HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Andersen, H. J. Articles by Chriél, M. Search for Related Content PubMed PubMed Citation Articles by Andersen, H. J. Articles by Chriél, M.

Evaluation of the Surveillance Program of Streptococcus agalactiae in Danish Dairy Herds

H. J. Andersen^*, L. H. Pedersen^*, F. M. Aarestrup and M. Chriél^*

^* Veterinary Department, Danish Dairy Board, Frederiks Allé 22, 8000 Aarhus C, Denmark
Department of Microbiology, Danish Veterinary Institute, Bülowsvej 27, 1790 Copenhagen V, Denmark

Corresponding author:
Hans Jrgen Andersen; e-mail:
hja{at}mejeri.dk.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
The aim of this study was to evaluate the Danish surveillance program of Streptococcus agalactiae in dairy herds with respect to 1) fluctuation over time of the presence of S. agalactiae in bulk tank milk, 2) sensitivity and specificity of the bacteriological method used, and 3) contamination of bulk tank milk samples with milk from other herds. From June to September 1996, bulk tank milk was sampled from 100 Danish dairy herds seven times, with intervals of 2 wk. The samples were examined for the presence of S. agalactiae by four different methods: 1) by the method approved for the program, 2) after ultrasonic treatment of the milk before examination, 3) after freezing down the milk before examination, and 4) after selective preparation of the milk. Selected strains of S. agalactiae were examined by restriction fragment length polymorphism of the gene encoding rRNA to discriminate between the isolates.

Streptococcus agalactiae was found in eight of 96 herds in which S. agalactiae had never previously been found during the surveillance program. Streptococcus agalactiae was not found in all seven sampling rounds in any of the eight herds. Comparing the approved method with supplemental findings by the other methods, the estimated sensitivity was (95% confidence limits): 0.786 (0.628; 0.892) and the estimated specificity (95% confidence limits): 0.995 (0.985; 0.999). Using all four methods on the same sample could increase the sensitivity, but by comparing the methods individually, there was no significant difference between any of them (P > 0.10).

In milk samples from three herds, the ribotype of S. agalactiae was the same as in milk from herds sampled just before; therefore, it could not be ruled out that cross-contamination could occur.

Taking into account that S. agalactiae in bulk tank milk reflects the presence of S. agalactiae in at least one udder quarter, this investigation gives further reason to assume that S. agalactiae can be seen sporadically in several herds. A surveillance program based on annual bulk tank milk sample examinations will only detect a limited number of S. agalactiae infected herds. If the overall aim is to identify herds where the infection is established, annual bulk tank milk sample examinations combined with the information of number of colonies of S. agalactiae in the sample will be sufficient.

Key Words: surveillance program • sensitivity • specificity

Abbreviation key: BTMS = bulk tank milk sample, QMS = quarter milk sample

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
In 1950, it was estimated that approximately 30 to 40% of the 184,000 milk-producing herds in Denmark were infected with Streptococcus agalactiae (Danish Veterinary Service, 1981). Therefore, in 1954 a nationwide surveillance program, based on bacteriological examination of milk can and bulk tank milk samples (BTMS), was initiated, hand in hand with an eradication program. The sampling from 1964 to 1995 was carried out with changing intervals from quarterly to every second year. From 1995 the BTMS were examined every year.

The eradication strategy was based on identifying infected cows by quarter milk samples (QMS) and subsequently treating or culling these cows, in combination with improving milking procedures, milking equipment, and hygiene measures to control spread of the infection within the herd (Danish Veterinary Service, 1992).

The eradication program was compulsory until 1988. After this time it became voluntary, but still with prohibition to sell cows and pregnant heifers from herds declared to be infected.

Results from the surveillance program from 1992 to 2000, in which bulk tank milk was sampled from all herds with an interval of 8 to 12 mo, identified 0.4 to 1.1% of all Danish dairy herds as newly infected during each examination (Danish Dairy Board, 2000).

Many studies have discussed the importance of the human and the bovine reservoirs of S. agalactiae in relation to bovine mastitis (Van den Heever and Giesecke, 1980; Jensen, 1982; Finch et al., 1984; Denning et al., 1989; and Jensen and Aarestrup, 1996). The discussion is particularly interesting when the prevalence of herds infected with S. agalactiae is low and when the focus is to find sources of the new introduction of S. agalactiae to dairy cows.

Human strains of S. agalactiae normally do not ferment lactose, but the ability to ferment lactose can be achieved after 8 to 10 passages in substrate containing lactose (Jensen, 1985). The same change must be assumed to take place in the mammary gland due to the content of lactose in milk. Records of the surveillance program in Denmark from 1992 to 1994, in which herds were sampled every 8 mo, showed on average 1.3 to 11.9% of the isolated strains to be lactose negative, which could imply a significant influence of the human reservoir, e.g., by continuously causing infections of cows in the herd. Furthermore, the chance of detecting S. agalactiae in BTMS and QMS is mainly determined by the number of colony-forming units (cfu) in the milk and the amount of milk cultured. The same records from 1992 to 1994 showed only 1 cfu in 12.7 to 20.2% of BTMS with S. agalactiae (Andersen and Huda, 1995).

This gave reason to believe that S. agalactiae could be found in more herds if the milk was sampled with shorter intervals, and if different methods of preparing the milk sample before culturing were used. Finally, during the surveillance program, it has been presumed that bulk tank milk with a very high number of S. agalactiae could cause subsequent samples to be contaminated.

Therefore, this study was initiated to evaluate the Danish surveillance program of S. agalactiae in dairy herds with respect to, 1) fluctuation over time of the presence of S. agalactiae in bulk tank milk, 2) sensitivity and specificity of the used bacteriological method, and 3) contamination of BTMS with milk from previously sampled herds.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Sampling Scheme
The use of BTMS in this study was approved by a board of farmers’ representatives, and the public authorities gave a written exemption from restrictions if S. agalactiae was found in the herds during the study. The owners were therefore not initially informed about the project to avoid deliberate influence on the results. The study should mirror the reality of the surveillance program, and it was decided to use BTMS collected routinely during each delivery of milk. Furthermore, this way of sampling had some obvious logistic and economic benefits compared to manual sampling.

BTMS from 100 dairy herds were collected seven times with a 2 wk interval during the period June to September 1996. The dairy herds delivered milk to the same dairy plant and, based on results from the national surveillance program, were chosen in a region representative of the national herd level of new infections with S. agalactiae.

The herds were located on 10 milk collection routes. The contents of the bulk tanks were from 126 to 7997 L; the median was 1184 L (25% lower 816, 75% upper 1902).

BTMS were sampled with the same procedure as under the national surveillance program (Danish Veterinary Service, 1992). During weighing-in the milk from the bulk tank to the lorry, the first 30 L of milk was routinely flushed through the milk hose and pipes in order to avoid contamination by milk residues from former sampling. Hereafter, 90 ml of milk was successively taken out by means of "Spentrup automatic sampling system" (Mark IV, Spentrup Machine works, Spentrup, Denmark) and stored in plastic test tubes.

The test tubes were marked for later identification and immediately stored on ice. Within 24 h the samples were transported to Steins Laboratory A/S (Hjaltesvej 8, 7500 Holstebro, Denmark) for examination.

Laboratory Examinations
Bacteriology on bulk tank milk.
Each BTMS was cultured using the commonly used ("approved") culture technique according to the Danish legislation (Danish Veterinary Service, 1992). For primary culture, 120 µl of milk was mixed with 9 ml of a streptococci-selective media (5% bovine blood agar (Blood agar base CM55, Oxoid, Hampshire, England) with 1% wt/vol esculin supplied with neomycin sulphate (Pharmacia & Upjohn, Copenhagen, Denmark) 0.212 µg/ml agar, polymyxin B sulphate (Sigma-Aldrich, Vallensbaek Strand, Denmark) 0.11 µg/ml agar, and sodium fusidate (Loevens Chemical Factory, Ballerup, Denmark) 0.25 µg/ml agar, and Staphylococcus aureus ß-toxin), plated into a petri dish, and incubated for 18 to 24 h at 37°C. The next day the agar plates and the remaining milk from the BTMS were transported to Steins Laboratory A/S, Ladelundvej 85, 6650 Broerup, Denmark, for examination and further analysis.

Colonies showing ß-hemolytic activity were counted (1, 2, 3, 4, 5, 5 to 10, 11 to 20, and >21) on each plate. One ß-hemolytic S. agalactiae suspected colony per plate was selected and cultured on 5% bovine blood agar containing esculin 1% (wt/vol) for 24 h at 37°C. Following this, isolates being positive in CAMP-test on bovine blood agar (Barrow and Feltham, 1993) and positive in a Lancefield group B latex agglutination test (Streptex, Murex Biotech Ltd., Kent, England) were identified as S. agalactiae. Finally, all S. agalactiae strains were tested for their ability to ferment broth-based lactose (Barrow and Feltham, 1993).

The remaining milk from each BMTS was split into three aliquots of which each was subjected to one of the following three culture techniques:

1) Preculture ultrasonic treatment of milk followed by culture using the standard culture technique: The milk was treated with an ultrasonic unit Elma Transsonic T 700/H (Elma-Hans Schmidbauer, Singen, Germany), 35 kHz for 20 min, according to the manufacturer’s instructions, and plated as described above for the standard method.

2) Preculture freezing of milk followed by culture using the standard culture technique: The milk was stored at -20°C for 24 h, thawed at room temperature, and plated after the standard protocol.

3) Preculture incubation of milk added a selective inhibitory substance followed by culture using the standard culture technique: A 10-ml sample was added to 1.1 µg of polymyxin B sulphate (Sigma-Aldrich) and incubated at 37°C for 4 h prior to standard culture.

All S. agalactiae strains were stored at -80°C in 50% glycerol and kept within the strain collection of the laboratory.

Testing for Cross-Contamination
The milk collection routes, the order of the samples taken from each sampling round, as well as the order of samples and examinations at the laboratories, were recorded.

Cross-contamination could occur as a result of milk remnants in the Spentrup sampling equipment from a previously sampled herd infected with S. agalactiae or during handling at the laboratories.

All recordings were scrutinized together with the results from the bacteriological examination and 29 subsamples with S. agalactiae were selected from the 15 herds. These strains of S. agalactiae, which could be suspected for cross-contamination, were sent to The Danish Veterinary Institute, Bülowsvej 27, 1790 Copenhagen V, Denmark, and examined by restriction fragment length polymorphism of the gene encoding rRNA to discriminate between the different isolates as described by Jensen and Aarestrup (1996).

Statistics
Sensitivity and specificity at the level of BTMS.
The commonly used and "approved" laboratory examination was evaluated by calculating the sensitivity and specificity of the method (Table 1), and classification of the results was performed on the level of subsamples.

(a/(a+c) in Table 1)

(d/(b+d) in Table 1)

Sensitivity and specificity at herd level.
A herd was classified positive, if S. agalactiae was found by one of the four diagnostic methods during the sampling period (Table 1). Herds in which S. agalactiae was isolated by only the approved method were defined as "unconfirmed positive herds." Herds in which S. agalactiae was isolated by any of the three other methods in at least one or more examination rounds and not by the approved method were defined as "false negative herds." Sensitivity and specificity was calculated as described at the level of BTMS.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Cross-Contamination of BTMS, and Bulk Tank Milk Subsamples
Ten different subtypes of S. agalactiae were found in subsamples from 15 herds. In three herds there was a marked coincidence between the same ribotype and a higher number of colony-forming units in the BTMS from herds just previously sampled in the same examination round. BTMS from these three herds were categorized as possibly cross-contaminated, and the number of herds suspected to be really infected was reduced to 12.

Ribotyping is considered to have a higher discrimination index than biochemical subtyping, and has been used earlier for the discrimination of strains of S. agalactiae (Denning et al., 1989; Jensen and Aarestrup, 1996). Together with comparison of the number of cfu/120 µl from the suspected contamination source herds (Table 2), it could be assumed that S. agalactiae from herds with a high number of S. agalactiae in bulk tank milk can be transferred to samples from following herds. This, in spite of flushing the sampling equipment with 30 L of milk before sampling, is initiated. The number of colony-forming units in the contaminated sample seems to be low, and by using only the approved method, only one of the four cross-contaminated samples would have been detected positive (Table 2).

Findings at the Level of BTMS
Seven hundred examinations of the BTMS by each of the four methods were performed during the study. The findings of S. agalactiae in BTMS are shown in Table 3. Three BTMS were positive in the subsample examined by the approved method but negative in the corresponding subsamples examined by the other three diagnostic methods.

A total of nine subsamples were negative by the approved method but positive in at least one of the other three methods (Table 3). The number of cfu/120 l counted in these samples were ten or higher in two cases and below five in the remaining seven cases.

Studies made by Villanueva et al. (1991) on QMS, have shown that freezing of the milk samples before culturing can increase the number of cfu/120 ml of S. agalactiae by 2.5 after splitting up the chain. Schukken et al. (1999) could not establish any similar increase. Murdough et al. (1996) demonstrated that the viability of S. agalactiae in QMS was not influenced by freezing. Ultrasound treatment was used in this study based on the same idea, but bulk tank milk is mechanically affected by both milking, pumping of the milk to the milk lorry, and in particular by the sampling procedure. This may cause the streptococci chains to be split up, even before special laboratory preparation. We found no significant increase in sensitivity by one method compared with other methods.

Improving the sensitivity of a surveillance program when a small number of bacteria are excreted normally requires improvement of the method or repeated examinations (Martin et al., 1987; Noordhuizen, 1997). The results from the surveillance program from 1992 to 2000 showed an annual new infection rate from 0.4 to 1.6% of all Danish dairy herds (Danish Dairy Board, 2000). The high percentage of herds pointed out as infected during the project must be attributed to both the repeated examinations and the use of supplementary methods, but the repeated examinations could also reflect the dynamics of the infection in the herds more than just a matter of diagnostic sensitivity.

Calculation of the sensitivity of testing for S. agalactiae is quite difficult because of the lack of a golden standard. The golden standard used for evaluating the approved method was defined and based on the test results from the other three applied methods. Based on this approximation, the sensitivity of the approved method was quite low, 0.786 at sample level and 0.800 at herd level (Table 4).

View larger version (21K): [in this window] [in a new window]   Figure 1. Dairy herds in which Streptococcus agalactiae were isolated in the bulk tank milk during seven examination rounds with 2-wk intervals. Numbers in columns refers to herd number. 100 herds examined in each round.

Based on the results, it is most likely that infected cows periodically shed S. agalactiae, and/or are continuously newly infected and self-cured or cured by treatment. Cows infected with human strains, were demonstrated to have a higher self-cure rate than cows infected with bovine strains (Jensen, 1982). The dynamics of the infection could be attributed to a continuous infection by the human strains or a reservoir outside the udder.

In herds 10 and 11 (Figure 1), lactose negative isolates of S. agalactiae were found. Samples from these two herds were only found positive for S. agalactiae by one examination round. In herd 4 both isolates fermenting lactose and isolates not fermenting lactose were found, indicating both established infection in the udder and newly introduced infection by a human source. During the surveillance program from 1992 to 1994, with nationwide bulk tank milk examinations every eighth month, 6.7% (1.3 to 11.9%) of the isolated strains of S. agalactiae were not fermenting lactose (Andersen and Huda, 1995). It must be reasonable to assume that these strains have origin in the human reservoir.

Streptococcus agalactiae does not multiply within the milking system and bulk tank at temperatures below 27°C (Gonzalez, 1986). Therefore, if no contamination from human sources has occurred, cows might have been infected with S. agalactiae from the human reservoir within the past 2 wk (Jensen, 1985). Self-cure rates for infections caused by strains not fermenting lactose are, based on results of experimental infections, assumed to be high, and the difference between the bovine and human strains with respect to numbers of bacteria excreted per milliliter of milk are significantly lower by human strains (Jensen, 1982). This could be of substantial influence on the dynamics of the infection.

In only one herd of the eight herds categorized as newly infected, S. agalactiae was detected in all bulk tank milk examination rounds. The farmers were not informed of the project and had no reason to act against the infection.

To run an effective eradication program such as the Danish national S. agalactiae program, it is essential to classify the status of the herds correctly. This study shows that a limited number of herds are incorrectly classified (false negative). These herds pose a risk for horizontal spread of the bacteria to noninfected herds by either animal movements or fomite transmission. Extrapolating from this study approximately 1% of all the samples in the Danish National S. agalactiae Program are false negative.

In Denmark the herds are submitted to restrictions when S. agalactiae are found in QMS or in bulk tank milk on two subsequent BTMS, as a consequence, farmers must be interested in a high specificity of the method. In relation to the surveillance program, the estimated specificity at herd level must be considered as acceptable, 0.995 (0.985; 0.999).

Each examination round can be considered as one result by the annual national surveillance program. Therefore, it is reasonable to believe that further examinations with 2-wk intervals will cause detection of even more herds infected with S. agalactiae. Furthermore, the infection will be permanently established in some herds and only transient in others.

	CONCLUSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
In Denmark one can expect to find approximately 1% of herds newly infected with S. agalactiae by each examination of bulk tank milk with 2-wk intervals. Streptococcus agalactiae can only be detected once in many of these herds, either because the infection does not establish in the herd, because the number of bacteria shed to the bulk tank milk is low, or because the shedding is fluctuating. Using several methods for examination on the same bulk tank milk sample will increase the number of findings of S. agalactiae in bulk tank milk. However, in order to identify all herds that are newly infected with S. agalactiae or all herds where a latent or persistent infection is flaring up, it is necessary to examine bulk tank milk from all herds with a frequency beyond what is realistic.

Cross-contamination of the bulk tank milk sample can occur during sampling due to milk residues from formerly sampled and infected herds in the sampling equipment. If the overall aim is to identify herds where the infection is established, annual bulk tank milk sample examinations combined with the information of number of colonies of S. agalactiae in the sample will be sufficient.

Received for publication May 27, 2002. Accepted for publication November 3, 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 


Aarestrup, F. M., H. C. Wegener, and V. T. Rosdahl. 1995. Evaluation of phenotypic and genotypic methods for epidemiological typing of Staphylococcus aureus isolates from bovine mastitis in Denmark. Vet. Microbiol. 45:139–150.[Medline]

Andersen, H. J., and A. Huda. 1995. Results of the Danish program to control Streptococcus agalactiae - 1992–1994. Dan. Vet. J. 78:1127–1130. (In Danish).

Barrow, G. I., and R. K. A. Feltham. 1993. Pages 219–238 in Cowan and Steel’s Manual for the Identification of Medical Bacteria. G. I. Barrow, and R. K. A. Feltham, ed. Cambridge University Press, Cambridge, UK.

Danish Dairy Board. 2000. Page 16 in Annual report from the Veterinary and Milk Quality Department, Danish Dairy Board, Frederiks Allé 22, 8000 Aarhus C, Denmark.

Danish Veterinary Service. 1981. Page 9 in Annual report on the mastitis eradication program. (In Danish). The Danish Veterinary and Food Administration, Moerkhoej Bygade 19, 2860 Soeborg, Denmark.

Danish Veterinary Service. 1992. App. 4 in Guidelines from The Danish Veterinary Service for the mastitis eradication program and laboratory examinations related hereto. (author’s transl.). The Danish Veterinary and Food Administration, Moerkhoej Bygade 19, 2860 Soeborg, Denmark.

Denning, D. W., C. J. Baker, N. J. Troup, and L. S. Tompkins. 1989. Restriction endonuclease analysis of human and bovine group B streptococci for epidemiologic study. J. Clin. Microbiol. 27:1352–1356.[Abstract/Free Full Text]

Finch, L. A., and D. R. Martin. 1984. Human and bovine group B streptococci: two distinct populations. J. Appl. Bacteriol. 57:273–278.[Medline]

Gonzalez, R. N., D. E. Jasper, R. B. Bushnell, and T. B. Farver. 1986. Relationship between mastitis pathogen numbers in bulk tank milk and bovine udder infections in California dairy herds. J. Am. Vet. Med. Assoc. 189:442–445.[Medline]

Hovmand, H. C., P. Livoni, P. Madelung, S. J. Olsen, O. Roemer, and S. O. Koch. 1954. Recommendations from the Mastitis Committee of the Danish Veterinary Association. Danish Veterinary Journal 7:105–116. (Author’s transl.).

Jensen, N. E. 1982. Experimental bovine group-B streptococcal mastitis induced by strains of human and bovine origin. Nord. Vet. Med. 34:441–450.[Medline]

Jensen, N. E. 1985. Epidemiological aspects of human/animal interrelationship in GBS. Antibiot, Chemother. 35:40–48.[Medline]

Jensen, N. E., and F. M. Aarestrup. 1996. Epidemiological aspects of group B streptococci of bovine and human origin. Epidemiol. Infect. 117:417–422.[Medline]

Martin, W. S., A. H. Meek, and P. Willeberg. 1987. Veterinary Epidemiology - Principles and Methods. Iowa State University Press, USA.

Murdough, P. A., K. E. Deitz, and J. W. Pankey. 1996. Effects of freezing on the viability of nine pathogens from quarters with subclinical mastitis. J. Dairy Sci. 79:334–336.[Abstract]

Noordhuizen, J. P. T. M., K. Frankena, and C. M. Hoofd. 1997. Application of Quantitative Methods in Veterinary Epidemiology. Wageningen Pers, Wageningen, The Netherlands.

Schukken, Y. H., J. A. H. Smit, F. J. Grommers, D. Vandegeer, and A. Brand. 1989. Effect of freezing on bacteriologic culturing of mastitis milk samples. J. Dairy Sci. 72:1900–1906.

Spentrup automatic milk sampling system with autocomputer Mark IV. 1990. Spentrup Machine works, Spentrup, Denmark.

Van den Heever, L. W. and W. H. Giesecke. 1980. Experimental induction of bovine mastitis with human strains of group B streptococci (Streptococcus agalactiae). J. S. Afr. Vet. Assoc. 51:107–109.[Medline]

Villanueva, M. R., J. W. Tyler, and M. C. Thurmond. 1991. Recovery of Streptococcus agalactiae and Staphylococcus aureus from fresh and frozen bovine milk. J. Am. Vet. Med. Assoc. 198:1398–1400.[Medline]

This article has been cited by other articles:

				H. W. Barkema, M. J. Green, A. J. Bradley, and R. N. Zadoks Invited review: The role of contagious disease in udder health J Dairy Sci, October 1, 2009; 92(10): 4717 - 4729. [Abstract] [Full Text] [PDF]

				L. L. Tikofsky and R. N. Zadoks Cross-Infection Between Cats and Cows: Origin and Control of Streptococcus canis Mastitis in a Dairy Herd J Dairy Sci, August 1, 2005; 88(8): 2707 - 2713. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Andersen, H. J. Articles by Chriél, M. Search for Related Content PubMed PubMed Citation Articles by Andersen, H. J. Articles by Chriél, M.