SEARCH

SEARCH BY CITATION

Keywords:

  • Escherichia coli;
  • Klebsiella spp.;
  • Septicemia;
  • Uriscreen

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References

Background: Bacteremia occurs frequently in newborn calves. The predictive value of clinical signs is low, suggesting the use of calf-side diagnostic tests.

Objectives: To investigate testing of urine catalase activity (Uriscreen test) for bacteriuria and bacteremia detection.

Animals: Five colostrum-free calves and 3 colostrum-fed control calves.

Methods: Controlled experimental trial. Colostrum-free calves were inoculated PO with Escherichia coli O78+. A clinical score was established to define the onset of the illness. Blood and urine (cystocentesis) samplings and cultures, and Uriscreen tests, were performed 4–6 times from inoculation to death. Three control calves received the same management as 3 inoculated calves, but with colostrum and without inoculation.

Results: Bacteremia was demonstrated in all of the inoculated colostrum-free calves and in none of the control calves. The E. coli O78+ strain, E. coli, and Klebsiella spp. were recovered from 4/5, 5/5, and 2/5 inoculated colostrum-free calves, respectively. Urine cultures were negative for the 2 groups at the start of the experiment; 5/5 colostrum-deprived inoculated calves were positive for E. coli thereafter whereas 3/3 controls remained negative. Concordance of Uriscreen tests with bacteremia and bacteriuria was 0.86 and 0.88, respectively. Kappa value of agreement between Uriscreen and bacteremia and bacteriuria was 0.73 and 0.76, respectively. Sensitivity of Uriscreen for bacteremia and bacteriuria was 80.0 and 86.6%, respectively, and specificity was 92.8 and 88.8%, respectively.

Conclusions and Clinical Relevance: The results suggest that Uriscreen can be used for detection of bacteremia in neonatal calves in connection with a constant bacteriuria.

Abbreviations:
CS

clinical score

SD

standard deviation

Septicemia is common in calves during the 1st days of life.1,2 Bacteremia was demonstrated in approximately 20–30% of diarrheic calves during the neonatal period.1 The bacteria isolated during bacteremia or septicemia include Escherichia coli in >50% of calves, and Salmonella spp., Campylobacter spp., Klebsiella spp., and Staphylococcus spp.2 The rational use of antimicrobial treatment in neonatal calves with systemic illness with or without diarrhea depends largely on the presence or absence of bacteremia.1

The early diagnosis of bacteremia in calves is difficult. Clinical signs and scoring have been proposed, but their predictive value remains low.1–3 Blood culture is considered a definitive test, but time and cost limit its practical use.1 In human infants, septicemia is also difficult to diagnose. Leukogram and C-reactive protein seem to be the most accurate criteria for sepsis diagnosis in humans.4 In 27% (16/59) of human adults with Staphylococcus aureus bacteremia, bacteriuria (>105 CFU/mL) was detected within 48 hours after positive blood culture. Bacteriuria appeared as a consequence of bacteremia in two thirds of cases and as the cause in the other one third.5

Detection of urine catalase activity by the Uriscreen test is commonly used in human emergency clinics as the 1st line of urinary tract infection diagnosis. Catalase activity is present in most bacteria and somatic cells. According to the manufacturer, the Uriscreen test positive threshold is 50,000 CFU/mL of urine,6,7 and moderate specificity (45–80%) and good sensitivity (70–100%) are reported in humans.6–10

The aim of the study reported here was to evaluate the possibility of with the Uriscreen test on urine for the early detection of bacteriuria and indirectly bacteremia in neonatal calves.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References

Animals

Eight Holstein male calves were immediately separated from their dams after birth. During the experiment, the calves were housed in hutches with straw on the floor; the ambient temperature was 15°C. They were fed 2 L of a high-energy oral rehydrating solute (Energaida), twice a day, throughout the experiment. The control calves suckled 1 dose (60 mL) of a colostrum substitute (Locatimb) between 2 and 4 hours after birth, before the solute feedings.

Only 1 calf was included in the experiment at a time. All experimental procedures were approved by the local Institutional Animal Care and Use Committee.

Experimental Groups and Inoculation

Five colostrum-deprived calves were inoculated once by oral route with a septicemic E. coli strain O78+ (kindly provided by E. Oswald, UMR INRA 1225, ENV F, Toulouse) before 12 hours of age. The calf suckled the inoculum poured in a bottle. Vir plasmid carriage (conferring septicemic properties) and the dose inoculated (5 × 107 CFU/calf) were retrospectively confirmed on the inoculum. Three control calves were not inoculated.

The calves were numbered from #1 to #8 according to the date of birth, and calves #1, 3, 5, 7, and 8 were allocated to the colostrum-deprived inoculated calves group whereas calves #2, 4, and 6 were used as control calves.

Clinical Evaluation

Clinical signs were evaluated every 2 hours by the same operator. A clinical score (CS) was calculated as described previously.3 Briefly, the CS includes 5 points: rectal temperature (1 if 38.0–39.0°C; 2 if 37.5–38.0°C or 39.0–39.5°C; 3 if below 37.5 or above 39.5°C), thirst, suckling reflex and ability to get up (1 if absent; 2 if decreased; 3 if normal), and general behavior (1 if comatose; 2 if lethargic; 3 if normal). The threshold of the CS associated with the onset of the septicemia was 9/15, as suggested by previous results.3

Sampling

Inoculated calves were sampled at inoculation time, at the beginning of the illness (CS of 9/15), and 3–4 times thereafter, according to the clinical progression of each calf. In actuality, samplings occurred at 4, 12, and 18 hours (calf #1), 6, 10, 16, and 20 hours (calf #3), 6, 12, 18, and 23 hours (calf #5), 4, 6, and 10 hours (calf #7), and 4, 6, and 9 hours (calf #8) after the beginning of the illness. The results were categorized into 4 periods of time with reference to the beginning of the illness (time 0): 4–6 hours, 9–12 hours, 16–18 hours, and 20–23 hours.

The control calves #2, 4, and 6 were sampled at the same time after birth as inoculated calves #1, 3, and 5, respectively, so as to reproduce precisely the samplings that occurred on inoculated calves #1, 3, and 5 at inoculation, at the beginning of the illness and between onset of the illness and death on calves #2, 4, and 6. Fecal samples were collected at inoculation time, at the onset of the illness and at death.

In order to minimize environmental contamination, sampling of blood (jugular vein) and urine (cystocentesis) was carried out by aseptic procedures: hair was clipped, skin was scrubbed with povidone-iodine (Vétédinec) and alcohol 70%, hands were washed, sterile gloves and sterile sampling materials were used.

Cystocentesis was performed with a 18 G, 40 mm needle (Venojectd) attached to a 20 mL syringe (Omnifixe) via transabdominal route with the calf in dorsal recumbency. The puncture for cystocentesis was 2–3 cm cranial to the pubic bone, on the medial white line.

After jugular venipuncture, blood was immediately transferred into Hemolinef bottles containing culture medium with growth factors and anticoagulant.

At death, liver and kidney were collected by aseptic procedures and macroscopic inspection of the bladder was performed.

Microbiology Procedures

Blood samples were incubated for 24–48 hours at 37°C in Hemolinef before plating on blood agarg and Drigalski agar.f

Urine was plated on a Drigalski agar,f and simultaneously cultured on a BUSPf medium for 24–48 hours before plating. E. coli was counted in urine after serial dilution and a 24–48 hours incubation at 37°C. Bacteriuria <102 CFU/mL of urine was considered negative.

Attempts were made to isolate E. coli from fecal, liver, and kidney samples using bloodg and Drigalski agars.f

Identification of bacteria was made 24–48 hours after a 37°C incubation period, according to Gram staining and biochemical results (API 20 E galleryf). In order to identify O78 E. coli, an agglutination testh was performed on 5 E. coli colonies per blood agar plate.

Uriscreen Test

The Uriscreeni test was performed on urine immediately after urine collection, in accordance with the manufacturer's recommendations. A complete ring of foam in 2 minutes was interpreted as a positive test.

Statistics

Concordance between urine bacterial counts (>102 CFU/mL) and Uriscreen or bacteremia and Uriscreen were calculated. Concordance is the ratio of the number of concordant results (between bacteriuria or bacteremia and Uriscreen) to the total number of paired results submitted.

The sensitivity and specificity of the Uriscreen test for bacteremia and bacteriuria, and the interrater agreement between Uriscreen and bacteremia or bacteriuria, were calculated.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References

Inoculation, Clinical Signs, and Death

The CS of the 8 calves was 15/15 at inoculation time. Inoculation occurred on average 10 hours after birth (standard deviation [SD] = 2 hours) for the inoculated group. The mean duration between inoculation and the start of clinical signs (time 0, CS = 9/15) was 33 hours (SD = 10 hours) for the inoculated calves. Death or in extremis euthanasia occurred between 9 and 23 hours after the onset of illness (16 hours on average) (Table 1). In control calves, CS never decreased below 9/15 at any time. No lesions were detected in the bladders of any calves at death.

Table 1.   Blood, urine, and fecal positive cultures and results of the Uriscreen test in inoculated and control calves.
 Time of Blood and Urine Sampling (hours)*
InoculationOnset of Illness4–69–1216–1820–23
 InoculatedControlInoculatedControlInoculatedControlInoculatedControlInoculatedControlInoculatedControl
  • Numerator is the number of calves with positives results and denominator is the number of calves submitted (blood and feces) or with successful sample (urine).

  • NA, data not available (fecal sampling occurred at inoculation, at the onset of the illness and at death).

  • *

    Hours after the onset of clinical signs.

Blood
 Positive culture
  E. coli0/50/35/50/35/50/36/60/33/30/32/20/1
  E. coli O78+0/50/32/50/33/50/33/60/33/30/32/20/1
  Klebsiella spp.0/50/32/50/32/50/32/60/31/30/30/20/1
Urine
 Positive culture
  E. coli O78+0/40/20/30/31/20/21/50/32/30/31/20/1
  Klebsiella spp.0/40/21/30/31/20/22/50/31/30/31/20/1
 Positive Uriscreen test0/40/21/30/31/20/23/50/33/30/32/20/1
Feces
 Positive culture
  E. coli5/53/35/53/3NANANANA3/32/22/21/1
  E. coli O78+0/50/31/50/3NANANANA3/30/21/20/1
  Klebsiella spp.2/52/33/52/3NANANANA3/31/21/21/1

Bacteremia and Fecal Cultures

At inoculation time, all blood cultures were negative. After the onset of the illness, E. coli bacteremia was detected by at least 3 positive blood samples per calf in all 5 inoculated calves. In 4/5 inoculated calves, E. coli was identified as O78+ in at least 2 blood samples but not in calf #7. Moreover, E. coli O78− was isolated in blood cultures from calves #3 and 7 in at least 3 different samples, and Klebsiella spp. in at least 4 samples from calves #5 and 7 (Table 1). Cultures of liver and kidney showed the same bacteria as in the blood: E. coli O78+ (calves #1, 3, 5, and 8), E. coli O78− (calves #3 and 7) and Klebsiella spp. (calves #5 and 7).

In all 3 control calves, blood, liver, and kidney cultures were negative. All fecal E. coli cultures were positive. E. coli O78+ cultures became progressively positive in inoculated calves (Table 1).

Urine Bacteriology

Urine sampling was successful in 33 of the 42 cystocentesis samples for the overall experiment, and in 27 (15 in inoculated calves and 12 in control calves) of the 34 cystocentesis samples after the onset of illness (Table 1).

At inoculation time, urine cultures were negative in all inoculated and control calves. After the onset of the illness, E. coli was identified in the urine samples of all inoculated calves, and urine bacteria count increased with time, whereas cultures of control calves remained negative and urine bacteria count remained negative (Fig 1). After the onset of the illness, 5/15 urine cultures of inoculated calves were positive for E. coli O78+, and 5/5 samples of inoculated calves #5 and 7 were positive for Klebsiella spp. (Table 1). None of the urine cultures of control calves were positive for E. coli O78−, E. coli O78+, or Klebsiella spp.

image

Figure 1.  Urine culture results on inoculated and control calves. I, inoculated calves; Ct, Control calves; death occurred at the last sampling time; time refers to hours after the onset of clinical signs.

Download figure to PowerPoint

Catalase Activity Test (Uriscreen)

The concordance of the Uriscreen tests and blood or urine cultures was 25/29 (86%) and 29/33 (88%), respectively (Table 2). In 4 of the 5 inoculated and bacteremic calves, Uriscreen tests were positive at least twice. In bacteremic calf #8, only 1 Uriscreen test was performed and it was negative. In none of the control calves was a Uriscreen test positive.

Table 2.   Contingency table for Uriscreen test and bacteremia and bacteriuria.
  Blood CultureUrine Culture
  ++
Uriscreen test+121132
313216

Interrater agreements between Uriscreen tests and blood or urine cultures were high (κ= 0.73 and 0.76; 95% confidence intervals, [0.48–0.97] and [0.53–0.98], respectively). The sensitivity of the Uriscreen test for bacteremia and bacteriuria was 80.0 and 86.6%, respectively, but the corresponding specificity was 92.8 and 88.8%.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References

In this study, bacteremia was detected in all 5 inoculated calves and in none of the control calves. Colostrum-free calves are very susceptible to bacteremia after experimental inoculation.2 The pathogenic strain E. coli O78+ was reisolated in blood and organs from 4 of the 5 inoculated calves. However, bacteremia in calf #7 can be attributed only to bacteria considered as opportunistic, E. coli O78 and Klebsiella spp. Those bacteria could have colonized the digestive tract before the experimental inoculation, in accordance with the lack of colostral immune factors in this calf, and as suggested by the early bacteremia detected at inoculation time. These 2 factors may have played a role in the polybacterial infections detected in three of the calves too. The blood cultures demonstrated the presence of bacteremia, which corresponded with postmortem bacterial isolation from liver and kidney.1,2 Finally, bacteremia was present, enabling us to evaluate the Uriscreen test.

The results suggest an association between bacteriuria and bacteremia in all of the calves. In all of the inoculated (bacteremic) calves, urine bacterial counts were positive during the clinical phase and negative before. In all of the control (nonbacteremic) calves, urine bacterial counts remained negative. Moreover, the bacteremia was associated with bacteriuria in all of the calves for E. coli O78− and Klebsiella spp. (4/4 and 2/2, respectively), and in half of the calves (2/4) for E. coli O78+. In human adults with S. aureus bacteremia, the simultaneous detection of S. aureus in urine was shown in only 27% of patients.5

The bacteriuria appeared after bacteremia in all of the calves (Table 1), suggesting a contamination of the urine from the blood. In humans, bacteremia as a consequence of bacteriuria is reported,5 even if bacteremia as a cause seems to be more frequent. Nevertheless, mechanisms leading from bacteremia to bacteriuria are not well understood.5 The formation of microscopic abscesses in the kidney after a blood contamination has been reported.5

Uriscreen tests were positive in 4/5 inoculated calves, but in none of the control calves. Early death and low urine bacterial count could explain the negative result in inoculated calf #8. The false positive results (Table 2) observed in this study cannot be clearly explained. Catalase activity of erythrocytes or of inflammatory cells because of repeated cystocentesis7 probably can be excluded considering the negative peroxidase and leukocyte results on urine samples (data not shown), the lack of macroscopic lesions in the bladder and the negative Uriscreen tests in all of the control calves.

Kappa values between 0.6 and 0.8 are considered to represent good agreement, suggesting value in the use of Uriscreen to detect bacteriuria or bacteremia. Moreover, the sensitivity and the specificity of the Uriscreen test for bacteriuria and bacteremia are close to values reported in humans.6–10 Nevertheless, these evaluations were made on few calves.

Successive samplings in the same calf clearly show that cystocentesis was easy to perform on calves including males and without harmful effects on bladder. In the field, a single cystocentesis can be performed on an ill calf and used for Uriscreen testing. No data are available on the Uriscreen results when voided urine is used instead of samples collected by cystocentesis.

Our results suggest that the calf-side Uriscreen test can be used for the detection of bacteremia in neonatal calves in connection with constant bacteriuria. It would enable reduction in the use of parenteral antibiotics in calves with negative Uriscreen test results.

Footnotes

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References

a Elanco, Suresnes, France

b Mérial, Lyon, France

c Vétoquinol, Lure, France

d Terumo, Leuven, Belgium

e Braun, Melsungen, Germany

f Biomérieux, Marcy l'Etoile, France

g Oxoid, Basingstoke, United Kingdom

h Biovac, Beaucouze, France

i Savyon Diagnostic, Ashdod, Israel

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. References
  • 1
    Fecteau G, Pare J, Van Metre DC, et al. Use of a clinical sepsis score for predicting bacteremia in neonatal dairy calves on a calf rearing farm. Can Vet J 1997;38:101104.
  • 2
    Fecteau G, Van Metre DC, Pare J, et al. Bacteriological culture of blood from critically ill neonatal calves. Can Vet J 1997;38:95100.
  • 3
    Navetat H, Schmitt E, Rizet C, et al. Septicemia in the calf: A practitioner approach. In MailardR ed. European Meeting: From Research to Clinic. Paris: SFB; 2003:142150.
  • 4
    Malik A, Hui CP, Pennie RA, et al. Beyond the complete blood cell count and C-reactive protein: A systematic review of modern diagnostic tests for neonatal sepsis. Arch Pediatr Adolesc Med 2003;157:511516.
  • 5
    Lee BK, Crossley K, Gerding DN. The association between Staphylococcus aureus bacteremia and bacteriuria. Am J Med 1978;65:303306.
  • 6
    Berger SA, Bogokowsky B, Block C. Rapid screening of urine for bacteria and cells by using a catalase reagent. J Clin Microbiol 1990;28:10661067.
  • 7
    Waisman Y, Zerem E, Amir L, et al. The validity of the uriscreen test for early detection of urinary tract infection in children. Pediatrics 1999;104:e41.
  • 8
    Pezzlo MT, Amsterdam D, Anhalt JP, et al. Detection of bacteriuria and pyuria by URISCREEN a rapid enzymatic screening test. J Clin Microbiol 1992;30:680684.
  • 9
    Hagay Z, Levy R, Miskin A, et al. Uriscreen, a rapid enzymatic urine screening test: Useful predictor of significant bacteriuria in pregnancy. Obstet Gynecol 1996;87:410413.
  • 10
    Teppa RJ, Roberts JM. The uriscreen test to detect significant asymptomatic bacteriuria during pregnancy. J Soc Gynecol Investig 2005;12:5053.