血清半乳甘露聚糖检测对血液病患者侵袭性曲霉菌感染早期诊断的价值

目的 评价血清半乳甘露聚糖(GM)检测对血液病患者侵袭性曲霉菌(IA)感染早期诊断的价值.方法 前瞻性采用酶联免疫吸附试验(ELISA)每周2次测定患者的GM水平,并计算该诊断试验的各项评价指标.结果 共有来自92例患者的113例次感染进入研究,检测血清标本472份,以确诊IA感染和临床诊断IA感染为真阳性组,排除IA感染为真阴性组,0.7为阈值,连续2次GM结果阳性为真阳性,该诊断试验的敏感度为83.3%,特异度为91.1%,阳性预测值为78.9%,阴性预测值为93.1%.GM阳性结果较痰培养阳性时间提前4 d(1～7 d),比影像学改变提早7 d(1～14 d),较抗真菌治疗提前6 d(1～15 d).结论 通过ELISA方法进行血清GM检测可以快速、灵敏的为早期诊断IA感染提供有力证据.     

False-positive results in premature infants with the Platelia Aspergillus sandwich enzyme-linked immunosorbent assay

The sensitivity of a sandwich enzyme-linked immunosorbent assay (ELISA) for detecting Aspergillus galactomannan was tested using 783 serum samples obtained from 247 patients (1-15 sera per patient) with severe underlying diseases (haematological malignancies or intensive care unit stay). We selected 146 serum samples from 50 patients for retesting. Serum samples were frozen after routine testing at -18 degrees C until retesting. All patients charts were checked for signs of Aspergillus infection, such as pneumonia or sinusitis. Adult patients were divided into four groups: proven (5), probable (6), suspected (8) or unlikely (25) Aspergillus infection. The results of Platelia ELISA were 100% in proven, 33% in probable and 50% in suspected Aspergillus infection. Patients with unlikely infection had no positive results with Platelia ELISA. Group 5 consists of six paediatric patients with prolonged ICU stay and a birth weight of 400-1320 g. In five out of six infants we found positive results with Platelia ELISA. All positive results in this group of patients are considered as false positive (83.3%).     

Perflubron as a gastrointestinal MR imaging contrast agent in the pediatric population

A microplate enzyme immunoassay (EIA) for the detection of lysergic acid diethylamide (LSD) in human urine was developed. The assay kit is designed around an LSD derivative coated on the wall of microplate wells with preservatives and stabilizers. Sample and rabbit anti-LSD are added to the microplate well. The immobilized LSD and LSD present in specimens compete for the opportunity to bind to the anti-LSD antibodies. An anti-rabbit antibody labeled with horseradish peroxidase is used to provide the assay signal, which is inversely proportional to the concentration of LSD in the sample. The assay requires a 25-microL urine sample and three consecutive incubation periods of 60, 30, and 30 min at room temperature. The assay was tested with a variety of drugs, including ergot alkaloids spiked into drug-free urine at up to 100,000 ng/mL without cross-reaction. Nor-LSD was shown to cross-react between 16% and 28%, depending on its concentration. Of the other compounds tested, only ergonovine demonstrated slight cross-reactivity at approximately 0.0008%. The assay is designed to be used with a qualitative cutoff of 0.5 ng/mL. Precision testing at 0.5 ng/mL gave a coefficient of variation (CV) of 6% based on 20 replicates. The CV at 0.375 ng/mL (cutoff, -25%) was 5.2% and at 0.625 ng/mL was 6.6%. Precision at other concentrations within the range of the calibration curve gave similar results both intra- and interassay. Clinical performance of the assay was compared with that of a commercial radioimmunoassay (RIA). Comparable performance was observed with both methods, each screening a total of 458 samples as negative and 17 samples as positive relative to a 0.5 ng/mL cutoff. The EIA found an additional three positive samples that were negative by RIA. The EIA is suitable for the screening of urine samples for the presence of LSD. Preliminary indications are that the assay is also suitable for use with whole blood specimens. The assay can be performed manually or be fully automated and without the need for radioactivity; it can be used in any laboratory.     

Quantitative evaluation of pork adulteration in raw ground beef by radial immunodiffusion and enzyme-linked immunosorbent assay

Quantitative estimates are important to establish whether pork adulteration in ground beef is accidental or intentional. A standard agar gel radial immunodiffusion (RID) test using forensic-grade antiserum to porcine albumin and an enzyme-linked immunosorbent assay (ELISA) using forensic-grade anti-porcine glycoprotein immunoglobulin were used to determine from 1 to 75% raw pork in raw ground beef. The RID test, which incorporated 1.5% anti-pork serum in 1% immunodiffusion agar, formed precipitin rings with pork albumin in agar wells. A linear standard curve was obtained by plotting the diffusion area against standard pork concentrations ranging from 0 to 80%. For the ELISA the endpoint optical density increased linearly versus log % pork between 0.0625% and 2% pork. In spiked samples, the RID test had a detection limit of 3 to 5%, a coefficient of variation (CV) of 22%, and a recovery of 105%. The ELISA had a detection limit of 1%, a CV of 18%, and a recovery of 114%. The mean recovery from the spiked samples by the ELISA and RID test was not significantly different (P > 0.05) from the known sample amounts. Quantitation by RID of 28 ground beef samples (27 of which were DTEK ELISA-positive for pork adulteration) revealed a wide range of pork content, with values as high as 48%.     

Comparison of an enzyme immunoassay and a latex agglutination system for the diagnosis of invasive aspergillosis in bone marrow transplant recipients

The performance of two Aspergillus antigenemia systems, the sandwich enzyme-linked immunosorbent assay (ELISA), Platelia Aspergillus test, and the latex agglutination (LA), Pastorex Aspergillus test, in the diagnosis of invasive aspergillosis were compared by testing 364 serum samples from 22 bone marrow transplant (BMT) recipients. Sensitivity and specificity for the ELISA test were 60% and 82% respectively, vs 40% and 94% for the LA test. In the two patients found positive with both methods, the ELISA test became positive earlier than the LA test or remained positive after the LA test had become negative. These results encourage further evaluation of the Platelia Aspergillus test, to assess its role in the management of invasive aspergillosis in BMT patients.     

Interlaboratory study evaluating quantitation of antibodies to Haemophilus influenzae type b polysaccharide by enzyme-linked immunosorbent assay

An interlaboratory study was conducted to determine whether an enzyme-linked immunosorbent assay (ELISA) with an antigen preparation composed of various-sized fragments of Haemophilus influenzae type b polysaccharide conjugated to human serum albumin could be standardized across laboratories and whether the ELISA-derived results from different laboratories are equivalent to those obtained by the standard radioactive antigen binding assay (RABA) for quantitation of anti-H, influenzae type b polysaccharide antibodies. Twenty coded human serum samples were quantitated by ELISA in 11 laboratories and by RABA in 5 laboratories. The mean RABA-derived values served as the basis for all comparisons. While the overall correspondence of antibody values between the two methods was good, significant differences were found among some of the 11 ELISA data sets and among the mean RABA values. Seven laboratories generated higher ELISA antibody values for low-titered sera. Four laboratories generated antibody concentrations that were not statistically different between the two assay methods. The results therefore indicate that the ELISA can tolerate substantial variations in protocol, such as the use of different plates and different antibody reagents, without affecting the quantitation of serum antibodies. However, attention should be focused on low-titered sera, as some assay conditions may yield spurious results. This ELISA is a serologic assay which can serve as an alternative to the RABA for quantitation of antibodies to H. influenzae type h polysaccharide.     

Calibration of ion-exchange HPLC measurements of glycohemoglobin: effect on interassay precision

To investigate the effect of calibration with lyophilized calibrators on the interassay precision of glycohemoglobin (glyHb) measurements, we used an ion-exchange HPLC system equipped with a Pharmacia Mono S HR 5/5 column. Calibration of analytical runs substantially increased interassay variation (CV), from 1.7% to 4.4% and from 0.9% to 3.2% for control samples with low (6.5%) and high (14%) glyHb percentages, respectively. Standardization of glyHb results, though essential for interlaboratory comparisons, should not be done at the expense of assay precision, as may occur with thoughtless use of lyophilized calibrators. We therefore recommend the use of carefully determined conversion factors for standardization of glyHb results obtained with ion-exchange HPLC systems that are capable of excellent long-term interassay precision.     

ACP Broadsheet No 152: March 1998. Clinical implications of plasma homocysteine measurement in cardiovascular disease

Enzyme-linked immunosorbent assay (ELISA) and agar gel immunodiffusion (AGID) are the most widely used serological tests for Maedi-Visna diagnostics. The purpose of the present study was to develop an indirect whole virus ELISA and an immunodiffusion test and compare their sensitivity. A total of 747 ovine serum specimens were analysed for antibodies against this ovine lentivirus. The number of positive results in the ELISA was 430 (57.56%). In the AGID test, a positive result was found in 380 samples (50.87%). In the group of discordant results 78 (10.4%) samples tested positive by the ELISA and negative by the AGID test and 28 sera (3.7%) were found to be positive by the AGID test and negative by the ELISA. The data in this report show the ELISA to be more sensitive than the AGID test, but accurate serological diagnostics should be based on a combination of the two tests.     

Liquid chromatographic determination of tilmicosin in swine feed at 200-400 mg/kg level: interlaboratory study

An analytical method for the determination of tilmicosin at 200-400 mg/kg, the intended use concentration range, was evaluated in an interlaboratory study involving 5 laboratories, including the sponsor. The interlaboratory study evaluated the intra- and interlaboratory precision and accuracy of a tilmicosin feed method. The method procedure involved extracting tilmicosin from feed by adding 200 mL extractant to 20 g feed and shaking for 1 h. The extract is filtered and analyzed by gradient liquid chromatography which separates tilmicosin from feed matrix in 30 min. Each laboratory assayed 5 replicates of fortified feed at concentrations of 0, 100, 200, 400, and 600 mg/kg. The mean recovery among fortified samples ranged from 81.4 to 98.8%, with a percent coefficient of variation (%CV) ranging from 0.3 to 4.0%. For all blank control feed samples no significant interferences were observed. In addition, each laboratory assayed 5 replicates of medicated feed samples prepared at 2 levels (200 and 400 mg/kg) with either a horizontal or vertical mixer. Along with the medicated feed samples were included 5 replicates of a blank control feed. The identities of the medicated and blank control feed samples were blinded to the analysts. The results for the medicated feed samples ranged from 95.8 to 106% of label claim, with a %CV ranging from 2.1 to 6.7%.     

Evaluation of a commercial enzyme-linked immunosorbent assay for detection of immunoglobulin M antibody in diagnosis of human leptospiral infection

The PanBio Leptospira immunoglobulin M (IgM) enzyme-linked immunosorbent assay (ELISA) is a commercially available screening test for the diagnosis of acute leptospiral infection. The ability of the test to diagnose early or recent Leptospira interrogans infection was assessed by testing sera with known microagglutination test (MAT) titers to serovars pomona, hardjo, copenhageni, and australis. The IgM ELISA detected all 41 cases of early or recent leptospiral infection (sensitivity, 100%), with a positive ELISA result seen in many cases before MAT antibody titers reached 1:50. Thirty-eight of 41 patients showed seroconversion (fourfold or greater increase in titer by MAT, 2 of 41 patients had a single sample with elevated titer, and 1 patient from whom leptospires were isolated from a blood sample failed to show MAT titers, despite a seroconversion (negative to positive result) in the ELISA. Follow-up sera obtained from 8 of 12 patients (67%) for 3 to 48 months after the acute stage of illness showed persisting IgM antibody. However, the range of levels detected in these samples (maximum ELISA ratio, 2.0) was lower than the range seen when infection was recent. Reactivity in the IgM ELISA was observed for only 1 of 59 serum samples from asymptomatic donors (specificity, 98%) and 16 of 233 serum samples from patients with Ross River virus, brucella, Epstein-Barr virus, cytomegalovirus, mycoplasma, Q-fever, toxoplasma, hepatitis A virus, Treponema pallidum, or Borrelia burgdorferi infection (specificity, 93%), with the majority of these patients showing lower levels of IgM in comparison to those in patients with leptospiral infection. We conclude that this ELISA is sufficiently sensitive for use as an initial screen for leptospiral infections, with subsequent confirmation of positive test results by MAT.     

Glycine antagonism does not block ischemic spontaneous depolarization in the rat

Two enzyme-linked immunoassay (ELISA) systems for rapid screening of Listeria spp. were compared for their use in analysis of spiked foods regulated by the U.S. Food and Drug Administration. The Tecra Listeria kit is a 48 h visual ELISA that detects Listeria spp. through colorimetry. It has been approved for first action by AOAC INTERNATIONAL. The Vitek immunodiagnostic assay system for Listeria (VIDAS LIS) is a fully automated 48 h ELISA that detects Listeria spp. by immunofluorescence. Fifty-two food samples were artificially contaminated with high (11-42 colony-forming units [cfu]/25 g food) and low (2-8 cfu/25 g food) levels of L. monocytogenes and screened by the 2 protocols. Unspiked samples were also assayed as negative controls. Six unspiked samples were found positive for Listeria spp. by both methods: 3 were identified as L. monocytogenes and 3 as L. innocua by official methods. Both ELISA methods detected all spiked samples. One unspiked sample was assayed positive by Tecra and negative by VIDAS LIS. No Listeria spp. were recovered when the sample was tested by the conventional method. No interference due to background fluorescence of food matrixes was observed in the VIDAS LIS method. Results suggest a modified VIDAS LIS preenrichment medium may be used in place of the VIDAS standard medium in the protocol.     

Evaluation of PCR for detection of DNA specific for Aspergillus species in sera of patients with various forms of pulmonary aspergillosis

Pulmonary aspergillosis is classified into invasive, saprophytic, and allergic forms. In this study, we evaluated the usefulness of PCR for differentiating between different forms of aspergillosis or in monitoring disease activity during treatment by detecting DNA specific for Aspergillus species in the serum. Nested PCR was used to detect Aspergillus DNA in the sera of 30 patients with various forms of pulmonary aspergillosis. The results were compared with those of latex agglutination tests for detecting galactomannan antigen. We also examined the serial changes in the results of nested PCR during and after treatment of a subgroup of patients with invasive pulmonary aspergillosis with amphotericin B. The highest proportion of positive nested PCR results were in patients with invasive aspergillosis (10 of 12; 83%), while patients with pulmonary aspergilloma had the lowest frequency of positive tests (1 of 9; 11%). These results suggested that the sensitivity of the nested PCR depends on the extent of invasion by Aspergillus species. Serial assays showed that the results of nested PCR became negative shortly after commencement of antifungal treatment and that such changes did not correlate with clinical responsiveness to treatment. Our results indicate the potential usefulness of nested PCR with serum samples for the diagnosis of invasive aspergillosis and the detection of a shift in the status of infection from a noninvasive type to invasive aspergillosis. However, the results of the nested PCR did not correlate with the response to antifungal treatment.     

Direct determination of benzoylecgonine in serum by EMIT d.a.u. cocaine metabolite immunoassay

An EMIT d.a.u. immunoassay for urine testing was applied on the Syva ETS Plus analyzer for the detection of the cocaine metabolite, benzoylecgonine (BE), in human serum. Serum was analyzed without prior extraction, concentration, or matrix modification. Calibrators and serum controls were prepared from EMIT d.a.u. calibrators that were reconstituted and diluted with EMIT Tox serum calibrator. The assay cutoff concentration for BE was 50 ng/mL. The within-run and between-run precisions of the assay were both less than 5%. Analysis of 162 patient serums yielded 43 BE positive results. All EMIT positive serum BE results were confirmed by gas chromatography-mass spectrometry. All patients with positive BE serums also had BE positive urine samples. Serum bilirubin and triglycerides as high as 38 mg/dL and 319 mg/dL, respectively, did not interfere with the assay. Modification of the EMIT urine assay allowed for a simple, rapid, and reliable method for the detection of BE in serum.     

Human T-cell leukemia virus type II infection frequently goes undetected in contemporary US blood donors

Serologic screening for human T-cell leukemia virus type I (HTLV-I) infection was begun in US blood banks with the licensure of enzyme-linked immunosorbent assays (ELISA) in December 1988. We examined the donation histories of the first 60 Western blot (WB)-confirmed HTLV-I/II positive donors to one blood center and found 8 had made 16 previous donations that scored negative on the screening ELISA. All 16 donations had ELISA absorbance below the cutoff for a positive assay, but still well above that of the average donation (17.6% +/- 5.7% of the cutoff). In a more extensive study, 17 donations from a total of 61,752 at six blood centers were both ELISA-positive and WB-positive for HTLV-I (4) or HTLV-II (13), and 218 samples had ELISA absorbance greater than 50% of the ELISA cutoff. One hundred seventy-eight of the 218 were tested further by WB and 11 were found positive. All 11 positives were confirmed by polymerase chain reaction; 10 had HTLV-II and 1 had HTLV-I. Thus, the HTLV-I-based screening ELISA missed at least 10 of 23, or 43% (95% confidence interval, 23% to 66%), of HTLV-II infections, compared with 1 of 5, or 20%, of HTLV-I infections.     

Diagnosis of paratuberculosis in dairy cattle, using enzyme-linked immunosorbent assay for detection of antibodies against Mycobacterium paratuberculosis in milk

An ELISA containing lipoarabinomannan (LAM) antigen was used to detect antibodies in milk and serum for diagnosis of Mycobacterium paratuberculosis infection in dairy cattle. In experiment 1, milk and serum samples were obtained from 25 cows, and subjected to LAM ELISA testing immediately, and after 1 year of storage at -70 C. Milk samples, with and without a commonly used chemical preservative, were tested. There was no significant difference in LAM ELISA results between fresh and frozen samples or between preserved and unpreserved milk samples. In experiment 2, milk samples were collected daily from 30 cows over a 14-day period. The day-to-day coefficient of variation was 0.19 for milk LAM ELISA and was 0.15 for serum LAM ELISA, with no statistically significant time effect detected. In experiment 3, single milk, serum, and fecal samples were obtained from 764 cows. The fecal samples were cultured for M paratuberculosis to identify infected cows, and the serum and milk samples were subjected to LAM ELISA testing. Results were compared, using the area under the receiver operating characteristic curves. The milk LAM ELISA had specificity (+/- 95% confidence limits) of 87 +/- 8.1% when the cutoff was set at 50% sensitivity, and specificity of 83 +/- 9.1% when sensitivity was set at 60%. The area under the receiver operating characteristic curve was 0.85 +/- 0.03 for the milk ELISA and 0.75 +/- 0.02 for the serum ELISA. In this population of cattle, the milk LAM ELISA had comparable accuracy to serum LAM ELISA, although the milk LAM ELISA was slightly less reproducible (higher coefficient of variation).     

Amphetamines in hair by enzyme-linked immunosorbent assay

Human hair was collected from the occipital crown region of the head from several subjects; these hair samples were presumptively positive for amphetamines by a previously evaluated immunoassay. Hair was washed briefly with methanol to remove external contamination, then extracted with hot methanol for 2 h to recover the drugs. The extracts were evaporated to dryness, reconstituted in buffer, and analyzed using a new enzyme-linked immunosorbent assay (ELISA) technique adapted for the detection of amphetamines in hair. Gas chromatography-mass spectrometry was used as the reference technique. Cross-reactivity of several related compounds was evaluated by equating the inverse of the ligand concentration at 50% antibody binding to the affinity constant for each compound. The ratio of a compound's affinity constant to that for d-methamphetamine was used to derive percent crossreactivity. These experiments yielded values of 30.8% for d-amphetamine, 7.4% for I-methamphetamine, 4.3% for phentermine, 2.9% for I-amphetamine, and <1% for ephedrine, methylenedioxyamphetamine, and methylenedioxymethamphetamine. Cross-reactivity of unrelated compounds was found to be non-existent. The optimum cutoff concentration was determined by receiver operating characteristic curve analysis to be 300 pg/mg and the observed limit of detection was 60 pg/mg. Intra-assay precision at 300 pg/mg was 3.3% (coefficient of variation, CV), and the interassay CV was 10.5%. The sensitivity and specificity of the method were 83% and 92%, respectively.     

Assessment of corticosteroid-induced alkaline phosphatase isoenzyme as a screening test for hyperadrenocorticism in dogs

Quantitative determination of the corticosteroid-induced isoenzyme of alkaline phosphatase (CAP) was evaluated as a screening test for hyperadrenocorticism (HAC) in dogs. A series of 40 dogs with HAC (CAP range, 96 to 14,872 U/L), 30 clinically normal dogs (CAP range, 0 to 38 U/L), and 80 dogs with various diseases (non-HAC) and without history of exogenous glucocorticoid exposure for a minimum of 60 days (CAP range, 0 to 1163 U/L) were used to evaluate the test. Sensitivity and specificity of CAP was calculated at various cutoff points for absolute CAP activity and for CAP activity expressed as a percentage of total alkaline phosphatase activity. A cutoff point of 90 U/L was selected as optimal for use of this assay as a screening test for HAC. A prevalence survey then was done of all canine serum samples submitted to our diagnostic laboratory over a 3-month period, to calculate the predictive values of a positive and a negative test result in a clinical population and to determine the relative frequency and magnitude of CAP activity in dogs that had received glucocorticoids. The predictive values of a positive and a negative test result at the 90 U/L cutoff value were 21.43% (95% confidence limits, 8.3 to 40.95%) and 100% (95% confidence limit > 96%), respectively. It was concluded that CAP isoenzyme activity, determined by routine biochemical analysis by an automated levamisole-inhibition assay, could function as a screening test for HAC; however, the predictive value of a positive test result was too low to recommend the assay as a diagnostic test.(ABSTRACT TRUNCATED AT 250 WORDS)     

The use of whole blood absorbed on filter paper to detect Wuchereria bancrofti circulating antigen

The Og4C3 enzyme-linked immunosorbent assay (ELISA) to detect circulating Wuchereria bancrofti antigen uses 50 microL of serum. In this study, a whole blood sample absorbed on filter paper was tested as a substitute for serum. Serum samples were obtained from 60 Sri Lankan subjects by venepuncture and finger-prick blood samples from the same individuals were directly absorbed on filter paper. Og4C3 ELISAs using serum and filter paper blood were compared. Despite the fact that the estimated amount of serum available for the ELISA with filter paper blood was only one-fifth of that available when serum was used, the 2 ELISAs gave almost identical results. Of the 39 positive serum samples, 38 were detected using filter paper blood. Employing the ELISA using filter paper blood, 619 people in Matara, Sri Lanka, were examined for antigenaemia. The positivity rate was 22.5%, 3.1 times higher than the rate of microfilaraemia detected by examination of 60 microL blood films.     

Serological association between spirorchidiasis, herpesvirus infection, and fibropapillomatosis in green turtles from Florida

Serodiagnostic tests for detecting green turtle (Chelonia mydas) antibody responses were developed to test the strength of association between exposure to spirorchid trematode antigens or herpesvirus antigens and having green turtle fibropapillomatosis (GTFP). Plasma samples from 46 captive-reared green turtles, including paired pre- and 1-yr post-inoculation samples from 12 turtles with experimentally induced GTFP, were found by enzyme-linked immunosorbent assay (ELISA) to be negative for antibodies to adult spirorchid (Learedius learedi) antigens. In contrast, all 12 turtles that developed experimentally induced GTFP converted within 1 yr from having negative to positive antibody reactivity to GTFP-associated herpesvirus antigens, whereas the three controls and four turtles that failed to develop tumors remained negative. Plasma samples from 104 free-ranging green turtles from two Florida (USA) coastal feeding grounds with different GTFP prevalences were tested by ELISA for antibodies to L. learedi adult antigens; and there was no statistically significiant association between antibody prevalence and sampling site. When a low optical density cutoff value (0.15) was used to interpret ELISA results, 98% of the turtles from each site were spirorchid antibody-positive and there was no association between antibody reactivity to spirorchids and GTFP status. When a higher negative cutoff value was used, however, a statistically significant association between antibody reactivity to spirorchids and GTFP-free status was found. These results suggest that spirorchids do not have a role in GTFP pathogenesis. All 20 of the tumor-bearing lagoon turtles had antibodies to herpesvirus antigens whereas only two (10%) of the tumor-free reef turtles had detectable anti-herpesvirus reactivity. The strong association between antibody reactivity to herpesvirus antigens and GTFP status in both captive-reared and free-ranging turtles is consistent with the hypothesis that the transmissible agent that causes GTFP is a herpesvirus.     

Comparison of a radioallergosorbent (RAST) inhibition method and a monoclonal enzyme linked immunosorbent assay (ELISA) for aeroallergen measurement

BACKGROUND: Mouse and rat urinary proteins are potent occupational allergens for exposed personnel. Methods of measuring airborne allergens differ greatly, and reported levels of allergens vary considerably between laboratories. OBJECTIVES: To compare the values obtained using two different methods of allergen detection. METHODS: Air samples were collected in rat rooms in Sweden and the United Kingdom at 2 L/min on to polytetrafluoroethylene (PTFE) filters and extracted in buffer containing 0.5% v/v Tween 20. Airborne rat urinary allergen (RUA) was measured in all samples by both RAST inhibition using a polyclonal human serum pool (UK) and a two monoclonal antibody sandwich ELISA employing antibodies specific for Rat n 1.02 (alpha2u-globulin) (Sweden). RESULTS: The two methods gave values which were correlated (r2 log values = 0.72, P<0.0001), but differed by several orders of magnitude (median [range] ratio of RAST inhibition/ELISA = 316 [7-26(80)]. There was a systematic bias: as the absolute values increased, the difference in the measurements increased. The rat urine standards used were antigenically similar. CONCLUSIONS: A large contrast in RUA values obtained from the two assays was observed in this study. This may be primarily due to methodological differences, but variations in antibody specificities or composition of allergenic epitopes in the air samples may contribute. The results demonstrate that standardization of methods and antibodies is necessary before interlaboratory comparisons can be made.