10000 5000 Dilute x 8 Cut-off 0 0 8 16 24 32 40 48 Hr. The same can be true of opiates and amphetamines, simply because of the large doses of these drugs which are commonly consumed by abusers and the very high resulting urine concentrations. However, after the body has had some time after dosing to metabolize and eliminate these drugs/metabolites and their urine levels are reduced to only a few thousand ng/ml, then dilution may allow a user even of large amounts of drug to escape detection. In contrast, for cannabinoid consumption, the urine levels are typically on the order of only a few hundred ng/ml and accordingly a 5 10 fold dilution of urine may easily allow a user of cannabis to avoid detection with the conventional 50 ng/ml screening cut-off. cf. the chart below.
ng/ml Cut-off 400 350 300 250 200 150 100 50 B Dilute x 8 Dilution: THC B 0 0 8 16 24 32 40 48 Hr. B B Measuring urine dilution There are several methods which can be used to identify and measure the extent of dilution of a urine specimen. Clearly diligent collector observation of the nature of the specimen is important. For objective scientific assessments, the two most common tests are for creatinine and specific gravity. Although some have argued for their preferences between these two, both have been recognized as appropriate markers. However creatinine is easily and accurately measured on automated analyzers and is the analyte to be measured first in assessing dilution in federally-regulated workplace drug testing programs. Creatinine Creatinine is a metabolic by-product formed primarily from the breakdown of protein within the body and eliminated in the urine. For a given individual about the same total amount of creatinine is eliminated each day and so its concentration in urine can be used a marker for the extent of urine dilution. Typical urine creatinine concentrations are on the order of 150 mg/dl. Current federally-regulated workplace drug testing programs have established 20 mg/dl as a cut-off for indicating a dilute specimen. Again, this cut-off reflects an approximately 8-fold dilution from typical levels. This level of dilution does
not occur with normal water consumption (e.g. after an 8 oz. (240 ml) glass of water) but rather after consumption of 1 liter (~4 glasses) of water over a short period of time (i.e. 30 minutes). Creatinine may be roughly measured on-site using specially designed urine dipsticks. More accurate measurements can be made on typical laboratory automated analyzers. Creatinine is to be distinguished from creatine, which is used as a dietary supplement by body builders. The difference between creatine and creatinine is a molecule of water. The urine tests for creatinine do not cross-react with creatine. However, oral loading with significant amounts of creatine (e.g. 5 10 g/day) can increase urine levels of creatinine, but not significantly so. N NH H 2 O N NH O O H NH 2 + H 2 O O NH Creati ne Creatinine Specific gravity Specific gravity measures the density of a urine specimen relative to that of pure water (which by definition has a specific gravity of 1.000). It can be easily and reliably measured on-site using a hand-held refractometer, which provides an indication of the specific gravity by measuring the refractive index of the urine specimen. Urine specimens have specific gravities greater than water because of the many dissolved substances and are typically around 1.025, with pure
water having a specific gravity of 1.000 by definition. Current federally-regulated workplace drug testing programs have established 1.003 as the lower cut-off for specific gravity. This cut-off reflects approximately an 8-fold dilution from typical levels. Typical: Creatinine ~150 mg/dl Specific Gravity ~1.025 Dilute x 8 Cut-off: 20 mg/dl 1.003 Assessing the significance of dilution in a population. Before implementing any policy addressing dilution it should first be determined if in fact dilution is a significant problem. This can best be determined by objectively assessing the level of dilution in all specimens collected over say a one month period. Automated instrumentation can easily and inexpensively measure creatinine levels in each specimen as part of the routine drug testing. After testing all specimens for creatinine it will also be important to group specimens according to their immunoassay test results. I recommend the following three groupings: 1. Specimens which test positive
2. Specimens which have immunoassay rates equivalent to that of a drug-free control 3. Specimens which are not positive yet clearly elevated in rate above the rate for a drug-free control. It is this last group that is of most interest. These specimens are demonstrating some level of immunoreactivity consistent with the presence of drug/metabolite albeit at a concentration below the cut-off. The presence of drug at levels below the cut-off could be due to two possibilities. Either the drug use was far enough in the past such that the urine drug levels are naturally below the cut-off as a result of the normal elimination process. Alternatively, the immunoreactivity and drug concentrations are artificially low because of dilution. There could have been recent drug use which would have been identified, but because the specimens are highly dilute, the drug concentration has been artificially lowered below the cut-off level and accordingly allows the user to avoid detection. Note that for those specimens which test positive, drug use has been detected whether the specimens are dilute or not. Urine levels of drugs and metabolites can be so high that even with significant dilution they will still test positive. The extent of dilution in each group should be determined, but further analysis must be performed on the group demonstrating elevated but below cut-off immunoassay levels. It must be determined what proportion of this group is thwarting the testing process through dilution and thereby escaping detection. By adjusting the immunoassay levels for the extent of dilution one can determine whether a specimen would have otherwise tested positive. For example, consider a specimen which has an immunoassay rate equivalent to a cannabinoid level of 20 ng/ml and a creatinine of 15 mg/dl. If using the standard 50 ng/ml cannabinoid cutoff, the specimen would be reported as negative. But this specimen is clearly highly dilute given the very low creatinine. Given a typical creatinine level of 150 mg/dl this specimen could be considered to be about 10-fold dilute. Thus the measured cannabinoid level of 20 ng/ml would more likely have been 10 times that or 200 ng/ml if the specimen had not been so dilute, and accordingly would have been reported as positive were it not for the dilution. To be conservative, it may be more appropriate to use 100 mg/dl as the typical creatinine level, so that this specimen would then be considered only 5-
fold dilute. Even in that case the measured 20 ng/ml cannabinoid level would likely have been 100 ng/ml had the specimen not been dilute. Of course it would be most appropriate to adjust the level according to the donor s own normal creatinine level, but this figure may not be known. Thus to be conservative, I recommend using 100 mg/dl as the creatinine level to use for adjustment in these calculations. After performing this calculation for each of the elevated but below cut-off specimens, the extent that dilution is thwarting effective detection of drug use can be determined. If only a small number of these specimens are shown to be dilute and would otherwise have been positive then it may not be of sufficient magnitude to justify development and implementation of policy responses. However if the number of users escaping detection through dilution are significant then it clearly makes sense to address dilution with a sound documented policy with appropriate responses for dilution. # Subjects 100 90 80 80% Drug-free Cut-off 70 60 10% Positive Double detection rate 50 If 5% highly dilute 40 30 20 15% Non-zero immunoreactive equivalents 5% Positive 10 0-10 0 10 20 30 40 50 60 70 80 90 100 Immunoreactive equivalents, ng/ml
In the chart above, a hypothetical distribution is shown. Here most of the specimens (80%) are at the level of the drug-free control. Another 5% of the specimens have tested positive whether dilute or not. However 15% of the specimens demonstrate some level of immunoreactivity, albeit below the cut-off. They are reported as negative but there is evidence of drug presence. If 1/3 of these below cut-off specimens are highly dilute and would have been positive by controlling for dilution, then the detection rate would increase from 5% of the population to 10% of the population thereby doubling the detection rate with little additional effort or expenses other than testing for dilution and have a policy response for dilution. Policy responses Once dilution has been identified as a significant issue undermining the effectiveness of a drug testing program, there are several areas where dilution can be thwarted with appropriate policy responses. These include development of an appropriate written policy, addressing the unacceptability of dilution when subjects first enter a drug testing program, providing for assessment and documentation of dilution at the time of collection, objective laboratory assessment of dilution, and policy responses to unacceptably dilute specimens. Advice of rights and responsibilities When donors are first instructed about their rights and responsibilities in participating in drug testing, they should be put on notice that highly dilute specimens are unacceptable. I recommend wording such as: You have been ordered to participate in a urine drug testing program. You will be expected to provide a fresh, clean, unadulterated, undiluted specimen of at least 30 ml. You will be allowed a minimum amount of time and minimal access to fluids in which to provide a specimen. Any failure to provide an adequate specimen, absent a documented medical condition which precludes your ability to provide a valid specimen, will be considered a failure to comply with the requirements of the
testing program, will be reported to and may include sanctions up to and including. Each participant should be asked to read and sign acknowledging their understanding of these requirements for compliance. Of course, donors are not expected to be kidney physiologists and accordingly know what level of fluid consumption will generate an unacceptably dilute urine. Accordingly I do not believe that donors should be sanctioned for their first highly dilute specimen. Chain of custody form The chain of custody form used in the collection of each specimen should have a check box for collectors to easily note if the specimen appears unusually dilute. Collectors need to be trained to understand and recognize dilution, and to be diligent in making the appropriate notation on the chain of custody form. Specimens arriving in the laboratory which are clearly dilute and yet have no such notation on the chain of custody form should be called to the attention of the collector. Of course the laboratory should note any specimens that they receive which appear or are measured to be dilute. Furthermore the availability of simple dipsticks or a hand-held refractometer can allow an objective assessment of dilution at the time of collection. Collection procedures Note that donors with normal kidney function can generate urine at a rate of about 1 ml/min. Accordingly collectors should not need to wait for several hours to obtain an adequate specimen. Current federally regulated workplace programs allow for 40 oz of water spread evenly over 3 hours. I believe that these allowances are overly generous. Typical urine production rates during the day are on the order of 1 ml/min. Thus donors with normal kidney function should be able to generate a 30 ml specimen in as little as 30 min. Thus allowing 1 hour should be sufficient. Of course there may be donors with a documented medical condition which requires special allowance, but these conditions must be formally documented by a physician. Furthermore I believe that there is no need to monitor or control fluid intake during the collection process as donors
should have been put on notice that any excess fluid consumed may result in an unacceptably dilute specimen. Criteria for unacceptably dilute specimens I recommend using the current 20 mg/dl creatinine cut-off level for dilution established by the federally-regulated workplace drug testing programs. Although those programs do not currently have any effective sanctions for a dilute specimen that is still within the realm of human kidney function, other testing programs are not required to follow those policy responses. In the DHHS and DOT federally-regulated urine drug testing programs, both creatinine and specific gravity must be below their respective cut-offs before a specimen is considered unacceptably dilute. However the Nuclear Regulatory Commission (NRC) drug testing program specifies that either of these markers is sufficient to indicate a specimen as dilute. I agree with the NRC that either creatinine or specific gravity is a suitable marker for dilution. Responses to unacceptably dilute specimens Providing sanctions for a highly dilute specimen is an area of controversy. Under federally-regulated workplace testing programs, 20 mg/dl is used as a cut-off for reporting a specimen as highly dilute, but there are effectively no sanctions for having a urine creatinine below this level, as long as it is not below 2 mg/dl, which is the level considered not consistent with human urine. This lower level for creatinine indicating a specimen not consistent with human urine was 5 mg/dl but some challenges to that cut-off has the DOT to lower this cut-off to 2 mg/dl. However the 20 mg/dl remains the cut-off for reporting a specimen as dilute. Under the DOT program, if a specimen is shown to be highly dilute, the only sanction imposed is that the next time a specimen is demanded of that donor, it may be collected under direct observation. Unfortunately, this response does not effectively address the ability of a drug user to again provide a dilute specimen! The NRC has proposed that dilute specimens be subjected to testing at a reduced cut-off, but this has only
been published as a Proposed Rule and has not been established as a Final Rule. One proposal oft cited to counteract such efforts at urine dilution is to adjust urine drug/metabolite levels for the extent of dilution by normalizing to the urine creatinine level. This is already done in occupational health when testing urine for heavy metal and other toxic chemical exposure, as well as for other clinical measurements. Thus there is ample precedent for adjusting urine concentration measurements for the level of dilution. However I do not recommend making such dilution adjustments to immunoassay levels. Rather I recommend that there simply be sanctions for a highly dilute urine, whether there is drug present or not. An approach like that proposed by the NRC could also be implemented where dilute specimens are tested at a lower cutoff, or even at the limit of detection of the assay. In its Fitness-for-Duty drug testing program, the Nuclear Regulatory Commission has proposed re-testing dilute specimens at a reduced cut-off or even at the limit of detection of the drug assays. However caution should be exercised whenever different donor specimens are treated differently, ensuring that there is a sound and non-discriminatory basis for doing so. I recommend that the provision of a highly dilute specimen (creatinine <20 mg/dl) be considered unacceptable and a failure to comply with the program conditions and should be met with appropriate sanctions. Of course excessively dilute specimens should nonetheless still be forwarded to the laboratory for testing as they may still test positive. I recommend that all persons subject to drug testing should be informed at the outset that they will be expected to provide a fresh, clean, unadulterated, and undiluted specimen; that they will be provided a limited amount of time and limited access to fluids in which to provide an adequate specimen; and that any failure to provide an adequate specimen absent a documented medical condition will be considered a failure to comply with the conditions of the testing program, with appropriate sanctions, depending upon the testing context. To conservatively allow for the possibility that a donor simply does not understand their responsibilities, kidney physiology and dilution, and may have innocently consumed excess fluids prior to donating a specimen, perhaps the first time a donor submits a dilute specimen they may be reminded of the policy regarding excessively dilute specimens being unacceptable. Since
modern immunoassay instrumentation can provide a quantitative readout, the laboratory can indicate if the specimen is consistent with a drug-free specimen or rather demonstrates an elevated rate above the range for drug-free specimens, albeit below the assay cut-off for reporting a result as positive. This information can be provided to the donor indicating that the specimen was unacceptably dilute and that the testing demonstrated the presence of immunoreactive material (drug/metabolites) but insufficient to be called a positive. Furthermore the donor can be told that had the specimen not been dilute it may have likely registered a positive test result. Although I recommend against sanctioning a donor for a below cut-off test result, the donor can be put on notice that dilution can be detected as well as drug use below the cut-off. The donor should be instructed that a second highly dilute specimen will be considered a failure to comply with the program requirements whether there is any evidence of drug use or not. A recurrence would then trigger appropriate sanctions.