Edited by Daniel M. Baer, M.D.
Absolute neutrophil critical values
Q: Do you know of any laboratories which are using an absolute neutrophil critical value? Would you have any references on this parameter? Our infection control practitioner is requesting that we begin using an absolute value of 1,000/L. We are concerned that this value may be too high.
A: A useful reference on this subject is found in the Merck Manual, a source which presents consensus statements based on authoritative sources.1 The latest edition defines neutropenia as a neutrophil count < 1,500/L. But it notes that neutropenia in blacks is defined as a neutrophil count < 1,200/L.
Based upon the <1,500/L definition, the manual notes that the relative risk of infection is L, moderate for count of 500-1,000/L, and severe for counts <500/L. For blacks, these limits would be slightly modified. Absolute values for reporting leukocytes are much preferred for patient care, and current automated cell counters report these counts in absolute terms. However, if differential counts are being performed using blood films to determine the neutrophil proportion, it is recommended that they also be reported in absolute terms.
A meeting between your laboratory director and the infectious disease practitioners and oncologists would be useful in determining the responsibilities of the laboratory for reporting alert values. A count of <500/L of neutrophils on a newly admitted patient would, of course, require a stat report to the patients physician. If the patient is undergoing chemotherapy or is septic, counts being used to monitor the patients status may not necessarily require stat reporting. On the other hand, if there is a sudden shift in the count, perhaps a stat report is indicated. These details should be worked out in the conference.
John A. Koepke, M.D.
Professor Emeritus of Pathology
Duke University Medical Center
Durham, NC
Reference
1. Merck Manual. 17th ed. 1999:931.
Clot activator tubes
A: Clot activator tubes are made both with and without serum separator gel. The tubes that have a very viscous separation gel interfere with DAT and auto control testing in blood banking by giving false positives. To the best of my knowledge, it is acceptable to use the serum from these tubes for serologic testing, but the cells present problems because of the separator substance. The tubes without the separator gel are not recommended for antibody screens or DAT testing because the silica particles used in the tubes may cause interference with the test.
In a survey of hospital chief technologists presented at the AABB Meeting in November 2000, 12 percent of facilities surveyed (n = 137) accepted serum from serum separator specimens for antibody screens, but did not accept the red cells for ABO/Rh testing.1 Seventy-six percent did not accept either the serum for the antibody screen or the red cells for ABO/Rh testing, and 12 percent accepted both serum and cells for testing (with many qualifiers, such as only if necessary).
Many facilities are now using EDTA specimens due to the fact that they are the preferred specimen for Ortho gel and Immucor solid phase assays, as well as being the preferred specimen for the DAT and for most fetal-maternal bleed-screening tests. Therefore, incompletely clotted specimens drawn in red tops without additives are becoming less of an issue in blood banking.
The 13th edition of the AABB Technical Manual lists a procedure for the treatment of incompletely clotted specimens using thrombin, 1-percent protamine sulfate, glass beads, and epsilon aminocaproic acid (EACA), which may be useful if testing facilities are still using clot tubes rather than EDTA.2
Diane Avinoso, MPH, MT(ASCP)
Manager Transfusion Service
Oregon Health and Science University
Portland, OR
References
1. Butch, S. Personal communication. AABB Annual Meeting; Nov. 6, 2000; Washington D.C.
2. AABB Technical Manual. 13th ed. Bethesda, MD: AABB Press; 1999.
No microscopic on trace blood
Q: We do not report trace blood
on reading our urine dip-strips. One of our physicians feels that we might be missing some clinically significant data, as we do not do a microscopic on these specimens. What are your thoughts on this? We are an internal medicine clinic no pediatrics.
A: Im not sure why your laboratory has chosen not to report trace blood on your urine reagent strips. It would seem to me that such results should be reported to the physician, along with results of a microscopic examination of the urine sediment. At that point, the physician can decide if further follow-up to determine the cause of hematuria should be pursued.
It is true that the reagent strips for blood are quite sensitive (they are capable of detecting two to three red cells per high-power field) and trace values may fail to uncover disease. Normal values for red cells range from one to eight red cells per hpf, depending on the particular laboratory or study. However, when viewed together with a microscopic examination of the sediment, significant hematuria might be discovered. When hematuria does occur, it might be the result of bleeding at any point in the urogenital system, from the glomerulus to the ureter. Discovering the exact cause requires a combination of laboratory and clinical information. Causes may be renal (glomerular or nonglomerular), hematologic, or urologic (involving the ureter, bladder, prostate, or urethra).
Several papers, including a study by Lynch et al., have documented the need for further urological assessment, even if a single urine specimen shows hematuria.1 McCarthy states that, Hematuria is often a sign of underlying disease and therefore should always be investigated, not simply treated or ignored.2 The paper outlines causes and diagnosis of hematuria, factors that guide a hematuria workup, the role of urinalysis, differential diagnosis and urologic workup.
In a study by Carson et al. of more than 200 consecutive patients with asymptomatic microhematuria, 20 percent of the patients were found to have highly significant urologic lesions, and 13 percent of these had a genitourinary malignant neoplasm.3 Carsons findings also showed that the degree of significant urologic pathologic findings could not be related to the grade of hematuria.
A study of cancers of the urinary tract in healthy men over 50 years of age discovered by home testing with reagent strips for blood was reported by Messing et al.4 Of 15 men who had positive reagent strip tests that revealed diseases requiring therapy, eight showed trace as the most positive result, even when multiple tests were performed. The study confirmed two characteristics of hematuria that need to be recognized. First, the seriousness of the underlying cause of hematuria is unrelated to the amount of blood found by reagent strip and microscopic examination. Second, the occurrence of hematuria is usually intermittent, even when caused by a potentially lethal disease. Therefore, a full urologic examination for all men over 50 years of age who exhibit even a single episode of microhematuria was recommended.
All of the papers cited in this discussion define a trace reaction for blood as a positive test. Although the finding of trace hematuria by reagent strip may fail to reveal serious pathology, it seems that such findings should be reported to the physician who will be responsible for the decision about further investigation.
Karen M. Ringsrud, MT(ASCP)
Assistant Professor
Dept. of Laboratory Medicine and Pathology
University of Minnesota Medical School
Minneapolis, MN
References
1. Lynch TH, Waymont B, Dunn JA, Hughes MA, Wallace DMA. Repeat testing for haematuria and underlying urological pathology. British Journal of Urology. 1994;74:730-732.
2. McCarty JJ. Outpatient evaluation of
hematuria. Locating the source of bleeding. Postgraduate Medicine/Hematuria. 1997;101:125-131.
3. Carson CL, Segur
JW, and Green LF. Clinical Importance of Microhematuria. JAMA. Jan. 12, 1979; 241:149-150.
4. Messing EM, Young TB, Hunt VB, Wehbie
JM, Rust P. Urinary tract cancers found by home screening with hematuria dipsticks in healthy men over 50 years of age. Cancer. 1989;64:2361-2367.
Reporting in SI units
Q: In our laboratory, we report test results in conventional units. Some of our clients prefer to receive results in SI units. Is it acceptable to apply conversion factors and report the results in SI units for those clients? If so, is this true for all analytes, or is there an exception, for example, for enzymes? Is there a good source for conversion tables and factors?
A: Some lab computer systems can individualize reports for specific clients. This is seen in some European computer systems that are able to produce reports for individual clients in different languages and reporting units. At least one LIS marketed in the United States
(Molis-Sysmex Infosystems America, Tucson, AZ) has this capability. However, if you need to report in SI results with your existing lab computer system, it may be necessary to do the conversion manually and that can introduce errors.
There are books containing SI conversion tables. One such book, available from AACC
(www.AACC.org), is SI Units for Clinical Measurement by Young and Huth. Most analytes can be converted to SI units. Generally, electrolytes and enzymes are already expressed that way. Exceptions are usually complex proteins, such as hormones.
In deciding whether to provide individualized reports for clients using SI units, you must weigh several issues: Can your LIS do this automatically? Is manual conversion economically feasible? What error rate in transcription errors are you willing to accept?
Daniel M. Baer, M.D.
Professor Emeritus
Department of Pathology
Oregon Health and Science University
Portland, OR
Daniel M. Baer is professor emeritus of laboratory medicine at Oregon Health and Science University in Portland, OR, and a member of MLOs editorial advisory board.
© 2002 Nelson Publishing, Inc. All rights reserved.
