CLINICAL DECISION-MAKING - URINALYSIS

URINALYSIS

The urinalysis is a fundamental test that should be performed in all urologic patients. Although, in many instances, a simple dipstick urinalysis will provide the necessary information, a complete urinalysis includes both chemical and microscopic analyses.

Collection of Urinary Specimens Males

In the male patient, a midstream urine sample is obtained. The uncircumcised male should retract the foreskin, cleanse the glans penis with antiseptic solution, and continue to retract the foreskin during voiding. The male patient begins urinating into the toilet, and then places a wide-mouth sterile container under his penis to collect a midstream sample. This avoids contamination of the urine specimen with skin and urethral organisms.

In men with chronic UTIs, four aliquots of urine are obtained. These aliquots have been designated Voided Bladder 1, Voided Bladder 2, Expressed Prostatic Secretions, and Voided Bladder 3 (VB1, VB2, EPS, and VB3). The VB1 is the initial 5 to 10 mL of urine voided, whereas the VB2 is the midstream urine. The EPS is the secretions obtained after gentle prostatic massage, and the VB3 specimen is the initial 2 to 3 mL of urine obtained after prostatic massage. The value of these cultures for localization of UTIs is that the VB1 sample represents urethral flora; the VB2, bladder flora; and the EPS and VB3 samples, prostatic flora. The VB3 sample is particularly helpful when there is little or no prostatic fluid obtained by massage. To better obtain prostatic secretions, patients should be instructed to attempt to void during prostatic massage and to avoid tightening the anal sphincter and pelvic floor muscles. The four-part urine sample is particularly useful in evaluating men with suspected bacterial prostatitis ( Meares and Stamey, 1968 ).

Females

In the female, it is more difficult to obtain a clean-catch midstream specimen. The female patient should cleanse the vulva, separate the labia, and collect a midstream specimen as described for the male patient. If infection is suspected, however, the midstream specimen is unreliable and should never be sent for culture and sensitivity. To evaluate for a possible infection in a female, a catheterized urine sample should always be obtained.

Neonates and Infants

The usual way to obtain a urine sample in a neonate or infant is to place a sterile plastic bag with an adhesive collar over the infant's genitalia. Obviously, however, these devices may not be able to distinguish contamination from true UTI. Whenever possible, all urine samples should be examined within 1 hour of collection and plated for culture and sensitivity if indicated. If urine is allowed to stand at room temperature for longer periods of time, bacterial overgrowth may occur, the pH may change, and red and white blood cell casts may disintegrate. If it is not possible to examine the urine promptly, it should be refrigerated at 5°C.

Physical Examination of Urine

The physical examination of the urine includes an evaluation of color, turbidity, specific gravity and osmolality, and pH.

Color

The normal pale yellow color of urine is due to the presence of the pigment urochrome. Urine color varies most commonly because of concentration, but many foods, medications, metabolic products, and infection may produce abnormal urine color. This is important, because many patients will seek consultation primarily because of a change in their urine color. Thus, it is important for the urologist to be aware of the common causes of abnormal urine color, and these are listed in Table 3-3 .

Table 3-3 -- Common Causes of Abnormal Urine Color

Colorless	Very dilute urine
	Overhydration
Cloudy/milky	Phosphaturia
	Pyuria
	Chyluria
Red	Hematuria
	Hemoglobinuria/myoglobinuria
	Anthrocyanin in beets and blackberries
	Chronic lead and mercury poisoning
	Phenolphthalein (in bowel evacuants)
	Phenothiazines (e.g., Compazine)
	Rifampin
Orange	Dehydration
	Phenazopyridine (Pyridium)
	Sulfasalazine (Azulfidine)
Yellow	Normal
	Phenacetin
	Riboflavin
Green-blue	Biliverdin
	Indicanuria (tryptophan indole metabolites)
	Amitriptyline (Elavil)
	Indigo carmine
	Methylene blue
	Phenois (e.g., IV cimetidine [Tagamet],
	IV promethazine [Phenergan])
	Resorcinol
	Triamterene (Dyrenium)
Brown	Urobilinogen
	Porphyria
	Aloe, fava beans, and rhubarb
	Chloroquine and primaquine
	Furazolidone (Furoxone)
	Metronidazole (Flagyl)
	Nitrofurantoin (Furadantin)
Brown-black	Alcaptonuria (homogentisic acid)
	Hemorrhage
	Melanin
	Tyrosinosis (hydroxyphenylpyruvic acid)
	Cascara, senna (laxatives)
	Methocarbamol (Robaxin)
	Methyldopa (Aldomet)
	Sorbitol

From Hanno PM, Wein AJ: A Clinical Manual of Urology. Norwalk, CT, Appleton-Century-Crofts, 1987, p 67.

Turbidity

Freshly voided urine is clear. Cloudy urine is most commonly due to phosphaturia, a benign process in which excess phosphate crystals precipitate in an alkaline urine. Phosphaturia is intermittent and usually occurs after meals or ingestion of a large quantity of milk. Patients are otherwise asymptomatic. The diagnosis of phosphaturia can be accomplished either by acidifying the urine with acetic acid, which will result in immediate clearing, or by performing a microscopic analysis, which will reveal large amounts of amorphous phosphate crystals.

Pyuria, usually associated with a UTI, is another common cause of cloudy urine. The large numbers of white blood cells cause the urine to become turbid. Pyuria is readily distinguished from phosphaturia either by smelling the urine (infected urine has a characteristic pungent odor) or by microscopic examination, which readily distinguishes amorphous phosphate crystals from leukocytes.

Rare causes of cloudy urine include chyluria (in which there is an abnormal communication between the lymphatic system and the urinary tract resulting in lymph fluid being mixed with urine), lipiduria, hyperoxaluria, and hyperuricosuria.

Specific Gravity and Osmolality

Specific gravity of urine is easily determined from a urinary dipstick and usually varies from 1.001 to 1.035. Specific gravity usually reflects the patient's state of hydration but may also be affected by abnormal renal function, the amount of material dissolved in the urine, and a variety of other causes mentioned later. A specific gravity less than 1.008 is regarded as dilute, and a specific gravity greater than 1.020 is considered concentrated. A fixed specific gravity of 1.010 is a sign of renal insufficiency, either acute or chronic.

In general, specific gravity reflects the state of hydration but also affords some idea of renal concentrating ability. Conditions that decrease specific gravity include (1) increased fluid intake, (2) diuretics, (3) decreased renal concentrating ability, and (4) diabetes insipidus. Conditions that increase specific gravity include (1) decreased fluid intake; (2) dehydration owing to fever, sweating, vomiting, and diarrhea; (3) diabetes mellitus (glucosuria); and (4) inappropriate secretion of antidiuretic hormone. Specific gravity will also be increased above 1.035 after intravenous injection of iodinated contrast and in patients taking dextran.

Osmolality is a measure of the amount of material dissolved in the urine and usually varies between 50 and 1200 mOsm/L. Urine osmolality most commonly varies with hydration, and the same factors that affect specific gravity will also affect osmolality. Urine osmolality is a better indicator of renal function, but it cannot be measured from a dipstick and must be determined using standard laboratory techniques.

pH

Urinary pH is measured with a dipstick test strip that incorporates two colorimetric indicators, methyl red and bromothymol blue, which yield clearly distinguishable colors over the pH range from 5 to 9. Urinary pH may vary from 4.5 to 8; the average pH varies between 5.5 and 6.5. A urinary pH between 4.5 and 5.5 is considered acidic, whereas a pH between 6.5 and 8 is considered alkaline.

In general, the urinary pH reflects the pH in the serum. In patients with metabolic or respiratory acidosis, the urine is usually acidic; conversely, in patients with metabolic or respiratory alkalosis, the urine is alkaline. Renal tubular acidosis (RTA) presents an exception to this rule. In patients with both type I and II RTA, the serum is acidemic, but the urine is alkalotic because of continued loss of bicarbonate in the urine. In severe metabolic acidosis in type II RTA, the urine may become acidic; but in type I RTA, the urine is always alkaline, even with severe metabolic acidosis ( Morris and Ives, 1991 ). Urinary pH determination is used to establish the diagnosis of RTA; inability to acidify the urine below a pH of 5.5 after administration of an acid load is diagnostic of RTA.

Urine pH determinations are also useful in the diagnosis and treatment of UTIs and urinary calculus disease. In patients with a presumed UTI, an alkaline urine with a pH greater than 7.5 suggests infection with a urea-splitting organism, most commonly Proteus. Urease-producing bacteria convert ammonia to ammonium ions, markedly elevating the urinary pH and causing precipitation of calcium magnesium ammonium phosphate crystals. The massive amount of crystallization may result in staghorn calculi.

Urinary pH is usually acidic in patients with uric acid and cystine lithiasis. Alkalinization of the urine is an important feature of therapy in both of these conditions, and frequent monitoring of urinary pH is necessary to ascertain adequacy of therapy.

Chemical Examination of Urine Urine Dipsticks

Urine dipsticks provide a quick and inexpensive method for detecting abnormal substances within the urine. Dipsticks are short, plastic strips with small marker pads that are impregnated with different chemical reagents that react with abnormal substances in the urine to produce a colorimetric change. The abnormal substances commonly tested for with a dipstick include (1) blood, (2) protein, (3) glucose, (4) ketones, (5) urobilinogen and bilirubin, and (6) white blood cells.

Substances listed in Table 3-3 that produce an abnormal urine color may interfere with appropriate color development on the dipstick. In our experience, this most commonly occurs in patients taking phenazopyridine (Pyridium) for a UTI. Phenazopyridine turns the urine bright orange and makes dipstick evaluation of the urine unreliable.

Appropriate technique must be used to obtain an accurate dipstick determination. The reagent areas on the dipstick must be completely immersed in a fresh uncentrifuged urine specimen and then must be withdrawn immediately to prevent dissolution of the reagents into the urine. As the dipstick is removed from the urine specimen container, the edge of the dipstick is drawn along the rim of the container to remove excess urine. The dipstick should be held horizontally until the appropriate time for reading and then compared with the color chart. Excess urine on the dipstick or holding the dipstick in a vertical position will allow mixing of chemicals from adjacent reagent pads on the dipstick, resulting in a faulty diagnosis. False-negative results for glucose and bilirubin may be seen in the presence of elevated ascorbic acid concentrations in the urine. However, increased levels of ascorbic acid in the urine do not interfere with dipstick testing for hematuria. Highly buffered alkaline urine may cause falsely low readings for specific gravity and may lead to false-positive results for urinary protein. Other common causes of false results with dipstick testing are outdated test strips and exposure of the sticks, leading to damage to the reagents. In general, when the sticks are damaged, there will be color changes on the pads before their immersion in urine. If such color changes are noted, results with the dipstick may be inaccurate.

Hematuria

Normal urine should contain less than three red blood cells per HPF. A positive dipstick for blood in the urine indicates either hematuria, hemoglobinuria, or myoglobinuria. The chemical detection of blood in the urine is based on the peroxidase-like activity of hemoglobin. When in contact with an organic peroxidase substrate, hemoglobin catalyzes the reaction and causes subsequent oxidation of a chromogen indicator, which changes color according to the degree and amount of oxidation. The degree of color change is directly related to the amount of hemoglobin present in the urine specimen. Dipsticks frequently demonstrate both colored dots and field color change. If present, free hemoglobin and myoglobin in the urine are absorbed into the reagent pad and catalyze the reaction within the test paper, thereby producing a field change effect in color. Intact erythrocytes in the urine undergo hemolysis when they come in contact with the reagent test pad, and the localized free hemoglobin on the pad produces a corresponding dot of color change. Obviously, the greater the number of intact erythrocytes in the urine specimen, the greater the number of dots that will appear on the test paper, and a coalescence of the dots occurs when there are more than 250 erythrocytes/mL.

Hematuria can be distinguished from hemoglobinuria and myoglobinuria by microscopic examination of the centrifuged urine; the presence of a large number of erythrocytes establishes the diagnosis of hematuria. If erythrocytes are absent, examination of the serum will distinguish hemoglobinuria and myoglobinuria. A sample of blood is obtained and centrifuged. In hemoglobinuria, the supernatant will be pink. This is because free hemoglobin in the serum binds to haptoglobin, which is water insoluble and has a high molecular weight. This complex remains in the serum, causing a pink color. Free hemoglobin will appear in the urine only when all of the haptoglobin-binding sites have been saturated. In myoglobinuria, the myoglobin released from muscle is of low molecular weight and water soluble. It does not bind to haptoglobin and is therefore excreted immediately into the urine. Therefore, in myoglobinuria the serum remains clear.

The sensitivity of urinary dipsticks in identifying hematuria, defined as greater than 3 erythrocytes/HPF of centrifuged sediment examined microscopically, is over 90%. Conversely, the specificity of the dipstick for hematuria compared with microscopy is somewhat lower, reflecting a higher false-positive rate with the dipstick ( Shaw et al, 1985 ).

False-positive dipstick readings most often are due to contamination of the urine specimen with menstrual blood. Dehydration with resultant urine of high specific gravity can also yield false-positive results owing to the increased concentration of erythrocytes and hemoglobin. The normal individual excretes about 1000 erythrocytes/mL of urine, with the upper limits of normal varying from 5000 to 8000 erythrocytes/mL ( Kincaid-Smith, 1982 ). Therefore, examining urine of high specific gravity, such as the first morning voided specimen, increases the likelihood of a false-positive result. In addition to dehydration, another cause of false-positive results is exercise, which can increase the number of erythrocytes in the urine.

The efficacy of hematuria screening using the dipstick to identify patients with significant urologic disease is somewhat controversial. Studies in children and young adults have shown a very low rate of significant disease ( Woolhandler et al, 1989 ). In older adults, one study from the Mayo Clinic of 2000 patients with asymptomatic hematuria showed that only 0.5% had a urologic malignancy and only 1.8% developed other serious urologic diseases within 3 years after identification of the hematuria ( Mohr et al, 1986 ). Conversely, investigators at the University of Wisconsin found that 26% of adults who had at least one positive dipstick reading for hematuria were subsequently found to have significant urologic pathology ( Messing et al, 1987 ). Obviously, the age of the population, the completeness of the subsequent urologic evaluation, and the definition of significant disease all influence the disease rate in the group of patients with asymptomatic hematuria identified by dipstick screening. It is important to remember that, before proceeding to more complicated studies, the dipstick result should be confirmed with a microscopic examination of the centrifuged urinary sediment.

Differential Diagnosis and Evaluation of Hematuria.

Hematuria may reflect either significant nephrologic or urologic disease. Hematuria of nephrologic origin is frequently associated with casts in the urine and almost always associated with significant proteinuria. Even significant hematuria of urologic origin will not elevate the protein concentration in the urine into the 100 to 300 mg/dL or 2+ to 3+ range on dipstick, and proteinuria of this magnitude almost always indicates glomerular or tubulointerstitial renal disease.

Morphologic evaluation of erythrocytes in the centrifuged urinary sediment also helps localize their site of origin. Erythrocytes arising from glomerular disease are typically dysmorphic and show a wide range of morphologic alterations. Conversely, erythrocytes arising from tubulointerstitial renal disease and of urologic origin have a uniformly round shape; these erythrocytes may or may not retain their hemoglobin (“ghost cells”), but the individual cell shape is consistently round. In individuals without significant pathology with minimal amounts of hematuria, the erythrocytes are characteristically dysmorphic but the number of cells observed is far fewer than that observed in patients with nephrologic disease. Erythrocyte morphology is more easily determined using phase contrast microscopy, but with practice this can be accomplished using a conventional light microscope ( Schramek et al, 1989 ).

Glomerular Hematuria.

Glomerular hematuria is suggested by the presence of dysmorphic erythrocytes, red blood cell casts, and proteinuria. Of those patients with glomerulonephritis proven by renal biopsy, however, about 20% will have hematuria alone without red blood cell casts or proteinuria ( Fassett et al, 1982 ).

The glomerular disorders associated with hematuria are listed in Table 3-4 . Further evaluation of patients with glomerular hematuria should begin with a thorough history. Hematuria in children and young adults, usually males, associated with low-grade fever and an erythematous rash suggests a diagnosis of immunoglobulin A (IgA) nephropathy (Berger's disease). A family history of renal disease and deafness suggests familial nephritis or Alport's syndrome. Hemoptysis and abnormal bleeding associated with microcytic anemia are characteristic of Goodpasture's syndrome, and the presence of a rash and arthritis suggest systemic lupus erythematosus. Finally, poststreptococcal glomerulonephritis should be suspected in a child with a recent streptococcal upper respiratory tract or skin infection.

Table 3-4 -- Glomerular Disorders in Patients with Glomerular Hematuria

Disorder	Patients
IgA nephropathy (Berger's disease)	30
Mesangioproliferative GN	14
Focal segmental proliferative GN	13
Familial nephritis (e.g., Alport's syndrome)	11
Membranous GN	7
Mesangiocapillary GN	6
Focal segmental sclerosis	4
Unclassifiable	4
Systemic lupus erythematosus	3
Postinfectious GN	2
Subacute bacterial endocarditis	2
Others	4
Total	100

Adapted from Fassett RG, Horgan BA, Mathew TH: Detection of glomerular bleeding by phase-contrast microscopy. Lancet 1982;1:1432.

GN, glomerulonephritis; IgA, immunoglobulin A.

Further laboratory evaluation should include measurement of serum creatinine, creatinine clearance, and, when proteinuria in the urine is 2+ or greater, a 24-hour urine protein determination. Although these tests will quantitate the specific degree of renal dysfunction, further tests are usually required to establish the specific diagnosis and particularly to determine whether the disease is due to an immune or a nonimmune etiology. Frequently, a renal biopsy is necessary to establish the precise diagnosis, and biopsies are particularly important if the result will influence subsequent treatment of the patient. Renal biopsies are extremely informative when examined by an experienced pathologist using light, immunofluorescent, and electron microscopy.

An algorithm for the evaluation of glomerular hematuria is provided in Figure 3-6 .

Figure 3-6 Evaluation of glomerular hematuria (dysmorphic erythrocytes, erythrocyte casts, and proteinuria). ANA, antinuclear antibody; ASO, antistreptolysin O; Ig, immunoglobulin.

IgA Nephropathy (Berger's Disease).

IgA nephropathy, or Berger's disease, is the most common cause of glomerular hematuria, accounting for about 30% of cases ( Fassett et al, 1982 ). Therefore, it is described in greater detail in this section. IgA nephropathy occurs most commonly in children and young adults, with a male predominance ( Berger and Hinglais, 1968 ). Patients typically present with hematuria after an upper respiratory tract infection or exercise. Hematuria may be associated with a low-grade fever or rash, but most patients have no associated systemic symptoms. Gross hematuria occurs intermittently, but microscopic hematuria is a constant finding in some patients. The disease is chronic, but the prognosis in most patients is excellent. Renal function remains normal in the majority, but about 25% will subsequently develop renal insufficiency. An older age at onset, initial abnormal renal function, consistent proteinuria, and hypertension are indicators of a poor prognosis ( D'Amico, 1988 ).

The pathologic findings in Berger's disease are limited to either focal glomeruli or lobular segments of a glomerulus. The changes are proliferative and usually confined to mesangial cells ( Berger and Hinglais, 1968 ). Renal biopsy reveals deposits of IgA, IgG, and β_1c-globulin, although IgA and IgG mesangial deposits are found in other forms of glomerulonephritis as well. The role of IgA in the disease remains uncertain, although the deposits may trigger an inflammatory reaction within the glomerulus ( van den Wall Bake et al, 1989 ). Because gross hematuria frequently follows an upper respiratory tract infection, a viral etiology has been suspected but not established. The frequent association between hematuria and exercise in this condition remains unexplained.

The clinical presentation of IgA glomerulonephritis is alarming and similar to certain systemic diseases, including Schönlein-Henoch purpura, systemic lupus erythematosus, bacterial endocarditis, and Goodpasture's syndrome. Therefore, a careful clinical and laboratory evaluation is indicated to establish the correct diagnosis. The presence of red blood cell casts establishes the glomerular origin of the hematuria. In the absence of casts, a urologic evaluation is indicated to exclude the urinary tract as a source of bleeding and to confirm that the hematuria is arising from both kidneys. The diagnosis of IgA nephropathy is confirmed by renal biopsy demonstrating the classic deposits of immunoglobulins in mesangial cells, as described previously. Once the diagnosis has been established, repeat evaluations for hematuria are generally not indicated. Although there is no effective treatment for this condition, renal function remains stable in most patients and there are no other known long-term complications.

Nonglomerular Hematuria Medical.

Except for renal tumors, nonglomerular hematuria of renal origin is secondary to either tubulointerstitial, renovascular, or systemic disorders. The urinalysis in nonglomerular hematuria is distinguished from that of glomerular hematuria by the presence of circular erythrocytes and the absence of erythrocyte casts. Like glomerular hematuria, nonglomerular hematuria of renal origin is frequently associated with significant proteinuria, which distinguishes these nephrologic diseases from urologic diseases in which the degree of proteinuria is usually minimal, even with heavy bleeding.

As with glomerular hematuria, a careful history frequently helps establish the diagnosis. A family history of hematuria or bleeding tendency suggests the diagnosis of a blood dyscrasia, which should be investigated further. A family history of urolithiasis associated with intermittent hematuria may indicate stone disease, which should be investigated with serum and urine measurements of calcium and uric acid. A family history of renal cystic disease should prompt further radiologic evaluation for medullary sponge kidney and adult polycystic kidney disease. Papillary necrosis as a cause of hematuria should be considered in diabetics, African Americans (secondary to sickle cell disease or trait), and suspected analgesic abusers.

Medications may induce hematuria, particularly anticoagulants. Anticoagulation at normal therapeutic levels, however, does not predispose patients to hematuria. In one study, the prevalence of hematuria was 3.2% in anticoagulated patients versus 4.8% in a control group. Urologic disease was identified in 81% of patients with more than one episode of microscopic hematuria, and the cause of hematuria did not vary between groups ( Culclasure et al, 1994 ). Thus, anticoagulant therapy per se does not appear to increase the risk of hematuria unless the patient is excessively anticoagulated.

Exercise-induced hematuria is being observed with increasing frequency. It typically occurs in long-distance runners (>10 km), is usually noted at the conclusion of the run, and rapidly disappears with rest. The hematuria may be of renal or bladder origin. An increased number of dysmorphic erythrocytes have been noted in some patients, suggesting a glomerular origin. Exercise-induced hematuria may be the first sign of underlying glomerular disease such as IgA nephropathy. Conversely, cystoscopy in patients with exercise-induced hematuria frequently reveals punctate hemorrhagic lesions in the bladder, suggesting that the hematuria is of bladder origin.

Vascular disease may also result in nonglomerular hematuria. Renal artery embolism and thrombosis, arteriovenous fistulas, and renal vein thrombosis may all result in hematuria. Physical examination may reveal severe hypertension, a flank or abdominal bruit, or atrial fibrillation. In such patients, further evaluation for renal vascular disease should be undertaken.

An algorithm for the evaluation of nonglomerular hematuria is provided in Figure 3-7 .

Figure 3-7 Evaluation of nonglomerular renal hematuria (circular erythrocytes, no erythrocyte casts, and proteinuria). CT, computed tomography; IgA, immunoglobulin A; IVU, intravenous urography; PT, prothrombin time; PTT, partial thromboplastin time; R/O, rule out.

Surgical.

Nonglomerular hematuria or essential hematuria includes primarily urologic rather than nephrologic diseases. Common causes of essential hematuria include urologic tumors, stones, and UTIs.

The urinalysis in both nonglomerular medical and surgical hematuria is similar in that both are characterized by circular erythrocytes and the absence of erythrocyte casts. Essential hematuria is suggested, however, by the absence of significant proteinuria usually found in nonglomerular hematuria of renal parenchymal origin. It should be remembered, however, that proteinuria is not always present in glomerular or nonglomerular renal disease.

The American Urological Association (AUA) Best Practice Policy Panel on Microscopic Hematuria has formulated practice recommendations for the detection and evaluation of asymptomatic microscopic hematuria (Grossfeld et al, 2001a, 2001b [16] [17]). The panel concluded that, due to the lack of specificity of urinary dipstick examination, as well as the risk and expense of evaluation, patients with a positive dipstick test should only undergo complete evaluation for hematuria if this is confirmed by the finding of 3 or more RBC/HPF on subsequent microscopic evaluation. The mainstays of evaluation, according to the panel, are voided urinary cytology, cystoscopy, and urinary tract imaging using ultrasonography, CT, and/or intravenous urography (IVU). The use of these tests in an individual patient should be based in most cases on the relative risk of significant urinary tract pathology.

An algorithm for the evaluation of essential hematuria is provided in Figure 3-8 .

Figure 3-8 Evaluation of essential hematuria (circular erythrocytes, no erythrocyte casts, no significant proteinuria). CT, computed tomography; IVU, intravenous urography; R/O, rule out.

Proteinuria

Although healthy adults excrete 80 to 150 mg of protein in the urine daily, the qualitative detection of proteinuria in the urinalysis should raise the suspicion of underlying renal disease. Proteinuria may be the first indication of renovascular, glomerular, or tubulointerstitial renal disease, or it may represent the overflow of abnormal proteins into the urine in conditions such as multiple myeloma. Proteinuria also can occur secondary to nonrenal disorders and in response to various physiologic conditions such as strenuous exercise.

The protein concentration in the urine obviously depends on the state of hydration, but it seldom exceeds 20 mg/dL. In patients with dilute urine, however, significant proteinuria may be present at concentrations less than 20 mg/dL. Normally, urine protein is about 30% albumin, 30% serum globulins, and 40% tissue proteins, of which the major component is Tamm-Horsfall protein. This profile may be altered by conditions that affect glomerular filtration, tubular reabsorption, or excretion of urine protein, and determination of the urine protein profile by such techniques as protein electrophoresis may help determine the etiology of proteinuria.

Pathophysiology.

Most causes of proteinuria can be categorized into one of three categories: glomerular, tubular, or overflow. Glomerular proteinuria is the most common type of proteinuria and results from increased glomerular capillary permeability to protein, especially albumin. Glomerular proteinuria occurs in any of the primary glomerular diseases such as IgA nephropathy or in glomerulopathy associated with systemic illness such as diabetes mellitus. Glomerular disease should be suspected when the 24-hour urine protein excretion exceeds 1 g and is almost certain to exist when the total protein excretion exceeds 3 g.

Tubular proteinuria results from failure to reabsorb normally filtered proteins of low molecular weight such as immunoglobulins. In tubular proteinuria, the 24-hour urine protein loss seldom exceeds 2 to 3 g and the excreted proteins are of low molecular weight rather than albumin. Disorders that lead to tubular proteinuria are commonly associated with other defects of proximal tubular function, such as glucosuria, aminoaciduria, phosphaturia, and uricosuria (Fanconi's syndrome).

Overflow proteinuria occurs in the absence of any underlying renal disease and is due to an increased plasma concentration of abnormal immunoglobulins and other low-molecular-weight proteins. The increased serum levels of abnormal proteins result in excess glomerular filtration that exceeds tubular reabsorptive capacity. The most common cause of overflow proteinuria is multiple myeloma, in which large amounts of immunoglobulin light chains are produced and appear in the urine (Bence Jones protein).

Detection.

Qualitative detection of abnormal proteinuria is most easily accomplished with a dipstick impregnated with tetrabromophenol blue dye. The color of the dye changes in response to a pH shift related to the protein content of the urine, mainly albumin, leading to the development of a blue color. Because the background of the dipstick is yellow, various shades of green will develop, and the darker the green, the greater the concentration of protein in the urine. The minimal detectable protein concentration by this method is 20 to 30 mg/dL. False-negative results can occur in alkaline urine, dilute urine, or when the primary protein present is not albumin. Nephrotic range proteinuria in excess of 1 g/24 hr, however, is seldom missed on qualitative screening. Precipitation of urinary proteins with strong acids such as 3% sulfosalicylic acid will detect proteinuria at concentrations as low as 15 mg/dL and is more sensitive at detecting other proteins as well as albumin. Patients whose urine is negative on dipstick but strongly positive with sulfosalicylic acid should be suspected of having multiple myeloma, and the urine should be tested further for Bence Jones protein.

If qualitative testing reveals proteinuria, this should be quantitated with a 24-hour urinary collection. Further qualitative assessment of abnormal urinary proteins can be accomplished by either protein electrophoresis or immunoassay for specific proteins. Protein electrophoresis is particularly helpful in distinguishing glomerular from tubular proteinuria. In glomerular proteinuria, albumin makes up about 70% of the total protein excreted, whereas in tubular proteinuria, the major proteins excreted are immunoglobulins with albumin making up only 10% to 20%. Immunoassay is the method of choice for detecting specific proteins such as Bence Jones protein in multiple myeloma.

Evaluation.

Proteinuria should first be classified by its timing into transient, intermittent, or persistent. Transient proteinuria occurs commonly, especially in the pediatric population, and usually resolves spontaneously within a few days ( Wagner et al, 1968 ). It may result from fever, exercise, or emotional stress. In older patients, transient proteinuria may be due to congestive heart failure. If a nonrenal cause is identified and a subsequent urinalysis is negative, no further evaluation is necessary. Obviously, if proteinuria persists, it should be evaluated further.

Proteinuria may also occur intermittently, and this is frequently related to postural change ( Robinson, 1985 ). Proteinuria that occurs only in the upright position is a frequent cause of mild, intermittent proteinuria in young males. Total daily protein excretion seldom exceeds 1 g, and urinary protein excretion returns to normal when the patient is recumbent. Orthostatic proteinuria is thought to be secondary to increased pressure on the renal vein while standing. It resolves spontaneously in about 50% of patients and is not associated with any morbidity. Therefore, if renal function is normal in patients with orthostatic proteinuria, no further evaluation is indicated.

Persistent proteinuria requires further evaluation, and most cases have a glomerular etiology. A quantitative measurement of urinary protein should be obtained through a 24-hour urine collection, and a qualitative evaluation should be obtained to determine the major proteins excreted. The findings of greater than 2 g of protein excreted per 24 hours, of which the major components are high-molecular-weight proteins such as albumin, establishes the diagnosis of glomerular proteinuria. Glomerular proteinuria is the most common cause of abnormal proteinuria, especially in patients presenting with persistent proteinuria. If glomerular proteinuria is associated with hematuria characterized by dysmorphic erythrocytes and erythrocyte casts, the patient should be evaluated as outlined previously for glomerular hematuria (see Fig. 3-6 ). Patients with glomerular proteinuria who have no or little associated hematuria should be evaluated for other conditions, of which the most common is diabetes mellitus. Other possibilities include amyloidosis and arteriolar nephrosclerosis.

In patients in whom total protein excretion is 300 to 2000 mg/day, of which the major components are low-molecular-weight globulins, further qualitative evaluation with immunoelectrophoresis is indicated. This will determine whether the excess proteins are normal or abnormal. Identification of normal proteins establishes a diagnosis of tubular proteinuria, and further evaluation for a specific cause of tubular dysfunction is indicated.

If qualitative evaluation reveals abnormal proteins in the urine, this establishes a diagnosis of overflow proteinuria. Further evaluation should be directed to identify the specific protein abnormality. The finding of large quantities of light-chain immunoglobulins or Bence Jones protein establishes a diagnosis of multiple myeloma. Similarly, the finding of large amounts of hemoglobin or myoglobin establishes the diagnosis of hemoglobinuria or myoglobinuria.

An algorithm for the evaluation of proteinuria is provided in Figure 3-9 .

Figure 3-9 Evaluation of proteinuria.

Glucose and Ketones

Urine testing for glucose and ketones is useful in screening patients for diabetes mellitus. Normally, almost all the glucose filtered by the glomeruli is reabsorbed in the proximal tubules. Although very small amounts of glucose may normally be excreted in the urine, these amounts are not clinically significant and are below the level of detectability with the dipstick. If, however, the amount of glucose filtered exceeds the capacity of tubular reabsorption, glucose will be excreted in the urine and detected on the dipstick. This so-called renal threshold corresponds to a serum glucose of about 180 mg/dL; above this level, glucose will be detected in the urine.

Glucose detection with the urinary dipstick is based on a double sequential enzymatic reaction yielding a colorimetric change. In the first reaction, glucose in the urine reacts with glucose oxidase on the dipstick to form gluconic acid and hydrogen peroxide. In the second reaction, hydrogen peroxide reacts with peroxidase, causing oxidation of the chromogen on the dipstick, producing a color change. This doubleoxidative reaction is specific for glucose, and there is no cross-reactivity with other sugars. The dipstick test becomes less sensitive as the urine increases in specific gravity and temperature.

Ketones are not normally found in the urine but will appear when the carbohydrate supplies in the body are depleted and body fat breakdown occurs. This happens most commonly in diabetic ketoacidosis but may also occur during pregnancy and after periods of starvation or rapid weight reduction. Ketones excreted include acetoacetic acid, acetone, and β-hydroxybutyric acid. With abnormal fat breakdown, ketones will appear in the urine before the serum.

Dipstick testing for ketones involves a colorimetric reaction: sodium nitroprusside on the dipstick reacts with acetoacetic acid to produce a purple color. Dipstick testing will identify acetoacetic acid at concentrations of 5 to 10 mg/dL but will not detect acetone or β-hydroxybutyric acid. Obviously, a dipstick that tests positively for glucose should also be tested for ketones, and diabetes mellitus is suggested. False-positive results, however, can occur in very acidic urine of high specific gravity, in abnormally colored urine, and in urine containing levodopa metabolites, 2-mercaptoethane sulfonate sodium, and other sulfhydryl-containing compounds ( Csako, 1987 ).

Bilirubin and Urobilinogen

Normal urine contains no bilirubin and only very small amounts of urobilinogen. There are two types of bilirubin, direct (conjugated) and indirect. Direct bilirubin is made in the hepatocyte, where bilirubin is conjugated with glucuronic acid. Conjugated bilirubin has a low molecular weight, is water soluble, and normally passes from the liver into the small intestine through the bile ducts, where it is converted to urobilinogen. Therefore, conjugated bilirubin does not appear in the urine except in pathologic conditions in which there is intrinsic hepatic disease or obstruction of the bile ducts.

Indirect bilirubin is of high molecular weight and bound in the serum to albumin. It is water insoluble and, therefore, does not appear in the urine even in pathologic conditions.

Urobilinogen is the end product of conjugated bilirubin metabolism. Conjugated bilirubin passes through the bile ducts, where it is metabolized by normal intestinal bacteria to urobilinogen. Normally, about 50% of the urobilinogen is excreted in the stool and 50% reabsorbed into the enterohepatic circulation. A small amount of absorbed urobilinogen, about 1 to 4 mg/day, will escape hepatic uptake and be excreted in the urine. Hemolysis and hepatocellular diseases that lead to increased bile pigments can result in increased urinary urobilinogen. Conversely, obstruction of the bile duct or antibiotic usage that alters intestinal flora, thereby interfering with the conversion of conjugated bilirubin to urobilinogen, will decrease urobilinogen levels in the urine. In these conditions, obviously, serum levels of conjugated bilirubin rise.

There are different dipstick reagents and methods to test for both bilirubin and urobilinogen, but the basic physiologic principle involves the binding of bilirubin or urobilinogen to a diazonium salt to produce a colorimetric reaction. False-negative results can occur in the presence of ascorbic acid, which decreases the sensitivity for detection of bilirubin. False-positive results can occur in the presence of phenazopyridine because it colors the urine orange and, similar to the colorimetric reaction for bilirubin, turns red in an acid medium.

Leukocyte Esterase and Nitrite Tests

Leukocyte esterase activity indicates the presence of white blood cells in the urine. The presence of nitrites in the urine is strongly suggestive of bacteriuria. Thus, both of these tests have been used to screen patients for UTIs. Although these tests may have application in nonurologic medical practice, the most accurate method to diagnose infection is by microscopic examination of the urinary sediment to identify pyuria and subsequent urine culture. All urologists should be capable of performing and interpreting the microscopic examination of the urinary sediment. Therefore, leukocyte esterase and nitrite testing are less important in a urologic practice. For purposes of completion, however, both techniques are described briefly herein.

Leukocyte esterase and nitrite testing are performed using the Chemstrip LN dipstick. Leukocyte esterase is produced by neutrophils and catalyzes the hydrolysis of an indoxyl carbonic acid ester to indoxyl ( Gillenwater, 1981 ). The indoxyl formed oxidizes a diazonium salt chromogen on the dipstick to produce a color change. It is recommended that leukocyte esterase testing be done 5 minutes after the dipstick is immersed in the urine to allow adequate incubation ( Shaw et al, 1985 ). The sensitivity of this test subsequently decreases with time because of lysis of the leukocytes. Leukocyte esterase testing may also be negative in the presence of infection, because not all patients with bacteriuria will have significant pyuria. Therefore, if one uses leukocyte esterase testing to screen patients for UTI, it should always be done in conjunction with nitrite testing for bacteriuria ( Pels et al, 1989 ).

Other causes of false-negative results with leukocyte esterase testing include increased urinary specific gravity, glycosuria, presence of urobilinogen, medications that alter urine color, and ingestion of large amounts of ascorbic acid. The major cause of false-positive leukocyte esterase tests is specimen contamination.

Nitrites are not normally found in the urine, but many species of gram-negative bacteria can convert nitrates to nitrites. Nitrites can readily be detected in the urine because they react with the reagents on the dipstick and undergo diazotization to form a red azo dye. The specificity of the nitrite dipstick for detecting bacteriuria is over 90% ( Pels et al, 1989 ). The sensitivity of the test, however, is considerably less, varying from 35% to 85%. The nitrite test is less accurate in urine specimens containing fewer than 10⁵ organisms/mL ( Kellog et al, 1987 ). As with leukocyte esterase testing, the major cause of false-positive nitrite testing is contamination.

It remains controversial whether dipstick testing for leukocyte esterase and nitrites can replace microscopy in screening for significant UTIs. This issue is less important to urologists, who usually have access to a microscope and who should be trained and encouraged to examine the urinary sediment. A protocol combining the visual appearance of the urine with leukocyte esterase and nitrite testing has been proposed ( Fig. 3-10 ) that reportedly detects 95% of infected urine specimens and decreases the need for microscopy by as much as 30% ( Flanagan et al, 1989 ). Other studies, however, have shown that dipstick testing is not an adequate replacement for microscopy ( Propp et al, 1989 ). In summary, it has not been demonstrated conclusively that dipstick testing for UTI can replace microscopic examination of the urinary sediment. In our personal experience, we always examine the urinary sediment whenever we suspect a UTI and subsequently culture the urine when pyuria is identified.

Figure 3-10 Protocol for determining the need for urine sediment microscopy in an asymptomatic population. (From Flanagan PG, Rooney PG, Davies EA, Stout RW: Evaluation of four screening tests for bacteriuria in elderly people. Lancet 1989;1:1117. © by The Lancet Ltd., 1989.)

Urinary Sediment Obtaining and Preparing the Specimen

A clean-catch midstream urine specimen should be obtained. As described earlier, uncircumcised men should retract the prepuce and cleanse the glans penis before voiding. It is more difficult to obtain a reliable clean-catch specimen in females because of contamination with introital leukocytes and bacteria. If there is any suspicion of a UTI in a female, a catheterized urine sample should be obtained for culture and sensitivity.

If possible, the first morning urine specimen is the specimen of choice and should be examined within 1 hour. A standard procedure for preparation of the urine for microscopic examination has been described ( Cushner and Copley, 1989 ). Ten to 15 milliliters of urine should be centrifuged for 5 minutes at 3000 rpm. The supernatant is then poured off, and the sediment is resuspended in the centrifuge tube by gently tapping the bottom of the tube. Although the remaining small amount of fluid can be poured onto a microscope slide, this usually results in excess fluid on the slide. It is better to use a small pipette to withdraw the residual fluid from the centrifuge tube and to place it directly on the microscope slide. This usually results in an ideal volume of between 0.01 and 0.02 mL of fluid deposited on the slide. The slide is then covered with a coverslip. The edge of the coverslip should be placed on the slide first to allow the drop of fluid to ascend onto the coverslip by capillary action. The coverslip is then gently placed over the drop of fluid, and this technique allows for most of the air between the drop of fluid and the coverslip to be expelled. If one simply drops the coverslip over the urine, the urine will disperse over the slide and there will be a considerable number of air bubbles that may distort the subsequent microscopic examination.

Microscopy Technique

Microscopic analysis of the urinary sediment should be performed with both low-power (×100 magnification) and high-power (×400 magnification) lenses. The use of an oil immersion lens for higher magnification is seldom, if ever, necessary. Under low power, the entire area under the cover-slip should be scanned. Particular attention should be given to the edges of the coverslip, where casts and other elements tend to be concentrated. Low-power magnification is sufficient to identify erythrocytes, leukocytes, casts, cystine crystals, oval fat macrophages, and parasites such as Trichomonas vaginalis and Schistosoma hematobium.

High-power magnification is necessary to distinguish circular from dysmorphic erythrocytes, to identify other types of crystals, and, particularly, to identify bacteria and yeast. In summary, the urinary sediment should be examined microscopically for (1) cells, (2) casts, (3) crystals, (4) bacteria, (5) yeast, and (6) parasites.

Cells

Erythrocyte morphology may be determined under high-power magnification. Although phase contrast microscopy has been used for this purpose, circular (nonglomerular) erythrocytes can generally be distinguished from dysmorphic (glomerular) erythrocytes under routine brightfield high-power magnification (Figs. 3-11 to 3-15 [11] [12] [13] [14] [15]). This is facilitated by adjusting the microscope condenser to its lowest aperture, thus reducing the intensity of background light. This allows one to see fine detail not evident otherwise and also creates the effect of phase microscopy because cell membranes and other sedimentary components stand out against the darkened background.

Figure 3-15 Dysmorphic red blood cells from a patient with Wegener's granulomatosis. A, Brightfield illumination. B, Phase illumination. Note irregular deposits of dense cytoplasmic material around the cell membrane.

Figure 3-14 Red blood cells from a patient with Berger's disease. Note variations in membranes characteristic of dysmorphic red blood cells.

Figure 3-13 Red blood cells from a patient with interstitial cystitis. Cells were collected at cystoscopy.

Figure 3-12 Red blood cells from a patient with a bladder tumor.

Figure 3-11 Red blood cells, both smoothly rounded and mildly crenated, typical of epithelial erythrocytes.

Circular erythrocytes generally have an even distribution of hemoglobin with either a round or a crenated contour, whereas dysmorphic erythrocytes are irregularly shaped with minimal hemoglobin and irregular distribution of cytoplasm. Automated techniques for performing microscopic analysis to distinguish the two types of erythrocytes have been investigated but have not yet been accepted into general urologic practice and are probably unnecessary. In one study using a standard Coulter counter, microscopic analysis was found to be 97% accurate in differentiating between the two types of erythrocytes ( Sayer et al, 1990 ). Erythrocytes may be confused with yeast or fat droplets ( Fig. 3-16 ). Erythrocytes can be distinguished, however, because yeast will show budding and oil droplets are highly refractile.

Figure 3-16 Candida albicans. Budding forms surrounded by leukocytes.

Leukocytes can generally be identified under low power and definitively diagnosed under high-power magnification (Figs. 3-17 and 3-18 [17] [18]; see also Fig. 3-16 ). It is normal to find 1 or 2 leukocytes/HPF in men and up to 5/HPF in women in whom the urine sample may be contaminated with vaginal secretions. A greater number of leukocytes generally indicates infection or inflammation in the urinary tract. It may be possible to distinguish old leukocytes, which have a characteristic small and wrinkled appearance and which are commonly found in the vaginal secretions of normal women, from fresh leukocytes, which are generally indicative of urinary tract pathology. Fresh leukocytes are generally larger and rounder, and, when the specific gravity is less than 1.019, the granules in the cytoplasm demonstrate glitter-like movement, so-called glitter cells.

Figure 3-18 Fresh “glitter cells” with erythrocytes in background.

Figure 3-17 Old leukocytes. Staghorn calculi with Proteus infection.

Epithelial cells are commonly observed in the urinary sediment. Squamous cells are frequently detected in female urine specimens and are derived from the lower portion of the urethra, the trigone of postpubertal females, and the vagina. Squamous epithelial cells are large, have a central small nucleus about the size of an erythrocyte, and have an irregular cytoplasm with fine granularity.

Transitional epithelial cells may arise from the remainder of the urinary tract ( Fig. 3-19 ). Transitional cells are smaller than squamous cells, have a larger nucleus, and demonstrate prominent cytoplasmic granules near the nucleus. Malignant transitional cells have altered nuclear size and morphology and can be identified with either routine Papanicolaou staining or automated flow cytometry.

Figure 3-19 Transitional epithelial cells from bladder lavage.

Renal tubular cells are the least commonly observed epithelial cells in the urine but are most significant, because their presence in the urine is always indicative of renal pathology. Renal tubular cells may be difficult to distinguish from leukocytes, but they are slightly larger.

Casts

A cast is a protein coagulum that is formed in the renal tubule and traps any tubular luminal contents within the matrix. Tamm-Horsfall mucoprotein is the basic matrix of all renal casts; it originates from tubular epithelial cells and is always present in the urine. When the casts contain only mucoproteins, they are called hyaline casts and may not have any pathologic significance. Hyaline casts may be seen in the urine after exercise or heat exposure but may also be observed in pyelonephritis or chronic renal disease.

Red blood cell casts contain entrapped erythrocytes and are diagnostic of glomerular bleeding, most likely secondary to glomerulonephritis (Figs. 3-20 and 3-21 [20] [21]). White blood cell casts are observed in acute glomerulonephritis, acute pyelonephritis, and acute tubulointerstitial nephritis. Casts with other cellular elements, usually sloughed renal tubular epithelial cells, are indicative of nonspecific renal damage ( Fig. 3-22 ). Granular and waxy casts result from further degeneration of cellular elements. Fatty casts are seen in nephrotic syndrome, lipiduria, and hypothyroidism.

Figure 3-21 Red blood cell cast.

Figure 3-20 Red blood cell cast. A, Low-power view demonstrates distinct border of hyaline matrix. B, High-power view demonstrates the sharply defined red blood cell membranes (arrow). Berger's disease.

Figure 3-22 Cellular cast. Cells entrapped in a hyaline matrix.

Crystals

Identification of crystals in the urine is particularly important in patients with stone disease, because it may help determine the etiology ( Fig. 3-23 ). Although other types of crystals may be seen in normal patients, the identification of cystine crystals establishes the diagnosis of cystinuria. Crystals precipitated in acidic urine include calcium oxalate, uric acid, and cystine. Crystals precipitated in an alkaline urine include calcium phosphate and triple-phosphate (struvite) crystals. Cholesterol crystals are rarely seen in the urine and are not related to urinary pH. They occur in lipiduria and remain in droplet form.

Figure 3-23 Urinary crystals. A, Cystine. B, Calcium oxalate. C, Uric acid. D, Triple phosphate (struvite).

Bacteria

Normal urine should not contain bacteria; and in a fresh uncontaminated specimen, the finding of bacteria is indicative of a UTI. Because each HPF views between 1/20,000 and 1/50,000 mL, each bacterium seen per HPF signifies a bacterial count of more than 20,000/mL. Therefore, 5 bacteria/HPF reflects colony counts of about 100,000/mL. This is the standard concentration used to establish the diagnosis of a UTI in a clean-catch specimen. This level should apply only to women, however, in whom a clean-catch specimen is frequently contaminated. The finding of any bacteria in a properly collected midstream specimen from a male should be further evaluated with a urine culture.

Under high power, it is possible to distinguish various bacteria. Gram-negative rods have a characteristic bacillary shape ( Fig. 3-24 ), whereas streptococci can be identified by their characteristic beaded chains (Figs. 3-25 and 3-26 [25] [26]) and staphylococci can be identified when the organisms are found in clumps ( Fig. 3-27 ).

Figure 3-24 Gram-negative bacilli. Phase microscopy of Escherichia coli.

Figure 3-26 Streptococcal urinary tract infection (Gram's stain).

Figure 3-25 Streptococcal urinary tract infection with typical chain formation (arrow).

Figure 3-27 Staphylococcus aureus in typical clumps (arrow).

Yeast

The most common yeast cells found in urine are Candida albicans. The biconcave oval shape of yeast can be confused with erythrocytes and calcium oxalate crystals, but yeasts can be distinguished by their characteristic budding and hyphae (see Fig. 3-16 ). Yeasts are most commonly seen in the urine of patients with diabetes mellitus or as contaminants in women with vaginal candidiasis.

Parasites

Trichomonas vaginalis is a frequent cause of vaginitis in women and occasionally of urethritis in men. Trichomonads can be readily identified in a clean-catch specimen under low power ( Fig. 3-28 ). Trichomonads are large cells with rapidly moving flagella that quickly propel the organism across the microscopic field.

Figure 3-28 Trichomonad with ovoid shape and motile flagella.

Schistosoma hematobium is a urinary tract pathogen that is not found in the United States but is extremely common in countries of the Middle East and North Africa. Examination of the urine shows the characteristic parasitic ova with a terminal spike.

Expressed Prostatic Secretions

Although not strictly a component of the urinary sediment, the expressed prostatic secretions should be examined in any man suspected of having prostatitis. Normal prostatic fluid should contain few, if any, leukocytes, and the presence of a larger number or clumps of leukocytes is indicative of prostatitis. Oval fat macrophages are found in postinfection prostatic fluid (Figs. 3-29 and 3-30 [29] [30]). Normal prostatic fluid contains numerous secretory granules that resemble but can be distinguished from leukocytes under high power because they do not have nuclei.

Figure 3-30 Oval fat microphage, high-power view. Note the fine secretory granules in the prostatic fluid.

Figure 3-29 Oval fat macrophage. A, High-power view showing doubly refractile fat particles (arrow). B, Phase microscopy of the same specimen (arrow).