ANNALS O F CLINICAL AND LABORATORY SCIENCE, Vol. 29, No. 3Copyright © 1999, Institute for Clinical Science, Inc.

Urinalysis: Current Status and Prospects for the Future*

ANDREW E. LORINCZ, M .D .,| DAVID R. KELLY, M.D.,$

G. C LEM EN T DOBBINS, B.A.,§ VICKI S. CARDONE, B.S., MT(ASCP),$

STEPHANIE A. FUCHS, B.S., MT(ASCP),$ and JANE L. SCHILLECI, B .S, M T(ASCP).|

fDepartment o f Pediatrics, University o f Alabama at Birmingham,

Birmingham, AL 35294,

fPathology and Laboratory Medicine, The Children’s Hospital, Birmingham, AL 35233,

§University o f Alabama at Birmingham, Birmingham, AL 35294.

ABSTRACT

More than 300 million routine clinical analyses are perform ed annually in the United States. Methods for routine clinical urine examination, including detection of bacteriuria, are briefly reviewed. Prospects of some newer, better techniques to carry out such analyses are introduced. A preliminary report is presented on the use of supravital microscopic fluorescence technique (SMFT), employing acridine orange as a non-specific staining fluorochrome. Results of examining 218 unspun urine specimens by SMFT are compared to traditional bacteriologic culture at a large pediatric hospital reference laboratory.

Introduction

Analysis of urine has been considered an essential part of the examination of a patient since 4000 B.C.1 Each year, more than 300

This paper was presented to the 117th Meeting of the Association of Clinical Scientists, “Clinical Science in the New Millennium: The Past is Prologue,” in Philadelphia, PA, October 23, 1998.

* Address reprint requests to: Andrew E. Lorincz, M.D., Professor Emeritus, Department of Pediatrics, Uni­versity of Alabama at Birmingham, 1825 University Bou­levard MJH B-70, Birmingham, AL 35294.

million routine clinical urinalyses are p e r­formed in the United States. Urinalysis in its broadest sense includes the physical, chemical or microscopic inspection of urine. This paper addresses prim arily rou tine clinical u rine examination, including the detection of bacte­riuria, and presents prospects of some newer techniques to perform such analyses more effectively. The biochemical inspection of uri­nary constituents for detection of disordered metabolism, eg, aminoacidopathies, mucopoly­saccharidoses, etc. or the testing for drugs and drug metabolites, is not within the focus of this article.

0091-7370/99/0500-169 $01.75 © Institute for Clinical Science, Inc.

170 LO R IN C Z, KELLY, D O BBIN S, CA R D O N E, FU C H S , & S C H IL L E C I

Routine Urinalysis

Routine urinalysis, as in all laboratory test­ing on humans, proceeds through the process of subject selection, ordering, specimen pro­curem ent, specimen transport, preparation, analysis and reporting.2

Fashions in urine testing have changed con­siderably since 1679, when Thomas Willis described the taste of diabetic urine as “won­derfully sweet as if it were imbued with honey or sugar.” Many older physicians and clinical scientists have lived through the era of gently boiling urine in a test tube over a Bunsen burner, or they have used Fehling’s solution or Benedict’s solution to test for reducing sugars in urine.

In the early 1940s, a major advance in the laboratory analysis of urine occurred when commercially-prepared copper reduction tab­lets were first used in the laboratory. Copper sulfate, sodium hydroxide, citric acid and car­bonate were pressed as a tablet. The tablet provided its own heat for the Benedict’s cupric ion to cuprous ion chemical reaction with urine. Reducing sugars would change the color from blue to orange-red, depending upon the amount of reducing substance in the urine.

The introduction of urine chemistry reagent strips with complex chemical reactions incor­porated into a paper “solid phase” matrix dra­matically and forever changed urine testing and screening to a simple dip and “read” pro­cess. At first, in 1956, only glucose was mea­sured by this dipstick technique.3 Protein and ketone testing was added to a single strip in1957.4 Later, pH (1959), blood (1961), biliru­bin and urobilinogen (1969) measurement and specific gravity assessment (1981) were devel­oped as dipstick tests.

In 1972, nitrite was added to the specturm of urine dipstick tests. A positive nitrite test indicates that bacteria, which are capable of reducing nitrate in urine, are present in “sig­nificant” numbers. Unfortunately, not all bac­teria convert nitrate to nitrite.

Introduction of the leukocyte esterase analy­sis to the paper dipstick array of screening tests in 1984 provided a means of estimating the

white blood cell content of urine. Unfortu­nately, neither the nitrite nor the leukocyte esterase dipstick test is considered as reliable an indicator of urinary tract infection as detec­tion of bacteriuria by traditional Gram’s stain or the use of the extant “gold standard” of bac­terial detection via bacterial culture of a cath- eterized urine specimen or a fresh midstream “clean catch” specimen. In practice, the ease and simplicity of dipstick testing frequently outweigh the complexity of and time required for “gold standard” evaluation.

Routine urinalysis can be perform ed in vastly different settings: by large commercial reference laboratories, handling thousands of tests daily, by smaller commercial urine labo­ratories, by hospital clinical laboratories or by physicians’ office laboratories. A laboratory technician, physician or nurse can perform such testing.

For detection of bacteriuria, specimen col­lection ideally requires a midstream urine specimen in a clean container. Cleansing of the genital area prior to collection is advised. Quantitative loop inoculation of a well-shaken urine sample to bacterial culture medium soon after collection is desirable. If storage of the specimen is required, collection in a closed sterile container, preferably with storage time not exceeding a few hours and temperatures maintained at 4°C, has been recom m ended.5

For microscopic examination of urine, 10, 12, or 15 ml of properly collected urine is cen­trifuged at 450 g for five minutes. The super­natant is removed and preserved for use in diluting the sediment sample if microscopic exam ination is req u ired . M icroscopy of the sedim ent can be perform ed by bright- field, phase contrast or polarized light vis­ual analysis.5

The diagnosis of urinary tract infection is based on a quantitative estimation of the con­centration of bacteria in the urine specimen. An assumption that each bacterium multiplies to become a colony has led to calling these “colony-forming units,” (CFU). In an asymp­to m a tic in d iv id u a l, m ore th a n 100,000 CFU/ml in an adequately collected specimen usually indicates urinary tract infection; in

URINALYSIS 171

patients with symptoms, the threshold can be as low as 1000 CFU/ml. Although uninfected bladder urine is sterile, samples collected by the spontaneous passage of urine are never ste rile because th e cleansing process of the genital area is not completely efficient. Therefore, bacterial colonies generally will be found in the urine of uninfected, as well as infected, patients.6

Automation

Urine chemistry dip-and-read tests are tech­nically chemically complex, but the process itself is certainly convenient and easy and can be perform ed by anyone with normal color vision. Since 1956, well over 20 billion dip­sticks have been manufactured. They are used extensively in a variety of situations, eg, in hos­pitals, clinics, physicians’ offices and homes of patients, and by individuals with diverse levels of education and training.

Several types of automated and semiauto­mated urine analyzer instruments have been developed to “read” chemical tests on dipsticks by reflectance spectroscopy, to standardize readings and, very importantly, via computer interface, to store data and print out results.

Since 1997, such reagent strip urine analyz­ers have been advanced to measure human chorionic gonadotropin (hCG) levels to serve as a fast, reliable way to test for pregnancy. Microalbumin reagent strips have also been developed to identify patients with early stages of kidney dam age. The hC G and m icro­albumin reagent srips, however, require the use o f a sem iautom atic dedica ted in s tru ­m ent analyzer.

Many automated and semiautomated urine analyzers are currently available. The majority were developed in the mid-1980s and are used mainly in large hospital and comercial labora­tories. These analyzers can cost more than $100,000. Their specimen capacity (through­put/hour) can vary from 25 to greater than 250. Microscopic analysis is not available in most urine analyzers; slideless microscopy is avail­able in a few. Slideless microscopy provides a profile of cells and microorganisms present in

a urine specimen. However, if the profile indi­cates an abnormally high level of microorgan­isms, routine brightfield urine microscopy is still recommended.

Prospects For the Future o f Autom ated U rine Analyzers

At a meeting of the American Society for Microbiology in 1998, a fully automated com­puter-assisted, optically-based, in vitro diag­nostic instrument intended for rapid diagnoses of bacteria in urine was reported and is cur­rently still under development. Optical prop­erties of single cells are used for quantitative measurement by computerized image analysis techniques. Apparently, this instrum ent can provide antibiotic susceptibility testing in a m atter of two to three hours and is capable of performing bacterial screening and microor­ganism identification via monoclonal antibod­ies. This may well be the prototype instrument for use in large-scale reference testing labora­tories. This system has the potential of auto­mating completely the evaluation of urine specimens for the presence of bacteria. Such a process would be time- and labor-saving and, very importantly, would eliminate the need for culturing negative specimens.

D ata Interpretation

In the 1950s, Kass conducted studies which resulted in the concept that a urinary tract infection is likely to be present if bacterial cul­ture of a clean voided specimen grows a sin­gle bacterial species w ith a colony count >100,000 colony-forming units (CFU) per milliliter. His contributions, along with those of others, led to the universal adoption of quantitative cultures for the diagnosis of a uri­nary tract infection.7-9 Varying criteria for “sig­nificant” bacteriuria are indicated for catheter- ized and “clean catch” specimens.

There is a tendency now to move urine test­ing away from the physician’s office laboratory. With the transport of specimens to a commer­cial laboratory, an era of urine practice has evolved where broad-spectrum antimicrobials

172 LO RIN CZ, KELLY, D O B B IN S, C A R D O N E, FU C H S, & S C H IL L E C I

are im m ediately prescribed while awaiting results of urine culture and sensitivity testing24 to 48 hours later. The time lapse from the initial patient examination and receipt o f test results frequently blunts the physician’s inter­est and, consequently, the attention given the laboratory analysis. More importantly, it is generally known that 30 percen t of urine samples submitted for culture are reported to have a “significant” urine colony count of ^ 100,000 CFU/ml. W hen more than one spe­cies of microorganism is detected, even these elevated counts may be interpreted as being contaminants, rather than specific etiologic agents of urinary tract infection.

To accept such “quantitative” colony counts “blindly” does not necessarily represent valid data interpretation. To ignore the specificity and virulence of the infecting microorganisms involved in a urinary tract infection invites therapeutic failure. Neither should the sugges­tion tha t cu lture of the urinary sedim ent requires, “prior” washing of the sediment to dilute inhibitory factors that may be present in raw urine be lightly dismissed.10

M icroscopic Examination o f Urine

The microscopic examination of urine has not changed appreciably for over half a cen­tury. Centrifuged sediment from 10 to 15 ml. of urine is examined by brightfield microscopy, using a simple glass slide and glass cover-slip preparation. Detection of an excessive num ber of microorganisms and/or excessive numbers o f white blood cells (WBCs), coupled with information derived from dipstick analysis, fre­quently forms the basis in the decision to cul­ture or not to culture a specific specimen.

E nhanced Urine M icroscopic Examination

In 1982, Corman et al, advocated the exami­nation of fresh, unspun, unstained urine for bacteria on a counting chamber. Using a Neu- bauer hemocytometer counting chamber and a high dry objective (450x), they quantitatively counted WBCs, rods and cocci in chains (they

were unable to differentiate a single coccus from amorphous crystals). They advocated the use o f this m ethod as highly accurate for detecting bacteriuria. Currently several labo­ratories, particularly in Europe, employ modi­fications of this procedure for counting WBCs and bacteria.11

Supravital M icroscopic F luorescence Technique (SMFT)

For over a decade, the use of the supravital microscopic fluorescence technique (SMFT) has been advocated for the screening detection of microorganisms (bacteria, including myco­plasma, spirochetes, fungi, parasites, etc.) in a wide spectrum of clinical applications.12-14 This technique requires less than two minutes for sample preparation since it is basically a simple wet-slide preparation involving the addition of equal amounts of unspun urine and a fluorochrome (eg, acridine orange in a sterile phosphate buffer solution). The edges of the glass cover-slip preparation are then sealed w ith m elted paraffin . U sing an epi-fluo- rescence microscope, a 40x and/or a lOOx oil objective, it is ready for im m ediate visual inspection and interpretation.

In 1995, Lorincz, Baltaro and Adamson reported detection of significant bacteriuria using SMFT. U ncentrifuged urine samples from 77 patients ranging in age from 54 to 99 years were examined by SMFT and compared to standard urine cultures. O f the 77 speci­mens, 71 (92 percent) were accurately identi­fied as to w hether “significant” bacteriuria was present. Only one of the 17 specimens that had significant bacteriuria was not detected by SMFT examination.15

SMFT Urine Examination in Children

The authors are presently engaged in a con­siderably larger prospective study to compare the SMFT screening of uncentrifuged urine to traditional reference laboratory urinalysis and bacterial culture. The preliminary results of 218 samples are reported here. All SM FT evaluations were made by two investigators:

URINALYSIS 173

AEL, an experienced fluorescence microsco- pist, and GCD, a prem edical student with minimal brightfield microscopy experience an d no p r io r fa m ilia r ity w ith an e p i- fluorescence microscope. The SM FT slide preparation and examination were executed individually, and the elapsed time from slide preparation to reporting of SM FT results rarely exceeded five to seven m inutes. A simple decision was rendered for each SMFT evaluation, namely, whether or not urine cul­ture was indicated. Criteria were semiquanti- tative, and if microorganisms were present in greater than 1/hpf and/or the num ber of white blood cells were greater than 1/hpf, then cul­ture was considered.

All SMFT evaluations were conducted on re fr ig e ra te d , w ell-shaken , u n sp u n u rin e samples within 48 hours of collection by the hospital laboratory. The SMFT examiners had no medical record information, nor did they have data with respect to indications for urine testing, use of antimicrobials, results of dip­stick analyses o r o th e r lab o ra to ry tests. Interobserver error for the two SMFT evalu­ators was insignificant.

For the purpose of this preliminary data analysis, cultures perform ed at the hospital reference laboratory that had greater than100.000 CFU/ml, including a combined colony count of potential pathogens, were considered positive. C u ltu res w ith counts less than100.000 CFU/ml were considered negative.

R e su lts o f th e f i r s t sa m p le s e v a lu ­ated follows:

McNemar’s test comparing SMFT evalua­tion to bacterial culture as paired samples failed to reject the hypothesis that these two methods differed from each other. P value = 0.28.

Figure lb demonstrates the overwhelming capability of vitally stained material, as viewed by epi-fluorescence, to reveal exquisite detail of cellular and nuclear morphology, as well as discrete individual microorganisms. Since the organisms and cells are alive, their mobility patterns may also be observed. M oreover, microorganisms less than 1 jjl. in diameter, eg, mycoplasma that are beyond the resolution of a brightfield microscope, can be readily exam­ined by epi-fluorescence. The same field is viewed by brightfield illumination in figure lb , and here even detection of the large epithelial cell becomes a major challenge. Since a high dry 40x objective is currently used for most urine microscopy, the severe limitations of such brightfield microscopic examinations are even more evident.

Prospects For the Future o f SMFT Urinalysis

SMFT screening of urine has the potential for autom ation for large-scale use. Even though sample preparation for the supravital microscopic technique is now very simple, the day may soon come when self-staining, readily disposable, non-glass slides are available for epi-fluorescence study.

Hospital Culture greater than less than

100,000 CFU/ml 100,000 CFU/ml

c M .2I 31 Üw -g H «5 fa o> o§ s l w « a

38(17.4%)

14(6.4%)

8(3.7%)

158(72.5%)

Predictive value o f a positive test = 73% Sensitivity = 83%

Predictive value o f a negative test =93% Specificity = 92%

Additional Advantages o f Supravital M icroscopic Technique

SMFT is very inexpensive, has a rapid turn­around time, can be perform ed at point of care and is relatively user-friendly for anyone with minimal microscope skills. The screening tech­nique is scientifically more accurate. The test is reproducible and requires less than 100 |xl of specimen. The procedure is easily taught and easily learned. It can accurately elim inate unnecessary urine cultures in 70 percent or more cases, which further enhances its cost-

174 LO RIN CZ, KELLY, D O BBIN S, C A R D O N E, FU C H S, & SC H IL L E C I

a

F igure 1. a . Oil objective xlOO view of an epithelial cell with adjacent and adherent bacteria, as seen using an epi-fluorescence microscope. Acridine orange in PBS is the fluorochrome. b . The same field as in la as seen through a brightfield microscope, at the same magnification.

effectiveness. T he ability to view vitally stained microorganisms can achieve m uch o f the same results as G ram ’s stain preparations. U nfortu­nately, as we have reported here, SM FT non- specifically stains all cells and does not detail the potential susceptibility o f the m icroorgan­isms detec ted to microbials. As docum ented, the re is a possibility tha t SM FT may be sub­stitu ted for darkfield m icroscopic examination since spirochetes can be readily visualized.16 Similarly, yeasts and fungi can be easily iden­tified, so that the traditional KOH preparation

could be rep laced .17 Also significant is the ability o f SM FT to determ ine nuclear m or­phology of living cells.18

Today, w hen we are attem pting to solve m anaged care problem s with yesterday’s tech­nology, the prospect o f better-developed tech­niques to carry out m eaningful laboratory test­ing looms fluorescently b righ ter than ever for routine urinalysis.

Ultimately, as in all laboratory analysis, “Die m ethode ist alles” (the m ethod is everything). N or should we forget that the real reason for

URINALYSIS 175

urine laboratory testing should be to improve the outcome for the patient’s health. Now is the time to establish a solid research agenda to measure and monitor these im portant out­comes and, most critically, to change practices when results so indicate.2

R eferences

1. Kushner DS. Urinalysis. JAMA 1966;195:163.2. Lundberg GD. The need for an outcomes research

agenda for clinical laboratory testing. JAMA 1998;280: 565-6.

3. Free AH, Adams EC, Kercher ML, Free HM, and Cook MH. A simple specific test for urine glu­cose. (Abstract) Int’l Cong Clin Chem New York 1956 (Sept.).

4. Free AH, Free HM. Urinalysis in clinical laboratory practice. Cleveland: CRC Press 1975:51-77.

5. Henry JB. Basic examination of the urine. In: Henry JB, editor. Clinical Diagnosis and Management by Laboratory Methods, 19th ed. Philadelphia: W.B. Saunders Co., 1996:411-6.

6 . Belsey R, Baer DM. Diagnosing UTI. MLO 1996 (FEB):34-6.

7. Kass EH. Bacteriuria and the diagnosis of infections of the urinary tract. AMA Arch Intem Med 1957;100: 709-14.

8 . Hoberman A, Wald ER, Reynolds EA, Penchansky L, and Charron M. Pyuria and bacteriuria in urine speci­mens obtained by catheter from young children with fever. J Pediatrics 1994;124:513-9.

9. Hellerstein S. Evolving concepts in the evaluation of the child with a urinary tract infectin. J Pediatrics 1994;124:589-92.

10. Fugazzato P. Diagnosis of interstitial cystitis. J Urol 1993;149:199-200.

11. Corman LI, Foshee WS, Kotchmar GS, Harbison RW. Simplified urinary microscopy to detect signifi­cant bacteriuria. Pediatrics 1982;70:133-35.

12. Lorincz AE. Rapid method for the identification of mycoplasma organisms. Manual of Procedures for Applications of Nucleic Acid Probes and Monoclonal Antibodies in Human Disease 1986(Nov):163-65.

13. Lorincz AE. Rapid fluorescence technique for the detection of toxic pulmonary microorganisms, e.g. Legionella pneumophilia. Manual of Procedures for Clinical and Analytical Toxicology 1987(Nov):129-31.

14. Reque PG, Lorincz AE. Supravital microscopic fluo­rescence technique for the detection of Tinea capitis. Cutis 1988;42:111-4.

15. Lorincz AE, Baltaro RJ, Adamson DM. Detection of significant bacteriuria using supravital fluorescence microscopy. Ann Clin Lab Sci 1995;24:196.

16. Lorincz AE, Reque PG. Comparison of supravital microscopic fluorescence technique (SMFT) to dark- field microscopy for detection of spirochetes. Ann Clin Lab Sci 1990;20:281-2.

17. Lorincz AE. One step on-site epi-flourescence detec­tion of fungi: a possible alternate to KOH screening. Ann Clin Lab Sci 1993;23:307.

18. Petcharuttana Y, Cutter GR, Meeks RG, Lorincz AE. Fluorescence microscopy of DEs-induced morpho­logic transformation in unfixed, cultured cells. J Oral Path Med 1989;18:451-6.