Originalia W. Graninger, D. Fleischmann,
B. S c h n e e w e i B , L. A r a m , F. S t o c k e n h u b e r
Rapid Screening for Bacteriuria in Pregnancy Summary: We evaluated a bioluminescence assay as a screening test for the detection of bacteriuria in pregnancy. A total of 1,000 urine specimens from a randomly selected group of pregnant women undergoing prenatal surveillance was investigated. Sequential dilution of urine specimens on C L E D agar plates served as a reference method. Set against the reference group, bioluminescence screening scored a 93% sensitivity, a 78% specificity and a 99% predictive accuracy for negative results. All urine specimens were also analysed chemically for the presence of nitrite and leucocyte esterase by dip sticks. Dip sticks proved to be insufficient because of poor sensitivities of 54% and 59%, respectively. The biotuminescence assay is an effective, time- as well as labor-saving but questionably cost-effective method for the detection of bacteriuria in pregnancy. Zusammenfassung: Rascher Nachweis der Bakteriurie bei Schwangeren. In der vorliegenden Studie priiften wir die klinische Wertigkeit eines BiolumineszenzVerfahrens zum Nachweis der Bakteriurie bei Schwangeren. Insgesamt wurden 1000 RoutineHarnproben von Schwangeren untersucht. Als Referenzmethode diente die sequentielle Verdfinnung der Harnproben und Ausstreichen auf CLEDAgarplatten. Eine signifikante Bakteriurie konnte mit der Biolumineszenz-Methode mit einer Sensitivitfit von 93% und einer Spezifit/it von 78% erkannt werden. Die Voraussagewahrscheinlichkeit ftir negative Resultate war 99%. Gleichzeitig untersuchten wir alle Harnproben auf Vorhandensein von Nitrit und Leukozyten-Esterase mittels Teststreifen. Die Sensitivit/iten lagen nur bei 54% bzw. 59%. Mit Hilfe der Biolumineszenz-Methode kann eine signifikante Bakteriurie bei Schwangeren rasch, verl~iBlich, aber nur bei relevanter Probenanzahl auch kostengfinstig nachgewiesen werden. Introduction In the thirty years, since E. Kass [1] emphasized the relationship of asymptomatic bacteriuria during pregnancy to acute antepartum pyelonephritis, screening for bacteriuria has become an important aspect of prenatal care [2,3]. The most common reference method for bacteriuria is conventional urine culture with a cut-off limit of 105 cfu/ml [4,5]. This method is time consuming and patients are often treated before results are available. Therefore a fast and reliable screening method is required. The bioluminescence assay of bacterial
adenosine triphosphate (ATP) presents a rapid screening method with the ability to detect bacteria in urine samples from hospitals and health-care centers [6-10]. In the present study we investigated the bioluminescence assay as to its clinical usefulness for the detection of significant bacteriuria in pregnant women undergoing prenatal surveillance. As the exclusion of negative specimens from further labour intensive processing is a major aim of this screening method, we paid special attention to the predictive value of negative results obtained. The purpose of the study was to investigate the diagnostic efficacy of the bioluminescence assay compared to conventional urine culture as well as to dip sticks indicating the presence of nitrite and leucocyte esterase.
Materials and Methods Patients: A total of 1,000 urine specimens were processed,
randomly selected from pregnant women attending a prenatal outpatient service. None of the patients had received antimicrobial treatment before attending the service. Urine sampling: Urine specimens were collected as clean void, midstream samples. All specimens were processed within 30 min of micturition. Bacterial cultures: Bacterial cultures were done by sequential dilution of urine samples on CLED agar plates. The plates were read after 18 h at 35°C. The quantity and type of bacteria were recorded. Bacteria were identified by standard laboratory techniques [5]. Bioluminescence: Measurement of viable microbial count was performed with the Lumac/3M Bacteriuria Screening Kit® (3M Company, St. Paul, Minn., USA) and the Lumac M2010 Biocounter®. The test procedure consists of selectively releasing the ATP from nonmicrobial cells (red and white blood cells, epithelial, etc.) using a patented reagent (NRS®). The nonmicrobial ATP is hydrolyzed by a calcium-activated ATP-ase enzyme (Somase® reagent). After a 25 rain incubation period the sample is placed in the counting chamber of the single photon counter (Biocounter~). The subsequent release of microbial ATP by NRB ® reagent is assayed by the automatic addition of purified firefly luciferin-luciferase reagent (Lumit® PM) and by measuring the light emitted. The measurement is in relative light units (RLU). Specimens containing 200 or more RLU were considered positive. Results were obtained in less than 1 h. Dips Sticks: All urine specimens were also analysed by dip sticks, using the Multistix ® 10 GE, Miles Laboratories Inc., Great Britain. We screened for the presence of nitrite and leucocyte esterase. Each stick was placed into the urine sample, excess Received: 12 August 1991/Accepted: 18 September 1991 Prof. Dr. W. Graninget, Dr. D. Fleischrnann, Dr. Laleh Aram, Universitgtsklinik ftir Chemotherapie; Doz. Dr. B. Schneeweifl, Dr. F. Stockenhuber, I. Medizinische Universitfitsklinik, Lazarettgasse 14, A-1090 Wien, Austria.
Infection 20 (1992) No. 1 © MMV Medizin Verlag GmbH M/.inchen, Mfinchen 1992
9 / 13
W. Graninger et al.: Screening for Bacteriuria in Pregnancy Table 1: Bacterial spectrum of urine specimen with significant bacteriuria detected by conventional urine culture (_> 105 cfu/ml) and bioluminescence (> 200 RLU).
Escherichia coli Enterococcus faecalis Staphylococcus epidermidis Klebsiella pneumoniae Proteus mirabilis
115
109
5 3 2 1
4 2 1 1
Total
126
117
Table 2: Comparison of the detection of bacteria in 1,000 urine samples by conventional culture and the bioluminescence method.
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria > 10~cfu/ml (n = 126)
681 9~)
193~) 117
1) No. of false-positive results, z~ No, of false-negative results. Sensitivity-: 117/126 = 0.93, Specificity: 681/874 = 0,78. Negative predictive value: 681/690 = 0,99.
urine removed and the development of the indicator chemicals was allowed to be produced. Statistical evaluation: Sensitivity, specificity and predictive accuracy for negative results were computed according to standard statistical methods [11]. Results
Using conventional bacterial cultures 126 out of 1,000 urine specimens had a count of 10~ cfu/ml or more and thus were regarded as significant bacteriuria. The bacterial spectrum of routine culture findings is given in TaNe 1. Using the bioluminescence-based ATP assay we were able to detect significant bacteriuria (i.e. >--200 RLU) in 117 of the above 126 urine specimens. In 193 samples emitting 200 or more RLU by the bioluminescence assay no significant bacteriuria was found by conventional urine culture. The sensitivity of the bioluminescence method was found to be 93%, specificity was 78%. The predictive accuracy for negative results amounted to 99%. False-negative results were only found in 0.9% of the cases and false-positive readings in 19% (Table 2). Sensitivities of the dip stick reactions were comparably low for nitrite (54%) and leucocyte esterase ( 5 9 % ) a l o n e (Tables 3,4). Sensitivity, however, increased up to 79% when the two tests were combined (Table 5). Specificities were between 87% for the leucocyte esterase test and 100% for the nitrite test (Tables 3,4,5). Discussion
Our results confirm the diagnostic efficiency of the bioluminescence assay of bacterial ATP also in prenatal surveillance. Many investigators have studied the sensitivity and specificity of the bioluminescence assay [6,7,10]. The diagnostic efficiency of the existing screening methods has been studied by HaUander et al. [6], whose study showed that the ATP method had the best diagnostic efficiency followed by bacterial count in sediment, nitrite test, dip slide test, white cell count in sediment and granulocyte esterase test. Our results correspond well to those published to date in the literature, where the sensitivity of bioluminescence is indicated between 97% to 92%, the figure for specificity is 14 / 10
91% to 67% and the predictive value for negative results is given at a 97% level [12-16]. Because of its higher negative predictive value, bioluminescence is more effective in determining which sample should be further processed conventionally than are dip sticks. Possible causes of false-positive results obtained with bioluminescence could be attributable to bacteria which do not grow properly in conventional media or to the presence of antibiotics in the urine specimen that prevent the growth of bacteria in the culture medium. The most important reason could be insufficient destruction of somatic cells, e.g. urine granulocytes and erythrocytes or epithelial cells [17]. False-negative results may be mainly due to threshhold levels in bacterial concentration (104-106 cfu/ml), to the presence of antibiotics impairing bacterial metabolism or to inadequate storage of the urine specimen [8]. Chapelle and Levin [18] indicated the difference in average bacterial ATP concentrations of various bacterial species. Several working groups [8,19,20] all doing clinical assays with bioluminescence reported a higher rate of false-negative results with Staphylococcus spp., Enterobacter aerogenes and Pseudomonas spp. These pathogens have a much lower ATP concentration, which might explain the difficulty of detecting them with the bioluminescence technique. The major advantage of the bioluminescence screening method is the early availability of results - even during consultation with a pregnant outpatient. Conventional methods for the detection of bacteriuria entail the quantitative or semiquantitative culture of urine Table 3: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating leucocyte esterase.
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria --> 105 cfu/ml (n = 126)
762 522)
112~) 74
No. of false-positive results. 2) No. of false-negative results. SensitMty: 74/126 = 0,59, Specificity: 762/874 = 0,87, Negative predictive value: 762/814 = 0.94.
Infection 20 (1992) No. 1 © MMV MedizinVerlag GmbH Mfinchen,Mfinchen1992
W. Graninger et al.: Screening for Bacteriuria in Pregnancy
Table 4: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating nitrite.
Table 5: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating leucocyte esterase and nitrite.
! ii iii! i i i ii !ii i i i! iii i i !ii! i!ii ii iii! !ii! i!il!i i! ii iiiilili! Ii!iiiii!ii ii iii i!iiil;!ii iii!iiiiiiii i i!i!iii!ii:i!!!iiiiii i!i!!iii!i!ii!4 ,i! i!!iyiiiiii!Niiiiiiii;,iNi!!i!?ii iii!i!ii:!ii)!i!iii!iNiiNi!iNiiiii2!ii?ii!iiiiiii Bacteriuria < 10s cfu/ml (n = 874) Bacteriuria --> 10s cfu/ml (n = 126)
874 582)
0~/ 68
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria --> 105 cfu/ml (n = 126)
762 27z)
112t) 99
No. of false-positive results. 2)No. of false-negative results. Sensitivity: 68/126 = 0~54. Specificity: 874/874 = 1.0. Negative predictive value: 874/932 = 0.94.
l) No. of false-positive results, z) No. of fi~lse-negative results. Sensitivity: 99/126 = 0.79, Specificity: 762/874 = 0.87. Negative predictive value: 762/814 = 0.94.
s p e c i m e n s on solid c u l t u r e m e d i u m , n e e d o v e r n i g h t i n c u b a t i o n a n d a r e m o r e l a b o u r intensive t h a n t h e b i o l u m i n e s c e n c e assay. T h e m o s t c o m m o n s c r e e n i n g test for b a c t e r i u r i a is t h e dip stick test for t h e p r e s e n c e o f l e u c o c y t e e s t e r a s e a n d nitrite. I n o u r s t u d y d i p sticks p r o v e d i n a d e q u a t e to i n d i c a t e significant b a c t e r i u r i a b e c a u s e o f p o o r sensitivity. Sensitivity, however, i n c r e a s e d u p to 79% w h e n e i t h e r the p r e s e n c e o f nitrite o r leucocyte e s t e r a s e was r e g a r d e d as a p o t e n t i a l i n d i c a t o r o f infection. F u r t h e r m o r e , it is w o r t h m e n t i o n i n g t h a t in o u r investigation t h e nitrite test h a d no false-positive results. T h e b i o l u m i n e s c e n c e assay as a s c r e e n i n g m e t h o d for b a c t e r i u r i a is s u p e r i o r t o d i p sticks a n d far m o r e r a p i d t h a n t h e c o n v e n t i o n a l d i p slide test w h i c h n e e d s overnight incubation.
W h e t h e r b i o l u m i n e s c e n c e s c r e e n i n g for b a c t e r i u r i a is cost effective d e p e n d s m o s t o f all on t h e s p e c i m e n q u o t a a n d staff costs [12]. T h e costs for a single assay a r e h i g h e r c o m p a r e d to c o n v e n t i o n a l b a c t e r i o l o g i c a l m e t h o d s . M i n o r costs for staff a n d c u l t u r e m e d i a , as well as initial c a p i t a l plus r e g u l a r m a i n t e n a n c e , a r e all factors to b e c a l c u l a t e d . T h e m e t h o d d o e s n o t s e e m to b e s u i t a b l e for small institutions. I n p r e n a t a l c a r e centers, h o w e v e r , w i t h t h e s p e c i m e n q u o t a e x c e e d i n g 50 u r i n e s a m p l e s p e r day, the b i o l u m i n e s c e n c e assay o f b a c t e r i a l A T P m i g h t b e a r a p i d , r e l i a b l e a n d inexpensive s c r e e n i n g test for t h e d e t e c t i o n of bacteriuria.
References
11. Benfari Ferraro, M. J., Kunz, L J.: Predictive value of microbiological diagnostic tests. In: Lorian, V. (ed.): Significance of medical microbiology in the care of patients. 2nd ed. The Williams and Witkins Co., Baltimore 1982, pp. 248-251. 12. Smith, T. K., Hudson, A. J., Spencer, R. C.: Evaluation of six screening methods for detecting significant bacteriuria. J. Clin. Pathol. 41 (1988) 904-909. 13. Bixler-Forell, E., Bertram, M. A., Bruckner, D. A.: Clinical evaluation of three rapid methods for the detection of significant bacteriuria. J. Clin. Microbiol. 22 (1985) 62-67. 14. Males, B. M., Bartholomew, W. IL, Amsterdam, D.: Leukocyte esterase - nitrite and bioluminescence assays as urine screens. J. Clin. MicrobioI. 22 (1985) 531-534. 15. Martin, E. T., Cote, J. A., Perry, L. K., Martin, W. J.: Clinical evaluation of lumac bioluminescence method for screening urine specimens. J. Clin. Microbiol. 22 (1985) 19-22. 16. Szilagyi, G., Aning, V., Karmen, A.: Comparative study of two methods for rapid detection of clinically significant bacteriuria. J. Clin. Lab. Autom. 3 (1983) 117-122. 17. Johnston, H. H., Mitchell, C. J., Curtis, G. D. W.: An automated test for the detection of significant bacteriuria. Lancet i (1976) 400- 402. 18. Chapelle, E. W., Levin, G. V.: Adenosine triphosphate (ATP) concentrations of 19 bacterial species. Biochem. Med. 2 (1968) 41-43. 19. Drow, D. L., Baum, C. H., Hirschfield, G.: Comparison of the lumac and monolight system for detection of bacteriuria by bioluminescence. J. Clin. Microbiol. 20 (1984) 797-801. 20. Welch, W. D., Thompson, L., Layman, M., Southern, P. M.: Evaluation of two bioluminescence-measuring instruments, the turner design and lumac systems, for the rapid screening of urine specimens. J. Clin. Microbiol. 20 (1984) 1165-1170.
1. Kass, E. H.: Bacteriuria and pyelonephritis of pregnancy. Arch. Intern. Med. 105 (1960) 194-201. 2. Sweet, R. L: Bacteriuria and pyelonephritis during pregnancy. Semin. Perinatol. 1 (1977) 25-40. 3. Krieger, J. N.: Complications and treatment of urinary tract infections during pregnancy. Urol. Clin. North Am. 13 (1986) 685-693. 4. Clarridge, J. E., Pezzlo, M. T., Vosti, K. L.: Laboratory diagnosis of urinary tract infections. In: Weissfeld, A. S. (ed.): CUMITECH 2A. American Society for Microbiology, Washington DC, 1987. 5. Lennette, E. H., Baiows, A., I~usler, W. J., Truant, J. P.: Manual of clinical microbiology, 3rd edl., American Society for Microbiology, Washington DC, 1980, pp. 68--70. 6. Hallander, H. O., Kallner, A., Lundin, A., Osterberg, E.: Evaluation of rapid methods for the detection of bacteriuria (screening) in primary health care. Acta Pathol. Microbiol. Immunol. Seand. [B] 94 (1986) 39-49. 7. Thore, A., Ans~hn, S., Lundin, A., Bergman, B.: Detection of bacteriuria by luciferase assay of adenosine triphosphate. J. Clin. Microbiol. 1 (1975) 1-8. 8. Kolbeck, J. C., Padgett, A. R., Estevez, E, G., Harreil, L J.: Bioluminescence screening for bacteriuria. J. Clin. Microbiol. 21 (1985) 527-530. 9. Schiffman, R. B., Wieden, M., Brooker, J., Chery, M., Pelouca, M., Norgard, K., Palen, D., Reis, N., Swanson, J., White, J.: Bacteriuria screening by direct biolumineseence assay of ATP. J. Clin. Microbiol. 2O (1984) 644-648. 10. Gfistrin, B., Gustafson, R., Lundin, A.: Evaluation of a bioluminescence assay for the detection of bacteriuria. Scand. J. Infect. Dis 21 (1989) 409-414.
Infection 20 (1992) No. 1 © MMV Medizin Verlag GmbH Mfinchen, Mfinchen 1992
11 / 15
B. S c h n e e w e i B , L. A r a m , F. S t o c k e n h u b e r
Rapid Screening for Bacteriuria in Pregnancy Summary: We evaluated a bioluminescence assay as a screening test for the detection of bacteriuria in pregnancy. A total of 1,000 urine specimens from a randomly selected group of pregnant women undergoing prenatal surveillance was investigated. Sequential dilution of urine specimens on C L E D agar plates served as a reference method. Set against the reference group, bioluminescence screening scored a 93% sensitivity, a 78% specificity and a 99% predictive accuracy for negative results. All urine specimens were also analysed chemically for the presence of nitrite and leucocyte esterase by dip sticks. Dip sticks proved to be insufficient because of poor sensitivities of 54% and 59%, respectively. The biotuminescence assay is an effective, time- as well as labor-saving but questionably cost-effective method for the detection of bacteriuria in pregnancy. Zusammenfassung: Rascher Nachweis der Bakteriurie bei Schwangeren. In der vorliegenden Studie priiften wir die klinische Wertigkeit eines BiolumineszenzVerfahrens zum Nachweis der Bakteriurie bei Schwangeren. Insgesamt wurden 1000 RoutineHarnproben von Schwangeren untersucht. Als Referenzmethode diente die sequentielle Verdfinnung der Harnproben und Ausstreichen auf CLEDAgarplatten. Eine signifikante Bakteriurie konnte mit der Biolumineszenz-Methode mit einer Sensitivitfit von 93% und einer Spezifit/it von 78% erkannt werden. Die Voraussagewahrscheinlichkeit ftir negative Resultate war 99%. Gleichzeitig untersuchten wir alle Harnproben auf Vorhandensein von Nitrit und Leukozyten-Esterase mittels Teststreifen. Die Sensitivit/iten lagen nur bei 54% bzw. 59%. Mit Hilfe der Biolumineszenz-Methode kann eine signifikante Bakteriurie bei Schwangeren rasch, verl~iBlich, aber nur bei relevanter Probenanzahl auch kostengfinstig nachgewiesen werden. Introduction In the thirty years, since E. Kass [1] emphasized the relationship of asymptomatic bacteriuria during pregnancy to acute antepartum pyelonephritis, screening for bacteriuria has become an important aspect of prenatal care [2,3]. The most common reference method for bacteriuria is conventional urine culture with a cut-off limit of 105 cfu/ml [4,5]. This method is time consuming and patients are often treated before results are available. Therefore a fast and reliable screening method is required. The bioluminescence assay of bacterial
adenosine triphosphate (ATP) presents a rapid screening method with the ability to detect bacteria in urine samples from hospitals and health-care centers [6-10]. In the present study we investigated the bioluminescence assay as to its clinical usefulness for the detection of significant bacteriuria in pregnant women undergoing prenatal surveillance. As the exclusion of negative specimens from further labour intensive processing is a major aim of this screening method, we paid special attention to the predictive value of negative results obtained. The purpose of the study was to investigate the diagnostic efficacy of the bioluminescence assay compared to conventional urine culture as well as to dip sticks indicating the presence of nitrite and leucocyte esterase.
Materials and Methods Patients: A total of 1,000 urine specimens were processed,
randomly selected from pregnant women attending a prenatal outpatient service. None of the patients had received antimicrobial treatment before attending the service. Urine sampling: Urine specimens were collected as clean void, midstream samples. All specimens were processed within 30 min of micturition. Bacterial cultures: Bacterial cultures were done by sequential dilution of urine samples on CLED agar plates. The plates were read after 18 h at 35°C. The quantity and type of bacteria were recorded. Bacteria were identified by standard laboratory techniques [5]. Bioluminescence: Measurement of viable microbial count was performed with the Lumac/3M Bacteriuria Screening Kit® (3M Company, St. Paul, Minn., USA) and the Lumac M2010 Biocounter®. The test procedure consists of selectively releasing the ATP from nonmicrobial cells (red and white blood cells, epithelial, etc.) using a patented reagent (NRS®). The nonmicrobial ATP is hydrolyzed by a calcium-activated ATP-ase enzyme (Somase® reagent). After a 25 rain incubation period the sample is placed in the counting chamber of the single photon counter (Biocounter~). The subsequent release of microbial ATP by NRB ® reagent is assayed by the automatic addition of purified firefly luciferin-luciferase reagent (Lumit® PM) and by measuring the light emitted. The measurement is in relative light units (RLU). Specimens containing 200 or more RLU were considered positive. Results were obtained in less than 1 h. Dips Sticks: All urine specimens were also analysed by dip sticks, using the Multistix ® 10 GE, Miles Laboratories Inc., Great Britain. We screened for the presence of nitrite and leucocyte esterase. Each stick was placed into the urine sample, excess Received: 12 August 1991/Accepted: 18 September 1991 Prof. Dr. W. Graninget, Dr. D. Fleischrnann, Dr. Laleh Aram, Universitgtsklinik ftir Chemotherapie; Doz. Dr. B. Schneeweifl, Dr. F. Stockenhuber, I. Medizinische Universitfitsklinik, Lazarettgasse 14, A-1090 Wien, Austria.
Infection 20 (1992) No. 1 © MMV Medizin Verlag GmbH M/.inchen, Mfinchen 1992
9 / 13
W. Graninger et al.: Screening for Bacteriuria in Pregnancy Table 1: Bacterial spectrum of urine specimen with significant bacteriuria detected by conventional urine culture (_> 105 cfu/ml) and bioluminescence (> 200 RLU).
Escherichia coli Enterococcus faecalis Staphylococcus epidermidis Klebsiella pneumoniae Proteus mirabilis
115
109
5 3 2 1
4 2 1 1
Total
126
117
Table 2: Comparison of the detection of bacteria in 1,000 urine samples by conventional culture and the bioluminescence method.
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria > 10~cfu/ml (n = 126)
681 9~)
193~) 117
1) No. of false-positive results, z~ No, of false-negative results. Sensitivity-: 117/126 = 0.93, Specificity: 681/874 = 0,78. Negative predictive value: 681/690 = 0,99.
urine removed and the development of the indicator chemicals was allowed to be produced. Statistical evaluation: Sensitivity, specificity and predictive accuracy for negative results were computed according to standard statistical methods [11]. Results
Using conventional bacterial cultures 126 out of 1,000 urine specimens had a count of 10~ cfu/ml or more and thus were regarded as significant bacteriuria. The bacterial spectrum of routine culture findings is given in TaNe 1. Using the bioluminescence-based ATP assay we were able to detect significant bacteriuria (i.e. >--200 RLU) in 117 of the above 126 urine specimens. In 193 samples emitting 200 or more RLU by the bioluminescence assay no significant bacteriuria was found by conventional urine culture. The sensitivity of the bioluminescence method was found to be 93%, specificity was 78%. The predictive accuracy for negative results amounted to 99%. False-negative results were only found in 0.9% of the cases and false-positive readings in 19% (Table 2). Sensitivities of the dip stick reactions were comparably low for nitrite (54%) and leucocyte esterase ( 5 9 % ) a l o n e (Tables 3,4). Sensitivity, however, increased up to 79% when the two tests were combined (Table 5). Specificities were between 87% for the leucocyte esterase test and 100% for the nitrite test (Tables 3,4,5). Discussion
Our results confirm the diagnostic efficiency of the bioluminescence assay of bacterial ATP also in prenatal surveillance. Many investigators have studied the sensitivity and specificity of the bioluminescence assay [6,7,10]. The diagnostic efficiency of the existing screening methods has been studied by HaUander et al. [6], whose study showed that the ATP method had the best diagnostic efficiency followed by bacterial count in sediment, nitrite test, dip slide test, white cell count in sediment and granulocyte esterase test. Our results correspond well to those published to date in the literature, where the sensitivity of bioluminescence is indicated between 97% to 92%, the figure for specificity is 14 / 10
91% to 67% and the predictive value for negative results is given at a 97% level [12-16]. Because of its higher negative predictive value, bioluminescence is more effective in determining which sample should be further processed conventionally than are dip sticks. Possible causes of false-positive results obtained with bioluminescence could be attributable to bacteria which do not grow properly in conventional media or to the presence of antibiotics in the urine specimen that prevent the growth of bacteria in the culture medium. The most important reason could be insufficient destruction of somatic cells, e.g. urine granulocytes and erythrocytes or epithelial cells [17]. False-negative results may be mainly due to threshhold levels in bacterial concentration (104-106 cfu/ml), to the presence of antibiotics impairing bacterial metabolism or to inadequate storage of the urine specimen [8]. Chapelle and Levin [18] indicated the difference in average bacterial ATP concentrations of various bacterial species. Several working groups [8,19,20] all doing clinical assays with bioluminescence reported a higher rate of false-negative results with Staphylococcus spp., Enterobacter aerogenes and Pseudomonas spp. These pathogens have a much lower ATP concentration, which might explain the difficulty of detecting them with the bioluminescence technique. The major advantage of the bioluminescence screening method is the early availability of results - even during consultation with a pregnant outpatient. Conventional methods for the detection of bacteriuria entail the quantitative or semiquantitative culture of urine Table 3: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating leucocyte esterase.
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria --> 105 cfu/ml (n = 126)
762 522)
112~) 74
No. of false-positive results. 2) No. of false-negative results. SensitMty: 74/126 = 0,59, Specificity: 762/874 = 0,87, Negative predictive value: 762/814 = 0.94.
Infection 20 (1992) No. 1 © MMV MedizinVerlag GmbH Mfinchen,Mfinchen1992
W. Graninger et al.: Screening for Bacteriuria in Pregnancy
Table 4: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating nitrite.
Table 5: Comparison of the detection of bacteria in 1,000 urine samples by conventional urine culture and dip sticks indicating leucocyte esterase and nitrite.
! ii iii! i i i ii !ii i i i! iii i i !ii! i!ii ii iii! !ii! i!il!i i! ii iiiilili! Ii!iiiii!ii ii iii i!iiil;!ii iii!iiiiiiii i i!i!iii!ii:i!!!iiiiii i!i!!iii!i!ii!4 ,i! i!!iyiiiiii!Niiiiiiii;,iNi!!i!?ii iii!i!ii:!ii)!i!iii!iNiiNi!iNiiiii2!ii?ii!iiiiiii Bacteriuria < 10s cfu/ml (n = 874) Bacteriuria --> 10s cfu/ml (n = 126)
874 582)
0~/ 68
Bacteriuria < 105 cfu/ml (n = 874) Bacteriuria --> 105 cfu/ml (n = 126)
762 27z)
112t) 99
No. of false-positive results. 2)No. of false-negative results. Sensitivity: 68/126 = 0~54. Specificity: 874/874 = 1.0. Negative predictive value: 874/932 = 0.94.
l) No. of false-positive results, z) No. of fi~lse-negative results. Sensitivity: 99/126 = 0.79, Specificity: 762/874 = 0.87. Negative predictive value: 762/814 = 0.94.
s p e c i m e n s on solid c u l t u r e m e d i u m , n e e d o v e r n i g h t i n c u b a t i o n a n d a r e m o r e l a b o u r intensive t h a n t h e b i o l u m i n e s c e n c e assay. T h e m o s t c o m m o n s c r e e n i n g test for b a c t e r i u r i a is t h e dip stick test for t h e p r e s e n c e o f l e u c o c y t e e s t e r a s e a n d nitrite. I n o u r s t u d y d i p sticks p r o v e d i n a d e q u a t e to i n d i c a t e significant b a c t e r i u r i a b e c a u s e o f p o o r sensitivity. Sensitivity, however, i n c r e a s e d u p to 79% w h e n e i t h e r the p r e s e n c e o f nitrite o r leucocyte e s t e r a s e was r e g a r d e d as a p o t e n t i a l i n d i c a t o r o f infection. F u r t h e r m o r e , it is w o r t h m e n t i o n i n g t h a t in o u r investigation t h e nitrite test h a d no false-positive results. T h e b i o l u m i n e s c e n c e assay as a s c r e e n i n g m e t h o d for b a c t e r i u r i a is s u p e r i o r t o d i p sticks a n d far m o r e r a p i d t h a n t h e c o n v e n t i o n a l d i p slide test w h i c h n e e d s overnight incubation.
W h e t h e r b i o l u m i n e s c e n c e s c r e e n i n g for b a c t e r i u r i a is cost effective d e p e n d s m o s t o f all on t h e s p e c i m e n q u o t a a n d staff costs [12]. T h e costs for a single assay a r e h i g h e r c o m p a r e d to c o n v e n t i o n a l b a c t e r i o l o g i c a l m e t h o d s . M i n o r costs for staff a n d c u l t u r e m e d i a , as well as initial c a p i t a l plus r e g u l a r m a i n t e n a n c e , a r e all factors to b e c a l c u l a t e d . T h e m e t h o d d o e s n o t s e e m to b e s u i t a b l e for small institutions. I n p r e n a t a l c a r e centers, h o w e v e r , w i t h t h e s p e c i m e n q u o t a e x c e e d i n g 50 u r i n e s a m p l e s p e r day, the b i o l u m i n e s c e n c e assay o f b a c t e r i a l A T P m i g h t b e a r a p i d , r e l i a b l e a n d inexpensive s c r e e n i n g test for t h e d e t e c t i o n of bacteriuria.
References
11. Benfari Ferraro, M. J., Kunz, L J.: Predictive value of microbiological diagnostic tests. In: Lorian, V. (ed.): Significance of medical microbiology in the care of patients. 2nd ed. The Williams and Witkins Co., Baltimore 1982, pp. 248-251. 12. Smith, T. K., Hudson, A. J., Spencer, R. C.: Evaluation of six screening methods for detecting significant bacteriuria. J. Clin. Pathol. 41 (1988) 904-909. 13. Bixler-Forell, E., Bertram, M. A., Bruckner, D. A.: Clinical evaluation of three rapid methods for the detection of significant bacteriuria. J. Clin. Microbiol. 22 (1985) 62-67. 14. Males, B. M., Bartholomew, W. IL, Amsterdam, D.: Leukocyte esterase - nitrite and bioluminescence assays as urine screens. J. Clin. MicrobioI. 22 (1985) 531-534. 15. Martin, E. T., Cote, J. A., Perry, L. K., Martin, W. J.: Clinical evaluation of lumac bioluminescence method for screening urine specimens. J. Clin. Microbiol. 22 (1985) 19-22. 16. Szilagyi, G., Aning, V., Karmen, A.: Comparative study of two methods for rapid detection of clinically significant bacteriuria. J. Clin. Lab. Autom. 3 (1983) 117-122. 17. Johnston, H. H., Mitchell, C. J., Curtis, G. D. W.: An automated test for the detection of significant bacteriuria. Lancet i (1976) 400- 402. 18. Chapelle, E. W., Levin, G. V.: Adenosine triphosphate (ATP) concentrations of 19 bacterial species. Biochem. Med. 2 (1968) 41-43. 19. Drow, D. L., Baum, C. H., Hirschfield, G.: Comparison of the lumac and monolight system for detection of bacteriuria by bioluminescence. J. Clin. Microbiol. 20 (1984) 797-801. 20. Welch, W. D., Thompson, L., Layman, M., Southern, P. M.: Evaluation of two bioluminescence-measuring instruments, the turner design and lumac systems, for the rapid screening of urine specimens. J. Clin. Microbiol. 20 (1984) 1165-1170.
1. Kass, E. H.: Bacteriuria and pyelonephritis of pregnancy. Arch. Intern. Med. 105 (1960) 194-201. 2. Sweet, R. L: Bacteriuria and pyelonephritis during pregnancy. Semin. Perinatol. 1 (1977) 25-40. 3. Krieger, J. N.: Complications and treatment of urinary tract infections during pregnancy. Urol. Clin. North Am. 13 (1986) 685-693. 4. Clarridge, J. E., Pezzlo, M. T., Vosti, K. L.: Laboratory diagnosis of urinary tract infections. In: Weissfeld, A. S. (ed.): CUMITECH 2A. American Society for Microbiology, Washington DC, 1987. 5. Lennette, E. H., Baiows, A., I~usler, W. J., Truant, J. P.: Manual of clinical microbiology, 3rd edl., American Society for Microbiology, Washington DC, 1980, pp. 68--70. 6. Hallander, H. O., Kallner, A., Lundin, A., Osterberg, E.: Evaluation of rapid methods for the detection of bacteriuria (screening) in primary health care. Acta Pathol. Microbiol. Immunol. Seand. [B] 94 (1986) 39-49. 7. Thore, A., Ans~hn, S., Lundin, A., Bergman, B.: Detection of bacteriuria by luciferase assay of adenosine triphosphate. J. Clin. Microbiol. 1 (1975) 1-8. 8. Kolbeck, J. C., Padgett, A. R., Estevez, E, G., Harreil, L J.: Bioluminescence screening for bacteriuria. J. Clin. Microbiol. 21 (1985) 527-530. 9. Schiffman, R. B., Wieden, M., Brooker, J., Chery, M., Pelouca, M., Norgard, K., Palen, D., Reis, N., Swanson, J., White, J.: Bacteriuria screening by direct biolumineseence assay of ATP. J. Clin. Microbiol. 2O (1984) 644-648. 10. Gfistrin, B., Gustafson, R., Lundin, A.: Evaluation of a bioluminescence assay for the detection of bacteriuria. Scand. J. Infect. Dis 21 (1989) 409-414.
Infection 20 (1992) No. 1 © MMV Medizin Verlag GmbH Mfinchen, Mfinchen 1992
11 / 15
