of Urine Dipstick, Microscopy, and Culture for the Detection of Bacteriuria in Children


Barbara M. Goldsmith, PhD,



Campos, PhD

The authors compared dipstick tests for leukocyte esterase and nitrite and microscopic examination of urinary sediment with urine culture to assess whether the former tests could reliably rule out bacteriuria in specimens from children. The authors studied urine specimens from 1010 infants and children younger than age 18. 5 colony forming units (cfu)/ml, the sensitivities of leukocyte Compared with culture at ≥10 esterase, nitrite, and microscopic examination of white blood cells (≥5 wbc/hpf) or bacteria (in at least moderate numbers) were 76 percent, 29 percent, 82 percent, and 80 percent, respectively. The specificities of the same tests were 81 percent, 99 percent, 81 percent, and 83 percent, 4 cfu/ml, the sensitivities of the tests were 64 percent, respectively. Compared with culture at ≥10 21 percent, 64 percent, and 59 percent; the specificities were 82 percent, 99 percent, 81 percent, and 83 percent, respectively. The negative predictive values of leukocyte esterase and microscopic examinations of urinary sediment for white blood cells and bacteria were all 95 percent 5 cfu/ml). 4 cfu/ml) or 98 percent (≥10 (≥10 The authors conclude that the leukocyte esterase test is as accurate as sediment microscopy in 4 or 10~


Table 1 compares leukocyte esterase, nitrite, microscopic WBC count, and microscopic examination for bacteria to culture. Of the 66 specimens with >_ 105 cfu/ml, 50 (76%) displayed positive leukocyte esterase results and 54 (82%) revealed ~5 wbc/hpf. Only 19 specimens (29%) yielded a positive nitrite reaction, in contrast with 48 specimens (73%), which showed at least moderate numbers of bacteria on microscopic examination. Analysis of specimens with 104-10~ cfu/ml revealed that 15 out of 35 specimens 215

Downloaded from cpj.sagepub.com at Purdue University on June 3, 2015


Comparison of Leukocyte Esterase,

WBC Count, Nitrite, and

(43%) had positive leukocyte esterase results, and 111 (31 %) had ~!~5 wbc/hpf. Only 2 specimens (6%) demonstrated a positive nitrite reaction, and 12 (34%) exhibited moderate or greater numbers of bacteria upon microscopic examination. The sensitivity, specificity, and predictive values of positive and negative urine test results compared with culture are presented in Table 2. For colony counts 2’:: 105 cfu/ml, the sensitivity and specificity of the leukocyte esterase tests were 76 percent and 811 percent, respectively. The predictive value of a negative result was 98 percent. Observing >_5 wbc/hpf or at least moderate numbers of bacteria in urine sediments was a slightly more sensitive method than the leukocyte esterase test, and the specificities of all three tests were approximately the same. The negative predictive value of observing 104 cfu/ml, the sensitivities of the leukocyte esterase test, and microscopic examinaTABLE 2.

Sensitivity, Specificity,

and Predictive Values

Microscopic Urine Sediment Analysis with Urine Culture

tion for WBCs and bacteria were 64 percent, 64 percent, and 59 percent; the specificities were 82 percent, 81 percent, and 83 percent, respectively. The predictive values of negative results were 95 percent for all three tests. The sensitivity and specificity of the nitrite reaction were 21 percent and 99 percent, respectively and the predictive value of a negative result was 92 percent. Discussion

goal of our study was to determine whether negative leukocyte esterase/nitrite urine dipstick test results were as effective as negative microscopic examinations of urinary sediments in ruling out bacteriuria in pediatric patients. A large number of previous reports have found dipstick testing to be an acceptable alternative to sediment microscopy.2,6,8,9,12-14 Comparison of leukocyte esterase results with WBC counts (~~5 wbc/ hpf) resulted in sensitivities of 88-94% and specificities The

of 90-97%. A smaller number of other reports reached the opposite conclusion. In these studies, sensitivities ranged between 82 percent and 94 percent, but

specificities were only 42-66%.3°5

of Leukocyte Esterase, Nitrite,


Downloaded from cpj.sagepub.com at Purdue University on June 3, 2015


Microscopic Urine Sediment Results

We could find only one report dealing exclusively with pediatric specimens that compared urine dipstick results with microscopic examination of urinary sediments.’ The authors reported a sensitivity and specificity of 91 percent and 98 percent, respectively, when dipstick indicators for protein, occult blood, leukocyte esterase, and nitrite were compared with urine microscopy. They concluded that microscopic examination of urine specimens was unnecessary in asymptomatic pediatric patients when urine dipstick

negative. study, we compared dipstick results urine microscopy with urine culture, the &dquo;gold

tests were

In the


and standard&dquo; for detection of bacteriuria. Because there is lack of agreement as to which colony count is the best predictor of urinary tract infection in children, we compared dipstick and microscopic results to colony counts at both 2’:: 1 04 and 2’:: 1 05 cfu/ml. For specimens yielding colony counts > 105 cfu/ ml, we found lower sensitivities for leukocyte esterase and nitrite (76% and 29%, respectively) than those reported by some others.3,4,8 However, our sensitivities for microscopic detection of WBCs or bacteria in urinary sediments (82% and 80%, respectively) were not appreciably better than those for leukocyte esterase. The explanation for the lower sensitivities we observed is not entirely clear; they may be a result of a higher rate of contaminated specimens. Our study was conducted without the knowledge of our clinical colleagues. Thus, extraordinary effort was not made during specimen collection to avoid periurethral contamination. In addition, specimens from children are more likely to be contaminated than specimens from adults who have been properly instructed in disinfection techniques to be used prior to collection of urine. Contaminated specimens could very well be responsible for cultures exhibiting colony counts >_ 10’ cfu/ml in conjunction with negative leukocyte esterase and nitrite results. The predictive value of a negative result is an excellent indicator for calculating how well a test would perform as a screening tool. Compared with urine colony counts of >10~ and 2’:: 1 05 cfu/ml, the negative predictive values of leukocyte esterase and microscopic examinations of urinary sediment for WBCs and bacteria were all 98 percent (h 105 cfu/ 95 percent (2’:: 104 cfu/ml). Thus, leukocyte esterase and urinary sediment microscopy were equally efficacious in identifying the subset of speci-



or < 105 mens harboring < 104 cfu/ml. Of course, the leukocyte esterase test has no value in ruling out noninfectious diseases of the urinary tract.

When asymptomatic children are being evaluated for urinary tract infection, we recommend the leukocyte esterase test as a screen for infection. Microscopic examination of urine for WBCs and bacteria in these asymptomatic patients is unnec. essary. Specimens yielding negative leukocyte esterase rasults do not have to be cultured, and < specimens yielding positive results should be out~tu~e~. For ~chi1dren displaying symptoms of uri~~J1~ry traèt, iRfeotion; urine cultures should be perthe leukocyte esterase re~~‘.s~t=~ A com plete urinalysis; ’including microscopic is indicated when positive results for

fl§8$iti,dd .r8gardleSs .of ::~:e?câfl:Bin~tiOl); ~~~t~~r~~cli~~tl~l~c.’t~~t~‘~~:~., protein, blood, ketones) )4iflfO b$]ain8d.; ... &dquo; ..... -


complete urinalysis, including dipstick testing

and urinary sediment microscopy, is time-consuming and costly. The College of American Pathologists workload recording system credits laboratories with 6.0 min for urinary sediment microscopy with the dipstick test and 4.0 min for a dipstick test without microscopy. Assuming a medical technologist’s salary to be $12.00/hr, each unnecessary microscopic examination would cost $0.40 for labor alone. A laboratory performing 100 examinations per weekday would save approximately $10,000 per year in salary. Looking at the savings more realistically, the laboratory would gain time equivalent to 0.4 medical technologists to use for other testing. Our findings indicate that the leukocyte esterase test is just as accurate as urinary sediment microscopy in identifying specimens containing < 104 or < 105

cfu/ml. References 1. Christoffel



Campbell WD,

Kohrt AE. The abnormal uri-

nalysis in pediatric practice: assuring proper management. QRB 1978;4:5-10. Valenstein PN, Koepke JA. Unnecessary microscopy in routine urinalysis. Am J Clin Pathol 1984;82:444-8.

3. Sewell DL, Burt SP, Gabbert NJ, et al. Evaluation of the Chemstrip 9™ as a screening test for urinalysis and urine culture in men. J Clin Pathol 1985;83:740-3. 4. Christenson RH, Tucker JA, Allen E. Results of dipstick tests, visual inspection, microscopic examination of urine sedi-


Downloaded from cpj.sagepub.com at Purdue University on June 3, 2015

ment, and

5. 6.

microbiological cultures of urine compared for simplifying urinalysis. Clin Chem 1985;31:448-50. Morrison MC, Lum G. Dipstick testing of urine—can it replace urine microscopy? Am J Clin Pathol 1986;85:590-4. Sheer DW. The detection of leukocyte esterase activity in urine with a new reagent strip. Am J Clin Pathol



Analytical goals and clinical relevance of laboratory procedures. In: Tietz NW, ed. Fundamentals of clinical chemistry. 3rd ed. Philadelphia: WB Saunders,

11. Galen RS, Peters T.



7. Doern GV, Saubolle MA, Sewell DL. Screening for bacteriuria with the LN strip test. Diagn Microbiol Infect Dis

1986;4:355-8. 8. Loo SYT, Scottolini AG, Luangphinith S, of


urine-screening protocol.

Diagnosis and management of acute urinary infections in infants and children. Pediatr Infect DisJ

10. McCracken GH.

al. Performance AmJ Clin Pathol et

1986;85:479-84. 9. Hamoudi AC, Bubis SC, Thompson C. Can the cost savings eliminating urine microscopy in biochemically negative urines be extended to the pediatric population? Am J Clin Pathol 1986;86:658-60.

Use of Blood

Penelope JG, Kunin CM. Rapid detection of pyuria by leukocyte esterase activity. JAMA 1981;

12. Kusumi RK,

245:1653-5. 13. Chu SY, Macleod JE, Aterman K. The use of BMC Chemstrip 9 in the macroscopic urine screening procedure. Clin Biochem 1984;17:249-52. 14. Mariani

AJ, Luangphinith S, Loo S, Scottolini A, Hodges CV. Dipstick chemical urinalysis: an accurate cost-effective screening test. J Urol 1984; 132:64-6.


The FDA November 1988 Drug Bulletin discussed recommendations for the transfusion of red cells to increase oxygen-carrying capacity in anemic patients. Adequate oxygen-carrying capacity can be met by a hemoglobin of 7 g/dl or even less when the intravascular volume is adequate for perfusion. In deciding whether to transfuse red cells, the physician should consider the age of the patient, etiology and degree of anemia, hemodynamic stability, and presence of coexisting cardiac, pulmonary, or vascular conditions. When a treatable cause of anemia can be identified, specific replacement therapy (e.g., vitamin B12, iron, or folate) should be tried before transfusion is considered. If volume expanders are indicated, fluids such as crystalloid or non-blood colloid solutions should be administered. Red blood cells should not be transfused for volume expansion, in place of a hematinic, to enhance wound

healing, to improve &dquo;general well-being.&dquo; Platelets may be transfused to control function. In the average adult recipient,


prevent bleeding associated with deficiencies in platelet number


unit of platelet


increases the


platelet count by at least

5,000 platelets/ JLI.

Prophylactic platelet transfusion may be indicated to prevent bleeding in patients with severe thrombocytopenia. For the clinically stable patient with an intact vascular system and normal platelet function, prophylactic platelet transfusion may be indicated for platelet counts of less than 10,000-20,000/ILI. A patient undergoing an operation or other invasive procedure is unlikely to benefit from prophylactic platelet transfusion if the platelet count is at least 50,000/ ILl and thrombocytopenia is the sole abnormality. Platelet transfusions at higher platelet counts may be required for patients with systemic bleeding and for patients at higher risk of bleeding due to additional coagulation defects, sepsis, or platelet dysfunction related to medication or disease. Platelets should not be transfused

to patients with immune thrombocytopenic purpura (unless there is lifethreatening bleeding), prophylactically with massive blood transfusion, or prophylactically following cardiopulmonary bypass. Fresh frozen plasma (FFP) may be transfused to increase the level of clotting factors in patients with a demonstrated deficiency. In the average adult, each unit of FFP transfused will increase the level of any clotting factor’by 2-3%. Laboratory tests should be used to monitor patients with suspected clotting disorders. FFP transfusion is rarely indicated if prothrombin time and partial thromboplastin time are less than 1.5 times normal. Patients deficient in vitamin K-dependent coagulation factors because of warfarin therapy and who are

bleeding or who require emergency surgery may be candidates for FFP transfusion when time does not permit warfarin reversal by stopping the drug or administering vitamin K. Patients with rare conditions such as antithrombin III deficiency and thrombotic thrombocytopenic purpura may also benefit from FFP transfusion. FFP should not be transfused for volume expansion, as a nutritional supplement; prophylactically with massive blood transfusion; or prophylactically following cardiopulmonary bypass.-FDA Drug Bulletin. July 1989. 218

Downloaded from cpj.sagepub.com at Purdue University on June 3, 2015

Comparison of urine dipstick, microscopy, and culture for the detection of bacteriuria in children.

The authors compared dipstick tests for leukocyte esterase and nitrite and microscopic examination of urinary sediment with urine culture to assess wh...
412KB Sizes 0 Downloads 0 Views