Effect of hemoglobin- and perflubron-based oxygen carriers on common clinical laboratory tests.
Clinical laboratory assays play an important role in the care of many peri- or postoperative patients and trauma victims and will similarly be required for patients receiving blood substitutes. However, both hemolysis, because of the strong optical absorbances of Hb species between 500 and 600 nm, and lipemia, because of light scattering, are well known to cause interference in many colorimetric and spectrophotometric methods used in clinical laboratories [11-13]. After HBOC administration to patients, there is a dose-related presence in plasma of soluble Hb and a marked red coloration of plasma. Plasma Hb values determined by CO-oximetry can be as great as 50 g/L in these patients, well above the concentrations of Hb described as interfering in many laboratory assays [11-13]. Current Perflubron emulsion dosing concentrations (3.0-4.5 mL/kg) result in a dilution of -1:20-1:25 of Perflubron in blood, and plasma samples from these patients can have a lipemic appearance. Thus, it is important for clinical laboratories to determine which tests are valid when performed with samples from patients receiving these blood substitutes.
Materials and Methods
Instrumentation. Biochemical tests were performed with the Hitachi 747 (Boehringer Mannheim Diagnostics), the Vitros 750 (Johnson & Johnson Clinical Diagnostic Systems), the Dade Stratus, and the Abbott IMx and AxSym analyzers, according to the manufacturers' instructions. Coagulation tests were performed with the MLA 1000 C (Medical Laboratory Automation, Pleasantville, NY) and the BBL Fibro System (Becton Dickinson Microbiology Systems). Total plasma Hb was determined with the Radiometer OSM.3.
Patient materials. Human plasma pools were prepared from excess nonhemolyzed plasma samples received for routine biochemical analysis by the clinical chemistry laboratory at Barnes-Jewish Hospital or were provided by Biopure. This study, as part of ongoing clinical trials of oxygen carriers, was approved by the Washington University Human Studies Committee.
Sample preparation. A stock bovine HBOC preparation containing HBOC, 130 g/L, and diluent containing no HBOC were provided by Biopure. The highest possible HBOC (as measured plasma Hb) concentration in patients receiving HBOC in clinical trials at our institution was <50 g/L. Samples for the interference studies were prepared by adding to 6.15 mL of a plasma pool 3.85 mL of the 130 g/L stock HBOC (final concentration of HBOC, 50 g/L) or 3.85 mL of the HBOC diluent (final HBOC, 0 g/L). Thus, any constituent in the HBOC diluent was equally included in all samples. Mixtures of these two preparations in 10% increments were prepared and deviations from the value of the plasma pool containing no HBOC were determined.
Interference studies with Perflubron emulsion were performed by adding 250 [micro]L of 600 g/L Perflubron emulsion (AFO144; Alliance Pharmaceutical) or isotonic saline (control) to 4750 [micro]L of a plasma pool and comparing deviations of the Perflubron-containing plasma pool from the control. This 1:20 dilution of Perflubron is ~1.3 times the maximum amount of Perflubron emulsion expected in the plasma of patients in current clinical trials at our institution, who receive a dose of 1.8 g/kg (3 mL/kg); i.e., a 70-kg patient receives ~210 mL of Perflubron emulsion, resulting in ~1:25 dilution of Perflubron in plasma.
Table 1 depicts the values for the plasma pools containing diluent and no soluble HBOC and lists the concentration of HBOC (as plasma Hb) that produced a significant interference in chemistry procedures on the Vitros 750 and Hitachi 747 analyzers. Fig. 1 shows examples of the interference patterns observed for several tests from each of these chemistry analyzers. In general, the HBOC concentration invalidating an assay value at our institution was the concentration resulting in a positive or negative interference for enzymes and proteins >10% and >5% for electrolytes. For instance, on both analyzers we established HBOC concentration cutoffs that produced <10% interference for [Ca.sup.2+] assays, because we consider any interference that alters [Ca.sup.2+] values by >5 mg/L to be clinically significant.
Table 2 depicts immunoassays that are unaffected by HBOC concentrations [less than or equal to] 50 g/L, and Table 3 describes immunoassays that are affected. HBOC interference in coagulation testing was method dependent. Prothrombin time (PT), activated partial thromboplastin time (aPTT), and fibrinogen values from the BBL Fibro-System fibrometer, which utilizes electrical clot detection, were unaffected by HBOC concentrations [less than or equal to] 50 g/L. In contrast, PT, aPTT, and fibrinogen assays from the MLA 1000C analyzer, which detects clot formation optically, produced a "no clot detected" error flag at 20 g/L HBOC. Finally, manual latex agglutination procedures for D-dimer and fibrin degradation products were unaffected at HBOC [less than or equal to] 50 g/L.
The 1:20 dilution of Perflubron emulsion in plasma had a slightly lipemic appearance and produced a "lipemic index" of 35 on the Hitachi 747--lower than the index value of 50 we use as the minimum value to flag a sample as "lipemic" at our institution. Thus it is not surprising that none of the assays performed on the Vitros or Hitachi analyzers shown in Table 4 exhibited any interference as a result of this concentration of Perflubron emulsion. The 1:20 dilution of Perflubron did produce a negative interference of 30-50 U/L in the amylase assay and a positive interference of 20-30 mmol/L in the ammonia assay on the Vitros 750 analyzer, and a negative interference of 20-30 mg/L was observed on the Hitachi 747 iron method. An increase of 71 g/L phosphorus was observed on both analyzers; however, the Perflubron diluent contains a phosphate buffer. The Perflubron emulsion-containing sample exhibited no interference in thyroxine or thyroid-stimulating hormone assays on the AxSym, nor in the creatine kinase (CK) MB isoenzyme, troponin I, or human chorionic gonadotropin assays on the Stratus analyzer (Table 4). Thus, at current dosing values (1.8-2.7 g/kg), Perflubron emulsion does not appear to present any significant problems to the clinical chemistry laboratory.
The most important observation from these studies is that neither bovine Hb-based HBOC at plasma Hb concentrations as high as 50 g/L nor Perflubron emulsion at current maximum dosing rates interferes in assays of most "critical care" analytes that might be urgently required peri- or postoperatively. This is not surprising because most electrolyte determinations are performed by ion-selective electrodes and should not be affected by colored substances or by those that scatter light. We also currently report electrode-based results from the ABL 505 blood gas analyzer for samples from patients receiving these blood substitutes. No interference was observed in the Vitros 750 enzymatic creatinine method at HBOC concentrations up to 50 g/L. However, a marked negative interference from HBOC was observed in the Hitachi 747 colorimetric Jaffe method, which measures the picric acid creatinine reaction at 505 and 570 nm. Other "critical care" blood analytes such as glucose and urea were unaffected on one or both of the analyzers we examined. Nevertheless, because many laboratories may not have more than one chemistry system, it is important for manufacturers to examine the effect of HBOCs or PFCs on tests that might be needed during critical care periods.
Another test that may occasionally be needed in the operative or postoperative period is a cardiac marker. The troponin I assay in the Stratus is clearly free from interference at HBOC Hb concentration as high as 50 g/L in both positive (troponin 15.6 [micro]g/L) and negative (troponin <0.4 [micro]g/L) patient pools. In contrast, the CK-MB assay exhibited an unusual interference pattern in the positive patient pool (CK-MB 15.5 mg/L), where HBOC concentrations of 5-20 g/L produced a 20% negative interference and those >40 g/L produced a slight (5%) positive interference. Although a previous study found no interference in this assay at an Hb concentration of 3 g/L , repeat analysis with a separate patient pool confirmed our observation. No positive interference was observed for the negative patient pool. Given these observations, we use troponin I instead of CK-MB as a cardiac marker for patients receiving HBOC at our institution.
Taken together, these results make clear that tests of water and electrolyte homeostasis, acid-base status, renal function, and cardiac damage can be performed for patients receiving these oxygen carriers by using the instrumentation at our institution. However, it is important to note that although HBOC does not affect [K.sup.+] values the way that true erythrocyte hemolysis does, it is not possible to determine whether hemolysis in an HBOC-containing sample contains a component of erythrocyte hemolysis. Therefore, clinicians must realize that a [K.sup.+] value reported from a sample containing HBOC might be falsely increased if true hemolysis is present and, furthermore, that the laboratory currently has no means to determine this.
[FIGURE 1 OMITTED]
Electrolytes not measured by ion-selective electrodes exhibited various degrees of interference from HBOC. Both magnesium and phosphorus show >10% positive interference on the Vitros analyzers at HBOC concentrations >20 and 25 g/L, respectively. In contrast, on the Hitachi analyzer these assays are not affected. Calcium values were influenced by HBOC on both analyzers, which is consistent with previous studies [15-17] in which increased Hb concentration influenced calcium measurement on many methods and instruments but is not mentioned in package inserts. The marked interference (~70 g/L) we observed in both the Vitros 750 and Hitachi 747 assays for phosphorus with the Perflubron-containing samples was caused by the phosphate-buffered saline content of the Perflubron emulsion.
Previous studies indicated that Hb dramatically interfered with bilirubin measurement in both total and conjugated bilirubin assays [13, 18-20]. Our results confirmed this effect, regardless of whether the Jendrassik-Grof diazotization method or the Vitros method was used. Hb interferes with a large number of colorimetrically detected assays, so our results are not surprising. Most of our results are consistent with manufacturers' package inserts, stating the limitations of performing analyses on hemolyzed samples, and with our own internal data obtained by assaying samples containing soluble human Hb. Examples include all of the tests not exhibiting interference and the amount of HBOC (as Hb) causing negative interference in the Hitachi alkaline phosphatase and [gamma]-glutamyltransferase methods; the Hitachi methods for uric acid, albumin, and bicarbonate; and the Vitros method for conjugated bilirubin. Exceptions to manufacturers' package insert claims on Hb interference include the Hitachi methods for alanine aminotransferase (ALT), aspartate aminotransferase (AST), cholesterol, and calcium and the Vitros methods for calcium, phosphorus, ALT, amylase, AST, and magnesium. One likely cause for this is that the package inserts usually state no interference up to a particular Hb concentration such as 6 or 10 g/L--considerably less than that tested in these studies. Interferences probably were not examined at Hb concentrations as high as 50 g/L. In other cases, bovine HBOC may behave differently from soluble human Hb from erythrocytes. Finally, for some tests (e.g., ALT, AST, and calcium on the Hitachi and calcium on the Vitros), the manufacturer may not have deemed the extent of interference to be clinically significant; indeed, in some cases, the interferences we found only slightly exceeded our criteria for determining whether or not to report a result.
While many assays are affected by HBOC concentrations that can be present in patients, it is important to note that HBOC and Perflubron emulsion have short half-lives (2-12 h) [1-4]. Together with the lack of interference by these substances in most critical care analyte methods, patient care is unlikely to be affected if some enzyme and electrolyte tests need to be delayed for a day. Indeed, in clinical trials at our institution spanning 3 years and including >125 patients, we have had only one insistent request for a test result affected by HBOC. In settings where such a result is absolutely required, dilution of the sample might decrease the HBOC concentration to an amount that does not interfere.
In our current studies, as well as in others (e.g., [211), interferences by oxygen carriers have generally been examined at only a single concentration of the analyte in question. Thus, reporting the extent of interference as a percentage might be misleading. For instance, a 30% interference for a sample with an amylase of 95 U/L might actually be trivial if the maximum interference is a constant 30 U/L at all amylase concentrations. Assessment of these interferences at multiple analyte concentrations will allow laboratories to determine when an interference actually invalidates clinical information. For example, an amylase value of 500 U/L in the presence of 50 g/L HBOC would indeed still be useful if the interference was only 30 U/L. Similarly, a "one-way" result might also be clinically useful; in other words, if HBOC produces a positive interference, a low result might be informative, if the laboratory and clinician discuss the situation.
As blood substitute products approach clinical use, it is the responsibility of the laboratory to assess their impact on clinical laboratory values and, along with clinicians, determine what amounts of interference are acceptable for patient care. We have established guidelines for our institution based on Tables 1 and 3. All samples from patients receiving these oxygen carriers are identified by colored stickers that alert the laboratory. If HBOC concentrations (as plasma Hb) are greater than depicted, the resulting assay values are not reported.
This study and that by Callas et al. [211 are the first step toward assessing the impact of these substances on laboratory values. Future studies should examine their impact by using multiple analyte concentrations, determine the effect of changing sample matrices when these samples are diluted to minimize an interference, particularly in immunoassays and dry-chemistry methods, and assess methods to remove these substances before biochemical testing . It will also be necessary to examine the effect of these substances in nonfluorometric homogeneous immunoassays, which we did not examine at our institution. Thus, while this and other studies will be useful guidelines, individual laboratories should assess these interferences in light of the myriad testing methods in use, differences in the various hemoglobin-based blood substitute products, and institutional-specific opinions as to what constitutes significant interference.
Received May 16, 1997; revised June 20, 1997; accepted June 23, 1997.
[1.] Jones JA. Red blood cell substitutes: current status. Br J Anaesth 1995;74:697-703.
[2.] Dietz NM, Joyner MJ, Warner MA. Blood substitutes: fluids, drugs or miracle solutions. Anesth Analg 1996;82:390-405.
[3.] Zuck TF, Riess JG. Current status of injectable oxygen carriers. Crit Rev Clin Lab Sci 1994;31:295-324.
[4.] Scott MG, Kucik DF, Goodnough LT, Monk TG. Blood substitutes: evolution and future applications. Clin Chem 1997;43:1724-31.
[5.] Feola M, Simoni J, Angelillo R, Luhruma Z, Kabakele M, Manzombi M, et al. Clinical trial of a hemoglobin based blood substitute in patients with sickle cell anemia. Surg Gynecol Obstet 1992;174:379-86.
[6.] Feder BJ. Race for artificial blood heats up. New York Times Feb 14, 1994:D1.
[7.] Roberge JQ. Search narrows for blood substitute. Biotech Lab Int Sep/Oct 1996;6.
[8.] Wahr JA, Trouwborst A, Spence RK, Henry CP, Cernaianu AC, Graziano GP, et al. A pilot study of the effects of a Perflubron emulsion AF0104 on mixed venous oxygen tension in anaesthecized surgical patients. Anesth Analg 1996;82:103-7.
[9.] Vlahakes GJ, Lee R, Jacobs EE, La Raias PJ, Austen WG. Hemodynamic effects and oxygen transport properties of a new blood substitute in a model of massive blood replacement. J Thorac Cardiovasc Surg 1990;100:379-88.
[10.] Keipert PE, Faithful NS, Bradley JD, Hazard DY, Hogan J, Levisetti MS, Peters RM. Enhanced 02 delivery by perflubron emulsion during acute hemodilution. Artif Cells Blood Substit Immobil Biotechnol 1994;22:1161-7.
[11.] Sonntag O. Haemolysis as an interference factor in clinical chemistry. J Clin Chem Clin Biochem 1986;24:127-39.
[12.] Guder WG. Haemolysis as an influence and interference factor in clinical chemistry. J Clin Chem Clin Biochem 1986;24:125-6.
[13.] Glick M, Ryder KW. Interferographs. Users guide to interferences in clinical chemistry instruments. Indianapolis: Science Enterprises, 1987.
[14.] Chapelle JP, El Allaf M. Automated quantification of creatine kinase MB isoenzyme in serum by radial partition immunoassay, with use of the Stratus analyzer. Clin Chem 1990;36:99-101.
[15.] Chan KM, Arriaga C, Landt M, Smith CH, Ng RH. Interference by hemolysis with various methods for total calcium and its correction by trichloroacetic acid precipitation. Clin Chem 1983;29: 1497-500.
[16.] Porter WH, Carroll JR, Roberts RE. Hemoglobin interference with Du Pont Automatic Clinical Analyzer procedure for calcium. Clin Chem 1977;23:2145-7.
[17.] Frank JJ, Bermes EW, Bickel MJ, Watkins BF. Effect of in vitro hemolysis on chemical values for serum. Clin Chem 1978;24:1966-70.
[18.] Shull BC, Lees H, Li PK. Mechanism of interference by hemoglobin in the determination of total bilirubin. II. Method of Jendrassik-Grof. Clin Chem 1980;26:26-9.
[19.] Karkoski DJ. Hemoglobin interference with the BMD total bilirubin assay in the Hitachi 705 analyzer, and its relation to the hemolytic index [Letter]. Clin Chem 1985;31:791.
[20.] van dW, de LJ, Guder WG, Schleicher E, Paetzke I, Schleithoff M, et al. Studies on the interference by haemoglobin in the determination of bilirubin. J Clin Chem Clin Biochem 1983;21:437-43.
[21.] Callas DD, Clark TC, Moriera PL, Lansden C, Gawryl MS, Kahn S, Bermes EW Jr. In vitro effects of a novel hemoglobin-based oxygen carrier on routine chemistry, therapeutic drug, coagulation, hematology, and blood bank assays. Clin Chem 1997;43:1744-8.
[22.] Balion CM, Champagne PA, Ali ACY. Evaluation of HemogloBind[TM] for removal of o-raffinose cross-linked hemoglobin (Hemolink[TM]) from serum. Clin Chem 1997;43:1796-7.
ZHONGMIN MA, (1) TERRI G. MONK, (2) LAWRENCE T. GOODNOUGH, (1) ADRAIN MCCLELLAN, (3) MARIA GAWRYL, (4) TERRI CLARK, (4) PAULO MOREIRA, (4) PETER E. KEIPERT (5) and MITCHELL G. SCOTT (1)* Departments of (1) Pathology, Box 8118, and (2) Anesthesiology, Box 8054, Washington University School of Medicine, and (3) Department of Laboratories, Barnes-Jewish Hospital, 660 S. Euclid Ave., St. Louis, MO 63110. (4) Biopure Corp., Cambridge, MA. (5) Alliance Pharmaceutical Corp., San Diego, CA. * Author for correspondence. Fax 314-362-1461; e-mail firstname.lastname@example.org. (6) Nonstandard abbreviations: Hb, hemoglobin; PFC, perfluorocarbon; HBOC, hemoglobin based-oxygen carrier; CK, creatine kinase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; PT, prothrombin time; and aPTT, activated partial thromboplastin time.
Table 1. Maximum concentration of bovine HBOC (g/L) for reporting clinical chemistry values. Vitros 750 Assay Analyte concn. Max HBOC Sodium, mmol/L 140 >50 Potassium, mmol/L 3.9 >50 Chloride, mmol/L 110 >50 Glucose, mg/L 1800 >50 C[O.sub.2], mmol/L 15 >50 Creatinine, mg/L 45 >50 Urea (BUN), mg/L 485 >50 Calcium 71 4 Phosphorus, mg/L 47 25 Albumin, mg/L N.D. (a) Alkaline phosphatase, U/L N. D. ALT, U/L 111 2 Amylase, U/L 55 3 AST,U/L 141 20 Conjugated bilirubin, mg/L 5 Any (b) Bilirubin total, mg/L 33 Any (b) Cholesterol, mg/L N. D. [gamma]-Glutamyltransferase, U/L N. D. Lipase 1560 45 Lactate dehydrogenase, U/L N. D. Magnesium, mmol/L 1.9 20 Uric acid, mg/L 40 >50 Vitros 750 Hitachi 747 Assay Interference Analyte concn. Sodium, mmol/L None 139 Potassium, mmol/L None 4.3 Chloride, mmol/L None 107 Glucose, mg/L None 1820 C[O.sub.2], mmol/L None 14 Creatinine, mg/L None 50 Urea (BUN), mg/L None 520 Calcium [up arrow] 77 Phosphorus, mg/L [up arrow] 47 Albumin, mg/L 20 Alkaline phosphatase, U/L 61 ALT, U/L [down arrow] 103 Amylase, U/L [up arrow] N.D. AST,U/L [up arrow] 146 Conjugated bilirubin, mg/L [up arrow] N.D. Bilirubin total, mg/L [up arrow] 32 Cholesterol, mg/L 1250 [gamma]-Glutamyltransferase, U/L 65 Lipase [down arrow] N.D. Lactate dehydrogenase, U/L 274 Magnesium, mmol/L [up arrow] 1.8 Uric acid, mg/L None 38 Hitachi 747 Assay Max HBOC Interference Sodium, mmol/L >50 None Potassium, mmol/L >50 None Chloride, mmol/L >50 None Glucose, mg/L >50 None C[O.sub.2], mmol/L 30 [up arrow] Creatinine, mg/L 9 [up arrow] Urea (BUN), mg/L >50 None Calcium 9 [up arrow] Phosphorus, mg/L >50 None Albumin, mg/L 3 [up arrow] Alkaline phosphatase, U/L 2 [down arrow] ALT, U/L 7 [down arrow] Amylase, U/L AST,U/L 14 [up arrow] Conjugated bilirubin, mg/L Bilirubin total, mg/L Any (b) [down arrow] Cholesterol, mg/L 3 [up arrow] [gamma]-Glutamyltransferase, U/L 5 Lipase Lactate dehydrogenase, U/L 8 [down arrow] Magnesium, mmol/L >50 Uric acid, mg/L Any (b) [up arrow] (a) This assay is not performed on the indicated analyzer at our institution. (b) The presence of any visible "hemolysis" negates reporting values from the assay Table 2. Immunoassays unaffected by HBOC [less than or equal to]50 g/L. Analyte Instrument Troponin I Dade Stratus Thyroid-stimulating hormone Abbott IMx Digoxin Abbott AxSym Phenytoin Abbott AxSym Lidocaine Abbott TDx N-Acetylprocainamide Abbott TDx Procainamide Abbott TDx Quinidine Abbott AxSym Theophylline Abbott AxSym Table 3. Immunoassays affected by HBOC. Analyte Method Interference HBOC, g/L (a) CK-MB Date Stratus [down arrow] 5 Gentamycin Abbott AxSym [up arrow] 25 Vancomycin Abbott AxSym [down arrow] 5 (a) HBOC concentration above which values from thes e assays are not reported. Table 4. Clinical chemistry tests unaffected by the presence of Perflubron (1:20) in plasma. Values (a) Test Vitros 750 Hitachi 747 (b) Sodium 141 mmol/L Potassium 4.1 mmol/L Chloride 107 mmol/L Lithium 1.1 mmol/L N. D. (c) Total C[O.sub.2] 21 mmol/L Magnesium 1.7 mmol/L Calcium 91 g/L Creatinine 10 g/L Urea N 170 g/L Glucose 830 g/L Bilirubin, total 5 g/L Uric acid 59 g/L Lactate 3.8 mmol/L N.D. Total protein N.D. 67 g/L Albumin N.D. 36 g/L Cholesterol N.D. 1620 g/L HDL-cholesterol N.D. 480 g/L Triglyceride N.D. 1080 g/L Lipase 392 U/L N.D. Creatine kinase N.D. 85 U/L Acid phosphatase 0.7 U/L N.D. Alkaline phosphatase N.D. 126 U/L Lactate dehydrogenase N.D. 197 U/L ALT 25 U/L AST N.D. 30 U/L [gamma]-Glutamyltransferase N.D. 59 U/L (a) Values obtained from the plasma pool in the absence of Perflubron. (b) Values not shown were within 2% of those obtained on the Vitros 750 (c) Not performed on the indicated analyzer at our institution.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Oak Ridge Conference|
|Author:||Ma, Zhongmin; Monk, Terri G.; Goodnough, Lawrence T.; McClellan, Adrain; Gawryl, Maria; Clark, Terri|
|Date:||Sep 1, 1997|
|Previous Article:||Blood substitutes: evolution and future applications.|
|Next Article:||Interferences of o-raffinose cross-linked hemoglobin in three methods for serum creatinine.|